From e2d270e10c7aa8298ca98fd90f1dae60386e08c9 Mon Sep 17 00:00:00 2001
From: Github Actions <not@mail.com>
Date: Tue, 25 Jan 2022 16:47:44 +0000
Subject: [PATCH] Eddie Bergman: Merge pull request #1379 from automl/v0.14.4

---
 master/.buildinfo                             |    2 +-
 .../example_classification.ipynb              |    4 +-
 .../example_sequential.py                     |    3 +-
 .../example_extending_regression.py           |    3 +-
 .../example_multilabel_classification.ipynb   |    6 +-
 .../example_sequential.ipynb                  |    4 +-
 .../example_interpretable_models.py           |    4 +-
 .../example_regression.py                     |    4 +-
 .../example_multioutput_regression.py         |    3 +-
 .../example_extending_classification.py       |    3 +-
 .../example_successive_halving.py             |   12 +-
 .../example_interpretable_models.ipynb        |    4 +-
 .../example_extending_data_preprocessor.py    |    3 +-
 ...example_parallel_manual_spawning_python.py |  153 --
 .../example_classification.py                 |    4 +-
 .../example_extending_data_preprocessor.ipynb |    4 +-
 .../example_regression.ipynb                  |    4 +-
 .../example_multilabel_classification.py      |    5 +-
 ...mple_parallel_manual_spawning_python.ipynb |   90 --
 .../example_extending_preprocessor.py         |    3 +-
 .../example_get_pipeline_components.ipynb     |    6 +-
 .../example_parallel_manual_spawning_cli.py   |   10 +-
 .../example_resampling.ipynb                  |    2 +-
 .../examples_python.zip                       |  Bin 110298 -> 105390 bytes
 .../example_extending_classification.ipynb    |    4 +-
 ...example_parallel_manual_spawning_cli.ipynb |    2 +-
 .../example_resampling.py                     |    5 +-
 .../example_random_search.ipynb               |    6 +-
 .../example_successive_halving.ipynb          |   12 +-
 .../example_random_search.py                  |    5 +-
 .../example_multioutput_regression.ipynb      |    4 +-
 .../example_get_pipeline_components.py        |   17 +-
 .../example_extending_preprocessor.ipynb      |    4 +-
 .../example_extending_regression.ipynb        |    4 +-
 .../examples_jupyter.zip                      |  Bin 157773 -> 151577 bytes
 ...hx_glr_example_inspect_predictions_001.png |  Bin 16399 -> 15967 bytes
 ...hx_glr_example_inspect_predictions_002.png |  Bin 307238 -> 301152 bytes
 ...hx_glr_example_inspect_predictions_003.png |  Bin 39740 -> 41262 bytes
 ..._glr_example_inspect_predictions_thumb.png |  Bin 14594 -> 14304 bytes
 ...sphx_glr_example_pandas_train_test_001.png |  Bin 45025 -> 44871 bytes
 ...hx_glr_example_pandas_train_test_thumb.png |  Bin 37631 -> 37306 bytes
 ..._parallel_manual_spawning_python_thumb.png |  Bin 26794 -> 0 bytes
 .../sphx_glr_example_regression_001.png       |  Bin 54864 -> 54882 bytes
 .../sphx_glr_example_regression_thumb.png     |  Bin 54239 -> 53963 bytes
 master/_modules/autosklearn/estimators.html   |  182 ++-
 .../autosklearn/experimental/askl2.html       |   38 +-
 master/_modules/autosklearn/metrics.html      |   83 +-
 .../autosklearn/pipeline/components/base.html |   18 +-
 .../pipeline/components/classification.html   |    6 +-
 .../components/feature_preprocessing.html     |    6 +-
 .../pipeline/components/regression.html       |    6 +-
 master/_modules/index.html                    |    6 +-
 .../20_basic/example_classification.rst.txt   |  590 +++++---
 .../example_multilabel_classification.rst.txt |   57 +-
 .../example_multioutput_regression.rst.txt    |  103 +-
 .../20_basic/example_regression.rst.txt       |  139 +-
 .../20_basic/sg_execution_times.rst.txt       |   10 +-
 .../example_calc_multiple_metrics.rst.txt     |   78 +-
 .../40_advanced/example_debug_logging.rst.txt |   12 +-
 .../40_advanced/example_feature_types.rst.txt |    2 +-
 .../example_get_pipeline_components.rst.txt   |  897 +++++++-----
 .../example_inspect_predictions.rst.txt       |   22 +-
 .../example_interpretable_models.rst.txt      |   59 +-
 .../40_advanced/example_metrics.rst.txt       |   64 +-
 .../example_pandas_train_test.rst.txt         |    8 +-
 .../40_advanced/example_resampling.rst.txt    |   51 +-
 .../example_single_configuration.rst.txt      |   86 +-
 .../40_advanced/sg_execution_times.rst.txt    |   22 +-
 ...ample_parallel_manual_spawning_cli.rst.txt |   52 +-
 ...le_parallel_manual_spawning_python.rst.txt |  268 ----
 .../60_search/example_parallel_n_jobs.rst.txt |   14 +-
 .../60_search/example_random_search.rst.txt   |  959 +++++++-----
 .../60_search/example_sequential.rst.txt      |  510 ++++---
 .../example_successive_halving.rst.txt        | 1233 ++++++++++------
 .../60_search/sg_execution_times.rst.txt      |   26 +-
 .../example_extending_classification.rst.txt  |   61 +-
 ...xample_extending_data_preprocessor.rst.txt |   87 +-
 .../example_extending_preprocessor.rst.txt    |   48 +-
 .../example_extending_regression.rst.txt      |   41 +-
 ...restrict_number_of_hyperparameters.rst.txt |    2 +-
 .../80_extending/sg_execution_times.rst.txt   |   12 +-
 master/_sources/examples/index.rst.txt        |   21 -
 master/_sources/faq.rst.txt                   |  639 +++++---
 master/_sources/index.rst.txt                 |    7 +-
 master/_sources/installation.rst.txt          |   42 +-
 master/_sources/manual.rst.txt                |  411 +++---
 master/_sources/releases.rst.txt              |   48 +-
 master/_static/documentation_options.js       |    2 +-
 master/api.html                               |  438 ++++--
 .../20_basic/example_classification.html      |  590 +++++---
 .../example_multilabel_classification.html    |   70 +-
 .../example_multioutput_regression.html       |  104 +-
 .../examples/20_basic/example_regression.html |  147 +-
 .../examples/20_basic/sg_execution_times.html |   23 +-
 .../example_calc_multiple_metrics.html        |  106 +-
 .../40_advanced/example_debug_logging.html    |   37 +-
 .../40_advanced/example_feature_types.html    |   27 +-
 .../example_get_pipeline_components.html      |  935 +++++++-----
 .../example_inspect_predictions.html          |   51 +-
 .../example_interpretable_models.html         |   75 +-
 .../examples/40_advanced/example_metrics.html |  109 +-
 .../example_pandas_train_test.html            |   40 +-
 .../40_advanced/example_resampling.html       |   88 +-
 .../example_single_configuration.html         |  109 +-
 .../40_advanced/sg_execution_times.html       |   35 +-
 .../example_parallel_manual_spawning_cli.html |   69 +-
 ...ample_parallel_manual_spawning_python.html |  326 -----
 .../60_search/example_parallel_n_jobs.html    |   37 +-
 .../60_search/example_random_search.html      |  975 ++++++++-----
 .../60_search/example_sequential.html         |  519 ++++---
 .../60_search/example_successive_halving.html | 1286 ++++++++++-------
 .../60_search/sg_execution_times.html         |   37 +-
 .../example_extending_classification.html     |   75 +-
 .../example_extending_data_preprocessor.html  |  101 +-
 .../example_extending_preprocessor.html       |   62 +-
 .../example_extending_regression.html         |   55 +-
 ...le_restrict_number_of_hyperparameters.html |   31 +-
 .../80_extending/sg_execution_times.html      |   29 +-
 master/examples/index.html                    |  227 ++-
 master/extending.html                         |   37 +-
 master/faq.html                               |  433 ++++--
 master/genindex.html                          |    6 +-
 master/index.html                             |   39 +-
 master/installation.html                      |   88 +-
 master/manual.html                            |  281 ++--
 .../20_basic/example_classification.ipynb     |    4 +-
 .../example_multilabel_classification.ipynb   |    6 +-
 .../example_multioutput_regression.ipynb      |    4 +-
 .../20_basic/example_regression.ipynb         |    4 +-
 .../example_get_pipeline_components.ipynb     |    6 +-
 .../example_interpretable_models.ipynb        |    4 +-
 .../40_advanced/example_resampling.ipynb      |    2 +-
 ...example_parallel_manual_spawning_cli.ipynb |    2 +-
 ...mple_parallel_manual_spawning_python.ipynb |   90 --
 .../60_search/example_random_search.ipynb     |    6 +-
 .../60_search/example_sequential.ipynb        |    4 +-
 .../example_successive_halving.ipynb          |   12 +-
 .../example_extending_classification.ipynb    |    4 +-
 .../example_extending_data_preprocessor.ipynb |    4 +-
 .../example_extending_preprocessor.ipynb      |    4 +-
 .../example_extending_regression.ipynb        |    4 +-
 master/objects.inv                            |  Bin 7957 -> 7329 bytes
 master/releases.html                          |  321 ++--
 master/search.html                            |    6 +-
 master/searchindex.js                         |    2 +-
 145 files changed, 8904 insertions(+), 6709 deletions(-)
 delete mode 100644 master/_downloads/6507de4916ad481ba20cb60e6dc3309e/example_parallel_manual_spawning_python.py
 delete mode 100644 master/_downloads/986156f70b0b7eee4d669b791b35bd6d/example_parallel_manual_spawning_python.ipynb
 delete mode 100644 master/_images/sphx_glr_example_parallel_manual_spawning_python_thumb.png
 delete mode 100644 master/_sources/examples/60_search/example_parallel_manual_spawning_python.rst.txt
 delete mode 100644 master/examples/60_search/example_parallel_manual_spawning_python.html
 delete mode 100644 master/notebooks/examples/60_search/example_parallel_manual_spawning_python.ipynb

diff --git a/master/.buildinfo b/master/.buildinfo
index 298464334c..cbee93a719 100644
--- a/master/.buildinfo
+++ b/master/.buildinfo
@@ -1,4 +1,4 @@
 # Sphinx build info version 1
 # This file hashes the configuration used when building these files. When it is not found, a full rebuild will be done.
-config: 13a06760a87ac8d7b659b4aa6efbd203
+config: ba5b8e6ad6e1c8d0fcc11a6163f03902
 tags: 645f666f9bcd5a90fca523b33c5a78b7
diff --git a/master/_downloads/16fb4037a0b549292dbf3df70af70372/example_classification.ipynb b/master/_downloads/16fb4037a0b549292dbf3df70af70372/example_classification.ipynb
index 7cc4b7cd39..934f27b5c7 100644
--- a/master/_downloads/16fb4037a0b549292dbf3df70af70372/example_classification.ipynb
+++ b/master/_downloads/16fb4037a0b549292dbf3df70af70372/example_classification.ipynb
@@ -26,7 +26,7 @@
       },
       "outputs": [],
       "source": [
-        "import sklearn.datasets\nimport sklearn.metrics\n\nimport autosklearn.classification"
+        "from pprint import pprint\n\nimport sklearn.datasets\nimport sklearn.metrics\n\nimport autosklearn.classification"
       ]
     },
     {
@@ -98,7 +98,7 @@
       },
       "outputs": [],
       "source": [
-        "print(automl.show_models())"
+        "pprint(automl.show_models(), indent=4)"
       ]
     },
     {
diff --git a/master/_downloads/1a053e45a20a2c15032411b9fee890a3/example_sequential.py b/master/_downloads/1a053e45a20a2c15032411b9fee890a3/example_sequential.py
index b991802470..fad088396d 100644
--- a/master/_downloads/1a053e45a20a2c15032411b9fee890a3/example_sequential.py
+++ b/master/_downloads/1a053e45a20a2c15032411b9fee890a3/example_sequential.py
@@ -8,6 +8,7 @@
 sequentially. The example below shows how to first fit the models and build the
 ensembles afterwards.
 """
+from pprint import pprint
 
 import sklearn.model_selection
 import sklearn.datasets
@@ -48,7 +49,7 @@
 # Print the final ensemble constructed by auto-sklearn
 # ====================================================
 
-print(automl.show_models())
+pprint(automl.show_models(), indent=4)
 
 ############################################################################
 # Get the Score of the final ensemble
diff --git a/master/_downloads/23ae4950352edc8dd9ea5443ba77886b/example_extending_regression.py b/master/_downloads/23ae4950352edc8dd9ea5443ba77886b/example_extending_regression.py
index 7ee53cc975..3bdc008d4e 100644
--- a/master/_downloads/23ae4950352edc8dd9ea5443ba77886b/example_extending_regression.py
+++ b/master/_downloads/23ae4950352edc8dd9ea5443ba77886b/example_extending_regression.py
@@ -6,6 +6,7 @@
 The following example demonstrates how to create a new regression
 component for using in auto-sklearn.
 """
+from pprint import pprint
 
 from ConfigSpace.configuration_space import ConfigurationSpace
 from ConfigSpace.hyperparameters import UniformFloatHyperparameter, \
@@ -137,4 +138,4 @@ def get_hyperparameter_search_space(dataset_properties=None):
 # =====================================
 y_pred = reg.predict(X_test)
 print("r2 score: ", sklearn.metrics.r2_score(y_pred, y_test))
-print(reg.show_models())
+pprint(reg.show_models(), indent=4)
diff --git a/master/_downloads/2991959d1e025c5f9f27e3b4d3265a81/example_multilabel_classification.ipynb b/master/_downloads/2991959d1e025c5f9f27e3b4d3265a81/example_multilabel_classification.ipynb
index 341f5b333e..bf999ddaad 100644
--- a/master/_downloads/2991959d1e025c5f9f27e3b4d3265a81/example_multilabel_classification.ipynb
+++ b/master/_downloads/2991959d1e025c5f9f27e3b4d3265a81/example_multilabel_classification.ipynb
@@ -26,7 +26,7 @@
       },
       "outputs": [],
       "source": [
-        "import numpy as np\n\nimport sklearn.datasets\nimport sklearn.metrics\nfrom sklearn.utils.multiclass import type_of_target\n\nimport autosklearn.classification"
+        "import numpy as np\nfrom pprint import pprint\n\nimport sklearn.datasets\nimport sklearn.metrics\nfrom sklearn.utils.multiclass import type_of_target\n\nimport autosklearn.classification"
       ]
     },
     {
@@ -44,7 +44,7 @@
       },
       "outputs": [],
       "source": [
-        "# Using reuters multilabel dataset -- https://www.openml.org/d/40594\nX, y = sklearn.datasets.fetch_openml(data_id=40594, return_X_y=True, as_frame=False)\n\n# fetch openml downloads a numpy array with TRUE/FALSE strings. Re-map it to\n# integer dtype with ones and zeros\n# This is to comply with Scikit-learn requirement:\n# \"Positive classes are indicated with 1 and negative classes with 0 or -1.\"\n# More information on: https://scikit-learn.org/stable/modules/multiclass.html\ny[y == 'TRUE'] = 1\ny[y == 'FALSE'] = 0\ny = y.astype(np.int)\n\n# Using type of target is a good way to make sure your data\n# is properly formatted\nprint(f\"type_of_target={type_of_target(y)}\")\n\nX_train, X_test, y_train, y_test = sklearn.model_selection.train_test_split(\n    X, y, random_state=1\n)"
+        "# Using reuters multilabel dataset -- https://www.openml.org/d/40594\nX, y = sklearn.datasets.fetch_openml(data_id=40594, return_X_y=True, as_frame=False)\n\n# fetch openml downloads a numpy array with TRUE/FALSE strings. Re-map it to\n# integer dtype with ones and zeros\n# This is to comply with Scikit-learn requirement:\n# \"Positive classes are indicated with 1 and negative classes with 0 or -1.\"\n# More information on: https://scikit-learn.org/stable/modules/multiclass.html\ny[y == 'TRUE'] = 1\ny[y == 'FALSE'] = 0\ny = y.astype(int)\n\n# Using type of target is a good way to make sure your data\n# is properly formatted\nprint(f\"type_of_target={type_of_target(y)}\")\n\nX_train, X_test, y_train, y_test = sklearn.model_selection.train_test_split(\n    X, y, random_state=1\n)"
       ]
     },
     {
@@ -98,7 +98,7 @@
       },
       "outputs": [],
       "source": [
-        "print(automl.show_models())"
+        "pprint(automl.show_models(), indent=4)"
       ]
     },
     {
diff --git a/master/_downloads/2f293025f5b75af329925f2c3ff58c06/example_sequential.ipynb b/master/_downloads/2f293025f5b75af329925f2c3ff58c06/example_sequential.ipynb
index 4b401c5b90..89111256c1 100644
--- a/master/_downloads/2f293025f5b75af329925f2c3ff58c06/example_sequential.ipynb
+++ b/master/_downloads/2f293025f5b75af329925f2c3ff58c06/example_sequential.ipynb
@@ -26,7 +26,7 @@
       },
       "outputs": [],
       "source": [
-        "import sklearn.model_selection\nimport sklearn.datasets\nimport sklearn.metrics\n\nimport autosklearn.classification"
+        "from pprint import pprint\n\nimport sklearn.model_selection\nimport sklearn.datasets\nimport sklearn.metrics\n\nimport autosklearn.classification"
       ]
     },
     {
@@ -80,7 +80,7 @@
       },
       "outputs": [],
       "source": [
-        "print(automl.show_models())"
+        "pprint(automl.show_models(), indent=4)"
       ]
     },
     {
diff --git a/master/_downloads/34d1d3a5e629625f2f22f39d7c3a650e/example_interpretable_models.py b/master/_downloads/34d1d3a5e629625f2f22f39d7c3a650e/example_interpretable_models.py
index a9a4e015c5..a78695082c 100644
--- a/master/_downloads/34d1d3a5e629625f2f22f39d7c3a650e/example_interpretable_models.py
+++ b/master/_downloads/34d1d3a5e629625f2f22f39d7c3a650e/example_interpretable_models.py
@@ -7,6 +7,8 @@
 The following example shows how to inspect the models which *auto-sklearn*
 optimizes over and how to restrict them to an interpretable subset.
 """
+from pprint import pprint
+
 import autosklearn.classification
 import sklearn.datasets
 import sklearn.metrics
@@ -70,7 +72,7 @@
 # Print the final ensemble constructed by auto-sklearn
 # ====================================================
 
-print(automl.show_models())
+pprint(automl.show_models(), indent=4)
 
 ###########################################################################
 # Get the Score of the final ensemble
diff --git a/master/_downloads/42784b4d4739b840b1eeb6159bf08f4e/example_regression.py b/master/_downloads/42784b4d4739b840b1eeb6159bf08f4e/example_regression.py
index adfc390dab..6b47607db0 100644
--- a/master/_downloads/42784b4d4739b840b1eeb6159bf08f4e/example_regression.py
+++ b/master/_downloads/42784b4d4739b840b1eeb6159bf08f4e/example_regression.py
@@ -7,6 +7,8 @@
 The following example shows how to fit a simple regression model with
 *auto-sklearn*.
 """
+from pprint import pprint
+
 import sklearn.datasets
 import sklearn.metrics
 
@@ -43,7 +45,7 @@
 # Print the final ensemble constructed by auto-sklearn
 # ====================================================
 
-print(automl.show_models())
+pprint(automl.show_models(), indent=4)
 
 #####################################
 # Get the Score of the final ensemble
diff --git a/master/_downloads/49577cbea6f814a2c9a5167f35be2814/example_multioutput_regression.py b/master/_downloads/49577cbea6f814a2c9a5167f35be2814/example_multioutput_regression.py
index 5db733da0a..a2e345fcac 100644
--- a/master/_downloads/49577cbea6f814a2c9a5167f35be2814/example_multioutput_regression.py
+++ b/master/_downloads/49577cbea6f814a2c9a5167f35be2814/example_multioutput_regression.py
@@ -8,6 +8,7 @@
 *auto-sklearn*.
 """
 import numpy as numpy
+from pprint import pprint
 
 from sklearn.datasets import make_regression
 from sklearn.metrics import r2_score
@@ -46,7 +47,7 @@
 # Print the final ensemble constructed by auto-sklearn
 # ====================================================
 
-print(automl.show_models())
+pprint(automl.show_models(), indent=4)
 
 ###########################################################################
 # Get the Score of the final ensemble
diff --git a/master/_downloads/515ab036d01801cb08e4878be1aef556/example_extending_classification.py b/master/_downloads/515ab036d01801cb08e4878be1aef556/example_extending_classification.py
index 3c6c880a0c..b6132f4c18 100644
--- a/master/_downloads/515ab036d01801cb08e4878be1aef556/example_extending_classification.py
+++ b/master/_downloads/515ab036d01801cb08e4878be1aef556/example_extending_classification.py
@@ -6,6 +6,7 @@
 The following example demonstrates how to create a new classification
 component for using in auto-sklearn.
 """
+from pprint import pprint
 
 from ConfigSpace.configuration_space import ConfigurationSpace
 from ConfigSpace.hyperparameters import CategoricalHyperparameter, \
@@ -149,4 +150,4 @@ def get_hyperparameter_search_space(dataset_properties=None):
 
 y_pred = clf.predict(X_test)
 print("accuracy: ", sklearn.metrics.accuracy_score(y_pred, y_test))
-print(clf.show_models())
+pprint(clf.show_models(), indent=4)
diff --git a/master/_downloads/54625156807f9d21fa14b78034b1df5c/example_successive_halving.py b/master/_downloads/54625156807f9d21fa14b78034b1df5c/example_successive_halving.py
index 4f95296aef..fdb29da6e0 100644
--- a/master/_downloads/54625156807f9d21fa14b78034b1df5c/example_successive_halving.py
+++ b/master/_downloads/54625156807f9d21fa14b78034b1df5c/example_successive_halving.py
@@ -14,7 +14,7 @@
 To get the BOHB algorithm, simply import Hyperband and use it as the intensification strategy.
 
 """  # noqa (links are too long)
-
+from pprint import pprint
 
 import sklearn.model_selection
 import sklearn.datasets
@@ -110,7 +110,7 @@ def get_smac_object(
 )
 automl.fit(X_train, y_train, dataset_name='breast_cancer')
 
-print(automl.show_models())
+pprint(automl.show_models(), indent=4)
 predictions = automl.predict(X_test)
 # Print statistics about the auto-sklearn run such as number of
 # iterations, number of models failed with a time out.
@@ -143,7 +143,7 @@ def get_smac_object(
 automl.fit(X_train, y_train, dataset_name='breast_cancer')
 
 # Print the final ensemble constructed by auto-sklearn.
-print(automl.show_models())
+pprint(automl.show_models(), indent=4)
 automl.refit(X_train, y_train)
 predictions = automl.predict(X_test)
 # Print statistics about the auto-sklearn run such as number of
@@ -177,7 +177,7 @@ def get_smac_object(
 automl.fit(X_train, y_train, dataset_name='breast_cancer')
 
 # Print the final ensemble constructed by auto-sklearn.
-print(automl.show_models())
+pprint(automl.show_models(), indent=4)
 automl.refit(X_train, y_train)
 predictions = automl.predict(X_test)
 # Print statistics about the auto-sklearn run such as number of
@@ -208,7 +208,7 @@ def get_smac_object(
 automl.fit(X_train, y_train, dataset_name='breast_cancer')
 
 # Print the final ensemble constructed by auto-sklearn.
-print(automl.show_models())
+pprint(automl.show_models(), indent=4)
 predictions = automl.predict(X_test)
 # Print statistics about the auto-sklearn run such as number of
 # iterations, number of models failed with a time out.
@@ -245,7 +245,7 @@ def get_smac_object(
 automl.fit(X_train, y_train, dataset_name='breast_cancer')
 
 # Print the final ensemble constructed by auto-sklearn.
-print(automl.show_models())
+pprint(automl.show_models(), indent=4)
 predictions = automl.predict(X_test)
 # Print statistics about the auto-sklearn run such as number of
 # iterations, number of models failed with a time out.
diff --git a/master/_downloads/5f58bfc8fd90dc731fa26fb9996d2aa5/example_interpretable_models.ipynb b/master/_downloads/5f58bfc8fd90dc731fa26fb9996d2aa5/example_interpretable_models.ipynb
index b4b7eb5b6c..04a8cc6aba 100644
--- a/master/_downloads/5f58bfc8fd90dc731fa26fb9996d2aa5/example_interpretable_models.ipynb
+++ b/master/_downloads/5f58bfc8fd90dc731fa26fb9996d2aa5/example_interpretable_models.ipynb
@@ -26,7 +26,7 @@
       },
       "outputs": [],
       "source": [
-        "import autosklearn.classification\nimport sklearn.datasets\nimport sklearn.metrics"
+        "from pprint import pprint\n\nimport autosklearn.classification\nimport sklearn.datasets\nimport sklearn.metrics"
       ]
     },
     {
@@ -116,7 +116,7 @@
       },
       "outputs": [],
       "source": [
-        "print(automl.show_models())"
+        "pprint(automl.show_models(), indent=4)"
       ]
     },
     {
diff --git a/master/_downloads/60abb5e0c4b0861f5ecbe4ae9c2e51dd/example_extending_data_preprocessor.py b/master/_downloads/60abb5e0c4b0861f5ecbe4ae9c2e51dd/example_extending_data_preprocessor.py
index 6a92fa2bc9..7fdd72e971 100644
--- a/master/_downloads/60abb5e0c4b0861f5ecbe4ae9c2e51dd/example_extending_data_preprocessor.py
+++ b/master/_downloads/60abb5e0c4b0861f5ecbe4ae9c2e51dd/example_extending_data_preprocessor.py
@@ -5,6 +5,7 @@
 
 The following example demonstrates how to turn off data preprocessing step in auto-skearn.
 """
+from pprint import pprint
 
 import autosklearn.classification
 import autosklearn.pipeline.components.data_preprocessing
@@ -89,4 +90,4 @@ def get_hyperparameter_search_space(dataset_properties=None):
 
 y_pred = clf.predict(X_test)
 print("accuracy: ", sklearn.metrics.accuracy_score(y_pred, y_test))
-print(clf.show_models())
+pprint(clf.show_models(), indent=4)
diff --git a/master/_downloads/6507de4916ad481ba20cb60e6dc3309e/example_parallel_manual_spawning_python.py b/master/_downloads/6507de4916ad481ba20cb60e6dc3309e/example_parallel_manual_spawning_python.py
deleted file mode 100644
index 8e050537b7..0000000000
--- a/master/_downloads/6507de4916ad481ba20cb60e6dc3309e/example_parallel_manual_spawning_python.py
+++ /dev/null
@@ -1,153 +0,0 @@
-# -*- encoding: utf-8 -*-
-"""
-===================================================
-Parallel Usage: Spawning workers from within Python
-===================================================
-
-*Auto-sklearn* uses
-`dask.distributed <https://distributed.dask.org/en/latest/index.html>`_
-for parallel optimization.
-
-This example shows how to start the dask scheduler and spawn
-workers for *Auto-sklearn* manually within Python. Use this example
-as a starting point to parallelize *Auto-sklearn* across multiple
-machines. If you want to start everything manually from the command line
-please see :ref:`sphx_glr_examples_60_search_example_parallel_manual_spawning_cli.py`.
-To run *Auto-sklearn* in parallel on a single machine check out the example
-:ref:`sphx_glr_examples_60_search_example_parallel_n_jobs.py`.
-
-When manually passing a dask client to Auto-sklearn, all logic
-must be guarded by ``if __name__ == "__main__":`` statements! We use
-multiple such statements to properly render this example as a notebook
-and also allow execution via the command line.
-
-Background
-==========
-
-To run Auto-sklearn distributed on multiple machines we need to set
-up three components:
-
-1. **Auto-sklearn and a dask client**. This will manage all workload, find new
-   configurations to evaluate and submit jobs via a dask client. As this
-   runs Bayesian optimization it should be executed on its own CPU.
-2. **The dask workers**. They will do the actual work of running machine
-   learning algorithms and require their own CPU each.
-3. **The scheduler**. It manages the communication between the dask client
-   and the different dask workers. As the client and all workers connect
-   to the scheduler it must be started first. This is a light-weight job
-   and does not require its own CPU.
-
-We will now start these three components in reverse order: scheduler,
-workers and client. Also, in a real setup, the scheduler and the workers should
-be started from the command line and not from within a Python file via
-the ``subprocess`` module as done here (for the sake of having a self-contained
-example).
-"""
-
-import asyncio
-import multiprocessing
-import time
-
-import dask
-import dask.distributed
-import sklearn.datasets
-import sklearn.metrics
-
-from autosklearn.classification import AutoSklearnClassifier
-from autosklearn.constants import MULTICLASS_CLASSIFICATION
-
-tmp_folder = '/tmp/autosklearn_parallel_2_example_tmp'
-
-
-############################################################################
-# Define function to start worker
-# ===============================
-#
-# Define the function to start a dask worker from python. This
-# is a bit cumbersome and should ideally be done from the command line.
-# We do it here only for illustrational purpose.
-
-# Check the dask docs at
-# https://docs.dask.org/en/latest/setup/python-advanced.html for further
-# information.
-
-def start_python_worker(scheduler_address):
-    dask.config.set({'distributed.worker.daemon': False})
-
-    async def do_work():
-        async with dask.distributed.Nanny(
-            scheduler_ip=scheduler_address,
-            nthreads=1,
-            lifetime=35,  # automatically shut down the worker so this loop ends
-            memory_limit=0,  # Disable memory management as it is done by Auto-sklearn itself
-        ) as worker:
-            await worker.finished()
-
-    asyncio.get_event_loop().run_until_complete(do_work())
-
-
-############################################################################
-# Start Auto-sklearn
-# ==================
-#
-# We are now ready to start *auto-sklearn* and all dask related processes.
-#
-# To use auto-sklearn in parallel we must guard the code with
-# ``if __name__ == '__main__'``. We then start a dask cluster as a context,
-# which means that it is automatically stopped once all computation is done.
-if __name__ == '__main__':
-    X, y = sklearn.datasets.load_breast_cancer(return_X_y=True)
-    X_train, X_test, y_train, y_test = \
-        sklearn.model_selection.train_test_split(X, y, random_state=1)
-
-    # 1. Create a dask scheduler (LocalCluster)
-    with dask.distributed.LocalCluster(
-        n_workers=0, processes=True, threads_per_worker=1,
-    ) as cluster:
-
-        # 2. Start the workers
-        # now we start the two workers, one from within Python, the other
-        # via the command line.
-        worker_processes = []
-        for _ in range(2):
-            process_python_worker = multiprocessing.Process(
-                target=start_python_worker,
-                args=(cluster.scheduler_address, ),
-            )
-            process_python_worker.start()
-            worker_processes.append(process_python_worker)
-
-        # Wait a second for workers to become available
-        time.sleep(1)
-
-        # 3. Start the client
-        with dask.distributed.Client(address=cluster.scheduler_address) as client:
-            automl = AutoSklearnClassifier(
-                delete_tmp_folder_after_terminate=False,
-                time_left_for_this_task=30,
-                per_run_time_limit=10,
-                memory_limit=1024,
-                tmp_folder=tmp_folder,
-                seed=777,
-                # n_jobs is ignored internally as we pass a dask client.
-                n_jobs=1,
-                # Pass a dask client which connects to the previously constructed cluster.
-                dask_client=client,
-            )
-            automl.fit(X_train, y_train)
-
-            automl.fit_ensemble(
-                y_train,
-                task=MULTICLASS_CLASSIFICATION,
-                dataset_name='digits',
-                ensemble_size=20,
-                ensemble_nbest=50,
-            )
-
-        predictions = automl.predict(X_test)
-        print(automl.sprint_statistics())
-        print("Accuracy score", sklearn.metrics.accuracy_score(y_test, predictions))
-
-        # Wait until all workers are closed
-        for process in worker_processes:
-            process_python_worker.join()
diff --git a/master/_downloads/650cd19a8dfee813a7d7fb5b90341ed3/example_classification.py b/master/_downloads/650cd19a8dfee813a7d7fb5b90341ed3/example_classification.py
index 86fc09a5f4..fcb99b65ef 100644
--- a/master/_downloads/650cd19a8dfee813a7d7fb5b90341ed3/example_classification.py
+++ b/master/_downloads/650cd19a8dfee813a7d7fb5b90341ed3/example_classification.py
@@ -7,6 +7,8 @@
 The following example shows how to fit a simple classification model with
 *auto-sklearn*.
 """
+from pprint import pprint
+
 import sklearn.datasets
 import sklearn.metrics
 
@@ -42,7 +44,7 @@
 # Print the final ensemble constructed by auto-sklearn
 # ====================================================
 
-print(automl.show_models())
+pprint(automl.show_models(), indent=4)
 
 ###########################################################################
 # Get the Score of the final ensemble
diff --git a/master/_downloads/6dea5849db1f35abbefe123cd0eb49fd/example_extending_data_preprocessor.ipynb b/master/_downloads/6dea5849db1f35abbefe123cd0eb49fd/example_extending_data_preprocessor.ipynb
index 490787f633..bb3899d5a3 100644
--- a/master/_downloads/6dea5849db1f35abbefe123cd0eb49fd/example_extending_data_preprocessor.ipynb
+++ b/master/_downloads/6dea5849db1f35abbefe123cd0eb49fd/example_extending_data_preprocessor.ipynb
@@ -26,7 +26,7 @@
       },
       "outputs": [],
       "source": [
-        "import autosklearn.classification\nimport autosklearn.pipeline.components.data_preprocessing\nimport sklearn.metrics\nfrom ConfigSpace.configuration_space import ConfigurationSpace\nfrom autosklearn.pipeline.components.base import AutoSklearnPreprocessingAlgorithm\nfrom autosklearn.pipeline.constants import SPARSE, DENSE, UNSIGNED_DATA, INPUT\nfrom sklearn.datasets import load_breast_cancer\nfrom sklearn.model_selection import train_test_split"
+        "from pprint import pprint\n\nimport autosklearn.classification\nimport autosklearn.pipeline.components.data_preprocessing\nimport sklearn.metrics\nfrom ConfigSpace.configuration_space import ConfigurationSpace\nfrom autosklearn.pipeline.components.base import AutoSklearnPreprocessingAlgorithm\nfrom autosklearn.pipeline.constants import SPARSE, DENSE, UNSIGNED_DATA, INPUT\nfrom sklearn.datasets import load_breast_cancer\nfrom sklearn.model_selection import train_test_split"
       ]
     },
     {
@@ -98,7 +98,7 @@
       },
       "outputs": [],
       "source": [
-        "y_pred = clf.predict(X_test)\nprint(\"accuracy: \", sklearn.metrics.accuracy_score(y_pred, y_test))\nprint(clf.show_models())"
+        "y_pred = clf.predict(X_test)\nprint(\"accuracy: \", sklearn.metrics.accuracy_score(y_pred, y_test))\npprint(clf.show_models(), indent=4)"
       ]
     }
   ],
diff --git a/master/_downloads/77f2a85d2877495effe09ede1c7a1d31/example_regression.ipynb b/master/_downloads/77f2a85d2877495effe09ede1c7a1d31/example_regression.ipynb
index 3bfacf545e..acb7673ace 100644
--- a/master/_downloads/77f2a85d2877495effe09ede1c7a1d31/example_regression.ipynb
+++ b/master/_downloads/77f2a85d2877495effe09ede1c7a1d31/example_regression.ipynb
@@ -26,7 +26,7 @@
       },
       "outputs": [],
       "source": [
-        "import sklearn.datasets\nimport sklearn.metrics\n\nimport autosklearn.regression\nimport matplotlib.pyplot as plt"
+        "from pprint import pprint\n\nimport sklearn.datasets\nimport sklearn.metrics\n\nimport autosklearn.regression\nimport matplotlib.pyplot as plt"
       ]
     },
     {
@@ -98,7 +98,7 @@
       },
       "outputs": [],
       "source": [
-        "print(automl.show_models())"
+        "pprint(automl.show_models(), indent=4)"
       ]
     },
     {
diff --git a/master/_downloads/89647a1665eba015b7197cfe70420e4d/example_multilabel_classification.py b/master/_downloads/89647a1665eba015b7197cfe70420e4d/example_multilabel_classification.py
index b46caa2233..835b110ea6 100644
--- a/master/_downloads/89647a1665eba015b7197cfe70420e4d/example_multilabel_classification.py
+++ b/master/_downloads/89647a1665eba015b7197cfe70420e4d/example_multilabel_classification.py
@@ -8,6 +8,7 @@
 `here <https://scikit-learn.org/stable/modules/multiclass.html>`_.
 """
 import numpy as np
+from pprint import pprint
 
 import sklearn.datasets
 import sklearn.metrics
@@ -30,7 +31,7 @@
 # More information on: https://scikit-learn.org/stable/modules/multiclass.html
 y[y == 'TRUE'] = 1
 y[y == 'FALSE'] = 0
-y = y.astype(np.int)
+y = y.astype(int)
 
 # Using type of target is a good way to make sure your data
 # is properly formatted
@@ -65,7 +66,7 @@
 # Print the final ensemble constructed by auto-sklearn
 # ====================================================
 
-print(automl.show_models())
+pprint(automl.show_models(), indent=4)
 
 ############################################################################
 # Print statistics about the auto-sklearn run
diff --git a/master/_downloads/986156f70b0b7eee4d669b791b35bd6d/example_parallel_manual_spawning_python.ipynb b/master/_downloads/986156f70b0b7eee4d669b791b35bd6d/example_parallel_manual_spawning_python.ipynb
deleted file mode 100644
index 7851c9efc7..0000000000
--- a/master/_downloads/986156f70b0b7eee4d669b791b35bd6d/example_parallel_manual_spawning_python.ipynb
+++ /dev/null
@@ -1,90 +0,0 @@
-{
-  "cells": [
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "collapsed": false
-      },
-      "outputs": [],
-      "source": [
-        "%matplotlib inline"
-      ]
-    },
-    {
-      "cell_type": "markdown",
-      "metadata": {},
-      "source": [
-        "\n# Parallel Usage: Spawning workers from within Python\n\n*Auto-sklearn* uses\n`dask.distributed <https://distributed.dask.org/en/latest/index.html>`_\nfor parallel optimization.\n\nThis example shows how to start the dask scheduler and spawn\nworkers for *Auto-sklearn* manually within Python. Use this example\nas a starting point to parallelize *Auto-sklearn* across multiple\nmachines. If you want to start everything manually from the command line\nplease see `sphx_glr_examples_60_search_example_parallel_manual_spawning_cli.py`.\nTo run *Auto-sklearn* in parallel on a single machine check out the example\n`sphx_glr_examples_60_search_example_parallel_n_jobs.py`.\n\nWhen manually passing a dask client to Auto-sklearn, all logic\nmust be guarded by ``if __name__ == \"__main__\":`` statements! We use\nmultiple such statements to properly render this example as a notebook\nand also allow execution via the command line.\n\n## Background\n\nTo run Auto-sklearn distributed on multiple machines we need to set\nup three components:\n\n1. **Auto-sklearn and a dask client**. This will manage all workload, find new\n   configurations to evaluate and submit jobs via a dask client. As this\n   runs Bayesian optimization it should be executed on its own CPU.\n2. **The dask workers**. They will do the actual work of running machine\n   learning algorithms and require their own CPU each.\n3. **The scheduler**. It manages the communication between the dask client\n   and the different dask workers. As the client and all workers connect\n   to the scheduler it must be started first. This is a light-weight job\n   and does not require its own CPU.\n\nWe will now start these three components in reverse order: scheduler,\nworkers and client. Also, in a real setup, the scheduler and the workers should\nbe started from the command line and not from within a Python file via\nthe ``subprocess`` module as done here (for the sake of having a self-contained\nexample).\n"
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "collapsed": false
-      },
-      "outputs": [],
-      "source": [
-        "import asyncio\nimport multiprocessing\nimport time\n\nimport dask\nimport dask.distributed\nimport sklearn.datasets\nimport sklearn.metrics\n\nfrom autosklearn.classification import AutoSklearnClassifier\nfrom autosklearn.constants import MULTICLASS_CLASSIFICATION\n\ntmp_folder = '/tmp/autosklearn_parallel_2_example_tmp'"
-      ]
-    },
-    {
-      "cell_type": "markdown",
-      "metadata": {},
-      "source": [
-        "## Define function to start worker\n\nDefine the function to start a dask worker from python. This\nis a bit cumbersome and should ideally be done from the command line.\nWe do it here only for illustrational purpose.\n\n"
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "collapsed": false
-      },
-      "outputs": [],
-      "source": [
-        "# Check the dask docs at\n# https://docs.dask.org/en/latest/setup/python-advanced.html for further\n# information.\n\ndef start_python_worker(scheduler_address):\n    dask.config.set({'distributed.worker.daemon': False})\n\n    async def do_work():\n        async with dask.distributed.Nanny(\n            scheduler_ip=scheduler_address,\n            nthreads=1,\n            lifetime=35,  # automatically shut down the worker so this loop ends\n            memory_limit=0,  # Disable memory management as it is done by Auto-sklearn itself\n        ) as worker:\n            await worker.finished()\n\n    asyncio.get_event_loop().run_until_complete(do_work())"
-      ]
-    },
-    {
-      "cell_type": "markdown",
-      "metadata": {},
-      "source": [
-        "## Start Auto-sklearn\n\nWe are now ready to start *auto-sklearn* and all dask related processes.\n\nTo use auto-sklearn in parallel we must guard the code with\n``if __name__ == '__main__'``. We then start a dask cluster as a context,\nwhich means that it is automatically stopped once all computation is done.\n\n"
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "collapsed": false
-      },
-      "outputs": [],
-      "source": [
-        "if __name__ == '__main__':\n    X, y = sklearn.datasets.load_breast_cancer(return_X_y=True)\n    X_train, X_test, y_train, y_test = \\\n        sklearn.model_selection.train_test_split(X, y, random_state=1)\n\n    # 1. Create a dask scheduler (LocalCluster)\n    with dask.distributed.LocalCluster(\n        n_workers=0, processes=True, threads_per_worker=1,\n    ) as cluster:\n\n        # 2. Start the workers\n        # now we start the two workers, one from within Python, the other\n        # via the command line.\n        worker_processes = []\n        for _ in range(2):\n            process_python_worker = multiprocessing.Process(\n                target=start_python_worker,\n                args=(cluster.scheduler_address, ),\n            )\n            process_python_worker.start()\n            worker_processes.append(process_python_worker)\n\n        # Wait a second for workers to become available\n        time.sleep(1)\n\n        # 3. Start the client\n        with dask.distributed.Client(address=cluster.scheduler_address) as client:\n            automl = AutoSklearnClassifier(\n                delete_tmp_folder_after_terminate=False,\n                time_left_for_this_task=30,\n                per_run_time_limit=10,\n                memory_limit=1024,\n                tmp_folder=tmp_folder,\n                seed=777,\n                # n_jobs is ignored internally as we pass a dask client.\n                n_jobs=1,\n                # Pass a dask client which connects to the previously constructed cluster.\n                dask_client=client,\n            )\n            automl.fit(X_train, y_train)\n\n            automl.fit_ensemble(\n                y_train,\n                task=MULTICLASS_CLASSIFICATION,\n                dataset_name='digits',\n                ensemble_size=20,\n                ensemble_nbest=50,\n            )\n\n        predictions = automl.predict(X_test)\n        print(automl.sprint_statistics())\n        print(\"Accuracy score\", sklearn.metrics.accuracy_score(y_test, predictions))\n\n        # Wait until all workers are closed\n        for process in worker_processes:\n            process_python_worker.join()"
-      ]
-    }
-  ],
-  "metadata": {
-    "kernelspec": {
-      "display_name": "Python 3",
-      "language": "python",
-      "name": "python3"
-    },
-    "language_info": {
-      "codemirror_mode": {
-        "name": "ipython",
-        "version": 3
-      },
-      "file_extension": ".py",
-      "mimetype": "text/x-python",
-      "name": "python",
-      "nbconvert_exporter": "python",
-      "pygments_lexer": "ipython3",
-      "version": "3.8.12"
-    }
-  },
-  "nbformat": 4,
-  "nbformat_minor": 0
-}
\ No newline at end of file
diff --git a/master/_downloads/a23bc40b83b60b7e97a3eb188a82ad24/example_extending_preprocessor.py b/master/_downloads/a23bc40b83b60b7e97a3eb188a82ad24/example_extending_preprocessor.py
index a67528007d..9ac93a45b3 100644
--- a/master/_downloads/a23bc40b83b60b7e97a3eb188a82ad24/example_extending_preprocessor.py
+++ b/master/_downloads/a23bc40b83b60b7e97a3eb188a82ad24/example_extending_preprocessor.py
@@ -7,6 +7,7 @@
 discriminant analysis (LDA) algorithm from sklearn and use it as a preprocessor
 in auto-sklearn.
 """
+from pprint import pprint
 
 from ConfigSpace.configuration_space import ConfigurationSpace
 from ConfigSpace.hyperparameters import UniformFloatHyperparameter, CategoricalHyperparameter
@@ -130,4 +131,4 @@ def get_hyperparameter_search_space(dataset_properties=None):
 
 y_pred = clf.predict(X_test)
 print("accuracy: ", sklearn.metrics.accuracy_score(y_pred, y_test))
-print(clf.show_models())
+pprint(clf.show_models(), indent=4)
diff --git a/master/_downloads/a7f5a98ecf82e30e146cc9bfe484c397/example_get_pipeline_components.ipynb b/master/_downloads/a7f5a98ecf82e30e146cc9bfe484c397/example_get_pipeline_components.ipynb
index b26c9ab89c..33f2d673d0 100644
--- a/master/_downloads/a7f5a98ecf82e30e146cc9bfe484c397/example_get_pipeline_components.ipynb
+++ b/master/_downloads/a7f5a98ecf82e30e146cc9bfe484c397/example_get_pipeline_components.ipynb
@@ -26,7 +26,7 @@
       },
       "outputs": [],
       "source": [
-        "import sklearn.datasets\nimport sklearn.metrics\n\nimport autosklearn.classification"
+        "from pprint import pprint\n\nimport sklearn.datasets\nimport sklearn.metrics\n\nimport autosklearn.classification"
       ]
     },
     {
@@ -87,7 +87,7 @@
       "cell_type": "markdown",
       "metadata": {},
       "source": [
-        "## Report the models found by Auto-Sklearn\n\nAuto-sklearn uses\n`Ensemble Selection <https://www.cs.cornell.edu/~alexn/papers/shotgun.icml04.revised.rev2.pdf>`_\nto construct ensembles in a post-hoc fashion. The ensemble is a linear\nweighting of all models constructed during the hyperparameter optimization.\nThis prints the final ensemble. It is a list of tuples, each tuple being\nthe model weight in the ensemble and the model itself.\n\n"
+        "## Report the models found by Auto-Sklearn\n\nAuto-sklearn uses\n`Ensemble Selection <https://www.cs.cornell.edu/~alexn/papers/shotgun.icml04.revised.rev2.pdf>`_\nto construct ensembles in a post-hoc fashion. The ensemble is a linear\nweighting of all models constructed during the hyperparameter optimization.\nThis prints the final ensemble. It is a dictionary where ``model_id`` of\neach model is a key, and value is a dictionary containing information\nof that model. A model's dict contains its ``'model_id'``, ``'rank'``,\n``'cost'``, ``'ensemble_weight'``, and the model itself. The model is\ngiven by the ``'data_preprocessor'``, ``'feature_preprocessor'``,\n``'regressor'/'classifier'`` and ``'sklearn_regressor'/'sklearn_classifier'``\nentries. But for the ``'cv'`` resampling strategy, the same for each cv\nmodel is stored in the ``'estimators'`` list in the dict, along with the\n``'voting_model'``.\n\n"
       ]
     },
     {
@@ -98,7 +98,7 @@
       },
       "outputs": [],
       "source": [
-        "print(automl.show_models())"
+        "pprint(automl.show_models(), indent=4)"
       ]
     },
     {
diff --git a/master/_downloads/baf53fc945368668a0cd202acebc6220/example_parallel_manual_spawning_cli.py b/master/_downloads/baf53fc945368668a0cd202acebc6220/example_parallel_manual_spawning_cli.py
index fc03d52655..41200cd78c 100644
--- a/master/_downloads/baf53fc945368668a0cd202acebc6220/example_parallel_manual_spawning_cli.py
+++ b/master/_downloads/baf53fc945368668a0cd202acebc6220/example_parallel_manual_spawning_cli.py
@@ -11,11 +11,17 @@
 This example shows how to start the dask scheduler and spawn
 workers for *Auto-sklearn* manually from the command line. Use this example
 as a starting point to parallelize *Auto-sklearn* across multiple
-machines. If you want to start everything manually from within Python
-please see :ref:`sphx_glr_examples_60_search_example_parallel_manual_spawning_python.py`.
+machines.
+
 To run *Auto-sklearn* in parallel on a single machine check out the example
 :ref:`sphx_glr_examples_60_search_example_parallel_n_jobs.py`.
 
+If you want to start everything manually from within Python
+please see ``:ref:sphx_glr_examples_60_search_example_parallel_manual_spawning_python.py``.
+
+**NOTE:** Above example is disabled due to issue https://github.com/dask/distributed/issues/5627
+
+
 You can learn more about the dask command line interface from
 https://docs.dask.org/en/latest/setup/cli.html.
 
diff --git a/master/_downloads/bb7d59d9ddc2ff29f0d6eb99747a3347/example_resampling.ipynb b/master/_downloads/bb7d59d9ddc2ff29f0d6eb99747a3347/example_resampling.ipynb
index dbe33aaa65..55c4cb4f14 100644
--- a/master/_downloads/bb7d59d9ddc2ff29f0d6eb99747a3347/example_resampling.ipynb
+++ b/master/_downloads/bb7d59d9ddc2ff29f0d6eb99747a3347/example_resampling.ipynb
@@ -141,7 +141,7 @@
       },
       "outputs": [],
       "source": [
-        "resampling_strategy = sklearn.model_selection.PredefinedSplit(\n    test_fold=np.where(X_train[:, 0] < np.mean(X_train[:, 0]))[0]\n)\n\nautoml = autosklearn.classification.AutoSklearnClassifier(\n    time_left_for_this_task=120,\n    per_run_time_limit=30,\n    tmp_folder='/tmp/autosklearn_resampling_example_tmp',\n    disable_evaluator_output=False,\n    resampling_strategy=resampling_strategy,\n)\nautoml.fit(X_train, y_train, dataset_name='breast_cancer')"
+        "selected_indices = (X_train[:, 0] < np.mean(X_train[:, 0])).astype(int)\nresampling_strategy = sklearn.model_selection.PredefinedSplit(\n    test_fold=selected_indices\n)\n\nautoml = autosklearn.classification.AutoSklearnClassifier(\n    time_left_for_this_task=120,\n    per_run_time_limit=30,\n    tmp_folder='/tmp/autosklearn_resampling_example_tmp',\n    disable_evaluator_output=False,\n    resampling_strategy=resampling_strategy,\n)\nautoml.fit(X_train, y_train, dataset_name='breast_cancer')\n\nprint(automl.sprint_statistics())"
       ]
     },
     {
diff --git a/master/_downloads/bc82bea3a5dd7bdba60b65220891d9e5/examples_python.zip b/master/_downloads/bc82bea3a5dd7bdba60b65220891d9e5/examples_python.zip
index 63d76d25a987cf4bfbfd4fdef3c8b104bf0817c7..fc04d551d2f70b3dc43be612cefc52fff2ba00fc 100644
GIT binary patch
delta 3456
zcmb7G4NM!?9lr-QCMJNt%2#OmlCYW>f23JifrhA=R5dAU0>f+5b}1b8flt^z&v%%m
zqXe2}S*0mz`qI39Y%6G9QYb-PZ`~<rw~$GTl&mDnTJ5_jlcsf*npIQPC?GZ3|J{MX
zI3e9xf9vkO-~aQy|HofHD45g=*0i?dn_L`wz8Go@Zay_I_qv7S7<tG)>@G4<{dr|~
zQi|b3A}PidEXER2Qknl!M{XZCk!|KGs~3y$un<?8?<S+ZHZo!ESL-%-ko|kZjll`a
zm49z8;y6Y=CX`MeL)Fy!6k4mkU$6#M-BiAn%)DH$))ZdMQ?FXX%QeyOd<*SXe^+YG
zv%dqjpZHsUU0VUiF<Q^m91gs_vgl1}^xy$RMur;57FQXaaiaaIl>Z*Ou}9~M?*(U0
zYHIDMmE%Uq2lmsq5Kp29UFwI9pXKKYtR-XShg3&-pV@R}e`8Sn(>8Ot=?724Tjd)i
zcGCs+_LtEuC5{=`HTTYKzvm8)W8~*3{~9Wz7muSVmDX&^yM+x4E|#dz*8bGAFD!->
zQHt})y|_OjBn8~r8I!_7v_lMcc4Dd9x&aG(D1!AxEXtV2y~18E=Hp?U;-h_nv9?4^
zNQx`G7#HI`a0R-hWQ=DDK#tUnl?bmGqWpLpdwE2Nx@F9`@jRc5MMVZH+zV#7J3GDX
zJITj;**i#ruaG1w3p+ww7Ghmdp`%|Adm{QWk$hzhI2aK!h67uVm=fZ+YcH6={B(zT
zh3`ltg+x*c39>9D=LM3(*UI)~G$n<eq_G=thoZbJi`}BYpn42IjNBU)_+-3eDR5ye
zi-E;5t|Uc4_T#VjDHwvYU`i;(1cDwumWZ;@k?EtBeGq46Dl7^*SU-tS3gl#tLsq1u
z5Qa!END^d4gpf-~nMsU_GNh=n$)XGg7nMMKzo<kQhY`$_q=3#2T_|w*tyVp`$^GWC
zC5fz#`Zl{vQ*h_ix-(C$a@gP|-eH!1G2rAlM(d25s7@XDXNgl5qC!X!!W~csM2IeK
z#$@(LZLy!1mAwg}7D~rPD?vw{L_5-CUyioYnIiNU#hcJ7GWbLTnR?+hvVFFMl+3xv
zNn5L`9S!E0e#<N=ed%nW=`72R@Ao@Oiuc1>{YY2Z3#G-Y-WT0~DsEOxNL%4oXy8#~
zp}(@Djp|#mR|<2}l2xu>lb@Won#@ec%<gulGw(oS@RDzqA79IHjPBw%+BpFQY?lv}
zsaMY1@{pBY_M&n!KAj@h-+jQ++1Zp7x|`_W-Kd#fo<OU*^&opZJ9h=|Z}NEXwk|0p
zVBx#$^!1Dii!u*4I*h}RVTuGvCc{faQ4(@fAkYJ+*Vp9_NwGkfmwN+%hHKT;2k{Q*
zNI6jdwaxcf)wLI@*O;DurZK3U^dO(b(GI5X8T>>0?42CP$j#B%>)?Ff$Bz7!>IT+Y
zN$j9Pb{@B=?N_cAEen)pFGa;jmFBK3uWahj^{wV*XjmISTDuutT8=j45b1QQ_OcZ{
zvAn*s?Q-NS11CP3di0px!Eub%L;P(MXoGgX0{wl(@`ts@ep!u13YHl(uXgJX+qCi;
z6xHo5T5Mn&$y(cwXrl)W6rX`m{?W#N5-jAnpjK}~{EE%W2-_h#bQD!-haBk5<sOm4
zTfPZ5EJX_DME#bX1>ZF1^T_9_Txi1=VaZ)$<?(GC$7mD9&!Y<M(*fj1H}Aw+^&q<B
z+^YBJ+EBFHFGr+)Bg5s|jZ2)fE;jsgpy~6{x%00;GBffTngSxKT0s*9O(K16HzdHw
zAgZAgyHT^|??oy5ymqLIUq-tM+8;sbW>+!CG5V^Zecuk`xpfk|>5(?%(9A7JD*FGN
zejC)jm;U&H9dMHv`3<$FA4cM>oYuhUpzXdd6=Rg=%D>52MIS$g*27#?Nw1$kt8vbR
zqz_}pUwZg63+0ca3i4_}IX!d?;5m~{7Chl(aOdHG-9TqeEHJe|`N}-$VN|7~Y3*@T
ziP;#^{B8HYrpq{vQ8x5)jWKrA!R(>WDfGc{RF(J9z4|g~e#1qd8%L}4!G->J47O&!
zVQhUdT27l_3+GNU8B@6QaO2~dx#mv2>Ch3T;GHK8DZ9?tysa&ZQO#{eONBK*V{lC!
zD9++yE#%F-w}c_cXGY#dCNSX|HB2}+hH5bDF&4|&miOUmGOEdf{$|8bFy3#Y4P)S3
zRwIIDRF?(+^CW{;kUi1cvg~0^rsUnoSSe&5BTp9BL#G&5!@3(ZpbRcOoSt6Nj(kgF
z#2Ufyi&sy43Ij2tY!`m~iOzQy&TSGI=@66)pRtxvk8e<gGqCjV>NWJ|6UaqRRiNy1
z%Br9cahPB;PkMNd&tThWL&dhGrn9<8qmys}{!>qXe*%@$S0|C%vv~eFRWc)$1Ebv!
zGNyI3WfI+vbIM#6^o0%vZNfR_C=2zZYzTaI2RON%rEN|f$-;j)WZ=&|j7rN|Tg(uB
Pj;nxQ4a7hp2)_OY55F(K

delta 4343
zcmZ`+3vd+W9p5jxJUEhYmxMenuP+bDUA&wjL5h*65vEFc2!^(F0y#E!d)$`0-Sc)Y
zN3cv9M`_15{0R$aHCS}os)#*1*AZp}JHv=65PS?%MzppSh^8X7($;7H-)`=5+t|%~
zncMw-|Ht?HKX(5Iley>4=a#Qto^2`R*cE$R?D_eTtNX9b<+y9%XC-e!5y^vh4%^|A
zd7UtRIh=T;FPp&4S6bkxcO884N_XP(vNs8Q(w+kYfzrg43J0;b|3d89@znaqZL>L!
ztyb+xAeEYJ7b#ENlz#((8@Jiuy$v*RY}Q*@*;2%l7%A{uQaWG_zgGLco46AD?0+VO
zpQ1-2!=IhI+sbiwLxtm&43DnWs{2WC;<hTtPD}H}PB>J&GNIJPa_qlD;os~!F}yvG
z<Jf9pp0q-j!?`2!lEyD0wTboQfoaoMDqc&xncG5&T1Lel+xGciwAXN4dre|ZuRjak
zK3tK(!ZSI&AyHAO9H43}5)u?Q6?k9JxS{=mC~dMfQZ*>hu<RE?G%kpnf+`gy{h1&s
zLbyL9P+sz9fJ9XhLIID}3Xy;Vu2if^d=PlQ#IhK@NlZ57lw1CQ)Fygg?kOqi#9yty
z?0hK~3C^7j_LYu=^@V(*{X-=#_>aSt7&`P=eu@>0%Nzoc607CDCs4uoXmL^TdHi+i
zFKyimN;r<KdQRIhLdp{Iy?-mx&J>UmI0&8j7@c@fhzbw@XHIY12CEDB2QI_Jt{&L%
z`#TZ`N7iOpwlSx2_Z-i+>|omSRxU2GEM>nwyRpb_*}`^T-@0v9;lO^er_|XwzND7p
zmLv|JygiFh*fD+^tQqfs^CxXkeZ~r>KdykP@qxrcXB<`bb10!)Gxp%G=W!fc&G5>}
zxrwhoZO+!ZYss3OeQF~6-(Stv#+!(qfAH!f&B^5!@~#E$3REHy-MeXWyomIY<OwTz
zHQW5uox|=^G7Cv^+g#G8OR{Gg>bbx;voAm|jy!#=`?`FNW2=sNp&i^y3X}U<$&Ez2
zwU(^j+4nGv#vSln%hi!bNm25nHgYgCD__%Ivk+0&H;VdBM82`SZ|69+8gzYqdx<T%
zqKdqel|+4?=casPeq2mS^T-k1-Q+4asU((5EG3?uY|XPjh}v?M(eKoja~xX>P@4y-
zoCTKs9UAqMI=KDRyu!Gs1~K_o52`^~(#|X-3*gX7I}HAP8O%97Y~F^aPxtAe@xaJw
zTY(kRn^y#?2m+moRT<fsj?|ZD@^XsAa72!(lvf5NpD088Y#sdStR0>|`%zK5SJ=Xb
zQHNIzN1CC&uculFh>}19G0CTjGFF646_rQvxE$RK-DP#Kad9qe{F5cT)IhQ_nu+GA
zpI0{PLPC@VqH>r<^y7Kx+MuXl`H2eUX}_paUo6}&L=`z~RGbo&V<A5k{Q@5f4Pv!Y
zzbpyx$m#M`DD^{jOHqm?K9waz1mq|cL!p?WMtSBkAEJ?1G$Jbk7HcD25)^!!^*Yu)
z^2<Intl|cYotZaeIMvGg2YAUR_&q^29MUxgVo^L<lsPJ)g|My~75asML5SB7^)g<B
zs8jI;1%E6gM7_M<9~BhE)oG=8F{2)zECs|(9z{@{cQ^a7HATh#n2P3x3~7&_7s9gC
z+)2O7hZJF(%c{%j5u?naU)Jq7r<5|q#wJT0!$zjsRlFn(I&~ck%C2eu#7I~AW+>&Y
z+XR->U{v7!O4oez^N<)2R52`cbzJYJw2|^LRc6fi^iV0mn2MgpjW%kWj4Cp6kFgHP
za)b(!UojaP3gfw>gWixB7S*mcUFlL$;aU4NwrH4__)S7skXQz@$fGir5ww4hcBB1P
zWiysQRKhLRpwsGI3~o4KM8e<!iHq~7#n`#cgV`=B7&>Pr=c4S{B&c3t0IhnN=T4U=
z8k4*+Nfkq0pUf&o6`bj)xUh2WgmV|>!sVJ`c=T9K2=kQ31eN4CWlRoE*HH^2E#-8J
z{*poVjB<$zAs*9$MxwG0Q%F!e)<(?0wK9!ijb{q7msNz1qNEUlp)2LHD9Z%IZq#ZU
zNY&6r4=WYSU}>s+rwY=?>I5qix#20Mr7CPu-6$3hiui&I3%sPD6pt+HME_5KRXGwt
z_hc*;%A96Q#8kaBQep8}#Q^n!YqARJFTOa`NauU#5~M&y3FDCI<fG0NGG@*a!zGs<
zxo^+sOx$!@OH;L{u*6AI*0nYo6Wn+R#`}=sjbKR{!A@nfo*7d_bQ+b$a-)&9duWf|
z>sU`SWTw-#Q4eMuTWuPm#^n^ujig{7WPL8Po0mz<6Kqs^FG*!!BfR%Uc|+>XFyu`i
zfOgT|4LTX{;a}_EmdQ$(Jm0h~1xM?cn_apW<{o6ORfb2X(&d~wswYi+y6CCPZH3XT
z4p=%_X45rd8Kexs)^``d!FMZmen8>C`I45@^G4c{$&)W6VwI$&*y&;Ah029fZLsVA
zTAvC7lb$Y6ICar&VV2;Q=UNL=CiMsQ1=Y5WHuE91`f^fd-~*Q$Vfa!rj9zlUZ!ax#
zq+})UkL7+-Cxe`cURXDQr`lQu-8EISr%GS9^7ur3WlD6=i?0+>nl=f?C)(iSiDoGL
zumPGstSd<&Gd0|G|26rk=g?i;kiOCOUr$!fN;3tEK58;t=O<e+%=J~%-}<*cCH6eD
z;-9-N7jPU~9oiu;Y0#XHk<#RtLJCOoib}?2Yj*-Et8rPU1y`Zqi}9w-O~o9?mPZ#%
zRu2;|(VkjPg2}J$A!}{AU}dJ@dK9ETYQH*wpPK{8=e%U0f!7dUo%|_J&f43eqNF;b
z0=~$Ca@?zbJ1b6CrWQYH`R3w#V_)FN@<nO#=I@bZB#CCnOLU92Ozb+uCcnI%w3b$p
ztkUdn8FFyYY0!26X@Dn7?C#aer(rY37=1!A{v8<JoU4rjsbEu0BQ*^vIH=$UZu&Yf
zeB2H-4yyeTNCBNOk)&}PcVKv1HB-8Xn#T`h*o0c%q<^XJ-c9CFHX3xTZ@zG{jN{lc
zTRzdPBf)l!Rhu(H3iSCw8`p><>y35B*3ipkTE!mp%sf!w_-C@|3K?+g3+38T4dHB%
z$jdvv{H;QcW2?yo-xAllHPTLJ4jI$11H%vBXDF5LEz}P0M)%kxp^fcE_sCv-OZ#d!
zX`*bbxbE#E@GrE(mf5wj{S118K5}SvBPcxkkfFD7aj7jcqHGk=TA2EpV+P{mtu`%$
zmZ@nrK@#Ur`-yRR;&nKBvcr@eGe#4WK-&Zgl)~4a&edx7l7jjfBZLWEYa?*%WC<H7
zD(h!V2PWwK6$TWih3a}X1Pt%F`ut@~Vz$gNc(#Q>tF@zh$lUrFQ-VqBJM$UT&87y^
zA!oD*lhlf3+8z5yJ?j{h&S(lI=z~8npx7hkXp0|1!cDyZ`+?hp|1xOcPwgX>)Lif!
u*Jgqq`3ZyGWMQSQjqSzY=Imro%C)celIGGJ<b~rZ@ozhhZZ|zc@b16IJa78|

diff --git a/master/_downloads/c4ec137bb92db1e6b2219eb13818b898/example_extending_classification.ipynb b/master/_downloads/c4ec137bb92db1e6b2219eb13818b898/example_extending_classification.ipynb
index 08d5d78bf2..2a97dc7eac 100644
--- a/master/_downloads/c4ec137bb92db1e6b2219eb13818b898/example_extending_classification.ipynb
+++ b/master/_downloads/c4ec137bb92db1e6b2219eb13818b898/example_extending_classification.ipynb
@@ -26,7 +26,7 @@
       },
       "outputs": [],
       "source": [
-        "from ConfigSpace.configuration_space import ConfigurationSpace\nfrom ConfigSpace.hyperparameters import CategoricalHyperparameter, \\\n    UniformIntegerHyperparameter, UniformFloatHyperparameter\n\nimport sklearn.metrics\nimport autosklearn.classification\nimport autosklearn.pipeline.components.classification\nfrom autosklearn.pipeline.components.base \\\n    import AutoSklearnClassificationAlgorithm\nfrom autosklearn.pipeline.constants import DENSE, SIGNED_DATA, UNSIGNED_DATA, \\\n    PREDICTIONS\n\nfrom sklearn.datasets import load_breast_cancer\nfrom sklearn.model_selection import train_test_split"
+        "from pprint import pprint\n\nfrom ConfigSpace.configuration_space import ConfigurationSpace\nfrom ConfigSpace.hyperparameters import CategoricalHyperparameter, \\\n    UniformIntegerHyperparameter, UniformFloatHyperparameter\n\nimport sklearn.metrics\nimport autosklearn.classification\nimport autosklearn.pipeline.components.classification\nfrom autosklearn.pipeline.components.base \\\n    import AutoSklearnClassificationAlgorithm\nfrom autosklearn.pipeline.constants import DENSE, SIGNED_DATA, UNSIGNED_DATA, \\\n    PREDICTIONS\n\nfrom sklearn.datasets import load_breast_cancer\nfrom sklearn.model_selection import train_test_split"
       ]
     },
     {
@@ -98,7 +98,7 @@
       },
       "outputs": [],
       "source": [
-        "y_pred = clf.predict(X_test)\nprint(\"accuracy: \", sklearn.metrics.accuracy_score(y_pred, y_test))\nprint(clf.show_models())"
+        "y_pred = clf.predict(X_test)\nprint(\"accuracy: \", sklearn.metrics.accuracy_score(y_pred, y_test))\npprint(clf.show_models(), indent=4)"
       ]
     }
   ],
diff --git a/master/_downloads/c6746d1b897496495baebd219e94d74e/example_parallel_manual_spawning_cli.ipynb b/master/_downloads/c6746d1b897496495baebd219e94d74e/example_parallel_manual_spawning_cli.ipynb
index 0da3e8120c..0c7b01d234 100644
--- a/master/_downloads/c6746d1b897496495baebd219e94d74e/example_parallel_manual_spawning_cli.ipynb
+++ b/master/_downloads/c6746d1b897496495baebd219e94d74e/example_parallel_manual_spawning_cli.ipynb
@@ -15,7 +15,7 @@
       "cell_type": "markdown",
       "metadata": {},
       "source": [
-        "\n# Parallel Usage: Spawning workers from the command line\n\n*Auto-sklearn* uses\n`dask.distributed <https://distributed.dask.org/en/latest/index.html>`_\nfor parallel optimization.\n\nThis example shows how to start the dask scheduler and spawn\nworkers for *Auto-sklearn* manually from the command line. Use this example\nas a starting point to parallelize *Auto-sklearn* across multiple\nmachines. If you want to start everything manually from within Python\nplease see `sphx_glr_examples_60_search_example_parallel_manual_spawning_python.py`.\nTo run *Auto-sklearn* in parallel on a single machine check out the example\n`sphx_glr_examples_60_search_example_parallel_n_jobs.py`.\n\nYou can learn more about the dask command line interface from\nhttps://docs.dask.org/en/latest/setup/cli.html.\n\nWhen manually passing a dask client to Auto-sklearn, all logic\nmust be guarded by ``if __name__ == \"__main__\":`` statements! We use\nmultiple such statements to properly render this example as a notebook\nand also allow execution via the command line.\n\n## Background\n\nTo run Auto-sklearn distributed on multiple machines we need to set\nup three components:\n\n1. **Auto-sklearn and a dask client**. This will manage all workload, find new\n   configurations to evaluate and submit jobs via a dask client. As this\n   runs Bayesian optimization it should be executed on its own CPU.\n2. **The dask workers**. They will do the actual work of running machine\n   learning algorithms and require their own CPU each.\n3. **The scheduler**. It manages the communication between the dask client\n   and the different dask workers. As the client and all workers connect\n   to the scheduler it must be started first. This is a light-weight job\n   and does not require its own CPU.\n\nWe will now start these three components in reverse order: scheduler,\nworkers and client. Also, in a real setup, the scheduler and the workers should\nbe started from the command line and not from within a Python file via\nthe ``subprocess`` module as done here (for the sake of having a self-contained\nexample).\n"
+        "\n# Parallel Usage: Spawning workers from the command line\n\n*Auto-sklearn* uses\n`dask.distributed <https://distributed.dask.org/en/latest/index.html>`_\nfor parallel optimization.\n\nThis example shows how to start the dask scheduler and spawn\nworkers for *Auto-sklearn* manually from the command line. Use this example\nas a starting point to parallelize *Auto-sklearn* across multiple\nmachines.\n\nTo run *Auto-sklearn* in parallel on a single machine check out the example\n`sphx_glr_examples_60_search_example_parallel_n_jobs.py`.\n\nIf you want to start everything manually from within Python\nplease see ``:ref:sphx_glr_examples_60_search_example_parallel_manual_spawning_python.py``.\n\n**NOTE:** Above example is disabled due to issue https://github.com/dask/distributed/issues/5627\n\n\nYou can learn more about the dask command line interface from\nhttps://docs.dask.org/en/latest/setup/cli.html.\n\nWhen manually passing a dask client to Auto-sklearn, all logic\nmust be guarded by ``if __name__ == \"__main__\":`` statements! We use\nmultiple such statements to properly render this example as a notebook\nand also allow execution via the command line.\n\n## Background\n\nTo run Auto-sklearn distributed on multiple machines we need to set\nup three components:\n\n1. **Auto-sklearn and a dask client**. This will manage all workload, find new\n   configurations to evaluate and submit jobs via a dask client. As this\n   runs Bayesian optimization it should be executed on its own CPU.\n2. **The dask workers**. They will do the actual work of running machine\n   learning algorithms and require their own CPU each.\n3. **The scheduler**. It manages the communication between the dask client\n   and the different dask workers. As the client and all workers connect\n   to the scheduler it must be started first. This is a light-weight job\n   and does not require its own CPU.\n\nWe will now start these three components in reverse order: scheduler,\nworkers and client. Also, in a real setup, the scheduler and the workers should\nbe started from the command line and not from within a Python file via\nthe ``subprocess`` module as done here (for the sake of having a self-contained\nexample).\n"
       ]
     },
     {
diff --git a/master/_downloads/c6beb850ad22be83885d2737cca63b33/example_resampling.py b/master/_downloads/c6beb850ad22be83885d2737cca63b33/example_resampling.py
index 39e76cb481..124316a60a 100644
--- a/master/_downloads/c6beb850ad22be83885d2737cca63b33/example_resampling.py
+++ b/master/_downloads/c6beb850ad22be83885d2737cca63b33/example_resampling.py
@@ -98,8 +98,9 @@
 # data by the first feature. In practice, one would use a splitting according
 # to the use case at hand.
 
+selected_indices = (X_train[:, 0] < np.mean(X_train[:, 0])).astype(int)
 resampling_strategy = sklearn.model_selection.PredefinedSplit(
-    test_fold=np.where(X_train[:, 0] < np.mean(X_train[:, 0]))[0]
+    test_fold=selected_indices
 )
 
 automl = autosklearn.classification.AutoSklearnClassifier(
@@ -111,6 +112,8 @@
 )
 automl.fit(X_train, y_train, dataset_name='breast_cancer')
 
+print(automl.sprint_statistics())
+
 ############################################################################
 # For custom resampling strategies (i.e. resampling strategies that are not
 # defined as strings by Auto-sklearn) it is necessary to perform a refit:
diff --git a/master/_downloads/cd4c6485e901a328467de871ee3487d1/example_random_search.ipynb b/master/_downloads/cd4c6485e901a328467de871ee3487d1/example_random_search.ipynb
index 753e1b54fa..1748e1bda2 100644
--- a/master/_downloads/cd4c6485e901a328467de871ee3487d1/example_random_search.ipynb
+++ b/master/_downloads/cd4c6485e901a328467de871ee3487d1/example_random_search.ipynb
@@ -26,7 +26,7 @@
       },
       "outputs": [],
       "source": [
-        "import sklearn.model_selection\nimport sklearn.datasets\nimport sklearn.metrics\n\nfrom smac.facade.roar_facade import ROAR\nfrom smac.scenario.scenario import Scenario\n\nimport autosklearn.classification"
+        "from pprint import pprint\n\nimport sklearn.model_selection\nimport sklearn.datasets\nimport sklearn.metrics\n\nfrom smac.facade.roar_facade import ROAR\nfrom smac.scenario.scenario import Scenario\n\nimport autosklearn.classification"
       ]
     },
     {
@@ -62,7 +62,7 @@
       },
       "outputs": [],
       "source": [
-        "def get_roar_object_callback(\n    scenario_dict,\n    seed,\n    ta,\n    ta_kwargs,\n    metalearning_configurations,\n    n_jobs,\n    dask_client,\n):\n    \"\"\"Random online adaptive racing.\"\"\"\n\n    if n_jobs > 1 or (dask_client and len(dask_client.nthreads()) > 1):\n        raise ValueError(\"Please make sure to guard the code invoking Auto-sklearn by \"\n                         \"`if __name__ == '__main__'` and remove this exception.\")\n\n    scenario = Scenario(scenario_dict)\n    return ROAR(\n        scenario=scenario,\n        rng=seed,\n        tae_runner=ta,\n        tae_runner_kwargs=ta_kwargs,\n        run_id=seed,\n        dask_client=dask_client,\n        n_jobs=n_jobs,\n    )\n\n\nautoml = autosklearn.classification.AutoSklearnClassifier(\n    time_left_for_this_task=60,\n    per_run_time_limit=15,\n    tmp_folder='/tmp/autosklearn_random_search_example_tmp',\n    initial_configurations_via_metalearning=0,\n    # The callback to get the SMAC object\n    get_smac_object_callback=get_roar_object_callback,\n)\nautoml.fit(X_train, y_train, dataset_name='breast_cancer')\n\nprint('#' * 80)\nprint('Results for ROAR.')\n# Print the final ensemble constructed by auto-sklearn via ROAR.\nprint(automl.show_models())\npredictions = automl.predict(X_test)\n# Print statistics about the auto-sklearn run such as number of\n# iterations, number of models failed with a time out.\nprint(automl.sprint_statistics())\nprint(\"Accuracy score\", sklearn.metrics.accuracy_score(y_test, predictions))"
+        "def get_roar_object_callback(\n    scenario_dict,\n    seed,\n    ta,\n    ta_kwargs,\n    metalearning_configurations,\n    n_jobs,\n    dask_client,\n):\n    \"\"\"Random online adaptive racing.\"\"\"\n\n    if n_jobs > 1 or (dask_client and len(dask_client.nthreads()) > 1):\n        raise ValueError(\"Please make sure to guard the code invoking Auto-sklearn by \"\n                         \"`if __name__ == '__main__'` and remove this exception.\")\n\n    scenario = Scenario(scenario_dict)\n    return ROAR(\n        scenario=scenario,\n        rng=seed,\n        tae_runner=ta,\n        tae_runner_kwargs=ta_kwargs,\n        run_id=seed,\n        dask_client=dask_client,\n        n_jobs=n_jobs,\n    )\n\n\nautoml = autosklearn.classification.AutoSklearnClassifier(\n    time_left_for_this_task=60,\n    per_run_time_limit=15,\n    tmp_folder='/tmp/autosklearn_random_search_example_tmp',\n    initial_configurations_via_metalearning=0,\n    # The callback to get the SMAC object\n    get_smac_object_callback=get_roar_object_callback,\n)\nautoml.fit(X_train, y_train, dataset_name='breast_cancer')\n\nprint('#' * 80)\nprint('Results for ROAR.')\n# Print the final ensemble constructed by auto-sklearn via ROAR.\npprint(automl.show_models(), indent=4)\npredictions = automl.predict(X_test)\n# Print statistics about the auto-sklearn run such as number of\n# iterations, number of models failed with a time out.\nprint(automl.sprint_statistics())\nprint(\"Accuracy score\", sklearn.metrics.accuracy_score(y_test, predictions))"
       ]
     },
     {
@@ -80,7 +80,7 @@
       },
       "outputs": [],
       "source": [
-        "def get_random_search_object_callback(\n    scenario_dict,\n    seed,\n    ta,\n    ta_kwargs,\n    metalearning_configurations,\n    n_jobs,\n    dask_client\n):\n    \"\"\" Random search \"\"\"\n\n    if n_jobs > 1 or (dask_client and len(dask_client.nthreads()) > 1):\n        raise ValueError(\"Please make sure to guard the code invoking Auto-sklearn by \"\n                         \"`if __name__ == '__main__'` and remove this exception.\")\n\n    scenario_dict['minR'] = len(scenario_dict['instances'])\n    scenario_dict['initial_incumbent'] = 'RANDOM'\n    scenario = Scenario(scenario_dict)\n    return ROAR(\n        scenario=scenario,\n        rng=seed,\n        tae_runner=ta,\n        tae_runner_kwargs=ta_kwargs,\n        run_id=seed,\n        dask_client=dask_client,\n        n_jobs=n_jobs,\n    )\n\n\nautoml = autosklearn.classification.AutoSklearnClassifier(\n    time_left_for_this_task=60,\n    per_run_time_limit=15,\n    tmp_folder='/tmp/autosklearn_random_search_example_tmp',\n    initial_configurations_via_metalearning=0,\n    # Passing the callback to get the SMAC object\n    get_smac_object_callback=get_random_search_object_callback,\n)\nautoml.fit(X_train, y_train, dataset_name='breast_cancer')\n\nprint('#' * 80)\nprint('Results for random search.')\n\n# Print the final ensemble constructed by auto-sklearn via random search.\nprint(automl.show_models())\n\n# Print statistics about the auto-sklearn run such as number of\n# iterations, number of models failed with a time out.\nprint(automl.sprint_statistics())\n\npredictions = automl.predict(X_test)\nprint(\"Accuracy score\", sklearn.metrics.accuracy_score(y_test, predictions))"
+        "def get_random_search_object_callback(\n    scenario_dict,\n    seed,\n    ta,\n    ta_kwargs,\n    metalearning_configurations,\n    n_jobs,\n    dask_client\n):\n    \"\"\" Random search \"\"\"\n\n    if n_jobs > 1 or (dask_client and len(dask_client.nthreads()) > 1):\n        raise ValueError(\"Please make sure to guard the code invoking Auto-sklearn by \"\n                         \"`if __name__ == '__main__'` and remove this exception.\")\n\n    scenario_dict['minR'] = len(scenario_dict['instances'])\n    scenario_dict['initial_incumbent'] = 'RANDOM'\n    scenario = Scenario(scenario_dict)\n    return ROAR(\n        scenario=scenario,\n        rng=seed,\n        tae_runner=ta,\n        tae_runner_kwargs=ta_kwargs,\n        run_id=seed,\n        dask_client=dask_client,\n        n_jobs=n_jobs,\n    )\n\n\nautoml = autosklearn.classification.AutoSklearnClassifier(\n    time_left_for_this_task=60,\n    per_run_time_limit=15,\n    tmp_folder='/tmp/autosklearn_random_search_example_tmp',\n    initial_configurations_via_metalearning=0,\n    # Passing the callback to get the SMAC object\n    get_smac_object_callback=get_random_search_object_callback,\n)\nautoml.fit(X_train, y_train, dataset_name='breast_cancer')\n\nprint('#' * 80)\nprint('Results for random search.')\n\n# Print the final ensemble constructed by auto-sklearn via random search.\npprint(automl.show_models(), indent=4)\n\n# Print statistics about the auto-sklearn run such as number of\n# iterations, number of models failed with a time out.\nprint(automl.sprint_statistics())\n\npredictions = automl.predict(X_test)\nprint(\"Accuracy score\", sklearn.metrics.accuracy_score(y_test, predictions))"
       ]
     }
   ],
diff --git a/master/_downloads/ce2e8ffc7b9780e0d18638781ffd952b/example_successive_halving.ipynb b/master/_downloads/ce2e8ffc7b9780e0d18638781ffd952b/example_successive_halving.ipynb
index f285520243..e62b893576 100644
--- a/master/_downloads/ce2e8ffc7b9780e0d18638781ffd952b/example_successive_halving.ipynb
+++ b/master/_downloads/ce2e8ffc7b9780e0d18638781ffd952b/example_successive_halving.ipynb
@@ -26,7 +26,7 @@
       },
       "outputs": [],
       "source": [
-        "import sklearn.model_selection\nimport sklearn.datasets\nimport sklearn.metrics\n\nimport autosklearn.classification"
+        "from pprint import pprint\n\nimport sklearn.model_selection\nimport sklearn.datasets\nimport sklearn.metrics\n\nimport autosklearn.classification"
       ]
     },
     {
@@ -80,7 +80,7 @@
       },
       "outputs": [],
       "source": [
-        "automl = autosklearn.classification.AutoSklearnClassifier(\n    time_left_for_this_task=40,\n    per_run_time_limit=10,\n    tmp_folder='/tmp/autosklearn_sh_example_tmp',\n    disable_evaluator_output=False,\n    # 'holdout' with 'train_size'=0.67 is the default argument setting\n    # for AutoSklearnClassifier. It is explicitly specified in this example\n    # for demonstrational purpose.\n    resampling_strategy='holdout',\n    resampling_strategy_arguments={'train_size': 0.67},\n    include={\n        'classifier': [\n            'extra_trees', 'gradient_boosting', 'random_forest',\n            'sgd', 'passive_aggressive'\n        ],\n        'feature_preprocessor': ['no_preprocessing']\n    },\n    get_smac_object_callback=get_smac_object_callback('iterations'),\n)\nautoml.fit(X_train, y_train, dataset_name='breast_cancer')\n\nprint(automl.show_models())\npredictions = automl.predict(X_test)\n# Print statistics about the auto-sklearn run such as number of\n# iterations, number of models failed with a time out.\nprint(automl.sprint_statistics())\nprint(\"Accuracy score\", sklearn.metrics.accuracy_score(y_test, predictions))"
+        "automl = autosklearn.classification.AutoSklearnClassifier(\n    time_left_for_this_task=40,\n    per_run_time_limit=10,\n    tmp_folder='/tmp/autosklearn_sh_example_tmp',\n    disable_evaluator_output=False,\n    # 'holdout' with 'train_size'=0.67 is the default argument setting\n    # for AutoSklearnClassifier. It is explicitly specified in this example\n    # for demonstrational purpose.\n    resampling_strategy='holdout',\n    resampling_strategy_arguments={'train_size': 0.67},\n    include={\n        'classifier': [\n            'extra_trees', 'gradient_boosting', 'random_forest',\n            'sgd', 'passive_aggressive'\n        ],\n        'feature_preprocessor': ['no_preprocessing']\n    },\n    get_smac_object_callback=get_smac_object_callback('iterations'),\n)\nautoml.fit(X_train, y_train, dataset_name='breast_cancer')\n\npprint(automl.show_models(), indent=4)\npredictions = automl.predict(X_test)\n# Print statistics about the auto-sklearn run such as number of\n# iterations, number of models failed with a time out.\nprint(automl.sprint_statistics())\nprint(\"Accuracy score\", sklearn.metrics.accuracy_score(y_test, predictions))"
       ]
     },
     {
@@ -98,7 +98,7 @@
       },
       "outputs": [],
       "source": [
-        "X, y = sklearn.datasets.load_breast_cancer(return_X_y=True)\nX_train, X_test, y_train, y_test = \\\n    sklearn.model_selection.train_test_split(X, y, random_state=1, shuffle=True)\n\nautoml = autosklearn.classification.AutoSklearnClassifier(\n    time_left_for_this_task=40,\n    per_run_time_limit=10,\n    tmp_folder='/tmp/autosklearn_sh_example_tmp_01',\n    disable_evaluator_output=False,\n    resampling_strategy='cv',\n    include={\n        'classifier': [\n            'extra_trees', 'gradient_boosting', 'random_forest', \n            'sgd', 'passive_aggressive'\n        ],\n        'feature_preprocessor': ['no_preprocessing']\n    },\n    get_smac_object_callback=get_smac_object_callback('iterations'),\n)\nautoml.fit(X_train, y_train, dataset_name='breast_cancer')\n\n# Print the final ensemble constructed by auto-sklearn.\nprint(automl.show_models())\nautoml.refit(X_train, y_train)\npredictions = automl.predict(X_test)\n# Print statistics about the auto-sklearn run such as number of\n# iterations, number of models failed with a time out.\nprint(automl.sprint_statistics())\nprint(\"Accuracy score\", sklearn.metrics.accuracy_score(y_test, predictions))"
+        "X, y = sklearn.datasets.load_breast_cancer(return_X_y=True)\nX_train, X_test, y_train, y_test = \\\n    sklearn.model_selection.train_test_split(X, y, random_state=1, shuffle=True)\n\nautoml = autosklearn.classification.AutoSklearnClassifier(\n    time_left_for_this_task=40,\n    per_run_time_limit=10,\n    tmp_folder='/tmp/autosklearn_sh_example_tmp_01',\n    disable_evaluator_output=False,\n    resampling_strategy='cv',\n    include={\n        'classifier': [\n            'extra_trees', 'gradient_boosting', 'random_forest', \n            'sgd', 'passive_aggressive'\n        ],\n        'feature_preprocessor': ['no_preprocessing']\n    },\n    get_smac_object_callback=get_smac_object_callback('iterations'),\n)\nautoml.fit(X_train, y_train, dataset_name='breast_cancer')\n\n# Print the final ensemble constructed by auto-sklearn.\npprint(automl.show_models(), indent=4)\nautoml.refit(X_train, y_train)\npredictions = automl.predict(X_test)\n# Print statistics about the auto-sklearn run such as number of\n# iterations, number of models failed with a time out.\nprint(automl.sprint_statistics())\nprint(\"Accuracy score\", sklearn.metrics.accuracy_score(y_test, predictions))"
       ]
     },
     {
@@ -116,7 +116,7 @@
       },
       "outputs": [],
       "source": [
-        "X, y = sklearn.datasets.load_breast_cancer(return_X_y=True)\nX_train, X_test, y_train, y_test = \\\n    sklearn.model_selection.train_test_split(X, y, random_state=1, shuffle=True)\n\nautoml = autosklearn.classification.AutoSklearnClassifier(\n    time_left_for_this_task=40,\n    per_run_time_limit=10,\n    tmp_folder='/tmp/autosklearn_sh_example_tmp_cv_02',\n    disable_evaluator_output=False,\n    resampling_strategy='cv-iterative-fit',\n    include={\n        'classifier': [\n            'extra_trees', 'gradient_boosting', 'random_forest',\n            'sgd', 'passive_aggressive'\n        ],\n        'feature_preprocessor': ['no_preprocessing']\n    },\n    get_smac_object_callback=get_smac_object_callback('iterations'),\n)\nautoml.fit(X_train, y_train, dataset_name='breast_cancer')\n\n# Print the final ensemble constructed by auto-sklearn.\nprint(automl.show_models())\nautoml.refit(X_train, y_train)\npredictions = automl.predict(X_test)\n# Print statistics about the auto-sklearn run such as number of\n# iterations, number of models failed with a time out.\nprint(automl.sprint_statistics())\nprint(\"Accuracy score\", sklearn.metrics.accuracy_score(y_test, predictions))"
+        "X, y = sklearn.datasets.load_breast_cancer(return_X_y=True)\nX_train, X_test, y_train, y_test = \\\n    sklearn.model_selection.train_test_split(X, y, random_state=1, shuffle=True)\n\nautoml = autosklearn.classification.AutoSklearnClassifier(\n    time_left_for_this_task=40,\n    per_run_time_limit=10,\n    tmp_folder='/tmp/autosklearn_sh_example_tmp_cv_02',\n    disable_evaluator_output=False,\n    resampling_strategy='cv-iterative-fit',\n    include={\n        'classifier': [\n            'extra_trees', 'gradient_boosting', 'random_forest',\n            'sgd', 'passive_aggressive'\n        ],\n        'feature_preprocessor': ['no_preprocessing']\n    },\n    get_smac_object_callback=get_smac_object_callback('iterations'),\n)\nautoml.fit(X_train, y_train, dataset_name='breast_cancer')\n\n# Print the final ensemble constructed by auto-sklearn.\npprint(automl.show_models(), indent=4)\nautoml.refit(X_train, y_train)\npredictions = automl.predict(X_test)\n# Print statistics about the auto-sklearn run such as number of\n# iterations, number of models failed with a time out.\nprint(automl.sprint_statistics())\nprint(\"Accuracy score\", sklearn.metrics.accuracy_score(y_test, predictions))"
       ]
     },
     {
@@ -134,7 +134,7 @@
       },
       "outputs": [],
       "source": [
-        "X, y = sklearn.datasets.load_breast_cancer(return_X_y=True)\nX_train, X_test, y_train, y_test = \\\n    sklearn.model_selection.train_test_split(X, y, random_state=1, shuffle=True)\n\nautoml = autosklearn.classification.AutoSklearnClassifier(\n    time_left_for_this_task=40,\n    per_run_time_limit=10,\n    tmp_folder='/tmp/autosklearn_sh_example_tmp_03',\n    disable_evaluator_output=False,\n    # 'holdout' with 'train_size'=0.67 is the default argument setting\n    # for AutoSklearnClassifier. It is explicitly specified in this example\n    # for demonstrational purpose.\n    resampling_strategy='holdout',\n    resampling_strategy_arguments={'train_size': 0.67},\n    get_smac_object_callback=get_smac_object_callback('subsample'),\n)\nautoml.fit(X_train, y_train, dataset_name='breast_cancer')\n\n# Print the final ensemble constructed by auto-sklearn.\nprint(automl.show_models())\npredictions = automl.predict(X_test)\n# Print statistics about the auto-sklearn run such as number of\n# iterations, number of models failed with a time out.\nprint(automl.sprint_statistics())\nprint(\"Accuracy score\", sklearn.metrics.accuracy_score(y_test, predictions))"
+        "X, y = sklearn.datasets.load_breast_cancer(return_X_y=True)\nX_train, X_test, y_train, y_test = \\\n    sklearn.model_selection.train_test_split(X, y, random_state=1, shuffle=True)\n\nautoml = autosklearn.classification.AutoSklearnClassifier(\n    time_left_for_this_task=40,\n    per_run_time_limit=10,\n    tmp_folder='/tmp/autosklearn_sh_example_tmp_03',\n    disable_evaluator_output=False,\n    # 'holdout' with 'train_size'=0.67 is the default argument setting\n    # for AutoSklearnClassifier. It is explicitly specified in this example\n    # for demonstrational purpose.\n    resampling_strategy='holdout',\n    resampling_strategy_arguments={'train_size': 0.67},\n    get_smac_object_callback=get_smac_object_callback('subsample'),\n)\nautoml.fit(X_train, y_train, dataset_name='breast_cancer')\n\n# Print the final ensemble constructed by auto-sklearn.\npprint(automl.show_models(), indent=4)\npredictions = automl.predict(X_test)\n# Print statistics about the auto-sklearn run such as number of\n# iterations, number of models failed with a time out.\nprint(automl.sprint_statistics())\nprint(\"Accuracy score\", sklearn.metrics.accuracy_score(y_test, predictions))"
       ]
     },
     {
@@ -152,7 +152,7 @@
       },
       "outputs": [],
       "source": [
-        "X, y = sklearn.datasets.load_breast_cancer(return_X_y=True)\nX_train, X_test, y_train, y_test = \\\n    sklearn.model_selection.train_test_split(X, y, random_state=1, shuffle=True)\n\nautoml = autosklearn.classification.AutoSklearnClassifier(\n    time_left_for_this_task=40,\n    per_run_time_limit=10,\n    tmp_folder='/tmp/autosklearn_sh_example_tmp_04',\n    disable_evaluator_output=False,\n    # 'holdout' with 'train_size'=0.67 is the default argument setting\n    # for AutoSklearnClassifier. It is explicitly specified in this example\n    # for demonstrational purpose.\n    resampling_strategy='holdout',\n    resampling_strategy_arguments={'train_size': 0.67},\n    include={\n        'classifier': [\n            'extra_trees', 'gradient_boosting', 'random_forest', 'sgd'\n        ]\n    },\n    get_smac_object_callback=get_smac_object_callback('mixed'),\n)\nautoml.fit(X_train, y_train, dataset_name='breast_cancer')\n\n# Print the final ensemble constructed by auto-sklearn.\nprint(automl.show_models())\npredictions = automl.predict(X_test)\n# Print statistics about the auto-sklearn run such as number of\n# iterations, number of models failed with a time out.\nprint(automl.sprint_statistics())\nprint(\"Accuracy score\", sklearn.metrics.accuracy_score(y_test, predictions))"
+        "X, y = sklearn.datasets.load_breast_cancer(return_X_y=True)\nX_train, X_test, y_train, y_test = \\\n    sklearn.model_selection.train_test_split(X, y, random_state=1, shuffle=True)\n\nautoml = autosklearn.classification.AutoSklearnClassifier(\n    time_left_for_this_task=40,\n    per_run_time_limit=10,\n    tmp_folder='/tmp/autosklearn_sh_example_tmp_04',\n    disable_evaluator_output=False,\n    # 'holdout' with 'train_size'=0.67 is the default argument setting\n    # for AutoSklearnClassifier. It is explicitly specified in this example\n    # for demonstrational purpose.\n    resampling_strategy='holdout',\n    resampling_strategy_arguments={'train_size': 0.67},\n    include={\n        'classifier': [\n            'extra_trees', 'gradient_boosting', 'random_forest', 'sgd'\n        ]\n    },\n    get_smac_object_callback=get_smac_object_callback('mixed'),\n)\nautoml.fit(X_train, y_train, dataset_name='breast_cancer')\n\n# Print the final ensemble constructed by auto-sklearn.\npprint(automl.show_models(), indent=4)\npredictions = automl.predict(X_test)\n# Print statistics about the auto-sklearn run such as number of\n# iterations, number of models failed with a time out.\nprint(automl.sprint_statistics())\nprint(\"Accuracy score\", sklearn.metrics.accuracy_score(y_test, predictions))"
       ]
     }
   ],
diff --git a/master/_downloads/d6ccbec3202b9c83accb060d07132b4e/example_random_search.py b/master/_downloads/d6ccbec3202b9c83accb060d07132b4e/example_random_search.py
index 292f005da9..2c9cc76695 100644
--- a/master/_downloads/d6ccbec3202b9c83accb060d07132b4e/example_random_search.py
+++ b/master/_downloads/d6ccbec3202b9c83accb060d07132b4e/example_random_search.py
@@ -12,6 +12,7 @@
 as yet another alternative optimizatino strategy.
 Both examples are intended to show how the optimization strategy in *auto-sklearn* can be adapted.
 """  # noqa (links are too long)
+from pprint import pprint
 
 import sklearn.model_selection
 import sklearn.datasets
@@ -75,7 +76,7 @@ def get_roar_object_callback(
 print('#' * 80)
 print('Results for ROAR.')
 # Print the final ensemble constructed by auto-sklearn via ROAR.
-print(automl.show_models())
+pprint(automl.show_models(), indent=4)
 predictions = automl.predict(X_test)
 # Print statistics about the auto-sklearn run such as number of
 # iterations, number of models failed with a time out.
@@ -129,7 +130,7 @@ def get_random_search_object_callback(
 print('Results for random search.')
 
 # Print the final ensemble constructed by auto-sklearn via random search.
-print(automl.show_models())
+pprint(automl.show_models(), indent=4)
 
 # Print statistics about the auto-sklearn run such as number of
 # iterations, number of models failed with a time out.
diff --git a/master/_downloads/d9369ef2377e1dc469a08a83d0887d1b/example_multioutput_regression.ipynb b/master/_downloads/d9369ef2377e1dc469a08a83d0887d1b/example_multioutput_regression.ipynb
index f3112d78a6..7998b22d6f 100644
--- a/master/_downloads/d9369ef2377e1dc469a08a83d0887d1b/example_multioutput_regression.ipynb
+++ b/master/_downloads/d9369ef2377e1dc469a08a83d0887d1b/example_multioutput_regression.ipynb
@@ -26,7 +26,7 @@
       },
       "outputs": [],
       "source": [
-        "import numpy as numpy\n\nfrom sklearn.datasets import make_regression\nfrom sklearn.metrics import r2_score\nfrom sklearn.model_selection import train_test_split\n\nfrom autosklearn.regression import AutoSklearnRegressor"
+        "import numpy as numpy\nfrom pprint import pprint\n\nfrom sklearn.datasets import make_regression\nfrom sklearn.metrics import r2_score\nfrom sklearn.model_selection import train_test_split\n\nfrom autosklearn.regression import AutoSklearnRegressor"
       ]
     },
     {
@@ -98,7 +98,7 @@
       },
       "outputs": [],
       "source": [
-        "print(automl.show_models())"
+        "pprint(automl.show_models(), indent=4)"
       ]
     },
     {
diff --git a/master/_downloads/e2278309f0c9a94fb08745a1f915766e/example_get_pipeline_components.py b/master/_downloads/e2278309f0c9a94fb08745a1f915766e/example_get_pipeline_components.py
index 76132291fc..f7a97ead27 100644
--- a/master/_downloads/e2278309f0c9a94fb08745a1f915766e/example_get_pipeline_components.py
+++ b/master/_downloads/e2278309f0c9a94fb08745a1f915766e/example_get_pipeline_components.py
@@ -14,6 +14,8 @@
 the sklearn models. This example illustrates how to interact
 with the sklearn components directly, in this case a PCA preprocessor.
 """
+from pprint import pprint
+
 import sklearn.datasets
 import sklearn.metrics
 
@@ -62,10 +64,17 @@
 # `Ensemble Selection <https://www.cs.cornell.edu/~alexn/papers/shotgun.icml04.revised.rev2.pdf>`_
 # to construct ensembles in a post-hoc fashion. The ensemble is a linear
 # weighting of all models constructed during the hyperparameter optimization.
-# This prints the final ensemble. It is a list of tuples, each tuple being
-# the model weight in the ensemble and the model itself.
-
-print(automl.show_models())
+# This prints the final ensemble. It is a dictionary where ``model_id`` of
+# each model is a key, and value is a dictionary containing information
+# of that model. A model's dict contains its ``'model_id'``, ``'rank'``,
+# ``'cost'``, ``'ensemble_weight'``, and the model itself. The model is
+# given by the ``'data_preprocessor'``, ``'feature_preprocessor'``,
+# ``'regressor'/'classifier'`` and ``'sklearn_regressor'/'sklearn_classifier'``
+# entries. But for the ``'cv'`` resampling strategy, the same for each cv
+# model is stored in the ``'estimators'`` list in the dict, along with the
+# ``'voting_model'``.
+
+pprint(automl.show_models(), indent=4)
 
 ###########################################################################
 # Report statistics about the search
diff --git a/master/_downloads/eec25698bb5fb78402eba3ef5551d2a0/example_extending_preprocessor.ipynb b/master/_downloads/eec25698bb5fb78402eba3ef5551d2a0/example_extending_preprocessor.ipynb
index a44db2d221..fa090d2f9a 100644
--- a/master/_downloads/eec25698bb5fb78402eba3ef5551d2a0/example_extending_preprocessor.ipynb
+++ b/master/_downloads/eec25698bb5fb78402eba3ef5551d2a0/example_extending_preprocessor.ipynb
@@ -26,7 +26,7 @@
       },
       "outputs": [],
       "source": [
-        "from ConfigSpace.configuration_space import ConfigurationSpace\nfrom ConfigSpace.hyperparameters import UniformFloatHyperparameter, CategoricalHyperparameter\nfrom ConfigSpace.conditions import InCondition\n\nimport sklearn.metrics\nimport autosklearn.classification\nimport autosklearn.pipeline.components.feature_preprocessing\nfrom autosklearn.pipeline.components.base \\\n    import AutoSklearnPreprocessingAlgorithm\nfrom autosklearn.pipeline.constants import DENSE, SIGNED_DATA, \\\n    UNSIGNED_DATA\nfrom autosklearn.util.common import check_none\n\nfrom sklearn.datasets import load_breast_cancer\nfrom sklearn.model_selection import train_test_split"
+        "from pprint import pprint\n\nfrom ConfigSpace.configuration_space import ConfigurationSpace\nfrom ConfigSpace.hyperparameters import UniformFloatHyperparameter, CategoricalHyperparameter\nfrom ConfigSpace.conditions import InCondition\n\nimport sklearn.metrics\nimport autosklearn.classification\nimport autosklearn.pipeline.components.feature_preprocessing\nfrom autosklearn.pipeline.components.base \\\n    import AutoSklearnPreprocessingAlgorithm\nfrom autosklearn.pipeline.constants import DENSE, SIGNED_DATA, \\\n    UNSIGNED_DATA\nfrom autosklearn.util.common import check_none\n\nfrom sklearn.datasets import load_breast_cancer\nfrom sklearn.model_selection import train_test_split"
       ]
     },
     {
@@ -116,7 +116,7 @@
       },
       "outputs": [],
       "source": [
-        "y_pred = clf.predict(X_test)\nprint(\"accuracy: \", sklearn.metrics.accuracy_score(y_pred, y_test))\nprint(clf.show_models())"
+        "y_pred = clf.predict(X_test)\nprint(\"accuracy: \", sklearn.metrics.accuracy_score(y_pred, y_test))\npprint(clf.show_models(), indent=4)"
       ]
     }
   ],
diff --git a/master/_downloads/f9a47ce26ef58a0617f152bb7396464b/example_extending_regression.ipynb b/master/_downloads/f9a47ce26ef58a0617f152bb7396464b/example_extending_regression.ipynb
index 5783cf90e0..7e7efbf899 100644
--- a/master/_downloads/f9a47ce26ef58a0617f152bb7396464b/example_extending_regression.ipynb
+++ b/master/_downloads/f9a47ce26ef58a0617f152bb7396464b/example_extending_regression.ipynb
@@ -26,7 +26,7 @@
       },
       "outputs": [],
       "source": [
-        "from ConfigSpace.configuration_space import ConfigurationSpace\nfrom ConfigSpace.hyperparameters import UniformFloatHyperparameter, \\\n    UniformIntegerHyperparameter, CategoricalHyperparameter\nfrom ConfigSpace.conditions import EqualsCondition\n\nimport sklearn.metrics\nimport autosklearn.regression\nimport autosklearn.pipeline.components.regression\nfrom autosklearn.pipeline.components.base import AutoSklearnRegressionAlgorithm\nfrom autosklearn.pipeline.constants import SPARSE, DENSE, \\\n    SIGNED_DATA, UNSIGNED_DATA, PREDICTIONS\n\nfrom sklearn.datasets import load_diabetes\nfrom sklearn.model_selection import train_test_split"
+        "from pprint import pprint\n\nfrom ConfigSpace.configuration_space import ConfigurationSpace\nfrom ConfigSpace.hyperparameters import UniformFloatHyperparameter, \\\n    UniformIntegerHyperparameter, CategoricalHyperparameter\nfrom ConfigSpace.conditions import EqualsCondition\n\nimport sklearn.metrics\nimport autosklearn.regression\nimport autosklearn.pipeline.components.regression\nfrom autosklearn.pipeline.components.base import AutoSklearnRegressionAlgorithm\nfrom autosklearn.pipeline.constants import SPARSE, DENSE, \\\n    SIGNED_DATA, UNSIGNED_DATA, PREDICTIONS\n\nfrom sklearn.datasets import load_diabetes\nfrom sklearn.model_selection import train_test_split"
       ]
     },
     {
@@ -98,7 +98,7 @@
       },
       "outputs": [],
       "source": [
-        "y_pred = reg.predict(X_test)\nprint(\"r2 score: \", sklearn.metrics.r2_score(y_pred, y_test))\nprint(reg.show_models())"
+        "y_pred = reg.predict(X_test)\nprint(\"r2 score: \", sklearn.metrics.r2_score(y_pred, y_test))\npprint(reg.show_models(), indent=4)"
       ]
     }
   ],
diff --git a/master/_downloads/fb625db3c50d423b1b7881136ffdeec8/examples_jupyter.zip b/master/_downloads/fb625db3c50d423b1b7881136ffdeec8/examples_jupyter.zip
index 1b5ea38bd831e0811fa28a51d3af12418d0e2fa9..0eca0ae90b1780bf9d4f4eacc8e54b024a9d942f 100644
GIT binary patch
delta 3367
zcmai0eQ+B`6~A3ej?Gw>9VwQ-+OF)@mStN?itUaCY4c$yv{O3+{)2#sb&^ifjpaM_
z?rg`&w8klA+MzRvd%(g#Tf>+Rbs%MOkV`^7+CbB^lqQ{KIsuX<B{MWo-1LJOD1`F%
zB*#|d7<tySxBK4jz5VUJkKdgt{oYh*-KKksY;KPIJ~h$Zf8pVTZPdweY_#j|6SscU
zVK;Vm=(r}K$U`KPQA9~2;!s9bwA>e6<HBBJ{T*LkLEq9)?I+rZD8&UyTianwezX}q
zeux<D>rNL{J_PvsnQuK?=i)dvRx^C#sKZfoV{ComSf}xE#nVN`PuyFJugBNd9Jt?@
zsq3;=Uj15kf78Zqlny&Mj*Z*s<N)>=Z#p`VF^ZZIJ%2a0yO;9bvTzd56@XjL?>!X5
zj_u(1&+d_)lE7)nm?his@^Jb|yvo?q__qQ`ee$DS#>Uo$QrnKNclR6nGA|I@@g3d$
z=08`WBOY7DL$G-Czn$*#S(vS;Iq^<3z;V%l;p}nRmN#kieNHGDkBc!)lqFsnBcmxn
z5lAF5B*%sHfEbTNh@1#Z0v}5eYmJC1;mM#d)<$?KPDc3ju#jC_gc6e_jTa?RO2VBc
zWMznF_+d#-5G}=PS)>42pPh_1tAr_&oSI5RO$8sC=bdPdMB3Q5!b^i}9+u!ECac=~
zh9IfJP&6$Jj0$2hm0f1uX(?geYfTl>i2&(eSW&}LQXCN^(hwc9Bmp$eYy3b)5i*J#
z6I4}Jau)Qdbk+F0WkpCT+09^cEX}K`m=FafX)yvQ>R?*n6=`6x;QU&i0uU={iYTZ7
zvSC;wz|_1Wu@Qy`X855@n(?S=;|i|{N#LEG3f2Wn-nyCCNLaddJE|rtLL7*n#}ZUc
zgd30*m0_ku6#_B4$ykSbOv_+kRMb*T!y<P?*1+h1g$YW5FkGJg?G*<LA_=V+G<v@f
zzuEQzJa+S-A4$&23-AhmG;`&R*i9VA#;y8&yFK)96jvB;XCScVUaxSw;5vkuCd3CI
z=tba~tR;H*lCM0#tJ+vb@Ikt00h`5_dRJaA8kQ*8ioQzwZbc3BGDTj!<u5b(h4VJr
z`wLV|zq=jP(VxGEl1AvSGj^LD0>!v=`So(!t8A7$(&#RK4xIh<XPxCm5Sc|T<DT6w
zXG7DqM5OC$Dx0X|^SIQwbI+!->j}6P?Y2i4`uODG8dufB-Tg-ldiFpQ$Fb4Dag>ka
zPBRli&tnYkzxE<}jIMhT1!!a~sw<9!6d@6!-V>;u1{2sbo*Bz=?)Ufh^?xSh_mlNe
zc|;(>R`wpOfQySN4^KW$;t+tE3^Ay}B&BH?H53dc;Wme(ftWlLjPvSX5TxOuMTdd$
zpe3mWJG*YbgWYRb+AI%~7%vg))k46S19zfmgY9`pHYVPWK3(Q8;7B_k^o8ew95-ja
z8bOzrPb@C;6#-Q_uV>4AZyS05nOjHDUABU~Q`^w{_N7*zK<4NUw53oZi#GTOsxH)n
zy#{3TJu{0czX;Ax-+%haXf<R-7`IX9ezb}{C}5ZMAkBYGp)EzrCQ28D^YoZA(<oe0
z;9}N|AM8P0+05-&sATYjsdoZBX+HK#q~J;np6m&gg<dM>I5rxxo&<zkXqc-WN2iuA
zI+iD=jEb={h#t+La&zW8=$!&;auxJu77W`g7HltpZF64_-djsa2$eYAfbZY>cWpUW
z$8qP(13vs3u3qB#YwUqAZc^rKE8bg0`45n1QLQ%7K6DjTo&>Z%A6xgw4YeG{Mvxx%
z;RgDZB(5+Y{SE3s^xZ3{-aO;PccEp?$n@$z;BrdZaFcnc5}$K>t=7Q@xip}r<k4(S
zQhhB8B<EcC6}Yfp==)HHpk|}Xa=~=Gg}(p)qr1iY;3CSD-{5q|0s85=(^JpXa2y+J
zvZwn`0<Sa={vG{(dFEsjj@0R13i}FbLD*e%gGS0~ud@H9C%E~xO5fSxp@T_WMhZHk
zJcVOU{ilxG=_^dKpwG#ZY;ByZZ_Wzl^(;W5H&5{5YPvCkgH{_tSM#_Q@9NQecUS3u
z$T+CXgCW+I9PeuE9rJJ;8?AZzn*}CK>s`o0&&P0?)w<A741nq$u(r}M+(h!*7@#{3
zq<_*f+0t_pC1BF3JoGZdus>&Eyy-`-woUgg5L-}?^6*AaT3S$_>aAiy&qV>Ne%4xb
z#&9!Xg~R^DA>W^X5H_0g^gsN!<<B>dxajsMF7xIW7mkBo?AmX7qP7C*eY=Adi~VDz
zgN_58MPeh{_2@sI^w5_XpG(w|MPUB5G!n-ftV~URC;&ib9MQ`C$k)CYKQDVj2KQfs
z;JE3D;f$r?qHjk~HGMY@M<~dcc?!l<P60CLE#88hmM8nA136j9arIUXqGyxfCw~Ic
zmk(D_XA1jU7p^KVD?w)ZWFFcr7cH$0>#ceVa!#J?eIHx0gyo=mZ$Wa%lRaC5nKy(&
lvhdJn6F@;h$mc0M)t*y;6ySE-i`iy1{H05v*58fc>wmg!R`dV>

delta 4601
zcmaJ^4R8}x9^bc3g#zgZNxy-<656J<4J|FAKuT$Qw6xIHD|dL3Hr*!MW^J+?vl~K{
zTI-0Sa4PQL^MkK;@W#^{DFoed&Qlz@8D~TRz41mpJ&#kl({ubDQ5n7e+fCYpg2}wh
z=Dq*__y7LC{XcRj?X@#$3%c4<Ojd?|p4lgKzxp!!)kZVJ&{e{a?X7HCG&8dfMJJY&
zA#$h`<wt)mNuQ-2=}cAe)3eC?4<WN^J?~SyR@onX`#Y_^7%gYhSMC?Ow>RyozA&F*
zF3gX<ld&X4U3D^x94SR})SpZ8q8luKNJ$Zcj_6l8+pd}cMj9&EwCK+ZuSqp81ElfE
zqZig$7-n7chw58dT?qiGQXMPiM9<Zpo5>&TTGf`8^);9?_N{aK^%)F9SC#r+Mm~AB
zh0Q(MzMYMpV~4L~#($$1=Kd`uYxN<aJ6M~4XMZKb&{d=26S?HtHg;ija0FcP_XiIf
z+JH-Tv)XMhA{&bKJ@UG#(J#mfmIkmA4*Gey67!tLXZ*qaydVy`M8(JBfaK-<IKm5q
zJ_QRRrhB|7^MQUpucCuF>KzR)%<zr+`o1~QW8*!GO#cD(q6eD)icQT2gl;WjLx0K7
z+6Mkwd{<-5yWn}c9J<b5w6g`e&f23{s2{15>@0QU3mZB0II^qHx86iLb5Mr%p#>e8
zJ5yj{WKDZOR1{sbS<Mw7{@tUeFXk;|n7rtHr=@Il-A5g>A*c}_AL7-_kL)R1s(VPs
z)o4Qf=UYu`$3Lv<{*SxWwu-^%_VZ7snqCH7qZOaMnQnTIqLz;=x0-g)r$PROT+;|W
zwVV($%)7^h?&H>pr&~%HrbS!pL>F0v$*T=$J-NRDRguX-lsP9jtoS6+5ghI#YX?z2
z+1!8>?X|TiwAkD@A$0ftxO-$`DZ@-?yMyR+6HTIwy>67H_MCiCi)=;f(#abSp?r1X
zgL18C588yZ-n-BmlfJ3FbT9fcb?R|H(&PtGuW^>Z5PBHp8B}zW0Lu5rmVSoJTS0wu
z<+B;%=?pWjNzb9Gltl5TG)X{jBbq*+ny%pe^W>f#=oX5W^$wtYm9=kdX&%GSRYJPF
zD4QJL$XbjDS4^PZBmuSBg~yR=_9gbI89P(8AC91!s|>?BpGB4wQyrjdO<uHmUe+Oa
zTw;CmyOssubyzE?phdc~P1M;@_sD8QChJkDIU*=N$loqK1mrA_dPvh6wBwTf_tv03
zBW@1z+8UH;c8O3OoXlgH=P|jj0XZmo0cjpYIY#`}G@^!7H8{36jgyB(k07~3vbz-(
zSrj3_yF`~L1cFjX!Cp=tA}3qX*XhgL{5CECxNap7EK`r392(^Z1d+!BVbP-q5>%Ol
zWrYK#h!h&)LlmXiAffd)lZ?~RW7K#cBn5CR*YGBvAcGDv=5W8DU{5&E&xd3wz++DI
zV%aB!{a&0QczMq6ABM`uUP<K1nR=A7$l($<)BhOwXt`iX1d;(MgayApOb+fv+3IhO
zSE+qFA1G+@@tz^QJ5W1;$paD<muSOeULz_UHPor%yjwZZ!+RY*CE(Yg2Erk*2~4y>
zdjaeKM?C<dU80vCFx=oaxZQ>~?6Swld&7P{<mSBI5HHK+D_tUn4;Ah3NaBDn=#Y8E
zeruT*np{Ze505G!(_nXaIX)nXWh?PI&M)&L<x~{q*29OXP_Lwmvd3|!PiYnzV&cdY
zxPueLVY`7sY3XNLNg?QzM=w>mGMO5BZHVW*va{xr1HUl9(?oLCF0aJchPkjJQ5Sji
zaLK+fB$gC0`m$jkmL-S-1l%u4LFm<92tOV*l+u3#V18)W?H2-q;;hzDn+2JJhK`LR
z9N<K5kPq;p0#2dP2Q-=%xPKV01qG_)At;KF2uCvSADA|!oYEU=#6l92y9sj<4uE1I
zfNU3J2%<gN`GVva<P|r+6(qZ<`u1{1C@i|eq9XX+9*NeB!rSBFQH$sKN4xbpNz{tQ
z6Y`dWgcYR-rhXlsuBZy?!MFhxdJ%vtjrs}keh%V^gCWTSX~oNW8g7zs7#ct#AH4~K
zTnHG8_1h5#!f=@mhPjYeuK^F-nwaK_OY9>>x1d74&w)WF$aFDQ*|C!K_<@rTQNM8*
zQdi-(DUbpYp8&UUfagRRm^g@l9t5gGLi80W7zA}Cs3=U0AP(qCv?R&o5PH%?eXpL_
z=p@&;(D0}Y*Enz!ctZhV!``VJ#r6#nWK@%(3<QO;UOv|o2@_IYj1{L$Giu6z&P|~(
zUkMh3_)*AicMxjUNP($D*Hb3u1<0A{T(aS14s1l1Ha&xl=5p*~)1w=K&a7turBNbM
zoURhQ0-cg}LZkE7##%qs209u1K+l=R+A47R0VEk@zZd(VA3I3a4X9u)cH%8P1}$mc
z47UT<h3X6AtRNh6)~PEI>r>>&W(Y&WE>k(2$xeIBI>(f|<9(|Vm&eMT6n~2>6gYpr
zLx-Z35f?$s*Q1rB?FMumv1~@!SI75lxHi#49=`x}GtEZrCftoEe?_cKI<KfmEC`f8
zp>D}jE_ib};{xH6Um_E=i0n;6#p<=Y=BaysYaydMQ5yY2I(yIpb@i^<WT*#W`Y;==
zORsaqiWO?jE(<x(gBmmAT8r+Rq<(00`pMTlXcJ*?Lsqk)eWE*z{LQQ=Bn{<k>g+f_
zxw{uFCgZ)RK$rOEUR16>eAsJ<CMGv3)OU*9C@(wCHq~k9hWf9^N_BYBE25G)(>_zz
ze*f!fNiBV-#2}m14;McT5BZ5hv$UpubP-uP;GOMZ;fGsW;Ju%&YZ!){?PM|eeJ7i%
zRTr>#vQuxdr%xfQ6)t3h>6VZnDv%Qc4!C?HZv6#Y%{p(PZ%&pQ0CM=nH&2h{GYnld
z2IMk|)`hhCrR+V3?0*&AL?&4_n|#W$_1bt1d(QHUs9r!lkorb%K>@?i1@D-KqSc*j
z=F#>#^vvt*76VrU+`Mc9EyTk$I#b60p#AaX1+P3)2;&&6+Bn+j4%Vs#|BVPsRrF|g
ze~gah{E}iO7eHy!E9h83A!4m5R~Yfg@fLP5xxbagIC*Gem}-3}G=2_`Jzyq_TG@J>
zG>}1Hij%0`!sK)-yA;ouy^^T5)T!@h<k3;ghGz^}NtE9;sl_L9=*(4)>AZ8Q=%xI7
zF~iW6EcuC`PFc5}&DZB4Qm}z7KsS@(Ha3@R*ua+H8KY5>q-9U({4gn%Vmjoc?tk8X
z4(y|=EQ#`WFX)tywX;PyX$XQrXh{;~?QiIm|7&9x=)(?~2Xh$v&gc4WeLGvG4?g6{
z^{{*NC%UU{94W}`YM%}morlQvK#=mue5BsICtKac=aNU)v-67Sfb^@|3#BSL+rmoL
z@=p~yIU#MJX3_)&Q&e4&nu}JFJsa3YoHQpv9#<q$X?>K+N?aS+B{*rOfw^XJ66yE@
z2I<b{P-bPKH8X~nB(^&a8f-Aj6k$4vSeMtI{c$eC(3Nc6HyR~fNHRCFd3eU8l7w1!
zl9E<k7kg<Jg2Q5eqE46TG*kSC$Ig6!ruvSp_%badfp+K{owlxGI((Pu4hama(#e@l
zwwOSxSb#56`U#{Hc7qi90LHYmbD#a>OAquLx|~yl40CwwiDpWgOK$97E3Bzk(X)K`
No}JAw18x@n{U6Ah(aHb-

diff --git a/master/_images/sphx_glr_example_inspect_predictions_001.png b/master/_images/sphx_glr_example_inspect_predictions_001.png
index 1d54b34313c4662683034d184142eff030fcbc9a..f067635f4462decf5ea5fdfce7bc0c323eedd605 100644
GIT binary patch
literal 15967
zcmd^mby$_#-t9sRL<I{`QWO&<l#o^s6cI$EB}4=%0qF({EJO)WkWd8a5RndLD=HvJ
zN~0)UO7|TTcb)T{bMHO(et+EOx%=7s*}z)sUGtsu7h{aycyA~v9$7`pL`$JiR>>Vb
ztW2RS@}y9x+Gv*IF9M&MzTtnuwlW&Fr!0+Z9rUdYDGK_wR_2zr<|b!1+Z$Tjm{?lu
z=N91R+q2o&*4D~Kgoo$CpKsu{v_8-CS^sAqF0$O}sHP2tLa$H$ql%GyW<sHem&zSJ
zq~iFfzu8T@+Ax3qhtL(;;OiTx&T?+t&=Rq8r^!Q2&U$-kt=R9T_iH1Wg|>QZyKvT2
zF1X+M*a?dZPW}(?)7Lep-@lU@pCzMpH+uQ8q>lc!=0%Hk3lGoEe$f}(qxEj(Fl)!O
zvjp2PZ&FrAFnwLt*u+@yq;4yvh<7Cwg;KcB+=vmYqYQ4LETT{}R#L6TAJ5Ww;!hi-
zDT^tTl#PovQz)KqR#WiyJzf+V3gryn;)8ho7RUeAN8TDxS<TH%-n?<+q?J{Qva<3E
zld9+yV>;=kfqwq}H7_rm*0FxS?bz|-)h>g1>hZq%8$W#bpgcX+U-R5hJ}NpoE;IA<
zsqbWQYM%NVy>hN3XrxL-9`UWcyktdypWl-#o6ep{Ux~^)tfK4L*mR%GXWR9T40_C;
z{_*37>8B^xY7;ag&xbk5vU|*=R#sMW%B2~VDcjq>+`yxo@cFZb!oz(sr!y>S#`+r>
zSXgFm2?&QOvWht>ad2?Z($H9qt78v_HhJlH{dQyB&6_v<%4$7l#fqq>#`?{~e*CsB
zDV&wgdkq7F;konY>s)T@<~eWT$sM|kg?_OW@p<>|XnT#io}IX!mX_(_B}+E3yAB=7
zd)t^{8P0KI89ker<51hHB1?AMZM;M6R6+22g_@e059Q_RMn=)`WkzYc1O!x_oSX_K
zoQrrz#tY_kMD6=35;W5|<#HSb?>Aeho0`VO>lquzU|$1*f}X9}DqFp&bELaU#BItp
z@!~d<)_gb9s%V8P9+EgQy;HinNnKsW*REdW?;5pOR)}r<@WpE9mMu@zTn-<;iu>c9
z$bSBOpIukQp=T$;)y|xWn42z`Kb_HJeT(zGMMlosiE=)(u;~-^_4UDm7qmjh1TQp@
z$nexN?W$`!o<n<plk;$hkY#P1o=0Mkp1bCzO`EhbEOf3Emz0*OXlg#}O|@*5C}&tf
zp|l-oYN1-WV#O(Z$+2U{Ow)q`0(S1(_w#acTD|$rTeti<w{71Zmy)7{@5=u1mRTp6
zYuB!m*eNrghYub&w*0t!`SSYp>%VnaQsZLtxR^Ok+t1H0AS9&M)J7!qLs^+>f%|Na
z_?@b5TQ_eWK|!^9_wJ>?-NVCE_tuA*$MiV1r26|9T?D7$xpT&)5BQELDym__1RMsm
zRn^o?`#z`oJ^d7UBwpH^?!(8A_ie|fo!`HIpWPD|Z#y+&;`a0X4(GwV@sW-Y%VQ$b
zPGg9W(MoxqSewp|Nd{-0Jb4lj7B+ChkK|3sRw}EBp)VrV->yG@{`_1<W8i}aF^?Xx
z)XA{8j1MH*$toBcMzML!T<mWx(A#_ZrHZR-?tN~ZxGzQCIa6ZOeQ9wC3DsMD#hz4F
zD(iU6j(&LmKDKU?CvG5<*Oz5+|E&)tB@9eVaX7oAMJQeFZ<-$!<$3-sW!Wa_Hj9;f
zu`(|5S|wuJb+;kIGMS2c=}unWioWMa@bELsDvA@d(vMxdc#(<U`0hZ#y!$F9{-gL+
z3#nQkg=<M3qKt6Ft0h*7k2ghLy}J1O_wU`fX?*u7DXGf?S)FWBn~^OvNsGqT(0NPa
zqsNXOt>~$VH)9QHw-pHt3ro7t_%b##^MUUbDoV-$yl)+|pm{}@xXZ}QaCsfAM`e9u
zhoO;?>G(i%LyG<tIR@mt2flTUHyx`Vr`=%7Y;Jkoi@FgUx$(|L5x&N)`FWfsjV~{B
zy5&dbMYB^V2c_3g+<x<<qF%plUF5_4=Rb7Y);fRew%si%dO9^Vm5KL^=U-R7bm@|F
ze|p^uqcWzbXV2=gX%C2t>zLIhe0n5olS!&yrq7iKw|5nfrPHmy9}^!g-QUw%P!KPU
zP~uWgs>a_H66h$Di4}J){W#RL93Plraj>Bi^_0NADhahXM%EkRj3nPU{M+iuc*#*t
zRW%T)*V8=z?Y)0P+IiozXU|qYH!N{3;FgfkO~m#ey@9H{g~n4?Sw-b3{*m<vpP6wT
z9ca$^)Vhp9;n!JwkmrW4ue!Utd)_qTh7G3GF(<XMZDX@-yRA<SVzY*9@Zrih)%d0T
zrFPww*I31oGba}x6k5;7*o!0d*=Mh>ub(K%EPeFobrFYwCgXN|s`a<THl{A`*NNGK
z(T9C~doQrhxut#0=6s-o5X+=#qW#Dz$vUOG#P_9O&dLl1wkciGMb7M6T)0%EX6o4J
zob<I8k9jvclgo^{>o+iTUs}diA)29MV!sQwQ$pjpN=383I>Y{1{)GB9#g}%@t}|X?
z)Eau%yjiAd2Le9r`P^<mRv}9<j|i3yE%s%%wB9zbeED*|9PD(PA-?9=i4#H9jt(ux
zUBaG>SK<-yDccvFp}uzQnn_dU#nAze)Qrn!J1v%AliZh6Rmd|rb7h{4l+{u3`0H=n
zV`^5b@@TG}enslo>q8YdF@@VX!o{^T0tos-?=}DS@cwPLblgMv7%r*D$HzxYN2hk|
z*iF<)+gJIP1+y=guG|=5-`^0clVg8OK_Rhf{?#&em4gQlCKe}VwDiPU|4}a9++_Ih
z?p;N7>?u_xr3j?R?!9{zyy@8`Ea&ZdYfZkryEj!VIrlg>S4`%}5i|4=5v#U?iS8LK
z&hd1tBHx@{hh8lU+ll60SSe=B(-si0ezd#l$Z$uQKPOro`ONF9%Y((7e)8nEe6fnb
za)N~{!(@Hg_ICHtoiQ<qRZG@wjtI_V<VE<XN4KuX%+A)DoSej>#f4(ZqN}S_kpDGi
zx8ve=&yG}`$R4SbH%m`GouT;U%a`nqV6&0W2EN^&Qw<nd+R>19^YA2(VrLKP{MWwI
zXYZ2t))Dn%Jp+R(*WniD-bAhE=WbZr*a(=_?n%@u(3SRDd9Ucpfx(Sv2DGbI*@-`i
zjAWjZ$6igO8W}MTec(zqp8WY8ttRyHl`A^^TV-!s%{rhcIgfNbj!()y)?1rcRVLJ`
zeg08t>2Z|%Om}V`o*j!(ULJWjih3gWX;-dPk&}~49P{&AD<dmA*JsvQ`Y=*m*GXo6
zSaLp+P273Lyj3w&WY?ZO$0cWmik!mfX=#;11TUB(mncd>jZ`Tzw>F@YDO*@1#h0Pt
z2nngrPL1ZY^%|9jC2rbpv~$OfBj_Im6B5(ogLSC}#nt(4E)Md@r4)W&7Ha={_a0}o
zx}Uzq$fKsO|Cr6CkMHkYh-?4;su~ZVnc#hfH(T7sb>95;+dGSl5{r!ZCr_mOR8F3J
zJU>1^7oI=W!_{A(b|fn+E4w!-2MPY{#S8uu0V@IAG|>gxhQ1WpeS5dY(9jSaor{m}
zq_D7X(hBLZfo86de(hS<;W8mr9i2p%fvjh>2TnaxthvRLA5&O(&~3cgK`YlW^_aXo
z0|P_Ut1HXQ-rn9sOH2Fn@O$jQ{qOf|C~j*y2`Kb1oD%lL-;A~XJCWr70r>oX<0I{<
zlmi}ft~jur2M%cL+O?|z7<kEa{_I#I=|>IeX3tO&fc4T9J&Je(0t2IfM&k`j?jzS!
zFJ4T2_F)NPWRLISLIy6iom;m`<5W}+9lGQ^))$MjWtDK9$epcAJpJt6)+0Ua^Ap}b
zfBsz0%9@BH>MguPog|TVzPz?KxgcrDiVeo?Z+)`A`N;>|zaNduI6XOsb1I?bU0G!1
zaQ8nb5=Fd7)+hq3{kJRlaRvqk?%J|N>ehz6mDks7b-!5p^QZMSFE7BB;};rUTpVOX
z#UA1q=wQikpv1?=2Zo0yS65ftTxl>T_o%C@oA)**F0S%~<m^*<o;(%57cS!#_5EX7
z23}rXySHuoe%T;Q!c7YWC!G!JjVbA;IF59k6lS55IefU<)c5-JyUiBKbsNjY#~%TB
zX<th2uw&eB1eh-+Ev@6*)zx)tRuv!I^J*mCQ_tx;!(P37ZI7v{@bxS#O>GzV?ca}7
z>u~OC%qYL2iU6MQH5#B>y<tN%;yy4SpxTrN*z?@EZ{_*+A0G)je-Bh2sXrecd&HM5
zX)L%7?}%}m9!r0F{P=OPL5)L)4k0T{+bmI@&ahexQeGN=l0s<#Ahk<J?fu@-5vNz+
z&gk0QSDz+e@OI_*Z{Kzb3uo2Z0HEBweLF$G$jK>7(4uJ~%>Zj@3(Cr)-YEwp&9vX>
zHsMSemU?BOqUzXlCQ<tYPLq+YPdycnuRKpn+qS@N)YR_Z-+Max!Gn!;<5Q@KiM5HR
zxi)V;xNe7HHFo#u$52r&K$ug}$4{!L1XNa4iJ#mD2x#cUNugtx&^~g9<s>fAXP>in
z-MT=3|Mu+b27@hm&+9*45X_(K2#HHdss#c(W%elc)U%#9jCui_;{Ot&UhH+bj7=Y^
z5cbT!Zi6oBS+H-G<skmVzxO|(kpFLF@qZ5;{{3FcO1UY<=BlcyyLa!7bC~zval*7b
zOq`J9lKXpmXR{|J9Mer};yAV~qO_^9t?(S~EU$}C&flHWoCGM)<0I6f1g28nI6ggC
zppL`(IXpbt?km{?j+nPy@sUP+IjP!U+BK#*=*8SUJk|&O05Ss*%w2>2NWYdm7on+X
zgctW;b!EkdJxTy89UnvEQCIpur%I!j+fAvfs^X*-v&UOpM$zf@tePyEv%AyEB}CQD
z_Y3a1_~UDPJ2F~ta1OwWNx(FKP#12UZ0%}Xzhb>3=gHXthk~k!AMFD6eHwC=VD2M7
zirIUc9a?Juc|q5!p7~CVbZ?dGbZb%7PBTIhj2ym^lA@NVC189$Nj`|XuZxN@<JZV|
zNkv_qOHxuVHa50%X?lG?aBv(DP{vzsZf-TivWQ*J(`9Uq2SKXx&a_rNI}x9on_CMI
zNwLdfrTDUm@AfWweEs@$t^PG@);w?QU&SW0Z5!J5US&naJ1wrm3jHm4r>U2&thljm
z$HzuVaE8Z^A158LwY9y{$_ywKKH0~|3o!C2Z@eWQkYFb;S-9)Z8|aCnU%c7FGCV-s
zp<ihGq8>P9okdnfzv>^PcwM`;3yfgfwr%MR^2d&eP0i-LywKRw6Q5j5FF9k?(s|_<
znN1zPV#&qLo!@4Wl0&;<h1qQVMky(&YG@p?%<Ld(?e>(vSZ@5th^?GoRksQNN0{qR
zIR|ft*qE4#B%Pd*UcLDS1A`|fySce*AwV>D4ekdgkY)Ag{1|%mAjNZ+m^FiytgIIr
zc;vTl=SKS*)xla-K|*`u+Wvv^zP{Jrn2|F#jXu(l>y$>SNoM8&xyq+<{<g)e4lxGB
zcSIbABBB*T<4`MoScEIrZHrtUUq6}i*+8o&{_(Oyb?Rk&`+d@{ET<Im{5VQsk#Nnt
z!y^17Az@E_zdbnq?j1Xd?eyOv9u3n%r?bDHYt_W7^VF5wbd(Yz#!ql9V5Q#+k9l`|
zfpYO3)~fG8dP(SLx_WM-f!qZ%*1$KC6%UGEXgf69`RwDSQ1S~6c5q2fe)DIXd6%)h
zy5{<g8~5}#N{Y7p{9Yoi$~1Gl@ZqT*29r}Im7$e~!t?HQU)H)mpptwvWYNb*`*PM(
zJY_DLt>PDc#o(ekWNg+wd?c#8EUDwG-8~ij%AnR4Sl!xZ4Uz0H<*>g^+-P9WtT%1m
z{M>X;<zmYMH4T-8FdS^MkE)MC+qp@(zWv$rhFHsfMYkWxw4vs`BNG!o?rYw0c;}~!
z8uG6goA?YdVj0zH1;vvm?{S*UO#Ilc{x$6G-PLLGkiOQF7s{OciKbN(SHm-wENmZ@
zs#BWE&$4(Gh4+-_5BZ>W!IS8ZK*y@kS%5GM?1gzxf?Ar@rARPMt+8^W2JITK7V2D(
z@`5)y*eF(s9$tmqu=w{%-bbp%bM(lWRm_52pu48v+l0X2sMAUoQ3@kJ<Mlf5N&+(v
zU0K>aH$7Mvap(&9RW)Aip$-#;!X`smY%4g^Y*Q`{IY3EQH%#%-0m56hadLKrxeRas
zh0o9~J5=<Oh0goy*RKJ0??xa1N{mFu2e&+{iba$|HBfl-=8Y@MO3Ki3+;OG}R1CtD
z!#w6DwZ5G2GZzd&4^IbZxbQjU)l*t3%Fv!goA(T2$$n2klf%a+gU5p1(Rl0@AJou=
zinL7*s)6X>ut2}SEP!ecjb}<O(L=Uxm)(E<V?b4R!ssKU77)Z)0|S9Gg%tnYiz$U^
z<ZZb5ys-u|tu!Ma^iDIKC66F%fT<V*8fXLl4l!eiN0~%yW7LwgV+ll>ogQ~cFMj*h
z^!tZE^7k7xRFpPN7HaAlcA^?otj`RwAv@`dOE){V`xpgLQ4U6a74^j7{5Q$4|C>G3
zcS+&oa3>db;nk~G$K>QbMJt4Ge<od~E=9kHXeyqb6o8Mqem<ZZB}umBo?L1PN+>0y
z0$jelXtb~Xl!=Lni0j0V)Aa9+gW#q0n>Kad`}G+rFa3=JbWrM^NVrWIx8yqcapIVv
zODa6zJ%mm;S6}z`x8IDv{&pkyz{MxUZv&qG1DG`V%md2QcBQFI7}!efU-g4zA14TO
zfk8oa&e*{S?hI;D)V^CCSife?$#`)9Rr{P~GXtaZ=ehX#t4FWjxnoh`f-W&3T98G1
zVE_J8AV*2X`EJvt+{;n*cL>-xc&4QAE?K%X4(f}Vi=5p1B|JmMxA}gR?dphed^lw-
zu<hv=#*d#q<pjjbZg`~l3>3B|Sx+*2(>ko<{C|MsLi64kAWD<wC9K*!1ua_=2!!M2
zPSPcaFhXB!Ed^3Q($*t)c5-pa+vn)qxN!rT;6<I^QXK|$>&-g_nV~MnK#tCG58HqK
zP6w#$CF*6S&DnONOk^#q+W-GxYgMIvizpMx&?vxJYjWn_`|n6UU(N<976X1Y>aGK!
z6#`q$-L+JOejuw%3Mjt_O{Q5lK&l$3?6My0lp#jIRiEup;~;>H4(MrU#6U$d>8_0G
zsf;R-$)l%4%q@Hv;waOL{dH0DjP@gxh&312U|jM~vlG{4s|#gzp)UhsFd6o}e*Id@
zL(mI_-fo~N>(kq1p3#NBCMM3KHR^^LjWucKPvk-W?rF>zTr%x3^63Z>0LJ?2BLU-~
zRcdDUziiYlmz-$30(O2evF>S9)GoBtQ`@LBkD_R1PxZtz&$`L_FxAAg4>o1_ucm<_
zw3ectpz-+j?Uj2qQVxQV28%i*VLI}{pqTEOw|8ClJQ}PiG$!cYwfz~bN!YAi&_<Um
zS<=&zmlNL#tt#%?3dU@AI>P_6C`>U%or7n6O=6S1*cR*P2^5#|*V1jL@H70cl0B))
z1Hw3ww}S=Dv@g`@x+H>iq@Q`UNbbA?MP6P$-FdWUX^jdd7Znvspgge(A%d%zs1IJi
zTFBhs`7$1%(J873FDQ2Xw+_br3v$`wq~{+Lq#j>jJJ#Mj-&bJKEP|Wt4K_>qWZ)|v
z4G5!?a_1kx;_RBCf2<f^FOO5w8~Z!^Ei)Q*BwHa-kwT8%kZxATEj~Xp%(U;^b;7$h
zZqztm+Sg}}WtKJTEh{XiIbd@vZ_5Q!&FoVmpGWA{7~=>0PJ6)SL!UCGP({LpVF;XA
zdp;D6516tmE<ULeLo8beF?HSH4<E+jnCi;W+i%{vGj=S|(~zY(_VSU_^5K%Z-7M_+
z7?W>3#jo)xXOo}K83j>OqA!a6GvGw^b!w?QEGw(4;vv^l9qW&<8!4ifd?Z9!x@5`4
zdo&gA1P7BVi#%d@q2#(9$PKCAM)8boqVwFV^+HKm*$;&b@;!HVpUh?mOU@X5lurk{
z&ra=v$OT$yo1UnXqqd-SP4{7h0f=@A<*Rw`ri0YYd?OOV=6lz(vnK=3JGssT$J;XZ
z?l$P|6ZX8wLPcHpRqW@Qe+@?8-eKjEV-k0Mfr&)~_EfI&)Cut@6tbT5I^C*d_famw
z<6B*aeV`fcl~b0HxrTlftzYD|Yu`TQ9Q*!Ckd%1f)u^Z_qI9exl>TD-TY;%*Td3gi
zwp0;o{w?XP_?`5+>F0XP&uP6hf2iyDWeGXzo~EocsH6QN)Vu)ot#0F?WSEH2jEszo
z36$7iK7*y~!zk35@3(eRrnUe?1+#tZruLlT`-A>UxU5`BiTDy17su_mp?8vhM9|g8
z?$Bz_w*X5NNtHijes+nZWx?AQZT6J4NzG9LyGox+;IV()bMp_eUuc9TY+_=<D_~jS
zIZ~sdALl4V%KBu-%&FC1Bj_Jhrj$GK{?yVLQerx0WD3t)G4OiI7L8YkJ<H70P}g%y
zuQ$vzTp$L8(rBuEltsq;m^x&4g=RrOOu83m(qL*o$JEQcYt^dO(UZeJo+tO~E>e@F
zEcV>~lyA^aYGP@^6+-VZA7(vBF<t}+tqdd$Kpj9QnCOv{v8Op!_nzRAN5Y1X{^uAW
zn+IQx6#WZWb9a2T{YPMZ9fM|t*B!H}&J*+uMnPZh(|bGYhwMLpi@Z?CDn005RQsgm
zNK3W0Q{(3z7gcKvF1{XEv~+0GBWfCo`}O@BspdN@J}w|ze!hW@N6~h>Hlgt~VcCC8
zwXqcdsNZ~6SY=Ku9U#+-wnI2XQdg?S-<6kJzHpH$Vj?2u%;m3N&xQhz=(?Q<6N`<K
z52|(PYtC^n`w*}T=k0|`&^yb9nl#$OymsxAkB?qmpD*OY)n{(wgF0@bHRLWrMIF{I
zn#FLz8yiJ&mwi*<8>9YIfhTW7InSYs`)zmCGrOUMCG6U{^Iar~z|&-&ynD{=@LH<X
z3($Q+_=p2YT5V;%`cpTb;SD#GBX2j$R9^6Xsiyx%S8i5(eyQ>Eg(crEn~onDV&7fL
zU9^{-ZqUen`Ow1X-u><G`}gUNF63+siy~jxb}M5r^rpSt`|8!5cLwLbND`Vbu?fmJ
zlaPgW-n1Ulb=50jPb+eK6naxkTo7n+4tgJ{uI7-hb5uevMbqqUE_m#)h^h(irTpJd
zQvShrKvWapv2{orq=?Gk+>R4r)zCu-E`{P%+bi&kAgM~rk8Jw4stM2Md14p{*mdd!
zpa9ygCM*G!7c!bVoLhKTDujxtk~s}1y<M|iT{or%Do74-RY+t*TmK#$>Np+dQSsbY
z81DW)t@+E7K&CYq)+;$WI)={axlcyP-eHMF32Szr?kBYb%z?~xz)j@+$b0|;a~-1m
zo0}&@AAk4+^2>$Rd>w%Bu$>SdCNlq%==+++tp0!AcQ<)Gj)~bxhMmFuCVrf;kZ>WE
zXuUALCwqJ2k6QwO)j-WEfXHwb_~aC9!M;3ar_Jce4_RAABJJ0Z(9u=f!R8KTEC15b
z(CMZCBDingKGSs6W|PsL8o!U*t)XsVtW-A)(FNS&q^@p@6LlNO3849f4mdEQh4VkR
z!@e^f89*}XwREez&uU8ZcYBJM^GFOR<d!X4Ow*4XKCC@->w$gFZ0>BoeNOLW@XHmW
zlYjUYuDD|$*yi%Lmgl<A@*V}XYA<>8@S&!edTMU2_NPyuwyU99G+ivNnEsZM9y1=8
z@k`-SiLCuSW1@83q@|?RcSg2S+5Y>sJ7+#0ED7NU6Q5}<HvpV+Q6nX#yRa}YO_?F2
z+1LGQOf@+ke<V@3!I`FtCo(Ib8IdO2-QDf?QFpOlNJ#t+C{DS_-F3;~WI}fF>tXdq
zj1UOVkW)tZY}vXs5EckpS0ILl%!`p^khCyma*qKeXLLJUVppeAceG|g(ARbGBMhDR
z4BkS1)`VhJ)0AZc^&l3rsbhW^1twyWxDz6n-$?bT6DDni3)AwEc)i(UB|$uOLanZm
z7&prMtzW%50)3_@Do9rusiIR{cvwazOKLlAjOp$F)cFutW<QW*1A1JCdH2Zc6+B&&
zo#E=Z22h_V3=V!HeQD)zH7NUv{k%>xa(1+iYsU`n{iNG8eft}S`EKzsb=I1AnI_gP
z+WcQoH+tj0b}Y5u6TLL?BzQQnq@+Y435Q*nX8Kb>da(0L2GtNI<9<&;O#Qj=b;c|^
z6dx%W8+XYRe6+M&(rGI~N;SMRJ2!7G?6c<9f@qJiWI7Bk7KWXMrFl*dT!wkaJM;(2
zsscJuS<pSCPM$pZyq<fG+08wN+BmJ=_6>vj10AnMG7$Q<A*$faKv_{!h<BC{y^Up9
zw8;3X`pNiLGPs?;WPk`?3>PXf7G_w#{+#`;`+vTlt)oRuEzG4Lj|O)^=`5zh$Nq_X
z{NpZEc<<59RDuR=6r?W)_*>UHKQ}JY4~!iw;g-FeL9-b1B<_hOP|rM#<!!8xNkuHD
z;i(9C_^`YA>>eWo1ewz+hKXUMAH9__br<xC3&;zaW=j6dPx-Wx)Br|3*ONps9dsYh
z(dq{agK;QWPE`|Gb6H|;qFtmPI^Ri%*$Of5n23Ih;OoJ_IK2UKZ5#%!4^;vMW;%ol
z7Bp9n`RUfAm#m`pJE7acO?Bx;;A8=4t_cK5sK(XMm_eKL9_sj+(0EFGgK9}emUnO^
zX3N}@=nh~RWU^5orTN-#n>@cw$8jH4Q8m;kqaXu}ST->Unm-^J6DsDUyo`>e0_8uU
zKJC0&{c}U3pw0}!yLj>^q@WSUjLsXN+_=yDII@mw*rU~bI&o~UH5tAbf`uX2lm5T}
z)C)h)1D!Cv5Kt^SkaEzb9?W2~zj!fMr<<vlJ=}i;IgX3#EOqs3wh#hWH{j&t+*lSO
zBs7rOPG^AP=vDNEH;2CfiFO7G{zlB7uY3Z@jKmm8R!;7Qen#V}kTN5CC->xth{CSg
z#E><NjOVqTyQALT<&eG&|I$O9`+D$Iy@fb@J7+Mz@%qi1``}49x6$iW@c3HqBK_kJ
zcweQ|o6Eu=-~+OU4cvyY*2eutrM#G^`*f6rFhg1mx)%~+X0NIHiY~(U=Jo3pqjh@8
zU;*!q(o8U6C-bSK?4I3FYd`k=oKZY@(37*9uo-~ND16(_UAyE6Sbra=DXp#jDBEE$
zN!(>D29>tRn_m6gxu^AplLAG&)!BAtz@Nv6$qxsZ4rr2A^8B{1<eUb|jygPD)o=~w
z4Sy6R%5%M;<_Ca&e!CtOqRP?I(-ZR?_>Jv!e});t!YHV2qEs;=rpzLPIXQnS?h`hB
z>wyEGH}ajmnrh##1t(<WlP9ko$Li1fa3$wfe%n2JqKxj0Pq0U*kXh~R=DV_E<sNfY
z5Rmzt#|JK$x8^rH;?h3A)9v}w1N-l=OIl#(ON`QLw1(ja3tPj&Vt$)d^u9prJ2t0p
zYqkQ(J%9#741qX_O-KelZU8<5U<RCe9l+*oVVjPP&}R0d%j&OW7A&1;Y@IVY5UQvi
z4&ilUkhb-W8>-IBs4Bdh03R042oso11@p6mK47mk-KUh5&n{lP_<>TW2NbA#7eVnv
z`Umr;lYX>ap$}HNXZOPR(`(|exZTrZs@{nFJfrQ>pDqhq)$xSVg4s?<=aoggBxSd5
z*>aik`5pMDk$FpQmg01-)9{`(Vx`r&PBvR5Bw8MbyS%&z#y&i@i;HW&MeYw-9`z14
zf;Nl>rM@P3v%70x!h8yc#QnLLcV&Nbjyke+<g8)LJ<!_|5|f9~RgHqa3e_g-C4<S;
znmCArI!}HN5-|L*eqG%2B-k%dB(ml#e`2yo`Vz*e+toWH6~iUmUQITo8hFFW$|P6n
zX7CvfF^JJaw`LP~#}^lFzrOk`5zH^tKfh+M?|5Oh&=I&ZUN{XK)#ffoDb~;*$f9|=
zoXE!RQ{6k#N+`KE?JViWXi8B+i)n36{GAy51v@k0J*xcThH_56<zU;EnJ@GnN^ncU
z+NEo7ji+dOHz~XiT<%;lxlG2cIRn}9<izTuCIXw9v>^qObhj1>O%HQ{g+d{YRzPFi
zr~Tc%tqr&t&Tb+@Ky9(GmZ^P3YX1((6pPe#VM1g9eYmg#n7|~b&CWlQoPGcFL&7Dy
zK{ryPdG#;1xD1M81}>)z4<%{WaGXgabM|FP&A%a+lJxfVbbE*1!tXeAhAgD(Q{)t}
zk@jACn+Aym<o+nLkVQ2tdZV~{Y%R5qx1Hr3=ncel<1tyHr=q152Vy%rS?*CgHQGB`
zy1$&%G6T}5!i(r4=;1KUBluNseyYwW=&W#%j{QTV`cs&0AaCp;mxFSesGX&Z#Tg`y
zdCau4XCpFgVL`E>O`~A%N@u}2r|u#$M+aM5`}XZ2-YtTW>kT#5_O`H~Lv`jw>=F~x
z#!uT-&asl#It{^mgoDic$w+F~sptnfH@}OGR-YmriMW2Cq{KK4cch2SHxIg>S2mpU
z#Zp2n^EJr|@9)uk($;oHt`ZGPz^HU1tE9VjxZBvn_Y6}!zg9%3M&~0DyJMDlqxU7I
z`}Rv+4<Zg27?lG!F^@>BwNX1j|KWIoi%ltJ@x*qOl=+pG$lrc13`;gFS+=aV>V%{~
zmdO0uUq``ukbR$VMZ^jTM!+SKy1EHVhv;haoWN#AFgJ;hi8<<5jjAp_R7h=F|2*n_
z;H2cjVl)wUl0;VVTfht2%PS6$2!2ysY*rR5015I|{dWyo8%ws1d_nF-KTg74<gjE7
zzXViu5<vj#NI?~7GQ@3N_JA+)W0tM4Sc1zS#P=+8SVQ|w_kSacO!lF4Nz-t2O!ixP
z#AEZi1J!ju!IKmZLZpVVNHv=1t2fJiZeuHVggHtP-UK{Q%KgTA0qb-Qtx2JOUs_r%
zI+z=Y;d3_}NdBC&!{PJQWE}mYy|@>VE{O0X4kEZwx;}~yCe+MzO3rJdyl2v}960m)
zNuf!dhSCJZI(_ZhwLMK|Eo<Ubx2tx{|N0^;h7@{DboRN4V#!)s4squZCDgK!iFV&%
zM*XcK?lYN~N%y`eE-vl`Bp^Y9gL-kHQ4NEm{W&Y;&;A*}%jxK99a`O<VY;5JHVqef
z0Ia!~LIkTkoR<d)!8}|Sj$5t%b=!|u5L+{P$cLvNX4re#D7n6a2wV_QQ%o`J8am5D
zIz^Icd<=jRoDnNE=rrQ6iaAew@g_f$bR!+8{`;grZJschn820(2@!vTAU+Rb96bjS
zZK@{OJqoQz89)1T0>%>?_tZE46Z~f|hvFKsgmt`nW^&jpbk!*Cm}~d$58pn_Fe6nk
z255MBA!I2XOAM*RxY<*n8EBD<NOa7xz%t8mweY6Kc*n$8qh-3)7b!Q`1VwB*ZiCxB
z_GWWF>NYjP2c%KF5VH_lN~Q*K2CMU29O}&lN&HJTX2A8L8%H~oAMQRS=P}!>w{V%M
z?q?9C>YynmEMs?70)r#Hjg(!tfh^k3=dP|E=G?yh0~}|_V=NQq`ijXpaUhz_Kp#@q
z=5E;yo?cm0K^#D+1R$ZXM44b)7wm5C(-&5;OEmSZo!x*Ni7yuy_1q3PleGG5T0ISi
z?g+6yzyk;$Wg(CIu1xeq-M9}-8sqsKl9G_DG_VPzK0^9~@~-GN@|M}Mi?9`7r}u$V
z?0;mL{~9URqRw!Sq`BGYc+Iqikgw-IKKcY7dn~g><|`cfy51I77*Kcb*@MCu{MMH}
zS#%&%`qHIK=~FD?&MI(AG~Ihs^k*)-pBvP(kOasr_gX!-#a!&l(p8m@rM<~>3h{a1
z)%g7^<A)*~2f}@Nq(YiL76^f{4{Wew++%t0r9Z=hlb3fWm*X4klmIw8uJUgQDT_v@
z)N!Bup56ezZY-wwAm$+mT7%}0_U>olYVeeXj(K`gP>C=(QY7e&z>als!<N4aC;;-+
z)nDE4je)jfVFf5*W{JZQfV4n0qB=HZ0+=ZA#UCyx&K4@hOjkuo2}kwVEGbL2vR;@T
zc1ciG`@YYQ@g11D9wPW9OKbl1pN;3B0?l%q@)dds$ClNV1s)!0ABN2Yf~J_g9-~dC
zo7Y#4YJ*eQ!hvmydACtuXB~C|*`bkX^%#t3Yj>YI8=U&j$e_Qy*tc$WyfqoU$qli#
zUA<TD*EXk6cTq3s#02OgR`W$T5NNV^zq*FT1I=wyc_>7L9uU_XD%C!G2XRZ)Ch5cj
z4cD4tXg?@`{>Xa3>LBzxCNakp@C8%lz`4J&^30dl*ZiKsP$hL;4ryy@zx;ktwvajo
z?ZhzE$@3bMxJ(IqOY{W{Aj)=@C-V{FSI$A1S}k=wi_C$6xs-_4e^_!3V^aUPpB|`4
zO`zh=U5~v-8qAWjEG~@#PZP*WY|WhT&B+mU_J-=zj$b{=b#b2os<R_V@F)PH8ok6s
z1QEUogCSx!sZOLU^AyZHuirL>8Z)k{qSA$6s0G)t?I_(Or;L)#kA^opH(pS;ElB$b
zScbfaOC5_N-5(W#p5y9+9T;^H>nA`81Jq0~m>B#DV@oA!xPWDgSepFEX{?`I9-Tjk
zllX&3t$xp-xr|&WjM8-qOEsy93xtoQCe`2wCN5_E4e34gY0vO<haC>R2hnU4)Rmzz
zPHA%!Oj}VXjQz1=b@_w`e+XChBQ_Q-Lh>F)P3GZQ8q%fxY%xrRZzkW56JDL^-0_%e
zohbuUh0_bsd^Rk^B3q6C7tT=euy5qf_hvLb!DAG?2n|8fjI)<f2HSU^VOL%!g~`%)
zS`Ommhm_GS!q4c_j=7d~^Yo7!JHjdqgLI;?lgbV)PS4ofMo#+24_hX<P=+M(sk@{x
z+r_`Z0-_<HBzj5Yeb9Dmp)1C1*sJwG`>_Wpszx6l#gd|`u6^+u!kTWOWl*}dVrR%^
z!2(1~dTH{h>ubd+53Zn@TsJBUHbMJHvboWwgH@#gCR?IC7h%~pLx2!i_UCIq^Esrw
zjDLQ6-;iw=k7;F3)BQelQvTR|^TiAIxAhGBbKpD7pcLrEhA14rm6*R=Vd=GO@#1%1
ztQ=FB%=DY7J;(8+1P#Yxwcp4`wo(?&e}Sq1zdFqP5N=j6UYa0>{!r&~)BzSczy@hn
z5nCnDJe~3DkBA4^sgu<|K8+4bcO=wgwN&Zw&k`Q9Pfuim#a-+ZZ8{0TlbrkEyIrkg
ze&LW8A7q!J;VVZlj?|vEB+mmV0lCKKoi7iA4{l%T@5_ghfUI!>8i>fu%oOQYe0EB5
zcElfw#IoOakbso#hV*zr^Uq4p+~BGJf(`ofbv(0iRDp(g&lByXj<%N|Y#;C$*tQi?
z!F@@XC3%Vpiy+T^q?HVtEu8A22om_6NjU?Rh-h_%a0wlP;_<jnb&iAO>}aw_4MOQG
zn4XUvJ*A~s*i*s?{gp^1FU{+p+`LJP=P^!eYHE@Kg2!ciyvW<gI0w(rNi6a*#Gp`-
zii&Eq#|6bsHfbpyy5pb(-v()cy(?JfvSBL1SbOn7G@(+^ZiGe39ugL}25~fIbxAr}
zD2r%{QS(#D^U98nnR8<q1>L0UWmq(qa??{NzF#&Qv!`JmnK$%$1(UG#QxvB?a#JW3
z(Srr^Ex-tk!0>_KVLr@4@DZuT#l+m@1YeTDITNJ15IkFc?f&)?VF^LH&T{zXL_5g_
zl2^oR1-(U*oH+Zh6Hl}2Rci(=uxltPDzbwmciryi3lZ`A>s3IU2%8wiX#yi5Cq4if
z{d>ORiM6h*&M<#G%LcgsIh8;0b%jDOzZ}@VV2)V9LcabU5G0*ojKs4G#PW_O7HU=r
zHASFT_g-Vv<^AK$?Hf0;E5pD}Dk0j7U(pvO6uoa&2W;nO$4PX;>Cc~YxF7>U-G%0C
zRX|Y^J$UNH++NPzmz>%W%E+Ost2Ne;X!EAf93`R<p7@LJ40B0Rn}%zb6zX0i`@UFA
z1;iFmAcl@6n8w>RAea;JC}Z8#n+fg1<&t$6ET}zmdv}*(3ePw}<P!N?DxAPPA#s0~
zDZgclHgMiAoQe|jD;Rfha2$fWCZ;Oib<*0dzo7<~yh7Gj2UmlYq5{qP4<F)c6lqLc
zVdecqo>IcuH``Vz_&o*jB*@NnKKULAt3RqEK`Zg<$+e{q_md}@(1Vo$oh#0VyG4VG
z2NPyT3d2Q)6yHf8O99AZ7g}=b8Kltkj?r9Nc+~;V!^r7*%%<b8@Ytbgx@A-5NkAYx
zJtz2UJ8m6xeMfo`qO8`W6%sUD!^{Yb1<ru6aD0_Ja#0+QPyW*lpT~<m7^;Pi5lIif
z{L||@cJ4e1t<0{0o>k<jQCX&19N*vX4$eV5C9z=Mc?r@=Mzz`B@1KUkF~9J(Oq$=k
zZ^J+crcfC^KXV=i^<yH;QW(WBu(2hfL05dEovp)0*cPo4>O%P-alvT<NsMS+(AXJR
zSua=<sQN2LIIga?fp5GAxTFruU;D^Ca&@BH0YU!}xPS~<FD+Y>@?6H_va>-kLLFL6
zlx1oQN!3NJZI-$|4iG(G2{&val(XG1YJ2GuO?fmIs$Z&z*%@*qL#xadY9|?h<FQqc
z`Q5N^6Y`FZOlUcApApEt5MA0!`lL;z-l6Xy%ejbsuD32(-P-y&wA}M3VS9OaPC)S`
z2np7?Mby*_!a9`xg2^(=QttBPzilfUKp*g$oHE`F{c@)q`gSC6pT3dFpV@+)7`7d3
z(K-Fn{2pi2)2GI0VN+jk@$lm@RJ)$)_uqOzPyXOb!K>FYGxySRfDTohw#<{EF*&y|
z<K*Jnwje_jzX_~(UC&Fnt8OxI9Z&jL@ay<!KYP0Z`w_DUf_)TRqYEsi4n7zij&IQ+
z7W;Az%L{}0;iS}iFE4gv`@LJV({QA0LDAyp{xun}`WLpGf0V{C;r!2@f-aq3mK<Q8
Vt*zyL2;w_MPDb%?lGK^Y{{ta^yIBAL

literal 16399
zcmd6Oc|4ZuzWz&tCTUzkvC2?VWR6Tt<{`6WsAQHQLuM*zL}?&%kuemBjEz)C2)*W#
z%p~*tyKc4DUcY_TK4+h^&mU)fJ}o}p_kG^`x$p0AUDx+|E*(;pqhH3nj6$K%E65+v
zpipRiDHNL4C5!PF;mXFZ_#x(WP|r!z&dllbu~ViLm19o!Hg--nmdAg0F+JsIX=f|Q
zE5s|f<97=uCwoV6KE9KGyn)y5lsR8n)PV|IWU0OUVMhvu;TZWxlPH~FNufx4Dje9W
z<r>k~<YAz#Jv%oj8bkZ;;FTPvrT*s`*QLjq>R<0XVc8w06MIKL$SQ)_NiobXpQGKP
zEqnO7re@e3v$jydrcUd!&O>*9bA3z8DL1sNS@znMn2z2Swpj^prpK#RnzOz4mShVW
zlCM4N5y{Y#_pPVFYt*Qrbk~agSk%K6cA3lZ)7N@6g@!`e;ZIpYp&aF3D1*OU*-W9O
zQ1n*NtfEk!9be*$KmE3!vJfBIK>IsB@_N-j`-rk&=5mW2E_sF*7}#`<9*xP*&oASp
z!&NL-y!_Brov@yXiQ8^~M>@LW5zY7SpI=_Sd|4oG1uiM;*!qmd(4#L)*E6DI@v~>o
zWE48Ka&6o8=BpLHJtE=pV=Z-c^}u|``w}7)qw#vtZQCxA6-#J%@j13}VWEFp_0?+H
z)-SKGtlzY0r?j**Cl{Ah>Fu2%eR(c@`Xk*nN~uS(by9RvZ(h3=bGq3rL6+_1uA8@S
z$6vg-WMXQHhnrjB{Q2|mw=-hN84+tJy6Hx#H*c=HpOC=A$EVhm>#TX|R7N1Hr2E@h
zT|?(v-g8qeJX^QQ?Q!Y-;J0+G(6JW_y+*0#fl@=?%By+O^Y3vg-L}B*RyE1Df^Xb#
z{#M3heiOHQ^stftu?_T_IQaS13p~bZ?;X6t5!rH%Z~Jyd&3NU?;jZc)A=w~SNvAr)
z)0U&XRDl)yuV23&ChnXZuM*Csu$`M*T~9Cc_Rb^m{3gZIy{`A~-(R+JrH+EaQqeQm
ziJ+<v5!J1NzBM&9@5;-yH8iST?=2{}b@OIiL_|ciWb>{SrlzKrb!mpVBeh336+FiJ
zwXwMY{O8V{yA>MhO65;UPS(=b*Kb>dlc{IkNC{$-<_#~8fB5hrbKkLJ$F^_Zp4;od
z$;r8F$&y1?e%ncw*;S)!ToZY^nOh;SE$u|v)vGHO(b4TOvZ1A?Q0y(^`xb22up#ke
z!;_>350nodKCGyqpguL)Cv4yP%sn6^EG!wjTvEX(B5!lqMkG8VGt>F&o7D#o97yz;
zoBc63$e*`^LTTNMrR?9g?`vCI?3XWJnEPJ8e!Yx=p{UoPxw-jrK)@{yP7aQ3($dCE
zOiawX?PJ<<zI-{3Ykb@wxOwyE;I@g$$#->i*=LFZCTi0RQ<5|jtXc}a0-tntTI|$&
zD8nv$byxZ>d<{SI^m=>T*2he9HEfhveXmPiMn=ZRY`YfmAd~C{ZEbB%h0B*Nsk^$m
z2JkB>DP_40edA%yOgfzBk}4a(e1Cl0wf*bYP8{RAw{O)wJ@d&Qv#jgF47X4y-zlmS
z%XV{e9&9TKLU<K?Oi*LnX^?ZMy}dnb*Riu4mW??MRt-;51q1{*B8!aD@t)k-$sT;^
zeKo0x0|WU(fkqyE!U`R1);~UKs#(5|5bOSM_uTR0$AwIaR~(d;wHp8a`TaI(-zOUp
zl}K^z{TEi&<U03!?emffZ*i(t(a*Mx^k-m8v#gG{{Pggc=O)A>^DAx|`{B+?ahKlE
zs;Vk2Ev+~|zlEk|X5zMAJ{9>Cv44=)aC3Xg>ND-INy7Ee?Ck92D_5#7ujdhRAGVN-
zPKfr_$MW63UYB#~{>a^Qs&wbOva*}e(b0m6E_<j{YJIj{v`(tNwn<vtka0*te0;KT
zk@vgz@6*PdgAuR6H%4ik<Wx?wJ=0Fn>6)Dys3WCf)v7~1Gt~)dJlnR(^XR61{Pf5~
zB@v09v%GqY#?R3phicmKJ~G5}Zr{{g0XK?vjMNfe+j|I@%YlKljZYm-vm31Q&9fVB
zUB7-eG<3trKub&9G8&3+!ta#arEF|$<j~o?hOc8cjM{ixKfk=t6B{H|H9ax(lm^94
zDAbZ=Z|sxR?6MZ+VZtQ&CP#WYEP^gwstgmh{zQkR>|P{#CTDZhLR3p{#r=*e%_&XS
zf`S;16p)JcQrH*w_Lqe-T*e^o-o1T`7A<n_H$Hqg$!ltqTU=Zp`|2#k%gd{}^T?A2
zYL`%hwk+|bKcuIZh;67#vh(&XWcB?1TsBQ!|IzXDI4r${!ZT;i;Q05m>te-?^fV8n
zmMvfI{E}WY)KGS(F0;Vtfo7E@3+Jo&h}LMdNjGbMs+i-qZCqT%U2VGB+FhgwU%h%W
zJw3g;q$TfkdXz`M*sYn4P!3Y*b@ovnY%R)|dFLS{BmVuKQ;dTY`(CTZkM<}|zK*4N
z$JfA@)KcM$`p)V{(Ni<0P16jLzSY%qx<$JrXLn}l)8iA17P+^cp&S0jmnN9E4!L!F
z;WKxEun9}+>xQ>d?XGQETgKaQS^GC?R@NKd*gMhG``9u@6E+$Axbk_$QAuqrE&1Rd
zKYlQqPEYnP(9@lmm{4U~i^HMEg>oMz4SjukLnSTl80~@u<-F))IUBGAI8ffC!Ki$P
zV{M&0)K3$Nk5&s&D76PD4-&3jTcr{vl=OZkeOTRo<aKN-Ho0OmC0Bm$-ZSo_y>72d
z47+Yfo^o`oiIofDncaw5y=kEg<E1NCG;M8dBd6-dxUS`J<!s%Is<xBw<*pT(S1eZi
zQGPF-J$v?7zhe$_oDY%zUjO_bwA_Ev@1}}b>`3vns*Vq+e{2~W7Z;~L!{*+(+0D(Z
z#+9B`;?v=F(&TSz(zbTjq}&P)zBfGVge*=vl5MM~s#;qaE5}FmdFfBjyvOCi>_mmn
zb~HM)31%USYb!U2YsI#-K0imQbcuCMviAFJq-9p+uXwqOSzP}3@q6xL{iz49{H73$
z;(71hJt5O~>r|qpX3T@KsWxOyzwOkIb7(785cgU9{bQ0Qs*+)W)%Kk`we9U6Dk>_j
zTenUtogY!qsJ`D(^;`u-M^~3eR8&_%L4lEOWK>5(Bly&*Q{6(c2M?ZqA0?TRl9rn+
zJLON$vW}TKnZ+(&JxFpiR5fC^1C>Adziej7)f*}#hOTa5XSeoy^ZXp0^~8_%ckkX+
zDe)D=1+a)(y}7nIgK879k)54=|MTb119ed#XMDJ^I)>V^URop1!Gi^3%^vj+kDZ&I
zp5EAJQhe26T1-mH$S}{PW~9zIot{OcGjLCTqOhu!RkDVL#zyIo+qc)yB3^m#J~#i1
zo;LT^q@yA_ozd7+eQv5~E+x${Z}vpXXkP;l2gd>F>F+N&TgFC4v=1FR6o7rFSXtXs
zY%1@aM+Bq4?vM*_venhmxji}B=Rs}zRvw-}%{1o`6x1wmADN_9h`D}`!PeZpc=4kD
zZGuKDRjG@)(2?x&2mPD9r%twabo4wbVbw&*=TZ*+U4dEDHbO&FbJL8mdlzSUDF1%D
zf|1+zVq=k<C}Uf0-MXa^jQtwUYnf3E7dcr~=rwio)~)-;Uix#3i0Hh1`}X}drg=bh
zioUh8GS<2G6F*WWDKSxg>C&a>6i$sBGJNK|-8_uT!qN*y!;(`{gsna%GPlJgB=k&l
zM27?g1-bNox|`kNeI!9OQscylxW)9$ak#0{y1F_h&IR*vOnOM~0)h!|bxjcf8)NxD
z#<Bl{Jp4cS5eI#WLB8uFoD@zp;nk}HC#f0j8D1@}ZS-h{#wY6^tK+zC-n^-mQgm#C
zqLPy4fdhWiGcyD!zId_E{1ynp(Zi}seN%MPbrcmZS5;Tv3JFQDZOY{q6FV#y#9DA-
z4xw$)`{~hShxA-hM8w8}2M=0)h}kRd)OjtoBHwLj_#IEW#oJ)c`h2%c{Vc2dU%#G+
z6mu;8b!H*QHpuR$Q&3jc8fY%avZ&mm5RAyx(z+S1eCHl8z+CU3NVp2tSqX$JWZSr#
z$^Y(~H}VA5RL85_;J`w+?bx9VfL|TTZ{qDzR#l||*qIU+cVO?{z4TLykYSDT|7rnI
z41Roiz-ayY^)(v=ckkA_bm>yarp)AIMZ~JG5wa{X^*PP=@xeCK0yNo_O4i?gi$`e_
z4;U;Htk@<XP_=rTcjIYJ6SEU1w(Z{iX%OTlx5^k8`q>)gqUGbS8I3=B3=||VNR7*Q
zdwY+tZWR{R28a%<JF|c;LXQ)ep~Ir~1I@nwfN=j*4WfN!pXR#sbz2nK8#p*T+OlPf
z)zv%KuGznR<mu(r-qUlWq6LxvOq|(s@k5l~Zfa%%pzFqypkzSZ-3{3Tra|GNc2Q;(
zQK@BR%3wq(#UA|p{J_!a{AQMxNg{U5(>F|Tjn)p~@f9%~_v7L$rpE{IzL3e`uGsU-
z8I$w#4c2lgU*kY&R0X2fmX-DA=(-<bq@k}Lrjurn1Uy~h>_tOE1A1^RBqSuty_>uI
zj-cEoai>Gd%2&<GLdhZ0vq?4nEp=)CNm+RH=x14wppMsDrzhp+>c`3kB=f7fyXQvv
z%(^L5RjWy-{`hf9^2Zx?ap&%vwoluC18NoN6yJ;U+HJ`ubNZE(FZM(8P9ekZ_*2+#
z{{?~lKbXb;t3@nf(X-cA7<m9T5rT@!${#@@c^k&^1`4a+TxTbwnU9YTbmXB<bVsz0
z5&jc4a&U0?ml!cJ2ZXACw8QfkE*QHszK`6~gR2JCp^tsOdJ4(7qnw-OdwYm+Enu54
z*s^o?M+K$ZJJgZEprUq$89pckBNwEn_o9bhPfs5(Ejd~H;LeMeFE^zFx|!c370Hr*
z<zG|V#Q}!E%80EjyA~q`iSNFn11&z(LQp3Fi_}2`waiI;wj_uxq})tXPfssk^x?x@
zB;h?8Q5}K0mG9iS<KCB5M+yNckqK(ig(v3zMrMJ2MrLggMrz-loTg!55;#ya-Og8&
zWv$D(b!*QJHK|H)%3<&q;|w1(dcc^C=?myq=HwhlZpYs}yq|8#(xunY%)p+kA2TvC
z&V8k4)lGKpsm=ZD&ocayLE7Mlm$7k6bbb^OG)Mv-{c9fTzjEM^j!wL!$LO{#TV7a_
zeAQD2N>bE7nJx}u1Fv*XQj3<-&#}MzuC$bFwL(>nL)*tECy$t(I3X|Y)AnYp#b;oA
z=<fatk54v;ELym58+vuH(8+qCCa<w3tGcwvm26TEfC(qY$4Q0%TMu7DLE%p|DGB7>
zxl;w@UQu5DkeOM0Tk*A-sj<?AsV{GfUtVBHfv6A|O)$*FP>1z1Um60AIJy8C37jP8
z!#eG$Y<A<RD<UWB<M8{RT7p0IppgInSINyn(BfN#=*z`yJs3gJDa>^LeQoNhChH=)
zoXpFYm#ts7?&DJ*Q`7qZR0*h7P=4wra&mHZXW39Qq&R6Qtr`nuT=UPKr5)*|W>!jX
z1@M#v(;cp$p~T!>;%gC1NEXVW_(<x52Oq(MyCp`O3p|8Dz{{f~kD#79OQmIGfHS)}
zce6nF0AT*;hRBI{3;FG3?-u6$``Ln5jA<iWS!{Ej_X;rEjP*D9$xvhpgoHZm7q3!z
ze_qgsN9VzzlE6)oQr=$aG!Igyr#&UdThBR5om)!r&0I~XJOL#2s50$o-o&?X7E=TC
zI+})O*qy&r?rdrI-e_~~znXi&qjc`0ZI^xZl<3vDq$V)Jo#~F|WxR_hN7Wa~2uYTl
z<{7A@bQuRJ&I#m&^3<toH2>my(bRUN@I}f+F7t-x%T%T{?@&#JK262$2{)vrJiCXD
zIn~eV5N*@Ea~&gN0xFBR*Yt80VbNOI>Z%F3=dXXPTSf6zAa6sD6SAy|lbwqwZnR*g
zPy~ewcsfxZKK%Z#36KvNewQ!;^k24YS$893zM^8Ep5#X&@rx<Xs7rjmyMkGx^>wUg
z8*4qsV5C0R*k~XUxR@fel$<LLy8YX?+JA@u<H`<v|2kJD|6`$s2TZp8nTnin5yQ5s
zR)2_isq#C1`%6;)X94w}Ag`jS{yhO3#WX4_E5T{2UYuXfBEg00br(Z_K`$n%7<wCC
z`<Qw}loS=f<&a)ub@jUl)gB-SEmzkZ&0=tUnWo*Q?477uBmGTBP$YsmsC_O3eZ9NM
z9VX>9sm98(X;W%Kl)Ivu<jBqH;+vnAV4q!it3}VIrKRcE+NK2u2lKA%?n{k}<A7R~
zV^fx#y#3bg+qJGZhnOde>7GHLDmBaG-m&9>tOw*qz!?J(Gc^0N=gw_PPfSd7_Yy%}
zOBf2x^VQVUgv|KGb<=M9XT3zyY1GK{U%6_{nk~>-)-y8JS@7}l7WUG!2Jpv2Y5-R0
z8F9ngbgCdz?$m#Ls3J-dG@{)iv!M@6LIb*7lbJH!5K)IW1m6TZ9c(KB3^nSmP6+yq
z*B}vHUnO+c-e7_XvrabXK=TWJ-|}^EW~^n;tY_zk%|Iq9z#z~CQ>#O9r9`%bh0j20
zd!Qo`sH?7iea)IRH}Bl(%DH4RH#_rVw9qwU5OPR&b3tBP2Ps>#(-X|Qj{1t&d^(iC
zk6k$=txBc)u;~$j65qAr2MF2ljX(eAga2P`*8e6}iN8Z#MEP!j{<eMFwnWFdckdMY
z8nV?$@Qsb75239)SWz=fy%ZNxl(tZ4Erfn%jd`q(D#F7ZdSGYt0!q|b%95DMSFc}l
zi;8BypBWk&8bMB3y}rD@`_rSg^<FC}G5t$?Da<P9;XTe~Y3K|Ql?!}jU;S&Da*SmI
zMK|u@!$WAbwl^Q)^48PY{0kQ^PKIz9AR<K(T12#RGsa3nP}XnU_#ilV4HkOGaAoUi
z?>BQgV@**<U6|-cXQiKTKQYOtnxcjZ&(~0y$VkZ_kv~}JcjaAOTq^w|@&IMw(PJnu
z&MsW7;v?glcbdJ%44rMx1{s&~ZTql*K_W)f#~~6Lly4I5sJqPKTSDqV7wsp7doL^R
z%8TI9Ieen7c1zw>GjFpHT1xyE-lK@CD}R~mTc_sQ`Fzw33?iTitYc$KhjtrRJjEb6
zd<`=ZtNg_bti2EB5m70(F%8<aJ~ubFe#VJQYu2vKvj4mfxo-`Dei=P|YRz0qdSMl!
zmgtyh*H-QCZZS)F=8!6JUx`t!h$t3aRrdp<y}Z04)p>p0lM@rKaX`uuem9P!BH}Bd
zfCB>q!-Fi29!wMq#fS`g<$o&D6zI$$Pf`H$i8>nF7awk`v2Wkm+dB?#>6&JGrkS8>
ziAl&v@}-kUDIHTo(Q^;p-q;#$D0{B2xgflNLL+1uZ?{0x&@dHM_kMi5czymK!v`hS
z8#ivmf^jmRge;KtqWPC%0}#+T+F(}~A?D~bf=R+DQjCDx#JqWd+&b(Qww@xQo&tcA
zCgKbo@uwELH&yS^zD;{voF>QDu<Tvn`_+%)Iz8?{wGqJ>r^bR|`Et#@d!KU#absd}
zVh&nGNyRxjhT4m(+1IUI8w>c<Ni~nIxwd)#%^Nq^+1R4>33;>lR(6Me#+fqh_e(9T
zlukfYcVkmxeO6NRZZ+605Hu^>P!iqN)<)2uq0h_+!8iRmZGrkTsD;)M+B?ucKxCDy
z!2c{1nLQcp<>fDWdU|RZ7#PGB<mBjm{`@&af=fbT`tqcb21X(4*Q~k6<~{y7cwlOB
zk|b;~8P8=2@1{JlsN9F8YC+M7Ew|0);o(t&U|jH(AOBCSG!Cv!H};{<20MLk-5a^v
zVc+W2tLb+Vy`?H5gW)LfU^DoB!OUpGyNZfBlOW0XKb6iLX3<L$6JKvYg!y=O(Q>Ur
zhhhlqZ}A)o@Sf>0R{A$(oz7qUtt@P2a-{m*EuN7^$BHg&x%P#Xn;zRXiI=eYN=Zr8
zKQg&Mbf-vhXGP4ph{6bdT5jZv4WR#e(~Huy8iLNZl*GiGURz3tM$5^g%+^PkDfYVa
zH8?~Q%Z`%tFdB3Kw{BfOQc+%R+7LQ->i*}vLbYR?Jr}5~rCB8t`-g}t%HbCiKj}Zf
zb8vid7*`?fVhY2nx&dGFHlCvtnVm#@ue&1_^@3J!0Z?YxMMxo$QeL_0#qTRB(g%&r
z+nCCHJy7Un))HwR`0Z|`bcdqdv!`qFP7iP`q_suGq^LXtfYWskFfyzeTO5&_WAAm$
zB(h+keLE`4hlLShUK&j=3u-dr4$UTbMo1N&*1lCAA@)G$e#h4%?Tz$f4H4Az{?E-n
z#7`F|wb3R@qZJ6zyl(Njn=iK@F4pH}pvZf34*4pYj-UqhDn3^F58iElLv(gu{$hCM
z;QfqlmWAAj@9dcBM>;ln(gB8s94E&C(rX+mBoqFa2xME;DAqX_z3gfeF-85Ufiep&
zB_(&n$?5TLJM>~O3<=fUu;UE5K^gf2H~YP&CuCWQ1o+gBT#3?l9XsKCf+bWo8?)8`
zW?@A#dP>$#Ayulc@u{X`$NQQ@_1MQ|<;?qjXI@Y8mF_sZ#h7U(bsY;!3Z`Zl&Z=UT
z7dQsdn9Qgm_nc0C`0!f#IfuhG3+X9|=R0=grT6yQkT!-PZu06`j<*dBN0O3~y6e)T
zF*U?IQO5DIUkN*f)|c*!SXn>epTTjwhPehmW|sFcPSOOBwX-Hl1)ZHo%}LDD#z>gp
zIDAz%3SIKZELf}c@B(IX_LQxYPkZ=|K(_*^lpHD()PH#N4D{rU=EkVC!d{b31bU?E
zXK5H2HFK7<)+-6EBNN!x13>;GxwAhM?fW=$mM+9JNC$MYLyoGgqeEJ!bi+VTdQmMh
zf{3AV4Z?K>u~P#@#ux_AQY&&Asn6$r+pyt{V5s_XbIjVumr8&2Ntv=$Bc04`j;&!`
z1AlX`K<j}B460&QM>dCh%<_Ax=*?aqz}#k*fdB{TV}Hl!Z+>F*Mxt)`M%#~FoH{XN
zTWN8+xps_uhx2+66umT>S3U&1d|&$un~RSXRh4|Np=)?XC=b+>>pZM4OGI{AY$I1!
z{d&3N``{8Op>^|&Lx>;TtU2HiCNq@A4P@v3jk!lPG<96Zpd8?H&z>W4fg9DKOHPgr
z@BvI_mY!mIk9AU%$h7siiHUQ=$?W-=;mPc#_)nkoQVsKJd#eom1+nI50%Xvu9xL0G
zme*f3;61hnR4)a@EZTcAi40|ht!ph_U0iJzLhU4@i!$DO{BQ$ohx$jR+sl>-<dF%G
z0A}B~@_9eMPPF04!4kGj0~qJ3))_bk5GsoTkv%7#b62dy(wK(NC|~bqf@z||KCz+x
zxG^*LBcx~>=6dK_mxF@pwnlquw0ZqA<AI*Ci^~(jOzjFMWHCX|R#gpDjh0G<y6W6l
zuT5(6Xn#`*(jZgmq|<+{&ZZYmDx!7Oyt=eD%ehAfnwtU0c^yK0lNXr_=o+~lg#J>A
z-;t?2gmIMJ5uB&`dS$>G918pWS5z9$P3l6ml|_5Hi*a*o(d<Y%v%m@Y0Pm<WpqH&&
zT*;U->OZ!)&hZxg>F+KE>k&3d_s0+*Mk?jl#OiHxySOl-CwlOI?r5-4@m#cm;+usz
zOJYn+YsLVGQ=a=odkDq?+r02o{M6TrZ3QrGjP%s0`Jhvg`4t_jYiqCxkZO`?>21~Y
z8AZTAp>eKj-Q8B;d)tMDYbw9>{f+M!{0PN{Lf!mmp@eA{CgdEyj1bpBj5ZF?(b4h1
zogpSB7P7(@NCa7bYT7_wzy6vr`f_OR`1J+vW?Rd}$E59BeL;0cZ+L-OTURhzujl2K
zn4a@oZMCka#n?LrReZAJ-FfmaL}K>t1y&8*Rb;nigNTNgzr?nzUDCT|v;PCz`edr<
zs;*wiImNhsy;yV62bZT(b2~A95;TARpl*O<nc66wU3YV=&x0wN!YVsfdW-zC917NP
zO2(gc8q_TZgzsMN63ypdp3E48;f}CyR`KE2810L1bwYXao^08p&gOY&Y!(&~afdIj
zOk2opNKN$$A|sS%y}iW^b-uas?b>w+08;#Pe+1kseBqnZ3MZ`~)8A7ad3=&3NvdFR
z%?zXcT`V3z2{KV4G!$76$kNd3_}r9TUB}f%i@vw~a$$7dAUZWa3MbL2TZ|WAT6^c(
z#cfSf<3p+}cEwkj>h*_{^7G$!r{yi~Dtvx!(-g(b%xwEqTaf7n|NhI}NhRhLF7+L`
zv5Q+4HQqb89O8^7MD0M4yrwivy+-;P#)3<R(Hmra6!eR{3u1ksHe)H)!;q?K5aN7$
zK(=KJF;%jLb6|2L-6!?Z+O6GnMYCCjVCI-?y+q|GMo1_8i>mkAt<fKSX%;*_`kc1O
zxz5mipkQpcT(qSVL%$FX^nk9Ugs3Wj+K+el)4hK6$^tiuY5Pa8%t&<s_^Co6{aYcW
zCZ~rgvOLCYOCq148|6aJA!SW$(oOtdAgzyJrqF+S>OL9WOHTD>5<5t3>X8I=z1@=%
zGsD%}`S|MCgCy<&xp&<@Vyh0h4y+{Nj}>ee64EmE96T?3b;DirXfGYow^1U5kH1<%
zk_{vz>j34ex;-hVyb{YlRaB@UECNRn_Q{woyv3hRLR&*%^^;2=gbYrlJdIy78*yg7
z3ZMFkN;cI|Ajld6@R@~Zo5xF|FH&NX1_uWpy~^?qd8zvhVf|k!D?bVL)sp@NuE%_O
zpc3BT0W-HxTRstASTXed6UB9M*ir#GcJu8mTu&&FI7oKy9={wk^hxeHG?GnHo_a7%
zfISHiN~f<6wzLhpg|8g-=>|tT)^E)cWd&}b9z#OyPeoVk2QAgXyfMEZefqLgkmQ0+
zR1oX*f?7y+I{?z(?o?EJ-WY!2sVKwG&#E~0_c(qVptU8_-M{<gnux;!^D~6LwF~#{
zmw*0<PLxMb0oe>nNHLQPe*9<tyBL<kxlT!NV3qN{3^pqdyYDmI#+FrI3X#{kcE{T7
z8G)$xhNA^`MKiVMZ&A`7LDE!u620fNDo|3|&f?>XS8RxBFAYH#tm@TJmZLbaV~2Dh
zD{-ktB%)uu>piuj9YfbTu%h7~rF@WXczCMO6;zVnF@I@m8$tW4vx=7XDTDxOvhQH+
zhe9t8BE_Ic3fIFamWJLHu60Xz`kQD`C7;iP(hO#U4v+ilvy?DW!w~S+rCV-={Y`n+
z@FigGREJqWmSr_prjh_X&4+-E;!g5d{F6Xj3E8n7ZIiEHI2h_v8E^TT+kD1pPaQj~
zUid9dPD>NX+gK{Epil{rH}c6OC<&b%#G-1`>BhUTWLbhJn}u{dBE6KaXj%jMbo$_&
z34+mkdQfgH&yXjq4mDuU>%q^_NUR6Hyavg;0B_l2uPK+z(jklGjJzM=3(9ip(%2;F
zu9e;Fo`N}39K@lNoSaX+KMYEJkKil>zEwoqy(%eTDGwJF?Ju0NH332PFE<m+6AJp;
z-fjvxlm%1!bCr)w<YMhw3T06v)wQ%vxO#Ym8%CAj@B~q0ZC(`@GZR}O|A{xN+Hg;1
zA?=mm&H*5Q8{jWUU(Lj1iQ{~|YsER?fh{88<)H$mmp}LQofNWbZq!2spmHA8z@_;;
zMr~eJR8&|r=Rf7Q`Lv6jho`hR!{Dr-ukCh>ONr_ZYXFa(ot@WIUpBL}cYa#YtozAN
z51;q`kO;)Yj-zX`C%q0{L)b9lz^uB%?F#vy3RqtZ2+$7UHGB+80Hx3h3K7mvG;c?c
z*#>`FTH411_hTryVDX9%5XH-QO9iB6$4!Vg0+Yvdcsx=?TRb)44He1Tp#<sa3Nj)K
zwe!5cKiaeQ=FOX##hnuIy=E?M0TsOqFNfjSuS!pUW{}0)e+7hIld@3$V)(G=U+%h^
zks;95l@KigY$tvrWmtmoJ;liWvvCL9T5UG+>{DN^@Lh$v4<iq2F@_y*Hx#qT!j$UJ
z;x)#!RV_*=Pa>$jug}n=ylCkC9)H-?_%SD`^<?*ek!^*sv9SpRK>sFN(ZG#j$5${3
zyvpfLPPtl>qKh+=<Kg8MA*b%t-$+GAJkNjXt9%>Ak;0%Zv1&kYmB3SLQ=I5&V_2AQ
zQ<eE{j<H}s#cXrqa%}{UO_Hd<fQ?J5gNjOOhBgteGj+tSz*J)R!+tKU6PqB6?uv5z
z#uS^ITVFYf@z4g*CZ`*fnBgFJZRTczSg{KvWtS{p&Q1}@K+R@pcycnOrVyiarG*TF
z!ibBDSFQ-6_nB|R1S8lBGpVb{s@xobx*-7|-l-iN@56UH+~ZOSi-T>$=t<DA4fSyR
z)n{3Eym)877v@E&iLYxREZ;G_T=FT$Ar|%HP)d_+cF%c6qYqb^1eFut<wBN$7o~N)
z=xvCxcM@h#Z{f@M`c6VlTf4@z!;v^mvG<((V*#51=#htWRBxS7589(^4=mwV!;w6(
zYk4oi#C5bOoE>gI+=EGZxbIWZp!MIP!ezo@AarzL5@QxZle4a;d1D0oE>dlI$|V1q
zX|!hW+61qxGwClkIsbGkg;uv!&1dq)ix)4-co*0+@ffTaNJD+sJJwHK9KV!wPN-1v
zpmDfGxwlZCA+*5byFIVTsI34gUmBNePt#NAuyK!zGHk@b98Iom0lDA8td-N=h~q-L
z#umzC?_a`Sf+9(ZFtHh<i)X3M`Fwv)$E9sX|G=XT{^7ct+?vDh_YAx-%Tx%46hx%!
zg30a_h_LtHu6O4pooD8t=b~O&v-c56d(SQAhSGy!Wu4FLP&6~2P!w?!BWsA4g7kNY
z9Wlkr11>n*1;RahvN1>92YzIXRJn<e;xpCI^4OtG7G{YWq%|{XPqSOzQ*y5ZHlt3d
z6;0Npm+`(nML->lAqVS>eF~`e^;Tkok%;=4W&w*=Zq$pNiJqHI=+7IFg3MMH@nQ##
zg~8EhHLQ;E&dym#MROOX+Zg6>q#8V}>J`{=3Cls&Dm8x~RxC{0(34jFytN1#Cw5}A
zLa88L)l8*F5rhhrhVbbkM2M1xfXy$SLb1+JxaB5QTObBWZKa{tNR57>=Tl%74g~-Y
zQm|B`B$6tm=Ni8*lVqH~vwLP{jJ{V0JHdSz!X_cM`PFM;8iJ8kp^9+&eDm`&F=elE
zMN2Su5qAB4l-SL?Yg5BHEFr%^5UB?CF+1fP74XL$z5%Cnp^ibZ9S-uDetEuzlJ-1T
zR1JeCq8ww4C?5<LW+I4(47{6My5A<|!=Ym}<;>hE$Z-l)S#?I%r#NNWc6*FIKwlXK
z4@yFVHVZM~CMaVnENAH}6?kettUmBz9HoPF^w!p+gk7RxYWCEo3VV$9>Q7bO9emht
z7>lkZ?)dF0)l_c{x$8nMRF}<up&esIa|OgvM+{@6sUbXsPxl*;=xU7k@J4U!o9mZL
zTZiFj>w+=XEabS@{I_2jU-l^=E<U~+nDrx~*LkR1R4KS{{L6C5x#^)W3AYR**UyV?
z@C+M3mLuwP|I>~r<~|Zy!{cp1JusefX@8r*<#SNuaWwaReJQU?O6uHSHKT&4B$o~G
zk9|AV&j+bo^aljqo(i8?$ADE;IBKGt1WC{C_E{}*@5WX&^!YR$59sAMSxcY_?6VxX
zdz978(uW*>4<K-8q8HlceD-H1u6`m-Z|C75-vzV(c5t@z{2z(|quiSDokU|nH&MoG
zkNRnzjpQu;5a85PtAcr1@_dmj@jhs=k(l*ScJHf3id%kpbtw$xkTiLa$Sww-@pII6
zHz7nKTGdEfkaU=#KkC^0J*|jgp`sx6=l1X#o^Em`Esld5CP})L8$Tef@6FCkg&W*8
z-{U#{@Hajq0~5Q%5;h;QlL!<l_Do)0{sV#8N1sttB6h2@%LNjUNHsl7`!iQ*W5PH4
zJ-du|e#&)Dxy@?`Elr1<AG4s@RaWouV;KE4^j1l*&c9uP_V*pum=<?e-M<3RkcB9V
z?rp$EGioo&X@*yB<nskKE6fQddowG@e3nojLPue^B)5ni)`|A$rxjt^{7M48Jir<a
z3yVa|$bew(E6sSkDJ@k3=I}$t^)wf3@V}dBUXct-j{mVlm2eTl{)w|kwApn(I<83G
zfuPro;U(RU6-+WW+&nz??7NtqExP&Dd4j#jb^(A-Qm>rF3Onv4KRDw&zzWyxCgbyP
z>07}9A0}iO40VT70sSo*f;=4m$8YeA%wzMZp(Vat?0z{H2=gb(Uzl);o5nPD2JxQm
z)RUSWrgg{scvCnZJeY6{6FVzFrt?_S>C%#t1ledzHhF}EZf?!k&Gd7t3u#5Db2FuV
zIH2jbt6muU1#~=Vu&ut%E{7tfA6o!J87cvb`Q3=C&UWC6d1?&O(+}&Pob;edp;)rc
zg~Tw;r#_<tLTr3|+(5t&=F@$5@80#hCi@kf0n0aZEEg88hf<PC+!>D_e~f!4!G>yE
zG}C8?X^nPD6Ly$LL;(Rc50AZcs?pZHb12drZR};E@c{sJzg2ZC{Kx&`hAO5}GVB9p
zVt{2yjU$FvI9JSWbq<nT?0eF1GbCgyR4ejA)r8pC;2;@Ax%Aa@!ro(QZk_@p!lm1D
z_~)7j*FR**S+$Rpxexmp3<dm#d+I74{E(<We`JRK<stNEm;xbrVGcn62$x!vjgA{@
zfU{lNa&s{fmoYKFD|GO=h2qSJ1xqDL;xGVL5uC2W5c}^Kq?#X_zlm2&GNQZYy(kTK
zMXA`vbe$!aVAjXsVEl1h=Mt+Ojyg=_WEg0IDR>g9fG*HLmiLT1`dX6g)ae$l1a!tO
zR36Am)iHa|Ibq&&BY9Z)XF=P&pmQ-+v|Rzu0C4$*a!5)x^{&~07Kw598bJ8RwZ?u(
z5=YOz=3HlMZC%}P^(eVN@33R);Y`_qqPapsReOA<LE^cz+ewMrAYv0rvJQ_8JR4l{
zAiZc-A11_VaCuIJBTsB9=|vll9={cJ>1X#arpaU4cTvp9t$qE>>{wCVeg?@n(wX3#
z`v62e427WQ($=UC%lYz;x?I9$4!Gl4lgU0ipA@jO9>j8R9GZb8ymymrY;%c#CzjZW
z;P=wPbnoZsW6?!Qr|go=-^f@4rh($XO(}S6>SRlyAtVr8c=1S&ga`1B;k^>%LJA(d
znZrsr70AmnGQMxuAEqPGA4TcZ!DTFgLeK<j(hT<ij$2`1LwX}Ix>T!(b^)-Og>*jB
z8>6(kH<M0_?HTwq`l4mtBXyf&rdzkD3iYXMM|m5}Up+X{;JBW&JmLoq=ICq68w;*T
zJU{f1RAD@&;kW8|6hgp|02a>3mZ-1UMKh`W`9rD!tdjbK>b$LS6RDM?<glKm$-BOM
za~!b~t;)dUzzZ!ZP{pxww~^0`b4ey!wCk6p1R%_3eud;%CpxU~Q}K%8Tslf!8?%h)
z@$3dX2vYNaI$<!|>@m=VVJabs!=14~$=Dq(-Of8G!U-Ui-6)W?q+z>=-F`_v&W`6q
z_Fa5Rn2lqH>OncVaKp>@$t`R&SkTFZhd|=dkHtng&ssM*lx(U;FF<2n*~3aHdoJv2
z`*q)v&pU4R(8DH19^Zg@IRyt_>%N03Io4>RVHy?W%s@#i(S2rn0)l85&_DD`ntd0q
zzz02hA_p`K*}XjF(Qotn0A{>u-_e5)JSg^(w7-8h2GU^%AIdC+hEi$2_G0)vF5uL=
zTy9nks9Xb*08iKMsMTUXS_Kf4xOB}okxvssLLdWy?-{10ppN;>Q}3O-styvEsq)}g
ztNe9HBo+~>br_ES4)tBmBiptKQ^KRie=VQr2Y#!D2x))x=uuXEpGzL>_}23c{O1m?
z-gnWBwA0?+Uj07n+6U?|I;kYS5uXnwn}vO{ynfeEm})Gx`Uw?2ek>HgCT)cCff2Y4
z5CTAmcc*dT{3Lq55tEph--|LqFk7$j&lhkGw~4hFpB?F}+)qSm5=n%<&+is|;evPu
ziM$Oo1GX#y@-|CLOEPDq-lqxhC+BdG0s-dXhC*6@Q-~3=;API;9zVXml}%erq10`r
zgk43*!u=X%c-E9)34?_)H&*W}6ZD>T%~X2017{R^#Tu~B4o((*C^UF#FLrJsdQKfZ
zRBC2ea~>fyKFy@I)4<7w`!*x*=!o)ACHg!@%bW^iC_!|{8z6TqqPAM#UU1_+z!oGU
zcY}#0fd>)84I(>Cnii1fHk+29Rg>5Pb$skSGs#>LOXj>pOd;yV&sh0++DNt<@bDuk
zlF<6VDMasX={bt4^!&OCCU3Rqc2-$-O}W~z&Itd*>#$>13RXzOVuX&02$jF5l&|RH
zW6NqnV*c3bpP^;m-}saa>Y>5G%YzZ()+YC!RMMx3Lixv=pPX@Ehy3ggVEWjGDu9_i
z(VpS=CanX6n2St%l!jdioFd7CgL58eHiV0KR~|=D$*%<6yl=^ai{t?DfGP*^R(VYh
zla*@5dNt>}O+v>Q2`QR>9rq3*p;Q_O2w>JnJlLU?GXFYE8U$?)ru)n&6^*~<v&tg2
z&bi6DBIUou=t3FMePF3jngb*|d5NY1Bg!8U{OeVml}@^mVag0P88U0+AD=ES<E1bf
zF4>HD6RuynIe_Tg%xL1|^hpH_CQnKL*E!4>vy)@}+kv3bWfPJ|#4ZRLaO@C{Gy-9c
zxG5oPRn8|+<Dy<OqOp`m?jeoR(eD}@oBNhXFrx|?T%jzHCp&=phvz!=V!eQL@i3)2
z>?X>|D_}6igR`PPklxmKA*)^#!G@VrvEStfZKO33Z90n$nXM)<=(iNzf?PA`XVaO0
z%auGLrH|5nx{jzT|B&*qcxl7xkTL@u%c}OO$9?9tfBHBRQV&m*5YL#!9vHWyyhaZ<
zxphQx=@2%79@Dt+i!AAAIIT`QpQ&(o&Nw3(He-0Di-c<Sne!s7gou5q+0Dsox2h(h
zLj=bma^Y7~?x{;BPZ1DO{rBQorfp=DhHcet$hJFeb5H<}&k=nD0D6A2aARg@gOj+i
zEV#}A4eBh4ke|xQZ<xuF^l{P$mN*oUmX(VZHPV;e0tlkye+<C5bg&R=-y?DB`JA#p
zpEX}%F&)st2m%s0ahrg1leS16BqYL-3qeq%0T@0btwPLwI8pzJO%mqNR&B)r#BUGZ
z*-^Y}Cn!axS-A%MBf$0Tc=+{*?bCzsyG@qcd7J$0GQSEj9{OCgc(GzIhPY1S-;XJ1
zfFV6b4HZ-xa{W_1NKgvltuOS-<KPBgBK(KwHit9K$dia%x*cFAI0G~}Qk?PVq}&oq
zNL-}j?vrQt014DGGU8Pr71e*`reyG(9^md;tXU=PedbT5wfs|nCmw|+qX)Qch_~6I
z^TXZx0*`EVzeew=KC=)DZfI~?zz0^1*C_KPjl}-O%?ov2<<j{U|0>A;prb=n|F<7i
bpYRbe&pUG_#Xb^RJ4NB3>VdR<N6-F0cz2V$

diff --git a/master/_images/sphx_glr_example_inspect_predictions_002.png b/master/_images/sphx_glr_example_inspect_predictions_002.png
index 668e4847894e1a350ae3d9704709b6cd8d3f854d..0e2f7fe773db978720b2c95d1b1a783a00faef1a 100644
GIT binary patch
literal 301152
zcmbrlbyQSu_%AwigMf5Os&scqN{KWK-7$1Gh=g=Xmkc2>^w2310us{Q4AL+l4R`pS
z^E>z4zwctPX69XQ?7g4-e4b}VfBc|`hfRqM008im-pgqM0LY;L0Ma@pI$}g%a(Nr^
zA>t{o@2TZt<LP7WZVgZ~_jGl1@pQDep!c?R_po<y=HnFL<YA|`^YnD}5ar@>`acsm
zUEFQC=FQJL5u0GSzBljy00_*V|B=2+mDvN3001SqciO(WhYx<Wjtd4Wv5&jCl_NG+
zWl}Tc%FI}o-;q$?4+zr7Z7>|(xt_j!C1GNwcpXW%8-_PO6n?Z$i6cWk%J(TAugM-a
z!S*!3ji+c5s%n`O{*g<kyh6LeWB<a|XQA48?B?-$*C^{QSF1t+@pu3B?U=K>ZM9LB
z$Xw<Ym1W>?z=^?3g)X#02jZYp37mnKUNi&&5MTb!n7<7O4WRdbCRF6xLXkuNuL<J~
zK%}1kz500(%m4yd|39k{J67oeT%Wi8Ut{}zx@>@!|2eKg_fou)1Yq=k9hX0Y4XFO_
zTK-!gG4AmH^Y|3oJ3v0pPVUxbe{2;5@-5B3iNN;xg7`kJ1`KTR*pB7CJK#F*Jaxb`
zoT<nUxxA@1X9Hjwb@(PPxpY4)NX9+i-MCDiR7n1kbNdluV7yWnG>^FUdq`PZBm_WT
z+th?Uq-11j8j_x9^WWAQ;K?tD)tz_uAv|;c-;Iq&tb&4qpuUKk9B;VQw9BfVt<{0-
z-x#WU;oWVLc2)hqHxJoYnu|gDUrz^S@Ee2^s%h5v<8{QW>AmXS^W$MP`Jab5-q6PT
zHcN;UCQj?Rx^=(xln0PUiU<CB!U5dOOC_#xJwE)g>#T9mF&#z7PraxL%C81@q#LN9
z1=q9R6(z#;5%(UQJ;O1jtS|lYHEMDBIMQgx75|u}Vd7O7X&wf$tLG{hP_x~%$|g>G
zf^_0Wzb3{UQv4Fz{8js)Lc|G~8SiBF&(}nF3ZOXMVg+!<CtE-sa1B?S7@}+Dp;Or=
zkI)O^*C!GgT&BD0>+1`WX*xj(e}M2XvW`&pk^CcBgd*loc?E|IAIk|gD3=5FP&3~!
z<K}6+4dgIWMk^$LN4WudiQI(QjMDh$H8l`qe%Hu?C+(8;z_($#&|Ho$jl3>C8vwK|
zH;jPeA6QE?jrmlS+@G`;1f0~-=Ejbe*5(pARsh5qwaQ;I0h7tqibxb>E%?iIGXauQ
zC2OoKkE)iq#T$YO2RZp`2<aMClK=vdNqX0!iCOoq?kWkcJZddo9G9qMv}`4a>DEjl
z=7>0eGq!T~dy-Fgv0q(;|ADU(Cu^-AA_ETJOSM9YdrCT_g`*VT#3~?>YMWuBQ0<9d
z1GWM7vX!n+lB<<z*aW)<)QXNe2aIv%R$ORbSSZsBLV3*5u&TZ(uqEmu<yQDQy3_rz
zdYcQ^x6KlAU!KLLzoezcKX^GS(h=p2mE?btF{Uo8yVujmgLC?Z;|vVopTuss@UoKD
z0=L83aQPVxDZ5S3n_|=NvD2#ecM-Na_y**J@#yXCoqyD4Fa!I(HJIpewkf~Z#H==D
zd*}LW!j3^RE3wc~8uKZJ_c&697LyDub!SbS0Pt~B{wx-bZQenbGo&#@<eVIt_2|#m
z&8dKjFtb?$8oiwK@+M2&SL)!>O(%(K%!EDZYK%p|Vy?@p+K5%Io^siQ{?-yp1JT;=
zSr;42{cCcCB~bi0J2R@U`;VG$B1|kkGFtARPD4s!qJmMrG;&fShZ2SQGY*Mkk7kU~
zxj&M;t!+DN*#j`LOM6b+vwX{|hVIXQBwjqFnV80zXvO^yulc&u^Fz8c=)PS?;trwd
ziCP-)xs$7c9Y%whEgi=PAV}}o9r=fqK(QTH8cc(+m=w4jT+HUS?{ZuM*lHK`k%}N5
zjigl`dCrrsu1Y!w16G1ur008zWkW&%uTM}$d{M7ag?F*croqr$4a=7-A2U`lN~cx;
zVX{k$82={GTJ|%>klg21;Kc#?$S=2oLS^1`ql#ugzgDC8V4E{&P?iK@RG<wFc)-db
zgCY@E_$LuOrIu<sfS$8Jx)o`;kDaRGxBY*OESz0XlNGa!&DF?4`NOnUc#4rfB-Vm&
z(I?6ikw2Cs>?;)1JezW@?!}Eq_y1DIpANds!s8wZc^b+<=&k{3aWY~2VKE+1jw}==
zW7}dTv*m<ZNCYSYd!1e3lj(~ma+gfL-b%XD{kFzIK7x@{gwnve?tCKv1muICvU_9{
zl(~7qgkaGV<RopCajno?uL;3Vzv9O!WO70%eD4a&wS_{ft+&+7umzY1eH}PV*&+nb
zJ^X0fP%1PqCW4N>dAWaBbal>|we0K@MY^2j5=0%LLh1JGpm~qS`oqWWVwQ3ae|Zf+
z$%x}H{Ufar_hTZmPP|6UPxdh_D~0&Be$X{;$iM#yAo;dlXJ%s^b)T@rDmnv@?YVn2
zV5jkCoFkm8J8M3sQe*wx4vYEmF>!l*er|Dv*1g{7rF^v!v4UbMW}dx~12s9blfsHq
zi-xPKH(++7X`KPOH-B?++GjgU6%C+FEwyNXfjfO(bPnj5SafLYXd!p)Y(Rzb{K5cs
zsG&`ei<M`F!NlIV0teMseeN&?3}W?(4=nE#DH6o+%-B~=QS(QnV!artj&jSqosn+)
zp5jwUACd2IxG$waWHkj+$l8h7X+^na4H@V0Tw<4v9plbb{}cOQ1|;PMZNh7T--0?~
zATrwg@}bQis`FLI4y={dWWe5R&_x*=G6SKIskNYL%T|FGUxzY%H^`)uI9mM**k*VN
z#sMfVm-fz}&5y4$-5zI62bZ9;T2OB!T{;x2f%JMkoR53U$Y%1cS<Jho=}r#QVaA4I
z@kS1B21$if*2JjCVygMV>gv54t4TyczZQ7)IRDKQD}2j}r4KM|xCoA^vh|GP>TZtF
zPR&kZYU_itC~2GF&oH!oZ#TWHa`yh;63USCJIT`K0QD<B-S~R>gXbvT<|}i6&eFYn
zi~~HC%9@wocAj|))tij1Y}PrSU1N(X)CJHIcKRp6vC|qcsVaMx##Rnxa=a<52JQ|4
z7G^Q&-3es3w#da2c_lI`=YKn{+nIazIvUe540mWw8x>vVcpPRY`Tq^wSx{K{*Erj7
zwb!^ftH(|OQ<<?LM2#sYmcD+u;lxN;xn-x_ddB?8G5l{lPgZ7Q2DAaCmL`Vg0^ooA
z4w5c*V8z*~sDB|=LP7x)AshXJ1p|7v8qdL%%gz9_^SK$@$^?-xOjTIB06n-tc6pYN
z_vBW;;K2C@x&7poH>nj8Kdx6WBhy2I<=DQHL7Iog?4YC06KK3#xka&pCu9uS+v)3P
zn3WCJ#w(n_c&`Im;_H%E!~<J(v*sM5bt_IlI<9|;ugb%!p}yhP_e3ZO>ZKM`5(4Vu
z4})q70Vu9U=$TFv#@IccG<%87_EiTmzY`s8a!(c)6~9@VK_9}tO5O(JpQC}_r&?{`
z60A=xTQDR<@8*>J^Ek)tQr88Xs&uSA9Q(k*Qn%p1`z^Xb@;<87Q}p25vXJb6<JPE}
z$Nie!c$YxNlR%hQ`;uw9cRZi%*lGS%N0d<KUo~x=euAe{0(imWRLJJ~dT2)X{Z0ow
z?_~R+>-Ye0ef9LX2Jc+E>%1R)x~9ANMtA#4^5L&KeEl<l`_ZDn<A&6IRN$U0U6DPd
z$-}-Loa?c_dtIQ$Pwh(oYQ=|c?O|o@cfrF#fg9o%oa<?)dz-3jgX;SC+P&K9d3rOz
z_3_@X>*=n$;CZGz=r`5cwZeaw-hSM6Kp`Gm>0dqF(RBFR;a`~m)C(6L9&9te$83}{
zE$-{|1IHVkI)b~htu9htCOMmNv3h5Ay544-{n9<d(lA}XVWc#A)-Tsf@d6}-)HD<$
z{>jp~D`}tV9SkOH;}+vE8{0TC*{&l$2(mQ3ewsC13%%o3zOZI#`>q3^gEYR!s@{mO
z-<a*;!J&lmC!hYsWGG1NIg8dEw2)K-z!jwZEMo(+COb$lNgC9yyDoj#BJYZ1lW0!c
z-|_Mc;pRNp+GO@(10*BPS7SqO4ZfQ=<WE=oiRPULtz89moPVvc;>|i`lsab&ylV~-
zqkBA!@;fkG8%}(>PK1v?-AJ7r!{<UC=0rSpaa{tpsF1G3LSQGSztn?MOz)tkTccTg
z_q#5SENd5PYd;4i|M47mUYP)F8YrISeAE2&Z>_$ep%0PbT#wpH_ARYFdI#^?2BV)y
zocPbRUn5r9(E6?;&Cbr&&LRZS5g~Tm^7*RsBq(ms@>w8HLv=!GeK~9QHmj$RLZ?w7
z*HKJLKN27J63^8TM+RPi>5SFxeQC94fd9(y0;aO-p%<Ec{>inI>cY9u+_fAx?n;Ub
z4zR&*^M!M-E9yb<s0-uqe*g91%s#&*5jk-F+ep5V@d()7g^>!oVp*LQcFAYBXY|pR
zNotn8;xtVw;I~x6&(Iy!Rw@b-NDnw&m7v+?uFbpQ^Kk32%*vuX^{9JLb-GJ3OCnB>
zMP+?c0!wY2*MZDkq8YaDp`gcfaKD_C`D^AaF5dJ>=~|<@AU{S|et$opl51gg&z4rj
z(3k|xjx*}l;N)+Ab3+u`$=CqbFCA9(O>P;x;Z@@QSfBpp{IylTx4t_Fd8nbgc_a(o
zAQKM07cRKpX1z%`xyJ9h3h2&2n4Q!<VC~t=&hyr8@?Mt+`y5kuU&nS2wR*sW5z+&M
zoSZyJ{AZ-Gi1GV|kbmO<BnPRx2C0!slgo-@x5q!Gw<?OXS8Mk<A$_~MEKge@_XJ%#
zhFuW|%Uj#}7Dc;>kPKJZhv==pPuEh91PB}I`gWIN`gCbJC^Fdn(EQYd^8LRv5vxyv
zP6BQQT<!;018H?H`!(Xd4#yGpgE!U2tj<r>nf5z)*cV@KfP}y#k2`;kddv_XCV&A)
zhnZ(m2?o4jyX@VXYjGqWH;mJ@SU-Ig`E&4n(mb-EGfbf~?Rc?A_p>)PSx;PDjIQCY
z_}IT;`@3<%%|FBCGD6$F&CKdZ7b^CYn#6e+0?G|}B9sO3I?)P8kmvb;D=9~8Il9s-
zFo4-vSGr+`UlGNw(iPm$X}a@jdLvlu)K0?5x1;ScPszNs=(7Np8>4Z6=9|e$148wI
z)1v{-p8j}}%Tz)pPLu)Pr`S5*N`qn3J^!|)E%nm>oZz!I2X89|hd(Q^QFky6nf5Gf
zEN^RvD@0`NI(}`a=6<;5bU>&pIf^bgQ|cyFYQyyDAOx@H*><X+9eeUzqV5GcIy6}^
zDb^G<fjex8%e%v0**dmpPS&5FzFv(<KXAlRaBI2M1F#(O^spR!cSLiIavf819m9Ba
zmYf(EjzJfg1cgEwrMN~?qG<j1;NpCh{+!w%8Vq>kgZ=Mf2c>wM&Wu`Ja!qgx?^}^>
zF-s%CxZ%jY*4j^A2fv7Z+emHvgCI>+n7inD1Zhq-<&%iqw;nv*56N@JztyPs@$Xt!
z)=pyT?X}x0m9~mznaXAtD5m@~{gFKWj~i#Tj33&fPr`EMX9Y7{usNdxTD!pyMRGYT
z^;x7`xh9_d8Vh;LXw`p7fNe}eumW2>o2J`rEy27J^$!d5kBSyD=W-3YOp4MbHrw7~
zH;jO~x`Ya-Y9-ot(9!Rsqn$fb^dvuLWcc;G)4m|GHml<nB=jKOk&(T}WKP#4Ta!q`
zTU>87U>AC@m`!B6>EhQdo#EnR6_;nQz2?;&TeB4tA2^0!0VE^#OCr@Dr-Y<Hi4Ru$
zhTWHjt?)6io120szp-4A&mIZSL1@|wug7rhR<{cl&=pm1lU1;x7jmiR*Ux6s%Vu(h
z4PZ&d8X%Xv?j^7oogAdWkBYV2zjR)F)%9@XV(a6RQSkV;V0TRHglXL4*@1Z`-pM&2
z{O#ggSGwuVSJT$NK7)x;e4OPvLFOYO<L>}&TlHAg&1p~uW_4V>0cJe~=2T_CT5pxP
zizZ*CVcef8vdAeak$9cM4{tdSGa9`UG`U@X)rl(kne~S^#->-B#MEgVS+m9mFPTcg
zm+4K(1<()r<rQAjv%DboC~G$9pXd;}Iy^Q!OPJMlrVNISI&yq_?J_rwtOZZFno1*!
zI$rr%F))?lv!IIQRG6@2lR$fJThib3oc@P0bI5*?{Fkj2ix&;r)AYB(=fAm}%f2=`
zxHzhVpa3yZKp}frtK1ZUnKQ+DKlDtO-dWd>a><kaIJ&94L5^?e_f}K+gFBTtpbhhI
zbNo=V{iY6JWreBL;QKgo{>d)EOWIPxdWE+2#IC-jP88*XFTT*2T!OT_h?!r%LhuHH
zsVR_oQ<ASu^>3kKWySco`zbj15f;(~I(wX?dzegK3S4Rqq6#^1dEofZ%~f{aR(2dY
zxf}#v?U+99=!07QFCDI~)EghK6Q44l^MRe5?wcGW9uI_}cHTY2n%-NC6-bSLWz!S(
zJ7)sWUH4!lEDwTA;FQ4+b0K-etRHH-x*m&ZLv+UD-N*Pi4gc8)M8VO~94|D{TVJX3
zBjV4vQ2}oa$Cl<OO?yN3WEI-as@j$Dc?v7)QOas{N=_PGI~3pfJZ&S%Tn{gPTjz+a
zg}iYJaao3auG!mPYn=u5PD*hsd@23u6!KC67ubFDYUD4mK2Luv#(EGpXI1(&70#}4
zp&37k-l6zcn867N+sM^SaCNc?XMz#jPYs0&vmv#qf$Vc_#cNQ&=&S7$+4ygNj&U=%
z_Ju2vEs8Wwz7AhFf3Y}~&9qO5a%cU+3fp5F=2^&#Q&ko!_-e@W;~oc#np|iAf62!>
zRLwL6s%}iR58Iuz4M+avcDQ_CC1QClmBhx<3^-N>-dl(<cVgmPX@Vl@LN{7VjP!@8
zac7$YXH>lr=cR-*+#Y8GLyAv35b0H3c&jh_3X)?Iy!>lN;^xD+Ot5MD1-bgP@P#kQ
z4zls|Nb+C0S6eCI-rY07{m6p9weKb!HujF3Yh1?4QArcMwuSTk%<h&Ic`E&f5w5L&
zYz+TtckQ0)LxJRPM1Z?Sq=erXn<;S7J9ao+Hs~I}AwSC`FW+>*#kw2rR$@&q7*=m*
zo6nn?JL7#i4~zqEGD1Y=s*SfKPW%X-z3fe8$YqV>uH{<5N$33^t_Px;fRzLCc!m3i
z?x(B2m;Ng!-H2fK={c2c?tVDBM`)1n)>)<x!Y}u1em<RT?)qAA9bK@#)OECUUG{WQ
z;J)P4d^W_GcuLiMMs?pqa1Ur+@m?Z3_k`@@JTrqjBf(#6&U*ZlxkriJ!3za*xoZU)
z0_jRK?%a@8!SeY60b$tRo$TKXmL9B@;x6-fvqp4!yh*EeHeP*S-THSPu%b#z+zJ~l
zl2~$&SR1RYIgCGb4_2Nv+F0~jx?nq)boDswN#}(fhp|A6XmGKRnPLE*=h{;vxSj0Z
zGVCJdf6xoz!sATds;*6BPGB3O@_}Ch173;(hIe8+sY`GY7XJV%bi1l4RcLy?;8Yt@
z%{L8;mU>ZTO126ma_8au#bKsxwpr?rPF=Exg~s!eZ}pMa%Eu{=Cbtv51}L&v?;SJw
z2l}pg?`AO{6>$efvSUGf7-5a?apf!CXv!wXtU@uR)KhW{2a_KRZ8jK^Du5$ZCU$6B
zCtEiL)!!dFl1oO2)TH>33){U<ezO}7a*M9j70&wYLOcK3HY3-c8IhGA7Xt|G-XtKm
z_>Q^U8(+aogl;z%DiVHNm~ixTno5#M+UbyaBp9+vCEhLw-w8XIfQjeZ(e8kTbN2jX
z0DmM`2)v5qmjlsIh2#^0Dg<@^0|ky>1rl(D+>H30#hQ+G-)B6f)ZDezh`F~ncU?jP
z;X)yNF8)(F&lFw4{V`d{8QBfO_h|eF9&pGwT@UAZy7nu(Vq9+0U2bv{A5I=uy0=^&
zAl?1XZp69!!TWJR>dyAQv-|3*ZH>zEy5{M!`+Nm~J31aOYhcIiC%de{o2=KfE)Tjf
zMS#M!6F5f5DaQRe!6SXhQO$$EGo?J9zw-=YIxJ8NWW6fcyPymk)Be4^9UJ>Co5?E_
z@KS4WmzF^Nda=rKD#dilChR^ZLo}U(Rzc32L(dqQ<S3l=-m2}*-rR%Kn2N7;8qpfK
zv1#v0Gj$RSrrfO^CVN^O=8LXc1|I^e&osY$>fTt>r<u?)Cqj<J`F_QF(}%2fy_M`D
z#!bQZrvn3-FA8j_x7n}HW`NTe=ICdk2%xZMz_sQQKn~?a60<iuVu($%lTWd8hbT3^
zQY$8*eiBqitNoePFezYUplHeAb@UoVR;1D}d)9|+7t`!5$c;nT*4#e0#&q8Jg}O`J
z%?odB@MlR49LoyS`(H<Brxhh0$T+0GUdQtiei?+7zlXl7{Js{a#ro2oVRToffUGe`
zA%e6$Q>XqkB;SrVbwZ_YjRR@%`$8<f4pnFz6er%y>hB>Q{cauqXE$CID_S@rHjNTu
zz<KB666uh%=b8w_NazgjCbBX_XES$B&-!zBg##8uZ)CBoyoKZ_<+n7a{rnw3WKQ?w
zkIM>?{9*cYY6EV@+<v|iVn@t4uC1uNTkv^WagTTZhVKlek8N8YXS#3o?tiM^qdwEq
zIyL?zQgA*gkJNYcx>Nl3^NA1SPY4K$!D-ae^Bja$Z^nH7!#C>JX9{3KDk>^!TG~}q
z_Tl>NjVQWZG%nwF7^=FL7t?f4XLN8V0*HxUuYX?l7?)7LJ<+R~;nQj!-dx)IRV2V^
zh;Jh><w{8No2@#*p*l{*2FffD19>+<uV$CGv)E{(*(3~8B9={lqmUY@_^-@}_IFmg
zd+)nwT}KXe?qMDXjrG<?U@B`2`^+?yJFFWn;j+E`Z{4k65O%v9Qa6(t#Od=e;UWPU
z1OrgW8J`>)ti&;&KwNPqsJJT;6NLV*Lg?SGduspotLo!L|FHw3ZZkzPHF{`q#2J?o
zr)k*jaEl_HOe%N*ritUYzNr~7DTLbA+a+FU%5u%%o`;JI<Oxm&*j3@thfVByk5Z{D
z2v?Mnc-MRBH`{@l$9y<;u2rF$NJ*ISOg-*MZ#&88w~aEXDqoI`M7{@3xF=i^0gT9f
zUq)lUnu#7jU9nD<+lo9knk5Y9EVX0LlmL^_bE=Ob%SDEWy(lkgwVAZQK(ZO(+_45L
zS)nk_caSjN$mJB{d?ayyozb)69-ed=t53Q{QG~pXZ|V~VOL^oG@W-lUBb1I5s<S4P
zSrH9kIyG{OJsdO}QzAy>F@pXK&-I>h5%kk%5Pj2iawi#l3=2+pMgiMp!TV*t@SL^L
zhpV-51Vnajdk|Ya!w*KprmTXen*ta!mmgL)B7*fbhuk&8rJi6<+aulgBLxw*wH2&D
zMC#K0%T;)LHQ;2Y7lUs5+0~591^%8BXFhGND61JBh@r3qYC+Q(1RO85K0wI+4CKui
z(R|g1QGCUl8)POUP#&<3osBsP9xjhZ21(q(v)Q7Up_j!#9(LT%)2_}T1K@ToZy1LY
z$jt=x(V*z>6~8Z1P?S!r#z&^?yyXdhR};=~jwH&{O7cwGPVot3Co9UMa~IkIkN0^Q
zO(3GI!53;%Pyor8L9UMWDt0A|tVN?9Uq;2}4Jk`lAN%A5k-ZjK7GY;0p-tU1sioXg
z8I+XO%`DIn0HK}XQ6uc(Kq=tYu>t}&c`0NX@)3BtERtQos!nIW_YIxGJmI{6`cjN;
zy_4}^sNWucNL>luBq5FMGV-d|&6in&VPGbg_Jn|c6#Ol@8EKCqunbwma^9U3*uhgQ
zs{Yb1C57oNU^`-IiK&fmOd8M-jCJMl@;!^9bcx9xw^oqK`Wwf>xTN;Up89+I*speK
zw19*74?FyOFaFl%oXmIa)g~+?jO<b$^g@r{>-#kS^~E30>C%ya0oK(Yx7E*uo_5!^
z+q(|h0oSbscOt<D&B3vz*Ab>IO5NzED%}?<_ooC;uC{{LOCfhIa0JtF*KuHcvY8Qb
zlTk2fX*!cydFDGx@VHLke!MJ_KqEHrtU&N;MDnsASp^H4foz%SHV9Q_&7_M@Iuh`H
zsD_O?j67EDZX9S$p7C?Er)I+g`2LmR(X%8HgG2@+MZm3g_nfq8g-n$$0ZDFN+Gp!C
zJQ<424Hf-2^Y||{0&V0RO$F2YW>d)#8K+<Rgy08G<gwJ({Z;lNRdORX9HW(#zJchj
zI^!!2p#xD|3Dfa_a~tlg7see(f#Ep@1cImn>Tl7ydfAY~cwpRkrFiybZ~YgOVA&?w
zDU|B@Oxu2ob*4&1O4_wCb`yIO7HRZ*v*^DqR_dj`O@%^DeiSk8#Y$^pG<(KvM>GCR
z>7le9K&Y9xS~yP;_?u{1y(V_<Ad=S*rlSlM)S-BXqcr||1}?{}?%_o2WDqUrn=UQ5
zgF6=KH-#};5?NoL4w>^|UxUUc1D)iOnWU(u;ADsMpPp?4;sP{#2@nes)G`f3MS}T<
z_-=G*1`C2T@|HRX^|tzG*|h0CwpryYWh*{j4{pz(m{n^dKe=Mwmkj=Z)YDddvyLG(
zBz2APRCt#)n_csCTLXKRxATpVj?T*T@xi6PP^+=aeNIJ1+leB066UvE>Hqc;k#`<_
zIq=mP|FH5l)&-R0X2DVBNR7Y?idVZtSbv27*^<Ei?HW;JG|uvk5|24w%+qeG4o8Ag
zvn#<Ow1mzsS6U7Se%H6*1QNU(W~<(iWH}rZOF)`YRmJR|#&B=956Q3J;^>e!I*T_#
zh|jt`Myc!<n9%FBlXGCwKlZ1LDg_SfD$!0qHg8}E-kT{MeNT!trOmBHUDq^_w`Tq^
zz^155qWun5R-_^ShR6w77Tqs9iE8Li8iUToRJchx_Ykpv`QC}YfsBQxZw)rv_arRV
zw|l><uF{}o{XPn$2mSAi1<r4%GMr#-@n>)IznkM6StT`98*ry0+u-^Yk#9$u<foMU
z*ee-G3U3ls{uY*1+QeOBSTU>7r2AOIp~ILxWb>=%mD6FwfllsBO8J*$_zJ@SBZ#=P
zQNJewhWY8Y?Y9-J32RJ;FmK+DIMGbIKRuK-8@V<tji>RXJy|jMjRnlOyt2v6G<$qP
z^`A{qt5(SB^R)n{x0eU{pmzNveOC4S?XJ7y_MNqko8K<GtRd&Dep5NCKP9e~`vO+s
zw82|K!6+cQ0PUs3hxx=a^~YaQH(|7Qh%#)y=IjmUM-iJdr%S#BJP$$tH|<l|BN*`G
zO+HSlv7@mTUwvK(`@^WKO~asyI>b#MSRJn-8GQ5N1hvuWA8!Jl7U%Kro9Pl+&ZCSp
z2%ZzKoA5tbl_8*_^*7HSKQ0>5whpj-?WjkW(o56+=@47U0@=u9==k+JU-}zhukx^K
zrd)_Xt0Xv}z&^?Dm~9I!DN?Xqgp%Av>>q{nQMnT4wU1MgsxtBayj(BSH%v+RliDn6
zM8UF-Wh}^)0S0eHbuIUHHcyY3Uoo%<6tgg^XRvr=MhJ2#EWPG9%(~2PBaHG~XfZRQ
zkE_H1RJ$FH8*-}en|aqYIum*@+QbaSWsm*MqkNCj0%X(lCYn?%jsA&2lEorq!5<wN
zXW*2jm#$=4PB~KLoHa^gZLGOOdhpH5EDvYQeMaP)d@9!JYxg|Q(qwUCE4qrp&oA-v
z2V^m{x2Cg%VWOh#0z9AH7ohAl$+|o*jlDkY{E1P}U(=Sh&S#I4m$z^p=DoKt$LZeF
z8~%?<ae4gR{Y%B<oG4B6Io%1I^9dsBKE>%i%?yDPJoZYh-yuTmv*8gouLa-10&`)7
zdNw*j)B6HIZO<`|W|`8j5-I)75<>|3{*`IX=4LN7<mFV`#*g0Is2VkqOo8p_-D!p1
zkq;2QYD&&{W!etjBJZTvlV>Z?6<b|H$sp^|J1LL6zF~{JezW&P{oLx4z(qx`DP9m-
z2eSfy*n`pMKl~6|pg~mzj^4DK0js0z(r`bgVH^EFc(~qL*RQ{qxlnOyP?FSA*^_-u
zS$!l&hPhIDbA_QU(~G)|r$Y8PTTYWrye7*P1{_j>dzE}LBucxqbX|`k=20Ls_K^Tf
zveH}S=m24%<QeRZKhdS{N#s+KY*|#T2^C$*B8ucm&3|~NX1dIe5NW1*vBYO;kZ-A^
zGinx%%!#@w7{baORIDX*mNaLJ-sK7<Xe|5@r3&5tF}m@&U+3GY@~R83Uw<LAYzX&&
z<%GPQU8h*N$5^~MzxJIYGjC~Qy3$Id5*w!-997(VA@1GQeE*EyBoV7Zs?X7~Sqrz@
zrmES)4?CHePar5Jm)<80JSwCYwe!NiphiEtQ5Qa$gkF#X^EoMOVwh3X@|9J@n)({#
zq$~mFB4~g_8nH;8NH)D{>YEz`|B?`gYgmnF9e_WRqb2Mr-abCh6$q&dS*f2ZK5Gw|
ziK#FVzsH}PMD}lw2NoQ?nFgJ8#yexQ6r6K`=lw<++a1(D+Qh~9vC~~}&AuAnX*4e=
zaG8VO3ug*Zk5W~~v!#VqK#fE)&nAxJKw#{&&23<B><7GFKU#^`t*#%>5y4R*T5@-E
zHgr^&wp9=|F4aEr)lF??-qnromC28!*J@6N!%ZdSRug_w0@PH+R#^jn@>9UluV80N
zp!jN->?dM$SyF?^caST*&!eGXTL);SeslP1w-r<OPm=D@J|UY#8UJ8`QzPT9<K<Ap
zf(fo<e_qf$q~g8#Dm(q}iu}O-#O34<9@i1ri?YN$1s*9PU+R6E0Wl*0Lavmuc^REy
zF6S*v&$#8<oZKoou1p6SRf-?E)<E)~;&KC+r&CPWo_1(#Cp;?2yoA?sNkl8{lem9W
z1gJV0LKWSpKum|3oPzSe1xfOcixb06vM*XZ!Y>`nckS}P*-;d~Hf;Yal7f}Rr#QtU
zHIav6Mw2iyygg@a%prUVso{GhFd7|3zhvu!a>Ef%7N?`OuUif%W|zU=gyN<?ptj(=
zpB5dMAKKNULIR;j@a8XP;;-+NQ|FA1>S{x*t24Y(dAbx#Cdr!{Oo{_XD#yf~PL_83
z5%HCM#P<kfI`Zs}|0~RGI{J?XtoE(lAgWFtk4_pdb0q@OOYN_p;qNKU+8K>$k|pmz
zZENeVnMysRU!`{z-+T(x5#3meP|(F)P|CK*WfS_|CSE#5uaGyz=&y0-ZSHpAC`LXx
zlF#h@b|gX&VRe0Jc~4GhjYVlls!A%3DC>dv94kpdtLO#-AlF|~Q<Zj{UW;-`bPn8f
zroivC#*RLGM3Qs9<TI(Ag1PFRovBzp?oH`Y%#p&VfzhirkQbAUvoRI+nc8MfvJSX8
zopI)`D8CEtrI%XwHF`D1Ti!6vL>@w#@OZe&5iES&r24{sNyqiHs7p&g&V#|oAQ$Z^
zLCBV6;O#Kxfth6z_h`4LAV6KuT474xhe@|@(8bx8IOb09XTvp>U#-qcg@xq;%~x7)
zQ>wk5$S_XENds1PY8+DbZ2c5D75+HvEdR&^ofwb#eTBw4;_gfnqJ{GBHgUMt)Znu8
zO_NO&hd}P)P~BZj+3zoCimeES`M;90rP{|&68%jhN6IbdVZyR7Q^EroF!*fiV49Xr
z-9dwUiz&S>MA&;xdce?S-ZrnCSzZXoWye>0zE>%Vi%Xo#8RE(0cUo4Fm%)qn$PMf6
zy5G5%{0Z;b-6Zqtq5P{Fz(l9T2~kt(GMr=yo+tIO4At+wm1tDd`uzzd+U~f6QR7A!
zTOvLs_s@W7_3>Xbylt*C0a4(nS$_sinIFpX*sh&i0;ry;kip=?L4fbt-RfWj4w>&|
zrRi_P=k;@;cq?}cBX}L7%WUxF!XPHH-sLhoM8Id+ZEy;GA_L|yAo%kfcig)4SocE%
z{*YI@M<1}d-8gu(EQY==fSPqT>K7h|Gac#g>UeM;GTFy(IDUp>DlJF0HR>YGLblq1
zRb*IKx#Z_5TC&yD!@^BcetVRp2J>?PZ@46<27UX>Y7(yd)J?Z;i;J8&?9@sv@iQBX
z(o$xR9#y4nl3#zQ-nroYJWELi6o|6!t8%oj!tt|qCv14N`mNm1Xnw;5AKQ6pOr&Xc
zYumrqH1G_|dF}7n5?$wL$m2R#+|yd-Kzyp+I<v$PPjl?#Yol{NNW#Rc!3d>L%QgJr
z3+~q|&o~`aB!v;ESai=OrltWP{r##Z+{bcR!FQBu#%H{2zXxvFp+ge;hFuv0u!=35
zhA3*12F`c`)ER}70580_bJ=U6-g=5cH5tNiLu|D%w`X~=<k`@M-)&6{b2cWt7yQPP
zt=r1U_JTwa*}o$L<1Dmq!wB>2ObB2>lKFv#m+@=3pREEKx(iPcrwz$4(@Xi}V)8A~
z(67uy2lcu`PFT`HlgQbO?OEZ27ZW=#7Lzbu`&%jM#A>xUa6rgdJ#h<lH-c!!?QoZ7
zKcVZymd#4oCjMZYd}raU^{U|Gm)PM?E}Q6ijLCxPKF_dg?V2j&Gia{*?75n4)D@Ve
z*L^$Fakp}E9QO)Qg2NdytzcJKJxlO8%^DAk+;ANd3V{p3pP`tevvY87g3_D*zf)x*
zZkwoT9$K8e75R)`trwE724EGKtLZR%L7_$+7lF(m9zM@xq?wUoH1l-M{KT2+yvU(c
z@YO+kU_*=S#Rm3%m?W!%5mUx6Zw+u@<H^qnfzS-B%d3vUY<n@viW;reUA6;{ItRal
zMuPNM1267howxus8Sz`i5(yR%Jq2{*HxrCe_m8YLs-Rd86Vf&WIi(dt|L8sIt}lYW
zL|O9k!0!i66Q3t#DII+(sX{}Hfpy*LFDn!d3VUV3Mp=%guk$es>Mb6X<FVsa4Z2p#
zO*31S$Ez%pF+L%Oyi%6|SA5k#8?TY_OXL#?_`n~Pmj$hY@@&rjE4p4D0!Y5^Ke}WJ
z6Al#FL(aw`Dszu{VMbp7y*K!mB0eZRIAxMeDD9=zX2zKQh%P%V;}DOt<`!nb6oB*!
zQRP5J@WG-7D5=E`8RIkdCxe)7{+MtyMlwd#IIICw*j^+HqXO0eiQ4qE0o$uznDo(g
zyl&t(^xTa#Q0&n6UV|j>MdJPx22E^5w2mqbt>HLnAo=4kMs5rLa+LwmKwZwD*a>Db
z)#hdwX^I%NIjFmwIFKVbwt8)&#qvldo^Qq~LiOsRY@%^DsX{rLM0#2mYRye$(F{7u
zVLw0r8fcwQNxupKG0zURk^^XpT<^osoFO<oBKRjiTQ)x`hSfq+e*25(+ZN@grb_4k
zz17UdJPoYc`SjQARb@u#a{Rk|S5N61hEVPBHvhU1Lx3=AVTD&hThO65WY1fB&~vZk
zz}nGExLf#mzU@13u)~=mBi&$yyOJZ77g^L#Rh_zw=_u`;v;GM)ZhiPAgY`#pbo-de
z`WNwRBdd)z{4QDAyeUpbf_9MCapnO@?iP?9{}EQsgzM>k!T>&sE?VMysAe%Gx&ijE
zW;5?Wa~&_SSA)y^Yt1y?jhm6;B#jl#*?1#dRm=>XP4-4kRdr09Ql9JvDv7T1W2Z6a
z%QIT%wPSe9lAy1r%h*ZaJuh9SYft~@6!U5dD=aQy!R=2VOHS%h#E;a9(rOyrH>>bY
z)4^Ym!MTIrH$le*haNXg1ve<w)am@@;<_NnUJ`rD{INCC@Fsa>i^ik_lDM{<i!vC`
ztpCkH5a7k4^hCrj+hO#7JA0OLM0Jd=GbvdNkFOuudCfn4i=*k^xpP<HIN-YU_8KIW
z@p0C6<niG!ZKLiDb{<M)QqBG1n6>xx=jTUq`RwR~Y6lMMgYQ3B8XRe$4ex3cF^AJ>
zo8abK6+b|q2|xw=qBO$+zpiEZdt;lNg>`8ZCzWFPo60<BK0#F3mEzLvd9fD?uZQ+H
zn&U|n0{Gp4aiRStZyVA}A~-+nlAF-P3llnw_(MLkX!N)Rj{hyO9E*E}S0`3ClIqYp
zPzVJq^qh-u9oVA&7JVOtTRS;^eVZX!XV7Eno+_;ieG4NUHCjP0tw!-WW|{gN#`%-4
zv~KWUMJ;6t#|%#(grIO(L-1ev(V}m%#z^Jg6yG^JFVs;a<AX_ml@(tWoxVngEbVmr
zg_zF}x3rBqr}?9-7d#W=xS$BDsFm4q{ezN=dO3c{*8@7gAB(NEXH?ln8XZ~#^(W{3
z0OjY)7F=ArR?)kP3@pdXfBA-RVOKVRd&xnZw4+(g_%P#)-GY)xajlC6)D!BhKkh-g
z8@Z@;&IG)-<ybs;=!r1bS?=<PUs^y~J{%F-I(Zt?W|r`;PI&ICuH*PM?I4^J5~$a*
zmt)V7p~ULbDd)-kk^0%`C55%+r1IGxK2Yw$;n@(BBINdz6MOvZUBaYm5c>&n$CpmM
z%hrQWmBbYas~tc_3Ri_>=5SZb86<zI1z3r9COm&O5;Fm?B2H>8n}cglFw8uI_Rm>F
zMP19z{E<||t1ah<jfzzr^vw&Ie8h&woui4^$&pRbQr<WC>_=P*Ed50So6#Jjbeujs
zHcvy`{XTe7#@i&{2j?{ob2=fZl(XnO^Sb)%1j8h5toEoSi^EeGSl`dz@5ipRXH@5?
z=Rs<R$)|;Vg$^v?#!s)<KntVky@9Xdz-4&gi|`e#>lYL=oz)IX1QdcZl8x?Q(m$`&
zN?)m^_Ppj)K&2!B`=;$NRz}Va2WZc?M>xJht{-|kAl^&<BM%*{MrkyQ3aCW%XmaSV
zfvT@!hbk&ASN2*l=|9sMdFHZ&8ljYHQd4cKlLLag5dF24{)-1J>(>2#GvPHZ)1&tC
zgTwt+|K2&zPsBD?Ua}*}wO;E%EP-^L*K5MflZyGTYKaYs+9#FkEZ*0R4&XZJ!7|f1
zz<zzj4p}-J;QA6rj)p}IXf^m`OTX|POQExVaqj$(pE*xqLCLE@WgZ9+AI7m&Z>{<3
zDGDA5+RAQ&>T}_I!K=dta-lAvzVpm~dz<x2osXCCxWVKtF}7+UO_{mT(t{rB{aIO!
zt1!r3f}FX1LYu9gDVe=y8qs{f0=48N%WzGmNT{)X^=a)rw+Bx9RE&YKHzM>$_f4=7
z@(1jcSM@hB{Lr<#HTW?y=6IUXeJIQ-6ktT7<GVicp=P54)_G;R16yq${HhVJo4?hU
z%m1duvU0H4!T01ruiS8)W=Vn<-0I4`b^HCRv0%Ouo{;RV&Wip6dv6=P?HQ`B?pRKZ
z<Gq_+oopGS*;I_&T_NApr=93*)&*Sq5<3(|R8Q3x*8n1tbyRKSuV1q!G^9u=WWkn)
zk&PEj-D<1NN(^NF6qJOr;Sf91t=9r<p9VO|@dM?^>S&J0dffbQ&&V1+oV~{pNqYsb
z>@)*%n9IaO{P7=v?%a-gJ30pbPJLmlRg5KH`U{ub$%+KH?;g(S`K2&<{-yp9PDjSo
zIK9d|g&Yef*49qkp2}x)+0xALxp5`wz?Jv!??-{zi(aWZl(5%SiEKGz9Ea(iXd0mJ
zGKvfuv8n^?$!>1)=}8pt8++f)<>#g2fhQuJA7o%;6kR{<ZWk;}Jd>G=6$CF2I{6)|
zRbjPb#OBzR)y1`OF>PL>+35^^yQ+@QZKkX`t9SAN0=4cV^oI&Zxfb7YVmamz4E$6v
zG3(%pX2&=tE~|BbirIWbD8q7?=1`%tldDOnnK*h@r9H}vm#5n=6x>V<6m<GFs%*3k
z1`tT5eiTHp&N;?EL*v24^T*~l(QLpQ+vYn&WMkkVXKedw&vC`aDPJZCqShN^o~h>c
z<UIV6BW;i=t>p<W{K3F8l!^}^D`!vhvuTS)@>+7|ZjRecUs38mM%1K~4}5i-%|}y7
zbpKSknv>cV@WxONrH01`%M?jOmF(0yEO0GNU~5)gxx0fYY_R-FDZX+bPJwzc0WDsk
z`!O83a<$u(h=27MMcGqH4LA8F(y4JOz_q-=b}Xr|B+8j+rF&ZGytcl49A|phBO>YL
z_OrLcJd+gx^g{E@BQ;`Fz$u9)!rQ|ex#B#5Go6Nr&QHm}$E3V2a}F))H1715^#51{
zxLW)iuPf5`FZ>mXwyN~)3}f|vw8*a$O9j%9<HQ*Tr9!=U{Pa*n4H}s&5A?hAgdxRF
zWZ!hm<O9nDh;{qF%Uga#Yhn%iNWe=2M$!}M&=cM{b^cZs-#H-%s^>3R)7sv!h$m`3
zV&~TUA<n^X=r-VNG@xxogQ6U#E|vdjs7qq*(`LL|BxN+pa9Hvyf$TjgSIJSTf3zp4
z*ZMFKcB+3c6<B2Ln~&y1>M6se{9oO?DAkH+{2bci>+p>UUI6$j5!J!^-KLf4ba~YT
z4Y#6bR35T7j^7)pzBTXpP4$n@3%~0ZS)82`rpi|&@1Gk!Flzpxs(h%rBGF`5qhpt+
zQYjd(AW0*t28>!$YPJmlPU)*yGKw92G`am5P1<Q*u}%HbUWSW!&NGGG2Fq5#&0#Am
zHpdXDhpX;f(&jY>P=GV=b$SRH<UnKEf9q&-PRL>9WpbXO9n3$=yfB6SWkQHgqm!Ya
zg|>cI!OlXglIuq%(Y8)ip=&W@RoFL&2&dS?6?aEFCz^&~-fBwx(iAV&!ZD2|ym)2+
zi64IFH209T_yq73928~nhq;5T<h$WWOPcpVmHit~tLx#ILq#Po0oe~T_XCt6>_X05
zTgXd4mc$IE6sENiVt*t=g=us^#KwALwaUf?7#&>GIGPh#U#A^^7?}tE^gD@!E2d8E
zg=+slDARt<@g^kHoXfmnVtUp?U$i>m!1dLO!VEk4WLC$2M2&xp3be)RrTYB$O0qAP
zCvL;*owM{3*hxM?`}r!1b(U1{mOh8G^K|S@Ckcft*pE40wunDB=577-?^PMEg7=Oe
z!>iX?3*dJizypz{^t0Pk6u~<RWqSl0A<`|SxREuNkV++C60{^rU|it7w)3W@)#LLW
zmm2wRkV0&^og-d-*}>K&vzFoK+{NK0SwXDw^}5rW_&Vxv@9Mq$FwQ2Yy5Mgfm+!`5
zovL1TOTHABh+^Ulk~}K?@PKOYE6Io#EYv5A0CjM>es3r~nBvt;Hq*>Iu~Me355)Ne
zW%fX2Rzn3yc>FKi3%ItRUi3SVJctH9z*`QiP{{l_5HPBlLBH>}WjkDp`ej<fb^6oa
zoUtDbrudlx+#m*1-!#6s<8{oAc#W6k$U9<AhRZ#(NY{;Hn!hD5_}*C==NMFStLOuN
zA<9Q6Eq^F)?PVO~ycy!w8JI*SD?o&O4AiSWPFvh}LZ0SD=t=ZXlZ&b4s4n8s8TYa|
zzQCV1RK|bDBEFL!oGM4nEm=}c3QP}G+1`{BfUih@`;g$WxLFw!{pPacfzTIZRUi->
zsT7vKO!ttmoS*kCMaUi3>YZojp(3IIt-EI@m;L4Bt8M2q=w;VZ`Lbs_^SaH-rnlSK
z7qRuhw<|~KI;@=W?XcISI^W_MkwYS|+Uu`{CU>SIXJ*^b4#;}vi$&i;1eG~12$9Hl
z|1yAw14v8bCh*fwtT&>2czFZpaW!Wm(OMlB3b4NC=;Ad1Rf*5Yt>0w~9>K6cS>~PZ
zMcQX@oX}T(2RNd%3q$(fi4nk^VHgaq|IED-opvxCw_1HnIE{E(B{c6k`M+gQ@G`<$
z@BqHRk^AKp``_Ok$$wL2l~L@1njl7Ih6ghGq4%&eCQ(jy${mU5N#Tb1WB-%(o4jZS
zrqvN1rs$uHnJ@b4g*ZeqaWcQwDaSN&+m%r4Vu|&ol|qd~t6D>}zkRjMWX@f9{T(<`
zTu(jfR7P-)xn!4Skpwzkq_y~3lwsF(jwTlPprSsZ@4I)KcE^?2>Qt$iZ1;Hj2^4s}
zEIG#_;8)QC3g}sYNPaf*;C5;C)aS_;Li9tqxkGva<`3B>pN_^`JwHA(Ezi{!Nlpr`
zPf{pVS9&9hTQjxqUoTaS1g8cLO-Np0pIFycnf}=lsp*OOi(?)XLp5w&9)|ZYC9J8F
zn^=t^TVI#j)PPd@yB>9$r>BU>^)2t`by|hb8S#2kz>k70J4md}gIvplMC+B@qN>iW
z*-Z|ab2zqLy7eD$J^HdBl41{O?`+;T+N?HFt5V^kw3`ZQ46;CM$fS$By7=mU4ze`9
zXY?SaJ82!I6?OjNgthiQXO74c1+2m9EQ*zvI>sC)i`HY+DI4(J(13AAFs9CJ!(S%?
zGiRb+msdnC>H<JqRBdskn85LK+%wJV@OwDE*0M6Y5)Nia4VIHJs~K)o+^wuCd*Nwx
z?dYPsbT_xh&*kE=z?Ub3BLS9v#PS%lD@6RddBg^YSEXVYqPqibK~PP}h&E4Q*f8(T
z3*|l687O;P!dO}VTpKZLP(%E-&V}ra-NR2?98TLAc|6%I(pLA8EU)I_%FpsP#$*Xu
zn2bV0$dx7AebmrE9NCEbknP|L1{nYeSqRxE;SX=M;dJ{MUI^;T{=6m@)V%D|>sE7c
zhPwl}5Hpl5;bKyG61_QFuBdU8Y6Kj>X9Kkol~<^7Fw$?)o|EYLJe#teAD$T&|H!b(
zj3-@Hhpb#yVt0k0=M}7fH#)w7jBdOvbd+V2v6%%bmc9EVzuB!>7&mae$B;0{7@Loe
z^%SjZ)mxIwR1~7=8|`X$D5HOro5_>lNNIJI#guJsa|%QqI_M0CFjyN27}5TE-J;DV
z=pRW&OZQPQ{j<UE9XEMeBcCy7TD&4?P=RYpXAqqn(jo~FX;^`A)QFDYBoMUtemwuB
zLfojk++=XNFtHnicy&N&0Rhg}FMXWc96F<P^F8i7IB7+(E@<`mXW=A1hpzVGgLeIO
zm$^y{^+M$fv<N0=r1gMrc#(#%N6d`ZMYyRcdlBBczxaK3xcKLS@PP7ci7I0%38E{G
zC9hJ0&HTg1Z`Ra;bhD8-v?f+ZK+aO-1G803P)f;^yTI58yKVQ7p2sS<ld04m^2NhG
zXQGJI27S2p1UGqCF<<K^u<n5xePGCTIbuxiK&-hIjsvxW?c#!5DmMO@$4_$F8Ku$(
zaO1{?H{=8f`v1_zs$`@pwr4(%6BJ%q(p4qUIVg8jSrTP8OSf<em4D$W{O`Jf+z)QV
zQ!eMVSdUc3xB6N-fQW-tI3WZMj0VmA8kj*}-~Ey2OYl~eFosy&!VFqw@PZtuBxqK&
z#~KOEC*O_qV^|>p)3?QI(44*a)O<-~{nAozk^8*b3*Yt|){(io6?u@Y38-CUyXzoD
z03x?5M^seTg`-NA`s)wq@ZFoGx_Gx?l_*Gt$@;Ix;B3?W=eK>ezE4p}e2`KLx}ATt
z%LBx>G$bSxv2-9&9V8fSgYj1DzF<fDDdt5g;6ym3F(~1QYOQQiHKa4J88j#S>5FGK
zwa~^eN}z7`8D$&%{xTgzgutVrH7ygLYZHFp$0!vQP&8qPJh0<02+~lsd?6(yJE|hO
z4fHYMfpEYFypi(Nw)Rp!G8&i3wNI*)^f%0^h(MFWM_<#7LCdhOq{`Xv8nWG(>FXV=
zUuF?}Xlkqb^N39*KsTdm9Pr+U<8=wY61J~=KD*?heapN9L}|)ZQk&c~4b`XUGHr}w
z@#Cz<tT~J#;3E*t+y_%jAo~ZoCn+Pj)w_$g3EE*t@07V_4%HkGC||@Dwq0&h#x?oY
z+Ok!$+SkfQfhc@*Ol&v4&v%|;yW?V-(y%WMqk1e;AVH{ek38g(r0k91ID*8!S!cYv
zfQP3Zh|r04kGlk}YjDzzx%wVUO0MTy$_iqLGz_2j%(&?n%PXeZ_Z=+-5kXqGvpsAY
z6fIsNAl^t6bLS{GgoD5|XXpon{xZcLXY=~dy3+dSY(GC$HUUyXZrS!1W44#6Wd;Vd
zGryuz++wC0*wDaFOHMY7)?h-vHBe6eVyWpjgC;GOGKRO`Nl%^JrA(Tw$@}}=KvFcV
z>hibr#dlMSjAW){7j`^mdiM5n2F!;!4CIs@nUwq9XIX|O4ccaCd?+SLV@739@FKS=
zj+RmqR*u5&F36e6I3sT;H`0P_^iSR-;NisZ_u9o@ekk>5<c(An6{~k^+PL{r)gPlE
z?EQK9PaAath=S>-&P|pLz6*c)wgVx?Hi>z#55NSX2wJn2jyce|uZdO=RS#bnRNjJs
z-AG|Oo#cjHx@b*fZzy*r;%&n=hxJ1<87M!51-_p8M4wcQ9MLg$_GcaCtu|rT+Xdh7
zT@Ik%IX|r-y;lb^jZ2y9C+3WfZnW5AjTNRBCzzJIq?6`QXd|TT9}f55+`fwMrbgNC
z9&o1}O74+oEs$_Re-5LGqr(Ng&)}uTKP`T)RDYbV08gvAHm~q!iUxQ8KTMrvTa@4T
z_6aE|rCYkY8>Jg*7)E5I8M;eRQo4ts8zhGAP+@3rC>a{1yGudze|>)s?&E$19H9H!
zd!6xF14MLe$QHJ#LoA^znnC=yq?hY|)pv{}T?F}cOM}L1-LkXZ6c(Dmnuv=8hq$Ga
zMZ82KL(k&AH#BDZ>%L0OuEU|XRKuCT+RKUd`~fYEGyK)nYE6Y2@|!GPZsdc|#D7oW
zEZIJc(F|ud{PiT(2s9bE)}RE1qbX;ri(NK8GEeVHAj;jT#T=;xf;A{;4~sU1wZ-|@
zbQ<>GOfx@<cOl{ztw@kwreLZU*IO3$C?-Mt2->H${Lz}Zi876X)j3f~SiX75X{S*Y
zi21m_mkM}2L#T`yZ6(Uer=r{LPs%pL<|_hs_bCY+rM&%daWZ|tvY$Hsjd8!nR4Mpc
z3yYttN(|2`wBAfQRwM=IUhMtNW0&~j5q9ow9rz|c9?wTH7^1o=+jd9WIsIq-t@-co
z+fRYF2NHaFNA2npT%#d02Yxmm`m;HNhN9{`gX*qxJ+s~8ViD?-5e~Fw%zHg!V(Jn)
zCEc^71Y&105^X(<jHSFAb;UDu%*(%j?Fp_s<Ct8%`-+^G<`x*Ld9ut_>zVV(UHTz7
z)!RX=SaWc?u60OOWA3AGu{%R^(J^Bd=3dmvdsbhJG~o?i0&)gXl{ZGe<pZ;4rexy-
za)vCkH>)LZm^8N)7?)i&$@Qw>SWg&AKK!PJO>4(534p@UaSm;6uJdSD{sl>45s+^m
zNTot|ij_AnW@$+{7e?~YuM`7Iq7p=EnwW!iUj+AqRvBF|QRWx38mQ4Pf2&f17^x{y
z^Wd&UffWoQDxHx{$N9S7&#`EhelFD^%H7z<wifY~s^vA9EU-JH#b>#N+>+ZUA%whX
z@83)ao7J&A)f0L>-=u1>E1o*#QXlGg;)ycu?x20L+nxw>0C5(}P|L&x>P@6Y@)L8C
zAf{v0g9v|NoViZ<-EsJv`GJ*>-o=K#|JAQa0<mibD!ci&`TLx0HH=v5#vC73MfEiZ
zy`FTvU4C%XNA>_s^kT)kJ<P(-qKd3HiGJ)az(QA^sa^-WSVnJ#rbK%g{V>MwyX<|`
z52>(A*$8P=l4{h4=EWf`-!?q3&WMuQ9b(7aYjuMf+>3Q0e>r63>%#;~EaFUjT;VZ$
zsX)ZZ7iEnZ*5~`avX(KudY@HaFnO%#2f3q5&-=)qzw|BKlyjS@k1n8UGx^OOmpA<V
z(WH*V>0SbEYfo}q5gJjXF3V#lHU1<}#zZdZ0sh9IVr-pH$_a1D-kUn;(!8+ia*AM=
zpO??{a#|xahGaULbr&69W*l$$#nnlXgUDqEp3sy}0a3KkIx*3#*fc{;*@HW>rv@R6
zx}w@KUrk4+a-^)Uw93<A#+z~uMr-<U;Bl`XA#V~9?|6^TA7rRWHE4)?Rns|MqLa{k
z?mrIBbL84OLn?=VBjXIf3|`LQ^NNt<%ZMLWWOTbSU!*o4NKY|^C&aTzcK+N93to(z
zWLO?cH(Q>4`Fz6<BuZ$c+vRqP56Nd##Zn4nfw~?fRk(&>UY(K?yz}KoZMo-k9h<^a
zy54IoH8(Pdqs}pkO(bWuX$u41+MvYKdAM~(_$&6`OwA?|ON~)d!7%KI2Rhl};>lf`
z8xzZ1?K41FM2>A%ph>dAEa7@9+GG-N2ydBowQ!WT--KttRF?fk6xQy4mSuc%b6{6z
zn<P>weWBBmqOqP2C#rVBhvqvG^{P9@v64q?DOxp=@MT2mhfMagCGc53TM{-(oDy-S
zx0C}BiamS&W=dH5YkkSy{*x;{sPZO*-)2>Qc|u(X-&YfH8$%UN8y~d2aoa13cthW=
zRT>ymt80l5x$;+<s6M=kiD3u1UR{yrrFbcv(Yk!qXpt(=PG;WyjKFKXg1_EhSNd1F
zPPphs5*$;s6OxT{(0o1D@Buu_gae}It=CgpAGW|FPSTjs53KflB1EEWrJTqgre<X_
zrpyr6m5CU!pYkTHmHovT426a%F?%qG%U*dLzn-0qw$rE%Re8r?DJ@4+qu@<IvF+2|
z+gBLvu?{Jo57a4@-Ov4YzqY}GbZ0!1Ab%(NngEo@<Lf3d`kp@%>XoLk)OYx0u2WP4
zoppN6p2(zKfo`-vs7LcUAUfq!;>$)v_tdojVLgKU=B0nR7km1tYKOI2=LzeU*fiD-
zZs%h*cU`;=)|7Z;m2oN7B?X=xlwsN)8fBV`PS_zXZ35oXEoS2HMW|FkoS&^m>|H@X
zm&EnPC{CAbca!8bD$CKRok8Eq-;u*~)d&P(+&Hp|5e_}S@#B+QukgzsLjG=O?vD}=
zDt_==@Y>~6JA70sUj>tNg%Lz#bBfkhHx^X1WuRj#8j@^7crCXm^I>9^bSfFQX&4pT
zB|KOm)D{Y>jjTiX%W1e~LT&Tg;m$nt5YAh!4Xir)`KXIvj^B<VC7e1GNPZRpX3`4k
zMJLUb_vnJd&1fJJ%k)_>m9bHTeaYAovoOG&G!RLcd|pD1Ey#!tJ?dw<eZ}N6k9z-a
z{&gtJZJx#EQrU)a)-V~W7N_Ig)*qH<itUrzhEeD^^ZeDHtpRSL2(?EK*-&Gq#r!E9
za(%=ucR9B4Cux5AT5JTv>{yeSF6|BvDY)PbVi$;*eS5cL@1x)?T&XFMKE{X{7N>St
z=3T?QH(aA-pDj^mn<g=XqTF?4EmEm*Z^>%ujl6+NkvS<M`0*~OpsP<~HOPn@W|*RM
z>6^aWgu~jh=!|fj7$r5EcC?J%59>pF5(~{cKUKsgJkdcqI@D;|r^ClWln7=B57nR+
z65fXS8N{VC8<_xGV6U7Yh!vXZQ($PHjcc>lPa|~pf#b>~apXB?`5@k2^&i%FVUe6x
zR^rz{f+Y^$L=B5gH~&;Qn<dW}r!{IxCN$-3)GXP8aIy&QX^$(oF@IWhft!llDi3Z8
zlsz~%c3#J?avWCmmbw(3Ps(%i3uIi<i-JBRel9N5p(>hTIxdf@Mv6Z@RSmX{V5x6q
zqO1_f)YB0I$&+^u;d6se?Y$LyPDO2V(E*Fqh|Jq2Ld3#okApa&PayJ|+x=I(Z?TF&
zf&)C(2Z#sHeIsL=w-7Mz30xt2&)8+rcg1Cw=w$n<GAI)kSnzoBt-lW?_GMn(y<l?S
z*;rZpOT!oR)XN!$BSIdf%-YvF;tfg%CD44UEovM(LBYx{9Z02*J6*4Ew(5IM=H~$z
zDAgy_El=|+^o8~gx>_exYDZh#RieRzwQUoC6ZbUBaTQkp<XLm@m6w*hL_PenO_RLs
zjw~)4HzbVW2O81RM>+<Cc+Hr)uf-FO3&J#7`I3nRE`|uZdDvoLEn)>EPwt9#g&~1~
zsZN+#&UueZIEuvdp2t4vab;+3a+nF`M3h!BBI@l7TSbx${EG8?`60d3CSxTz!$EU;
zKT|XbM_F~TejRW;eW}eyBcI;o03Q3zI7tuKv=awrRFy>r`~BI*dP>toMo4`n!r>CR
z)(|S1op*H+W|Lwy+a;mW2w=q3e(tv*%$=cX@AL_jd<&yypueDg9h#O^u=;c3Z{B@f
z!Y`q)cd6E}Gt&T%%L?lJqCyV&E~U-M$z#0zVn+|q`G?7q^Qk$W(k4Km3LMWWT(8T=
z%u?qlZQB08M~*t3w3sBf_H`lNE`D7G2q{qyGK0cdyKuUb<@{)iyo>>ErVd$q@Lb;7
z6Y79PS*FjjeU{@L{u)c4J+H!HhmL}p*YT}sUq21}(RwCne1VxzTL@+^Z8r3t-twtP
zEUCvy$Uall^|M&ci<E6oEWIhHzMk0COEG!k8?NY-u^m{3!%4ru-T)Da)8f;y4Dwy|
z7p3KS@q%}&c2vzoQo)k-@GB_QJGgmRU#N71*xpShPwGgIug<`whx&J4m%hl08V7|5
z7kXc`GgtRV#Bi5Oc3VMLRkl=8P%q7_CI(CDxh0?Of^H3lEM!d_7$^A~6wzArbO>QZ
zIJU`=U|?QAKF^fXh#uL7hcAXSR)Q*;^=`fT+6g?h8%3ZF>W?QxOj%EGU*2WU<q4P*
z9-;WV#R&NmNEL`93Z6C;(A}xO2^K}rWWgSqAY29upz+aGSQ0+g8K?cXGJo9~ZBZOH
zW)knn&swmJ$TC7RE0Lp-u4iz$Co1H=mLKlZ>9BmDEyBrkZzc|m?xp21QX?%@Yt-J_
z{h23Q*=%ff<oLW-`0k#sFm>MMzmDy-(Kk@4FDKD%DHR7?yywfl2kUcTh`3&f8Yvs*
zX!Y!{+Sio5(w_ncTJPOkV17xKz4*`8@=Zhy0hK2g7WCwqdE%a3WN(^UNwY?jngLI*
zrl<J}9jXe;t6&04r-z-PM**1?*8*;Qa@$&Sp2F!z{L>4IDEn4<l!im!*dV9iJ3S{y
zsl?31vc7x<pW#ra>6{`C(74ek6A=jcqF_o#+`5n5iE7xM{Ed!Fj$}g_Cc7Y>oX54%
zG$T*!d#4EQ_O^ncP1%Y$SuX-dP<nZEYkx{q6@9O+xZeG2&V%K*7MNtY9UdrW=rMG^
zlu2`C9NF*b-D{IFe)(j{{CyoEe*`-21t*D9Vu=ewsnLcf;H*vX4)-`WVm^;w(@u3J
z2X(7c$xzs{Vf;PWR9XAeWZZL8CjgWyLO1$Z{{6<@b*kOHIad5}J=7gEIx11ox($5%
zd7Ua<4}WiMt0?-uD+GpTy~B>fU+@3zn*WH^C))LQly8#TE`}Nd2@F6n3cV(5tE$zY
z9~_0C9Q_@yjAQp2Q9hH6EdRDApKaG2W451OE#!X(<WrW?>h%*_>|o3Ozb)mBNtbwH
z2SovQU_ZYwEQWjKw<}-R58J*m3GV%-BPE$$41QKg&YY2U`KyfFV;u{+0fK;rFL9L}
za+LE94Tp*nISr>s-DD916i=o4o_K#Z8M(Z&wy0-kY4U8#oYY@&`aLD+tFKN%x6UaN
zyDxYE;>l21IcX$z^mX*#FDHm#UB6OKCO@PFvXyKyFudky!#Po>S|@b4kIVsMYmB@B
z%@u_utMjABlPzAlI|8#d!8JKBOAb0D_rptwPo2tg2%Daoeo0hRtF^_HvrRlZ@>Oc2
z7T+TbH$*u^BtYJUe?FKesH71sLf{jU`_<{70ixS~j!x{Ma>i01H^S*DcP(})RCtLm
z93WiC^o>LJy?o~UOvvC`i~6hF8J*3JE@rmJtIba5p89p8D8WQxyBAa3tsf6!mqI9Y
zL(xCvPntvWI65<4iL<(o)+#!;UDIX=*Z(IqSqhYFiidh85C{xB$x7q8!w-!2n9}r8
zL)W`=kl34P5&8l7oq5Pp=d!mi+mf1`KOIc*ek3Vv_?Y9KIva0RM%8Og^la^@+#5WU
z;>eOK^qJ4}Bgl!sZ6D{O8$wt+ea1+_&&gK9oOj{LWs$Fw65F#y&aTzkkDqe)9NdsA
z!wdbyT&gRB8!Jl_Vt3gN0tU2pqN1O&I5YIsFB14-6KE&t{ki-;Hl_YcpgOjB6y-Q?
zeAi2nZ6GvR>CQN-4mUVth$-ioPWsMmbPUvrudPwcIy6*R6SbEN6HpQ^0X%!;i5&_#
zZyTbah#)bjj`BOQ&v^m!9yTik#265y!P*aT-{tjzm}HthR{mDCrpUw9{x4wGih3xI
z>)mNg6J@wK%MlT(D5b}{OTDST{@$hrg}i8smT{B}4&7c|4m~DJ_VF{iXuOR1GgesO
zXe0sD`|5$n>-=gbGPX63uQblUn&E?y#cqcr@9)b?Jo!NTiNCd+I^wT`ZY9IEw=03n
z%f(`^jOp^l&rG0z5hvGn|0gr6@J$$cDp<NV)h7s0=J+e4ak7Rv>U(|S@#^(|wcZwP
zCJL=Uz?O9kCi213jsYN_Vg2MpYt**kr~~8|BKcLZ+gq+ON%^+D-d7`6=GQcS>z*ix
zy_~}9G)_p@hd%gE*GPRE50!iVwyBjF9N~9Ql?YjFJzEgv{VJ4#aM>;=WXOUhO#|s$
z!(Dk|n?rS``Eg%T6C#QEewvuA_pB05;es_$OR-FCuJXiY7<J^EnhxTv(K9pv?Zm}Z
zx!Z5LbKkKk+e(NmiMZK@dzM96$Ba*;gy*1PG^4?PUW>I2k2Ye(*Qa5;!ai&*TJ-~*
zcs}}5I*fO^&xUh&UxhEES5t&neFOhfO@50Y4t9*6Ql957&R2bHzNInMW`zkJGG{-7
zYM8Sv?WEOoyK}gMm1GOrvf8ZaK92t@*QrzgZx0X5$F_gQ#XgvWirj24f0XgwcWyH^
z)_iyVX6bYz(Dxk}hc2c5+?QMZho(7L5YGDDfB@iD73FrQZ->9?rOm02a!dVTZJCU*
zgyH6A`QPLdBPv^4gv11VsHO1YUP#&?G-<b@<x}y=5vB`TLKQE?5upkm>`eA%@ky>m
ztqc=Zv8MR^_UxZjNd1Zv@nvZa)rzV!`q{c~xZM-e73t^;D_hb_F_D9URW~#gmcEoX
z<hVh7!LSCjj^(&L19)-i$=DHXCKd&71T^9!_V@t%7|wY`h-`$Q0DCB*_yxvlPgBAe
z=)`f&TQ~r27%PG`aaRQqFpz09Y>?30=ZbFfVt@g`m<{HM^|yNm89^MWkG!TemAh*S
z=_K!XD$#^(xU^8UXi`GY<+wGr0=&awuYz7@wiE4AIVv$MLxNT+PX@=jfj_kanz?U3
z8HZABkm08iMCOU}wSOZYykXwGQtu;7A`a?1#}txRv@6XrYD)qPh__ioe=Z8&Nirnw
zT&bg8%Z7w!<u5+x3-k>`U28>|y_6ETea(28szKTrC=B#Zfb!AW-)QQy13AgxzoMRo
zT}%l!cWoEf7y~h~cmu?FmUU>}xGh1i;%Asxy+G?_9!TidRVIOuTsLZndE4OG+V)sq
z-hVZea%9}Q$p)o*z+#5LD|>7l6MLP%#kOYryP4+uj!vhCN7j7X1>KDMZ)3gd>4G{w
zSE`9PG?L7?e9|qNKMu0*<)~gPH5>!%?TQz;WaYfo<2Y9@0-ufJ3@};s6bKiHI2G~j
zHX!<=z8$m9V0-63a_p9Wmw+piXH}QHYQ)6y^DSJ7PyK_u)*dzOgRtQUwDVgF8EpJc
zvW!}lJh#F%^7@Nn0r_Geq8R^Y2lY<xkQS*L%+p+m{F`NJ8tu1Q_YTpjPRZ)UT+{D~
zYz9R1YRX?UH8XWNX=6G!QbjUcW#$v@e6$x{_Oxs$c%8V4$B*Rj1BJ6F&C$!fN?Zg3
zBBr{ri1NX@e}|b|eL}^uHI31Q@lyCcYih$>I-EsDm-^xpI~>%nQFE^Gp-}y6<c#-f
zV4Z8<hTQb(*+X?1!203&ej0Z0`1C%Z<5{WSD|*dV>#gPTXdt8FPKJKW2FdK4NY!hd
zI#YyeQrP4t*u_2f1`)i^m4F#8dcQR?loV|(X{(zh9+52%$98^NTdk5Svu9-g0YCib
z9w#g)M)p1bqE<VLu7!ArWZCL+iV`>BXTluHSi02J6JY~dDjs?T+8P-B=xPn3h4oR2
zWCGvtns0P36}DYO9mgDy!g*2P?ayYP0=-}#49PAN5rhuek9V|zb2T~hV)S=A%-Cxw
zdD=aQ7mv7Vw|SgPKG9E>xp&A`cZk+?1WLnJ1D%wF93FV@EXVwngaZ`3o>eaWG{xva
z8KJ$nJ2!36O-fLfhGa`t8AnhIKGUNoD7u*T88i3%L-BY;{!iLGWXI<tYN8?~mPfk;
ztdo_tzbK|A$$*pOh$cKy6xl-ME6PPMl-rmQE`-<xjCq9pX&U%pg6MI;MPEh=eX&i?
z0_JN##|7qx^~2e^2QN31i{eYX(7Ov3JfHZ%^E7g3=<wmG9i0B}pB2n659c}nU<DxJ
zaBu%^GCo`haTx)Ets11eKr^<-oCy1?QoGdVfbgzqJK_>ZFM|47T&L6xt%e&kvtbg?
z6u(TOn0M)4`x}9F*@pl5{V?V9Qz?e5tmL?Tv~+8nvBSZ5R{#0n#1<?M_mcLYH5JX~
zG#K=Tw78m;sMq@|+FktCk6B(&n|be7m#bcy_y{(f>PnIv&(;;LA{wgs=n1DfXt=CM
zVNJ%j=<(nn5A>C|!ia-z`VLE_Ud%UNJXzC}mw`Z#`_NrPZ~~NidUX-`j(17^N!Cf3
zYt<`;uwAW|6+ss1SOkqW*hvJ+OdO09KlrQy6_$ugwC;a3d<XoNXi;RDdnfRZCtD5A
z>PMZ{Zs+L{0WaQufZ{|kWTDSH{$krHqF9T^z=5JyoWp?Gpt%7vL6n`Kt&V<jt@ebS
zC1@cqxYFsoOkFW+7>|~Fu*DUqiTfQkqO_mC#lQClfv!+|89uCGlA<R<1j$v{KYboi
z{xBkn(=?K%%WtmUHQ_mZxMZS8R%erxl$4g22Qb7BULIk;JyA|;k$?Za;Z;A}3V+>7
zMgtz|sVyR{r6<5fESX`kzJkA|p^{5Rc6JmsPE|zZyODv)W2Xl?KP!elU&dHk`YG&Q
zOL}ZezvDAzEBP+f&oQ66Jq-)1BmL<4>b80L*h{+{2uo3Qmh)r8K;JI%vWhXi>#64w
z{>XmqxpvcCa1{hj5F$BY?|3S2f*v<X3_?OUd^|R#xf02+)e9O59BXgMB=315_=`vE
zXl`5FW%!|WN9IQ;-t2t(fVl$w`<Y2%AT?tQl&E8P9^^HK&c>o@AM0;u&G{z@G~Vl^
zG%0&eq<OltLg{S@VZ^+qhhBYO->8kKFfxoaK4j7^R*?o1EcYh7t2=i(FvI$EC`|Dd
zp(!mnA=AAGSQVI7NQzUNH46y821gc&4p1EB%#`=A6{nAa)D(*G_g5gDfMnfk|KR*Y
z>#=fV+JmKv(yTaQmo1s;EcFr(ESo0=<&XG&mU(-0<_1{S9WFutkljamk&DZPtpKS|
zNKRqvNIh96hSDBtd0Ng6^4e|0G$++zTFaj{kf@XjmT!{UTwpG3gOP^%KZ7#1yWIlm
zQn{+%Bc=B%ovWKrYu_{o>@1|PV}eVN(bLT+{bv>b_x7)TQsGMP$)P-_|JFVJgom9J
zj$hjqj=#PS`^U1`RQ~!S_)V|oo}=i^L~yEmOP_QG9&T?|TxW$}<n<8l&6i`z2C}TD
zqT+0PG_SlOpZqQoJMq?i3r_Cu7^V5kK%=Fjss6DxolFR3<Q=9SLS`$AXfc0$Z2uER
z=!_YQZUOp6uSUKbiwitK%Ts!b=?P~HSfmb(iWdg{5ydq$W{>|=F@isoU`%c&{RpYO
z(vdA37Ftv29SS!ri$({N>{52mMmuZLz61^Fl6WWj%&05!TisMprW?6)X%Utj>oB(r
zaaghtGuOH2+f|!MN*^nz@m{beV{`1pv8sY(lAUPB)vfP~EX_c8Y$-+B<j?)dy+Jj>
zeIptXQDuh$yjcB}T@vK`UlG~=0)J@aB45jwXML8W!_D47{`*xgU;O1EbiRQS8+c6_
zQ<MBYi&A+yFU9hGKK^Em+;}_UvtEtx3}4$z_HY|`B^g7ltdM#jwuVZq<Jx0*HmW?+
zYA7QV+UPdYaMAR$>#IZY3mK8Xz%TIOD(plGfkR8rj0&CKR%4H_KjIx`?%nicb!N5N
z8fEs&2775bZ@r8AE<7Qo@9K0HyH7QR#{2ea(ObB_Q{K5V(8B-HUC*hWNEAYB;C4;B
zbMZE3gw;nn^X3||=m~(cD+R|pooe9{EmVB(l*m_63S8*_-rcS#>6|$Lw@4bgSSY1?
z0IP1xQ>t}ZCewi5$CDt~*gr`Kdp54JiqcT3A1prc+eBt0Ve}xeoby#*({O}ewptIV
z3)wrSG%BR1{dh5D)}~t%Jw)IH9;cLx0E}*wgOUmlcZACQUD-w-l-z;kquh|=f3`Ig
z2Z*@oR8*B!*oqEB!5h|eEcwVBGPnQyO{m$_&TI8UhMX>M8fU2|)KEdoP45}&mHOt_
zeDWqw^nWWU1*fh$5B9b6g)A^QgdXqv`A5r`b4es0ecRiPK9=9Kca)Td&;1#V=mv;u
z&Ga(n8k8Z;Il#CQ|L^c~jeMwEoKJx<y-XlbCzTKM>t~dE`}uEq<rMe3)AX!gXQg-v
zyAorgzv^@SWQ`PvXX)#B9VNdbJbJ3@4yLpfxN`+J4e%W6NT$_;@T1T#vMMs)m;M*V
zRh2Y-NLUA<8nTS+N(AbZyPjthPaL+5Z@3~FLSaR;S0XE~l2{A|yy~Aet5TsgGkh8H
zs5#%#oir14%?An4OHFfC>+2V%N_xDZ0D6byP=Z;|Ms*sdc;DVOLEl)SZPKClPtj5>
zY^ND@E!hv36LbbvT8ib$@xO7xRaMZTwvojA;XRf066iKCy`>=IL{}AuSJH209d6uc
zRvIisz=V~?YkCnxEj8v(vNT$T*)IDqkk?Deqe=QDuQ|ji5-fQ<9F2nt*|A4*4+P4{
zXOqx#J36OD_H!o@lj+wxW*`8R6-~@}-q?81cm5{lRwJUzjQkJ2pAUva3g&fPp0TU+
zKk;W-dL+g^!wW;rlM^=F$OB5s>z`TAvsR|F(+j@-;3@f#=5F1p>}~Z5|1r@)r03_P
zwfb;DJhAEO@zZbS&F>4BDY;D9)y25QbG3M1<b#|LXaYYb@_jZ<CuO-7SN?LJ?|z>B
z!3N7q%uOWSuS))?`<=+fhNZr$6n)VUzh4nLLNBUIL~zc}DZ;Y$oVkLFzXVb4V&7OW
zm-LW*is9sn*kYLqP8t8yw!Qe7%Igs2GF<CyZ!$j;)u^4uLuFaVsw3UJXp0!x+iv<A
z)Q^RZjH+k>(`PYGC5ow+c2-ai4ZdC6XL3aQNX{FH+QZER<|`@ZhsX}k0LOYVTGEz|
zAuNbBIh$6kzR=(Ka=PCG`6kmTwx?muu&rd4HVbm0wW+L|EKG790pv#M6{a!Vu8e`n
z03Ga19cw5z{I|tHz8%bz`5`mo%b{a*K1AKYm-?xRL{dyZWsCXHb5C2m#F_oXaKt+l
zfQ5pp#TmtU@yOVNs@QCV#IW9xIuKKn?tDj&;&9FZ^)$?|yKDtOl=@_);qbG(Qt}tt
zJbwpgIOadI<lnl1;&`i;nx!0WP`_KQg1*P~0)yRA5+5M=Y9=-$)PMFNKl{&K1eg{l
z3nyOZ4P}1B9)as5=EIa<e(xrG`<WQFC3XAbu6&mGIUPYudaB=fA6RH!u)1YqcIjBS
zx~Y<Uq%)L77#%)rv1;s&8-}sGlbu`sdk_@hN^Bv-E9y3tqi-!*v&mqkMFW4dBr&8v
z`Vm<!;y$v&tz}mRPOuqENZ2B?)aXY&C!*>#SsD;jg_-odk|Xubs(!k&hNIr=vgFRI
zVfVKu+T?|h0g@tFn3N$3&9rOpV<}NbSR#}Q1FZk3hwKioM6BJ#y-HV7AJNio+Cen7
z=9Tuc><)tXa=;#ubbLJE<gI!@maS(gXYZ4ccC%O8)?z<SY2|<Ck;tW&F8sraQ-wkQ
zFa?tv8n5dh@=CLZV8U6+z=Dq~iBxY`ooV7fP}61*a!%{pTA7^(`ls^<nHHJFxN%!F
zt+R_09kmf59q-}S`B#tC$V5-u?g;{piA0vfc&%<K1)95_BIC5DKH>(gX^t<+Obs=N
zWk`$%!%=+I?^jABd%WTs{pqIVD=}$7Twnf$^5rN#Uz!@4`oK?X-Yy(9Fda4hl>ez*
zwb+B|<+Qr{!T}e1)uWlo1DK@mL6h*vg39otN?F}5f;-O);RD3=1fsH*j+SX^DaD@8
z4kZT1?ghO0a=Jt5Je`^*FEZ5Pw*x<{23IG-A%9NI?ztM_ACfCs=8z5b669@lJmjTz
zOccjQSHA^QmuXsus*|)WO5TZT;@KCW3>K?KF=mk$MS8!?p=cky9J?7bzK2;)DE^=!
z7bdK92;xSrT(=#;x)>2RFw*bsjVLM{6B&*1SZ7W8&<K!ZO`VK~8B@``PF%#p=*N;5
zk+z4~!e!b6+xGzW)NAmhcBdF>s~B|N;L|}8pJ9tkXtu%;QfN+U^__^g?w1`n`_BGx
z+!nvNV~6f$S0cTW@>P)^w}5O##}oG((KtV^XLZ{INLE|6JtEU7Jbmc+dXGZ@FSEET
ztKXCI+nLB#RxTag+a5JsW?GYJf*$L#mF(NW>AbP<bWSpV!`660UpIwA1>||_i(M1N
zJOD%BB6u|EwHAOqbD!OZV*GL+s}p)Pjm=q(vu!zUyykmK<?-&f7CJuRHq1Zy$K682
z(Ij0$^d?k>UXBO&x+lIH``fbKzeZF}0X$FVR~iR{Q_ucAjHm(K&a8aUCOL5P?FhW~
zo0RMoU*DuA`87KU-W~hoP?~iycYD;ZHr{bM!(y&Z8Q^wHO952=wo8+lqHh|A4b}Sp
zD(vi@iM)H~fA;%b|CcLhfDv$Xk3g-wbn;}#;DQD;d43+6?3s9#73XF2l8$)%HE)ah
zAp4FL{gu!Ai&Y$9sT_W>S5%d{^Qo;vxbN$+uwgWVp78jjB~8uV7jME2%p6MRy?Pu*
zbkhYB2F?2U!(DR_mE+e|ryRB01ad0OW1`A(w+Ej#`XRA|CrjdRIkc!s+Q=5EEk!R_
za@w*)m>WcTS?vJ27gi_hY2PRysmPXYW)q9!H?G~z)(vAlz$yO*ZZGI;l`dyzrg!en
z5&lagRAk9#P(q$ihmY>T0NV*DN=|gRbx*z;j@HG}5&>u*&&(ih@KVYq_Z!vRcP^zG
zsdO%f9kMw(d|glHFPyA>+&u(UH3XqQ)P+cMTs0OY%hY|A(eGOewz!$1Y4VYldbSFj
zT8sQ!G`VPzde96zL${S`vx!+*@ng(LS&T+Wuu|d2o2;{Tt&f<dk2=GtvjXEuy<BH#
z$s0I#1?S?d>uAbok@66Q+HT7ck9;ZEG}vM%$f92AJ_vu4>Q(5bg5D+#&T}rI^7O+F
z*li-ofXbnLX<CKsL{~e-`qJX08PQLluDT@cx+LCWSAw0#l`maVf!8GdypuB<@Ah=;
ziQr$6TA6qG1Lrm`xXAK93vO-A_5A_NY{v-<b{ZMqg?uQZdX+P#qlh*$$>*dG6_-n*
z0C8^u$`|zZm#_f4^vhQM0i<$_A*72i!(X3>$<&kK8H&D6kM{trz4R78@5b8C_N?tr
z2Du{4Zxr<ILuSHP2NOwfwu!tf@3}s%cReM~@<T!hpi{L%We$6Zl5NA5n&iljCwiK|
zv~%FXvG5FfMo!le!C*A#hKn*wL+r*;qFEgUaT)m4xM;Ajgz750m7fqoDVr5ecZn?|
zj7K{=%Y1nq-^SVEDc=9MBGv&4pB@)~P`&sIo>J2)C_eEGQgcwIABgE3x`2a5#Um^?
z;SvQO$OL}IUgy2ze6}c>cd6m;IC0Qs%UPz!iMWlA@lo{CQXc2Br*C-~*SxIox4(Dz
zR}M8(Wn0qiT0kOc;Sa~P7C<@y1~Ts5Nhis;3_#C0x6wb8my1Z4U+V-!w1gl3zMGzW
z5#8ZCS=k`Z8tOT14hg+hkjq=#^MQm7{R{mm8}{p~UmGrukc`Au6!3_Kw}YQ$fff0q
zSOm%bi(e{*{bWc!dhEk@{I-h+Z``j&?cb2B&oXs;!D;rV`IXLZ3igTWD_=H)#<?5p
zOS|kn)*_t=AXKe}z`Gd2y4iy-npI61-QyP@w2NZfB<O`F^W*z`G99^$luw@ooX)0c
z9d5GUN6td1x%1jgO<pdscbl$1*j1U!LDM=!omwho30p`t;u1pLcW!EquVQ}j;Cv^Z
z0F7Tq;FeP?KV+Cd{i8Sb3bO#tZU6V&=PJ}y2~6>`ObM4)7YX9mA7Cc0aPxhA?RsIt
zj^iy?3~fHsLoy<1WmUkskZhXIZ>Xx2heRaNQ7w2P?`hn+Qr_UD`OyR;{5FBi(I#LQ
zw}eAdo8}BG$Hki5kR2}U;|*z}dgdg21o=`v!Br)Nob0hXNr-TMtxt%7mx<%B>a@V7
z=~@nvxEHCO4q}eI^Nwbgf($4C(twnqp-4jn-&_U5RZbQQO&L&#_mkqJv?1qRKzL5r
zP@}?X(d@mIE3HPANrtVuzX8pjTF63p3dJLbX?$#_{6AK9Ikw2~=3JS8XRj+YTL(;W
z6hdt|$eY{J5LS|DkSz@AH{W{9ThRVZA2`cM%GAa7IF#+K06l>=Ho5`;o`Xns^r)#b
zT>9|mhd45uYdyAJ#IfuXO`80dU*_K5zsGPL(5nd7Lb8)xF?dj`4ZZrVLvy<=AOF$>
z&j;!p@f2dctWUvBGCdV;7ze2{+_U+al3x<{ED5#GNBHBDtIAAgr8_$-=MV=)pI^7w
z$i7=FnsfTGtuBqayIiV=Er;T|4cgZyQX#ePvP^rZl2S`qal%f$;W0@s^O@i>O}s5_
zrXJjdm1tRAX%mpB8zrVcZ^!n9K<4U_`EDvCvNUG=2l(Q2F_wg-R@zcaTZ`T4b`YDM
zYu@WNsOv`jio=vz>oQUEUt7V+th2BFW}2nd{urmiT%qH(nyht^&Xe+9OUYuNr7KlX
zd$YhT>TFOcx*=H6%f}HPASAzoVrDbg?onqhbkZvr(!|TA1hrC?Qa-Rk-U4oHw%}EW
zQR|Vu3lltPG54`k`^W$V<@47<VgUU5<cs{yOmV~eDyrw5GC(C}dQ+(P>XDFA?`W&m
zX#Lxt@<5KKICEnw7*&((fqR_(M!*0r@uAg?p9@WB9WXChi56Y^xA^w2(zDQvtbAJU
zUEqfOy+vsqHcthxU-dl!Jl^i^&96<ofd6WzegW>Y<2btdd^$>|f7+8@ZvuQMAo_1(
z()Q<2=@*gS+{Yae{)&f{hA`wF_cz4hknYzSbB))ZlJBxyAI??G7fGqF<R5*$Z3ZIu
zIL>cA(bHQz#<Xi(Hf$B}X%`ly>?~tcy^K(Y<i8)Qb_c@QJ?>Smd3S7w0777aaQ4|&
z_-Bge^-m*AAT&EX(4aApQtG)CL`Cmc?h<IoRL4!_QKFB|`mlGm316^f_;e0h1|63{
zK4vRLX}{bgSLc+!JN-Z#l+ysCN8j;a7vg@)t=$G>!GM{UQBTQ5ma4M2uE<n`(o=mC
zqna`;YJ!&EkP9S8Zp0a2_fo1{a$Uj>H`S9Yg}lkqng4N<vEhmdeDUZl@&6p}G)kta
zOufGqjHzuX1p^$sF25^L<o>rjxprJ9M5iL9)Wv@p;Tl?VSg&O~T3AcGx0Yj$l}AQo
z#E`qfZ+xxCfsBiWabgP8e<l?|P%qv5LP-pqYXXQ`KJKp<hOA)?i#pPZ6NYrvV<K*#
z(8CepE8|2JE%n;V7CpARYZJeqP87kXh@wtH&U7ae6FT6TFeh8ls+ws}2i!yS5A0>e
zWnkJ&-5t^R-!q@ygtg5lUh&cIV>rwAvJ+rGR)xADm<>e)vY%{os^@P)dAOo!f&U<1
zz5P}^EhK>v3RoZ!6Kj<7&3b5c_@1V`!y9lQ6&5>4MO6y6_8MZH5soGbJ|Zjs^Th^&
zQL)!=ioY4hi4)7{$?ERo1Dy*6d(UkYJw2<5NoD&-P*j1)UIuDOsJpnRozRW2hT^m7
z_qxGuwUrMo*+~9LloywHfRG+_PM&Hy9a}g^h`(f@cOaDCIlqlHJQYU<l_Mc4+>Ia6
ztmnJHf^33P=Q(qaq1Q85Cy9{O{le4@RE`c**`8NkOiilQ-}9Dp&wpmpwzfJEN616q
z<#}MC4%^D)R()S>y>Tb>+1E^3;LThbH0vBwmRL@+MKFvG?jNxCKDQw4ZfaEqCnx&7
zd6GSsHYnxM#eB>^?RsTAEwwM*$01A%us(K(LoS}kzjcKJ`B`As8B{Gy9D$fvGBuOL
z0u@Kt&R8IWCjm)@)<Mhj$@t2;hJPh60Rwq$noPa<0y-)JtA_;OCA%Dv$7%|MCg7y`
zKXzvzw1OqVn#9^r_vM3gz~Tzp&n5;|K>(58oyxUKXVj_Q;+}mE_rS=2OeRvk*o6@=
zfxYIugG-0ad#>0k=Lr?xou~Y`+f(zuld12tjrj4fWqP70`SaMLBOaO_mz<~cbE7YT
zG+*vH-{9-eX@VIOF&-;4Au<!0X7GyXJWh-rxL8)!@{~e<@>Us2a|oy<Iw0yhL|u~c
zyDw98{A=1y6F1J}P<0jW<*jK+*Tz7zk|lc4)jAlOiLOm|kmkq_6MQ>~rRM@|AH3MC
z4Pp(oRmEI=ZSKU85EcwE38Lmlw*3=7K+nH+xXKWTsM-7ek%bUCB5XyQz}}Q1P+^zm
z4c$w5@<cDC_w&{+G!#6R47=qC15?sJV$U<nt>Wo-8?0#%)qeSWK}PLw285$oMeJfR
z+P=M`%Nr}ncSpd5Z#ezn91Pp?!rZFyA;Gp^gOymMM0`x_=+djpi7yySDI=6I${#Nl
zxP8penJZ{&b1GHVh?0(N>W<E6%_1R>i|BN%i>ClsGx{Xn`~@|oaY}(_vTld8RMFkZ
z#6iCo{d#V}iWC9=)&f!Q7DaOie`V=w(|7Si%eCEIiH-`d7RHIxy8s^FU>S@iTjSlm
zf1zAJw39sPv3{qou&u{Z3Kwum13GLv7a7=Jx}FEB0aINqAUJ+!uo!2e88y;~vdemo
z=~OxmS}=wRRN~?cIiXHbPM~Zoi8`yc;F5n6#<5pjqIJ%B$Oib6boNKtiIwNAx}gNd
zo#LWWZ)-GVIxQPyH2ZjfWYk*Edhwx{qL(vK;9WMPStzY+g2tiMUK|>DWiwP+*8-`n
z!seJiZn`ECLps1P-@rrn-RLJ5WE<i>tdCk%g&meAU-W5^I4Pfb_?~i8vRT;$*tE2W
zzl@ZsFLk31?vr%FqMW9s6bxtCo5wTO^eyFpd`LuAl}492%O?itbWsa9#H+d+%X!tJ
zU=<0fDb%OUS#;4+5JZ*{O*WsLhh1uXs;C~HKrGG29tq9GqjJ|9<KQa(GxSq+)S#MF
z<d^7s=Oyl7zZbfKIT`p9CKy<T8loh$AAzPjKzE()Ar9l)uo8ik0q&1KuFUYf4}kd1
zv*44K)ZKR^K!x~MO6=Y`f@QgBvzm}}JHYW4mArS}1t+l`Y*;9b>SgRI;Iow_5kJ}k
z<RkCWh`;p^u=X-9o!6NcR_x)1X9>EhQBn0uv-`$|({{_jH2$*xz1yu1%AJzonf75f
z_4~`g2->heKLww4dyaItj*!PX!vWT?R>!VI;eQWd1+Mu8ZZeV`;?uCe<lA?5yfMwu
z#)miojEDb?P18LUF8saxgX`I*Gu6_76GD8;usHx4!ZzpX`M0jnVzlzv^bD`?GYRPY
zOV5Qkp8fb7GR=7okfBi#=A2hPNWMw;B609a{lZD5IO;q$o-H_g=fAqf-^rtMz%TE3
z+l)U;;X0D<ldy9#^7Rh#*mMM==jqWlT{V4h|G#rLp&dP;x9ik@eExy!xbDG!+h6|`
zt1ulGHsD!?^=i#(5tM`uE~&XvXiG9b@IzAQah<gbV%wl=BV9&Pb-C0eDZFYhtr+%M
z5mq&EQ^FLU+{$-3FUy*$Rtqg|b|hi|EED6fwjjejs#}oyWr;&S0BSjm70Kb3hJ<uJ
zJ<fmH#y`jD`^3(nousJ5uL-i8FPX3-fw&}U8VoLL#R!+zvtw$avW;Vx+jB?Mf%ZKU
zb_qXxX$c*qXf|ZeTRQS`DYfa&8uMl4nn`=eP6*7)1i&BbnP5WvXk}8C9Ig(Hi1T{l
z<ij%>U+T5OJsPc203?$KI`wiDu(h^C9e<O7&$B<)ce@E(k7WU-&Hyu?-wk#iMqa{D
zc!pc$Q$4KbNtZwHqKfM!(uaU0N;H#hH1k~PnxbsykV>?7TZ~MM>3AzL&GJ~2cq=xc
zV`BCZ4p4~!OIUEd%9Wp{h%LSrMu603;~s+88f9O&WYVPKU0t%nNjW0U?^Mw<{L|kN
zjwxKPR8ftPp-;8IvZfdA8(L)9Yb}bJ3;Xtf1!~gAEY^a_{<?6EwcN{E*1R_o;22j*
zg#g2sI$=?-VinK`*1jFVHBOCbN~Q8lEzh%_3|c>Df%axRSRz}`0l0n6I|rIAo-bFc
zudiD|PdtEicw1E%UkOA1OskAGD65x9P#s|4TYZ6l{E*gGjBfnK68v@V`OKa>JP5{%
zzRY09%p+sWCl8nkbkCMd+vajj!S})I?87McBbzfC=l?l0SY#XY2PplQW;M)Ba;ONk
zl#UOH3fi)3=^Qdv!PI_#4@Qk%=K%x6JBVgRN!#b;J~V{8hJ@$7s2{u~T%EKRoxaN`
z@m6i&Au|!!{6@HJ(gtRn><qSOmgAwF2cWpSJO*~+GqJ7V-N~dK_X8<%Z6SH944$qj
zY3xn2IddDqz@MQ*Qm|9B8g4?oR#QjWtuV4zt{x8QwKdYdc}=eRlTJ-&-u2qsKf1rg
zjaU>no5-fnq$E=|Pm6Ele#*Z|P*?4#nWXi~UH1O-w#Y~*a00A7f9lmxkuOBe(qn<=
z<66XUKD^%rqFLuk&$>`5^A+!pP=`g>t9#`=S66fZA}B^IH;R92Hb+UReHc%%c^L30
zSDf_Sw^w?_w!W(;*QBRaH<1zAi8n8K`+>cH^i}=mdm1jO$3SHM;`8KDczFJcp-RET
zARzK0)pPYoo2M=rdZ8*JNUs>Yp3<wOCM5P>xN@j`yMHY&B#E?Bw%`)z<ND(g1e_54
zDkMGj-naRm6BY2bzr(qY#{?f=5#Gu;MPISr@92RWY`E)Vi`@G0PszG}zIUrY=-2Mw
zPO!mEXV1m6FS(ZXDtpZ|_$SheHr<7rJq=UoDLPr7Xm;0m=(829+hJYf(<-vqA4C}J
zATW<TG7~o%0%x@->P^)WltoFUd}6XUwO-n%q$qwPOa@6;)f#ol6hA(?zOL(*we|c{
zv|VzjV2xSDl62W>hgn%ZeW&K&RwEVoPXrjWhz|Nssxdfl|GwMO^a0_jd~b@OOQ!s=
zJk{A3+h_pwx*JXmKIEX<^`rxjLKw4F+B1$n0BatS0X)tN401tDrJ+k0GU!{2_qzOJ
z!o`+&SeGsVnb}li0%dZ#h&N>HvG&Rk2pKE{jCCk=A;&N?pYx)lCi&N)Z$`#Zb(?v+
zz)L5lPQO!Bu$=s9%aB^h%UI4&aLtT_uO{3>_TT!zJN;I>Drp~lebI4|2<zx;EZ9zB
zZ5T92St$KSuBu33`w@zMCleW@LFaqXJk+R^st2W4$EnCJcH?(uJr0N}9U*u9Xml8L
zb}*y`mDVv%8?bHJ5_aJ*%I0)kv$fF^U!*BrIC0EdJ!zaBe;&el2?O{9kVRo01kHT^
z5RVQKw6nO7OL|ut9{;&llc?R}78rk;$GqQyMCzBP+;3d*H0}tKi35lKUBp#x`-Is7
ztC9`=vi}`c3WAfq^M8F6<Ra3fb7fZY_VPth4JD1+h!MwKW06m+@>j{02iGZtCe*95
z^-%4Wj1{LEZEQ(P<Bm|_F>uj^zv3%7KzyEPHhFQm!4F_`kev&^HdT|qJ4wc8h*>qU
zDKr+Ox*$Kfc1qSQ6iTipxi3z!&Z=+y(w%z3(HENf2td=l9qaKwnzS=-0j8gCbai^#
zfV~`Z2@4XqjB2G4SiJ%DDGUeq92`l$ez%M3*nS6m)l6x+rJ(%H)$4bvSA3;Ssm%LL
z<xNx(2?<twk(6LQGT*6@(T>i+kMh&?WqDxgowEgI;0wymBJTxi1wS(f7n%V-@Rn28
z!G<|yw{vk43p*P?M?;(<md%MM#1>!Z^XA>UUOhU>U(C?LEZLR;#a!k9f2TrH`JQ3G
z#3>jf2{jtsEVLL8nscM?xwz5Oue%)~7jRqfxewi%R;f$SeyL3pGo}1v&mPQ}1&D9o
zKZ^ORSqL)1R$6Cwp(ip!=hxvnF+OXz*$ORbYWLH<Jh!j&&(e&cc{EfEFycz7#{f%3
zLn!ILFIJ%NW)ebCNVYfnHtt?9@joSYtz&EbPz*O}{pJ|+*%!}|-@UYsQjhBl_|#2x
zW=E0~btJksqiHr~#QG!kb#%6=5}5IAt!dz^tPDrDe@_lOZ?9ti-|7BSuXG;+R(!JK
zzZM+3hR-D}!cx=n<N>X&RIW)RIDv;!xtGvMNJxRwRb<+N(2$nw*Bw8xtukV?<br<5
zO1xoDzZPn_3l%ouUTu}Ne`DxWe^69TVnoj&!^k@LB7ppDiS!rZLSx&3R7oED11*ML
zC*>{ef{z3xS<aHdw9sCYulXOmoOwi%ho=iRy@hxWZ^p|F7e(yLui;apBjNHDt)dt<
zW-)v9W}>QA8u+<X%3tM%%e-F6TaYW4&f2J2@v+N?DNyrFG8<_5IjL9~EVi^DY>irV
zd53G=;Zh>+t$kAf7|Mnpk8n?k6osRU3q^#yy(4_n3(N$lTrLzqp8cn3fa^t5N&_(I
z5ft|LejQZF)-5+oQ-_$$ACsN&85;qEa%AF;vUO1jtqPM*O%@~p=#G?elW)}GTJ|?&
zH7?FuRF<hJR84D(RBh3{X{jJ!Gmkg9ni-xK<;>iA%4>)H(sDn+jUuj-k}9xNGyFtf
za?9_KJ&iM!K&;oK8OFh01u<*abs_j}rv29%Pp<J?(e7j{@bj4LQn_yo{IFd?b97d9
z=`)KWCqdUet3=)!w{)vvgV+eb2O0i%d>tS!P<T%d?1vA!;5z^E_6qBnVi^9E{Rz(?
z>js-EE68A;DQWaT<c%uSiO#^;wy5f;&WhTq6!O$mdqwvH$ulYfdbOb&VM>V4uyQbP
zvz4XkR>)#dmWm_6ORZe<A%Gy;sGvEm?e!MqqcGui9Qh68_uF)Bd_fd}PbWLwIx_fG
zq^fLHgkV;RWISipi$ke!?r0KPQEqQEjyDnPoDU?J#d2|O3&5Lvrz<-^9jtMBvMPc;
z5d1iu%WW>&a52*VD%{KD(_T}tCh_4zMk{|`Bx8?se6HXZAp;;(lt7NmBR`Z`e$d5w
z1@#g!7?19)+ViPsEpPQ!Y%&pl?nT#s!))saZ=TlS>9>3I1m<p})eRDtpfugw+l>Yi
z5RQ_B`iOY-Z8MJUV}jF=z+|B6X)Fn|c6f;r<5i=1yKI&W7(?H(@-1y0Q*2u{mO3`r
zc5i?G8hnb^jGKH)H63D5>*Yvu+XgI>_%48p={MaQaNEf8I^<LY2t<UW$0@xH%o+*?
zdTlEKK@*O31`GV607Md^(6hBgXK(v^3cx`c@G^BO+8fDz-icbyTj7=tdplG<od>Wb
zZYqJ@j=5gD`h~`aGXEa_LV@yNCECZIO9HRY`M`BN_{;ELUlindSns1~OXWj``}UD=
z%Sjuv9}|Budv-&<#D(Y|%b)E2`b~uK<=_0@JPlH9+-#@dVf$}$jqrGvWg}dK5i|=t
z?~*E%GrX?|nb>v^vM<dtWCsm>4qCK>0V>CuXhCbjAuMP%?rLz{aRE>n$>53vnL-Ie
zG>poZw)1wWPo-+;Vr85e_1FrH!v}`$;f9~E4&6t@ubL>z_e-4FlTjj7>=C}eZYtk>
zS$wEzS=adYJTwfrgpKKOtR6DNa^Lz#(^@@HyaePTuwre8WVW#sOSCLaf5Oup0iDDx
z@-~<j$Ez$$#v!9<d5UDDtlAuU0y@~^Tm^~=AT!7u2+U{SDy?va8w+A|_{|W+Wyj5A
zH)!m9$rXk_$7vhqmNAGj<7PW_)^!DTf`Rerqxh%umiErm(xEFpq0i@JnYt9WQ+sJc
z(m|jvpoA3ESk7KVu_Gcy9=)pdt<`JvlMA(I*;i%KU{Q`S*~QQh)jNt8XDDVzfD8y8
zU!*txb8GOrwATKWTwV~!&KAxq%+rDZ^NHfv!`Z(Z>;t<@)y7c5o4~_kfV<9s{)US~
zop1C)&5j5f)IY?Glx1N7h=Pgfo=0`q6q;nv;y7kS*OGx1mUbOuWTUmZ(0oa*p2^5q
zCV_m6Pps9w`M=wF#sBYEhphxe8Gz|9<5Q){NS@4IWOihnG1{6%HM*K^eyUB_!TcJc
z3zufu0I)kOU2VYph_s%OnFjGOdzg&wwMZuDiLww$s8{4!Ov*Hx0fN&RiNB-7e0Z+^
znuSRMzetI#bEu~*CX#jC!UmNvqw7NZ!q*KfRz+70cw1$!<UROHXGF8(%_<+KYljN|
z$N1StFbOHjY+VPOoB^oRQDa6<pN~U!K)C}gVDCg7X-tOOm5ZTi+MW`$h0Ra<9!q_5
z?-Ta`itHQuxpFq%Z{^pb9#pm)JBi0jSK6bja1PJ{x?zgM$C+4b*$MNe8m+B)^}jt`
zHr8VGjLxi$JpY~HI_j&4_}fMov)0gyi;uTq4q2BWKF?ct^*lWRYTE5NW_`C$e29K+
z<FpQDhp)xOG~vCxXHb8K>vWvH&+l+s>=Ok=yEOo8r#)T5Ygh7|GvUYYUYY-xu0lO~
z(!hu2yl4F_>8S`Wy|uT?2$x{OudTtSmb~d&K&rjI%Pj&Or1)+<1-OSl)ozkgtM?pT
z;nsJ!<@2S-MOae4;m1#BR@Dg<uK7J1bGyh?RtECgZy<85-s;E)3oYcEf%Eb6=cT6j
z$fC1%QXg_v^4lLHJ6hWRkEySYYw~@+w*W!Fw+K^8>25}2ARR+m28@y#A&d}^Qc_ZK
zG)N7p(E~<Fj!x;$(VY_hp3nEM-|NNy8_%BmzRq>dxz6>#@6Nv!zkjxS(sie`nJakV
zJJ-nz^*;vp`ej|?wnaa`zyaeU#!zC?sEFWa&-TG(cn2rN&e;)hB(F%AOyLyWcS3hB
z{b*LMudf81l!mZvgpISsAYu8ZZ=<ys4LDuj$KKJi;C%S>Ga&TaRm9Lz5_=?#2_51Y
zY(gI6h1i^pr=nu|p7!g-2<%vA5%4?_8|YqdC|6&0Y%m`I1N&xCma3zec)jrxz#+q*
zz)@suwPBqCjD~dg`%Be)0oI;$hL@vl^L8E5KT^83RvX?4G-gwHS|WYiNz|#4ZLs`s
z_?Pmh63Z78`px3|3?;|XF%|(#FgcAehdalgvIu@;Hj1nK0is3_QjfmVu9{zHGd7y%
zB$|SaGDYe7R;Ph@bl?^NB5W{!iLYFe>wE=@;3do=!M=Q&=5E|S|9B4mRMH=$co-bf
z82X^yiyvX2z^!Z`z&S0)B!3bOd7ku>lEO~gFy3EggHoTHE#{*g^?8Oqz@l~+BW9am
zP68#DJA0HRV7odqqiPSi(G2R20fvv2nW2C3F((vHtu$3XnY_5upy&K514&kLPRV9A
z4(?NdXFW@P7twI!-4J4oq6Sf`cq+II&vUJHKLr$mZ-gJ(M7~s7Ukv(PMZWLF-*y_@
zeVcKue7&EEZB^u_d;6VPL*+BFQEi&AQtM7<t*zzfkF4c<g0t==iDgiWw8`i>oI%Xb
zhKBCIS8-X3_re1A66twiqJa4Hu|5-^CUP=a^|vP<L%pxG@3PEa0X~-TlZraV_d#Q^
zo^4^Bo(jU$iEN|dEG%`!a?E06Y_ZC$A2+gZnN%Nmyf6W}|IoYqZz=~i`(GG@qV4_k
zgj2*|4I#wK`$^*bb<8dQ@>Ul*B@3qo(X7mciD5}zm5H%5bHsKfE{txl&Z)f6uRkQe
zPEFpkP2V_98TZ$lV~J*YZlLw_T}C(%*Sz~R^|3+{!bb4NI@wBL7D?ESvF#+ZR~+=b
zMOv`tOTFt_BYvxPwGB^xad3J6%o^hUB>4mH%5()XA4vC){$CviY=VS!?Y+R{EPmzL
z<BPsA&IC6L13^bti>pZn{Cu2jmDbLn?Re`(U0bq|5@E8L7LbCL>T-^_>zjHL;W>9?
z{R0SLE?d>DaRAayBF^XisB+|8Jgtgw+k3NY@4@FV>O<?t`{i#OYG0?vu@aZQu5X!d
zc#<9g3OW8|KgH25Qv0UVb7MLXkZLAon}D_(0>nh7V0PKE?N5b<ebouBsSHFJ8`Xdh
zX@Zj|k$x%pH2gR6l>N;C0M6#UX9o|DsqftG|Jxjt@H_Ztz26+*pgMY_eo)geHBaX^
z=;rD8x46$4N=QP&tx=;;C0DbN{2zzu>N^|CQ7wUA1asN<gNFp#1isSbzOfMe)~=Mz
zDyWp=f3N52<lM^GuVKp9Q_8PlVJbn#GLQ2YO>PfAQ;yL4Ij=JbyZzIWbFT&b<tst7
zC&*J@*k1BssM-jRRr$u5=ro@^uQQ?c$53H|Tp^n975`kG2Wy(#9BBX2OnEzM6^#XF
zUP-c4j9uZf;+<AZ9zeNk9VRNP8|lRdQ9h{tynZkt2)<0V3#zAh+sL-3q5bPV{X*_U
z8C(K{Du}UQ%@J4|Q0}_rdE)$aI?i4j_kpM<%6dtH%@X<)uf?E56G&MeC%)o=<f@8S
zC(|jsUx4~%4lZ|<Z0i2GIpXYQfMF${ME3p`<Ftd2=;jfD=#TJj-pfs(y@g~uiC5u-
zx+`4_oZsX5q>Jp-s`8-t;#!de_|Oo6_e&IkJe1SF9m*9F(I%27+{4|XWtkTtpK5Dz
zLb;i}X&{c2NkY=`Q~2k$4#`hfOu0bqyjJBqmUxzI7*X{Ng@Ph^;Z?3-*(`ed4sD=r
zwRmDITb~M`21;V|KK|-Xq1)pfvZcWc1j)oT6xO-X3B4C4%;=s6auXg9Tcpsy2IXX#
zIqf9Tp*hPky@giwcul)jo!u{>V000iKiCkZjJf9dV4L=dLs}gtSk9BKnIs(UY=Cb0
zxoNr~IrHjAhNE^o6{=v^*<mDm*_E;5P^Fh~fP@Q-&5)0#q$3j2a-1aG$Ijya9qg%^
z#9ZAznn)#JaJ*)k7f(T0Qwl~pN3ENSJ`Y)@E7~%^+CUOdfmXZ;r#I170|`@cwXuG@
zE+5Mep{H546&_pvydF)_@EuZ`#ne`qqMRm3a>V)=UC51SC|C{Al28j!3#_fCsMy+y
zH*TZQGSD6C_03RoTh*B_b1v`lFw9@muX+_18P%&dlRrawZjV_dAW>P0@WU5Clty!z
ztCU<22mWa~5Ni2$Xx_U40?@SPW!+71Pg}q07^0XG43+;gwLUe@J>8Yf(p|Wg)GhA9
zJ52~%h#gI~<hcyMFy3*9?q}6l!2{oTX;QwR4@`4`VbI4mRlE8$j;kEUO65eqpG$UH
zv5<+?y|6WU%H^OjQRA^{6oMOco7(vzjw(a#MSjef%B&C$yX!<6DgNwj5ejZZb7Aia
zOL$Z4Vz&5{g@8Z(!B5fOnn_v-PKdmgxMzP|X+@!$hkpj*XL6zy#b#`Nm-;w|u#`@2
zU{7d4j?F{XIopRL!~XB*UmRbzP--Vmwg~%XqQ8CIE_OGnD7K(jn{A$7c8xi>Zt+iL
zzw3ZoJ#>M+z8E34<V=4F3qV>EGyE?b(y~`@#a335OxajiS%v48;cwoyEf2-r%B=?H
z*Bgp{8$R|YCReX-4fyW(UFNtWb6PZ$D&9cElRA2SS#UI#(q;JTt5M?r@@+e0V$o%!
zaVT!O+QAhKt-O8+mBU-QI4oOEd2R9cN+Z+yLZLq*b%`)pTzLn0Y07!A5+Z_@&oAy-
zTN=~oSe$A(v*$6$3LDruCeqPP^xhXHx0wJo9E-Y!VV7-aDup59>Fj;6vEsd)V;h((
z@VEOWVO`nr$F>C%YR{<6jJxjY)H#qrPKGQav-#pzw~`q=o=X(}Mahg-QiB}^+-U93
z0{y-qy96H#qm<SEd*;8Ug|-H!B4iI0M0DwTjz7||>_nJbA)>M*Q|eO4tBoEFjvP9K
zqwI!0JAmK=hdh?ZRf{~Ku~jN)p3uC*{fYm0AVw-qZ}tA@qZCF32{=(ndGzniITg)1
zdE#v(7KHTSbw)H_PEyG|H3dnu)SnTv%c#eHNZKm1-W1AGsvyCQOxZ|)VR2qL-wkuV
z3-A^cl#$794#=P!MPP%iB5rqet*Lwl&FQS<JQ+Kx3WN{T=tfyexI($WuC2x9DXSOA
z|HQ0f<j65bvTn3w_Py;wtH-O<l7~~rKhdlZ;*CpCV{p*LMDAUS4M~T=W$*X1ExIs-
z0EENVrVbTG$on?n2Fl|~c4~TN&uD(VvhO!aaII_Tc*1l$3M{U7krKu>by(q+MpOd0
z+dxS8f7aP<$-&f>wG7=P6)Fm}n1Y;b5fodAg>2oT5!LJ~o9VCwcHVV)EhmcHzU(xP
zXZ?Y88A0W$pVz?1fj{6L<TTBP$h$_Wclwd2C)Nb)SxEv4W}V<33(B9OeU(qD95ybZ
zzBxi&X|jv#JVCsQ@0J|4E$>(lX5(jSksH2Nhyf0blM5}ZO9s=17lxA$tp=8{9+(Ms
zsK^F;iim{Qn&$4D$#bs$t3;H-ZB+Ajsh?_$vTUjOg2Ij4Q21`*(}d1aYgi)#@58as
z3pEd3+I$O%?ZS=SzgiKZ#vjwJjzq|{(J%Wh`x>c3j+1C)2EN+U$?H~5S9rKq;&Skk
z#nmv<pgUL}nxpHMh&edZ9m8IRo=0cwf||zVQ*NI2MY1!MoJA?VZJpyHpV9(%2AN!U
zkCP%W)`tgQQXk2NhI99X;UBLXN$EQ+*ajT?{P6p`R6i?C>|tIXj(<MeJiZ(gNgi?>
z_+ystFou?095rAy!91W!O8*&0{Os>*ZimE3bnmb<V?Xzvp~F!@{YT@;Oc$Z|9@pN_
zv`0N{9aWH2fu8a+m9^prwV(g{IVMtHNomPuJsEBOsz*g3i`2seeS1uLz#FNl^^qxZ
zWhZ>?oAix)Q}X9tum1B*ySh8qw|>_M1v7O+_{B6DGAjJNl41cFV&W_srS@n!caf#5
zt0&gC4YfV}Gr;n>>iu}QiXe-+2`9s7!=g<_*&EfQ2b?l`nz~~`_=ds`4rHNu)jMo3
zHiaP5@|q^a!Yfn0QgbZ6nyr@SPm^VMPR;%lm|pN%dvb*TafMuLRPM@Nl2*fCjB6^r
zY^M(bMV<=pWZ`-qA`vIfK0pG7;Hs-E6-ng`mJKCg3lwOWR&=fq+-B1GPIbbtX$&P#
zyek{8ooxq^AmskU$ny(wT3ATa{>a%{*_2bCrC>$^l?)S4o!qy-gKG4sl<Zv3?%mPo
z(YMY)ue7mil`Q!jEtP?7)6P74DWZuVQZ}%i^DI`@biG~rkS|t+h8#qn%iO{nxVj&4
za6OR_p`V(O(f7$<9*Hb2PQsZ_7xjdptjmWF_2Cj1m?%Bfh6Q+%b5)m8ILY82oBAU<
zaWbvt>}+_dy_=D=A%2`=*&;?mhXOz43i%}>Nf!6bF)o2BWHfLL9pUAe87v29B&J7q
zWvv%(Y8_rmo8MUg<vZzpb%47X(cV$*X3DNI0{Jlr5e<rR*hNOMcYG>--4+yfF6^58
zuER6bIyH<%lV8vZVZ1ag3FVtxZ*m+yTNOU=yAgixEx&igZ{!I2+He@|A^l@2okjJT
z?U~SV@JAXhjH3EIEStlnFu76h6Ag>DC-)UKb+^9dle`-9tRR!1rcIf=$k!ZTQKMGO
zCL0zpBO^EDs@D)aK_Lc)r9loIOdp?E_k3d`FBX=m{up*KBj{!jr94v!$D)SL9X637
z^Z^-;MSfxh+{38!;E}_$j#hhTwE)zXUWVEGLW6r{ZN)4Dd0ivbgYHC?@2V75G#L8a
zczN=X4kPd+E+yK{R^95JN1*I4&1S={C6;=di3Z&jL)=JaP;^KTN?Em8;oV9RRxgKn
z+bz<?q$#Yz+>zq=WIo=G$i=GMm5?UN^}c`ATiK5}+egR}zQ+z#y6)P{+$x_k0~?&G
z7HvQ)Nk<H+9R&4pNdmmjMIGin%Lnt!WFr+<S0zH7jl}_U<RTF%W&UULQgp^A6L{=F
zmbxx09xbEw!lLyretg1QoW9>bT1sV}_sr^5m{6Q>op=8i&BA@zm>6|=pvoMZy?8su
z_Oo1xgdBg*>!y<^c%w_!v6xg2i9Qak|M(gAq>=+K^#Z!$BBh-TAg+v95EY*2#ScgK
zOZWG?SrbkIgNc$&O*xhGZW0E$7M|V9RNn_5mtY&5(oF9)<vhtf<?d{Kb;G>yu70y=
z{pzgR=+13d^OvyHe8%HieFX~<d+o?W)t}ib)A~pmoXF(X1*w6kjMl-Qfrv(EgynaI
z{Rqg!DgC$sC;+3{_NU-K8>cL+3d2TX1;~2mkjavAb)=l%f+dvv3$K@zapA)nilaik
zVC>&x77P8>eF!A1<y}j<d?rVFYd@wp$V`@$(5+uh;%|}4i!(IqC+k7XpLyQ%zU21o
z;w@G$5hdp{t5BO>A+yip36j4ITKDWP?{)Ac@CkOW3Cy14YQVif7{<6bM3NKu>B!e>
zBdP?ZHy0$~BcDkPGLnQ0G9VL)QK-D_jbOBfnGhpqj{zHfaA%A3+H^|<A}MEE=+mF-
zFgQyEQdr@q!$y^NIFX)V+V)j0w55OaE0o9GItj_d8zyvA(fX$yS^}q0PKiqpYCM9Z
zN3qAn&>*3Qk~DDW6HZu;nbIL<+I8as#5>Vhqc&1u>Q3iE)0I`WZ&+MIzg)cPgj9+7
zQzr8?@e@q%Bn9*LhIIJPrewZZu!AiNwB-?z3ka?pMA@9d8k>^MV@a15nwEcWSV)xT
zvjSrr2M*lvHT278F?)JLG58lYs%T5rYPKscXxneGsU45vs#o5<8tXk_;rA08MrwVl
zOdTmgp}E0GW+F5eWaFL`=4wCbda>4gVMTOMgqh!v_x|kIv&W(*-<|BCr!9GR^S21S
zOvMET&Gfzfw_v2Ol_vU=-jj}uyD>!utyR`Ba(gVEoUU$)Of!VMsYqtil^oBWTd<WM
zO?WWN(=-L>Sg#V>2ROGk&M2FlPCLIq(%>eTc`_;3rgJ(Y(y{mA&?dvVR*9OZz_!Kj
zg)b-Z-Czpp531=u3RH(hh&x(R3lBd-8~bvzB(nm7`weDL+Y%eLZ9Wo`@4hCXa=-o3
zK1ehunErJ?!7>H3rUmxZI|u`5U<rcRI<!_uZHr!bJ6iBAr_r)M%9u*66pm6lqQFIu
zR@ZjqF3xRvN|(>Ad^=_9uSCn}IUTdC{vGz$OyX5+T`5~M1;tQ_@D48S;=QDO(TYi`
zq$aqynDdgjXj=BW@=!q=$zpdpgQ&=loAz$62qE3EJUwC0et;UB1gs}rJ#2D1LX2q+
zQwR54{;9;B&%arn=UD1i_4!K{zTkHl_+#OP3lO|$(!3eH&dN{E%fan2=;LEh`1`~m
zZiOL_i_2KtNPt{>FXG)%;kwTUBJZQ|2ZKW5N$d}l2MjTi4X5uiO>Qsfm#a)u;{9LT
zm5c&rLJ5f~1Ht^sP5#^GMiW+a7IFkt4xx*Zp(fCW7uWwzb}yPvQ}&l8obO)D{R1*w
zkKP?nV-<|Q`<9?>c`Q}#L>jkg&84Jp@N*WD;Dl`IQPPVd>ASx7VIlP^(KpJi@9uhZ
zP}`SJJz<WqOD^PQ8aNYv_YNBBX_xU`*_>jHN9UZFQl<019{&Q`4IRq28M^J3%Enzf
zZ_8*??hX09PJPwe!Lg{HEoDC)tGG8Kv^7u^27VJ>TRbM$Vx_z31!?>FOLbkdTQ2e$
zL^dBEe&3xU*GS4hAToFKJ0S$8#X+eY_iJ1Dd6SZ!8ib@Z*kQ!sdjNJVsaJMs09b%7
z7na3RB;E@bqtqT0wV)W)=%QQ@wUHFXYHAl~R-1(x<t)*p^$G7q-6O{nBlL#LELB+G
z@&uCESzxkF{ZFnBK&9I>M5BT5Br}89`}FMuSEGT-vDzS3B&&FQ?JsC9wI!$eBFbSo
znS#Ht0I+b`2z}BcFw^`niE7_J^-lF|vz)B3nFhsitYKmes3n18jm`Lp`IJt`v+Um{
zuhRQO4Eu45u8$(&lP+&4t(jgl6@ff!Oya8rBuBqW1P=g*RY-=f$A&&XXcqneYyKol
z^5^(|rem$->N=u*iOH&ZcjP>XL^GaBv4_~0i`=P~DhD?#ombp0?dnz?WM-R0Gb<ID
z%vN{6PE|Z3Ca-?rcO5)IIOQjCe)9}1_UP^Wg?xh#g|om9#GuB*sL*^m?L23{P_v)<
zo$;2o!bNu9Tu?z(Q-iZ5Shj&8-!`i}xazR}2Rg{pJ&vkQ9zG@KCx)q53|$?VPFBJk
zLZ@2Afr9_qG2OIHmkB0<q(-q@T<dS@__N@_|1r4yS}bb|j@(vS4zX7`I?g;KKGxD8
z9OYQ^h>2N!*XuZGZQ&;?0hf=MJ7Xa)D7Nh&;nw}2^Fb4_PSO51Pft&$_mlB2_3`SF
ziL^1t)2TNkojmfn<+Cg%`_IoGeSnV9^*=?w?63P=yJ(T2g%m;OElp=bLMtiShU$jS
zJ(s-`xr-q7W6GqXX|y=TnyCA12ozH=m~jtBEu=13=UClsIN;{Q8<*pmy?m{vVAfjQ
zMu4k~$5=ZrsKiC0RSlE~4?e#qWoAf^1FAmyCGW-+c`AL#cF5hA-iOZ@1TWDzv05qR
z$&Fegn8U%8mvF{hy7R7$dRw`NQ;9OPO%>+?QtmdasV}ZF;?c&J%+RCh2|bVls<^v@
zyMy|0eeaX;K%{lq_?dVTt1B!n9p28v1)Tr;FQ#pdOJ7&^&W(OmW26TEG_#u(vv;C<
z$X(2_&42H8Dvf;=7;u&~vBFYWib9(L$u6cBZl2ce+2tGg@;>|OcbVz)j@~>=$GbVZ
zu%M7+;mHRFO<gY4amu-LMy&J%CED1y+xVFk0mI{o-3-N<%5LXm%G|o=m_lLGvMn8V
zpVVq(E9aD?8*h+)JkoZS>*obLUXZ-QTO|BZg;Ault^zu5yEg`K6L*xSA!AjX2_rJv
zInA=G^hmC^QwXt)E0%&O%6RpO_Q~?JosI0>?W=6ny)UA{1#wiY2sX6QjF5ncV2_Pk
z|1P5TTH3Dg<JcjuBsvIX_W<4kri8!ORhGd|Rs0a6=X@*G1eu6y+ZXSfq*tUPqSoZ%
znGkMd-YT=BZe^hdv%aYd&YM-Z-|<}pE(wA|o$+7`3J_PL3bTq<s~Q_K@J39sq0P#s
zKi5`Fw}V|j<IJ<W<LSD~!P&X1ENZO%MDYN|a}A6ofs}~^N1=1?W|%EcnioriGeBGz
zmGr}Y4;`~2;CT*an{0~5NuvTeYX-Ao2g+2;N9X8FAx$gr;2#b<jxa58AFR4=0*gEZ
z+OyHPX^^%p&g9Vb;49pgh8E^{={>#kx47DVi{?|>1JV?78j*7a#`6-4$B9wMO#Tx%
z9}yFdQr67)&(7exum(0hu3+BOM?NNN%<L8b{HFesvhi~8+?%dV;}@9N$!1FQ%^B>6
z@BfFqpXX(-JskCivq)EZ@)TorszrEKj)(fQ(aOTzqJK_lJ`hByO3}n-uR4KT9O;Ck
zgj%bj2TXZs=b#y-4?X2;oP(ocCg~(B$x}sB9THvc187urRJ6P2%O*{@GiS<Mo|9G*
z9znNsDmQG;n6ef2!OQTU*uvHOFpi}BBJ3Ud_cHk+)K@9NwRUb0?dDWxX72PAX!&1T
z=JD*}+?>R2k*EqS=a2-Klj0nr<tU79)~3`EO;my3RM33yhwKyZ1i96$M5O+&^Nt!)
zI@z4@-0v8*^xihM_i5WsW<&ZFw^oSG{*`Ee6@$kk2#E|Wn6RJKzL~g7rxw}qVy<4~
z%V7SP<ptT?+mCwJQ@Gyxv&9f2Wn-h3V>G{yuHi|k%wFD{<9xrXj!2NY>OAzX&-w1#
zw!U3T%DN5at;LS7I%PPYvn&EY*-E*^aIm@@5;q5xO9B;x=|t=KGCm&uHV6-PIH;SW
z22TIu1}C-n#45VD1;5R}F{C0S(t)N}lm`G6Lhm2V`~4$H)JySR3}Swb@bV{TNq9I-
zi-~|SjftScQmWR&JF(BOqBt1^&DNimv8w>d;4Odu;qS^fN4($#y1UI=e}RnhwzNL&
z3a$1F`oR`K)_UoZ)vr^Z$qH3I#v}lBT&DP&IKOg*k;b!X7Lz{-HqJ6RMPO*g$Luk!
zE_lzToo$McZULn$gq4YNZyu_cm*x6}s^TS!e4Sj9T8=-0WjNZ5i~hHkTbCLHhsl$&
zqNTyWzYn7e3q-!j*h%VG_UlI)omR%6{AW~iQzYfMVO3dn6y2z!Df-@@Yc!-jv<nFl
zi{hoq$No?)0UQ0|8l$6%PIpoeqI(YU`_O1MBbc*!TW|1Vj>8j6AkR^}EC^eMe5YN9
zQA2zoOC1Z2fK<M_bJeQMn-t7xhFTmfyff;Sj%J@~DB>mB3aeiUL`26?t{o+t(Wa>7
zk=)%()a@7NdN`UEYp_`xWLiFb=wVUWZ_o==kfRaJ-%HYL{H#w6Mj6yq*5Sk{B=7FL
zv%FW#b?Fk!*v)52jMZEEg>a9wkROaQpK^_kFx#AdF{lkfs1zuVS6bQvzb|-9tV}JS
z8X16wN3_Lf|AN;N;wqEk7}zzZcCh5H7EG&0-xk?c&{XiS_pqv0qL*$@L66Zj96ezR
zKAVBcN#!H1Z|jDekB8j`9;eKT6A%5GN^NiVl<yUUIy}1FFhw4O8h%uhT^TMR$75fx
ze_JiqZc#CuAH>)Ry(fvX=7fB}m--yv|NjYZ-;;ID!IyoBX#N#TnK@-X<atH><L1wp
zHqEDbtcu5hh_v$Wtm?|6tL>}$03d1=ZC76XH5zLPtdk42=p!MtYFsIsb98FLuS~z=
zeiJSm9<NTS-x))VwV6ZNHh8lIqYG0;1wLk<kcfz6IE=XQ8f1hPaHnMH@l!j}a3$)>
z;~R|dHeA7fpda=lSWi`obkG|Dyp)vHMppQY&10~Ba8>3FpLNGP?eJz}hOVy2iognR
zsJ0a!MNyWjk6NVd8M5VtnYi$z6O9{4BbsnlA4rMZc>VsJ!8!7`IkW}UjwFV_h{LI1
z@yqdPKHzEovK1%RvZtbT6IY!~?NIZN|J1frG86k<h<q)&d<p@DOE6`Z+Mc1FG@mbX
z^3(Ng{uNDyh0H1UF>!j>SGG;S=$!;qr;_D*+mjGUhvljHv#-hO{7PDEdp}aDoakzf
zsaE}nh?<!dCuix-E%rDRzXQ(Z!H>GLk<yBK!T?<U6c+M+E<HApYQJ96B4_3Bia-5m
z*pOL`z60l<5FGUpN*VS;FTP@!DD294UC-UshI22|Qi(n5-Nz%=&gMIrw@c^38=}U?
zH~&X#?>=?+{NhgdJ!-Eb=PNXRv<kwb9e1Ie;IR8#Ql7Q0hzw_NBuFMtr_`%bC#TBl
z5|bncdkQlfeKMLfJ9@8aS7g@`=?M+%o&V!zuYDFLT+{4OIFR&lS&FSmNbtL;felfm
zzSg%WE6z9$*j6DJcEEcC*vLObQ}wl`HJ}f80{YRa&1R}6Fs=?6-Lb+h-o&a0Fj*!k
zk-V4VZiOmqLClxiyb~IptD^Fnh$^MT^ElLI+P+tDwGo$XYp$s~jf?nu;<z*eYA!-Z
z$SH<7xF*}G&|FBHHB;u5wc33nRcufbZ>ilDO}<QLqq1BQrI2LJNnY=VQSeec2Buff
z=c(yJ#}cT_IQ>f28f4k`l=Cfwpf3Y=R1HoH4cSiu?oOG}rEg=PbH^%<q7(r2?5;$Q
zGH9k5a;~=;g}KfQ4Y_tb3au@&<IMFj%r9i<V=W-o36SYdZ!Km~Q#mTM7;;6Z!9B!P
z*d1xJWz*RS%$WE#f|*f|Apsnv&o7X<njO?|J*{!7Kax!)!)Nwd9igZ6W=9gp7I5E?
zYh*RDO^WZt##40SF<#@5pQ&~1;xwTl?1wn_rkDfYV_K4P)~Vwh1}dm}+mh*vxK*k)
zl<cTrow)r&X7SP?SLDHI;mE~Hpml0u`X8o#gCtm7_;=-(nfbRMbNS#P8u!vKE%T}_
zbETU-^E%e=*4lU7`Xb2Wf?#j19_J*!THkzCf7zXRt!#8&U4Q4GlYbe%2l2QyRgZnv
z?g`OV5rZgjjBKWM_=|K`s4yxA(vFp}9&3n!3oAOo7PE_7*d`k(F!AFWS1eg-33^ar
zrlk^cq*WoFvvZ6Jt`w(MWsTTKsa=Q)h#!?#N&6u5LO)wK_Yq|t0hINv1(>=ZEz;KI
zv|eXx)kr7vW%U<K!E(TBlMT;O^R8dWAb<@S@I6zD{FeCGE65VM_0OL3#Mpu7Is}H$
za~25qJhW*^AJq`@?X@j<YF`hG?{0zZ!$ZWTrG1-Zn;9kRD+;LzIBV2p_tFR>0wOAq
zA&Ha)6v>u0(?S&+x7BR0g>~?Mn$_fK!O5nS*=gBabqyn{p}Tx2bfs9NpljrA6F(LU
z@7rTppBG;q;9wb0c_B4nm9usP6kmDWW3;NzxD+1%o+Y5UfYMPC6@qGUY<B894iOIK
z4Y?vD%g#2h-W1U0<aq4(*~y{h)`(8@Vn{XLh9-?qYZ~3{pga4zUaGCKgWzvexX))C
z<KRv|f;T-xbjDRW6@O;sy>s@EOWB#@+b&u@UR89PrH5mSMm5pmM+2c^fX=SUHYA{`
zn0PeoG^%aWNWBZU#69)K!e=X_ImWE!GG+hfe@*qd-<3Q_SsC^rqLmIeA~A8A6i98x
z<o{bBu??D((E3mN7mEy}`$p2fuDIGkO}0X(O-J8N#y}ucXjzB6-5T~<R$HRcGBWCU
zGYvY+*^MleVpK;B$wuIhR<<yLs#qV{)0oq4?Nj%3EhdX2>1?ev2j;=cPAN{srW05c
zG4JoEC3DL0vFp`ezaLZJL^o^M90twnw=0HEf7vB<QWFVx6B@ISl;nk$cMJtpyJJE^
z?^op(=N|lV)3>X0eU%Dl3PRv2ZM;N%<awRQ(v@S&g`>oi^5nP(+62iYGi0=rS=8t&
zvW+^w^?b(~tILtfy1FyvePiHEZs*iASecfD7zz#L+y2NI-?#|wS*H>Xu%kA5aZ!uu
zW>HI0APBv;qc}CCWc<6IaOg<BxpvyM7I+1KSHoe@iV{UPKmn|7WF(D+Nq-2rp$Tm-
z;u1{OaSQsGeB#q>5q0c8%oJCU%ofLw`=b;dc8Uo{+kxrffjqIXF0jNEhLt^z)jjcC
z%T!VtP<WLA&Eo~%f{?uJ>%CK;QRID?9j|q_r{;=`j}%s6G+8J1{v9dA5L;uI(NTvE
z`AFRto5s;SwRe@qqqW>%k+&ARYVKh_c|@I>&-y`+mXwCi5;k%SL!(76XZPZp91$xl
zrgZBk(%~H3WxN3^o;_c1OaG42AN&0uH22#f`E7DlZL;FkeC~DGVR9`$e~NZL6n=9o
zeBJAJ$zSJtBUw~rlD2=z-F(FD{qNlG&A*M8H_Y=_^XEs);rvI>Zd*+jy_)b|7n_;4
zn-ixg|5MUBtv_G>nJbqIU1Rms<TH!GWGPf5O;v4K{)0(ciDcxBy5`o8yONSZ0?Ww^
z>C)&IT7~nHdZ-5Z`Q^izT)n5{M8+5m(FXS>R*anaXruH;CuX#M?LeNMu{{|THwg=)
zSG6wTePW44J+YE_I%at<-VQ3-j<Vwmui&@r`eir8f|PtE-};93?zadhnD)sxelhcl
z(j%XLzs8e?u1!IeQc+Xs3Vm_@N)l5%FzwVzndtBHWxLRZH9)(@NP^=-nfbZZLB#wo
z_$(S|pQ#jC51rZ@sme|exH(}wKe<Ztk$W#^n^UM@J9q}-os85oqY93ZF0egAxImQi
z8T)Xor$<F}<TEgEmU#xcfHeIg2%BanaDOJgDYB%Qr40f)NGsgRmI3giTs$76KiHY4
zg9NDh{4krzSutRd?*>$u>fOWDyw_sy@9mk`#Za-ntQTpmW;s56w}7uHed!{Qu&h0?
z=yP5wU@zI=A&wu1v$XmxEWc}RQB@Rv{l!i$t+!36bkTc)h12bCQ}+Ovagf{l!-wHO
zRzi+iG>gA04d|F7|IbfLS`E8Zw+`^_2*Dz|-vuqyv+|Dj*lB?j^Cvn2pp(rHPg;@h
z>1OH>a=klk8+-o>=|WGZjZbbQ0Ki|UFnsCE=MgMGLr_u#_H&+_H!;JJi_D-yoD1lK
zGZS%t?(q)ju1-_0PW}P&XiMI3?OAt|fJwVG_eELA`&__V2JAf~Jr%8@Y9d@z5O7M_
z9I1%Nf&Efs2A>eQmEfp`!cSe^;P8z7p;Jh9aWrMT8pZpHhE5@~ytSi6gSba+Y`i!Y
zB^Nf$ffgQfN{hay)S=J58mo4e?u?1RiDhCn<6~KtQV63iJ;&k3#L_dOx&P5(?5D%%
zYoua)R3x3LknlxANmPL>0n&0YStx@~W%Db6a;Vw&rX3@k{HR++!>TDCfxRY)l;dQ@
zu#C_rVMMtBmU3)FkkSD$@oe1ASWh3DWN5EyjqJLu5{^oIDAY^Y)A1AT0#Gv{qLRg*
z=?W<TrD#ulSq}1qN1I)(H&0zdHf<iE7Y#Dq0{B|<#x>Ja(5Uztxa(I>tih<9u>Qk`
zv%+8_+-m&-w&N+j1AI=KImfK_h>x5m*Rp$w#|(t$YSqw2$xOM>tA=%Rlp@rg2T><%
z?PP3!|8)D;?5%Qthi{zOjFmejpxCB0{6f$Lu*B^x4Y_xbY<_CmUH(cWW)M5FHhF;=
z$lq80rBm3Yfa`EqQ(xyi7GH^T@3ipjzp*2`&3|(<`C+x`U+XCjeaCyEe!fF$;ZoiA
z7Vr1sKaIn)Gf}^r(tmEg7a@KZ;kVb1Zx<gQwLSv2ay<4ny{vA&6#K{a^4|sSawz2W
z3D)?S%;nA-*YNlA^eYl<86%-zSv1n}G}7Eg^-{<igF`<WjHGMl{1zrcK$Xq(b(Ukt
z{!C5^SsYhDl#j&kKhOMT;2>Ki@{|C99?&3IPrg-ihFc@}Qm72&2I&X0x0B<m?Z1zT
zGxOW6kzk+>bybvS9ik?6$7d=$jg@bA!wgu7Z9EE#E)^}?J#x0y<eKs&t%ycFB%(M9
zN$6SkKIsFuKDdu~B&`~QTft+c%jR2kpExNSSBQlIfM=iu%-GXox%7q2;>`?4uNFBW
z8E&;l<vtE9P9{8(ZTh75d!J6nr)W@db<l1;oC4GI27**mkezrAro()B0l_WS1xNs7
zcLJ(yH}0!pewKJ-4c65=yb<229PI=jvKXrJuyX9Q-cwKFoN6gsIf^=l*(W4>JJJoi
zLSE?G3ENCd`w8G*CI0o^SpSZQ?nx@fFC|O<J4?R*DLq(ioijbS-|ye5#2$nKkkigP
z1$hT8>Q~BN9bER!31(uY*p%dMOzzFRdbEJ})uSaz5A86(Ddbx@BDL>2JX$BYTRB~2
zBSq-WM*QjnJ15r*Bgf<9xeg)GgM{%%|Bh=`JRPxLO&o6e{IKfK342!~gQ&c)r;+K1
zR%>^*h*r5ycj&jQ?X>-?2Fv<3+tDS{6ellAP2KGNxCYbin-V*jaE~Y%T!KvrGDmw%
z;Dc3QRrC{w;#hyJgcdJQ%(_0)d@1)`sZ4l5oZ9bA>w}Nl-@eyJ&H7>4f<c?4B(}ME
z<PHV*=<#7n(cJd}N@R`UTUo`z+VVQa@!}+HB9YEka0O_0mPh5TlVx>BP*yV!Q(G0w
zcNY{w;b*LxPy&-o{}C{>DJ|odXBp)1P`zN+pbAF2Ff2pt0B|@nRREZ<NntD712|2P
z+7E}Y-de4N1&k|3@&%b9-Xo@5Dl{}c<6^n2Wo!&YLxK0$2D5N-8AP?%w#f*)wN&Y7
zB(P%gIC0g%_=j=^C!RuNGBS~=QKlxK0?NSp(If_B48{pWHZJudG2}J=@9PW))1QLj
z!cideYl_`PYG%>=Le@xTow|BUUCzopdE-Y-IFoxfZfxbbh=|q*jNMWSO?oi1LNwN@
zWF>}*3YDdKmu%(%I0`SQ7#@!;RO(GZek@mV)|)0jFR9XRmvF~rxzZbwU%3b(Y_Lf>
zd?z$&acPM7%HhvLWa)rjZ60hEtJiwHE#@Gt%B~`Ba?zz1?mMbSm=xCL*;jo-VCRq+
zC*KDixcu?2P0Z*25JIU#!W9>qr}yF{|4skG?Tnw-r3h@|a@_Go@Aco|A5F^(p8H-4
zz5F*L{D-0Mz5UrUy<-h)J{_huU#4DPzWIl5Hh*J1T6w?|npE%&!#yx()|<5p6(Z+k
zF=OUaT`p!*8et6?)H;;5<ZYcmJ1$MFJCpAG%HCfqLd%OG#8t)9c)vRjA4;Gi5+GC$
znjYr<2lVU-DA^ei=&xn_7pgna*2%?H_86iU@4?v8Fd0t?RmUvzUGoG!XVlXeQ2i;D
zpn1c*u5NXhZmdo~x@Om*G&7&&(;@J;aY<D4I;@vrK$Xp`1t?h+yiIFax_qVIXUHC$
zI`7V>JtGH2v{FI>nyiKGr#&!reRXQOG7>*rmZIDUJZN(79T~Iu|0pBf6qpVMl6s}c
z6NAaD5CcTU;!LHclVx=k(*QaXup#+m8MErWz862IEC<z{l_c)&33E5zI-GdXc#lT9
zQfwtCN8;526F-;l&FHy)E-R#$p_H3vZc*wbZN-T5R4OW0z6BuMkDOmFh$>fvrLr?V
zMps{qz27SvaorGm;B1Vf`gfh<Uh7Gt6)*DmliS6aqMoO(V@o5V>yRmDq0D39Miyn(
zg8P(lG<zGtK}E#zd+l4lN`3b&51Wz@;HfII3umsSR8O=kTm2`(6WQcgE*vTVw((@k
z6@lP9#+Dz;g<O3h`CYm;cFfx;({tJ;aa`Lj%G&tTY}<(L!v4&tJ6S=YLMkrlv9zCB
zY+|iS;)jWY#V>J0qLOMWX#vny8%E9e6^+^Yeg@R8eiBlqO{o2WmTa34+*9<&CpW00
zl$?Ca?&$Y9<@RnK=7C2>viVENI)!gtAli|}Z%T_^3)k^d3{}(22$l7RiH8Jusue!>
zz6owqP&Dr`%9BWCqrL(NIS!(=?BMYGKWgl&;nQp@;vfAev>EIh!@@k%%B3K+uVQyT
z4T{zY9C+9l_0I6036Q2El8AK~|6>_o;^a^@qa+NgFr{V6PsT|Aw9^~jTHz>;UXs!O
z{=OR&-V6rS=RZkd?63ip*otvVc1Jy@Dka@NDF^2J5${^7-c5)ALaAbn<c3@St@P5u
z8<)3wN;^~ae2?oKH_D~dF90CE32t#G16(#3GvP?q;~E~jF!>-ENGapu`51<qQ!p>d
zxS|lXZBQj9Wn>HoWNY<+_#EiFXix(%cesA|F3(8Zz6ypVTN*GvaR@6kryvF?fg|~<
z^#p&A{A~(AKN`0z95Sb$u&7+2SRd)In*{kxm^W@eVIxTVKqEB$LrdL1Vz?zDJOJe?
z$SX+T=AC-%lPS<&C#IZ*Tt3^f4KNt>_~PCg)WqL6&9K(75`Bv4&kkmYLG=LtQ2Mhg
z`}0>BQ|aYzQ~`yeRa<&C0AAch#hN1@y#2h`)jlf0qD$D#-0kOOrQ_V<8<><C<@&X|
z<HFy<-TQvUl3%Z`uIwEg4r6|J2c9?JU-g#$8+v^6A3!iP_u}5%9)7s?dUI9zrg!15
zp`YrVAM!0Gm)w2;6#@Lkmd2qt;VFFoyk2KkHa6a1i#E1Hgw>x>i~PqzaI9nWTT_oy
zpdvzS!hrk+B|DLz6B%W<lJwV32;Wo4)<e=M1K1;$oRHBV=h&w)W%U-MD9?_9p8tv8
z@HB+9YWs$n4bW4cM)fM-EOS<PH5pB8$Et1Q&a4F$kLBpRY*ZpjdO1(IZQ4B=xw%)}
zU%_%`Q{{Fd*H+^xDV7~*v8}7$^@gEQbaMZM4p{3ifCjM*WusD9Xzqt}I`(MU)i%4j
z1dJy&x^Ei;v&HDOe_g|6Pz)nvinE_n2yt7C;*l<-K)WiRmI$;rq>z=SF~T-fKS!5D
z3Tu^gI3Ns)O1*H%dw&n$p<iMJmpgJgoQs==5tmqQS%9dP%4Zady5CJHDzaS6qVmWa
zXI>2lxUzz;j<pr72Dr7A8l38?JlX&H8V>HM^mZ`4Y{$2};%fTw>DEu&g|V5pKPDtH
z*nua&&VMqwbG7rH2~Y^Kwk7djUM!p}(!AeS`yVs`GUo1neYSk3!Lh@#WOUI9BuoNK
zV{{nIw3qXcHm-Ln-~0XhTMU+W@Z88-6`!{P=oY_S>;X>=pis|2&k8dm-~Zcnv>3YN
z_j+(0GS}vNz3{3RIPk^AWp3%lZP{kfjbQO@!>CXPnm~BxEl-D_n;s1)%Hcn2U;;Ui
zp<f~m#p2ojrPnHNd&o|_7Nr3F%@f1q1j5@GKt{{sJxp>6E#E^O!;`x!*BEBY<VQc~
zI*F^+y3qh`g$&kEdW3Nym_f$NqA*pTQ^QqHZ&f}LHRgfr{%Kh`U(wBFRt(AulNx22
zM&#g~n|7C^K<<in-|zN!keX7*?xF#^&7x&RHi$U@C8#Q7Wu}1qAeUzmsr_ha$!Kt;
zrC-!>VGm)pp{*xER2grWe;3eg0PC`RY!RX?$^jlrm!&Lc{-3Iu7`c2iZ0y~a0|%Iy
zjT&iNG&+y*DH6lYQQ^szRHlN^%`wvdjN}|)(N2sw16B^HVo{ajpKJYKU-xT%JWC6v
z4mK;B!oV*v%eNhUTGSIXSSO<ohJ2X7hq5p;LqeYg^UCdER~~<zY%RxhxdxAhMb{XW
z7xk^nRj<O$w6%T?m~VOanr3u0JzvdJ3Q|)BDTk#rbV(#xZ~Ost;o(EIA=2fB#n#^|
zu8hGy$1tiBHA#e=gF66+Wnu3Io6b<yY+>rB`fbSGuhOX*oVCp}a!)aa$4YsijB{|H
zp|}=7F@#WiF>PP2B-fmg=ueFuavZXo-WfDkgXH{L8tw<djmwof7)}?zq>3lr|CR^$
z-r5V)j9-Eh*2hIR1fC*;_6@GXn~zRQ?OtB++djBnqx|~vdXeZdWB>An?={XZ&Ao9O
zHd~7bXg(CXCG<Pwzm@krQS>{ZGQscP&NcV`xS9OnGdy9kFMpkGrzl-7hQqi9lifZ1
z^!b&ZA$i8>->lZq<%1xu7d9v-p+^ao)29xiQcMc3xo5*F0B54JJ6%D!ae4dQT4amf
z;L4V5xXUYv#)f&mSsrAn8_4f<>ABHh{tDCt!Q5$AQn`oc8JNoK+=*a5%NI0bBc|o0
z;&I%<F8`zEvJ+$PlFOVr<$a;=;9^<QQntxc+RM~kClyueL+3>Cop6h%>NO`oA@LAG
zWL6_ujq*TB;|E`Gd7Juu9}zzE1EZ{w3tFiUMeIMn0(3)pXyu#m?y!_zZZe=7Sklkr
z&yH(Y9Y6G6sPk>@k9#X|3+5{Ud$?-<Cm_QW<Apn>gp>-pJ#m*pNdu%EIB~yrD2MR1
zmt&aamNaI{%9_FjG5He@1?Uf2+#{N(j_MnUMwEP8c`ev}sEQ2zlJBD9@jL@3Hz4o3
z{4>ygH?opu7vi!<`?u;<NNK&~qRTnT`+A&HEJ}{MQfc+u=d68a&+2}0C67Z627d18
zjBbM&%NNhm_$XTMVbSVII;(Fdh9%w{)cQWxVy4|hL?54p!*1Iu6<Iop9HE1^8#TlB
zlckq+e+L08Zrh&O|7-#NP^0gi6CwpeN=N?kOerhEqG{IDq%_$kBo+P@_Xd#`(Fqtx
zF1hXm`FgH=P*yj7tP(Pf=`tRvFbCn&cycTA?P9~Xq3aYM$4&{-8m5kG9VgqXBUz<i
z50D@(I;k_@Lv{w@J(tM|RG!99ED6GkS$bF;_4_DB5QZ#Tlyw-_#SpU8@BqR#WWPAa
zWM=5VfKFXqaDf@znP0cM3^EMXN^s}p)8sU*Xt1Eb^JJ$r6n2x;1uN@mZBGly`gkoK
zkGsI;zck@4n^dC{ytfA88TtWsl5XHBTh9ZlRLLS<LucHpN57`hQ9EkyqVTbm5zA?!
zFo~aGp$a+qBm#q?#1t*@YnDMM;<wnf(C%z5%XcD?E`WnS4M|A_b_n<+J0I@JlSJ3{
z2kJ0gTm#!8*@G~BW^=wHK$ecS8#*J9%UruRhJi$$qHzghC>pcg=eJJCgOk0$M8IGI
zD7z|49D=^cB;th6E_rK1PHJq3PMZwPF)B~Miq@6Dk(isEfMM6z@GNH2^pBX@@Q}ui
znN_YF6?kd;ZYTQ;3l{-j9ZpIMN<bYhw*Xr89gTGqruXR#3@9LikwO^k>-~mVyhlpe
zn;%oQS3CUXAZ7KzN@Zt3>kRo-%|v@w?-%nsBd(BfNj_i_z$@;%e2O6Pqac$?30N-O
zpf;WMM~Zd(vsb>m{C;!W(?qx7Tz4vY+RlA^ZzlZG_y3~z4Sa8|n)a`f)bp=99j`l|
z-K3v4?)x-c&F!z}`)=hUO|J4x{<7YVnEX|JAoCUNhgdoadG+dp?rGS<dBQkU?EQbd
zT02zM%T|!ZA}Awufc5xaUs?YanG>NLhP8PPD|ZUP_s(GzsO;81uA*OFJTAse?p^u)
zjP`iNt@emqcNkx&^Mhtmi@C3K{o(l<_lL(3%1O@dl$ByJ1H(|%av7YPKEx=q5cEo;
zuu8;JbJ%VYxMrOfY~vEf@pFqSZ{0}QDbkFK_F!#-s}GXCclivJ0b=eS5xPEY#4RVC
zT$>;@*d7FLsSc0;%w1K*lQvSo+>9EGy}QTqqK0zW{MemW_Qh*R_YsUm!fP}Mz}b7^
z>+8zD#ns^AY^kusnos9|Yyr??pV?!s>y!f>!$UcGEZU|vYQ>QQOFU23pV4^51@@zR
zSMW1*gTi|eOihPdP7+ApLmPG7Q1bG%6#K{CUz#8|k;{blvuiab;+>Uh^In!;Kf>t-
zI|3_&Lp_ene`FVW?yvn%p!gq;>4XQj*SM+Qd*bHkMmO(f*)^y6+r{gVD;H^k5JPeq
zftT=bP)QA#93(>M1mAd?OU)kc9~sqlf7{_xCZY6abZ@iI)*{u)f^pNQ>kIcVja!j%
zzX5{9msXFxY^{49|7*@<;<p2!0c-)K$7j+@S#kgsbxTr}Nm$R;ejRQuoDgAIvRgil
zK#<n`1BUE^(QzPFcbs6pj6b2PQFq2EEI>0!RZP4qaiClLX<TZlfHM7{Y0kC_8edsh
z(F#+=n`SuUS(thrr9?&W(5ou6g{J(D-+J92?i~7GXhp=RXQxawg>{Ee%dZs!BbeRM
z_#>EfFJ=FM`0_Zsi=)R0g81_V+46ZI@wwuHxiyW;vAB86y6dZO6tAF5xLg=28=a&m
zq(D|_UhqcxKK;3O`0=X`XZw-l`%Tin|A?rsQ@`JbuqX(|3}_Z+h&4G>j6}0!(A0Wi
z7dw?%>gwudUC^w3zfqQD(-=VKs~B<Bw_weG^zFu4KFEOvDAMR+%*J0fTzLE(qJ180
z2I^%T7nOfza9@TCksVAphiXHq{1GQvy-})ocej#EvMJPVqusMZN8;;MQ-8;|K|P?n
z8R}yPcn0cx#euFgs9l~5Vlg8~jiICKWl?cyEp8Q&*WQ3S-`KjU2LwL*v0h~q-pdUE
zU@JJ<EBlu%Az4U>*bHeZIkLc|Lg{j&U)x^RblrL*iGw@A_|Ix*jGB(LD&S3${(@H2
zTWxTqEE_(1j-hYOT|DE+UQ;1If)39+L#k*-3YnB2eFL^J^WseNHih0hw>IQ*$dNmt
z7ze{+8^+5{XmA3a-t`?n{w3Emp?wycf8G19{lU>{-1{S!H|H*2DGl+1nRp_<10s{*
z2S;Z$&FPLe<Bn_lH;v6T*VCo9Tct<$-yGcM0X-sJsSeM66GqhX?DO$*aW0{Vtwz=g
zvV7H!mtr_v`8qN}w@q!i52QQ*j|ODLWVyPqcd=yWJ10JfjnPAs`rbp8H~3*-O`MQO
zPrAH>F4>IFkG;ghSN{aI`y`|zU4q1jCDJVWmetm{4T>?alqhwBFe^5GZZ+9i2e_uD
z-bh}3#q{b(Svilf9aVCTaMWaGM?3VJrP8PJ{iGx=2UB#CP8+JN<68&{t~&AjVTb^~
zWmP5>YmEXUu9C8CsJ3Ecwa50p?1V%LyV`O@!!Ka$&S*z9eq@*_udv<c!~Ohk7E0l7
z4RPZlKwTG4%Xjrwl?>Q+u$Ns~SMKU&V%3R(SkHZj@{~;}4k~{*_*kP@iyE-)wFKdf
zHm+QK!liPO3elc6l^vHEgTh|I>i&UJ?2M?pBFiWBj(IXO)#54a*IT%~qh-Wz1>6!n
z-fL&zsYZHN8?;C4Rc$Mn%VXy9#o&(mMa_HqGO_SU8;MzJ_ALVg@AH{~!z(fMT49P^
z^O<URog_4G>AA?n56hGKF;R)6XF6{u#x}%^&e#MuXck=Eeh+}NCd!gKs0VilN-vYK
zw;%oP;YD)E4>H~FUIcANUi1CiCUQLRyDHiCTeQQpN>;<-jAhk`V^tl@wtn|^!ZeXj
zSy|0ARRR+}0)ojZ72s?Y)Dxk{VEG;wSL?&YBXZK_;O$F1Y`nHqRMuK{(N<J8A~IUe
z0+GqV%Urya>hPheP0>};WvQde(5O;2b=EKaLb;=&HpEDl9+lyR4k7Bpcq3G?g}pb8
z1>Jyo6exDPUWtTG+q!xeP%8c>J*T|{*SgUBx5ip2fS*SE9tJ*(`%6a0=LPGMHPh!Q
zt2Z=~9BC?g@v3bys#J|Ro?QP-bYrXtH2Ds88lqj}n)aR0`-p|YJKtx}Ez+FuI=16n
z3r89RRtef3TVO{+!;8yaN~uLf^qENOJC$`Q9s3*H*A{l-iMa=(xtPuGM1>s3d#yxk
zMw@NW(9(z0m^^p-_a(d#{Xbu7=dizQ$Gb7N$Y6(Tc=eag?q734;4fsGZ#zl;z_hg!
z?&W@x05$QVXP`z}nxkcqaRY$0hxCfT#X(rTN-;>SuQ*=GO=3{8B9X8u_r;pXmCJBE
zQnv#hEf_9zh9_>yJN~WkY$kepagxvtTCSwRXg5hnV;&`))N>cRP<3D*9*I$%dCIm4
zTwDMg^%{L1-F7R*cUCKHbC5-wuamH$=0@fz%6zVnlE}q0oV{&WMQKO1XiD=Wmz8Y7
zqH!&Id&rfnYT7i$a#n(vTGW17>-GO*>MWz8?xMEutpZ9aDMKg-NOub&B3;r93`h?{
z2ty;KAl=;{HAoHJB_Q1(3?bcJA|Ul1-)F67J!|=iFHZdDoV~AoUB7(~p>|YyGlMb0
zd368YiPyrJ*Oi~oA(7j0?~9A%%nQn!&en4ZSffjLb+tgtnMZ44NePGd@e3c~D77Bd
z+uy1uM7K*4u!v8$+wZT7ny&ZkHZb0M7(<M8%V?-@sAHG$ADjmFnk+81zX8u5BcF}F
zj^)-TC!tY&MGaYY&|~o5Hkr7OY!mJybyWG1i_Tj~VcMZdoT%}$Yeg=oXlqh)-M?p4
zSZ;0r06oOiiqzikht94YCMnTIq`AMSWT^*CWF%!`5_JRG$DF8P8t|+)Jtx^+Kdo4m
zpL4z48Mp1}vPVuCNh8jWvsaB0%J_ms(ilm#s;M8k)r66o#1~P)={?_DH^QErPX&<L
z12IF123M6%lqx-!S;Uax)6iPhaQp;0l%=6Xnw|2i?ddzq8KyI)$Fe7yQFzL!xc?#b
z1G*Fx@AT(9TUrqJ>bc=xU;YY2aBRPXj!`jlS7Hr}00#zkOcD!I*QcGgax|Y>wahpU
ze)-%+W8=6iYdN{f{IeT+<9JG&VwTkS?|7repNUNw3_q(C?ey-*5j!NJ?Z8lC-Sxeg
zF|`HE00GRe!b$l+<8qG-39<8X3r8k7A{S4BUz)zU8smAnl7x3JGF3KO>^IC$;&%VU
z^Kd38_Vv4$sWe0-g%5B=EX6AM7&)`{HtVpdWy#Ql9W_#q;9n4f@u%2!FaQ6@^rY3N
zTk!Vq%&xr0u;})zn&<9K`>Bj~A`y5MvbTek*~x?bjH<sCOHigG$h;BDMsk{XV|@*W
ziXJUP+u#FT5G($=p=%azxsFT5v7s9<MiQE!jbYHPKo8j;982)>3M9V}`7HW-SJ^aL
z;7*3=*pbp-<-jbvqUelW<Xz8%QTqa!TCexqy6y%%HgXOsQSsC&ozK+2UAZjq3&R*Z
zx97iHBj0F~KM9e**uPagcTNn1L9L?`8HacO-N!HFIy<c>vE}JFH_>+Gb0=j^>Eqlh
zmKi}SlVn3iT5SwNN&Yca#)Zotse3mYWE#o?P)oX^Xhx+*94)nN{ku1rG9o9sohIb*
zVd*;x|J@k-Fs=Hz$8h=mhi@=oPopZ+NrVpGT8iLZ41#s7A34&@J1#`Uqw3z5(hSu*
z67%9}94o}A-*?WTo|G=Y|J=}YU@rfkqIQ-k<TurFANiRpUDiF+$P3TzAJxu~gkW3w
z+WjHK_a&u@PyuCn9xohr4f@N!ErSe0K_}{ANa@adI~<;~tNYB)hS*?I-Hxk{jg!cF
zeoG@uNai%<3Y5vNal{<b&nL>sI^-<wo4c5|BBHvd6g*CBus2Obljw)=HSZ%`X)%t4
zpTViQV%9cn^2{qIP718UQPyQIgX_GAqe-z4+RHOc8<(^V*!jjo>1sha)|&xs3jGXn
zRyVV*Hb;q|;C*`Gb*r!wY4G2J!xM?y6U9frNmRw!hHrL<kI}Bpc%RN%Z?9oZZqYu+
zsy7esNUI%^NF0$|o%q}gPtVSpnw$GeoaWz<g1<>f*3Nnvl%C0OnM8#Y?Zj6hjagM@
z<#VT0QJjLV2Kw0V^b4`l<H-P<!HmwwC2wy;I|oR(k{uS(BH9a^tO_D@G+e~|RclYy
z7#=mt57$m_e|%4TdMsPLTx>>d0dlo}UbIu;X^avJ*!Tg#QhWFwgEV*mVp#iIP)|Yx
zARz4A{(Zx90iL&JvI!~wD7+hCs-n5$Gg5}QBu5v-lGBqD#l%3_B^SV)zF#WyG$M_e
z$bv^m<*tPqYB>yCK!P+*MsV`VAgl=o^AgFF_$DX%Lvw?otjQ-e)PfDP5iF+R*~Oa|
zf;D-us}t(76V_8tQo<H*=EeixLfdjSW^7yxP+4`%0?M!-EhUa;nhrt)%5cyNk2Gvz
z<*RjpTv$cL`sWGMT}DMUj0*>6sUV^r6_z>X>mE}n5shQ0zWTQQQ)Vf;ddrQeaV1H0
zj^<=KokuGV*>bQn6#(N}SntfiYDRTjqS0kRe8<FSn!$Q6v5n;GPyKAPwW@D_*ib=y
z{st55hnehWV25R=;dv2Z_@~pmf$p(XdDf*wVtg-SXX;m~xs}qlQ5V|?3YG-KA$-22
z7udM|AO>#t%eB1sT@T(h$y~$l2zei?ws2mwW30flVcL4ce#y5*Xe|m3^(A&K(bklz
zsQsTsB4}p<t)IHW33cx(vtuk3q&ftgoJ)O;A!1eYtI>`$kc4N6gyaupE5F3PUhIa_
z^?Yq2w>+VwI)x3JR_;02u^tCX$MGpLSf}fCWr!~j2Q`};QGuHx)wn;W&}>=#o^gyL
z#G($>Hc?HU81SlotO~qnq}r&HnQBS^<rx#`ZTgPbEmBC8vSE*eYUEpuku4jQlgOUq
z6hzF*?AXKbikO%A{~?*08qqHdvLN!0`y4JB5SygZz`7DN7Q>y`%sZqgMwW;|w0FNu
zb$Nm-Ip)F<U<-^KqZP_dom!{L=}=gOL}Zfe7_6s1MLNcTcR0v_a!rrbOJUsVbv`0B
zZAZkWN_<_utwilRJ8{WMr=;JRL|GrnMk8o0LId%l3-42gvgf0}qTUn<m`pHET8)PM
zMX7WM=$b8mN$)1^#Hb_a$>^hb`9}dwR#9BT8WJIm5t6G2*^&LPR`nhV*=04p7F{9;
zSPhoCKNT`B1~Q_EL1Q{wDe{nbthn8gfdK?X>x6&1n^pGf*Unz`y>J1Rhy{CRf5jqY
zZO>F-`sKu5J#3T+zc}(aqjpv8gzimagL9#>>Y0jSOtx9&-qGn<seMBbppl6A2*Zgj
zEK?#1`hTw4&U&$Hn7UFQwzv5{E2<kgE7(p+M~1<7?jt1iH1R!CLE>rA=LX%~`oAVi
z9gk~cPLu2pGkoi`L9~OszZvQCwiMpEt4&s94Un`eIcN$s(6-o?$>H)70Ox8v>cpJP
zf_R6?cMO;kYpC;Zq_#GBk@wkR{yse|EKEtxUO!OHn@TO>^fT`vXqi3*Ii6hUUNA28
zA!j?Pi%=30D1-pf%>tuO?ukMJ&BEmc&*gZMe3B`I`1Qh>ExQyoEcot_iYZ`0i)7n2
zirJX3(e!{h==@uiu7oT;%@-pH&qXuRkw5ksx_F%@Qv{(<M{Wv^aUe(*y_2B8App*k
z`-q`xzfaOBE3?1=m%A~m7>4&CWfg==4%>EK2q&~XPvroyP+_?>k5ex&j0TxrAVeIA
z<*X!wHo0Q{$X@B5L6N(K9nc-txAhd~51!?$K=D%|KK8se><Jd>Tg~!LXl2>0NvSp1
z<h2z28NX<{A4V?bY$r1=06lQ4f|(|-U5z2XcTm~RSIn6cTkNWWXz_B_Qkv>qDmWe-
z#8e*;2lM@HVDlPs_PQCzlN}iMb)*T5u&d-#xSyv+*5t(6`B!h8K?QHU{%s3^e$DF&
zKvjnY$_}5Ns}4UsPLD(%7rf+q)yTI}Eyy<f>7OpS(qQLs2Us6Y-(Qf9%cXro(#6E{
zCHK;EFosK!M7bx$y{+dy=RJ)y^Yqy|D7F43W*A*EwQaOE%QP#VXxy)5Y7$uz5w>_-
zxN4t*bR`8fny_JRm=v-oL7>VYHq<LK!Hy+2W|iXUuzG}VoO}|Ac?`yu)w0DdwA`!_
z^L39>H$W2FHY&y>ho?Qud2a!g&BXS$Lr87^Iy{1HuI7!=zOi1#%ZRd{lb+P(VSZ`+
zT#53k4aOs7PH411{ZG4?I7T%+6Ma!>sAEOc#@o2~%?3H0$&`#A3<&~SjGSu-=AoU~
z*K#<#w(9cFt*)#o@1|l1Xoh1Rw-mX#gBGE>Mpg?G=BqTu_EXLi#GzG5EaX+tT}K?<
z?bS|O`-qX>N`j)CXoS6Rq(~Z^)LnkO*3GD7P)MA6wPk|bV;o40<Y@yW>+<FNkCOG2
zljM;!D3Jzq<$R6Ef-}pAND@;TyUAF^O_Nw5^Meyy#qq6+?7&)lB!_jw+cg>`lUKDb
zpR*}){t*x0NPJ6&aWcq3vn>}}<g6Tyd<+HI)n7E-#Wf6$wP@)vl3d1v7JegY#$2qX
z6O%}8HWAFpgR#3B<G`ZBPIEH|Kf~HmPpF~}y~OClUcw%&VNFlJ?c<Bj?T#aKV(Dcf
zDi+i=n^4aONvMrLaFwi{T+<MhBA?{mSEFl^?qBU(!Fqa?Wp@$s>+F5qJo_E@6^bJ;
zG5|q%;Q{6%$b#Ue0O{|a0tj6ucP;tQSC8R#s0wNvf>h)2ED-#D<TmB>L~xJMXFt5S
z0iFEzf|v~8po(i~M$OC^U&aQ3+1&O={M0{B3(BAO`j>eJge3<gf8_qjs`eecTY%yS
zt_Zn?4h%8Vb7-QAY>jsQ^5Ov1wruW<QjiAUfaKMrW{Xc0xm!&eklvLMS4^W&rx{8~
zLzNzDnn1;S$jdHwV22KBTnR!ZsXuRsX@Q$2zZoG?-(r}jl2#WkGU*$UcJp+u?v=rq
zQ!9M-L2npSiDJM9EKF$6k<M1c0lsEi%-?&Rr=w!)_qeE!e($G!wi`QUk65Qn_>ZUr
zlq&86#FQHklK72B>_JpXF9lhVz5HBAme-2r<C7qC+18Wd8u0)Es{WrWq&9-;zIOZ%
z9iHNdbW64?^sko>a%R}tuP{90V14yxjK#H!PTycO-gh-2gPk~Ho#nl&d5=~1no|aG
zZabETg=;(9#lpr(c|F=QLI6xeY!3}oZH_{pHb@!vScrh;-n?_3eW~|E*?Y%7xQx%%
z-7-8orXm21x5*^0H)1z+6)Y>1{DSgnv%tFS&8pN@y3aNLN}Xm-2iJA76aM+ng&PY!
z<pTkU*7yBbi}@eI&v%<yxH8y-)T6nCnUv_a?pQ$-A+O_fauo=X<SW{mSV*SwM0b|1
z&<1qkhaMgMhJX-M71XZZuO4GBj5*N-hpt2wK5QaO?+4|}*e|KFj=-pOtPsVv{J?*r
zro-RR4`r=(7hzZ-$Lp($8&b&3vYk>~h|tBpMtv>UPN4-Y+TKJ$|3wRzUiX{(R~{=#
zfET*eVy8Aie4en*IR5SHO0pBAhPq-WL_a5qN*1r0OWrK);6d`zZdAj>iMxF>fO<Vb
zVPPWgq4$Q98U6Y3Y&4c7tr6s#4MRCpwnL}lP6lu|HFC$SZJ)t}I^^Q@_*YJWk+!im
z5L|3Q1G5g$o!z|*a@i&Iq9`g3(y>`QNO)q*{G$jU=+AbLO>EXP?J9d^%v+rG;P2>u
z1qNlJQdhU`PTHEh%Azm{K6C9Vf^<|435(MR!@z$?HX^;6`w%m-SmEbNiuh99e9jfw
z&-!vnepw3_*I;EOteTL2+_G<rP*8yI+tNy4%!@Wo+S|esBPh(IJImIN=j>*)XUopx
zBljGHRX5$RcWe_&4E^-;-$3h}SxM4sGj<%z4W1O+VwOis1U<&%E-@)rfAd`csA7*d
zlvrUgc<&WB+`#DEQWR%A$x%GA8Sz8@8A&+by9Qh6Z-ZFyywgtaA1*KX5-kWz=P(Uf
zi)pMZ4H8md(`L@5pV$(W90|4TeJ<_sy7k`~yc;Ft{rDi_t)0DvMyXYAMp~yx8+1l|
zEO=C_Nw<SX6TUsL&PBKS{=?NYWeuYo-p7`~nf~9OD}}*F{$JS~MO^<=7cH$p=7Fiy
zAJjvpOvW<VPqd!p_jq@JaSM;-)O~1{sBLe>y{UK8bqy%_fZ~I)X~)j0&!}fZJRqSL
zBOlXc(QRI`@(z>oNt`14VJ>;}IW*w#<7^(LiiaOrF#ifjkT1;4{quIt6I3*K{`9G*
zVZi;&pv1zaYds6D=h4&7^w|?zLD3j+(aI&J49LeJ?Wlp<Oh-EfG`)YbAA1>UH#t=;
zo*9Z6!>LWSv7yjAqLH@MAiOll-y83NPq@KTt7-xciGwC)^!o+DspKD?zt@90pg+pC
zj3A=2j!VA{3~ZNJEVR{w1V?J6gTx~pLlj<FpXlNi1`5sy<G%LFPK#<iy?Sg8J>Ikk
zl<cSL{jqj7aDy*3ei48Zkn7GUlcg)lqk5p?W(8NNMW~b61J<6%{xPOyjrDf;_BSyy
zVJV5++90tcVG!p%<fNs@%lq<zBnEdz(6T`$2D6$+YKtz=_(|4cZtp#{ytmPKwXc|O
zb0$VRy?u!Hv`JpQXX<<c^*sQ4_;${&13CZmPRq>S&puZ?PEs3VueoWAS><twg)AO1
z6C$+Zid`9d>Dbs@WE_C_CIZ!pPOK1mSM<!o3cuqkTA;6iT*WGetGl<eNb5lEmD6nH
zrqGYK4lP^`;l}+?Vlo%m$K8^BUzbLZ&Mv56`<IotvoWJcOk%OY$b-te-Ta3%^{P2l
zW~63jdBU4Cirs1}lO-9$vqP>T5|&1kqLML{e8K-jJAOIDRQ7=iZreCMBw~yVo`f2b
znO#}_-cd>5o4Z>1k|bjxlaEltxIhoP0ktxP-4XCThzlQ*_VoNoN*55?_y=@zV-EQV
zd1XF3-4i%vpWb)3aeC|EU`GvHcfpR5tXKs}++wq4o3s4Xeob*at#DjUMWI|iYs(nJ
z_|=%sMmfiQ>cC%=()Xc~CG^`V=+7mgVFiVE$fT{Puu+@_y;YC$k7c^oq6f*T#53iS
zkN}Xie7A(01lGF6t*E{quD;Bk*c)oCPFmfUnFv=XsKPvZNcXw@r;SS*<LY0d6kMv>
zHQ?X?^NbAmiY`R<B(AAg?l>mG?qo_k6#%zR7#NfsXEH!c$^{J8c~l)$)*ni1Io?`@
z4z_SzDiPeC>I=&E#iVxAaqTK!gy)})SZ`KS(=Ke=FX)Bd9yoS`X!mI>me#{WZOA6X
z(ped%`9nImU~4Q(z=?2OArXo&gQ}=~Mo^xkb1OXOCYCTC$PaHNCaIc(klH{HK}8Fz
z+iK)*C6Q_==Z4XVzh8dut90%SYWta(qou@`-d-X#d3(5ud+gia(B{;2Lg7cY0DyJd
zKSr2Q(txh_f>iJ&kSHf4#%<?hBVt~UonXa}opGR$Z6JDoIqsHhJ?o+*Cd=z>(EoL>
zPd964*7BKq?YM4v;g7?z8HXqQ0G7UP&pFpB&F*VV@Y=5;rno<Y%RNO}%R!hVToI}R
zW?nhA4Xlm2<4N5jmz6<t$aY~Bq;l+uf{trKFLa`8(#2AQ35O4flhMhMXQec$`J(=Y
zEeW5=xJ#qa*S=1Oc&RjtQmkxpm4?FOhxFhUecqZv*`=e0;~$HMz`1u`6Q~VEG=!~$
z?cnfb%p`LdwAo_l<s1T(Q{!tCvj-TY1m08<go8FtNL8v&U8N`VZPzGZ5N!%k1E1tx
zY!tZH4qDuPI9@u|{H3M%NC41l>_a0~=TqrCR5X!+g8G)0pmkqQzQzL9^yE2k0AM-6
zrjukP6%2a?F8km6JZ=3Q{1S|d8(P9msl=*A*Ii3#DyLn9afu^*V<K`w_wn;=S--AW
zex(9E)ALl2n@dd!=sIY2>Lc<VC-c76c<E{3dXv(5Pmbr}3Z!D5l;)(~{9_1p`us(H
z$n2U52q@^^#8hix2)LaMV!~m7!q6_<H?=DYB+GF`kEAsz3Y54JeqqlZn`&wQq<0&(
zF4Q9=jbOwnMLuKBPp~%L!bU)e*Rnt3jUQ{hY;3q&NR%oF-dqXV$&CU@2q~_)pBcZ(
zsAl*mZwxE)_-=oc6kJUU)P{gGa1bQI-pgtsFB#W=pH4&Mdjp}qLFT14RdID5$-)T-
z%DAh#<C8SKL582ZFFYAW{ZX@;I!J}QeV)$=xHv24j}i(%dXORwoAT$Fz+j*HRs@J@
zCK8(WTTD4YZAHQ7)n4KB%Eemg5rtB0ShM__q;<;_nSL_yuXa>UV#Y@_$|N7sIfeBe
zM1OZxn$(Zz%d8caI0W*#I#wBH5yT)c`_W1&8WE?}8!Ai3MLi@8&5g_Z>SLopc!18*
z$ckRWf)=Q0aIA1>2=Ja!gB)D{+5{6#w^_n!+I>oS!*RPAfE#FM*w}xoo`6jdYTQA`
zq6mhd<8mQR-Q1h!%L=Vt5Qj|Pb)StK>#OC1Ii^88>i*olCpN#u@Rc_@S3Xhvf!}Qr
zA~=QsG~D58CvASHQphp$%9ui|Y|Turd*=(g@TpYBVpp2^8enNxdR5D}_GZDfdFAlx
zvy#UvN7r@+r6cy(iHHuMhRfP1ml33&CJZi#WzCi)zZD5UnI?}oa*W{;<`bs^;ia4-
zwgC--nR?03pU!VC&WPR~Ue2GA)gQFlH$(3ngMcK%Ku-}ahCd;~=;DrIst<YS>-x2%
z;+kI)QTvB7nCT?vVwuOFT8V&jv}V3rHwCpVKm=(IUrI@)_7|?#UxV@15JRJ`f2BTZ
zWTE!*l`UqKf57d?erTS7t`85^pPNrTgc`dPP6+Gr(r|!v8#BGV(1jr)3W|_LNza)8
zDo{>Z%=UG6?WP-f1IoBCdQ_%6XDew}<@tWt{ZM;tm`CBggnw)c?9>iuyYQo+O^e|b
z;6!QQ(qh*L7t@`b<vbDIsQ06+{AM%~w@hyInBqL5*Ja_7zSc=i`<Dp8#jl|_ptl0k
z(>7%2%qyemS4S;J#0HD44)NOvavq@wsgS~T@iPLqg?KukBJ3FrjeGEc{Gt{?M}l9L
zJ%5X-19G!+`m}C@VnA7KVz+=lq5~0zwiZUoXv-lz5mV3D6jjKR>(ccJJ;sA>4|gBC
zLyty2Qt{XMl=P;nDJ4*Np05Y7S#xj_P^ImB6l<FP*N%{wi3}eMy1CeKGF)eyM6ZIA
zO0R$YCGFheR&#P$g(#<*;qf0l@r?O&5++L2f3Dq>5p>ZX%E$QP8J}SLQWh-rS>bi?
zyPd7Q%536e;$poSyF6I&xmn22!EeVH(qI<(B>wIJdR%Gw&1m3<a=s8@0%B2gFMqU}
zrB?rx4(|s>RN-34j$j|dW#dBGmYX1O4Y>@Pjy!VeJfa5(0B&4$UnP_0p-rf7Y{!Kt
z=3EW_a7T87{8D?$20~}7ScVr0<}fxnj^K1pj>HN}Jn73x2LeEm>#qGmH&rD9*5h%g
z;}hUZy-;!x%?O9@kTNcRaKV4~*b&M3Yy+zP2%cxwAY=nU2NzH&p1hAEXONc9BaA#N
z;qSb*hA7~0#QAX-ed~jae+jT1wRdXQ?7?@3n!YP4uMdC+V~h5YuIiD+6e*{k_+e87
zygF#+{V@q#V8iY%c3p!xQL)dmGy{_2QJo)WH<rIY?_~oYBkqXsE57mvtsg2^{$<o6
zZK}UL#@-el2paQkDhwXKYtT6f+V_n4Oa~oh39D)!4_rnI%P|B&Y>d*;T@J;Omk7Gn
zRNHMJpBZXE?nvqt-fMKSu52>PAk(bgq~szn|L82H3hbLm2+GD4W;0G&V*)+ei!45R
z$(2?%>Rhgz4ovXZ0fWz}Z1HJcA+442f<<hJR|i(Y9R094bY4%--pf2q=+^6A^7j+T
ztmm^JUAJuL`ieNo2=!`*iAF`*IJ?<553&!O`&|EV1|b@tH+6snZ`TO^hxGdZeRHFv
zL{|{=HZPBxA6P;C0;I;t2F+`i$|fD^kkm%o5Re=OSC!p$=ECj&{rw-&8@m#(q%80M
z<bNW2*K5~}yodunoqXv47(t5KndX+owdSy^V|rdTR+kqb$p}``Og<f&0;Trw-k{qb
zBN^zVZbR3Fu^*s{wB(iZ$^T0q&GXaC;bOH^-R~)u?;a66HJ>a=dbFBFzpu9>9gH5#
z+e@)ZV<nt#KyL|s5%dWF%ho&pc*?KUf$BmcoMk7aIw9wE57~>)PiU+du2-C*ALojI
zBG^5zH;=6C*N=Kj1<}6v4JdXj^F|l)>n~GZJwS6z<23UuA0w?hp&*~A6kgd%rvy2o
z^?hbD>NO{W*lxkb;xpK?((6p{9k7$1D5!-n9UPs|a1ucN*-i`6St@l0m__f-r@s}^
zdPrnja}iJ^Rde|Yg$7U;8ObjGp*nZ!GNEv80k7Wl82)<%$|U(lt;QOdWA|pEh&mVU
zoSo1~);`6S(>4L`gFFpDX?k9(0t_D=`opKXh(;K#;Kd-ENcRT|7cI7!N7>iC?OcR(
z%*Gci;73zUJ0M1Z&vW>RAP4gCnddrms9FAtVVitZbkPxTDN5pKj0YV<h9?=KV3-R*
z|D@^tM$gTzBtRlxgfiXknz}-Bv#GpFOpF?>Pd?u+rIzoN65gIf_J7Q2?U%Ujv5g>q
zsP)jJGN>e?+hrIYO%CWzAr0+(`1}NfsE0TwUl>Zul!JS%>X<SRUw0r0=yCPHZztkJ
z3j*>XA<}8ciXLG?byU`mpkTIEBjWz7`aRuZKv=z3@#a^zq_)w|MPY2^*=KBD9t%C>
z8>`SGhq*(eo4YpJ1fBHISzwTbLb(+tM4)zHn;yR*g0b%$EQ%Tuu8U)?cB^9Oi_xox
z|K!u^PXUa72cK~%5K{0K+Oh^6`R#DbI~R28&AyP6CUiXGUuBSc7z0y(t`KaU`hs|6
zx{QEb<Ne)@9leGGbFIgDj37P{O{%FiSf6ot_dwtF->E(NM_7-+bNNrw#@PnDreTR`
z$cMj6;ev^Qv|QBKo>zZ>0ge97a>ZMWleXSnW$%8MNYHrm+r)xG9g;zbR1laRed7NR
z9Mttf5xNYF)ebR2HnE^enQ(kq8Il7X5bpcN3e1a;=y@}D|17_fM?wwli$CE*#%r`g
zY!7(r1v^mv;8^E-sUZ(PX76i|!8p%*6m)kV47u&Nq}G0Ly;z-N(K0w$&7-TyIx3ZD
z43$xiQs6TuMm24;&>hs<eJ}O-7}<c{bQ7d`&mRnAx5DBnTK-<#Et?N6=By6<s`bvq
z>JwxDx3C*VZy_IH2P!H-Zl&;?<R2Cp?CL;iI-YU~lvY2!b+^dmrU)=IOwg?E@;IP6
z-mDi?(>B==Z9P7ExypDoPt=OIdr?>-^zm^ltc*TiQbzm1@k+n>@AQ)FxHs!=8QIp$
zA845Qe)Q|ak^Q3SW!VM8pb@8*UCez-xJ2{IrR2F}?h>W+6Y<P|oRfih-yHWh+M2$?
zmLm8N-d8*V1PjrVO?C>H19mhA7aQgDIk`1Q<Og7*r%b!g;wp33kE0Oac$b>CnL#v;
zp)VoMvwuW{;79pOLcYhBf;PZH*z=`k;b@iAxk+xC@}V7iB?b0GFd_Y2|H$p@Hlff?
z?#h?p4`S5Q{5!A#6c$v@L%K3kz)={#{!~c{uzvNs@w_kpK6F))YMcH^#zx@(x%_@%
z(~N`sF<{4qWFh*?^|~uZ1(1^w9h!s>0UC##53KD5-S)gYknr_w74jOXv3$E&LP97a
z)_;kIZwi3003jeZc#tm7VuNnX#?>nYEgzL|A39_f4Fc@I%I@I3sPS#H<lv)8tj}q(
z!G6S>3<=#s4x*Cd2t0A6j>Y4$wpO3@Pvg?<ptZ&G;LGDG{<ah2uFh*Y1>T&!ZEPap
z#*=`ydUQW?PkqJYB0w!642knZBw5iP7Dj~)9*9-HBNEZa860u%ujkY%0cD3@b8y(r
zZ9B$bSx3Sp5l%?{!`&Zl^`rFhy|Z(fSf3N0zX3u#+@<B-tM4T`o7LVLYVlKtN~<qI
zMJfIg(EhR+b>`?&W18Epu=pBA@**_uAA4rwA;4l=&L|d07s9F<&^$Ill(u<mMqAkC
z95f}|EDtKuq+!5Cx#|YaY03oSvI)of5xYBg6`%Tpz3B<fX3Z;0`l7MiD?q6ueZZU7
z!-jjupOn2XCHGs)s&QLwIAcN9)|1~}vE4QQsw4W|PXE+3lLY4N6qq?-WP4f_Gq4K-
z`(1y1grcTq*z>m3`@wR0Bm4~!l6t|hA1ikp0fb#|$L8bnh~=iZ<qxzJVL6pm`CoFG
zM<i*U8F=b5P!QYc@J8l+9SDJ7Vg^oX;J}Me-!QYj{V_Je+Tq%sj>H$r-Zb6RUXIUM
z$`5{;z_aFV0S5tX`XN?dO(wn!#-!9sj2`edAQu!!y7_w)>X~w0S8OOUx_<;K+J|(B
z`75&3J;G$VYwN{+Z(R!J8+R?iLf&hc#*$`JP4E+p-<jqAR;ahUY9)^aj#)NkDvj|c
z(pcy`4W0P0=+vJ-bIm(FxCqXF4~4m=9UQH$y#lv^yRna?^dR`L9v)=Q<WE}B&h6S?
z+qb>gSF$HaKQo}I;Cq;yj}@7|bNuH-91_uc(CXICti-~~<9J>pZ>Vka$CKRvPHL9x
z4l>iq^1PJYXP$v)bFB)qd>4n|8ZAdRbQr?Be=Ht%pL`K!b9bxk9AeJ?wiN7WKuGa$
z{M%ydM<>6RIiJ$y2Tb*CZxV0UmGK93Y@0_z-Ly%)>cd(pPyMG7;uB`L-y3@hg!8&|
z@icI*bPCNF7h8)M1QxtqpypBsat`v6pMisIQjZgOS~_c;Um@Xr%u4lBc4ec;BO4k5
z8Uu>umS^O-iWIrN2HgBvgG4EU1`F9+4<23*F`Mb|Y<?sNJ-No#r%BXA;tr2-z_5gx
zorxWazZckIKzd}%L8$`xolz^vag5vW$e<HnV@22}eeuWu>oF|@DGX`s4`w-$q&%e+
ze1maaTTvS#Gbi&+$jZB_SO7+uJ8RjT0gAo%rN1et5!EKiGFX;YP+8;FJ-QzoHzZ4t
z#3<bWfQ%2W3_+%4PmCFhKjd@zx}l#P->0agVR-@AHiPIz9^Wb>a&t5D68cr#j0FRG
z>#FaQ&eZz9+#1l1fFFJ~FZ6W}I<S&9{<`5lPGwL>!Rzus<<!8*S<k=oUx0Mmvh|tF
z`(nlQ;&<~yN%dp+LPM{AJfKM--KSP1#9Y76sr3x$>UwJtXmM-?tGydJ0S@#SRj-Dq
z?|)wmyPhBJJ@nZw7kFiV9oO_t82aPe7Llrr>-q1w<dJKdSRl4<Z9eIH^n1=uqFdH3
z(*5#+^J%@v^78kMYog+#%BOWhbLCHzscFVAi+8-wNOrD%W_rw~Vl7%1`y94WeMuG;
zI8ZjBHZUcI4@@vmRpWaq=3}a+EgMPmltii^e>qZ56Z+|g>Rqay^)M`8Mg*43(fr&5
z!Hj7|U&6r&zcLeZ72EG7#G$YbCUNIJ9<fQ;F~ysw5x<Dc32<WHj*{XY7iuD20f3EW
zeT~42&@lIu${@aF*!Le$Enqh-Fa9h||B(EB9lu!^@JbZfCf2^+te8ALr1|_B?q2Qm
zazjBIDI7Wq`V$~DU{Y*Yc|V~szeLE!<<JeTc^`X$V6$d?Z<wbCm&N2x{3U3j;)J_b
z{1vv6^6#^X4)F6E-zBdgx@+dY*>9`yqIVU}FQ6<l$&u(iZqW!#Yg0d<Aa!a%oeM51
zkcj8z*!(D{gMm1~QdhXmtK^`$W9IByN(H9DzX64ZTwk2$BE79}I4m-ou~;}sW_MVf
zaQ&e%yS@1mAp|7RISIws=Zs=V?;};>Hr_&|3wbkl^cKg^1*Th~SoAbtE{B<fGDjG7
z2!e>I(u%KQ2a~ZLuccm-+Y|c_i%XC$Ugg!9QQu3Qlx(#UtAA+??e3*S3Gc{Q={^h0
zT6<ZqqAcFLSN-tUEr%V3MJXX!;eakFDw-}e+-jLL3L9rHH2MPe5`TKP$NPF<C=SW6
z>d#h0<Gr}Ub=)fM{0YAOOALBAruNB}lEGcRW32wG(-2a7%t;BD-Fc;tZNUH7aeLkB
zG5Zg*kUe|@o)euj$wRBD1f9Q?;J%ncD9s9pU_6{nyP?r9>N~RE2Wv9WJjiG5RooAM
zzMmljMqTmMs0XHgX$$o(O6sIAN<rws$Gf@cCH6mcXf-v7H*twmRR2*6(e#csIYNM{
zPi&F5YVZeZ-X;|r_>i3FhT1F8a>psiZ)wea_Z=kjJkrA9epTNE2$+offZz8+`j@Zr
zhDZ0Tl{wM;i58E#<2Y9t2C$?bZ4fV6Y=L;vsA)r{tAgU{yK?@@rYmufKz~{<u)HE9
z`vuyjjmcA&-24e_ii>UG<jUP9g}nY==6{pBV{S!M??2J|&zd9kq!}AkP%ig_IoFyP
z%QE%vSc%RAD^){r5+mi)MjjK_lo4xPv5L{PrNIO+i4X1Xz>XdZS~r(YqX1<?k~Afl
z6(S5B&8sEj;s=&}?cD9-t0VfkymmQ<ca<#N$bFWV1cr!m&u8~9u7*yZqo1-1C%vV{
zerT+2#O$?nM4!Ks>SRE&faLz2(&64UJLC*5jW1^)oBc~oPCM?GI0287oMT<xvLk<}
z9NwW)__LC74&<&Ls*Pmh@;oA<K>qA3*7b$w$vx|V<uISy-~YYHvY-C8G;!xS3VY%9
zl;UFP!1LU>m#!rN?|bVJTPAozoF`t6d9x02RXJ!ZoLuYd_pBkf!;=UP!Jh)refxU?
z)OzTlH_{1M@@@0%OxnARHP)k51WE?^cc`UI^S120H$y$K-L~ZJ)H9NSbftToDJ2Aj
zF#0B3c3RV)mo?mB&9SdU@*y+eOR!N5dAa4}ouTa;bvoeqf8tPnm!)5>nyJ_N%^If{
z<5YE))Iue&-wVzVBR1H2d!c9CgNiPwH*f8gytX#UOjC?t4H?SXxNko~%&CX?j<dmn
zp?Yr1vbj^s2B2v=iX|e8%oXUwErstFIFZG|P~1etEKZtfl`kLGyAvZ<+$7_0cN_Nf
zrf?scQxY#sYR~MVte0LgX=!Q1wO@uF`%=B21OD<R)~(B$c7OSg@3eFj$2u*eV|B=Y
z@HJ-T7YOL(RQ$kk!_1M~MY;0sCK~2G>Y(DArWW^z5oETWaha7a?7HF43Zg9D(V6Ze
z;vVN7UnzvfRDGLxEC1%dZ`L;bxyIHYYSgmSrUhb4plQzwduw9jzTr;`h&@M*B%%?W
z{6mKvel>O`+oDLriXgn@;%7tF;B*}qc!I?O#?sCu9c9MA*Qx<@O3?0DizhOaJctR)
zk?sR0f%l?UBJ`l{YE1Uv@M)&_s!{af6UbR<t@~vj?d?(Wb1)dwVk}1?pr27E8U5m*
zLA<Q|0=_6f@RjGR_|#qd@+$VJ+u8qMRxJMO(b<sX4t0ni&+M)#uI;QkR48I*BeZzf
zjW_z$)@6?I`C)vBSTK8R>rRp=!7R**Wr%(Yw4*>gsB@`%#-WfgwSN?GBAzjqc`~Y>
zFZ|Cm#$=08kI)ZK(UT7NoQ2r_$zi+-Kiu|M2}(!yuya~A0C(go3Y>6rSu0oKatY<1
z1RRjmY*I`zypLu7*K1mA_?zQr9KdTjC*K1KpQ%iy^Mp3P9p(9tS;H4C!a7*QtnF(E
zhEX+lM!lE>YtH8wSpQeFHELTts3HdKn!s~)QSKGBF4oMqvf?TOh@l`S??{u`I3ASR
z{hu<}&9g_C7^r+)LEoc*|4f5NC4gyv7PRYV02C2&b1EA2m-K_iAy4>o(bggAjI^3$
zXxpz7DR*?8>$hL#1m&M?#F-Czl;WR9$W8*A&@OBD?-0oKP%QQANZrd2_aisSltk)5
z4GT4<>DFuPO5qJK#<iQ>KmJEfeEME^wM*^$$JsS;9c&mi|A;eBu;=n*(D0rTSo6c4
z9hUQ~qnZHdd_xRbTbq3d%aIeC{Rk;AK>+m+L3(`6pkg4Nif>#+%wrHvgy|v<cbi38
za<X<HYOIj23Gh)Do15L2Lr40<w1+o1Iv1<yJT%>_`m}}{@4Y@I!ig*fOhg0%OGldB
zFOE6FCC-LyZ-z2G@1Hn-`q>JOi_ELRjZUxkpM#T-zEuDm%^CY*r0pl;%UK>}8;$2m
zxXanOg}H2L@pt_v3n3jNTXp2K9oRb@sZC#`W3qD8tY=~#<MN_Pc5r2vzJlddY8Zg<
zGL!CoaS%35DJz`t86F(G>qIwa+y4`o6afAW?9I^^M{cca4Kfh-{-7@ay*jWOTqnQ&
zMH8R5pUiuA(wVYI<$15&a9;=%<|!YMl{9F>`5Vk4Xl7cltkUY`@YazkS`+-#bo^jZ
zlO!umKVVa*k+H|T>C`3<lX9P{K$kg7V(?Q0uLJx8y4Z)_VyriR7S6grP(R6RPQ_bp
zwxeoo&qn#DQoB5_)<N7I+1HHxlC@POwq^7;0od{4@|N_R#7X)`df9@TetxR12}?Nu
z`=#bmgAp>A&3baQ?W0_WN7Pj|V&g`n7E0}55FW*U!>eFNaj>hx-GGl!9gvNGiIV$;
zWLa;%&K1rxhnC5KZ98vf1vZ5kp4ROolaeXXOZJ#SCG~oIB7NhG$1O<Len2JrYtJKa
zFm6lCKPvGB8`0c|6<@o$hIT;R5zL%R<|X8N<qmHjf4RKqWt216Sc?C4M_d(b%&3jL
zN4Q>Y&%qY0E8aA*HnWrlG;H9|M2;AxtpCEC)i;bT0ksSaI6*@c*8at>-12UuAL%eZ
z)E4?nES%@j)*uiqX@KxKI!p2)l0HKivz|OhsaK*lA+Ui%0HTw(krHQ<|La63{pI&e
z4v!E%7QF42@Jz<g+$nXHv5_^)1XzGYQ1)MVDQ8=ndQNPm$lRG)%$rQQbC}uwpm?};
zhuv?CBG8DYv%XK4I#fVoPs(?J7uQg8mYlnxl0;asv@$xeEVK2k>zUBL_8HjA-$KWU
zF{4mB57_2~Ta5}#6a2Iz)*pseMeVAr5j!f;V0i7>ADe+um%*?}G<knE7ncIq+j9Z{
zB=Tt>*6xnUd!=efT{f0*UdupHvM4+t7q7Oud+(P6mY8Rkj@V8OZBcgV242;KfV0+X
zJl<uu=)-{_CF!WGVpfAS#cn)sDgsu`3$9aO6(0X)6umk&`mXI5j9w)Gob_$m&PCH%
z3mwsiDq!OWgDH0xp9t+<mJ`cp^@(gBF)5wRlr9E^e?xT3jEWhT^KrZB)$%=BXAa>s
zD%t<?9AwoM1DEH8El6{##QhLS<wA<UTH5Qy#sR(r>hZ-l3orT<WkTa901;Y@0BT*|
zbp0k0RF^4)2E*(1ek@O01pw(VhD_uij0s~=vSoj5_8Bhks)BKvoFN3E%kT_YOf?`Y
z^m%X+ogWT@DThKzAS&xWu=eD9V2@?<xSoJ-G!V_N)P8p!nbqHE?6>iUqUp9x>*h57
z%nh^lDiQlt>s2rfxaZBx2S*aikBUm-QN2DVlN>kQx7(u_XP<wB^;#A3U1i|QpjFd$
z02Dx3@un`kHWo5-X~Q;QB_sh7sV%27^3$2WxoeVfOYU9-1*J!2l6z9YDt5G~TY^en
zg*%Qio<2s`%t6BJO;p!dTRDCnLCy5`-zr!vxIz$dqfYntsz#CW6~6grs9n{yuWFhO
z+<$CiBeoBdQUVI5a%hH^Ei>LF2*fg&LdY5Fb3?&|vPn!a<PAWT;)lyKgPrkgn(&*@
z(yS~xY5fk9=9P+RXQXnvyLt=-8EWjw#|U}oPn$oTVDfXfA6*2QC9cUp2b10GR_6-G
zau(U)3HDX`GentT)yh5w|J>Dp7xbv5NQ!o0zkJf=VNBYB$5tFDjt6?_QN^G2(#8zl
zi~9QBJRWrDFHd5`%Hi(RG`#{1fKqJS@UK+lH8WUiAUWjbds=AKYv7`65RQ>f3}-@5
zf#2spzE!6r$lBGpLXLYINzgPy!DDa2INTW<-rfsQfJfuL>aJcD*SCF%pUMg0EeBkU
zU&0xVV#B(&(71S}iu=kdUOK>(?%Tv{Sx-(9?{IPxI0*)2pOb`Cq)=*oh^V7BYuTxU
zUf}d*YmwIs3PX@@h}BBm`&xi|0b-S;=56^0gFFJ^fwg*W3G$_M7vGg<7XfRgN1-1?
zlrv`1E94xPur}-^>yr<{-+;p*0XQ?mkv8d2%^+A-F`^M0Je_ko*jGI|2qKTlJFAg@
zBo7^ILN`faqBm?5Uh<qa^|8;^{vcjQkrr1mDz|H=VxZ9*(5{^T8AUqEqpKFDD-hYx
zr`5|jfO;<e<#CmLOo8Lzi{_W(l@2VmqVD)!=}bGfJ@SX05XI9-vL%)r@RjPX!b^bg
zKVgQa$f;?Pk|nTECBXYFIJ4GSTh&FrYy`iCZy?+F73%n6IOzmH;CS9UtN+>>xR&$L
z*1no5L-oF*IbELv#%ARCQ)LS8j;eOV<<yhL!__poRH35+lw?|Yixj^9L;@$N=N!2E
z;q+h*sy(|BJt!y-<sopSR)pu7n^Jc1E`M+(j!hstXqeharG@2f=?!FuMg9_78+S#u
zbFj0uH<^vtyMY-4pn}t`QFE$WAp6*37WM$9y^-q-(I_aXSH>6N_Q9uSmBjFg;J2cb
zjXxU$%lMzJ5$$dKOJgC!`ZWzoraS`xiNxE$#<nicWYdlu_-cHzshwI>WXofmgbhBh
zr!B~VC9%9h2Hoa8J=m_0o0VMvqFrX(B6<R-%}D{21ejG%%*Kjqmss8C;!v=#ED@Z?
zoQ2325U*fA2k7Nrjq-n%z<Z=)TDh>O|5SJ59wzc)9VY-Lfv;<Z1&&mEU{;T&dhWvX
z@XoBQU;0JS3nj6p`Ni#zshNNIB3rladWWA%-5yu=37`R+qE~3H6@X&8NBo8U?%sgw
zDG0F%2CGd@#P&N3u!t}M8d%47LFaaUIQ5#Qb2wN%l(HSfi1p4mdqq^#c9y<m_0hP4
zGB{z-dC{Z-z?u|<WtBl8P?J?Y4-_d%C@trYt%K>Y=9u=*+^;{zm7SzaY5OIoc?Bl*
z=<xv4LXAOm@fs19CBLc%)lO=NFjZ(vTs1)l9h)Q@joJl8oD25|r-UMv$N57FwI5w^
zwV-S0;y!yzRgI+5NP53hbh)saHLqmDt}bR5A^75&&sLdaM^z;jTZm1=W3~zTBHzXw
zGi0G2=azu>eSoh*&LR{kWuswQv{5<#2LN3c+rLVtQF#9D0<!sIHSL}$zEb9$lp1&i
zihrX4jptX5cnl{45rmj*lf13NLzz^=j`fRc(3*-H`#?;$AUg(urwOuZlet<2xKMc=
zR??B_F=vqax+LYl^&FNRe^mVQ#uv(<>E|`U49s{RD6Hb}OrFhdyA@Z9f5_wn54~A_
zg6Y>RKGja$Kl7IXzi9pWZDT=b;ZN6N<%Q)r3J<WWK@HcTa1A`IX}3>3ZYhg)Qs&Q$
zTk~scTMxFa_o_WaoQHlJQ^iOLC`4201+YKSChd^7xnCY7j7`mu88#JUe4jk~0guf?
z9vj){;;ESDF%`YW$z2Gzie2MD^WTpi?jHvOcc!gz3n*z*Ivy?7OAJoKhxInRCL0bm
z8cKI=Z?6|bl6hdk(cjn=os@HSQegUL53<8_h~RP3ouM8y>;})}@z&L3T8)pk*bEFM
zk~HENL1oX?T-V7UM_EF=Po#6Tt^YzZZxtZ~%*0d#-benn`{^7P$IUz;cb~TYT&jLS
zvK}nf2v;x59F4|9-TCj%9r?GCn)M=!3l}F(KYzWQntU2?ukPDQ>1+G@B=7Z>15g!{
zG0_3p%(?z%<1$dmw0Hq$gBu_D^qG+|XU?<cg_XZve@?k>Rz0>Q{1-0I+BGX!57JBB
z#0WQq*uD2BKNWvFKfdOXOsiaQ<#r~J7}|MUT9t=MA%_}P%puj{LtIr7z7&KxH%9m)
zzER9j<Z(ja#xiOPUuSC~the4=aw#Yo_>Nu@kQits1;(fxyDjs<n8YY?2HHt6KQDfG
zCk`*bOiK{at7IGfLF1wfulm<^u;jZ6U51jKaWo4Wm2h#To2S82vSfU**o5b%CjBE1
zVqrgqPr&8tO_XjhCjX6<5dPRy!ZZg_G3We^)G|r5wpOo-Jzdr`GY9b&jFYvbJ0>rt
z3<C!fjI=T$?#+v(IEsOMwLb8#7JrGz!6PYTK3wq)$cxq?gP$ZZ_ahpaF2xg7Xeh{j
z4mDI1)>Tjc5OAW6xc^&;5$USN+P-t$(hV;?&Eso)^tI5}x3KSE?PWJ!Kz~w6gx}MR
zrn0%Et^l3eG3K!ABrY%RWlyqL3PzmDNBWq$%IU@G94u)=dw6B%3gXAf>|iL*boZHP
zlRJM-F{xfn+cd8QrTKYzSi7=8-T^d?<8F*-MG-3<$Kpjg6V7<hFl^_f??dQ&caNUd
z8;!&Z)z;&r%%(AOyOK7$Ca*pZ_$w)i&Pgrm60<+g-+jt#9hK73yqRA(|1Aso{<Qq&
zX{%SJbR078z`>5xUO05K9^0e(i|I&N&{*2pZ~{00KIr^flQ(q8JK!eZeI=}qtO)7}
ziSd?Nt_*=#XET$T{})&pBp+24rmh|QqgQ}EU9#hi)@v=YNeGj^ZDc}8bOKos^Gfg3
zh8x}6`=8~DHl5>=;agR<^`u3G3Q-EXGkW7D?_M~D`Nc7CAVeWl9D(|&%7XfWMBT3p
zFea2Jv5foFN>W$4F^;69O3K{qMGWwGvsLJ4gyoY4-=}2GS^8sDj6`DS7Z242g|Ds?
zN{Xs%z-!1|gTS_hVRM!yR=61)U(c91gfslvpO3bXojO`Q5&h<%rf-+z1YwG6R}$Bk
z<NV9o%cGfRK^<Z^Er`!kObOAq4Mdgj!59a8bFp9d|2Et0ZyXTh7ffMFKu41SGprYp
zw7+alRP@u8WKo?-2<*LUGP{QN#!V8ARAx5&^!5~N-`&}m4YCfu7&T{<kkZn!SI*+y
zJrnzblbuL?`S1mwh>nh0SMv!OPUR<Nwo%q^q>}f|r6s$|zr}@U0iG3ZN&>0v(=@JL
z+;a-UdAgE8QScm|4U;DY3~kuqpR;8)e90=r*ExpHzce_W-|}6eU-zUOcW|22C$#PU
zuIlhvq?F$9C9Gn!>ipcBqWh=td)&!QKAlESd)lw`M_)7RXW)-5uKett8X<5OJb9&b
zSPK=@H4T3ukj{%pyC)U%yzYBjVf0+VAu$5OX7S`U)f{>WZ9ad6Ufq?vEiaM4$L><)
zWops9a%a(MDX`9!mX7ll66pBWjW;ZJnSkaYARf}Wi!*N}lH^Q9bKdgl6y7m-a9eqB
zv2fa|`-Ku`{1Z;crxo{}58X?Zn7{Zs4;Q?}p0nZpC5WwX(LIHc%#LgdD$Ufi^{86T
zVi)4N4tpL~tp2fx-K(jyGLV8<%bo;PN-L>_1i9vzXB8HfrH53!T4k4?{AQXnFP?s4
z&wNz`O?rfKU^duq+&zOSN&RN@-oxeB)QX~87;oEZV4AfJ9IOFupfRlb7RG%SymY;c
z{Z+P9cO8`RP@je8h7x<F^<HBHEJYlOxC#}Re~(9<sp=GGhxLOTbt(ZiL8_uxVuY!^
z&t_;27YR=lLN2D$qTeFNJT{T+#YvP0oUOfTaZ)Ch?OvkC#`3m<a&XOd15*w9*o;^q
zqO@StTFwI5YD_$_o)(yMM^VFlO+658eQdevSP{hI)H|TmPpqLeqPZoba+ha5PODR;
z+qm^0>l@jEC;!qY^Y=-C$yBhDyBt$l-!j<{w5ky5``&9O?nprf0q;*LRChymucT{a
zYQ-JZ7kSMK>m6Qc<8{BQ7y9?${95XFl}+OS_GYWYeZ;eXsQM%|NKM`M(2C-mWm3_a
z_7Se>ExmK(P<n}A=G4K_V7JDr@&&pfhk0Sxp@tCllfpiV1%U0)9eHIS3nyE<D*LZ2
z>-+EmR0^YcGt_$A`?NlANfm%4{>HMjx}SyRAlkKO*C)08#-){@-a_+sCB{hi#o(XW
zg?06#3`2_ZB6^R^u`rVmw}dW!IgvE~`{hQWg8GDotXnUovD^pCcZ8Ids}}dhu~cPV
z0iQqtYe1GdIb%q+VVY0|$5FBtv|_r(`a)hI%Iv(H`-oq<E1N_A&PeFHM~o=W#2GX^
zCRYT-xlx+U>O@0A>QCXKhYuk)*0vW87EZ$waDqx_z0rve<<IVb_OSMGJH-cN5nC{J
zR|fHOA=%sF_-*{MqrNp17wa;LjHhs%`|Tv})B!>tvh|>1z6z_rl=27EaAO%548Ad2
zE{rc}ew(tHwW+55t6XoSdiCWp=hKP%fN;>LK>P++2}$kuZ17<$weq;-3`ENxmNXvn
zY|tK`ru!Xu+hID{mo)tNq*&RSN=`aBw5c4aYL6U+OIV%|gcj{^D&!rO<n!%kmT*2P
z0SZt_K@v*)`TJEqEcwq0{)O^!X>WZ~iiK=boQ>YZP$Z?uq)_~N`uCT_d-?M>;b9ki
z!=)PWxUGLDgIA+8#_WWZv>(PkLmf~SkA<ME%SL(YJ@!q%L5A&W!2>d5n!hK9L6V<L
z;Fv-ZR9cM>8zWI>?rnP|VDi=6wxep%<iek?rT7@26DBP}BiJNB7;ykzN}y|}M?L&T
zVX>qG>+)WpR#I#N7SYbxr#v=hIl6NliRylj7i{w)(5MXu!+jSTd}X=;1vDY#%A@u7
za-{jfZN@@pB<^>j+eZr`hV}=Z(&K_)pRQ}Q+k$D+`F!+qIiYnkTAn`|;K`q**&s^4
zhri(bVBsmc&XsefJe5lMRrSJJbv|E<qj1MNGM5HI0{!1E*3KCg<K()hM`Fa!@=-@a
zy?3QF+|5fnEcA)F2~l)J0k2FY+D1B?F%_$?DbZsea|)9J>in;TSIp?5Y$S$Q3VIxS
zarZsQhxe-7-D~~f5C&;FAu+A2_l-f76=vjSt9h$3gejB$@;QaMWv>i=;y7GaxUYXy
zJoe@8O+KfTOaR3Y>0E%G>+bz|y@Um22Cxd7Kx3u8M9{S+bt=G;Ot$FDnQCH~b})6o
zsc`p*0CvWGOy<Dz43#%1$q?(UUxsJQYUgu-rd>r}(lh!`PB+cSQ33)j*%~W|$yvcv
z<y2uV=$H11W1DnU*f#MRR_W6Lk#t##cBF<5QYqa$DZ3zQ;yI%}#tl(BXwhxlq;MX7
zHYAc3;1<vqnkahrSkU=z2GljH6M-r?TxfCm!65hU-IG#v-{Qsibfz;Ix*YR?;W@pH
z9aSdSzL1-D$BHYh(uI)EAKcsDC48tBOR`85kAcp{($tF}ypL$f2w~>#q9iT#v8w+^
z)K^AD*@f*|prnFyD$)%OFof^|j&w7WbR(TIQc5cA&@J8FEjhr@Al=<5CE?kgb-wTX
zbuHF}&+L8QR}$Y;wNDVAf1<Gn0pmz{H|kvICTddYvep}94;~gq)YUAQHkvrw*DcLw
z5?=(~CwA~!Ix?iX9ll*kJW99~{a7=5BkXf`E=3kkTi)`=u+uVz`&f;LHCAQtlbX<k
z#~UeCtanJ9#$dG*ox+RqTbi`S8Pq88!Bfy&<6<?$X#<k?!6BT{^urdN4VA{pfWGRJ
z`;^)n!xxcJ6o18Xn97NO{{B`e5e&HpYkU1m=iP(Vm>%k?nYwf~?3f3z)Tr1D4)_k8
zRkM|`m9I}9wpM5LtLlQOZK1q^;B;p8mC%%MbaqR}J9N%E*;+kOf{7TzAj5<re7$|H
z3Zh8z-LyA{j$9@L`M85Pgu65<A{93HMFNH<&<!tOAr;&)5nI0Kcur9u(J^5)U7^W!
zruV4?ruyWs+87a)WcFg&5pI~{8#~<B^$@W!sq;4vVJZlb6U3p*w{(`d1y7R2$5tB4
zxpdRL!N~Ae7PIV;#~$dBo>SR)FRO49>8bn7gafgfMb+$_Pjy8k*F70zv<e<P9BLn4
z>V*I?&=1z(GSI;Wn<SFgTZm|QAg)1}je$^6PS;(-7mf|Iu$*$8kwx><i^9Kv?97`#
z9v0DNUAZN{5sr;rKIbv0A|$S!iX%tZGVjX)obykZi1cM0#<<7)hcsMIkjR1XuHpgy
z$5D~tg{R>9X*srh7ioHPlY+M(s&iyY_=EK*q414p{o-Z_ZL%Ze+mp&<CKFm^%PGuS
z0gB%gB$OPK7y$-|YybV&04p*P@@w5~VGwxE&9>HBxWSyGeM`gLX?1(!hC>{a)+GAh
z7*WfqOW)tyh;71amHRzGK404%RxXGTt&72xz8zLy)7DlruJ6|!1?dfCEQ&()jxE<J
zL!CJ<s3zX?ms0mb4H)*^Y)Xm+cwC9!lYV#8P$xu8)Id{kqE%Iv{3?loe7Sy~M8_-2
z6N>A8>JXkW{<xszJBb;d{{%TiVd6vPTa8%UIIPW(cxEn@@1baVE^>&xe*O6Bp9Z<e
z=zdA^YRE2?fdZWK=t;%2S!gmTQiePm!;#bEM4P&{cBEqZgErAmgr1F}=eLd5Rfj@9
zC}YFZ*`N(BWJ;t2sLCf|-wD=%L9Nov=z|bKJJJC9M9oE^=kVGquDE94iq42Ii_R^a
z{*>sa62q|YV3(<i>hX<R(rl7~G7*06%hd+u8LLL-;Pzj9PbY_=dfn~K50LVo8^F^y
z(1E}s<5Gslu(>{(Yj!5sFAVmzwuDfnY}U^1x#{6EiMF0hkaxiQ?IxF>U%xr9sgaf)
z5roIMKn6)hi|yiB{~p@Y^6D_S=dildh^?x{&4|Rm#U%xa`E%>duhVjhkL9@RPoK6r
zcN*Mwa+cDtgDgtzNafhF2}nHvTi#;Tl&GYaO;><Q!zfs2z6@mEJ?8YrEyU02XGkNp
zojqT@$42F(>(qY3&CIOD%_!UKnpyA1)PDCv)800bZe?Q8^2s$yr8qGcgCYSJJG+`K
z0r|TQgHPnMcO%N@VgndUKF5tpWY<`4hD>P~EraO?++*Kq$QG^32?HC8h>hO*!Bwzk
zTEF|gVPJOQo(mO44fiET)v4xtTG`F)N>$HJJtP?@x6OUvN*((>UVtY2LCp-mfIt^B
zu?3wT;@Z6pRH`$}xxixfv?UD>bVI`xR;o-%a{Jq(%hPzy%&?adFm0t!m98R^uS(d8
zV~YvK)dB}{0<0;)-p?+}#Tkz$*YLZ6)9x1l@K=8rRZKuw*y@=JS*ZpjXmB6;W<B%b
zA%6Pqer^aEN?FSqG${Ac_dpH0JGH;$;~={V&w3p7!PnoP8M>DCP~an8SWfZS#UoBJ
z%lNH^;B2gd+<jTc7)`KzSrC8AXY6AzsI%HBLPmq7B`33T(U}|at%<EKID_>#Dmgzb
zqaZlGNqN{MAk|`di*>WGj#Hl%$ZRTYSI8*PT2&Jb1r@()rtM~@0_*%XjU>MPa37u7
zFJ>;-;^Zst7GnXgm|4hYqn%ot!K@6=g2}*HEDM@WZ6v`K9~FnpD-Xt&CG!5G6Jal!
z!V{s|SCO?led&mux94ihi6@d6%NZ4i=<nstFrmEZimQ<c3GT4^SzJtlr?$Cp6+EO*
zdDL~Gh~UL?#0wLi-4E#<IrMZGbRg%!aduIV)MTsKH7z_f>(@Y=lVCq6GLw6>^$ESQ
zdQDBDGAQ=x>?^$j3js^0#fPS*$?um9figqoWPf<lnrAlu{*FuiTfTQqt4mA#`Nm!U
z5gNK`Ncf;^()DlsyGXIEPb9bJ6C=rBPhmz+f0)G;3CTpfMH0_x*$<zsksKzn;>pEj
z5rWnwTW667cL*|kAyehW@-JJ;v0-k$n2dsW43@-#qjonOI6RM)^U(ktqmtW%0PBhC
zUI?+lLth~Qmtw2qIdhQR1&?r*6H&{(eq;55DI^{iuUB^{@_ywruEO0kFTI7c?niQ4
zwM^!KC_Y`R#v$cw!##JpbGu*jffdm_{mYOQ7xLFSU8dFEoyx>ICE5}Bu2gDLaN>-&
z{v$Q%71M50A7--<ZM8H*WF>?`;m<bX%3W4B(tm1RH1Nl!6EL)^c{<iAAY()l9)TE|
z7@bvKJl;>Kjz-T$;V~IS&{D9QLpE(_^q`4J;e+0juh?W;&JFvIMT`q=PwTC{bgJ2A
z`CRoqc$e9pqOt@%&vhc+@6uQ+NVV^&46Bo$$!7R|wfdCWws_qqx#9Yuk4_pP6ms=@
z9bbu&AQRqS-GUb3c(Q`@S+yuRGVtF09I+2fb(l&_qF^}VKm&6&)qteAw?tx$dw6&{
z*5JyvrHuMd^NuCc7{_M&@2|L4(pu^VtG}QR=O~Z4e0B&8JP(tz*b#_aG^;FP*M0ET
z=x2|Tl@`S;UW!K)E$2N?@3Cy7_W(F*JuUTyJh)TYUGA|U2)?{Cq)TZD%jrx;#C9*s
zd()QUi-ZcTIOeVzlCv84K+K4<_b!@+66j3G<s5(6&#du1$YW>7&nRFe`R>xIKV{|Y
zk=dJsat7<TQ68&bPFKtqlupOnG4+E~4#e59e^AstF=Sh-^^Uq3EIN*`dRw|8!s3H_
zY<$In2w|pL@;8mCVReeYzycQ6M_Gd+eIrEM@c>Ps=c=pxm9bd16YlXPztFNkBAT?Q
zJ(Qb`M?IiAXDzn0*2d1BpWef5%h_k^xj6`II&z@m)@6j|^x4UAZseq&xa>LCgm#td
z(lMtL_{wT|itGMh!4}L|AM~LO`sOn3q7y#&vvTS{NOwo@w;qc!sbgT8G^uy#ZiVcr
z&VmZh>2qLa1Uvf4&+N@_xOoz75!uRKyjU_JgmTX_<&c_1we-a-@=k-ASSDimJe;bL
z0=7KY9Ip_Hv5e852WnitGl{ob#t>MLbIksjEZZ5<{Q(lAC7(-G<cMoatF?_o?srYT
z`^n5fUTh3zUZN#a0%zO#b<^)<(^Y~gLF`dfDD`a*!Y(UHRrwm-C@|xm>>fTP=cLnc
z>^7fW=4!5jq4~8L1yi#fMMB)6y2o;*d&^9MsMF7yX@g+tDVAY3P&vNaMlQ25%F2Pq
zD1I{<0V^~LURIujrb;mvysbO9c*A)?-E!sL4vEB+ghyu~VCvFkZO)V&{0$@#C}YFv
zeZ<Xm>&@uD6=ms9-ap<4#(n-g+?*ghy?oO7PO7Wp{$=@GqJeO{BYy-2X2bN4{e)!s
zSM5bh9~L`!-!|<g3Na9`##F8I!IwP~xOK7QtC(d=N(8R<et{)#(|7X1lBz6p6sI}V
z`U8X`Gv|CWOS_G3F#Zv_N~{X;$^*{AVDGWD53S;36I^6E;rS*Xg0qbLSwfg$;zq?R
zz2r{!@|9&C+Y)akbC@V@e)s=YXnYnn8cfdymf7&ku_}hn`C(JZ?u4&{7U^FTG`|yt
zndL`nGRS5RpW~L4@&VM-G{EK0P2LN(b<9Nwo{gFjF)7uc1U}<S*1oPq0|vnlaHB*D
z&1*dG6&+ewz~obs-;|6)w$)HQ>4lPXww0d>Y^jPc&FGjF*0~R}V8<=znwJ;rs~6V&
zMd#w!xbI4J1KFH^-p0`6l;6QMH@)Qk-2oC-h8vJ+)$`bbja_Y(OSDx6%?G2*>JitD
zm6$LHBoEIYvyKNj&GZ9TLoGL7p<Grvm@0|KZ4O;b7md8B*s6a@el{OQJh(ecYiYM{
z7JfSkRI4%6cfFKM085H3S<(`c)jn|`BxE2jNk_u5x%R8vxv|MP2`R!Alh?HZ+rWvj
z`h5roOd2*2!PI~%s;D9T-D3l<TNa<?01jl+(l-5C^BBX$px0=Iajh-Dv%DYO4@_$u
zc_q(wJ}Gj?1)c~g#z=sm9$N=#H3V9a|4=8@USf8r`8s_V{OWQ9Vp_U7YOm#6n0L&0
z%572=TvtC$XjFl{Y9_D$Q-2Y}%z#75Ea2XVb!IoaJn+GL4H?&(pcnF_>Kyf@Q;V4X
zyirpVH0Tzg`U6hDoR&~uS3IyzsS`4t7cGscix;rBhZgnZv&K(uL_xT$$JMc7d~@Is
zYV!S!e{r^ZF6>~pjt|r^(1le}=LlNJ$!zvGeOKBl2(-4csZIG*#r{f_Sdet_O@cEt
ztlTB$_6s^>*4FJ4zSLMe82XD=h-TH()#IgDDqomT(9<IVO8j)xI|Kf!tCSEjZcW?O
z(j-od9n?XuSlHoS&ixVvmuDNdAx%zEt<9$rndr@APUF<@b*F|&P*5`c`%Tmt0Nb{z
zZ16xeh{uL=OK@Hvwiu@t13C5NTEISkbD&C+4DU$>eQyZG9v7^k;H!-4=v#jj`137(
z_CWO4ppMUQ)v|@?5O<wu0P`3VuX&k$V9g_jy^NyQgCDWiuSIBLp&=#%d{=)~vA7Oh
z*H(1fw&~~;<#g3RFgzHlzL6{NW;Cs=H{<)|z#_A5CazzdWUO#CRrRL%ct>4#+*A18
zeljw=xoKB&y0aX;U(->W)(=FB;Y>nylcQY(Y3ognO)m<f9408t=cm=UT~Yw%pIiF-
za<rM}FolzolU?fdH_GPb=_N~&ku&y^QbaOmcHP3Z*3IVMnN4r@>$}I3QW!7(uF@K7
z2Wwrr5R9Gn>kMCIqB{OIl~#iqTyx)h2pyc8kB#_opEw;gj$8=8lVCI_5*?y2G`z+*
zWx+kZ_f*$Lfw=!adgHatn#886*?KZPuGLV~D;Cupao_a(y)#{1wN)Dy0`m~4$sSXY
z@t5R3j*2?D*1vThjy>a4NAp|G3fWF(F$Yxq4gEgp$}zG+?EBe0mQgj9*pV8aHXDO@
zl<(QCVnDjg%MaTW!GCPBUOX1<vCw-rebGe#f09|WNwN1-1nkkgeY>HmF*U_wjxHpQ
zgWJ63O~_<DlrqpB1TtV8(f4peVs2ZcI!31tpt=V3OHAag*ABj_ZRYdvluMS9{PHnZ
z=~LvXnXlCojb$e#x}A)orGXAXrEbv<A*en^Ls%k%8D)RFe-vPA;NWfo49bQji6bO}
zbL9x1>3(efvBg2rSe5*OsQ$Cfl`Urutj3G6<Wy@1#PH9`EN7<+n_jWY2{j^O2>b-H
zD57(6q8MTo&vT*%x~JA#I%-q<uhund$gi+2h92JSi+zmxd-5@@?5paEuCj>01m&og
zJWN+VnmDaxbaNqL=!%Z$admc??9!?S2axG}fFCDNJ_L^I##}bhkG4PC*q8jGh<(<d
zH&PA0`o|`T={ebx{kSLQSlls5Gw8FG%0qNH{oeZ0s{3@7`YyufxIO~%S<Lu$p!F}9
zQoA*%&D>mz#Z@#!QXbjVP}AVYsZ0O<H3BK&;>Vjv-1wMM;n>$9(`mu=5m-0J0!*Pw
zu{<d{&N<e`%u|+)5cZg&%Y)RE0AmnwE|i#2vG@^$-y3JcKs+@xCph1lbnOfM00N4S
z;3kZ}3K8Z25oOj2IJMzXBFR#Kb)%A(I3YZd__1v^Ki?k`Y!tTB7lO<xdYiUOmeJvi
zTO)iB(4xio5Gauh34-5$B$1X9Dy278R3o<0g9MKcx=)VQ`LwJZ(-&0!G*iEuw)UE!
z;0OPYQm0!P+)pZf8*}AoA|9Hnyo)44CKdh%kBK&eWiy-BcW&Q$m%HWUl+Bm_BjQMF
z-xncO2SxFOYGgXbwK~b|X=Ij1-CT%ZIWTL5q_7%>K-}49q*YHS>OE>!Zlj1*Vx4V^
z+crzit)N0r1MlaaKG9PvcO~b@lt%Ft&rcOHzbd>x9d}*wIr%pFappZqwGV<{#5JCX
z@yPBv>#A95VFq%3zD2}WPFJw>`998!6i!m`KD~7=DHWL75ds+Zp(xesv*$!G?9~p$
zTQo=`jnN-lQ5{-6hGv=UCUY>Tm{*!PdsWJeLFyVd(wUEx7KPTkQfj0LPmt-fm1Ad@
zYuWflI$%&}Id%`UW)9WhJ<<2aYREAp$K*P!0oNIzKq|9)D$f6K0*EX(>K=^XYHU9Y
z6WwrJw&*+p$B>)6y8c@T5K*z3vmmdp+A2<vka$e1{l1F7|D^;S>J4GHXfuyr<fl1s
zV+p=zBuk{yQeR@iSmhR%45rL}_fVInuFCcy44#GlMTnXMY$RasGVYb`eXOsXep@_W
z=}jlTr|x?nG~;j`RBhLR_Y(<YK9<}>$5%bu)X>Mabgg9gIaXYQHlH;ghN!4O^ZDFP
znH$O0KgH??x#4h+5C5VdqddLqm2J_Cjx;4^7~hj@9>k03s4eQeXZ&kA@)T-T6{-!x
z(1uTR<$@`5TYzKkldcF31K*D#S#_dVnZb;S!|4Vh>&XfK22q{PCZ8Jml8#Y(TS$HN
zv?+lZsa+48Jzx>=pP5!F>clxEj$TWSCYil-gYJjw|D}6)#l@O^rQ{w2yfic((_iMx
z1DD-$S8ZW>1%>P)*+Ehy10w%0gN3!jOdj<-aLR#L?E#lvt(BJ0qC4az3pT*p7AyY?
zU}u{yJp19uxO@OW5$g-7e&3?NYu{{eL~zzskGzoY65Q4nx~2{~q#C&T#`dO0nA70K
zu&4DbVE5+yu_V65OCwf?86|`5qhHr*qq@OY5?@kk>ZT(sc+gtLaT!I=5+ue}j-hJ+
zL;_BY{s$w8UVmiMd`?nKN{T!HUOyPi^VlTKqvLZXBGe@B=B8M*_0Ql|(80?|^~+0r
zQrH&*PsgOz^MqTm;f3olkgT!KwXohxU-S)gRDAYeN|)UIEuzhP5=8pB*ifT}>&0{&
zc|(R>Vp^=hBTEDem3}(t#XpOmRyVh-s;U12Aaq)tR&T<iNH-w;KOPKebP?O<Wsw6~
z`o+MAI^ZaDZuW{@6T=7F=+|6cZqUMG=EhHjI61glC@IY?jl=84oyi~^=M9`mwvXl1
z>cw=A9oBFTUk|zqk<f`_D7W+WlGm-cdqMD-iJ2=h29q_8aAQCquq*B-JU{&o^V;2|
zUrj?Wz0R2iMV)AR*Om*q9>GFXX#;kY&LNW)>j>7uv|ag_TSAvr^J6}{5<W`=W=D1>
z`VI8`@y2oT2&*}T`Q`Wjq8UV0nb>68<p;^duDlg<(pka7-c&Ybp8Klv!4S7HU(abU
zH`-kYzZ{QRVfBHyi9zVZ8O~1oKPW4%Ay2l8RE+Gn81;Ik$*^|ys82sAtftE{#z|f7
zF$5kel=|{$i=I@ceHl2%JICGfcZbj9yV($ri)-2WllU2kz3G1J-DHS-hxUyj`Uk==
zfe&NR{V=(t(78sQ3?#4W#lexs?+?@Da4ogVN6N6qlG&{BGTzEddgvwPm{j1trU-AZ
zrRAy<ak$!B7sW#@c_x&NRJrd<LqCtbXyb9cfp+k_+7$^vQ((-A?aTB(R>_i!>1c2n
z7+o?sES%{5of2!>(&{VQu2`4DI>=!v6~7$8XZ?2S8*UpJivg&sK(nL1Y|g*O2G`6u
zml3vJQ$b|ezI$IQeCLIzuAhF`>)D;{+@>q(x+4sU*<gqWy1(h|YheY!O(`l+{*uAE
z-9Yl5+L2N)r`v8aJ&$#?$+&cP*j@|BpJ%0B>w5{Ic=$v=Ru5HIcX#=`k0ueJxXuvo
zHSKpdTN-gxph+%`hiP-r$MP7feB-{{(eYz@xb2_bqh+{~o6PxY<I6q6*%DA?awJb(
zu;V))y*9ey!>jGc8v$71f`VZ;JV5=;+9u7XqJQ9%Jw+zJ;Mip@N5bhdml4rOVz|Pt
z=A-MCEqOn6Dr-khyH*PN54#jOC5970)1SsN4!?aBk_A!6cxpd@ofq)F0N7P#4Y-y}
zD+2jUP-w$H#$74r{<U75cSk4hI^(P>?mv2TVuWgFyk5k?N{}#WGl3%a^sBvCLV4(G
zD!ELnJ~=vBw9dsEuQo%ISH+87S#U+v&3#M=l=$@#mtd4DBO1Kdb$ve|mq_M%#h=J8
ztoIlNns<=lY|yXjWq;-UqBUui1odPmOfi!CUx}!#Q%2<Oxvqb-S_uh3WSSj2Y*qep
zyt<lwcsDz#bcN=!;SP3M5Q;ZS5Rw}EoYo?`)C*l3p?@J;L9zMcG~m7U&C|oDEGMB^
z>dFP!=y8NmZdZrVf3v?3c5=a!-_rg~Kzw}Oy9bdzXO(dX$VtMVZ=pUx_t8*`VeY!>
zjpJ(6DQ!dU1f`JXWFlE&U+Z<qM|bbW+s88@K8ZL#|IXTb{~UoB6KsMegWL-b_6+}v
zLC-$IhVa4M(sO7$_PzRjvPUxAQ>3rQc?T_LFujI=Pac4yeT3kVEmsKhmOO9kR4T)B
z#JbRx%zA=oJJuC)P|=BudvKc)z_J-qN8=}W(^-9A%&}Ps*t6ch1?@j2Z1q1?5>F4d
zkpPg5e}onO5}H`>0Ig15N0LPS56rf+uVMa@Y1PTvt4ti=C-VZ)F4KKXM<=sRl`19p
zDP^R20`g=O35(Y2RM;^4mctC$r&w0qOeQt^Vx6$(@+geh93wS9odp{JPV_y}AHtS;
z*)#I`8syA?>ZVQ3gW3hY?oPmi!|TY;hW(L0si|(vm;{SVR2F+sli+EfTzC!-`mNzY
zL&39L!Aj^W*#Mc#W;+x37r8fan>GHh+A-H7_7t;Za2CGS$<f^ro@(fGU?8wbR2?Z}
z;OkU+dtGE*Hvio!I@?-=>WWI~m27q=kIO_7e<O8n;0`J2r&f!ZzJVM+>%8PrLYh#o
zK~f$w1t|vw>eh*m20M*aw#^$7ebf3H4BGlU5rloR%Nyo?9ae9Vn_C6lmN6C6{Jn+2
zOp<BbF;>6UQ_+4ZJ;eizHR<UuX0{6bFfBrWo(aUK8d9>p=Xea5X1&h-c3gM>n;)jI
zZAe}ov_!BmJpa2XFXA;M=Bc8WJO^qNqduF3?wTBfC9q()N`do?$``V!ruoN3&ku-m
z?p6(2(O@+i=8G8!{Gzu@|2M>EKa+h4j44L6E0AZomb(|<Tnkx8vSA6PzXKfyq0;dI
zygBN5mO7YtQV^%y)A&Y7qY?lMh)6M^y8uW{N4Q_dC{Abobw2!+UR=S7D&A90bg%%v
z#mzwnmMv=ORA91h{NkRY@_?`Z)4`jDudE2}-`n+!wE{_@){`31eU^NnR83r!QpmrN
zi8eHZYO$%P%q1+AgGXwpC8FR7mcAPC4fCoYjgzQS51SG1{zZUd>qMl;H1zp?TG|>B
zOxwa)_V;|jflwj^bMWVg`=h!I_cqzu;)-h7&l~VHACC?=yTnv|&Oj1W<n;L{X+-z?
z0+3(#QK10)LsT?nOLk%Z1D=m_3pS&t@-62M{Q|TK)+&_&`BqgnU;Gzn|H%9KQ|@vA
z6-D-D0@U^5c%Xyaz|&qjdb3DCK3Fv)6!e@AgfUXuhScH%W8QmXOV%OVT6aWk#|Zn{
zK|+K|ZuTRY!7$ft?iS~!wcAtTNlY59eK$`1?o_j3CzpC$Q&5Vrd{;lzD`K2$BALw{
zQi;Q~B0!_K?I8q0LvRRz_d3_wxgoU2IRK0j)OXuS7IV56TLQKV(8SdbK-*XBZdNLp
zO5Eh#^cF5RD)Dki?k*r=|7iCxC7qS9IhPf?p}+w#&@lm_3HM5!MQY_05vdL953PrF
z2znz=L%3{?&$oSHkta)Yn**y1#78HJQI!Eh^?y4e@<{n9p{|N&truKPJ?67_mXB^;
zO^A7|e)W)~m7COh#8OH-&>Iqr_6o`|ae(3p8f4GuoVp*Y23bFhpxjjeIcJS1&}h2u
z&iM)wnLG!J=GDpe?aCW^FsVL56jlx7-e07m;g|VKaTKvGZL923;$;fV*w4i_&in*X
z&f9n7iL#5Y>sv+)i;Xg^WxbY9rgrOb<9H0}#kwIxNcF)|Sp0X=754o)Odz>v^oppl
zeMX-lXe}h9f^_DM5KIi<WreG!Dn1f};HU*xWUOwv$BMTaRg5{36&oLfJqXWv86xy|
zlE;N2$1T0SdHlKWXS=*21j(C8QyPoV*imAUL=2IHw?%{r>huXevoAhzTH{ILe!wOz
zce;M~bdrd|PgT72CH1&t<gv?|GdCVXRK4$uUb2#`)x<brN6wt&L%lbXhOLJu+iY({
zJ*tB;tvoj!2HE=oD%?4)j-PWsqFaKOdzW+R6WWwuz0*o&_Sdn^LpimCpnS_IWHMyv
zo3rEZPdTz<yj!1;j=H`?aW61MXMZa3hLgU`t&HYIj}Qe^q#l5drGZ%3801@Ua_;Iq
z7<{`ClLKYiXmOl^!dZAj^5&d%;3_+=7Mg%+LO@N3*ORYV1Sa)Fp--o*gy$CQc#q@(
z?wwwUeH*9{ODaNE%*@KUCNIFB6spOx3J=X-1@v);<LBZ+2zAOpjsmLkAL&NB9-ru&
z|I&#Rp8x8?-}gtxk;C3=2ZffS8*{vyq07;E9*;Lc<+RV4kYvBAT)ofiz3)&JhdvHo
z>i15TX^l0WO#RmWPMS?fkFR=f116*Xv8At|{WZha*$F-(xM6>XSwFbze;nP^+|yw4
z4px(lYfPQ{s^O@J;?uI9F8WZm<2KvP)AjqfyDhX-EOE5TRe!<C=c$%9FMF=Igh~(d
z75VF(b92A)G$#m7!sRzAW8y3rBHhT-tW*INGa#z_4SJ?sV}TlR<XAPXDpYo?k#_&<
z+3V+=rp>&~Sj$h*<RtT;otXac+=bvZj=y-eHERwd*N#~-56<gYzUN&1{=1~v>s_z8
z2t>M2T&|b>#M`*+nW3wQ&Z2NNCxx&aTVPx$q+hj3<<{+5%l<3}o}Ks9>CFN=Yse05
z=~zuL@=gVIYaZZTjW&PxZ>Y5u3YQt;%lof_iLI`3eq%;w)pwnYQ4nw!V68!A!Nezd
zp70Xz8_>AxlGg$2^QuZu8^pBP%&~rx+fqn`su=817)C}`{eH>bV?>+h>?^(Wrj;hn
z)|tJ>y6>0)YGky-9Q@Nb6T&3zgr9p~hO;3Ju39lxQ{S+m0w?}^HJP?T&I%qIDtd7L
zc>^0jHr0Uf-5-mOSEIM&98#+j8X;1f_oMz;OxTmzK&TJWuKleTWq9FT)n}N2IVXl>
zKU8@_oDy?<%@V*-dUV%|yls(dT^_av^G2JSRPsdY$vKNbU8T52bn?(EAn=V~y8l$_
zWnj)xoWab>Uy!IW9WfOl=FyC1WvdsmJmK>B5TVKGD#rWe=Ihx-%gNU<*jxSRPDJlJ
zwE0qkNJ`)u42I^}sz;i6H9RBgsa*Rf>0{>L!Y(W-f~H6j!MjX}PYub+CY?vr1D1IS
zUcE~Oa`r~O#&>8-Q!^}PP^D5G8&J^0ATK1;KIS2E&Jv8Qv3{0?;izg%@gYo1d{opX
zr8-td|H+H#@o|ctw2)}Cm>-1>^@~g0@BF+w8pz~^cD_o<X)>714b_c*bmzs|LXd=T
zuP!*qL`qAQ7D#=ko@^hbw})~8fL%gy^QAYEknu#mP*(ogE=NeJ#mL3mxH(6K0#&NA
z=VI#)J5g&4l`l5uA{<=8ssjUZzAbPKbV=9zs+syA3}UC6*vWsH^Ue-SS0LfZqZN3g
z5|;R8a7SJo!GbXfWD$LD_KTZ^bux-s>LlT$2jyZ0+28}FHSil2ve_;8Y5$z0k#~2X
zQHB#!Yohq!>TpKNh7w~|>_AR`zt~rYK~|9`4)ohf2o+;X?Asg`GHw_H%UUQ=ajk45
z$(gVWt}ue(Z;P2+!}9`NqJj$JI$0U(21y-h)2&VW-&X@qyZ<r#U{qdg@d(e(Z}MnZ
zQJw6H08z_bJTTN;Rlrz5p)tGXg-fr9EO{NH%(0%{mWX_OkRa{i;R*!>;0haZDkdwh
zc5cD&*y@ai<{EuCxyVcgQcbY^+Ea5$kNOsa_+Qz{5%Mb*pY_#7@Or{P|4Mz*busjH
z%s0)$&d=w9%aaJ>0#zfrCrMRIZ!y8+Z&g*$=$Q$Zh$e=*#-HhQv{(NQdB`Y>vjB}h
zJ;?trj_cwLkY>S<MtLhAY-faIsn32nD}Y)m)?RM)n3NJfTg$grgO4rbz}A#&wC(9+
zAn}v3MzN8EmdiKPF<v?PS&m?c@^7YBVjJ5~BK5D>?>Q(XHK#s&qcMqDv#IuvF1a_W
z7nBk%iQ?hXFYu0byzuP0O&>azZ;8#e-gBkapR;y0X_kqNOJ=82+#n>-c*&k?9;P3U
z34|NEsJH8M4kePqX&8+kb;0wNu}aZY#zEc?^<veMgh#iVIXKH7jC4)UPJ3QUIcP*z
z+4x!S;+5b%ia0IOLI(qMw%?Lr#Hxfv=D9u1fcsw;nOS~!yWD{@@rRQWTHPy#04;1z
zMcYl*bB|ID(Y4%J_FF4$W+A><SVc2b>84ow0RhCQ%!X0-2z!<ZnyeuYb#3hSeXC{_
zA&=&9cWeX-iel7<lhNb=r-UiH*_FLh+=>j4))uSMH%SvJm<)?AXxEutHu2!#*TD~p
z`ds>k%K2~(rA#E0+6IYquLeKYqR6S%*7IJy1<_+3yntOlOs3PX9SDEyPz@F?gAcEt
zs)Bn7xz13KVMu}Pb6&4-cTCAtguo0MlWyPT2(#W*s47^wgD%x`5bOwW7(pl!0Wj6%
z)3oAq{`}hW^f<bL8FAs#m0A8Zr!jmm&C5H==W=J{>3m27wwPD5QZljUhHTo<;=)pl
z+&nG1<!@Az=$^%ueT(y36%yB4K?{wJkxdnRI0&!M?#5t7a{n8ZPL^!C;%Gug;4$Lp
z46|KV(uG(Y&1m#hE}KZgI|zHdv<*K8HY+|8fjw1!{9AGk(D;{FR8PV{e9HZ5P>v#p
zo(K_pw3GK@z)A2G-+ud(*p2qOxqTUNB6>j+F*6dzB0hS!iH2Xu>K=K)J3{i**@L5F
zIlU5lh^hQ0)126;p!SoWcMhD=$)I$Y|47STxZgLv(L(f;Ide6hIXGEU)w8!MqhO7W
zjr0r~_5~Kn`}=h}ecYQ;=B>oxXZ-JLT=1zM@>IidMZ_CC9ct)$8GMKMTyu=tr|G(t
z;SDACgSU!iYs58TGiHIE4OcT%1B=VH)@lY*Duk?(e^{YC8kPAu6ZSTq$aMX6N7u)=
z{Cr)5(<w_F|4&QWdoDJ>;XYr2uI7ti=5PlJbl|~!RKDajk%%ua!7yz_BQ4-SlSLNT
zmmQAYA7D%@I5DhRTR{BG;{Q;!D{mn#9Sjgct5qE#smia{9WiA@){J|0lMIS-jN-nw
z1#e{rb<{XfXvpe8{*b{ZtB}Qdffc5b3kUg>zc+<S+SWT%iYDL<h>%9jD=G~h=j5N7
zQk>jPBbhn2g^;ls6M93jui_%USo|oUcqPV&mK6%`v!72X8!OLKw*H;?B}2W*jhjjF
z^?zoWW|g68SK70fmD~XZD}x$IO8N)K-OIl7;R-77CNGG|ip5R$-@E@$3(?-aY;lFI
zf~gh#R{3Eug^h@1{8Y}J7Zz<|#xr_yAXkK@b`|l-m*6&qJo|tmy6yfnmcuM}Qej-W
zsW;U{Y#OC5y?kQ2a=OrqJPRD)YN_kZZjplTvflt+PPzL&Oyvt*xumzwf7(ER1MDxb
zcTZF}o9cw$a{v%5(`+^IUoiF}>0^$cmc*xR+rhxcAoJoX*;Z5NQ6n{vACa`?sU?XX
z_npXl)Ce(BSDE9Kn?V{o145c!R+2FROUvkq-<&Y_OR0H<h^a4Y^3}qnDkxY=(6LR-
zi=XthM2*w?w@i>6xcDFZZC2bXAk>>{qc^NWw%18?6Upctxln*<hwGqb+H#uo)$_{1
zjeS1T@gi$hAZQruMcThfJ+j<eWx~KyL-DImyl0I2gQrvTpAO2q51g&ZiPd@5goO|s
z`oKZ{weuM_7r=ZrIX5jWLw{ss7&PWS#tQ*+!1Qk88Uhr*fU#!D@3oAI);qztKb&CE
znmR;0vg&TGBsORjTXrWm`oPMyR|VDk`nkoZps3bXr)E%qX5g|tg|EA1?VB6rKQ@;@
z8oSRUS5ANGL#Fl7pJm@7X5~mC|Frt@ZtI}eeHC=lY9!*lc0+H)*Z+L-s|}B9s+-MX
zhnJKfE*;ylQ#*0I_uJz4V+Yvc8k_#Ozf&cm3iZ+#)A_!}ATXf|#A)8Ex3qTm0Ob(p
zHnUaGXK*#n`LSkZZqD*f>-;ut?aW9FUE1W@f6U&kWJ+bRb<&uBPe#S5;feOC4JV#%
z%b8_+rsIp}_oUoo3kr~%#)(cGVQCEun-lLsxT9P`qY&(wk_tw{+`VkmjYvWLV|Tjm
zQI#*Ew90e3ycdh79a+#?S&sr)scW<&oer2RdoD+-T^dS27DQ-AXhDINbzk0@Cqv>A
zxLZrkF?O9=zFvVNZN8FW6a6T!j`!$tI?38|V}@c^Aq3k$NZ_Q4AqnfEau>uD6)MW+
z1KCAUMsiBYz^9JxYF8>f?B4t|*<4;W@~dNvnXU=58I_kMYvo*|OOVo?P*;pu&~?WX
z?V=kcuC{a}uIgZt*>?J#BULd|C_om)=Yo_7)heJ<O-LyT$?v#kV%sq4i}@l4I#(ZO
z)VkmTxtIcyeKq7T%$*lsn~~g#(?5`Q9Phw&o&B~y2LH{FIr%1#aR0Ecv2lTysUN0s
zkaKonczDgPZkQ+CmdM4g&k2L`@*P=l-zW5j>QGlT1c)kWlR3$L^;D!<tR#-UkUk-t
z9j~8@S{YGtYB_W77oZs`gVdAJ!@5!iFc};~B6_fUEe*N_)0*bi+muy>y$?3PJ(fJm
z4POv^zr0N-HsvM{cyv1Xgm)MKj}d;a(BfS7BV#4E^eGT5KvU%u&$Cm}?lIiIpX*QW
zpZt<CIVm!v%CX_MwX@|1j7rPp0lA6>5D&&bCYL&Wq!uJhsSqeGJ4vA0tv|`YDl3Y}
z@5uJ{GmAb^S`__4c(?0w%H;DuTr@azXa*#({<~n8*J~b5^g1kHmD#(T@xub*>0N6u
zIWb0{mb)(1N{?Bd8&o~@8tNS9Gf)5QSBg*J^5OIEj`X2ICG)L@c8k*Apy-5~!ygYW
z<M>e|WVZ?*tmX>OEaW?~28a98C;F>xAI=V60-jI5V{M^6qFBy!27Ig^MqvEB`c<o#
z@Q&Q}ed*u1w{Rj2(9umIv08e{X{)CSoeBSz&i3dVRjPKF?eS`80pVsv0qa_JlgPOT
z=cE9Id(I$$NK6SbAEf5v7+@<S0!g2}=`tS!x|ptQ5oD>br4ADDp41DWLYRe-31`nI
z9D4@)K2^UT?jo1#-M3m^=WK5Wnc9P@kAHvH&^|38<OI0)z#ONR(98M9&lkkrbL)R|
zL-A$*H@h9YfvV18t9bI9MoB`bM0|`_>n-^#^~)|=g;zDgBv$&^T!lyi)x<cPDyoS@
z%tpNtIei2BAB9NPlbPjEz!V1-US|BzwT|CAgVDM`BiyeY1|R3j`Br7^Z{2kMqA4*u
z3JZE3=q-;Lu@Ypw_6k~T5#G2e=5NUPoH6XIFi7%J0cF@AE2#apE~ndj5sYxaO7ySK
zAlWg0vtrum^1LlNEU>rvG_t19WCwe4GU>mr4wavBvm7pZ@$uM5RKo=1YNb)~E?eoq
z2^w+~1%!>nK(<_~q>+bx)4tCKX!XOV9kdOk0y+<xUGY+Uc1i8ZCc(`-Ef(wbmkHrD
zHFN@4jo13woe2HI%|o{ZVwjNR(9c!?oV=$FqGvl&TW(>{?$y`C_5~`_RYYiU$B*r@
z+3Nc8k*osNgQs4)YMX-^g(QDN@HJ12<b=Wp-;J%f5FEQxln3*ITNl7R;NR$b0Qno6
zv!adsacs%_=%HAQ%TqZuU=1wH>Wp{|fkm>#rdQ0nDXZVtuBGxzwNTw)I85O{+!}UY
z@roDbvH?&~SMPMnt`TV9z^EMM_@9pYo~sQjES=f^eO>VDRK*7xGK1cW>w_9#&_LOt
zUUJCAs&u=@IMjraaXwE~?3h`iV<Ajsu}r9W^^E{HP!aLGc%1`Dam0bx^MSqkpbkMU
z$4C*yQ{TY&=OCO7GpssCA$eUS!M|V9@i#b1eAzdb>z^t!JtG%M!3_0#q^5Q?zAjqJ
zPYIyC6X2#&GzY>*iO85oy&ugKhBRhJuJ&R?#e6KnQmvomjOxMF$wMAeDM>E8(USp|
zvt#zKfmqW*k637KN6~-zS<arWOQU6YjNa|FX@l<2-!jhMDS=Qh!xze?68WTI3`S47
zklNukx!me=9ZJu}17TKiaK(ca$787e`1s+d)r9qC=KW?}P2UKF5sf1vXVGX|DUUr9
z8(tt0sbN!PlUTzVN28Wml;uEMYJ5|*##G{_@_F)U+@(yP#_YQD<R|r*c3{)e_t$FZ
zW_^~q#u|IzI!j=io3=K$@F6mW+kxYz>P7F{l4llzlM^~O^M9{$|M$G{sp9g@C~}AB
z99KVn4*YS08qbHu5!@%otI195%0xzkS2BUtTXS>BaFuu>Uk$7Hta_rzL#m<rk2F85
zvD4-bL9gN$ix14q^su%u#~#0t-~!}|UF#<AnMuo;%OIUB4^fO=N^AYJTTciw)xyzr
zVAX}@xNSPP#>Bx)L76Iqdp)h$8LT5~x%Rj~e-dQk_{|KaP3kBydMF@|Sz-xak;iX_
z)8z#*N6mFq<drEXBs;}~+PVqxQ`%DM5zi0gwUo!jc-~g19<hV!$pSnwr?Gcm%F}pt
zyVR?nHE5&e0%XJ^H;mc8(MP#cV21;Ll;3O;2N1WUtO_2c6#+ej_?U64qSf}N_Ej{t
zQ97S3flS;#UHjnG`4~$FOd`^p{nh<UyBc;(KEI-U>1g$2ViB6hr0-IINf&5D#(3*I
zYW{;0g<HIg`p*Svh9#^Qq@?@d`8si8hC}`#I0|@=O<GP2+dY_YV#~{JMQbJ**_IqA
zzy$$>-eB_X&}Qaw18Wr;45gF`#g`MA<YU|>xg3wfGOe4+X4eG7$i^ITCz6a+>-ilJ
zpr9-4jPW8!G_c^vo$6@L4M%SBa`cwxjg(NLEo#l=TCe5GQno``FjZ=eSa(}yVyC*A
zaq$!HdZlKge0wq&UqBF)a1vg(clWmJ=i<MkOc(zQNLIo^_B8T>@iHTGWso~2F56?a
zQrhxA1oyT`YbnSY+CQP!4gPKy0!Fu+ze?=f>+j#g|2y+Edr8Dg-<9NBTNcQ(zN_AP
z!`3!Vj7$_xB*G}jD4VaTuQ=ejcnyqQdPCrw$9W&*v3cJcxjg>l1x7$%zJc^>ej$>N
z{4)@nlmb?)QRIl~5nyYR25S5YeuCe>4;fHvB-V7FRb$3IrqO<>et&3Ki@v?r85mvY
zmMv+JqzrnfpOv`py!+Z9d5w`2Z<Io%1o$Z+@=-<1@lxcO@OE~2>qpVDjc!Iet~a_m
z(S*(ky_d&*Kil|mT@<k6wIfe<qOc-R>Cc{HwyQTa|BAL0RLEw>U~CQL1%MiGqqzSa
zx1h<=YG{9l-snq|)5p4899F-PK2otLtn;>4Cmjf^V73$z^2?E|jJ?f*b(znm@9wCd
zO{%EZKPg?&<yM7t=`WhpZaP4*5Sykjef6e_$(62o5U<Ru|83)<*R(52rk{84NfeU%
z+Pf3zO+{*dt>xO`x|`KXYV=<4^5IF6_M#u>js|2+#7QB(<b1Z4c3h~!RWFuMgS!di
z&r(0099_qy#-LJ^M5Hkr=CfMC3EuL`8cgWUgqs|hQvG0MO`;9~b0ZWlTkrvdBP`>A
z<nxr(Y2F;$Z2%e<GQZ@4tt=%7F<?_mZ8|Nk0*6iJ3kF5r)Y<JLQ=(^C<nL`8`HQoD
z>!qZWpwoPeLO`5!viw63h>~QRl-ArlI95IscMxu=B~+m5SRrjSL!pUDLyVoM`rR}p
zbmjH-r-ER>@^8@IQHq_JOkarAiL&0i%0v)kA$pU7UIT7MUgZgzV*sp!?-IWqRUK;o
zrbNzw#9_@pG~UV1O<x^3Mx5{QAw)~6btuy6M}ZNGGoz<Z_CIt4O4LeY<dj~oW0)8i
zj=Hz$GHiEUC`ta=h-2&MPAgVs5`(hyJ-6?^BBC8FGG)fTYn%ztMAU!W7_%pT@Fo1w
z+MIp7O)d2&z8P^i(MnaP()u%vJpW7Djs8skqR6mZ5E2W41bBBC=dMcs)~2KA?y!<p
zkwCqk+O7jC4`5K407k=ZPA=KOo#-&`n1_ocjbE`d+Ej=I4ZB<XXOkNGvQs68_z)0m
zcVJFArhBJGbnbx<e002ldD@z>oV}PzxvHLfu6p42VE{4+?&_f(D_Jl4o!IqOR!VRw
z=BDictpVgQsLIIFIa(mm+@8V2n>B2-eQqw;dTmLEe}v60x!V2ip#s6yIqH1I(}D5W
z6IezaTw?4+6OJcFnhu`mfJ@-Gd>%jIov1;JKzg36FFljIa|T6{$rd)HLQ-t52{X1m
z)gbvFmWL?A(nB8n=LHo&lF11y=xtA5+u2q<sqh`rJvbI?@)-nke)uMjUtn9@VHn`%
z5r|T>&&3~t;0I<{aBWA|OpsdvWXhmiY#VCa$VM0l=7*DtB?ppI@cQ1WN9E1@h4i^<
zq`iOl=3k38|NB{-?6k=*O`k^ADr)q%hsx<CrPbGWFaF(l*-GkqeY<N>a&fepy!lc1
zM#*4$dG)$kYyu2EX8#7CkF7V6+yf&!<)!IXbvf@)zmDaXLkB&#Ct_mOJRX<V$*r<m
zG;f0)7GTn7X4Gpwpm)i!LXu$EN<mYHh807YUrFOoNv@lCygC1)sAmW_Mf3vgb2}t=
zGoi{xFH|?NkG0Poast>{b!MAd4&YzSDfzyx8V}qiPS%;?`Y|F<un_2cUpIc+(}#?*
z>S+_9Q^5Goo`Ura0Omu4;c81RB+fO#7zmTyqFO3&?wf;q@%QruQC+|ve3k$uG|LH%
zXZ6J-UPDV^9T<mc#XCgINhQ`;KW4_QgfRy!(&XUa*7URLsZEnd>{<@G$NaEnwZ<FK
z*oX-Ya|K8EU?(ZBU)lz}OgAqpc>oqH<<6!4(9yc%A?y_!4zelZF!84IO!VHxhXJ?5
zI3hS~@p*1V+FBE$%8=OX-$=HLphM@&{z;XIUvKg*TN)X*Anb9!Ha<}=wcy?aSB2+J
zMN^@RLRD~Z#<&cpO3DJg(+>!nbRi-3kxfcXaTFX|@wU=RkXPm#Z~YWgzJo!G>~fm}
zERvZ0;fT(`IaN+!bS2z3zfhv+hw;)Dt{?8~u+@|h^?Q@tRVSY<HFbtVDh`G|^?wFn
z09uNV5ruEOd?S_Pfrf5^+j(+AH+6B_sV4Mt{~~A&_9bW6b9s;gooA?J(P;k>z*3ZY
zDpwOxSc?0;X_$_-Gy@<5#Jq9Sfj8ud+B6%}X<j#7Ocj-1>$QOB05H4yto=W|uMd>8
z>}UtOyD<Kr*$jGOC5R~?kf_xlt1Im93XfL-5PVCd!SH^#f{r>pka7)~v>vXEi_bij
z5(QGc7c5gn`Fjlo8KBt<!HgNo&Q;(13y8`1734NQijtjFOFah>AsVxUng$?uBRYg0
zOAiis1ouOiuJtdd>_mo3_cefFWjwIvc;C;)_q%?(C#<3%t4HB9!<$7@-Y;}jTlqBo
z5sr!F4$R??`K)pA%i*<U-sR;fi#pjO`IOC?Bu~#$l)02dXIi$e$ZNjjev?2p#MN)>
zTRk+7g94sj3_<@C>FNM|q<^qtSP}VkZvdm|aG5($rGyCEHl&_;iJ>Zr$gZ7MV-&yk
zQ3g2!<a`x1+S`vt!gYX8Ds}PTC`#ZDz%~GVxcXu_B@?4mG=+ekTwsENEbwO}sY;(n
zbZBpJO_+6FQMK(&o?-1KP;A9Lew8NhloX^=<cNPdc5dc9s%zRPA9I|x(ed2vjW+N)
z^?$*()kvZs#nkIO7{@EyIDP?$4`eV<>4e5%fo~6FNj&KNJ!hi>4)KpladWgK?sk>E
ztSnavN!B;r9u_tqsg(cIB+~cd%`RZQ+p##mI@A|7UqgZL0E^Sefs&5wRMe~#4wc&a
zHdI1D-A$N3C+E1`Eore{Y$fXYO}BSD(}i#mGC65ifhP39r;@>0DY?5D3(n)6DDqgA
zArbl4E&5E}H{-wX=K}}Cymt434kRT5l=Kc%G>8Go)n-w0j<BjdHJQEeK~4tUvV&9k
zLWu;tSs6yhFcUGO4U4yk@L;F5vE5H}M$kl9GKaCeUiK<k6}4Y;%UWOf4$kg~F4XN5
znz>148SI<E3SkecVutE<`j@+u=}&E!O;ilS9nW$Qy^#Jb4@UUBm$SkxorV)Jj>{iI
z3Q%|%o;^}w`aoHo7>qdJ)({HcN{TQiP)RhQ6ngXCBEdYFa}7zBC0O#Z0Y%T&QxII>
z4@dGmRsFARgii$g=Ga~>6ftYRWf7Q20^|VTuOJFXFRb)OY@cj8G=&OkcmLM&)8ss>
zuUYu9OWv7KcN)@=G+6;)J(mvPJXu5tt#3nq6tPkS0y`{whA;O(Ic^N3@JS^VYWxV7
znuw>T$+co_TWgpvF{5mn@VOS62RH;lTw$s14sex%v=<SIp#IEgt4I)mQeVDcNRZux
zr~~ni05+MnnMp1t1fg8uQCj)uIPH9oa;AxR5N<)>UF>?ZFlP+=5~eeRu*mFvr-|!t
zjTZ&wdJ^TE!}0E06oTHuESVq&BN01EAt;rt%V%t2`e2GVxKBjho%fq%qY!ZB8X#8v
zRi2GG<Bm$i`_<$))daiS6)K>3uI7VQDGL7f7ju-|nz18(gL@1fnZ>}2eeD&|)NOV4
z#_#=mTcEwXtz-brngB6&xRC$6^{#l^MV@$)w;b}lR=Mrh7c{sIf#vC+Sz06t*qj_V
zdJBDc&-3r0o<-KC5#1OPoVVTvB;5=c)E!H;{4Zg?;y?@{j@+*;2dC*;mQ=dxLvg?&
z$~=4)+TehE-{<_EV|nh^s(*FP(i7m4cDnHFEy4^d5wPXFcvJKTv-nr5Rs`xk41UtC
z7}Q+S!8>@lbIa3uydcQHU|j<1_|85Vm|rgIL=oGT-6E;-b3_C{Mc#it(asJdZXLRX
zrEn@#A50j`Y=eX;5CH+)q}H$fEiW}>+UVqg{f27j6;d|?#ZN(!%0VThNANgLP3>6I
z6~)UTI}=W%EHDYH0u6LwoCp#eChjR_4sN`WwLMpuN1QnZDkQ!VMt`&?!5bFU)Hw>b
z(PJ3IiB$<N$orkdC@@L6_lw7Fu`xA494KwY>FTh^tGS?wQqcjS^bokXCaX~Dz@~l4
z@kuuR3*4S#)@5KgeVqB39!>5}q7COgRigl<z+)M^@qlvRC7jUL*U&YvKAJO+x)}2E
z-du=eXUkK^`mYrnKP@9aharhIiqYnt+YA5BdTu<K22c;EgknpTrz2ubU9@wqc%uCd
z<FQ&AB7V#XSAWV*1rC%d>FX4pe`HCA#;<n#cn(w9r+0-w@%zvm+0jf}G=1UJp~t1w
z;1%84g7(kWa}e<KCld?UO~~pPLDEaCiIa07K)TjEzbk7z*jb=;2TkK=XPDx;z{>zk
ziLX!8ymw|kkD0ZH;cdIMx<M@p%}y=S7S}Wvgp4Z9j{md)P3`)A6axsRAvKs<bD%gX
z_Un2&AB3m2JZKK6$V&t7JJz^s1<t2Uw6p)iZ@q7IjN|5rhp~9*MF1dIW7eqBYNX7?
zw`$Fk{Md!csp?6>V{+{joyxDO=Hh)i=(chH`ROVb;4MmP7J4!G*2AMTmhd+viIHlV
zEd9xdl4>4(rml)%r5fc!`4ni37CR9qBi*>-ZA#6}8}xX>6&8q?VlqiHOZI!<ot$3C
zTKossW}5I7<JC_VpvLQ!RevjHkwhU2mtnN)Fs>rZ-FF~&hT@U^VS&S!6NG@Mw9Ewp
zXJC`{-st|^+=6BY4pclAIR_3r$>)-3#raL%z@9e7pzh!`uo`pA$#vk;$yeM%URPwo
zo<<x1*FJWqa)3NlAqXoCg*m@#kv1EQ%@`m;6~WMaR)Q3Zo+4&VU!OSFJH!$He&aWv
zDhYP<1`s&?-o+*bk(|MXohvsY-B>f|oWvnd+~UP;(G*%=&oYt@Zudru`wjSM3vu|1
ztByZcRep|r8yd5Et6quO@%vh;%P+8Zy#?hxl)k>Tl_-)KZsL@GSTX;7F+B>_`IryG
zklZ}JK>|&naPw)*w9>Y6r0Oc=X<(CA!`k_~hwv{_QIP|&T?1!~0F{^2{BYDHj^cOc
zKAApj%nM$!!Yy9ejmVL|4UH`}CZk|F^=DV;RRQ;nWUG}Ee*~z;gy7B8d@E@yZE=*)
z%3bpZqKUDMW#~^n!^VBb4U|?vAA8@EsFeo<U`C^D4kG?<l1Nu_RqS#-?lOeS_V)YX
z_2TIpG1hxxH<CADABQfxGqFIf!w}))9ae%kYt-wVA1A>tIO<2^@Y6ySu_qEoV}#$!
z>aSn%rWdd}*6C1a{&6xH<_jo58t&VR{U4s*!Y|713-^9hLQ?4z5QgqfMPle41{_K{
zq)S3V5s>aq0Rc(r4(S*gk?t4}X{6)bzUOxie*hn5xbMB!y4H1lS0>*PpuPwh>eBZ{
zSv~+LsKKg@t>Bi6ta%3<Z5KwfncZDH9mU>B#SmH)=#wz;;(?(DoShlwO$Em{I^lnl
zhTj+BfUwt=Ts+UPXckWIrW-*Ot}Cvg1+c{aJlWXFQ0}lyPYbaY9ALoU5^&<w51bBW
zGJlVC_N!;Oq$IHDYZ$SG;w0~(iiuw2@BOT7g^?}2$xi;-s~-swD&HPgJtP6n4Imom
z+fyB2hP{mWa4`w(zGN$4V=pl29pG5$VEGZvPciAqWSjM>Sd-Xj!2xjT5OM)Yp_2i?
zoD4d>IAn-<i5V=(!oKv71*y~&m(!5XX}G@S9S^df`q70gXKr$7$^k=0qw8D2kk=}m
z#F8Vxj+HTN*6XJ-&Hs;@SM$yD+;e1H`$t249VM{d0K-T%9a;B>eAj$t>MoJ*<N_*w
zcO#StP3L<FI&N&B+_0sNYx`Usf2gEA>6nd+e)Btq1hnAoz8>8u$I?|jvG$X@r>eSR
z+EUM2E2fwRl(I&o!*Y#D%|+_eMyH%P<JI5nACE4Hg4_!}D58Thg|I<YLCytpD1Co(
zevol=sc*XtFix~4y=)+};np+GVOni-j53;86JC0`(S5)AXr(Nv1Be*wuWuaNoPbk2
z+F{S&-uch9<iM^>^M_@L)x~&p&^ICysurtQVj5~KIK<eiF?by7fF?JQigg@-Y4sA!
z$vyuX2gAjWPq5=S<YC%Ivj<M^tAp&V!L>5cV%}_TM#)xJKwv2v-TNS^z)<9MpC5=t
zf!guR8~@AC8$}$|AZUV1n*yLKn4^m%{g)(c%k(yd7}s}|q<k$xYO<480BjFff0&MT
z!n5f~>MTNXUXsshofeF3P;A&yFPsJu<he(l%eDCNfNGONRc4wbOiRPqRyDlbYAu!C
zBea?tHf7x?mc*6WRboqn3J>|dK*b-<C#0`)D{edHg!ZxOSIv$vF0LvgcX7DK05}Zt
zV#$rT{yr-|qXLs8fdr4<ZuLgVQ`TL4gK7TRx3ew&O-<VkZN4?;HA*bsc-$h9hB@dp
z26-dgI$nN-!>a--9<(YURaz=U!Gk;WrhlwX4dRo}=71jKp|-JWU-PUS&w!0sjqJJn
zWJuxU2az(gT_TI2fIrt_`l|W`grFA!7x|@D&BFDIXm(qgYr#BsN(h*3qd>o<r)@2`
ztE~y~zpcLZwX76)IKhD`Ti--))`I23B>tKiE$scG?wdl7cWQ{>g$Jn{>{N#F2O|^M
zZMl}cMJ#A~6T*6xK?-di-$NaMs<szDBU9EJ`$LY4P4hN+a3<2wZ*C38{Ff%+t3TPW
zzlHqVMR>oNpP1R1QMn1ct;M+i+kO=LzgWY?$WcUmV?+<QX4VxIMu}iQ=m!<U;`Dbx
zZcYevP~j|jCN|mt03edBj0V$unD|8jNPxp>5#LL+ZUEn-s%2yS+AwSWZH_J5K?cXg
z**fqmc7ox+nNGNVI;W7nP*-Ba%kyil|1c*t=;LnCkYge#KO+Q3>5x64E#oy<S~;t`
zX<E?;H~@FL3PrzmoU`+IRwpL(r6LScP`XJ8?f@lCbAvExO(k6rnS0mFmF%4x?aGO#
z{7SvsW@MFP&6<>;;GdDcx9714k`c|P)yA+v^kCX78yk4at6~BgcgTU!M`%9a`(+0S
zL=+^_WBRk*`PMV4E!<4r&h9q;Ips)8f@WZ!dgn2QB;*o4zoGvJN$G(k=$}v3VgS{A
zjhQdFb>bYPf&dX|Bw6cgW5I|+ALEEC2rBaag>lGiURi!Z3G0K+jjJ3RN3SyiE%&r>
z`Ny&M-JsbgAU*#nRNaWx;F%}zG50a-c%v0nusAvh%^oyfgj%~AE$zNSw?dLD!zjv4
zw={b%>3hetL6z;S8oBH%vUQm$WIqj2((=1qB(T>-efQtzyC41j^7#<oT-3mTgynbk
zxuFy2(11j)G}`6+7s))spu=7Dq0oFVje#fU*Qsu_>)NDp@iQOaF}oB^Ei3sZJbPw!
zdRg46dHpc)+RP5m?_?=ZfE#r6!=U55+jXRQdLI(Mldnn!oo-mLJ)HJG=Hoti$5oT(
zAM>}c@jo9P1KuMr=VACI5q|<#`EF7fb`pLwXNBKi6SKc45F<;Ws%b5b@vLuUbp}Or
zZHNn9{hNpW&jH1U{R+M<f&pSYQMO=KpKgz1I0Y`L;3Ciu*m9-@%mjl~vbf2yA{wqH
zxi7<8TiTYsvgCIqYy?Go-GxLIzm$1!TtygrClL&8Jf^Jtb>`0PKaCHc?|U&7(<kRy
zF*LHUr^=^_sHNjy`^Jjfx!Wa~4C>OVy_QRr<=O)COHfO7Yg<rLS%`u(#~<Z6kgr9Y
zt2o#Rfxw(8#iq~o0GcGHr%CmIRXHB`2;nSbx4KC%9qCurb7H%y<2jk!W=o!8r~!d*
z;onF(`;7k-0ysY$9-+&dR=|hDZy>WUfiQCRT?87Zkv|`Ruf)Y>YiMJOKTCHgt1tcS
zf7k8zq)P)lT=2Y>r>rA<OzcowKi@K7h<4$NPWfr2H!4^}9K=Jwa^J0P(}<yj;KRyy
z^{P{7`>wzzm^iOb=nBF$X6+c4)1KKNNl<=jxFt0*Ux*`Df;O+wFyUAs>Ew0dq-fL+
z+4JN&h!1Am>Mu=D*#ce^sjfB96_v1b6%B`Ew~MZ9S%w2HOBIXfiGpVnCM<NPgg7xW
ziSaa|DG#A@i23bk432I{((q14r^YUG*OsQZc*0p_%VcO=qM#o~PD(mODd}?a<L8{>
z8fxC88_f2b<Nt?u=#uOX&Hh(=Gq1<^b$vzkKcJi%nP~YSpGP7^U&%$6vm?dU7@3tn
zZkqx|U_fPw{ELxg5%QjE_;PEVb&_rqJTv9E5(QF&yO`h&)93q31Txu)%7#XByA{7G
zAZdRV3Qcf=*N<3ar-?~T23uWe4HLVDg#_K8&lOzjn(=M3@RYz+xppQOKO8j-f}4IP
zpPt*651<@yS&(5TaA|Gkk2Aqc!gpG747;iTq8o~{6SDWtPy)s<q7)#&tb(2741ku6
z3cH>{MqIly6606agoT#YU=$;t_aPoC3x{pG!*isYzJgizu4?1u{mSvW${}9-2afM!
z+6pd!2J>jAO4OZT`W*M>t8q=-h;*M|WA)S*6hhVp444_!nM~apDwkG{zjWY;cdmJ(
zJUfQ}V%XNo9UP4@b16|4sLhWL42IdHOw{2@=ogy&HZVGQ9cc6HqYg-s0c&s7yt}6a
zHaRA%sxnYTmtLlw7;nU*NUk3x=dv{V?-Q9fT7NUhL<yxEEL-#7pv^kHFZoDI20#ei
zsdT*HnrLP8%veYCJB+UPHuX8SN}A?@ByQeE?3=4*tBf+!yDtAD=-n9^Ezc7|6(pSs
zA2D<_aZv~~it7j!BZUN=E<SX>-zh%(O382Dqavtw1MUF;rBv|Q&nBArw|Y0fo2?|;
z7uWc8ul|vzN`+Mcx9lkj9cboI*ebU~d#L@loBHpWHyV#Z9{TL-C5x~FoXt!B{h(n<
zp8u|%G#}hUzCXD}fI+imPB-3?_!~nJ75}ZPFE6{;<ovX{=3t~I@1*|QZo(*jsHnq>
zLj>q{5d~pqr6x9j@wD1)l}I|&6~V$fpxl^oYOtHj5?+;|7-O<aNu^3O%G=nu|1sBq
zoH;Av(S9ai|A~6kZB{k^s<58ng90$`D%+|PDR`B?p^0QCGg{%qz6H8ao!&$lU_G^2
zt>_W%q>z#zp3knBWK32t4n``)TGw8A<wIG~t>gHy2=$@{)aKW$xd^{R77X)^fWr;s
zA(%uRX}N(5nUibAtN9r?`*;xfrE38I7%{q}hXM~a1|DmIuHria{T=2Y<1X#m{RAd5
z@+w<fyt;XO(4;cQ$K-N=`dC$&Ss^Hx2M1;nUGC)uw!y)>o56yxmqrTD4=87vlr&ji
z^*D@{^&Md#S<G7PoqbgW+Cfy;vIbB&gY<V*%gdUFx&SN;VxyQ9!_h0%x$gYRC;=$T
zJOW<dIl)Rh=un=s!s(0hH7grMVPRkqp|RJu(vj&Aj~X!=HLJF8sZ;S>cT$T>x@jb<
zpV@y63YtPTcKiZqDGhkTPj}!<ins{`)`fYhtcc-H3DJ9X;{u7H*&vRizN;Mg)$6vS
z$rG5G8n}LO%0NrlRAkHyFmZshIv~s<#&lb(`9*ST`6s0wkpz|PY73R1iYG5*P@Y_u
zW9fKAqm^J+GWW~#|2LD8p(9=rTU~MrD*r}3Ds!x50N1Vvaiy&51p;uh*{9sNt$<}7
zyu>J50})@Y^TqI}ploxWlko&_G!?M}V$F*YQr_<`Qh|5NY!II2Y+p8jd~ZlC!=U0i
zUY1B=*k`U&=pj<DA{M5n!a+$HX_I17j154=+7>;|!}(1=T@i8=nAIWB6cB(L5Ewkl
z%aH0Sg$Zb5Asxi+YV4gWRi$+lxrE&f83x;T@%e29RAPRO%qB&AxX~lCwhkz?(G%0u
z;la~llIoP`4>@@_9i#R!Kzed~)}gtyzI6E$*n6_m?9m32&mb_VVp!!bk4!PadnHJ`
zL%#3!OhU$j{A+}+i^!5ea>zG$#)%u@FIE<80A2u-uCIc$&Xiw$P&*Mw-(-EjI|rxV
zvY!AWs}kkk$VGlZYMH`i;x1t11Lj$fr3-FzBa?J*L;_lE*hBE3iloW|R%yIqf2-?#
zCxNF~7b6m9Niq=$tpA>MUyQl_N<^Vzd0MqYwFgg>7LtRv4YQmHs&l`VMvDCNXw@G?
z@wf*idx+V&8~AOOF)T^EK6bos+y$aeU_62)g?e;tKwtwL@o_aV#Hlub44~lh;Gn>(
zqYr4Ak_O|I?oD{{(1fF%l1EG>f&W~e!+k_TdZZvk>9>DISr{Hz%HqjG>@&W2LPDEy
z3BV}rhd2LN0Nj!(cVjQCzX5R@6CU)&&mx-aTx?<X$_aBNd+Xuh{SEtpmX4Ecbq7c<
z)TsOy1kD(C$CFU5U<ISg{c{Dj0Ekr$;iq3S$uqqzETS1;E%`4wA{pJ(DYF)6Ml~*M
zOEWlMkv_=ecSA56nHY?j+x6x(*EbuZs!4ZeJQBcgYh(hni~K!G;7A;yCy4Kto6I$#
zdM$|l#U!+=F4vU8f`guw1DMy7l5c)tqNqpBnxZ<`hEqTDB_bS^n1_U9*)M?h!-Erg
z@(?y@{d2^+FD|Okwpis;#``Wt4|UxH65@OVKS|%RQNKQhep$?UH{7~3A+RxHK*YEv
z3@oGb+ABDZ2YOWO_Gs`x;aT&|&&UVKyBoQ6X2v_mDmRe6==_Te3|9Hc80N`WLjj!O
z$>(zsx)hGo!viM)$VZor?+iXY8PVa$C5CG@Ha_oo0EU2`e@@(}wRAdP!$}U;*K$L4
zgk|ML!?i$n1IQ78Gj@>qSmwx_W)%^DbWu$l=gZrBYv)c)4BgzBK`q8Jdcluw<iM#L
z_~{!<!jHVl{uCyuE8q6?7o!`!-k-8tmaGcxPuwU8zBbCb3UPqscuiO^;>`;1<}=4J
z@f$IbHz^sx%@G3~D!3nA@b3<NbAGO3!C!<nPt>E&=CVcTVkKYWy=xLW$Oti!gMpk^
zPA<<ABAUoY#OPoN(^$=zdm@<l{B-p~S*z)3pGti6<X+1ttCN`*6iK;Y`H`n6J34+j
zc)vP&1B{^umn{Hf2Rwau{zfl0kmtdkbsfX*1Mh~|XLcJT+8O4O^AP_4uWVdww>*i%
z?)Iw3Fh)YoRUqpFlXkrjW8twi>5=g^{}b48E*&T2{v2o2{wEe0w%MD9qrgK8MuTi4
zJHi`d4f=hL0bi~e1~!el89pN>QHCih6F8hpgqN~_#2kp%DROxSXSWE&Xht%oHU^hh
zg}LH0KbMj0<#!jFKSOU_tVUCjzzK4ClSHBx8G$wN5S_rN{;k7u3#lY~h>_D;VS+_a
zdALKl&8}VTGmNazT%|OHI5HRV-hcrw3G5U!RybnyHCu4;QwRt?GO|JpjazM$KIiOd
zB$0sO5f6?FXq~y-dm4zW7y)qO=aua0RVn}_4kxK5Y=SAkmvwCJBV~`zKATG+xq*L^
z0In3I|L{~~7!;MaOgg(X6<#kBikEi)TO$}cN1AO@bJK%B_K`897a&jL9v6woPZ8@S
zNhP5&xY3_xSu&OvxmdI+f^C49#}n2+^^~5UH7AhZ-lFws<iyc~^bUti3G_(Jky-_q
zP5@#7kO;S?mtL-xJYl*S`YR|jCj=}r7QdN67-Q0_({g_6a*xz8n?scR+S5JZO_$IZ
zNsvR=n%R5%HXp}Moj%hQ#0H6v-S=Uq)QU;>fx!Sr$*ju7x=SFuoh<vHQh@38?%G2Q
zkVS8LuNi;<wKt3b=(wo#C-N;GcU=$0@L-!;azuSUFVfqcJ}s6rt8LM)|4-W$P5<2D
z*RawWFRF;|%yo=O^1vMg#>gmNK;bh0YELWh+E(%F^0_6aZsN94`<1b_S@Ab#&58&Y
z4ApReI&ZeZ0p^H~RS5k6j#!oyJjwlcl(b%gXMtJ_i2!aS(w4?|)VxJ0<Romd{F9l&
zTCmyMZ^cU07>SNk0w3WV45skdD4iH$>*7ujVK&h^T-wPM(!+}Sy);(%jR+E#3*lxH
zQ0<83lPdq2nG>8}ZN&TrSU@f1N7}$<hwqD2t~`+%Ke3gkZS?qtEAS!-I~q?VsAN8R
zWe^pLqF0Pp0e3PvC%xb9Qpz=|P2>Y-W}6kSbt>Ql0K92K>!N8h!z2lm0CmJS;wgad
ziLxH)isdGX7$*HZ8RMZwssRTRNC!uO9bpbID(Xu7>Td8%Af&2As72PZ?qg0T1O%N$
zP2x=6QYO+fbte_Uv`lu`#Gic>JD13`&_2AI4S=SMmof{Mh;;pggBD>fYv4#=ZK@Df
zOj!QCv7t^XjYK;n;YY0i7#RLz<+@QGw~~a~hsfPBe}4jS5ng=25ZpImL4Bni7Gb$w
z`a?Z1|2*$k^#%zn-X<fs(?Q*@j!4oZk=Qy7A{!Z?;+Iu!tvThLcE;USC8OEl*zPeX
zz~!V8srcgRyOMP?q0Ci~I8}>p%}O}P`La~(r$+n^>5NP|G>bO~3k0L|@&tir@o(;z
zegAJLs>P#DR+9}^)rztJls${(GDg}GlsGHZ-@qlG78Oj30$Kjy174PHn~v7iFrI<S
zi+nCn)vqJEZ*$a>lF@AB^j`JxUfh%ka{cXI6nO#zC!IDr7(+Q5xmcC-7)`GZKsI)t
z^u^U^CWr)&w2bt$)-C;`{zoSD0JlV#mh~l`@;W%#e4O^pY}kNj@)IzjgAr|>z4B9l
z?j=jVI{sl%zwW2&3QHh!+G61x6U=|<lVd~WwoBgZ#d$XU)_#m_09a(m%Vl5NQc>1D
zw+<{YGk|!re8^#k*bIRCtd#IZ)^v5Q7U@W^mBD4Eqse4E7s&)ipX9M9GXYGlqpfl3
z=`jFXKpWH40ZyhDqpwQV3xFZGkOt-P)@>BT0;Hb1QU9Y#Ouc7NVgpl{6W97*jT7e$
z;-}zr1krJ?lqw~ccTHT}cJ&~~E3HWkpw=_MDa6H#K=N2n;GEkKGCj<l(oTz#GWzi|
z`ZZ(FHpVn1E~Su-W-eQ2!vqN)QoHe|`)k?Mh$}%a(Z5vm7|VM~gM%~4)@9$9r~pvH
z)xEVted+ajW|!#wMI^(obkg?I$`^OD&CEkLXi$PbF|??g8`oX%zc>AjQ&GRZy^$|H
z`HN*RY}8=yq<1yrHcKWnq!w7C$BD-|mY!V!?Z;<tH4s&=XP|#_F_(u`o&zR?L1ngN
z4at(O?|<k_Z%!{s7~gd5O4g-gd@30BMy~qbo=i73j^FqWOZq2V->Dy8Rb+3r#xsgP
z`y5>6%9q^v1+_!1CcU3^P3GNfkrDc?q?m=EWm0d_xO%f<Hc6Idd}uJs``!%-;?|G(
z0dxZZgh4(YoW&B#48yW{=brzTCCu*3*wq>ekVIMXM)%Sq5IyMN?Gn<ahFO)^n^Cc0
zcU}kxkic|7!e^Y%g0f&cty66t&@9c$EOJ=m=!mTv=}fBPfcYYUw6y*YPG*+b+YD|q
z4#T{?ZI!Nw>L5DDtSFfq5yoQJdukbrdS>g`mT%+&_COd!@~k4y`~F*)vs>Hlgbx*o
z=tIDDj(ojP7F-ZZDr$iRMSnE9L6j_ZN=XXJQm@+`?Yoxe$;*w|hAE2x;POtM%Yu8a
zc6JwgXC*V9(J$i&<Vg1ZZKP-cIiT@=c^G;T7`CPd=ffUg62)>G{>C~v-GM-cvERMt
z$|?i1T#;0{4R~ldTX!Sfo3*A&0NLSsLYY7!z=0>DiwLAJ(fLxMK}xbt*PJ5jCZTxk
zd2Sw5!1l-OnRk{{0_k7ZsTqexH!6%Lwi3Utt=K0HShuqlMg6qN)W72;S|Cg)9;OPp
z!)sE(=4D<HeI^Z(?X(f~H?%P%jE=OU0Anva{eqZy=>>_aBK%{)k9?^|%DS)RD4^)W
zS1qaXHkmv4oPBhP%5r{6lX2V4rNriX2vYJiH3W7zFFVOxk{mS1VmpP6V=_a1|NOH3
zZc6w60P?bD9v&XtWVUztpVHpSXJlp`9V59<>hI5%w<fdiZr3&a&aC{CxP5n&+fk$b
zM`Zr7Z*HgF+{P~NcD?MDXUINROBUeFOfq{~z&FV~h^-8h`nVYNskmPOx)ubHe@8Gt
z?o4X_k3;#cZ>u$)<VkQ9ucef<5u05d2e80N)K7Mf1+ri44+M;Ztve{sx)EkBcZA~{
zh*$cd_JIK-nXJ;<UYWUs4Qv}&H%o2;H&utgMtU?8Xj~TN61qRP-w!i`_8pMPR{++q
zE{CW<)WIIGJXsB=jhojEBHm<^`1ZMut6PU@rHjh$>LifF^O#X89NFx$)v=1j^sRpw
zXX_b9r*2+KK>JV@rP?*YU85is$A1JXkm_U`LG(^6B|jE$_Ri$&R++4D15;&y((~lO
zO~GJ=Y94kpj(R4#fQ$_Ld(K=Hzp44C4sRsknDsExZX**QNgF~Z1bd%{r&*ktuTA6Y
zMRG3~p~$l4%_*#W)>i}b5P`*EPU3^uP30Ytk)(#xblhw=9P8GUCo4WZe1iie4~xTC
z%0vtU(F7l#UZu$!=`6g!A-^X4d$y(qHm^gqbiUej2=tG+vwp$EONVfa8p@Jo-(Pt@
znkLj)Q=2wfWA@QV*JUb6`TjD)1@xWF+~;H3LG8Ca?}<rnHlyuvmf!C4M=QN+#((np
zB?$I{(-q9aMKa{<YB>U0eoSnP%>m*Sjeys+lN^%2v2|(9nC3WWes@9q9PoaJ^Avw{
ztB?O9?NNd5F#)h4>a$dAQZZ;VuK#uVCN?)~I8@eQgOXl*A=~l>8*k=Zq{bUt7uVk0
z{iFf55hDYoyDz)yt`Abk7(rqfY$y3@vXkK6yXTY`^V70O_e>n{y5dS#JVEk;^4CrC
z2OE|T*TMP9Tn$ze&bTUu3`5vPOZNOF;Xlgix`5k4EZWPj8HyNyo*nh0fbV<|5Xcf-
zg6#zCiEB*@_f}Ha4yd^ivV^g9A$$TLu8VD2^HvNad+wlODoJy(U!CI8^uHZ17hcDC
zLUn}BnvhZ88r@!li<Dt6VP)NH6{4F0PZ_1P+%u6MLSAkPFv8m7vm9%H6f}ZUmqt2#
zb(3rpIx=a~e<aa80qN+jW?a7pVtjdyAML(5jb&gFd7h5qN2_l@>$1!<19SW1Uq}Bk
zp_NM;jl-mAOu;oLz@*ZrD5n`<_WO^A{n-_b|NbQ&O{5IS9^i=Ej{u5Bn=t4_S&HKv
z@D>k*IoR0n3kkrb2F#lOq|j)&2U(0(>ht4X>rMmlm7`!{`cu2$Bk??tcStQ?2X`7x
z=PmQ>yk|n-c^42Wv=`{8=&(9#PSoM)m3Itq0PWxvEN3U>PmR3*+(MIC(cSJ7Sia-;
zV$u8Bx`Rp{hg1_o;rTW~to%8&&i3~=?1dSp*}bNSBzk{oO$Z~}wjiz?GSDUgz{iCs
z#ZXd8T|X60u&1>ZUA%rFA$g^XznD)_a-T{lPwG_Sr3R~hifDN{yqVaB1iw1R=(U*P
zs4Ses(oac4-6`X0l-Vyp3oeq>U6RFZszf|kB@0{WS;WuEKRUQS?)%@_)&KC27Ko4i
zZ|-;fGw(MfFV!!GtX_64U+MdY`=3VN5qKOdS@`<ip5NT=xGAOi>B{q@s(4;*CSNxD
zpC23x-}NqE#4aCzldLHFL%<hI9jqj-kobn)_uqa@+H`xvCb+fXCcnc!xD#xG7Q(XC
z<fwN%VBJfh>z7Do&sU2coOw<=8Pr^w-3JIe5V}S_<wO@1fq4{oDq#=KM|E57w4HFV
z4?}Nhbf+cQHkghl)G%Sms@G+GFtA+7s{9E6gx{4oO04d!GkVme-LPjbTGg{hErK98
zc#~!iyujsgRqG_C;tGIB=km%0e!<1kC2B1^f<Y|oScF(k2i{D!c)J0Rdvj{1mXL9H
zVwM%~Vj;duO^u5A(s!$+UfB2|OH)1*Hd}804zfKor>Tr^1)k!$w7!#h`%Vtj?vq}!
zku|shz<|B<E0O`@4OvkDh@DH5pZt{?#OG&Z`+CMbVb&?OLJ3XbH?t@d@7nJ`<>|wF
z$+J80*OR9b2i<;-YH!q-KfiodXw}60AJKoful_S-f@<lPBqrO2vU!obRn{w4W<<Pe
z)Q<aA%z0m1(_O_n%T`?qU(;eIAZ8^*Q|zWt$;wL28d+q{KG{6o+Lp|6jDnP+JN*}c
zp6XLvph14R<Jv}s@>;rXJlMT5%x+y+ozO~M%;~d@mK*9y&>1k!x8zHa2TmNF$<A@@
z07##@?3EO`)pl{nr@^X-PtHH?W(xapen*i%P<@ocO4*2&#B_9_ml&20(}+Rwykj7S
zTDDrcw!x%z8J8q8&HDL94+Lf$M9~zv-1ObkSrjjC%D5c&q5#HT5kHBITTab_<_-JF
z^8N2?+=URU`rmGBOpbeiT@!3KtqL5}1@uRs)HzPl58hEU(Lc!9vhKEVPHH>8QHvKo
z>y@Z8FeS9y#xX>_>*^gNVG55q<z7Cho^`s&=}W$}`o4d%;hi|z>e^YG$k%>mlnnm*
z%erCa5n<6x44>6sEBchLvcz>6kk~I>^Si>y#5*r<o=?*P!Ro+gpD9jTL&!aAwm8#S
z(Oz}cob-6>E(Hjf>(e^c2@&_%{-2%t)Em9{)}8(kLDm=x6xIT&r)om^KYBFXJlO9K
zZ)`Kl-`}1|3;EJK?=G5pf!mq_5?w!tDz=)+I42~q^^$_i*-uft9i1uK|K)jgkE^`$
zWiJ4;6YCjzdaYPGB`D#z_ab;tH2G?vnq-xd$1$;<KCRM89RmWd;=V$Y^@(G2b=ech
zBsvi(hu^rdlj#OMQ3?>xncy%sq{_qC#(g=~FtjHaGzG;P9)X$;c_z*ljr6@)O;s#`
zMEHG6YgJU$!lzgh343y(Wl7MuuM0a^Wy|?$p~^@LfPPebr70I@;i)+&5<)Q)$aJ!%
z72227v$1N>+#QZeUl2(<Zmw^}eZlK4vh3o78`NYL(lST4nKzHeRr9Rv%Bf#MK~`nK
zGP1U*C;dIq{@GRnZeo5I7Yhl4b5f=n^rUM7p2YsJkCAVk2m--{8$~FXS(=2|f2*;N
z1d9YYLPAagIM|)sg3|I}8$5W>UY}+=r^@w}ZRv2&UD+?;LR5^F&H+?khq?lT^T5Y=
zU4+<xjw}*ISLIjfR9_soa`nY)yz2QkN5f*+#fwjumwuDEk-M(PONP<r^*GMy)Z1BY
zHU>*e<14lNG<u3S3Hslk>KY_D(W3<0b?Sz8E{?HE@g(S@nYJ<JPVMXq=fhQi|J(Z$
z@(%MQu)P`b#xi~*XKTNBbN^iEqs0BqiNy7v>YJNU;7+0IprWF>No+sXy`b`6cl1Gi
zzi-UG7QNeAzDdu>$XF3sro_G8Yxn@QOHnc+Msx1%4d^)gf?6Y&PR?b`-MiUe4fINJ
zcCS_Gn(#U??wpCXNp|1orcTcmS(cra9qef=7=}CzrF}0G9ac2XM)A|$x$Ac`0q_)m
z_D#N%SPRNzQn&wA^qSI)TTFcD`Xr<^rw@e55fcwWsW2IqUZ!aAY2V#2dw810X7HSu
z(%>1;oW97KEsfm*<0_s@bNZT(!*P%49cM5Uq@Nnmq0P{`{b#Odx(!`8X!;ZRVxshm
z<*wb&J*Ze{e;)^_Xrm0b(H9bPHxg%LR?v2sPeHmw8d`&9SRMH@i8`&0UnEO2R3{U+
z4;&J-D&5{Dk?-hm#)}|FWOvvF!z*cB#Yh%dVMNd)%pl{8uXKtHCSg>C+Nz0*ANyw1
z8|!qy7wviPInr7lo}*+ZXkVN(I5T><9(ZE^xykHlVZg_H?g+AAw#jsP+{F+~-{d?+
zg(M3A@KG9;jy3k!m`ulb-86WDq7+c-e71)njk4Zln-(r^aZ)3e|1LqP$tU2q9g4qK
z1$CssqA{5=UxE7Umoc=teOIR@b7K=L5ez?hhA9JpQUu%?wuk=rYrad8X;qV~C#gvL
zm-l|jZB(n<*ad8^JouNe(ee{;K(v1BgD4=%(qvP~F`+giL3J_%6WeMJfhOCyN9jmA
zbdh5V7I!yD_?Hjt#ZxGtn+dNT)v^`ShoXICF>%}m|IV8UDtQQ^BOFC#U{roQvYIk|
z#<c8qZ-Z^1U-pF^+kd~uHRBDNYn$O!()F46;`uUix63anndbEJ>h}2yVF>qAN*%b{
zG_}kI@@ydl?EREH+Q6RvwY$AcZ#EkZ{c_uth0Mz_&d}jVyfbiH6vJ7U%di9oX%0+~
z7mlt-%2E`XU9t2uF1U!ywOQPXl<7~6k~!idRU!ELK|npP4Y|VsQFKkSh~H}pEyjLU
zMP;GVb@AunP7ju*f$?uXTlaJ%(n2rlV@~MfesJQMm5{vC6FOl*8E+7D2dwZs;m3t<
z8O-Mx3VlnVM8vxCVDo5q?9Ee!{y7LX8)KzX2}nt%!PR68%?lmv4_D;>SW951Z5W~F
z-oZ*d^fFcnkU~ggw#+HVU1jt@zIuRK9ZgtB3A(W6e|f_+&xr|*O9={~@GRVQkvN@s
z=Jk~#&?|}wCv)-1i^^_(D%Jh0kr~=+3B3vvT76d6a=3KvN@OeVB+uqREp{1{%O?#-
zraLH~u$aS@??{nPDmaGmb<u0$jfI`21<_PKd@b{$FK>731W99kQ-7RphpCi&jpr7~
zlcKLlo21KmYhzu^gGEYg<ym7Ki*LbE<&(_DWRiwuTwUf|B3>wrGyCkKMp^8f^Z;fU
zPw}4DD{GZJK<2%~5pdE8AQexj72|9~*<{px{FQZCC@9}EBqLI!V$I(2;VD~rkcWhb
zKIhE0D>B#@`4cU(rKwVA>e^@zR6|d@d=*C~t#-A-2k!bP<L!9Thi8V>c(_nY|3B-T
zHA(7@zqUtY@Pl8fJ8EABWqaK<R~@uF`;L-)aHck!7izyjwhyV_{#4Jrz0ST&%{Jjn
zJ!Ee`W?w!&Xn40rNqarhck%fiREJq@D8O=tNrG>mus~EFk7{1SbLPDNH~=SeVmAE#
zYzJ3(uwjPP#Fo#IRz>k7W2$K&@|IO9?&MLl`-t8mo;<TkVCafD;VSc5vYCBHQ}|>?
z^P*_In&qh_0r1IX`?w^6>taA#V~yJ_Q2Z1i0ES&a<LCl6R;9t0I17F`J&n9U(av`N
zf_sBwEtcgce&0)5!SvcDR^y`e4|@f}a$V9%-6j+TAcH;T=JvKOs(jFjVYRDrv}2|Z
zPpf0w_I2qrFsZlwmv2e$Fh~^23|E%XH3z@tI`2P25=_|yOX|o+oe+88J+-TS3#$~w
zC%sJVi~mrZ?_6omFu+%>u@O<O-NX65*@oz===(<pb`sfJrYG;iE5m86`m3Ibk_83k
z{LsgQ(VTAmB*ovnlVBRasiPzXhfaT71X+QnqFj%LO+t6Ld6B7>KJm)=d(%=cfumGR
z)_5K%sa{_5k$Pl;TPn~37TdxL1|{kZ3r#5+G^==;H-0|LjNTKNbr5Uu<yn<lHfwb_
z`lOp2W1d<`J6YpS3^i>(INB9f_cZg52pn23VHA3cKU^i%qcz%;m{f2$@#$rDv-6Nr
z&qjQ5+Je}bNUB#vOM-0)eCcJz;et+PT+wkY<|E6t59amA;2#(Le2lQwqDV1e`~m~S
zyOawG5@oO~&8TH&%Z(75=mW<Q$m_qGU8&X;yV)9sWf^=w+C88F<xk_J`{l&qgEac1
zcy%tAkU*RF_t93VL)G&)h~ycY<ZrJj{r_TquUM+Exs`9yZqL5Hd;c$*@A6$-viZZ}
z?g+&-FO_@vRGMnn_#x@z<WD($&GP(7#TZnA>5&C9qr-8P+vY{m>F0Rs=R5>h^3fPH
z^PE)hq=sqN&<Sep!3MCe3KnEpYi`Y<>&})GMON^~P3dc62)1|@+9`^F@FAMrTQ=Io
zw@xNSZow^Azrm(HHI&K?IZW5nuBg)zS&(b%R?MU-Z?@6b;wrMa*u^Rzzw(FCs9u{|
zrYn+P0I^m%g-s@Ip=a@_!$1Yy4Qlcyw9>hdz9u9%H-?mc&$@`?G+2J}RmvjAGkBtC
zCK`t>C(Ws$0M`Dg;T1oD_)ec_aB+nezXoG!#6!a*GIY(MvFJ*{R5}8ORpc{XxtHXa
zH~V`q{J!T=IfoZYL77o3Yh2vP&H_Q_Y6S}-DKVC!%=|fKAerA^HI%qX2>z4>mV5_W
z3Car=Yy-V=C-;1%l&>v-z2HWw#V0FOgQy2})KqbU%95!2QEoycc)K{(+(>f=r0WE5
z_?H%Fokh+@d8{*`b<Ntu+fCBsIFIJFrhrYW1+uQWY#|Vh(A(aKH)yCJLC~uPRXT3;
zg}V`DB4V`oHi9;w`KHA!?<E*jaV8|XCIkix6uQKwT@)s$VKAq>zzzvZ>h6ez6dqMg
ziw9Qt6V%JmytsHWBpi~$gRj~2sc}&eUDPrksY6XcEN826EGfwy!*58dr*lG9+sctf
zQ%E2!O6{rDA9Wvdte5W_8MqWCpBlmd?6*VwGYrE)rLdGikudw(QSz_h?D=v-8w&f!
z`nuV#U%xze@{=!S`tA&=PQL@Y<w$bdDWj3sW}@TO-|uZ#Y@?F*65m^n2mQ7A$)N<A
z#F@9}vv$4U=iu@={=zoLx`*R7FHR3PVpc;mH3$b^zt^%$E^ehYZ&o_l;&`x&Kq})z
zM%_(hn(gG*T}lycUeQ`|g(TQLO~4NSR4DY@PVgBxPq>$9o$7Z9twcc$0jkNPZf`ut
zj`XWn+2bAzzideb<u{!mKiVA<4!;O9kE~pC)p}W$jNF2o`QzsTh0<T8=!%9t8Wu)g
z0;c(ICtTcGAd<VizAJW*LbN(YcZbo~?KT*#!@A`bO$%YUjs1(yyi99WdsqvK$Jj#0
zxANNhmipy&EyNf<v5bnO%0h~G#I}o72A<C4LXZvfuCuh48(tl6hjUGKBS=;9{|5I=
z?z#h2n2+$u6_-botL4Y1u?$$BO`hgM5Fc|w?D{##gfI6ca)oiW8+qX{vqn>m|LOqH
zzO~bPkx3!&vDg5b>w#9`kkS<W2=-YY&38%4Bc~O<Gs3UZ-5bN52?FdDf<ed<5<At^
zk)4S5hGgcU(RMmEi9`I%u00dq=Gaj4onwbz4hXEkrSiE@sHEG*FS!0NK0=rV4<tr-
z7ZD1x8lK)IBRX9<4!eJ9lpCDG^1~w5Mc6%mW}mpeX^>z1)lj1)rQ3aB@38`OXRzA>
z<^u6Mp5tpj@{67<dJ@;IH49PV7l514Kn#|@g*sTp#&4jHX4@Wue<+AbuX<7*Z@f+H
zviZO^@CW~e&TrCD`>BhCNTTx6giGOrjKiW6PQT4eB8SWoL`ka8mdj_3(aW3W&$wNV
zZTC7zO1wVfywyXpm%drdzlyE20-jXg!MOfwpX4tUe!{Cw0gWT~=H-?opGXhl(8UWj
z3NT|aK$KVLRd0_-x7``{TTX776Xvo6OfqX`?D%J*2j^R2?dO-_eD+=@skGROQ*`8_
zTMOeji5;eVazzuy5lbf8{3sjbXlKZ<URl7_L2s5{rTDZkV~8H4kfJeO9?+<LOQY1)
z8MLcKD^}Ds_XazqXrerKjp?f<l9Adfk<tML3>%b!%6${&YF3>wf`17+>~M(Gm>)di
z?$BU-Tc=jY!Q~#tVP;K;N{dp!dib_(+uTcYa3*}GnHNvqz`D>xX>uG<qQ4-P8a`s0
zLcsJ;&KErIjH1dUQj{7SPq^Adt6%-7V>JH5?1mzM#xgMQM)M`HT_KkKHh(6f!IN-<
zn>1VG^{)*rID}$p6q;*Px`EQHVcDb6gKu4JnNiJ^K%ZIpTb0VJR;nkftsvNNv{#{5
z3&zUyoQx{lqYT7|+@0-|Vz3ET&x#C&APf%Y8Q2~+Z4v4CDa&-T+Q05My6Hx0@)jFu
zk;yYWfW5lL{Q>d_9*Zc`b1SC2e)L8~&@R)sBV-C5+pEbd_%+v<kO`s~m2GZUXE#x^
z;`fF;SpcB`cSBgNV14uOpgB4^;rt!dtN-3CikW~JmQ5FyDksWr@klX=v`8?QTCbcu
z&?c%}LFHj3?O+oRF{K!BP(<?MoIyENE1J-o6%L}vAY2w(HVb_L^cE3u8RS>!M<$sr
zUQq7=C68qQ^g7Q~aO$W?{dWt?@3Tfeg8WIl$!$;j1(_ER*}PDh9(Q^ETaLG7?LEtP
z^ZwWVe-He>KdEHFRx5bUJQ))i44ot(y}pfAPq5%f#9fm>IZRs=*;Q#xULG$#H;Fr9
zi2h!oRqeUC8gm|1$N^Us+7+rod(V*lZeV2~k`mO74jv=P`(Vg-MP58KneI**p8Vd3
zctIj-JoB@*>NA<d#u@|DM!Un~JOcUfN=IUp1f}=urXeC*c!8g{PKA*HogG+39{CB2
zkSAK}-5kDAf^!?^-T9LwS{#Qd>j%NlyudkMtySGnWl&(qcSxo{CiN3#CqP}08zN4F
zNgk$B{gucnpdi~^)C*tH=nIQ!=@=UYwYty#xj~BvRM%oKc#J9-@sX>`tVA&DbxH0Z
z;b(^DY5OCxV3VX^Sj&%a!ranVNw4B9a#bEL$Q!9VEsXGW{vjGXn3#CH_NB_xQvdM^
z?HN=vnUum;xI@e0cb!^{InJ|K<ihBfz(+@L3olmk^jQ<y2`SG653RBod$9B}tCuFQ
zq*BU&exGl~L4;WABLu77C?+J`wb^`(i>g6NX%sKCDr3_5N8~!iqN2jA_N+tbm#e=`
zU0OXGKJz9ky>3=H{iIN8%jhfDyX!ZtK^z_L3cD@4)-t=Wu4VjJ)utP!-7j}g!k&Lu
zt;fp-K+%s%BfVaafsDQ`(*vE?0<C8*lG3V12?H1rB`^O_I{r~ZIKGy(8O+Oyw&|B>
z9bW<TId;BtZ!Q^b)1tS>+{+@Sf0!kj;p=nWK}!-_CkL5?{y2~NiIDhJ^e3@<7@!@T
z`;76*zv6!<)E^HA_fRd9S8VB&!*L@6khSr#O&HmH#BJbt6q8jR2OKz>%>1q{@xr8e
z7F1Mw$Ab!fg~M_o%mXeEH1B?m9?9xgB5vO~ba@55RxvWiS*bSAYmgY~rSJ#iU@<t>
z0kw+fj7Lz)5)s-10=z=k^34@_iuVj}3`%X8$mT3oZwzlPJ))ReD#gJ@WUd+|jwzMj
zNeU0fwSzo6dNVF~TM>bQ<gVqqz{r}x;{=2&Q<XFZfwu^w7|Wm)h#s5!{L0Z3A4=as
z8)_bQ5a}a;h!B)yG3PWiDKzYihE++BMNA{QyRwwy$V#UJ-<7__TC6H6I^BA{i0B`S
z1VZ4#-Ux*kfuhmMX9{=u^*KEf8QsR^Oy}<hm5UXZ+d7{VV<tTQO%2qLtCPB78IH*;
zVFB}nu5C`&37CtQ^lj?InN7uC;?{d8K<f)|Hj^&0$D<-4nxvVEHqIs$f{C;(b0)1e
z>%sJEZZu)%T^u1;any>0>+4ULESn!afyWk+WA{Jh!e+6E^<jKeoU2irP)&bzB^UYG
zcbpRkcGOJJEowv@zuB2tQY_sxQQ_VKawuQFLe)U)X@L&!TQTy8YUf$lNt5o;ZtxPw
zGqF`6EjL0pp?;9^87r7R{vkAp0a-O>fyymm(~L3yg4C#V*R#mHlgKg0?%n(mUWu1j
zz&LWbqypKKsMcJs{p_2olKYmuyU2+k{3ltE0FoCk(f{vq%f)*!%)9&Y|8CUeeq^C4
zo0<lfFU9>oXW!0d-$vgx9-N-}NMMcp+3TB(nY4M8MwZ~7=#}BZgAFL#HA-UbuFaom
z3w_*5q%?orDTr)y)hJ7}PriBxN*sE2?3YbbU+h3JDYCkWDPw7()HUw)HZR!zv8|ug
zM`6&rsz7rPLNnj=6I4yI{rAR0!WifFuNj=fkuI;l<Jfi8{{4;UrVMe|B6G;fs$GI3
z7X0ed@vAt1KiQlABacEA)VY4;iIHK^7)Pl7pD_}KxG>h4fD!ulN2@=GdP3+mYu&1H
zpHhT{D=<5*ewlX=GcaJ|#UWgpOCQS&C6L!tU<s2gfgG|FlpoHx1U2PNrbEHOSRqNs
zv)`1TQ-p28;S!>%+Bq(`)<PE&XlqZ(h->)}8N<<_P1`<3dZ;F2*7_DeWyqp_2LzXW
zf?!SC!oR$ZVtQ~b6HPaQDAyBEqSre(+lkm00K=bO#Z0#DpF1Qrj68b#m2DECWbin1
z8hy@c&3X~?tBX*lr6kqJw6dHM=@THH9GX7&q<zm@IQbX`znHnCBugHQxQ&Q=4tXfP
zM+h+R--X|_eCtMDNR6(y=CKx5e+cD14YPF*Z`J&<I@5#a>URo?q_cfi^|I-2DQU(J
zyq_m=WxQ>*2`NH)I*;(DKNQh-4@6!by9CugVe&Dr4u&YYu}M2TMTr8(a9ZBcEK>QV
zq(1uM&nEu;#qN8FYWO<1NKg)sZ;<LQ8>I{FTnfcEk>_oq|LHw^S%t`FSz*pLLZy~n
zA&w4Bru_%eS3JCfeJ|CrKT@srfHt#7n}on#<jKycY1{qV&qyQ*E}Ym76gwc9e%&ex
zP8HU4%DRFNXk>GYLX-hAmj{o+nf)SYt%yZW4G09i%6B*femIHYDTW6K3}a116biq!
z6U^+Pi6!^qZI%JqrF4$ZO9CtTm<&we&c+EFvZ4K)XJ#e=-gfyzeeclt)#6tEF7md!
z7WO&O<P|;CwAlSkanlMH(lAfm7uZnq;L>^tniw)fS7cQbnI*kDgzU%ed|_uzoTX?_
z?%+;>N)jnzx0wD%M`l$hRoK_WP}-C3w}}Y>T_!)?AuaPv0MwTT*WgtQ!|M#0QM+HS
z%kRX?E#fxatR6V-MK-$gNm50#R}FbDc8?L$;JwHNxkMaQF7Cw8iHLK2kazjWrd-30
zp<X`S&}P)5&e!H*^<(vC-xEH3=7;EcN>5SrQ73bH?#>-av(LQ1G>3<@hXyLO964^D
zdIF3qW={64gFR}Yh!pI=5ru79L8vqvb^)mBA}Jc4hm`_7b-obE$Ea-bhjD0)*q1<%
z<+(G@-`P^}n@V;tk>5TjN#fbWT}%-^m(#Y-v=aLLC(1W~#bO&v^~-;p=~gC>lFFUh
zV5ZHMN9pz625(9Z??ORC!yW;_@rjl0;8Ca~{wB`UFHP(47Q2q=@#81gT*VJb=8Z+7
zE8S8z!-9%GkXa_EkO+sQ_DOeH+k}lpY*7TWnaXv^c5}U~MHGH5p;nc?VOh;XQQj;{
z4rx05Yvo2?oKySXAKd%CN_hnqTwfynFS-3;ME%@%Yuv}lcN1^cwzoqpt1sW*`O%&|
zm$T{Zp!b%eQ-P959ut<ACybMX`d=F*E5<lg?rHG#a5kGiR_#hm#BQQL`TJQ0s7zL)
zQ@kXp!cUkMB+_D08rATmlC@V9p1hoWE8v0!CHOV7tI%&!(23)!zGOeem!m<#iG+t{
zXmN&KRhu8)f*a!?;r<gbp0JYF&+~@mbXa{*x^8Sff7aYs<jk|HZ+sC34?ODd2(}EX
z1%cSbp#nAbQ6v_+q=)$WZ}1-lu{T#eeX0H_!G^BM!Gx_hmrRV2=1m#MH2^mLhSYM8
z4oPv+*hYd%;#6@DBqY`%G_qK6nIr(!1tn<H6WmVVjKMyBIMmbcN(hD}<dQ?A_0G=m
zcxwS_2$UIFnl|T0zj0y;SR%z0uV$L7Lu@5=FHqzP%!BC8N7n-@*?a$@<{7o28=SOJ
zoz<~=g{D=S)hQQP2hY?+GodqSE6h5at8B!U&!cNt8h)~lv+N9YrF_cSrc*!%N(vmd
zZob8T#y?W8d^1yPQJgd~O^f$9K0{gAqP$!uhB0)WnIGsQQOwWt=uEd5aTfzCxqG?)
z_IM8RYIj1`jW?<!u|ahtbn(|i9qeS3JUYd0zuDqt5SG@3m?Kt?<GMOn%n7><5Je&8
z|8W;+l6L-1VyF_j_u$??>=E|`FU~-C0H$vTqd}RT)x_u8v>wM6|G_7cm7S<1v9;P(
zELHjS&)KhgH7p$dsC%v(U#XsF75qp<|L>~(`c4@Xr$Urx2VUR(ng6F#9ZR;hASINF
zwy{mp__1Jb>w4AZJMajX#IrPO1(TH4OkdZM84Rt9ekRIRfAcnvJFIIG5lChMl=H0F
zZ<X4$Bc_z$xpWGA3p|^>tK{5-mC>4M&yjq0^poC2c4Ho9u?Y$0DQTTr2=e8qf(OY-
zgQZg(VGsA+P}X6AOh_O}t!^C=8zvsE8>}RCCN!0WXPoBq{RJn{*A{;<8JH~bW`88v
zxCvnxeck{cTtDBk=)!qqEc|giE(~NuSmQTDA8{9$;F!UCCQ>kYJ>GM5d8v5NOr!^!
zL!WOv4&P5Mz=mZ*6#wA)z7xYoQIa-yX>453hgEZsp(p0|OUpa30I0cq9cAg%z08ls
z|IRW0L`wj%<A6rxaa=}XT1Q8}mB&-#UF=Z@VH>!BoI|JHaOazdiBAgFHQ!4UzNH6m
zJCUM8jSdsvKibb2NxdSd=E<`wlBm>MJ$6ZqD;moVMI#5I&Yl}v+_xll(xw8s2Vi>a
z=~JeJ<UL$V0Mmrl{S3tvI)>N{ilv7$hP7@tBAZg941;{&6|Cbrv0mXnCgO$jnq!=#
zq}f-1LJz@~S=`ZD4zLM3+!Z3jacU(HfkbutsmpE4ME|q72i%g>RZ|rMu18t>BMf2|
zpMvcx3<Nz<VCwb+bh^*_%>5~=bV#0waz|Jv27S7zl{Z`CATnFvkckbdk&B0HIm<2T
z#6l<#!6itzrAg;-9xEaX7Bc)H`$Lhkf%WW$f`MUC3@IT@l72+i?*w!Os5_)p8~gPg
z)%3afgA4Avp7s?hzb&f^%=;DE!v)vTBUj0D*P99MTf%zBWlI~IP{|XkJ6u~^usG|!
z3yAPOi_DUjia_SNK;7SD`+qw)2YZW-AfO9a>5;}HV-!tS#o->jn7e%mcH4mIH%=xY
z72fBtqT`AKO~b&vc3+Xr{UYSl6kiyJ^NDruZB{r3j=X)f`Qv1mBFdbwV#*EZ`GBCN
zQ7@47V)~U4HOmRzmfy0)Mh|$Oy^MH{2<7~`nYn8&FM@s{7N|oAkX>!^QXLEGB3T$5
z(;PxnLbih89NZ1lM9$`|UV-!sUoz`L7FnM<{gzWK{EBnzw@iX)GD?WsLQi4?&to;6
z#h<>fjpRo<_K|zRT|N1H&PlH8PXE40o9DfBj4+Fh-rPt3QAY%t5u=$h4`2@l_y1gU
zn0t6h7_#wEV3gt-%P&E5LzF>s&6ql)GEb?XcQc<5Ol|p&P0!@p8pe0n(&$%H*wO{V
zOZ!E$dKV5x`ojbMHpu|{y9)H(`qU<NG5ryEZVsiR`W_%Ojla?7iG;u$o~iFNVX>j>
z>XelJQb}T8d*{z{JW!S4JwP`ajvnx8BZ5m8d{(9x!j_J7y?3<DV;hsnV*OR!%1<yB
zUbur+dF|3wm6)Op;HTPsQ3*>NxiJtoC}b28N;fE3X=GBk`LqzQqNoW`=(mXe#r=qA
zyXnxifW_ERb%QsX`JW##8TJ1ls#I1lhc|ff0RVX%1XWZfhPfV7RshYCskS@(!Yq3y
z&WS0h#;YAfH!@4>H51&~H(Kv$>o}CZ$@ivdo>?&{`0tx`-|h~PjXAxbXoo3q^6JvM
zdGRQ<?JTkH5~kyQR_ocYYQN_LuGYbalSd~1zxUxyyI22=`s0ZRy1{R=lyH~@Ap`_%
zKazXR1rI4f-Bto1GP;mnsUeVj$=d;siJrwXM4aas*$fUm4pE-wHprx}`kp|M3D*ov
zc7EDt>L~Vw^KEH%f<2O%dAwTBS9`HgbBsz&-b@wKB)nRj6l}KcRL~@Fsc3-uB$!_D
z-+;rD@DMz_BEuKQiw^*86O4SP()#NS%Tqj^2bacV5}_~4%ll`VdV3h9>_sxrz|^KU
ztAG4i1A$S3i+gS7-&#sVF3*hMf(k7m4Q&cZuy%tWjDsti|ILTrU64&{>SODbEZ-<Z
zf~hUvuDia<6YAKU`SkE2S$`TAzDR|WjLl*ZVW9G>8`(Km!tS073C(u0xQgKbNx$XP
zm?h&Lo<99f{&XQGxU5>IntUjTVr+~A7fP>)(~Dz$i@C!9E=^}CxP_6Q$Oh1%Zh<<0
zP?#c|+jiX}Oe*JlLI|9AN43~~F~?i0fh{Z|?JKFIL4js$rBik^U@YnPAhK;f)3s3C
zYDF&4ISisW6FjIhDTe`u_b%Usfo3O^j*m#Hmc&N$q2RUPhX>gtR2Wq^G-z9{$l>HY
z#k~!TJ`iYV(?p#~JGN@pd;T3i<bbj&;)uaNl@$zs%0enlt%0^*G!q?Sqt+$F<33Jc
zl_Q_rk42;~5y8UwO^+?DnCD2ANI@5}*7C({mt#tmNs6th2%+XdI#D5bqp%_MWHIJz
z%<J#^<YBT2U(s0U!>>(&(EavgE&pC{fB)QU*?;{ii{rlses?JUJ6yEepD^{K^Sh<x
z^-ENc;1#gL3jBXOopn@{ZPfJ@1tpbEiDBrWhenYYU}zY+q)S9nKnbO!JB9&77y$uk
z>5%S53F!twQc6MI%jbQ+?{C*~F*EmlUFV#=_wUTk&R(v7GHoXNi^nF3L8<BTarU*e
z%)zZMwG7SoCV|rUd@si5V)_2-W8Sw65#A*1U~?ZTL*$)tS%pp&5RK70BdmNcR6tR<
zFik)9*=t4G+M9W0#S>vy%ti9?>5hmbWzqs{&4H|3SJ3xbL2E(At1t#K0Q!BJVMK!1
zk=5kRiaz|5{8W(<%uqS9P4o2rdSe&Fhh`qUwBJac<?wKMr{?>ju`Q2n>;c1|{K8y@
zz>yW$Fkj7069y5KFK~Hkk4^t8ZAI6uP(iau;VX-_lglRM;7nvbNh3<p@3U##W=7@)
zA1oYFh^ZNsM4@wYGF$@%IHHA{1Z+Q#e)y!XKIw$#WnyDo6xlhgi4F_Wrexz{)!d3Z
zGHum;Rdt_VP0%Z4GRX5fp+r(dY4hnB9z(R>cgo<BF!_$jyAzqr!^%-XuwZ5+71le3
zA)*bI>lDSHCXe#jZVC4%OOWMq!&DeojoKIqyt@pC@zgcxS~jlU-SS8V8!O{&KVl^H
zRBv=jv1Pr>uz)R1%{zu?+c}FkKYC@0v|zM}Xe<!^XL2YiJeqb&Z!BPZ5-7gpqPeD2
zsz;2mC7iuK_NoDL*l8S-SbAWwMat3cb@qENp1H5?m1NJPUcmWxvg;ou<bwU)o-Kb}
zds~T(iqIb<oWJtoxhQ)x9uL~3<i>*+Es;g?yPnB<cU!W|am6$9`3G)=tmY*9OqvO7
zq}ylh8o^KM-3ISTw{*(<K9Ya*$M7wnodsMB0W``|PZ@8Oj-a@gb&DTZ#`KG6#xttg
zc=BV4Cm+AT#RSvr#&*nqldi`9sy8WtKPDW)U}|h8x~*~R4PFFUe^_KrZ*(w?i2*)!
zbYEU%L=G}Oif?XOQ+$#1E~y@=#(8Cq5Mh#VhyI&Zyh1W`gGv`gS~qB<qOyQI+BC|B
z+#&!!l9<YNatwkD*)Of^%DQZrngAds!%p*djN$v>Y<QfA#ucp9-1zz+Mpcy(%tbj?
zUxSaN@SHKr_LhO#1F=X6nhOxc*BF3k5JHVs#&seFR4JWeV{77D=eCE>hrDsQ3sIs*
zg|u6dLDU?&YT6bO`2?X$gpuMB36T??A+ciN!h5H7o&_vD<V<1|!}?H8db-V75*|H5
zvn1q1rX$)_;D~n68m=y<D3@x>$Cm<nN894%@^9r8RQLBg+*<p-ZeC5$ddF@`SDSZe
z;);1~bvwh=WRpJ{)wwBV+A0txA<#C1!N1hvnsr3Lsbb&g2n;vB>{-h8VrhL?il9k+
zm2m_4D)$MGmnKX^z|M|uO01j~#g1cqCZDF%!(7xxV#X-bPa3-F_3?X>CFt(tVx>uX
zBQu8@y1sqm9?(l_EZ_qVm&F=Kl!I2NjcnV;8A&#jd}0Wm7B}=w#*x({@ZwT$?#4*Z
z3FM#z@mDf?47i&v%ST4l!Dj)4DZjs0d;yu*UWib&WwG?ug><!1myqs)6eg$F;-k<j
z`RoQP89S~o->>hM5YIAYTsC<$WfGjc)E=3;)6JPUcxL7T1gi)Q=MF2ZiyQp(a3vWh
zVh^7C{=Ku)hjjGrWqY>o>HnD~s9p!g|MTmg#pb`ZsP_N4{MtAEsDe=4Tzu+7gMj+E
zTmO9N-8%lrO#TJh@Q}#d$GL{BF?z$*D*wR99aQi?ID01>(Ntnl71SOTD7jQ9a7mKp
zNrZ7*I<B<#xhqq@x09jLew5w7*)7cM62TO0Gg2K=y8lf{p7F##VE4CBu?RiC%Ej*v
zmV$}+q7t`f)$OC2(@uA-{F3d6=FmAjYD~U-XuTw6mVu*to>gd!kDEi3;O=k8!?oe<
zOp^j(MRT=2Er}`ja(Ftu66zF>u@bTjgOg>%kc~Kh<20f}@!K&#>nz)+qYz8pnac=#
zWGc4*J+m-Ad61#_ZYno>N#$#`H1(UY{`(Fp+O4ZQlrJ!%i7?SDdRuf+bgr_E(eN0w
zQ&3OTi3%mtCaU+SQjY^k7y!ZzZ~fyqNl*X2p|4sHFe)>ilg^sR4poSlz8vt%fDa|G
zMQ{!|7!Vn=CE4{^)LZFL0M6^+O%}+z9$L^yet((5A@*asp<Nm<_s%cBOZ6~l%Jnz2
z!-j`%l*h6}^>UdZ5^SAPW1$glI>~E9e#;IoJkyhmqM0}t)Nx3{<me8*fE7qj{5Nl-
z49*jPKa``4;e9WzIXbtHAaK_`qfDA65)IMYs7T=kbdIMxdXaJt*?3j2D{T)}LaFZI
z@p<(;PGNIlRCi4F<f|Eg8g=*ShExTD(Ur{MAomyd)e!!LZR;YTz$4bDz;%XQ22TK8
zykFGB)?sYury{?_JkiDLa{^6&`WsvKCjWDpUY$M&5~qG|uD(*SPQa!rpYnb`lH|g3
z?B-6gx+s-Gu)YyTG~o)|q%~W(2GgD_8mCkFN!+6~j1Ze#+LsJZ5g$qC)6xN@#RL@^
zqE1R0PiaC^t6?7`c-Cp0@(mMO+6#>Gtc&|2zCV}mB2yFC5+qD1taj^+tF$@6N~~e1
zGEnQ)6O4)?an2^UqjTdG_G+PHu${rHklBZ4um*u4;UL|u!_;C(R<EEgcYZfCyxvWK
zy+4Gf*2VF*(Qu#mZT#<SmffoE2e~0I9!?X<N~n9_tZguK1PnW7=o>)yGnR$d<N>#2
zfC9d3(|z2qqTY{WA%&wyb_iLdftwm>fkCwhTs^-R*TEOy{eCGD%TO7&O)<rwL%C+3
z-JxZ<*dAgvW()3_0&poM*+l9P!iO*J+Y&XQUp3R3Cns#+s@evxM|LT8K)^Y8W@gNm
zCC=+3aB$b^Fq2q)xm_m$*joW2?EQW&4_RtXxw%JO13O<jIaroM#;21Nk`s%7^pzyh
zS|(PQ3)eRoug-W7&v7oNyk$KNV?@cYaJiqpWz25*Dw_xe%64q2<wt@6%A`HhrGeA1
zcfE?{MF*@?X*7f6;EwCmF)F{sL@Yg7ENZK|oOegQv_zjX-p3uw&NpS21{yKhWOs4=
zK5DGLNm4o<mm6PqwXM7zmTx*)Hc^n6ys7Jh8lJO!KB@-b*h>-x?$JdK>L3cFq9Sxi
zDat3fN$qJVg8a{;0IOrazva67L<yJq-{PfeF#sgnm(xDj9(Vwta_?c?pAako>6-R8
z6K_9h5MFZKt<VwnbuJ8osGcK-C*`Qe`DmTtorWE8IxVLbuR}txJZeES2r8d6{h9Yp
zMO#tPeFlug2~hDVirD!**a~>@c6Pp1l^@*8%>wH3&&0OimVqH_!S9zXQ6Su~F^}bH
zycMT0V^M84R)_|(gW>AkD$-II5}I!D<7AEo5r!$Td<r~{Bwt~-5D8n!A&x6oST64d
zK9^*zaXnFtDEKe8VQbA;d>3@3P8BFBzdC1?<afNBw+ZAV`Ozlb9fyjPJNN^Qs5<ag
zO-uhx%PJ$MXCrCb_FKm`os;q<A-@LguyucG`?nN=lSEl<is<IsVrtq{bP)L@H`X^F
zfnc^v?5Ae7fi)op0|Qm>-*R`*o0-Gb8r>`a5A-u^sJPL(+guH8lMG8K3F}>lWwf6(
z=~r`_aEIlOa?u6_zK~?qa95KzS}XeD4BtiE>}fwBJ({ik-4rO)HAeZPCLq;=K2L()
zi?HYZWo1fH8%(AC)ejxmV{9tBFrm(dc4T3{c$i}1W>cfgO2FdgAHG`jORh%Nxcx5;
zodfsYLLIvoIG0Q(WK&y5DcYgSJ`9K8MqzJx%wmX9oF5xo530Rqe`!DR$wU0Ka27&-
z+P!=mNhX=6`Q*nzW9=)y#dWId;~gb?Qm-$k!}wyF8>S9Yek*e12(qnr{PXky6aE}G
zrzW{SRt}E%`9*<O$#g6Ey7lzx?7vv!^>2EhEcoQ!YH{7KQNa^YEwC&8`ruYie8fuh
zndT<M#5AoCgZJ-pFrfoN@^BlGn95eNdVhX+9|7dBv7k?A_F0bqL(J`ul>+#->EU-8
zeTA@A#nTw&pIW-<(mdhE(G~RMP~5asl~0o}?ay4fC_Ys(C2Lu=c#3KIb1+!d_Et!(
zEgT3z25juc=!{kTNCxDYcZZXIZ>y4zxyl=en;b)cd$Iza(xb2Xffno&eld{0P1Bh!
zpdX3^$FRT=@a&;NT)vA^fJ?GnVQT}5e<^_7R8(t1GvsKyOw;Xn(d&_}9srA={)8p)
zh#Ym0g4ejmpcwG$LCR<}3Rf+Fk(8Y)5^7CPkX8BSSrM?r+E!We4YM;DB^dE#Y!&(3
zX?g|qjKe@YZ1V<(u(W!ro7+DnxtBO-nUB0=l>H(nqTf)DJwabL2<N<o5x5fJnxz&>
z&;i%BB8qK6BGa{O(@o$A44gfN$kGf{er-Tz5K|#68$VPikz`|DLy24E8dO*t1l3$~
zhjuIdE7opVxtaUco7BT%bqzY}W238MaxK41_RoB*@<Lz((U<7spJW)=;rG4U5kZ>f
zE@1Un9NxPdN%y2x!s-lF6GsorzQF*<X_ZIM<=r4?<z)i=Q?7f`_Q+v|Wksd}__)U+
z10F#iEJ~b&Tv=iAInLY4zHQQKuGFS8uz|yN993%aawO-C4FiKB(0RFxbk%rrhjqT{
z@c=^}J(B_}ZiZ+aPrc>rzo9anNan!Ne^rs#sbX)UE7JdcnQCPwwqj)V4tXwg71$M*
zx=l@rj}D`VV|!SFeFrZz@QtV>QfJlgn5Fa`1!X9q7N(i1w5w>&Nf9@RP3)_-nOR(d
zp6Ue~FviqRWC6qV1K|AzgS_du3$XlZKrhNlFif1`&L>@U{UV6)DUc1dlfMpDm0AV@
z>zJ+@P=D8#x@f&M?7zZ@XSLiqLZk{H&ptRkbryA2$(O@Kz{VVM-fGzj?oBg+rRKLs
z>rtW#2_Ja|gQ9~gGz0nW^J<nF%s5Qv-YOWdk{-_zcDJT5Y8TRlBN<e1+m)?$2+T7*
zqzU5*Z$9A7fb9D)vVU-t?I68KRU*%hEcCFguY8=z`ENR^*p_Pn2O~AkA?n*c>445K
zPrBFW7=Q5X&n@%C2MP8PNsmMN$|lu->IZhkbWTeQfi?7TE)k_>nd=~gWBsoe=<lWE
z+Y_h!)E$wnQw^uH4c0)~7aLg>J6eB)6kJU9Wlnj&-HpS6sRJ)L9knEhcBcE6O>H5O
z<wO{MhejZy+nJ{Z`onY>9<Gn2N}_#lWKf=pcz!pqM5*cg-RAiNG;JTN=Jqs~dv%{p
zfeF>TK?XYDPNWVQwh1^;NdjZ4)gG>)%3-^qWmm9ArCzcU0x%nwp$o^LX<n^v797wE
z2I|B9++O8axd25<R_svPKFK-woxU7&-MKtJ(OZy7KP+2(WChSTr|u@5(_LKm8c^o*
z*?|-O1%0k=g$mzvFP_RA(d>9-41eM9ICt6qv%mgtYs&Sy!Slp5iTwP;_4b&>Y+Q#(
zxnv3jBNg-T$(9@y1;;mq{hda9OAW*sOic}gDnn}li`d7x+YZbT(jH>{2iWKcrqyma
zoeEu1ZOWW`(u*%fZPLM+_DN5DL%Y9~GVPIiYP@a4UeGs#YItffRSo-5-!pl|6gG`~
zQUpn{2stvbZqWn5l0#T)XS~J`EvFJSki=B*NzD1(+{f|kuDPB1%%|T$;Gp?k_!eD5
zyYHudG84kru1%v24;#&T#&tn61n$Rx<XmpDJFE%Y&HBLqqRZZgsC6i`N?gP7^OvNo
zOdz_8B{EBif2}0{B%YdzH1d^ocH4^3HE6A2)SqO<)WoxXH+c!zdM^nPqaL^G;EJvT
zcR(?Z-ndo5=;!m11SWC?T4ZW&5qE}HhWJoCGy<5=fJ=^og2=4R3D<3mn;_x#$@~2T
zN?h$keCO0O7;`_Vve1A?u%_8AKw8Ikw{r!+iF-kG$*7dZhS0l?$<`STYuQ(!N9obl
z<2*%Gf3A{tYU?H6{a$<8Gbn#^>s=4m&%{3cj<qp?CD*konbW$@x;aio4ABoS0j1fJ
zG!J)Eii8q%$Nf=Y@qh&ts#<~7YB%qrN?Qq50@zUI>hV?)E?0Y_9b>pHoN79!VJumd
z{nOM*38~nzj2<9y7QwK65V_GPVIuWzU5Z>7IW!EaWIW0bDnWgtj!7?2T7~#-cdmcA
z+Vyque%KDD#Pv=Fw8R4;%}>`0qxylmH~v2Q^n8=WuKxa5Y%|&jEns+nX_$z2L>~<*
z#!&NruA9%A$O@INaeYdKoB13#nrQ5syrq{x7yzt`Z&`Phj592y0opM_*{9#VYt4r2
zpwauKsaeSp$bPtF4|mDt^#WP+0nF@S)W@fikeBMIZ2x?p4{6*BS#vA*PE7fDQx2)#
zrj2@VHNAqfjVxnly)dzDn=TAv8N3W&+xH>dqQ!N$t1$dFiV{~Wn)Ve}82f~I{MePa
zk^x(;Li9^JZIzj?;INk~cFq*UX`G11I=+s6zqzR+n@pjsBiraa`q_*9O?X;AA?Y)_
zZEkDLRvlYpu2mr{MXljD9*gWX9$EnsA8@=U!Z&%&{xHPifRnqVa=4d#vC)i-q#7At
zZdw64=yliK<4}_VV8O}h^`NL<awXHLB1+|<i&MnO$mxSbUv8gB9<EH9R(X|{8HA5#
zf_7ccgiS?w&l@)vv`az;sRN(BI*S4(cIM&UVe%dSxILJ~ZkJl{ZB>KwX^kXcGiN0B
z2X15Q9fCN1gHeif?fb&N$WGt#aiHEfB~X8fODzQ_$G@q$RbvRZYEz5Y3Yc{P1`Q$s
zz~|fk^q_L#n_YzJe@|k_f8@6>h}#rfqG}NV9`PT3ZQ}ErJGhIJZ0V}0<SRCQSXJVx
zZS;8LRS^n}nr<8lyRSP!Wih+?f-dR<d1wOt@o5`c=3>k%;O9K*k@xCt{PU{pX~iVA
zmte69<LOLB-lFG7idRH+16c96%pbV1LR))y&wVkltn{9L#QcN9@n9B-cEc2YcvDZ(
zeI<hmcsfyd5|NnNK~l&bTvod?87mW&9)=&rZna9Jo~t)*R4ip`#9g&O(Om)OIY?6b
zaR21tMwCYZC&3Q@XCHdM&zAWd-&n5iKE9A;CgG|9#^cMrPOS)#M`?T`wpNtzeQ#Q3
zE_ZVQzXKexVnYRC1!hH-D5Q{G)Uo1?c~nEyeZ2(7)M#7-zEu!0GDt<F6^fBtoHuK~
zhNx!#+m9O(?55l?RyWFr?|%XrK3dU9hmF)vdDPdAu1`9@39IyMqn`g5Mf%d44&QV@
zPRvYHuSp!cO~Vk4%kec%hmpjzl1P+JRP(vMQSpp;Y9xCU;dD#~X>HslgPI?^Iml8K
zn;P@El`b>)wQr5NdJ-X4G@cJ*-^AS}?Ea`&Wm7p`Zd7Op3eIwB-{vddkZV~%PGs~|
z_#R~7%7JXD3#VDwjrEk{rV5KT&o;<KA?GJ)2X8U{@~1GO$U(kKg5m~n>*zu8Ycklt
z8wKjlXx!-lI%v?cQ~08c5viLCjyNw_4tqc(tnKf#b?rX6y%j{K)@!`!{ifnBTs3nT
zN0)X%hLQ`Gmd0ah$|p=rMuCra9G>JD&P^>%6K1+?2y|Z12!%g$Fm*};T7m+y{}7&w
z8lxSCxFa8bQVr8+v~t_j-J0PwlQmO#-Jl(m=eR|1?$#Yz6P(B6xrv`$Um23e<#*_k
z7WI?NCGjv@rV8lXM0IZB>s4`HAJZkTGd64gSMvjiE5BEKOK<2Saznv*;0KsGX$$F<
zmVG<#eUX*X11bIa`=wsA_iD5U;nd6_eCY~HTZQX<G_I14b$blf^i^@`P_fI(%}=6n
zxz1jBGzqM}E#n7UiS~J?y@B6n#6<+H1iG<%_RIO>6r>}k)qifq7Tg5S4u#PT=~=>+
zOKA#>gc&7vgYYMriTWLWmGkSSrnitc)=nXAVNgoLhOnH-ifw54_HZuPs*V0I$eLq2
zY8#rsv>J+!;Z_*<7ePc(xMXY8&z^{5&M9#1^AnkefX59OYig|Bg$TuIE_y4~I;dz9
zlvZDJ4mpai)O;)yqQIT^Rcr8Z0jp`<JVWV~$jw{$&0XTk^g&_tCSw%l*h@w<e_NIM
zdrU5;&mW8apGjOwQNIXO4a8|)^(GG1hjq0<ZK&?{7UXclQoN4jJh_}wRf??=^KlD~
z!}B6F64x?^Syl4ek7s%Hi}1!F$L`)+6WO>5QEL6nq+yh1!1SB`Nf*XQ@^CM6kn^=h
z-+RL~e+~aJeAiXy<qs4tyTHo=+-U8B6kr@C7HikkAho1DrIZ@l@v`1dn2?;4U_k<N
zn6AqE&x3eEj%PmT)YTf}@oupcoOhWy^i}3!Mn#4U8h(#n`9?HxQjI=YuK8P7Fa-?P
zzyWo5?vx&Q(an;UIsif_*&swixbJ;2wli=k0~RxBblFr{9$#b>_+5>m^CH`T{wJG4
zNy5(er0HAN)drr0J~y3L&st#&UEmRxeo^$s@7ezs`UBxUD0!?tzlu?@@duE-+&{(r
zQeh^FJPu0F79i}Bx%Z0C7RA=@4jM`7Ayzj_<wOP)wIfAr4ul^{TnR<hQ2eL?38DCe
zaG^I3BF2g2(<*dRHfwgaFalW$go>t^br7$8)ht0SvyRR10zAweLRy7H%3qr-`wSY6
z$wb`w{a>k>S7I8F=*S=^x^9LTc~Wbm@dF5HePJ;{1Gk$v!La8@T#4?e-dO055o_Rh
z147I8NfGniX#ekQNnn%?=Mm2e1vNQ$D(CNp<M68-r8lF0akVw#^OK3)QfDxn?yE3)
zK4?2^)rRb-)0onTjPI$E0Tmt^edpo50j)wKI_8|hprk2dh3J!c^qqOGw}j8Kusp!w
z{u0?SDRRk4K9*54{Pmthrh)?D*Wo+bS=5qw2p1oG69dq8<WTcQwnclsd8ci|xy=m%
zwQZ19pXXa-of45aVfnxqz(r5D%G^C}Gx=2Xk9Cf`Ojg*hxo{YH_vf7*J)kmZ+2E?D
zaMBY{2pt<zY+xV+axx%;Bx$=61#!n(PH;Pd2Z>$9<DJaZV*`ZMx3Vv<+NB3+UXjkK
zzwMdg20?_QvPH=a^QpWKJb<S?oY-OVDk;>Qp53yYyjaiEY$L0KOBP+Z;5uDI?|0D_
z4@v8dBvP7kBU-+pn0SD_t(PF6i_hP$rAJ8#aE?P^L!*@4@2UCWU?<5hJ3*PNc05(e
zs524{zipT7$h<?8W{I0LNx6E8%-PDYx9vZL4vQ22e4m9e)<58+n-t8+fwGY%$yQ@b
za`vaa?{v_PSk+*AME(b%TN!gBq2#4vN4^;0NXR(Td=ZvAsMz)gfg87^Dvn>_al1wm
zXrr$dGLsxzqo`+r0r=jZ1zP2cV3*#p=}m)02{5HMH7Tn!ul4cwZkJX$z*5Mg>T2}!
zmbaw0UhRJFoN58)&p_Wl;nhVE(VZ^_sG!%J%Qg(-8A{OlC7;5`(cd%}xsJ&(@PYrn
zUbL;$uLZet@LjKZyPQe9!`rHSe9IwGp~hm~%&bI9@T)L!L>_*Dg)s$qCUbQ^XaWQE
zgaFHkhnkru%-L-D_sf(hRa;>10{!aT{ww&itda)AR1|WY0gA?z)2y3uE%yS--ii;I
zz#&5E=dH6<SGSCXk0uO!@8|VYZD>^~bf1yFopOasvL&Tpx|XLoMm#Q|)TEkvNm;4#
z@gIohZu6_*CnbyXgLe)|)wabZe>?A9b2xAyNe_yrNAf-H)9v%PNf<MZW5f*z6h(z@
zj`i*9DMD#!1A{m>G5+GN+e4k|^D$j4^}2$?zk-qMLPB5uT0jqOVFss|qFsp+4x4lZ
zjmk`-n?jJxo{8CYqbmW=hwV_&#$fykj51@;rGEE_pZe)nIXxOCuH!BYYKA4Wu#{(T
z^_A1U%+1=J%Z@T>upqU}y-{jYY(Z+>mFx(#c2oLAz3y$x;%moZjW&x`BW5lpG?;#-
zK&KSy7vD`yu1{}_F7~xto0~D6RyBEk(L~=p49JWW-tub<DeZ?k@+FcdI#<*pC=6x=
z%Zbv##NVTY%*Y%Zn!x7;YOD$X_*z=-$@Sr(<*q4YW=*2TMX>vLXk(Hnt;%e0R|0w5
zAA`oWN}oJ_4TQWE`yS)CgAM}>IYPh{!`yecN`VjFUCs<{R{k94BGCRP6zp`F8W-L5
zKai6GkTT#kFZ_-PHy}ji2dMSwa?x|&yTf?p3(^JF4nl4i!c>RkB}KV`lMaJ4x7BFS
z&gk-jR?7Cb3l5p75#NNnCrn*U9ei2aBiOEWZFVKoAE57#=~S+kx`&)#5Ywn{(O*CO
z#vPD&^qxOj+=MuT;Z3~cRHeispCGXgxNEJS(5$_zx`7|Ttd9G>1){~S4H<vf-++_Y
zDXj)BZ5PO0lA1#ILNmKH4j})g7E$r&CGG5p{D>8RZ$vp7EXIhZhcIZ~GG<FKkwL$u
z!L+v$`+9i?VO1!UKj{+nrz(0gj%D{ai#-#R6Qj5k-%Pn9|4&b9zK!5>NKt1rDJ`Gi
zVqxhXttqv15OQSiUYcND0fryAnvY*zd+0hyf)`y@O`X=i66z78%+(bS5J9Y%8>lfq
zs+m}7?x3}n*&wGjmZlr@9rZ?#cj_7z8T$QnVNy>y-%Vlrw<{qSXv$--D0sphf5Ch`
zD6Tvu=}TFNnlDa3_IOs87PC%)ZriO-Ms&c(E|lCAZe~^gscH;%yb{V@uX|tuC?0PG
zf$z1~o#o!2=Z`!N{}a`&v~Tvl4Ge$cbDazbaIV0=p+5nB+)On&Jo8SanjeOk_L2z+
z3=AbiNr9wVu|U}tuiwR7_7S~~$yXl`+~Pk9twLC?hRG>6&wQS7nx1=<<pDnhG9fS|
zjCosEHdz$>wAhhbB~#$~@_ShD?dof$=9>0U4SX?`i*NKj(aq?gu#PGz;z=heav3ns
z-d<r(axIhE>>f)VYKY0>GBx5o{MIPn{qA+%<g=_7h#oDQN*v%0sj(KoM`#;)d%E4H
z=XS|F`T@#f4SVI#kdgr!Oi*4iD0+QfV#8pnP1Q0RKHQKYm4@(Z+CiXQzXy#$A!aZ*
zIVZbZUkqH|K5k*T%BId~g@#40G}(jcc!D*20-P?H5LUP-j=32bUUL0wleoEXAUu%T
zbkazu36b>N%+DXq#;w!VXv{@bdmc&K7g||60^6T`vEu|^dXv9+t_=BQ8K#8bkx_hP
zLelL92<uIwo8aFY9j%KAw9R2byALa2)@1pa8mzhSTenf8=F!W>wn@jlN^`r!h_!nK
zkQf^}=vI-dG9!^c=$#ps+vl899n!qi)bkvlf?00BU+?`f&_p-RFL^&d<JuMi{-omv
z^fG<*(|pGk0`>xNGcNc6>wWMgnO85f9MOWrR&xTjbnKVCs{^|+enfi9Ge0G-kA_%l
zOh|gbukz9=-GGr~#z}l`Zx|%Y&-ve(t%@z~9I2KuB~l(ddHAK`u8C&44=G6XU1#h`
z;QKh&gUb6dMdn39){FF^joLe}|Gqf;zp-}N-zYKK3ocvv9Gj0Hi=rEq8I6l)ro+PX
zR6N>VBqbbF5%Kq;-$JnZkC(liYF^mIYT1Si2oHx>5+M^0+e3;<EO)rKAv_{|xBmzY
zd#peUzE~^~KP0U+R1W{hZ{!&Y42rY7QIUleQorgOxcLH%5>`XgU77i(u?X#Q?b^e{
z<mNZMRizm2N~ooiP%7UVt(S>sOD?XtjrUfTS0*TijPzY|Q28!TK|uvFd$o<i;6h3)
zJ)C=y<?-w-7Wo=m44Z=GLhTFK)EE<?yhjtbp8^9ZEOyB9xibe!61mib-TFJ@MreGk
zZWoOg(twC(Ak`SBBXH@rWlcz&-QhD)e+!5@aE4p*DL~Q&j-{qkhr#7(>)nWmt^&TA
zx>2sEFJCqzg*AhkdL`18JbPFYZ)Z15cWVmIF|Sut3$5e+aKy5>hwem0J+DCHfX5hQ
zi#+@Blvd;aO51>ATI$yfefZrkW4TykZ#d(LLQlD~Bk-)gFos-g|GW|erF8WCT_jbP
zT^9x4kv{SRJ7xGlSs^zVv~-Qx0uR(3#wQya%gJRAX@m}~fVh<1GRLk$*y8T}xa!k!
zr*XP;65s^!zkBX>?9=|`<sWn+OG73oqdzmmU?ajC{7%NuaNd6h+nL$3<52HLFlYgs
zOUpu`fXuNot9-B&$1zRq-NL8U9$GOHj_9VlV3pwO^++T-PdqHFun35V(gV}Z;<^vA
z+~VK&CdTh@l09h@v%y*(f(+@G1OLr0|2Z}Pb#3Qw&p+-9{jzcsCP=J8XX5}WiDH*d
zePy`zk0AtEhyFx&j1Ks_U&AQjDtPKj_H<C?sG7ezGZb!8QHMPjFYwJ0>is3xT^Y2h
zMMQs)SVFq{m~OZh1hj+1T-nc<^Q{+In<FCI!*W2wB*9}ex(Sjaxn|=M>54!-UA7@}
z&so(H@mi2#^Z=V+FYCi-Uez$PxE4oI+cx>+M6eG;Q~iz}%6Ek^#rP2*t(u)c^d+_%
zq&aN@w}f4-De$>d`&GVeY!A^*i8-lVg-o4wXKKiE=gW1C(_RGi^50AerJ*4JjX$Wz
zTDWykz<%roL8AbW)q9B?+8%E4)M`yT9@{Ha!$Kq~Vj?x*@D3L;^g#?i;y5Ug>xlKv
zyT@joj<PfbHuardY@``e*%71(O#-nny-v=vW-P5dfBU#CrpEPbxFe(}kC)_G!g-~x
zU_HnJIyp)kZe=^Aw(l?OdHnH=ZBi!ejl$7OHF}D67R^LdJkv7sRO|^Dbo%zeu&&~1
zK5)a$?GbED!PUx%LctL9ooTU7HRuVWv%&RrGLPSb$~It%@%d|-b?0H)m7A+5UXd|o
z3lq}rOT#AduT0wLOCC_Q6{RB~R$AQ{i+1E>$OhV0y}$HoyH0vu<nZ<7Z&LnUI8B|P
zK`Kwvkwa(P6O)bW)7i6Wm+xTGahxI6csW_ZSW|kC_Q;DwW>_LCltDw<eDP`npSg%=
z(*#V|`kwsB8E~#uxt7&9Vd@0ck~yfBF2{=-8Pp*l-9*EYtQM2cYO#OJNy)qc;V=|f
z(Jx}JDYbGC!ayX?JL(^IuD#CbMPp#Un#q!hQ+&}vgye4~!i7mVZX)$Z-905K$>o!f
z<3Y?w{FP`QLHz~#3p62It>l9s8o{6uBrL|AsE`=`;e9nC#Om8uAY(SbR@rD-`Rz(v
zDvU`HskQMRficrBLCsV85ERc}orCkS6}p0B6cN5?OL`$A%<<`1GIRg*cCukHfuTP@
z3LIIqnkWfZeg<?f%cr^GS`_JFShv!3*HWWg^Kt4@pEsx@Z1k$*jnEsbfv>xq_*xI<
zjar{KFLr2k32O+wshoM>0}lM*T87jjcG+aXUsl8X5~k)wnSE~k`TcFZn?P6S?_k+y
zR2L8dl{wp^fR;Y%0ZI+(rYZ0%CUCsI*2cIy77li4gkEXi&}n;tatNw5Apyzg<cMv}
zj>$qP97n*)xvfV<96QUgqE?J1c+R8&k}I#{JJ1-&8c)LLqaTsIX)lOkP8tyJsCQ#7
z5e|<vk_o0fhjfgJzU#t(*VKrcFtPJ%)J6MG#v;gZaGZBtlg=`Db~Qig*ZE3gor>Tp
z?Bx6_>u==&7FXzqd|{6M5KOY&AQq!l%*#AyVbU$e>?6w#yDMQ_vKVV8g68`3ER$07
z=K)Co41mlFu=Npwxg;+d?V`{%+2=0E{X9)!xEdn+E;!p32eU^5^$p|WacEG`RtAa;
zNM-bB0VjGt2fyfrNA@Ka=ir{RzZC8gI4~xd!P3GUE`nQPcJ<tqRGXIl%h2~$1e?3k
zYIPoxo1Rnz>eNvcciOib$I1`LWr@m+Aeq1W-O~Y+A#e@fPJ7<iGa(bPiy3{#5bJ+>
z=aF{m;h^0EnLd0g+5EHB<SWZoYaZKps4nd|5ymzvC9<zLt{Vj2a(l^SUOr0~=c%hP
z-UxiT{`xARN2!($uqXhCpv39Ob<0hN7{PKI1QoDgzwK`6Yh=8RAb3N|4*OJ}fi#9M
z{Fjflrq=JLsVei6p^n{A+H3;42!*=sTyjfJf|ANQk+Tg_YZ)VE&Sb>Y4VXIn2%!z4
zymd}M9<^jztS|&HHgY7?xL|<YYTzwcI;j@{_BHVh5gb7}qHu+*3iP@~Z5k$k(}k0s
zo5w3@y%LF82Zn@(s&Ie>GZu7Pr4So5EJA;MKs+#NSb{M%<_3Fv<@r~%M^<gvyR8BG
zK*L^+-hq=WpIrcF%HZjzDc*LNya9y`ez$D<Hq6xGxvsN35d421hkM^E__dLvO!(2T
z$+e~LTS}3@CMOxCDxhYpmjqYPAaBGeG8&P9d<oAP-q&X-d$0|^9u_tCnjfNIA=m2T
zHQlF`3L`NWgxS)7#CAcv!Qzq(n&@EYTgUtVn&~tqB5E_-f3tZ{ZXL^-S<&lXDgGd<
zM{xFv@kz_<=2enJTuJi9U>-@LsjD;ru!B9~IDVdT1E*LdgD^|z`lMjk!ylNb17F?P
zdhWd#$p3Gt%il2rCRaFBsIorAU=Ucwai5*5f&Fx)`WrW;DyxcuzH`@TJY@v@ZBuW!
z70WBbVk1s>S#xE==4ZgISUGdw{+qOl!*tk<b2@_gbOC4Ei#uKK-{f$u)eA%qUWRl;
z#6gin8>F#0uD*q89fIY=xcvRoO!ZC8GQdF&R9KYWV=8<KEoMr&*Uz^JQcGDZh;fz4
z00mU8hcpBiA1H2aiP{x<hYUd>e_>XP`Jemoc$KhHgz+|^>qT$aqhPDKoYV9#fhm%@
z4<Z@~i>x+E238jP`kk86alY-Zs(3Z*09boVjD5u8N31MBLX2`v@X?1LyTKd7hPssY
zO&|=f&N=$MfDQu8w}aed;0M*PW1$y`0$`nO-632GDVAtqHA)u-jd~AB>M%hg0N=>6
zB~7CKM^ssCdZh;3&!uR;#Zan9h=(|d06abQSv(akwUpe8qc?7BMqB3qwme{2W$Cu#
zIsCMdHML)Ig4s53&7;taeNafhm^~7vqu3pwkW2y7BXUYrOpYM*iC;@53SGw_6QVpu
z2~oO6?Qn`C(`duzS;L0BCeELo-z|JR2Czc`<oUq$E-z4tWzNqDfZ%euns(*`_uL+V
zOzx!8lod$W%G*}k9py>+>pQXE!N!Xt!-avR-pd4>U|epARWt4iQ&Xem#GzL?an_um
zMC_hojMiEYa*Su8lt|41xVgdk30m_^YuJ4yHhI}Zw$uTtW92P+o>X8>wcO$`HRnFO
z6~}v6eFylUvkwRZe`6O@{I6qn;s_o;0(&w6ONU&u{Oz20$0TJ9J;{W?vJKrh_crKp
z!7}M(C|G3FZaBSFC3g@zNk;|dUr_S;LSCghbdbVY%fe2>uZ^1^!Vab%z%d32gAv7)
zBxd&Te}o#=n><m5EJkt;CzY8A)Q7(mvoyLCaDb!XeChh<J=4!cu)~NbUhU}=;=f%_
z?|i7gL2x~NVtd5k!Z;ExoN&WUlF(-~gPcRcjs12n213U4_Z(FWX2C$h^YM7z-Un+P
zqx^2`Mo%|reN&Q`U`_y$$Oe8X%AH&4*)R0B)+wOKn-sl?w<|-qfj0syLzGeJOPF?=
z1_LEl{tcfMFu;(_1T&5CEJQEG)^|C=s_@f3q1&x+9#^826wDUnc7!WhBCOlJt5X-(
z+R1wfEGojpPM244TK=&A^o<udf`RB=-GIonVbGL+Srz~G)|;$DtD6I-9mzzT8Q)h%
zGT;mkW=BW$(o;(W;}qHg1Wx-rz|FMc*QoHO6TmcpwL}1)jKQqwzRaI#GFy}%Ajcpp
zGpAnk?8f=&8qwG&x+ruA`Z!?~0~wFTuUM3E=N%Oa?6fBJic&Su1g%brZu#lHFHzBT
zDh=KAe^`T{U6UgJF5u>c%#@-b6urp5Ht)D2jvr8;LYV3rI#ti^NVk};uK&3eX5N|o
zdkgRSbRHFOQNr59xf=K6Ne*gKmTZ0}&|?*SU~>0H+5n(StddiwomcGeELPmS{nwTH
z_4R(4bn7_y3wjq$tjER#C|Gi)nLrf30Z?t+yFTn~oog#Lf6x~4jFe_IH^tcqko!h;
z7(<(DlCdcPdL{Lq)~~jJeLXae5NrOBy-ob~7`^KI_?>*CbGfdlmR{&zn9Y(RXcJb!
zP_jiS6Cg3voz+{hCtc^BYdv3|5%+xM96oBLY*H5YRtgBoS4Kz5nTL;6(c5-9YJAgo
zOjXMaE%mA5B4AZD(6UWWLNcEsSkS(ASV_lx91A;jx=byMg<P`k?s0*ek}|cp{hQQ1
zVQ|5@dg6)=9wcQ#2g>2=Fn-^asH%_4Tc*bDJ};|eXLNn9ZHDvm$bk9K73P4GpP%W-
z1m3Uu{NNmw%i*7lAhgq?g==SzwbT~$(x4_y^?i3-z7Lv5;-1H;vPlS*BJLoiKjku;
zwa>jJFd?9XD-bQ3Zm1DR{RohWoB)@}qHa`X<jIZ4OcVyD*(qJP=Kza5J1-$(vO0N~
zj47Gmqy@33OY48oT}5!CA1qX;&S)7Hf<zzh;b0F-19#O-`LtsSFiuR=i;n<|#?V-O
zo#{sCPxd%Gxt+7g1vHR$8SPAsxZE~9WAnK$4|j?gV$A|hvQR1woR^Op=9j>Q=Pde6
z(Io_m1dT1LSevMg%YzZO)sWnHhcr(NTX1uZ5{4Rh&GXCz0kDfhSs378i1ZXwa5Jwp
z1D!JhPx~J!kqi2gxjv65E@qn(FZjT-hL8;;Wjg=$lG4mTg7MBJz*+eH(8*BTJ`1Y_
zL+pbyOaJM&=QrU~CgDN$vMP%)hNT!lL1}R5+}2$a(6|I_NWpHCLSvxcmyeZm5%`RH
za}i8C8%zKA^YuYaW8G91Xg!8u;TOvssYcu;Ks4Mtdw8k*8n*AfS?AUrwP_;V%w`_;
zd@oLdj_^ixNntWC=^~WHtyZjH{jS+OOpUf461D{tegLVvQ6|lxLjiemRDY}<%OPj@
zNy}9-N&<7<N!XK|+Y4qq=Z6LQ5>wo6J2u@ruULT}DweNGw|dkzqJ>)U5YaOfhw?;h
z_?dLl(iW-3Q^9zb#wH%q+<#ehU!4-mu8IN<vK)F3A;R#9pp?kh_L{+9wG;4blK~XZ
zk;5Cja(F0?ww5Xx1+9;pS|#&l*;jQ*Ho?hpoAefHGAOOBnOP@6UY#3vykP9GafcPd
z2YbQin$L8#^=_IzVjxs$PP1W%y9Ibb5T0`N81W(WxmC#N+yy(mNRknH!qt;IfAFvP
z+G(F!bknn}%DkiY7r{gM9gRxl0jtfZ7k6;ZJlqO9l|*+;jSKIyN%(4ORlym5Z8jab
zcDmloV|;~QlXDPp%=^frK8D^N%;}Evq|Crc9UqB<OcbAT2kIp~ag<4q@SON&zfJ^n
zg!Uy@;LkgJM+*_Em`&-ep#*3|R*igt^U?eLLoid6=ii8#7X-#a<wS*VlmBYpx21f?
zG6BmZASMku!{gA|g3F_fo?f!cYNn>n(Kxr@5b3v10*=EkO?JWWLt1+>G4lNImtF9{
zz~ZADzk^TS*5A17DtojB0x7<kI^J-)n%EL95Ra8yy+JXE^TtfuR@uGndN-L^6ho4N
zEAP3>R_7)+D~LRb79dTC56A?Mgc6DkhUkvD4V6$4o)hTMf=WiMjBDBCerhEqS2MFM
z;B^nvLwp!zmVM560QRBF^MBA2Du=V&k)V4Ogdc9v|K^6&4q?yVygZ?lE<^+lyQRu=
zDxni=#Q-&cVa7z~_19G>w;GnB!TVpt3mJK|r6KA}%vm>k98VCDQuL;<m^1OVd)u}+
z1#PFc+`M2yWac}bEkY)#MVkUM$jGt{JM)z+6h2f0TMZLbnuv$;7;<GuSDDEhPy$oa
zSouknPaX4cI<6p_s%^bhf{g?amK$J)M_3kb%}pF(DqE=U8;GhsW^E5gt@#_p-r{-@
zVtx|v&#d{E14g<CLv=ZZJigoBbWVjDZJiZSorIQ-Y1y{xnvSM%<MB`?fz689Z$A$}
z4!2yU@|+2#em1}B(eHpf@w87X#;b&SJZo;xTBYTG$ROlu{obP9$?BufVf&3l>bqf<
zsTDIIcv?>tfC1S!U_iY5Vba4ef-~tP8hIORqOK0>%6@go0Cs5dh{e*@3$pTCHJaIZ
zFO8Lp<UA0A57+KiEd}|ab5(K8jSqJ1vb^K$LH;d1LSuKe*cM29pc9x{kibKBKHy97
zdST4AcWN52(kB9Py;65j4KBSGH64F?Jm_L0ONy*IKMj0^z0s%=2?evdJGbts>YZO)
z|6u*!n&_<LhT23jf52gsw9moo?7!RD`KEjxetv3sIyt)BR{?%e``>?;->VQK9}@`3
z6vQV3lV@U+RU5i2${<Sl_2b%xxt*bZs7IfiE~jsDiA$~wqu#8;wYMpLyG*UiHFmbf
z0C&MgOf$ZTdPafh4>f8DbZn`P5Cy2UdZT8X4FP%Ac9#R32<`$@AFfKmI#j;3YP!Gs
z{?Q$9Q8zt@2{-A)!}_r~J-sm%#GyR~MgJ`N;&H$eHI|JyM|LX?*%jMXi5+Y;3R=r_
zi|UFxu@6=D0X8%kxPX!woN%t<_+vH^8|`;j6uCMOo*6Ktu5(QB0Og~A9Hezy;}P}P
zGjq;yMMd+$&NxAMf~QWE7lZopPb?Q;TQ*>;=3T?9GIOW_+SJx;eI5%u%l36|ECd}R
z6)@`*iE0T^Za$g#0#89M*)AEI^OV3517?j9=^;2Ph1?(NU0LZcy^qZczkjS$NCXfm
z7+C!%O(uJ2nYqdllo=>Z0&r_Z5R~~^3?5RY*ydnW6Lbi)O*K4A{l}bnQuuv@bq&TF
z^*`AwO<|t=5?@XN;^_a7x5GD&bRz6)=|4>(KAC4mFdu>ZX^9CqpVW~;V-<R~p&XO_
z|J;m|JECVVJr<_3wJ35`55F||0z4gn)YM>bkT$dgZZI`Aqj$%<FM_>+^t#b6H2af`
z0{8iY=)+tJ6pHG%j%BK^WkJ{V4$r-d&I|V129K8O3buvs*;=BV)4!({A8y1wtmy;_
z9EkjY{HRrl-I2Xa%TO<QY`eJn#Yx8E_Zz@sHmDzXrBqrz!&zZ^b$WAjaqs`8OijNX
zppx1R?R|!IND)0|6I~;KRa#M?aBIdbtFm%}2wL-*kC#*;kp=?rgz_8Z`n97JEMov^
z^yxJ~juu58TarQH45SgFZsZ#IEn9MZM=KmWTF``X5I;cTvB@9&q>FunjXJgzpuWHS
zkK<6H7g3cdRlsM7$yFh<O!N@K2h?3K=W5b2jdZF}>>91B(O&L%r$kYbzXe_b5b42B
ziiW*?L(ekMl{?GHqDqzuGHH%@da8(2n`P#j+rK}2-g?YVm^2C0<mjN3#5U-lI3_&m
zMWa=@ezju?pt^FFgF)rK;&}p1Nvwa)NHUwNXNr9d9ZU}~>7GgrjL}B%_wyuxR|VQf
z06236Kksu$fS^AZqP;MV<?L+q0Jn51dk(h{xePl$6vGyiZ+*~y!Oo)>1vG5Xw-tr8
zZ$9^`U-aNg$6@}d$c2*0izoQb7F^xXmy*K$R4)UG9?c5Xdj0Hf_cQ-C>_dh>z@BMY
zXeDq;R+O*OHx4Wf(+zUpbPJ8ar^=JMmwpH0zve?1nJvsOKJ^03CO|}<>6JTij}}*C
z&{=H$<!aeJaIiT@O8!h7hWDbC@Eagi^X>{1oVYx;Q9+pt((U6mbDRY>MW!bF$46KA
zwkQWvTYQiP4iD4ldjEmP?`jzh1pe)TXCMiSs)d8cMlP{)|7qi=32q~B3V;DA0Tht#
zH^*8F?7+J&o@3>hsc~yLvGj*7Z?NNiY%u{PO4*$OwOs!b@A?l@)6suS&44Eat<B93
z1J2(DqOY#}b2(S`<{Ni+cMqy${!}Swck*Ah^ZOMnt~540+Lf2i3d<ap6Q#mg3$7S}
zD;t&>0TV#k&)9zyMUZlpIs8iZRAgBn5WRpP%d14~Qj;*Upw(IaaLboL(?1Bh+4Ig2
zoHY4wMiGuh(d?>34`quCY~7gG?~dl@aeW?O*L?m}08UB{jJm%G4}VV3ivWZKAamcb
z&Q2_~;v@qpaczn>cor;8dX)I80GQ`GtkSjg!ztZ^_l};v76q-CGT4dMzb)N5DWYEt
z;nC5y++`<fzH5|NC`Cu_mBFE3?95A&6P!RZhAA36gm%V|U5Z=pb~zL=Z$5`B+_(T;
z&pEPxBlqBNC9|?{yi#OA0)6!TZ+LpP#Sem1cK$&MW~4*_toNi6HsR^{;yauE(_D+Y
z<CzQ4H6JR06Kf*9wlUjWA|myCyB%G1(yDxBMdMgLa|RbqlGJ-@lCsIduJRj}avg#*
zj<diT!kPd$2;Cqb2`&aOa%6w!y8FFh(7IUfky{I*z)I-Q8u-US5k5TomfWu8uE6zR
ziYPUPB#Bds{H|(iaU7oC&$e)ZVTZJnt`WPVievTC(o}$%t|j>owU%-dKUaL!lWN@C
zTBG36&r0Cs&EM?Kd^$5XoqgX{0$8(LJ3<`Mhk?Hncz{}?l)H4&OzLIQBdS|ks_S#+
z(9OYL^U;LvdI=n2g$4S)9)bVC2dtlW3TJo)tZs0*r5-==v&kB1_8;6_+;<m>I`0f+
zXPWthVy;Yk&J^qWh1zf8D!<>g)Ur~?`@C4jGhH>Sf98Bw$|w*`*c$lvdkc*}i{+1Y
zVHpy#=s^CC*-?sAD$)e9gYE^N1f~p+V?zaH?#hr5g+2=YwR<hyrsZ4x`9TmKaB(VE
z;n0q4KNoNRn7koZDfmln1>U1YDd-Z@HFKYB#V@=nC^AROBr^*GN#c7vXbF+z)H6{^
z$U#xfH}23=>KC_<iX4-h&pbE+kmF^T6;)K3un4V*9@ssH;k9BAsef-nkafHaeT*|_
zxMVfn#EcA;k_1$UI%9!b#i^jC&bM-~e=q@{)7Vg+8e2ySW#vkR%#F|eR-ZYwc)2JN
z&;LM4FT@YY@m!#M8OY&pDS(--dig0ej9V3O3|_o_#3Pafp2oFspHYz3pN{ig9}9<2
ze>YW6@Ued0{&SCrU+u8{whgINk+P7j>Uvkyvl<goZw<550X%*gx}&byhjRj!^>zRN
zym4;9CwD&xj&<Gwk=XS={QTN(x?ma_oM5r8i0tmtNVh|7g5X~cJ+^O&4X|o&nQ4os
zhk-=w%lWl_mCyugk20{s`Tc;9cIwe^Tjqr0YoJn_l<8D^HdMvGD&RX72k{kJdCKuD
z{XW388Hca##D&pdDj#P>IF1|a-@EPk1h{;(*MDAZ<ETR2U(PGj+|Pb!unE*+&euCn
zDMd5bT6_vdlAclqT-i~pCpgUDJv2CAYldRi9kz-hOIty&@#W7hZosBX_EiqYJ0;`$
z2z=XFi)yEFPI}O%*jdUI*u+yx6y+Xh{=EdNb6jOkP%?;m(+#F8Y@kBb%vyI0{l62d
zi$5Si=6ds5`fP9hkH(iSZ!gd=qBn8f07c;lh$ui6sIqZA^Xx8b@>k1TzbuS%DWBk5
z=3N0x&`Gxmi}g@6x99HmP_!>idG~<r1~4QT9xnr&Gxr2Hy?DbQz*M_r4Mt2v{|i@{
zmAqFb<4C#zMSd_%hLd#Pe9|OrQRp1c7t&FZ*w)7R7y`a@1~a{3GRdNvVd-#utL2wP
zvpBDGC=S5VU_f5X;c38Fi?OSDVA!%$v&f!$5gH`|Y4B9ZYWF+e_NPgOn)MeGIaY<g
z=EJdWuk-KHoGyz$PGXxgoI7$#9Zp#nip>EJFPi=>N*5o)MsAi+m<CT(g>)Q)P~ywu
zyYeZV84wcQJ`Y{w-1}8=Tx7oXBv9-Dznk+u&RI^{Xa-(p1B)X;D&Ojs9Qvsv0lmmu
z22%G$&UDTw=~?)I+iE5!67g$9&Kqw2_;DFb&*!gSQVQb!McJ3BJST$IAM>SA)Oov~
z8EU?hXg>Se87`NKt|Ps^Jh^4=4<|Ef59<e}4!_^-83MfGG=xrH3b`0!r5C0j2p?GM
za|cz>u+yF-wpkC@!i_8$*%+cj1IjHOm37*(P-#ZYgZy!2$>&37fDaJiQ6gjsY5nYy
zIi3|pff%j^yz<kY%RYOt8B+)2MVWVkhpcLP0@M>t7Gxj%j;x{<>*u@lai2I>ISQYk
z>?{Y5H-B7S?mt?y>RAp2J!#-^a^O+_Bj1~W+bF3I^`kGs-1^I!Uk!?%Z`@i+Bq`R*
z;Fx@RxgGcHx@GISCCmT1x#`ia{YB&T2K%4YC89}F*ilUmvj%Qt5lXZ|{f!uzeS{(=
zIHw;J<18BPFO*i~X1(M!*b`j8VcKyC0VOqonF9YSvcQIUAUER_Bz7@**r(Tg8pl2i
z0!IiIZq@s+d>BYDMK{xkQSDaaEW=-T5+nHTt|UWoKgfx%KRc}Y-VmPfnfjt?hd2t8
zx2#C(OA-mI>?THZMUiRPK?1!Q^Xs&^KL;1JyMF`kb3|fJ!UtUPC(1fN8osIzFaL0V
zKNJv;#y*#OQbL#G_l8&*PY~wbZ|ZCc2>G^v0si8c7z}e+HC>PKTtcV`vBvShEP}&G
z><3c^FuN*K!FaZWD3@(+fXIjQ1N6UkwHM%afrUaKXeX`WyYCQI+7IwfU6VHEi%i0v
z2zjfOh}hCnf<e`;aR7pq)q{dTV+4KPWxt71Q-xUfr0wKR?BgB7B7v$wc|d-h({ep7
zEz|Qo3~C+mq)xkVMBi=BWfzKPW7cs0u6GNC!p{v^kXA$?I)CVQSnhJ{l9)Advhu%o
zagwI1&(!Ej{^_|g!I=RwAacp71FjR*xu1<tBMmS=0I!GlJ$o=AeDTeu#L9i=Yu8xl
zsax@O{5r0`81o2%6{_M6x~Xg*yOi3~ZjgV2Dc9HO;vhXM0>SNOlv*+78a@$C{Q?7~
zXcbZig;F3lx$Hn!X(UX0&NcFUDxxX~$P2$qEv|oShZQM%s^GDiaKxbOkYPC={VwH#
zuUF$9{r&${mR(h!Dt~k5`al|2+JHUo(%v4QL792i+136p6F-Bg+N9e;p4OrnqPFtu
zfS?WR60YI|;$PMPBmS8inoo_;aTpMt1}LM%2LMA5t2gO8eJoBF72PWbhAN*O#Bt&d
z-W+)E_JE3vT8izRVTn*23UoEJb%K@93Vo4u1Q#nmJ5rb~_69qBB&7H1RvK&oe>@2I
z;u6eE*A(1_|4!Re+0hhruCELAB*p*C@A%vMB*fGHV`Znx&<arICdLpw0{5rW+~R>c
zJxMPLK>c;kGI|p`4;36tBjrBNZCa|#lg*FkaF`Bn8k5ZXc4joPzPQ8JP*iSOZCH}w
z*)Iu5{C+137M2_Ln;xN(kX-RcK^AXZ64<n1+5f}SS4TzFg?lTaNEvh~HS|ahji8it
z4FiJYz);eSl+r0ZG)R{qLntL6Fmxj&-5^M(fZYAw?_2jz*RmGp?0uf+mv2Qg97Nem
zJARVtw<YLx_dq(-aS_e%$LoZtI5J~a$;6La$clh0#aDRMiY_bK2i%%yGYv981)0)>
zRV)E66D;$bU)AE*n~oh@|8;|0<i|yKBXB6u;N{dp<yph|1T9{+(Ka$m-u&Cr>)gVE
z2&q3cp2(86=sf&~c+%z--FLm2KU87_<MfxiII7uEwKmmrB-Di{I4{v&)}UNOyMipl
z_r-KevnJ|y*VXACP%KdRdBPT5#tPz{^yk*EcQY82yqv#x4oBY~i@Ir#k?0I;jtuH>
zSlGqr35Qn&SPE@}f)NnVD+Zp>zuhl$%R_!X={&Q$ph=<P5V!?(-clGP9`-VSFam9f
z4twh7--OjU(m8K7=o{OvE4rXjB`E|g=e54^4o*1FL^-45u`ve5%VWvOfJD3;6n^es
zN7%|U|2U<HS1tA4GLBU-Q>ime2gp!(Cy-Mn0J|#}0OQgi!!L<_!3lVOt6$4a7bY|$
zEF62;b-3ZfmpacfT~yN}`%oZ{?qnT+2Oy!#4@%?!^dOXaS&`h5VbCQJRK10B1;^_m
zfIXL%YgP-YvEJT$I^bF7xdf6<;7L@M9KOQlt61seZ&V|QXT);x9FURv1arRK!6Q2G
z3q1l5|E_Li<<tykdX)l_0XKl4XU&9)myo0vt-#3tA0uA!TFebDn|iVilAcC0LqUO7
z2=^F<*e3Fr2@V&{3fXyRIQkh_(ZR~|w>ma1o1u>%1O!BfVD$%b5qu?SG<sQM>DV1&
ziFq-)$iX)mAg?%MZ)&qtI(A*<dr){bG-8~eT!!51JA%}ZTs|X?jbRdGJGMd07UiN0
zetd|oj$#-bT$TXw!eyo|*7efUWOD1@$2iXZf+7-9d}_Inf(XU&iT%8wZx0LUqH&x{
zUc{k*xC8VDIr`>xwl}vZc=m{-;)1_xbWA(Awm~(1O(ig89;%<Bh&#W2(>I@uH-OIe
z$XwjmQQl2;Sa0O<Fn-IE^tqtMjM}T+jFtx%mU1dfWYQ0djCQJY9U{WDV5OT<X`BnZ
z-Tpp$u1)%@t*v=UKBr=>?%W#><9Xj6AD5x(FGpQ_KMPpK&?p2SYe%>Sd#+N0m&5#7
z#?e*Q#)Y4xW?EdnH2Gw9uz^&@@+*#o>kfY52XwRSYj3!Rs-N$VAmZrklZou<R6)*I
z)GQFGJub-Irs)y|Zu>c%i`hSB&bZEnrf-9sI>(AO>bxb)-Yy=U>jz#m{Ftjyo|^#Z
z_qkJB2kC1}Em}3}L(qQ-Ws?sIK`w_r>W;7q75P%pS+*T!`b>lhFW)PK9#!{Jh5O~0
z7QPmcVMml4b=zdNTwo#{`h0JaCf{h?5~3l_^4=BYHflOY-Ce96lI-*FKl2#-<K8kK
z<N4;xmlNd-3aNbEGqzcAcMbD!SfA%#mCz$Xgi^><c>RJ$@O+uYZC+>Q-D8W*i2jp6
zjs%9~&ZVg8On%xwBzkF5kHSNpA_@tbwHlFHe8Jlu9#EWnOn&qN#B#aNA=y&)floGX
zAteg-2s>gamE!4l80eGX%%B?1d$a@m*4!$jFEcP2BVJCH{qF|M&6-sb?+@4LXdFLe
zyNzc?udIv9T$)d4WjU9t(`s@i|Ma`{drHeTn_NzIe(-TBR*<M-qG)h<s#@;?%iFlW
zHdj!<+2<B{=^JXfcREe`ZuP}D5H-`Ridv6k?we_*3?`}SY*TFiA#qbS(;;h`@r}u1
zK39!^RvQS4h#Jcy%vtS~u<;z!G-C`%Uec8$UgdDo+p*7EG)(iXmsmNzJ$UjlGYmRZ
ztVa`5A?RsUFAttp*4?Y!=%+lk#Ma4oaTGKmwlZsRYcyP{v}_7>ej-cQa>lcF4@lLc
zTAscsbMY!r%i3tF6Et7!fKCvB4PY13Hx&s6PHn~3g1*9D(zk9;^(!I=D|4J1g_kCc
zM9mxbcnOgKFw1h&gKniK$_d~zbawB-bxCDU)c4r2Km4K9u?^1%M(1I&@rzf|Nt36Y
zSNq=TmA}X;4+lQ<{pij!BFRZdm%D6R^}gbMHVVwKj{dFULi#%~&nGr|ZBOW)zWH)=
zp!-mR{5-2m_u{<FTf*PJyS$=(t$=gka+Lrm#I6-zjQ>2RG@iXEJ(~$HRGc0?SEdKS
zU)W!PyQ7odzRZ;G{r_dRv<^q}x}~O-jB#$gPK(Oz)+SEU!G1F2OS!;5i+R#kqrekS
zDv!jo^H^mjzl%&X?@uD;ER0}khbfd$6QmNsg*(-&a5|tb`u1KsjHb)csovq{(cvR>
zGnjr9nigLXe+NWB{P_@C!lC^vX)Xuon2PnpBA_{C{HBSO&-bUZ!<_7jO5yyldmmeE
zP@*`;=@zT|?`zXKsPRSh(pKd8>{H?k0eN5b>yh8A+BgzpKgvO^5)jn*98V@eub4*V
z>ioyjMD%iiHGAPb!8dF}a#Zq*5s(Z9MkSy$U>t%s2*gSHy4EWtuF<LCTq*1mq@yz3
z6?usgV6V_}!qpaKB7Q1kqjk=_EAG0LXKAQF^fPvXH@-1Bl>EAUuiqv<ZA~8PPgWew
zc=IaY$uDl&44Y*9crux!>8J0`P5*vsO$`R2%h?4D1Eg;SwdD%MiJfv&E}zFVJk#oy
zF_w_&3r<yxhZs%K@t;aV7N^3gDL?%yXux8zoO@#9%T^?Zsc)XY6IgQa^Q;ZYVaK)Z
z9c*nUgg0+#mwBGa?NEq`s>Y?(J^r#6bMY}W`C{oTd)Z{?h^M6v3sI=6dA(NO=sTin
zP+Bc;T;~QWu!)@82}Qqcy0J}fIqP{V7b+;1fEt7*YC=+>19k6s*PgJ^fQlXP)#9W1
zs0CGR!$&HG$S>eii!csE{CFr?+u0pG{e<AJj0eTyIbWG-!gtK#z2BuOjYxthY7+2H
zDL!IetDLU_;qHbBqN2LXMa^fJZ)?prz1z<?Znf-PHpl3fcMjcL4Z%ya>zvmJ{<x>_
zT@CJ;+u23?pAP+-p1geQuzgbN7Y;x(V3O$%N2?}A@OHoA?l+^&r3JFjPh~j8pYd;z
zdo&6FiOLl9Ik>DT`Z`4^(ID9)s)Q>gjcVxQ6_p`6EE1+4zeX4B8z3U11L15?8rKq?
z=796TK(I~pH9qkhGKy+|S6WnwA5YPEieT>-3KN5BCVI(*Lh;D+opQ5L=!$D&+e!O>
zy=csoVg~Fln)f6|AYlSuL%B8MF~4f`KGSqZNCUE6Ltum~RgYoRY)+BAlI?CqE3uqd
zgC2j%m(87zKJND(geHWs4~PirE-z~(eIPxY&Abhc`hPkf?ceAKrXFcX)_RZ=YeaLz
z59tB${$fU|90P;EiaonEK-GY+$a?y2or}VfqW3OVib;Va4)>V!l+d!U06*|mwXL&L
zbK6!h0RRBELrKAw{k4z)_OiMITsRbO(dTD)q+rCIrpJ4o)=%G9^Mc9}ST@ClXn<M&
zPZrh8UW`wnpRC$PPL7OB9+xUZBnf0zGVO5@wcs04Z-G0MdT8rp!C10m<igsPfD7~|
z^pH--8}7F@{yeKupd*q*DMVUt638&tz8B4~M|9L)Zv~RoQ&Dd1e3TARd!Ng$auw!2
z@*(RUzPc@JivI6rX}|H+F!e`EMr-t8N!hE6-9*rJ6;m;>HGqTO7D`UuzDiJESTcLa
zv;TKK$RnGC`BB-etea87#?LjjlCBj4MHO_14o0)>G4SSX3=0^s^|Gp^@TrC*Bz-N9
zFX!KqIF{OtLAS0ty11)!9C;TawrT5c&C=G-ie-TBDpegLsT7GqG6N}9A#!p^3p;Np
zr`!D=M6#M%uf_*eD0rI6$Fs3l^<OtfR)jf~lykMcwP9aXkkyPL4gdZ%jyBF5xhAV?
zKFhllSDWVeut-#sv$L)W7tNgo5}jf#>Zh>LS`%qPg38Q2&H*AU&`NC?VI)vK!5&-S
z36l*?V?h`d=X>7tjO;J5j7}m4(>x4g%sly0#G>v&Ta0)#Xf-SajW8UJaSetjA4oqQ
zez(!XiYViS+=d(P{INTI4uZ9vmiQkeG`;71Gu0458fHHWGFZfAvg31{-HjkSa|{-n
zUF_YPX)0Kal4K%V)$^`)Srx=O7i$F5NG9%1v;oDo(APrVQ?qW<)e!d5sFLp2X%V@-
zKw>rcOuJ;7Gc@nL)1w3kbs8sJSP&`$CIjx}4oZJr9*3B2m+(-X($7DibsF{m=qM5<
zZ`|LY<bsX|y|mes69JuwWr9i#y|wN5n(<?t+n}@LE8cB9_(L|bhJlBMR`5|l`=@Ht
z)Lv6*z4gKTngohz&Nox29E-zErA!#*QpP{%;>7@C1QBSmXrwn@pKiS43Zh2Hgf5>7
zNI`l~Xd5xOG1>f#Og14Q`GkG}p7)T<tJZ%v$Bh#g9$$9hqhuNEYdHm+>33oXmB}rd
zsffj2_78Sj7MPA^-U|v>U~j3Xf3C|nMg30Y3X}wRBFlg1ETAk`D+P!|S~QkMYVUpI
z`RBKHO83$7wiY_inqW_S&V0x6GtBN^m(;=V3^2d1ouJ9ydwYBP9Ou!q47u#i*0(39
zOZH@nd!(eNuefAi(q*b`Y+r|cJNc{M(g9SVt>K=#M%mFi;bWQnnrWcB7QF0Sf`odv
z*%UShq?=v+TVLi-@Wb#`<4S0j3%+YqTTjtAaqe>ke9Q(Af(|WMbw;v5#PHpfjtr_$
z5mV`m2$PQS@A!Xmc^evp<<W~e9-W}m)D{RIVULyqqR|6$X36Ev?Z^T168odkiCA#c
z`;7I7LS>dx-*kNz1fi3%?Lw~|rm<UwUc-O-2)`ZB;*B9^nWU*e2O}G`IfF7uHSf9=
z1@95EiM@T|E|+2gq`5|PS;s-?5ehW&eJy9^AmJdf<|L5P3((OkG8D3!R2g*VlvQjy
zI=kEP4#@qr)aLjj_FD4|C%R*|RArW7<WFYwWYb<2ea6hzSnER3Hie$NqL1JBCTEdq
z32G1W=;%4<ug1%QB+uN$?@s%lVoyQfI<1Sn=KD99xr^?3Dj8aHVa>E)9xsKu`|8k8
zu+0eJ<-Par;&w{QQQ|$uWwXg){e;m>6VMg%bS+maLXvb+3qLIM`8UrudCC^$o3qzY
z`5IM@SZ+fm^+h7UkG(rt=<=bo$aq_%=mDt>x>H;YHYMs_=ewg>t9_+V+z8D3=0$Bu
zbA0|=GL|AE+)jT?lpIJLSFCz%+Cl&M;4`$Bmv{d+shY6zva6Ma?|l!ydju{txPx1V
z=dj??SIrIAXz3rOy&c>iUMPo-7GE!VIrKS*Kn7$wow?$Uz2Ew;+cOCZ^tZ~DDo0>?
z4j_u}VdmPQGvYJf)y-1#O`eZ$xj$Nbj@r*O0Dt-T<)STyRYxO4O~vE43ygy2qM4Kg
z63`+_ND7mqD49aiiw(Hw{1%E&j6*Fie6J@x+gFNC3W<jHA^SJv48$||-I+CEUW_21
ztkf=SBeKMV2B#dzm=#&CemF77e=y~{3(f#Ii^whDSMHZY$^6hjXtP*Sn9u%Eou<X_
zAPrks0d{yb>0hYm0*mmp0*c`Bk0k^N3LZ53zfZ%;E=dcssIF<!*cl28n37F?vxs0`
z+kw8`-z;&z|7feM_x=r(*&mM9FyB<YOQJm4aYjm!F$Nu)rW2uiaK~0A2QdoR)0Nco
z!M>0->2dRBo6(AhIjf=+UO~GjaLSdV%nzIF&#Sb|YE4Hn<qgDB_#LGwXem&Ja;$9K
zTHw?8r|Lrm2YY;n!E`I3**znbag9yU$As8s478mqow@Ua!V^%mGXk}?J$e2@g%_fj
zYe<rV2G@kVy-K1*MhX`&w}1WnfKYJcT0D)Q=z5@?1MDbp^2=2B5!2JWMDRO*qknWL
zZw|;RU1zeTy)JEYFqAUSL#!JX>H92?KmYHY>Z#|a|1O~p+724Oz3ZP;QIT8!RzZKT
zOR5U4FvJ^EODdCIlFQ()j(LXbGxYka3&x_-OlBnc?EE##_Xeueu){BVt&@oV1EdC-
zfW-l2=$X0T^6umJoT1DWj;`uh(sB~E4XzHwUY!MIHp{ZTqgDA#Aa8uAB4PwNF44Ou
z&P*>i*b{fr%!a-XKd1lqR_S(NYwOh^&kg^VmVaU&SN6|WVvdUb{mgEw`uuG<*&nQk
z^>^w7mcg~VCpMjLeqVicqE!a*0=<eSP9rW8bLsnvj0csegEd)|muS-8P$|-aw#)ca
zEO#-fEC+E=CYJ(N##=nnUDqO&fJCe`AFHCYpS_K=@M(efop@OL&%15;sO=r1=U{^#
zNum+{h+E)?v3Ig>?%&v>1GHWQCt-@E{9r*aC!t$H*Cb67M#^+kwk^oCRQuVVz6M{X
zc&BT}-ts(-c-#ocl6>wR6Dxi=57L2*mzXJ@48lQaMJAJ%-!r>8ST~lr6hvJ}P#&+=
zqtO%FeV73m=ziZ&FjhhZ6sp@U?wY(0JncOCp$Zyg;YkfW(gX6zEJ}sF28xtvb#br!
zInl*#s;0H}PVbD0+kc@FqQ4IGM|)XYLLLb>^oEP3_bI#ab68p~k`{2Re&XPFdjl9z
zk(d2T5ZuMMBu=<-6A?nrgbTQZ@R+mdZo7>`!rA+V$&+#@DOR4~BjOkmfcd0PP`T+f
z^OC9ITXZbTe+_S5A}Ifz$3>!ORwJC4ig9aQ)n(cHzl`Uf^0_iYRddWa+22z}I?V)Q
z3BZ5^Lf_^vT}4JxkMd~XD2GLUd^lJ;tDiesClmiWZd8voM<#E=9q61Jm?<zk+!G23
zMCP96wLkU@q;gG5Hq$^EI(t<IEDWS$uS^<EcWI!Sw+~Q~2fuDlf~(TnrQ4=`+2P-T
z3kaP-Er&wKq5qPH=CP`NH{Ru+m7x>1^KMmGkLQ>}^xYSWw2il}>MeKUS=*dJ_6`ie
z4SkB!W4^t-QI0H?yNzDP1<s5m)ij@!!k^D>S1!<df$0l;U0y=P9KZuo8yv=`uWO|Q
z>-w|ELOb3=CYrBQ&(G#!b-#6)lhoLOn!4EAGsBgx)Ce5~XHYE~4>bCYEgL4}OCK8w
z%xH1eAd2rHjLbRgBvpw3O#k#u$vF)b9T_PoZ`3g|W!Y>|^{X<hN;uFsAk)&C0A*Ob
zV6>3!C_!w6P|;_J#w7DPwlBxh#}<l`E9aSqdGm7mRF*QCm6GW5-Q!N>)Z%933SjiK
zsQ^KdBBuR1#;+3q=&$)A_L(67wjD(dK|+M8CeR_r1M=HV`z|+hVpuzMC6zt@+h6Vo
zCh}kp{6}A7a=6@gISW}$rXyasmIUW=bBPsH5xJn^M(3%|p~9;1+I8!cv=%u_Dxz~7
zQZIb}MAO-O7_Rr-vyayj5LDwybh=K&ruBrohlAE~E^#yTZ&rGVNg!7MyUza3AIUkr
z7w9#)Z`{lS{1k5~V!li|wG#=eZK_GL8g&R8XLLoAWG&DUxay>tQX{YESORW2ec_>M
zq;G6${bKxlS>+W&2he^DIXt)xUXj6v-z@`$Pc5%bX<sQg4<g59$IU^a${btzaYpQk
z4*oZGB%Zwr8QadsJ5P<@@*a>fn0ag>L3d5X!Eve(=(zngpF)cV|F_d9RHy&hXKpqd
z=&<$4F#;_UWQ}t?^YkRBfWQk(D8)H_%XZxVTLoJ_e7gA6(m^SF3G#IW92_=*{v}h|
zI<K2VTqd~;58l1FG|nl?kCG`6Ke^by@IwEdxD#<*j+3~B%VD2bsa-uq?bXyIh@|QG
z{7nfMs4%Zdi<DH8@N)r&3nH2x>7h5t#7zaHEW8&br`$xd<T7k$QW_#CJA4lQg-K98
zTZd=zQ&MuShQUwUxxidHcs{!R$+3IgXaDLd*?Oa5SCy(R=hl09<SO8Jy9JD9G96(j
znuy@i9Et_G!71gS3j0lgBcNr-)tj(R4AK=IiX8JzgY#)ql(L@05BiL^PLLbTJUu)d
z1X${)GJ=hDU{pVW$4i)DY+h!${LzRqACt(lD9y;?ajAf}!(>U`<>@i0!7HBh*DYkn
zZHOKrBEjqi9e?&UgnRPLG9|5FsU2xn5meG+Fqh6xh&+_o&>33NR)#D#z;il|a)V|J
zLrdrZKqH-*6gJkFqvXoTTFTTw<;8ldu@oAojQm{AMNsxEy{gbr<Z;eJHkCLOuV6j9
zG`X5vZ-0`UQ;QS@T~B;r<!ind6y5<wkkXO_7Z#m#At&@oaEVDWmU_tHp$3W?@u9rA
zvgR|)wt`;mLYZD--~!9ZWBU>G67bB^?t`HN*mkD$1GySWn=h6nB8ekg&i1|>Bun{(
z_}~1xy>QyiWLziw3C@Rs1*<P?^6O4qoW4pvU!y!dA9x!Ivb*=5&n?e9HChE>tpa3y
zV^Wt}jXvbJ@BfLHE<+v$?!8{=A$2h@c(+YZ>#&Nme{p%I;)zO!DW3|)E!7?S$+oyS
z;_&^G)O8&F`6@SaJ*HO>mr#3I2MR|CQoP*0zBZXB5bBpqyFUPycB^eo9KSf7(>{ft
zK>DEiOK(0$|0I!3VF>ic{Og?`<=>z-Xrq+Cj+gyE$&^9@a@dO_g6IjGGjTmN1N5!P
zQ20>QXA{iYhizEe0_yKh-Anz<VOnB{guXsR0K29CGG+qm<XjO{*sJiq5M)8C#fEXu
zbwt*UyKIUxDkCDhA~0@=7PEiEz;L#R+xFXZixU9{rC;XMxy!M_?m*=l#MHOCx>?ol
zqw0EATKKUdV#8g2(1ZfH%~>RdQ|0$fe8tE(A1TOE^w}Enp0cwaLBxq3RGLF#kd8$g
zhiwzLRLK{w>#G#eYdg_~N+BVX5xnx~Nc=wJlIWf|^7$+2wn);adL`iCDGNrzhU%+?
z*LQr9!g+9<h~E_1TdQ29lolNJXG#`wQUSd@s)SXA5!~vs$|->vyi3!Kz5osFIl0G0
zOe>}4iSkxS%%7zf0mg;nLPkBbvsb|nunm6wD~ZP(ipkC`At@3=ktr_wT8f`*_I_z@
z>Ko}O(&HE5vjmQPaBCmH%=7G4LO6IFh^a65=tX82RkSc&8}h5v{km={hMP*s+pLoh
zEF|hjOu)W4o&LPgv`rej>S6C1`SJ71GN?|vt6VOE-ztgt0h=TOZtB5)s`kuq?YreV
zfV|As|ASG&NJeS^9JTJd@oRQl>?(Bom8?@pg_6(jFFuQQ^?NOc@01BzTW-#7B#(}c
zu7)384<}zA8J`CmFP@*L&|hNcFW*Tm+kJ1I;5l&j=zS}0=DcV=!>Z=I`=oJJ%6&`C
zrys|ua-wpdkiMMK68uMH;j3+#Tx8G7h$D?KiF?6-41EwbmeI#0IPYI*NJhqShYh(X
zo9_j>Qg@=Czaz!SHvQtQpK}34n2sxnAP6Pbj?WnMyoRIv?xqS5vb?Jf8xO$>8ymFZ
z!@u6+e)#2Qo&U~Zz4&vAGK$R7p3A|nP60vQU4wzSS~#qJ0R??l1hZRWvm&nDJ$lwt
zpD9BUuEX1-?`%l@_=K`b5~Sdw2&ssLh#G-ROJtW-j>RY^Er=T~Epc2iX^WgOGp!j$
za6i-Wbgk{&g?y4ClyfRk*dYqC6z+qp_tI=^aDD!-q22VwwAHkTX4|z{Ob#jU0}=oZ
z=?fK!qn2eqIRTL1Z^nQbmz$Vi!awLL_RnL#L}V$y?ioR>)qweCxa(4DG$2H*O8X7;
zUM1l&$?HO3a>x=>j1<A9+8r`I=>(LDA%*|Vd%gVoa~j2;uN3lm1ra|U!XctXHd_*y
z36tjBLt5R&*)9CFVdwT2y0YD7%UwZE9}D^<YP8J0BknnmLIVw$lz_xRDc)S7MKIsC
zc!$WK@Fl2)C*F0n%=n#nUePrPbcl`2*$ef4-TGq|G!U>P|3<5M&>a5kq39ZWzp;CA
zfmS92e|pi7ZKLk1)Y7bUlCDQ*)NCyLh@3N|E$KU7%l@p8=sq*OL@=aJVHG@0sNA~o
z_sTQGUhPT4%i8=S;g&TfKuz?We6!Epk2bi~{8gasr+IJTz58|7r!PdZ-=gJ<ke|x<
zs~I>>zcY<O^N==)#CKI6v)(sMij*0SYFAXiYOtqE(@&Ia)@gPs%}Jw8JT^f<d8{&3
zg9?qz0%5%utiXVdhN}>i*hRbtCZ1b4<ZDtaae{hnLS&_A`GHZ#wxF>i@>vE85=bKK
zu3j9K7|?}7`f}{81L%u;SE+Ym40q4uq-*TAKMSc`i4^R0LbZvxq1qTXa=o}uo-#@#
z-C^bj7RDmK>{QoQn+BW*C=BbY6<%+Shg3LThbRuh6u{IS&-kF*6r^|~E&SRbVIcIU
znhM1gA4FNS8lX5dOe}lLu69sg8TZpWvzT_hj-Qle&o0f3ap!GV<B6)EngXsM^x2B$
zeqKChLL8-hV#WztHBB){vlI46X-HmACK+}0`jcmfTCd1`z~l%juWgOF99sl5Zpd=z
zo^~E*#$rq521Z}ozXn6>`sqm2Q6%C)(d`B(;NYUaE4B`Y50j@R*kSs-Z>F%E$F}G~
zUP98}=w5^A|1CT8uI~DiM-AHFO5p&q!)R1rY!*B)lX@A)0y=U%c8+%D%tqU&MqOlF
z^>D44&%#;$oe+AvT1tkL?tfSR?090L)M&f^_cT`cBi&P@3gmABGoIy=Xi;;ysTy<Y
zua^OnZ-63o_G#5fl_;O?OE`-T4U(<Op0>`kAF4vt`1tG|cac8JynjWzf0MgxNJl{d
zL3`Wi@$bP{*j?4T2+h|eZ&q6=_fGUm^{&RG&IbPvQx_HzI-k7xYkaip(Q?`M@8W1k
zYUs@W=B&ED{^G>1W9FT2x$&}#M-13E$RC;U_zipM*{9Ul<|q;%<32a|?>M@@oO-G)
z7Ztj9W{9?7y(7H3K0+jepaRk-&WHr%K7@*4!CsKcH5H&4xM*RvN$IZyjDFEnO?v<W
zE;DWLd&T|C;;MeUp&>XQ0<akWA-4q$J$QqzBJR>z^)CI{p02598Cd`l&8qk9*Kc1|
zQ=gLN=GNZ&Nd$)HO*P-Rd@V@8Ybc6MPadEwEqhBLNX_%`Xja8ECm>agu8?t9CM^sY
zn5=P|SXW9Uj%HBp=>Dl5L6#|UBBXQ=&FbRj(6QYJFEouR;gF~?SPMT9o3rk>SYR2H
z6|-Mw>6gxVR5n9XG{b^^gGxgWBInHO?fD7eyC-yQvM&-Z)f6(;0si?LJ-B<P>0=L|
z$BH6JPjwd7k}@BUF}#EasaZLYRlXL+H_LH7T1lY;N(&560jIXgf#ckh@K1xE72FY#
z(g=k4D2z=ZsGXESk$>4h_9nPpkBL3vOl{l*WAAe5*#E^?6rk&$h3tN@sP5=88Od7Q
zu8?fe$V2EjF=O(`3#cQA{1QiaaCr6ZdAx&zpo(&t<c4SRmovV}Qa8<TiLXW*k!SZ}
zH?G!cUp1Ze(>2P(ywcsaedzn&9yn2Rk*dD1E>3T_`YWe>%RM*{6PnWlowW1Yct@!i
zu;=5Ny5O@C`~5>#QH2{o8E-bdpDq)gg(`#@j!b^VYpN2y^xYG{7a6J9{Hf}<9Ams_
zUL-xO(USG%y07BDS!?g6)9sW^ay7TIBrP|^W{Q)64?pOhQ?0d&6Q)-QPUZAht$PIa
zLf2fWK`CQR#8m4wQwMIYIAL7GpkH1-qcYMkPm@389$(Q$uS#3cuJ)q-vPi+UY@%E6
zC#oNRodN<DKV<z$?12DI(K;)E;>Sbb2>jfPx)(uV&-wqnq;#x6f~>O<?eH8*fkULL
z+~;BTsu!(v;*1{N0<-L>irngvN&v2k%h1E&TiR|!a*sDD5jjZC%?K6l+tfJg3m%%}
z;cE@paQD=z3P%_r*qSvRFYpE5d@^D6>u_Wxa~*pDiOGZI<#ST~O3!&ntg5-$cze!T
zVO&Eb4WGQIq}R3d9y*s~V|l28gWxOoSe4s@LdoU9o)NN_CSfa4Mla1hf1BSiM1WJa
zwu^}@SA=TYxgt7$LJkecCXG_~2v-71iX%K;#DnmZ-UjKS`_tdI%_A8Uy-?mA^TTXC
zE&vEEP7g@7+y-NMBq)_VDID0UC&^=Y$!g|L5|65%Vi{P{F6yw!ll7pflsXmtrm^JO
z*k8UIwXu8ZWoOxOCRN)zeZhINy&<IcR{uY^MnJfssaa}m85xXC>)ddNVNWH{j5pcJ
z{lM>TX0EtglJIOEGqSk86g4z2X5qbgxaoYiZx|HBs@uUlef>FKqcizQg;&2-JmJ$F
zJ;)=1C)8CMY~0xp0|s5c%Xi6LliL!_^9&JkGNv@LvwLEz@g2y4)n#Oox{>H7%~8*r
z=?5T~0hqCS%J62lML}hv65NqF#*$Gpe*eqO(4N<u-ODdc^YJ*tFjf26x*y@BY|T5_
z3oZWHk(4|nxIg}n2sCb*+4#Gp4W;KI;Tj;}B97dJfdgD6^?+A9mCKt{(z9X!X?RZa
zWhXlIFFuP5%5pP)gI8B5Y>enp&vaTr+<Kd&sdbzCum{tFPRX;6Pja;%^-aEE8GIIo
zy@K^|Y;o>M&8()*$W-h^gCNC)(_x(LKm6IAZrQZ21pXa{s#88;s%sElcAtEVv{T;D
zE&`5Dn6lWLJIOrrPlh?)Mm2@)YEvmxNfg<X8t7MO3LuQ;DcHU@b9FA!Fco;`ip&kP
z1z{)!+?mPbkj+SEWQA-Qh~nW8jwzfKb9KtaeUl^#2#7;xa-JSgTy4f1se!<+eb3oT
zvRPV&j@rI;`_?T**_YZ%?K*ndT$oXI@*eZd50ur@XJMqbKOj)ye{v#8S)n>muwByJ
zKR;Mo<Fs7oq6+TwDBIg_n~1%RyQPok)H?#n*W7RXn0cT2Al{%jU7KJn!H|j!y!^w_
ziY_@Vy)FvXAfivdm{)NGkra`mC4_CeFi>n4(*9c=7ETZ6ldZJ0+;lvfTJfaN;uD;&
z*UW=gW3$UP-TT|LclKVlEzzJv72`<?ViTNr2~uz|RKDL-Vj*BW?)QnKILi$R)v=lV
zQ~%%foR^#oK4$-=8>i0ruJ?&4K0wKs0#MX`Xen?*>~(3!kl(4~O*_4KJJ0vZu53@+
zaPW%jIpFgg0bI7>zprh@A>okyVcYuq8LZ`1t&}^OXLeiQlvK^5_v*@W&vEDa-yIXx
zfN?ao088Wl&$q}k|Ak5uVp7=o_W7RijOD6)am4!<F$D6U<O)cX^6$IjD#}xZ3j#rW
zwFW635T|HDK;Qk_XK!wWJ$oWUo>Me{bOw8~RHP!Sr;lQq3k)9&8B-y3Y{2J|v1&_f
zNYQ&Qgf&5#Wadt2pkf$33-G;64|m+g`S6Ajlunv#(P%PtmMIgPKEq!QD#>2P;(R~|
z!HDFh65^p<IxIH9`rL{F;CB7?k*-YyfZWbD&Ct_2JV|o+r8;Yl2=*Bfiz2JL3otSn
zf&4L8A4r-FE>(+-XzkW&hrD9y;TJ;%CZVHCIAa58S&msHg$WVqv;_|%a>qjkl-hIm
zf`Sut*Fne+P(;8J^wO&}Ax^E&)9!DFv>q0NKFgRd<5YFaw=&80qUT|)!{k{84$g&e
zi?@r;D2D(@xFtG=KvpL&A2VMQUL93#76SsfA&OKH_gl|3J9|WFr3Oc|OBAjn&af|Y
zh<$8p2m5B2jpnGGOfkJvF(5*$Xy~gV|I4A*Z}pE0r3t-uA_D&g2NUc`jxU>#IKv)`
zrxlh~3Ln=w@uF%`ntZO!mD83b7NeQC{Sh)@oQgGN_u$67k5TycbOZm%8IFCzYb{dC
zo0T6GUF{A$bs-OQ@b-+e+@l+-pn^~r6F?Nz^70S1XrJq(GvU??y8^lDhI+GEQu4P~
z;1)74sBkjPk)GD(pymwJ>*z9`Gb;gC36K=w{<MuQX!!BAkrWI;^vxne*JjB{Ap;{F
z8(`VZU|0{YrBTmLFS0Fex|zND|7;*#k*=<;*KYlWpZe`{&{~SAsX5m1>7XxpTVNeW
zzIqH)JZkK=^yTrbUO@OY$gnE@R1c!SpVjEyKR8TjP!chC(oRh%ND5DyrmO1@kllIz
znn01jdE|TQIfm-ef|^-)y~7|iBO~OQ!l`kcK<lu%82KGU+{Ie*DS)~hfI1{EnF4?>
z+5YQK750TC-=QzkG)C>tp7^c&ejF*70JL5J-!Yh>2NCbbg}23-v(e=T^cDDbr=-VC
z>}o%hD$I(d9Udf~r3OFNjL%URhEhv>)h{x9%yzY+*S*{QDh{&#mX*Gq-LKfYI+k{h
z<5KueuE>vJZM7OZwOYGFgJt)jrO7W(UyeLCDE@7g77!$xg{t9VwOlXDFebCf0e{J9
ztC{!aXpO6m4eQHzr$%wJRXJ@UVr4jfb<`0d`5+HiFoH3ZE_VKHN9i&Z^if+w9d#uL
ziaZ2$>evyBP|S0%uW1de%Qnj9?YpNnNRzSAsYu(?hwt7F8%ZoF*(?Q4$?p1FIov;?
zrRWeSp%jL+18lt_?l>xl7cHUoL|XESW1i*(oeA4+=5nU1oj|M$LP6(S<#ZuLa&!-_
z*in6yiV~2pc+db<SEGfEA#o3p)IGD*e)e+1e|>__FOg4|&yo3uf{(3g)BR>o|66*4
z-O*q%?#*OQ9?K-TPHm+GvIhUU`L%GtADuS3M5pqS@)&l1l*1u|ZOym``!>HgbB?F@
z9LHFuTuP}@B0<a%?A&I?-=sL#xZxLx)Hesz!CAE5lcAutN^>i@VZ|bgw?N5Kz)@Tp
zl%Wi%ue4RUPPxXkmhHq~f6FN94x67J*AlT&gvLj3bk?!RP>IZ6(I%kq4axofbm?rH
zdkyOMSx`Xw15y+Q0KDuXg3qqU>C1-Gk&<fgJ~shoS#OPeOMsHWj-o(K!KOUX98sV4
ze|kV3VM=T&iYlrpMMP20I;`Ciy#fB^0W+ub&znf%HDAjaB8%m$X_|L1J5U<H4@A=f
z$5v=pA^orHB5=N3;)FzYRVN~+%00C^A6Xga^xl_qYTG7RGq@kqpnI>)LX+J?HY1n<
zKA-{{3XXob93>J5D?Y0B+5tj{P4UM(c99hqb7qt>ann|MKF2wC?^VYFR5PN)*)*rO
zos0jw-U|m3etf0}^4ycBU7!K2QN7mFvvv?e>TKuGJGVle-8{0gmid_N4GT9E;Y65R
zUD0d%u;ccznn$O>V;vCzW7AbRI*Q9KkZU+S(}k@r9ndir0mTNd7X3hF0eyX4D}AHH
z(3hsVP^P{N<Inp&lMlfyLH}JueIa~UFj$#O@W6!v0-~b+O%5#-A#H$SM&O8<(=9zA
zlYhwJ<f7xP88^4i`Qo&Y)Xl}^s#j-8J;B_Ise%j<Y^?Jr(e6V7wXaGlcRqC1f0Z(|
z+;rxBP+zIByy}y*Q11>#0@5PUJGz^Xh4m@UVhq7#*7{RBt7U<*(oSk!*XWlV+#F1Z
zUbf|f<JTPIGN|vUBOnV?YG^~w{RgYr5*tYV8-05AG5Gn`uOEqjn!Ak87p;wEY2r8b
zKUBLI%+R{5nS_v9N;LmEzeD-nc<(<cgVa^KRKM{Bmw&Lx$BW@mN!->VscY+hoeSr*
zSK}>zbr&LH&Y9h;Bid3g0J9brLR%W8eNOA(QDhZia?406gZ~tX-exTtO7e1Y>Bf-q
z1ODC3JBX6yu6gy#viHTT)}0m$D^|Zvi*YhUPnpp+6!p02&qKHIB!N;|vy78)eFE5Z
z$R&^WCDbc~JprTcU@QfM3kPwLNBxtwf2-fmJP+lWCy#9S$}PEP>30SiT|kU}czl!^
za5V0w6sZsK5n=-mk325xOaoZ^s5rHWqu1{KSUFlr+?5sKKp-qiEDm-P?1&84|6%%$
z(wRey6w=nM2%rdJ3L8X~0n%aTsD9>-xgxS+_&!I3O4bI>`_k?Vn;D7vc>wHFi9Alf
zlMF~u_GB6%g~cv>9{Y!)beAwCrdpp;Lx7f2t7=Y3xL<tZu|N7f7!(+;9*_F@EVC9p
zFlPBHwjB&DTdH^Yb4}m+92IyyePhRqFDvcfu*yeB(gs2O#@8(=;L3#;zg;z#RaBcM
zW0;GmAhArHdgd&xNvD371X<6RsC}9DKE7n%yhx=$!%DAI+Jk11ZJF_x2Qy5uPR`Eo
zwxPW85#DE6smC8usu(5E9*qyaGUA?;SG^n`1#=D9(*afn=L6z4@Yqc%kP^32Tbcxy
z%^N!oplw2Oxqlo<_NPh}$F=LnGY=;JO;P*(sg=o+99utm&gkjE%vc#Snh_a^$f|b(
z5XJpvBH+B6>|t$Wekk|Y@Y4s<$2b2@g<k}jzz~(B3%w(I{*5M!@**#LVUIXYOH0No
zYo(*f<1CpJ(Ccl3pf9dSj2z%eAh`lV4-ULjVAR-AE}_*nKd>ik2Ccq`x4?~~VtGor
z4Mnpu_kmvSydK9YLyKk$EukQ1wWLLzNz*)3^#obv1%?tM_y{lad)%vWZR)zXR<+yK
zoG^DmWvupH*&Ae0R7Gi>CQzk0wxzt8<CuPoq$}fsDZk5!48kN2dq}t<mPcyEx~jv<
z%|dnE<pG1e7BVmuc^>mTG$IH5i>Bo$K3xnSB9(mrKfY$M(SmMBS}4YF_QmJVbgDk)
zddOPSNVP&-@MHiiSDl~BL14{_gNz_bD;=)9`>PBw3UaJP21L{D6DE;{AN)OlEqi=y
zgiff=dlU|o)boQx@fy-DQ8lWgf`P}zoTvVi=`4PWCB(D5>)<>CqDi9`LBV-X@19Y-
zBeJ63dCuGc$j9*O8-OcFPSe4P;OziOeWH_lP6y!|(u!Smk+G7>pS&R7Q|K29@JxQ$
z`FQ@S#Vz5ao@C*oo5}hCU}w&o*>qFx;%9qazsIKiFqoE20u>UvuS;jBu-VTCu>Y=h
zdDlQ<WE!XKv#%AJf-s{6+M|7?q`34-;nS&<YCc$S4rOj1Qr)e<6nqMS&SA!7-z>)w
z0SPyVuWOb<c|i9?$>Cl2q!Tnkmn-ityB_$y0<C&LP`&5p19+L$-hYW{1(iQ5L92w%
z&LPgFS&Vv<`}-WuE04YD#lJtIc-4AbK16^K2q@v_JuE(2g4U#LDVJoAOdE`bfZk_X
zP<K=#xHu8l9zOBrPQy*0fkV%Iks%kqt1kZx|EriAK25UQNa&TB)T!CkpO))yExQY=
zGrrfRzlx0xr0ssCSz#<ifq)IPJN;nhax3)fpPynI;J1Gw3c5Sc+Z34*FB8pywL0eW
zgq5!975kFf<dV^XTpUEL<`14oEbtE=8o0_xdT-ZJ-Hr@B6wb8eH5R;B{{A>V@#skm
z$YUn^*M7o1hvB)uPaGLvQ6+o|_^1*}m&>UoOp=&2+NEeiB5ZbHOO%~tCraX{#D>KX
ziweIAq7X6qLT4C~@+GyqN};SiM`43|0vBNfH~;#V`x3O+sxaJ<d#qnul@p&YI;Iho
z;)<Yn(ist++YPafh*DuiC8}n!P5#dKtxZ*TTRiU;ehSytl$_~ZUU%u{I#J|eL}ofX
zlYmh04mG25%cR>Bb%SR(!+7?eM7>Nd%@IL!y+tmg`K;me;Qk~xAgcmv|G`+4@F{DQ
z#K?!YQ_{4_`5a$%)f7fEN=r<ONsKZW6-fvYfjV*ut7A~+C`L7SoHKW))Dd~$ThN~x
zf~*09bwr_3O&^69+Uau2bVtsW?KY{DJX}`E8)H#5OZG`uRWlaF=bm<#Pc!}l#Plr=
z7XFPGEN%U|xRuAsXO{-k>5eOoN<~N$8h-oO1Ms6g7)Yk@wDwj=$@jTR@|E<d7|6`(
zd#4PoEvxF74(1Fsi`j?_Jp&D<hh01aN28}H)kox@1sIT53_a)YJ%cs>e>=Y5^fuRV
zP!#(A_I|qlZC*qcZ653V6{uJ*C-%$c&C1;3C~g`ToKnXv!rigw?F*|nsoS((xL0{M
z>!`1xMo#Prn8=TYMK+Zhe0;9dawPW{6s_!gdcsK$cJN6|o4P?ub!O%xHj+oDNezlc
zt@7i~qQP`TC>Xh7-CFH&3GNS=oQdT<Bglh?;k*_pX)073=3YwG=DEH64V-_&NJuD?
zIin07zXrA=DK*(**@O`S3looJ7Z@Q2Z5z%gv}7|vYtfp-93)zm6;?*8G~MS>Q`B-=
z2zY&Fl?3~k2`04AAi6j3*A;r*5go<sFUrvJ*hgXM5heV_d?mly%$dkD+rP)XR^4i3
zY}b>``&_McYvi&zY*$4^wiNwxT)PrRuJ=JgW^*Vb79@7Lkt-)@Y6rj4v3Y3l)r^PY
zT~8jh3wdVOWEx-uKC()Blg7ys6Uu~F^ww$0lNG^~Xa?fd{Aotdn1I9@D1cPc@CkoP
z4>0xJEihe`&sTsbMvxYsYep56!IWv3h2?RG5Lk7s^<FWcrOwwGJ&&362L2RK%tt!_
zf-^?B(|s<Pm8Ek;_dPx9JLjh?H#-epe+IS(g_`Y4(k7g^LM0LaeU_HHpgcMzT338G
z*6-&<UHT@AKM#GFmAyjfaZk47T=cqiT6Fh`H<-+UX>V1IDng~qe36X<Ww@?g)LV1Q
z<!0K$VX=$)rq6Ou)7AuhkS_KXJmN!jX#g%Ek7KT<GQZ9xUAHej+32a++x67w7NkmI
zML;-4&pM^)OeIujdNDQqIdB41!jibW&7Ointa*Va4_Fz?<0I%To4yKk;xFGcNHaJX
zeOz%|=_H2sE%&DiP<6h!Z1robkcuKa@10MPz++S4CA*C#dr}?5b!kQ?<)yI*en#q9
zR#oErspovkZjsT21E<3)7P?amYW+zf;>J0L7Mu<g_d8y`l=8Q8=pV-4(#B4m0#!Ls
ziP7Sbf^J%%``)E*1Y|7MTE7s^XRZ5R)>u*qHoBc7yUF>5qOORPF*XvJRWI-fPSGfC
z8V>}EGPAslea;j%tSESUbkM7B`H`VkvFo|)HG=D96jjg>59WKIg<Yxph)xe3_xRyY
zq7U92sRUnhw6yLTYfPHQ!(<XrOH@hb6EqYL%YW`OW0DgrHEOZVBT$8dl|=TQP4U|R
z?dotE3|`|*rm&d`F|qg)CGnv1>B~gW4gwx8Xezqx#7#1u?i>0~CBL+<q2B<DEHN>O
zCWya84&>^M0y(_5Z)mbmSCDnD^FSN)svtVP6N#u33VaQ%Y~j$EB&Sv+nUxZbs2_7n
zIp^D;D3!qIywWib;03VKn!85?Ww^a*mP~a%ey5(ICB`10bZMwKSfdxjZx78~Evd^%
zV5JH9%5^W)EE~{!vs<FLP7x)FAOb-7jf_s35D17WnxJjkKBkNLB6xnng*PjaL1HNi
zEJ?f2#qipD;En8>NQ3@V@!L)H_L*sqv^_OcL2whCMy-WllXo9>S7oS;uY%{a(B#|=
zG&s4!gkFF9>HL?j*}T?ILAucTtWU79&#q*89=MY%w}~GY|J|nl-WrVMZ1_8I>(fG0
za~xr-`hp13e}^;(IJI)cqp9i{|7-r0W}XJYbwbdQOq|1jH%brEI2rq5H%Jt6Me<%0
z(5d+TRa41Yqe@eWN6Ch78V%{*t+H8_0~=gqAA);u^0ke=?L!0i(5Ejm1P-K4@5G_x
zje<x!=)@m?SA1XE+YvsT_e@*W1|3lW+!Y!Fg22EFXH<64p=A-mTlV_IfwSrZBivHO
zb@dvY&~r$(N{+^BYy_hBMY-c5MkhWFfKL2_%Ekxtc$L@#9LySYk@F+qR#WjX77&!6
zVq&foqLh~p>eUiWesf+j#|(VvavoUSktTlwSEyErnkr(9fOuA^K}>GExMMY$d8FeZ
zz;sR1kw}zu(w1N#r2U~?ECs}ZQT)EgK>2|Z^HwB9?%YWUJNo*|be(nEQa6v!2}t39
z*SZ0-fDGHRpTB&(bl#M1MLNw%{0CI`D#vlg`*x@9-Kju{BKQcOHs2Fa**S+3r}GWg
zi|!o8FV+{c=47Y3!?t{IBBq|5xq=<w10|4ngJ0{NfVSXO2lit@4&cUp?-|K9=DY0N
zuMf!SC7M$Zi9vnXou;ps`1)&g7t`+CE-T+QdN;t1CM@%qZ33JD_PX5@2}$Ri_TBEg
zitC43(L8h80H*;;CGc*UrgKXDU8+8Q68Af+EaksV)7gw+BjF3ZMy)owl=aIWb-mKm
zwv6#gD;=UN$vJtnoauGX<Qd?B+qKP`YXnzqi~0d0@R1S=u^yafe*t&-*XWV`?tdZo
zf3$e9U#^vX%>4U-mi#NL3opI~DjX;ic8?A2ecUgXY|`(F462^^NxtiTH@4oUbSLIc
zp=tlSv32nNrk$+WlnL7ryc&By(5y_wLI-s07i|(wFR>4oqpYdOB}&DBQp2?r8PzmS
znb{bs`x|*<L$e?3eaHTZs^VA%+r)O01|*zbuN-t~@z_UGc|R`X)ch9B)X$JZxgJk<
z@InCgfsa=(m3JWZ9n;^Z<c5_HWqhROF@h+#AWd#utB3z~kWhqK5>wAq2-Ay^V3+~=
zlW?R$RYqC?STao0nAWr~w!|}cV$wsAbo(DqSj$w?n`x__JBm`_$tud1eRcu(-`oT}
zRN>(hr_s}us2T;bZ%P5KEfc&4D;d|AP$oi)Jo+La{aX^r7eB%N)11q<v8GB>s{RZY
zKtO`3)OA1EZYB~P7(P+GW2$<86p!s39Q~#}?Z`mAR!jTSx-oJl7kE;o;N99OEI%w&
zUU5U^Rq;*}JIZ^p`nfAcFc?7L1_s5;ow+HsuEDSY(Ufz#kJoL=oW-YpDY42X*8FqU
z?+e)>ik)~o(EUu1Hrx3xD312<K=;(^uWkBujyhIHuUOyHv{!~pP@CFsss5SYmD$F{
z#*p1X`NL<2SJ$_}lHdotbFS14_zQ=_IA3fZIt8ZqA4VzsyTLxpK$crO!N@C*A7<Gd
zCD8k4Ztd~hdU<z3p8|Y&4)3U!w-nsk6#nub>o+al^}3-h`ITHUN0`7ByXkTlmitX>
zH)(&vNeFf=Nv!M{A$|@s;R;CoyUAwgQHZP{jK42^p(*W8A@BhAZ#%ciiCS#@hF8uf
zQoOab2C8ucQFLiRZtv(FmT5R!8em<InW(+p96FJ6t5(ZLy@O%J_x0;@?pxq$D#b!n
zns8%9#<^W`?KF9)sUrH{m4SlPx+<x)-}6DdkIM)K;y-fsVO`WHs2<_}h<9NQV0xbf
zp>AMhWE_FM&F8?XmCN>MynZ`fV|zz6E)1@zA!f>i1Cdi|feSgUq`*jL2L+DEDM~36
zS%0R_mAv@v%E?N?3-se^{e!KhE}JW%$UTuL0ij<)1It5&-MQjk-}XSigwGKc<od1+
z#<j!3M$9p#t8yIV@SrVQRn#gW?dbe4a~PcBH$~BUFVS-hB{yT-dcD=XZZHl8hU)yT
zvCYi0ZUvtdFyRX-o08mjU;Ue*+Grlj+~(M&)(s9SV)0Nb2`GJ%0;#`ScWVvE#<xal
zNqRO9$S4bboJ&yyTI^tO9_1_U_ajs>-1`Lsl0ZoTRfshYFH-+j=L=i;wG;(JXhN6L
zeG@gcOdx{ohs(2yr~M#UlVQNiwV=10IrPLme9y--&i{Dk8z?sIyy|RDA-e8<rFInT
zEwfI*_V*R{Ajkr+r;EC<G@e3!H)ZZ2i4=@HB1HY0<H7?(=mtd@dsPq<eu?y>NbWP8
z1ycuz74;W;!pI)*P&eg%`#7u~l|Q`xQAO$xA$T$Ac7b=ZNWZ_O?w@A#<rUiZB4FtH
z^6dXi%1txrd?~>|VuP$VMZV<}mx+l~{*QwIQEMIz(zRcOkEeDTal_fg(7!2?Da!g0
z{XCa~cRdA{umg-dwh!v0_A6ae8*=yw1%N68wR>V%S|~p^ni7KCs<SBt)wDiiB@{Aj
ztAuCwv{f)I>s?^0iP{`uBDyQ-4XQM9a=7?o`3^}~Uj~rMrj+(W6pE>ku9`yRZLfXg
zvm#MD;9j92P<tehf)&w}PAlxd^4A2(Hhy+-&rul=+Bf?^-&2(V3xy2~3l>bd#68{2
z*Vb#I16w5m&Y~n2$HgdL97>956mezNv7u1Uw2jRd_4xSYf&6XYz%L;v^=m8z?y$KZ
zD|%s+Yzp;M#69M)K5%Ef5utBStRRgqq;^4>LT-nr(qn`VMKeP3z0}4=BOirD*ja6Y
z)W5(S0#AGFPleH_JnQbk;U6*l%Mca?=iZPuV3-ENzM;9TBcKIjQ(&j&7lh{_!ztzF
zMoY-(V0=o+j=4dBnypijfXkA=%rhu1sX6GR$Pj7k<;ONwzh6;&0rEB-%{U7=o7I(>
zx%|F<{b<<f*KhV#aNez9KwQ1%d|dZ*>d8BzxmXvlbph<OLC5kp+h;*Lifz^wm+#78
z)cmILpNL{_%F4M+d<9U$l;1+IePYW)&EA>W7yZvIjQZKWJmmv2(2_l`uw;);;6tAu
zUO9A5y4gw)4r}R!7kv)L`+e4AmUNb3@8VT|8rs})6Z5}<1eb5`W_Je&R8v(NKNEIU
zLE^{^pTU!Q2*iKPWwozscw(O!TGVPT1IoP9AUvLNm@5KeE9t~J<)IlbDf0P2=KJrM
zDUkP(;t>%<TM#w-NNO<U3yrpFG{a+0QoVp2S}D(Png*5jQF;?a0KBGr(HHrzaB|A(
zSO0QKIIRF7{OUC>fj!+|%02wA+<84!DNOlo>W$ihFGzyZ%DMWJTn7hr6%I236<fQb
zop6@oqvN9(6c5_Ty41TLfwWQbM@wqhd}J8@Ae5|*U5_NQq^d-#8fL4?nCuKhjG#e?
zL5bQ6X<21s^`AG7XF_CzBI^RYC5Wr$QG*wC)zBj0qv_w?D(zt-P^R7=8s@HYKYp^<
zs`5Vfc*T80o)BYXWKwSG(kQ$#J_kq)ig)Ud|CUUPA!Ho)06{VN&=U9;oPk`laLnVb
zNP|s8`_=ALJE4}V>XJ$k$X650U|72-|G-Wrld}jV-%AK72j5MlqTK&Ljn02$Y!TM4
zk#>c)>_C4F@=Z4OP7iW}9K`s#F0lNBh)5Sg#4>Cpb;b|~g`AKTK;TI>8uW*elA<(#
z(W-Pu9S7u$$hY4UwOnSQYG?~b4*&r<5S`@{G6k~s0U|EA^at%F9Q+(ql`Tt5tAz1E
zZZ{S=hcuwmX<j>IN=BDTw1}?$4^L+u74;WxeF;%o8U)FqN4i0g?i@xyN?H)<7Le{P
zVW=M<Ln#PIcXzjRcY}b`d%W*mcP;<X1uTb|?>SHG{n>qsp|$0$NsX9lwtQEc5obUd
zW8e(8XtF*XB2obknuo*5S3v0CoIk)&+{XF^80CSbYYy!z=(RXrP6Ef%1{e!_4K6=f
zsy(Tizclh_WPH4gf7^n8Jit4Aod0m1W#Qm(a{2AT=Unc~`TsZewX2^uyO$0#6uV%k
zN*CvxsW87wL9`8x5Y5*@UVg`7B_k~sYnpmr==;i3!=DoCEQAaRM4n_!UiU-XAleA@
zVwYKI4E^udJM|4qPP>|p%S!}42|VG@OkT^UAdy<@jl$rxjdovqKcH!_ZQip2!ngPa
z!INs#?ARkYyL@gE1T;K{Ld$U*Y@%pi5Bb`{0$sjYV4=5E2C$Yx9Kggh+ak2XYRPmv
zP+C`3Ry97f;}pm;2E2mekvzDBZPNP~jFNcDol^y=&ws@Gm4rv97P3uzf|Em>jk$M(
zQO#-pgcAZdKXHBmP-ZH0kHJ?MU+$xw@h4m$Rq{>j#W3An?AjJP7}Pq4%n~h3z8e>%
zTYY8KtMe`O#O&eH9%cP#qA`yN`y(f5pK0MAy#%#_YN<eUW#h6ftP@`*sm|uu6#4HT
z$=O`Ksw2u0>JN*@(bFzSdCsU}yxci|s4AR}!4m(TOyE<1Yv(r$><@=&;$+w)fc!=7
zNw5$WLzV{=RGY;ZAw=9fx&;X6lrmTlf-%7TNn~zV3?=5>959UMTd&HXYSSs0Xmbxc
z*-{OJo<%ZG%V;0i+I0=JY$RyodF)CjWK+g%2~#==7{)qzR4OivU8gQJ4#xOgL9;y9
zLMt+OUNQgo{o}-^C~Y@cGhsfPC_9nab&$Q)9-)WIecq!O_2p^-|HRJdw^Uqb0av7n
z&xz7!ugVbB_IRCwq^q)+%#8`Js7d6V%hHyS$3K!nR-I={M+6PAR|PHD&VAF3zGjLj
zA+IeCXK8VyRpdq5b}V*S?sk6j{lN+bEvmoR(^D!^+0*db7#!jg?vD()!_g(u|5;|;
zAdyrG$>UC5Z94n()$A6nG$h;c3vmy$VuU0T*zjbB+`uBnx?_!W0wgOoe*z*88)@9S
zqRhi>zOC9B_#cd<E1`_0LJ&*-ED_SXql%it2c&tx3<XA`92{`n@2Pw?8DeBX3yVEl
zhQljBM|lWsQI~$x9|3Rxo$8&c=95}7?#S&C8HFy$voLExWp#M%)tO~aygR2frm|O=
z+!D*qfOcRS1q4_tYp;3E__1Ly+J3NM^zc^OUdw^x!KTVoD&I&juItJ!5xnTR4sY$u
zV`H8(2ZsB<+3iCuAbcSCBhtQsY$@-1Xq%PYI?o9a>BnHMH;lqX5DkIOM6-2Su-@b%
zwZN8w>3m0+`R0LDAFA?;TdusL4lwhX4FbhsFd0*gM&Hl~Xf(mU8wgw-GJ|tG)$8;E
z4a}y&wQhyQyr~HCcy^Qh$t<rV+IZcfuUl%t$Q*A+)ms_9q9#_QoNn&5viJy=L{Z0`
zoc(hYTK4K($Le3k?bJKQDXjMo(9nqEsDYlSTDN^JBkt=ex#XHU@ob<HII>BjRrN6b
z7|PTSav95ConOstFcu}ZHD9d?Y;Cf8ou8uGrX26=y@DF7Xw`Q>6@|u2xc)F8CH%8P
zC;ost4EzC1mSS8W!Q|)Qoc4IA9r|C{f%Klwnc$a%rK_l=<yS}70j<~%ADh{NyJ|bJ
ztT1SrMhJXi;+tn~bpxmb_fK;Z!JeI?WTJ;Ru5PcMpj<kxfH1$VMILMvHeuQ^pm93#
zFbN6@S6GS)&BIYja5w$ZwvqQ72=+7Ur@OH=A(+&<B>KO$Y%gmk(?(f#pwm!56w-F>
z+Qync6sEUV=kAbqey&-iMa>HL_+v|h87KDHrc$C;9Wy?XG6Z$xW+=f0c$-Q9_&4{v
zjt_}pzijXOMk`o#Wm`Rv`b8}J(o;V(mohIgd0!a*Gh3Q9D*io6QFsSPL+ZN^6e0-p
z(v5#g$o(5+yu#4L{UNHeRH5rr`U=1bg1f0C@?^~ySWkXVx=HknVjr_UZEq#obJ;9r
z0+#py1z%Cz0tDMfCsN~hOqa#7b|LxRigZLF=A~KUsvK)nd8^j~X^xv-`xM+?=;bjl
zUOmH6tCCETIcX6FwLKp$P2@_xaU{<731!tmZFjFmiG>Mj`#(S)z{A6@XRi6}MF7Mc
zs7DsrR%BSw1Wg*vVU>KKurED|nxVYLr(FR#IAfY&MR|^VxU-ZVsJt4Zg$R+_L<S{7
zPH6_95q2~#3oq;i-M$0gMWOC$9zQnW(17;S_(+S+H_4Ea!>9s1Nr`QAZlLcU>M^Y>
z0iQiECIxI8w@jOpu|J`;@vVPa(3ZhI4=O(b4+Am^hr&K)Lkh3SRs0q$EvO1ogTP?m
z!L_k+dSylqMhune8qqp$^LW9Sv*spr0W^KOq|d7aKMKK!7XBtUk+eZtv58h!e_sAR
z3<a2--ipYgyfU2%p;?YHE#5e6l{7$|{~n<P{C@D0_P~rdwOWvxD!2y169D8FLApf-
z1=ky`YR#<6hp#yP(h#YbjMz0R;1M1v*@p7J<kQ`pAy{D*jI4D_Cm=)g3ibE$5NOs;
zr)ls2K}o+@-8&iWprdNrJ>U_!Gkqec49pvAc|QpFVa^f6x=5QMe7s;VeJl9QCWucZ
z5c9&3)pVT+e@eW8OE$^eH5ec{G%B2dQ_e-tGS@)zYo6pm>Q6z^o3FZ=<%{JJ8n%E2
zko&C=Ov2n&L8<UaB{`*37di>d$r#|$R&Lk%zIMg^Ilz|}+g#usW0|b8*qVhKlZd}*
zR@9$J*dW#!Ji$T%1o6ydp#y>qpjB9s^^qe64&R_3Dc=ohcy~#5zTfYToKKNeuypC%
zd{4+_FO?|vMByh^f|fE0(D-+A=@!j4tbH5GC;?Bl!%7a>=daA0edrFsWMNAW!#<lF
zj0P(!l_H?W@!cN!tI=jXP9F-(tRKx4#|Dd6($5S8_F}w7LC}redtzPE*Vh?-a=A9(
z%B>ks`?oAP`#toGB9*1LX3j~PZl?TdVp2-Asvme!o_Xo(t=Uem#26WxmD!$xPtLTP
z?7(!##L?_$?@XPKJb3i4PyTY_>H(2~;UabU<YfvwcJ&SLb}bYRW9y+MTl>RyI!!LJ
zttg?Bu(+nIzwR1;-at=`Ry~zOe3@0#mtk3diOD$h-PwKA9y&`)E5jFGY8rqdG$3pk
z5XFJnLOy6P+koF491Qu^nqxA@ty9vDM1l6V_EbOl?Q6<K|EbgDrgpuSZPLoJo^`Pr
z#^Kl|<G!}9Qpx>xc4W9<`Me|DW)?d2*j*$P27VrX%DX2FF}VsXvd;)14zjW9`uau3
zE7&i5F7jT1mjH16<_6g|g@Cg`+l%E8UoPsL>P~N7Esi(BpIXlG%X~0?=v{G#W4n)U
zY%jr^;W8Qq27tVISEafz^;q45^gV!mItgIWhfTUk^1&@CIogQFX<Km#wRBBD`yJgO
zHLpe77c4nQ8uZPu<xOX=sAQ5;5t3v3wy4jseSOR%3G4^Jd6SX($Z}dasVE!M*D8I~
z%^RbFT){s$FS`(nC;r(iGbTQ$LhIEfv`MUyvdNZIk>0>T48VeoVL)-xa~5ea`{pYn
zAk6B<s*iPsimLKJWWby-=`!QM-b%q&{pdn}OgS>*lM2$r`rLvJ=wmvyJe(`ROP~-T
z80TUNg#i!FK+H_m)jsZHqcQ`6s@Q~}wbQ7(UbL_0*Fji}QY4}XpFm28vn9c2hOZGy
z{PzR(Q#-0sK)iPXE+&qb@Gzd%8j)sox<Mz^U%)4b>vMvgCgxogK*o4(Nl@%W3V@fq
zXK<irQ_eRj=eNJkzD!?!xEnCg(%mG>oNLoE_c_(${I9V5=_ZbT(?vF;&#ksm-j<IG
z#<Hiu)_`Ob&GoS^z+%0saa-&Iq!zTR6Kz+IPj1SSBRbYPP%UNBJ;~muV(xTDQQjc<
zqk0>!OE;X-LOP6`?rnG_5E5{o()V<tXvfPi_s8&*<3Rsazu;jPUMP9Lvzyiy{^a3>
zEavb6<?-#5Z)I>$(No8kybU%d#(*2NuTd75#p<K1@xD6EHT;lX-Uco-*6tYT*diFW
zI`Iw)8&f1q)~91s+Sx0ei%#`F>q1Zi7UhxU_F<VTnUf+XRFZyV1C#epSE+<B;AlLd
z_w(w-ZjK>iHhEI}4JM#vrrS0_3^izdYYy^PG8jT=6=NBVpN?TiYUh5_7z<*r6~*p6
zH&?xCh_P&o(PzV<azHo=k-XzqNh)*U{`-=WO(ApqCr&z-EQ}HOZcT{k5?Z0F5tq@m
zb>fV@j=xC)cn^5?UZ%y-l<^*eCiF-AHhpGVsXY?b8bf5bK<#i~-!~IcfG{y=i6J6Z
z$k7l0u``~w9W&S`{)zg4G~kREjB*)Ft;+_L1K#45HqD?pIgZl9C=H3_0phT?3S{$h
zsFH_W6dC2w5T&cK|1=`gV@(Bfz88NNc==C?^vO~(k@5S)r(3R-jSN;sG}2v0YQsV&
z^~f+$5+a%8u(3q^^i<1=qvcM_m{P&33Ca+lyMdUw%29(I(Fi=w->5)W`4(M;V7EGs
zGP_VQ&b=C7)$)uZCjGVqpHM}j4Gnzd<_}u%3(;`qG{?kI`CuaRgzMW;Xyf^UU%%2E
z1MoV3R;|9SIC^#R#d*)i;POt}&`ts6s`P%cd6g*Z3jJZ#=oZ`N<_FDW!@~6RUq(aG
z6nk?wz%@R%q^l}XukOV4GA``vtv&tnCx{={1X!ZcnAbn-0%i3vXlA|LHI6U%g6`tp
zPc)W^Z-566-zc3qRP%nv>2j0=CG*d!wAgoO((wF#-1_Pz_o=jKlljM#;Jy|Dr3w7f
zw;rZ;eHCs`lp8>g+SU6uOm^1&*k6?}PRv!3JM*VY&~-S_`HhylfaManS-vedntRB)
zQSn5W^dY;vWcH5$!siRA<GwT)8OrLm<H8!a@<0|ei;fD|P#yAhXqo#%fDij=sAY<Y
z71lwyY`%4FkH^Xau7f|hj9*cS3B_|p7^qITjYxQ|%V?acMNOAwXAqJV-*XEILG<LQ
z35KeC{3htN<4cyiS;Og#i*g8>e*Y4l<j|-F$;QUTM_^`?cO(s7K(1BjT11F#C$T{S
z*I*Bfd((CX1HlF$QDkKQ@G_*>@Pr@P-Qb{Kg)oE?wL!WC0x{oAn<Zn9?2ht*b5KgX
zp(TfJ!~MBq^|O{zflt&1g~nEF2sVec_kDMM>ZnaOW(h!q(*vrcCqT$nHvJ9r!M;;V
z!O`8Ta!4=(0e}SyYv88fb*^pDUjrMQdJ~clltGn9nfNWB3wY(Ts>gAvmQ_D2swObI
zd-IE3g!XpYim9%erm?1F`1bIHs~o+(bCJR7!v!<VN_8oXF^govrJvJ_``?_+;Sca#
zaS|OXU9V%CkRx}S>ra@zErPBnS87>}oZoqT0?91sXH(H?<x5<PF#2)RDv{eQ*5#2z
zT{sy9kFuoV^Qc6*fl*Lsu<&bm<rBeGcT$rA`g0@;k1eE5R*KnTRSqIER-?<)arY7A
z>VgpYi5U0$dtW6bk{seca<7H$eX>D*SM5(72@csY2jd!Wdw)oFfpomyZ2|9P;s#P-
zVN78u5JQm`O*JKa%Yc~I)l`G47p%(3f?gylqv)6FJ_jyb3N9BJT9ZBz7bT&XB^<Yg
z9lr(2>jD3QHygf8TUyY7Jjsk&EkTNU0%JsirSL0evoGtsY|7;25CW3lm34WJuGy#_
zLoE>ws3`M>pWcB*W&@Y$cJR8qyNoZmduI@evs&aH1RsA8g8m}gmI5F3wYIAq*xO^d
ze3So>mC6YlGq-hpkt396K!On+R0M`KW_iXS6WYVuuH#Qj_ZTQHgY=0<LA_%}fnyN2
zd?-rSkBe=Pw5`VAiosEz@JQFc@@-hZ5{cQjC(yojg-~f#jL2}8M3zA=EkNyAy>{RG
zM^UyJn4N$I0O?mWShhKE5e%&(@*GQ#P8Y{t;&s#AfDi-@r&3N=9dUQt=%X*z8A~<4
zWwigH8q3l9qX-q^9oFpo>QyV<rwfIf?CpG|ht~F9W<Po10o*Ad_G*#&?LST;z|SYA
zkvhaX*_O3|5!rIbgwk_RlmR?RHeBE*--;Z50zguXK*J>8#dQXntp|DC|KWzi*6hA?
zZ~UJvkLL3}H^_-#{u3e~F!GC&X13w9kKO1<uHnbOqv#O#)#2p}Cc^<hyP<*r5gOPD
zH9yfwGp9fG5jyF_oFKX(3?ZU{#{pz#8Vkv1YtBzGDV%!^4ZNE>6;2JNxt)h28(ZRT
zG37wyQ3+I9shnx3V%NJU4s67h;Sg6@Gld+R0-&95r{~`4#<XHOmyh`&pleIsP|n+l
zO}{FZZ@YKSgkl{3Uk?CX33eX<E%S@)bqx|a=Hc?lHQkQI0$;6ZWvG-4J#%O$E@v+x
zC4dKH^WbVi;TE;VrUiteXfYXa%zRcYD-f33+SQ<PO+n#iolbQ#P_P_AA)$}sWnm;x
zcs7g|=u*h*wowassG`vh|LD_NmgE3Z!ma$zA|^xmSV>)N7K)<dr(44ON@6V}y6J7h
zoLV1dMCPQvPP+w>bjE15hB(rfL&0~I=~Ne{JqL^G2>N&Cqb`u)TBm(S$C8fEYQugI
z2S8_vx?@GK<;n{irdMbQf(2=;6bI(%{o;t}Dl=GKt3-*FS@NCIZz_U4Zl-CrIu4!%
ziXK{)0=bh_Szl)mTW;H!y%_dWkWkecns)+(YaQuPH2GqmEnD!#z5Nwpk9Q~y3~R)n
z`H@SR167_wp+7)Ys$^UjwSu+Z2yqxNIj&WgxLmi)3fS$d303QC^p&&-G?a;MNRiD&
zHpI@!;WgimQF}83#qQ=+^Fvtw?HQ*H9XlbGB;}v+X79ApsfP-eFu;EN^01fyX7XTc
zw}J0s^riA2#qPFtc+vRtI$7@mRX174YbkAo_(Yrz{0}F*?_QF-NyZfTL7wRtr-SiG
zlOB2gdHPdM<pMnseo(70om;d?&9PX~OQc)K2U*x|%AKb_6IbSTNdl7>lXg@`pFQiZ
zQNn3pm^wPK^(U@<B;u!5Q1{d>P!%X5qx0Cw*_@eC6DAA8IW~L3YFvfdV+aK{``V}o
z^T3`CfnDT>?1^mr$rT^^kf?%dkd)%iZ3sO#QB8?86V{Eh70{9?AdmQUZ#|B*wx*59
z=(8;g2iTj&!tN73$N*1|qm6-O8OOP^0^VB`QSn(0vSRYK4-P^X0R3v}w5=)VvRQ<o
zi@+W<NRiA0$$en7?ue-jXFdNU4EU>hXk<#-Dc4}a?P!tRCaP97^C1b?d;uu-Yjt;x
z@jnt&E!~scXD}j2pe{MALW>O<{kA4cMFlvYY1?NaggGLa!7kg6lA+(;x<t8C3xw?S
zcI9n6hbw_AXCxyG{j4V<mhr?ej^HDP5^h3Loy>v&P^(whgK%)VcdgQV&LoeGMSo`n
zg8x@K*XD|Oc7vYn4A474dcoa@(R|)UU6gLR^IVhxpQ5bP(<85|CCVL0KjzT>$8<h>
z{k)3ftPQD?z)E%$e3@xmCXT05lm;~Th{6-#-YVyCSR2WMPu2d?mxHZj=ZSz`6(1oE
z^d@06POeRDK=kw5Jm8S;KT+)Kk=m+le+KJ7S_LqT0wa-HAFh?|`>W3XJChz3mu@~c
zA2mOSe%U(mYNR=-GBo0SVh_Peh4i1aCGrCDuTvf;b~9+e7kh|I2^35=vCCY6odzIh
zfT!#`9~7{W8q4}!mrV1l0E+Xt=Uqm~*F7<(z0fb$od#f23IHBRYc{G#DS)x^aQH||
zXi4}XUR_{t8cMBK=ydSD4b*Ko@D>e51xZ-T7%?8!qIR!V3VSM;v6Q?}1yOs!g<~aH
zhRfFNn{(6#5LI5VjI}Uw=b^0HF6!-BZN^r&8K1IBu6z<+nuiFZSI`$nlzZ$GFtJu~
z5HSjOtA>(6QXWTr0i=}dzI{AzCO>ip);H<cnTzk*EwXLt6$mxx*Vl~`L2=yGgKaf~
zaS$F&$%r1;qS;~fHc|&-A)%OVK7n)KcMWQHM3p=-brf0&wek#*13Cn=DRc-2I)JEn
zp;0!lvx;MU%t@M?1uP(7C{goy-~Oh@xRKS_0Z5CR7&Jc&#&08|8dG2ygm5KLs3nkk
z6$|3hRlEWEV~FKyg!#JD8Jt?TnD->df9tD*5NY{^Qw9&RgzJ=$SU0ko4^H^vcN>#^
z9SmtwgtHJD#;khxPt?@3xGb(OYT~vHUNyM0+cdKOAW9K4b?+jtNbZ|KhTqnrcRX;z
z!MHoy#Y17qy8=daOIz#uVa1<e_qXHBR}W&J|CxaBzue$GdiaW;zTCdGb$x!p0ValH
zj#0x)!kdDn7fri&lYAt=`t#*Rl*6IOARETty!)_)@APku9b$^FOS<6Tr-)S}ZVx2A
zz|I5jbD+<Dsa#GK)-ZFCG!lX5K1=K+^n<Inqj34YQc;9rBm@0C<G8gzsGqJ^hQ+GZ
z)})}Vof0V=%oLYzFLX9ba}DA?#0G^BGC=Bbau5UM))qKvOL=h5?cb{>vX%+8sf6d)
zu%K%IM}bUSw*FO}#EkMH-!ZTguC421>L!QrmeF&w9meJA_V9(uk5N6=`93!Q=fh*B
z^Ok!Q=)?*DpO91?Yz3J8nslHD7WMniF|4I(Loc5pug>N7g!cLDI0TLa-*i5OTN=5M
z@|>BjU{&)w!@mXZJto{zReRqGsGcr*oZ$z)3x2~R5i)2kV-kLE6}I&ABR{ds?|4cp
z^g$zQ7hIkuBmzv^*JNv8t0R%hI||yz6I+E^=31=<g7psvn)be~tZvt=DzN0r^lJ4~
zq+6vB%CZ(Q85g4(8dOP_&89t$^frFdSDp?T)-8Jei;c;NQNjxRmV*q{HUI@Wz@g<L
zpil>2Q@Uj}jdL);;&{ICHMne~L!qk2Yw00m>^4zeKp>RBwE^DP!W1^62EpY*BHVKQ
z)p5ZXp-h<%<sI+(IYeqnhnHvwk;3?#@LC@s<7^8cp#rP~T&3^ANw}jspR}bFufH13
zY@luZ_655jF}KnfNNdQD+tsF0#INvG+b{cI9T=X6JwMxm34VYK#$VeGfapyKa36lK
zoZlQ?6lNN9TKw{-S^4RDv%{l*u&2B1exRhbMh;1_SEAC;t8)NjMf@rsKy<JvKLafU
z5Yhz=q5u`LRkHGP#w0QUITk|b<jY{&z)w$chnNlL)^=MH?P_O?%PcCy|IAQ69=J_u
z4^z{Hfy_nW)$Bi+u$Q)%7GC$NS>NTGM2!7^6nBpOdf(zw#B@di7<Lc^P1{Q<5qLy7
zQ-vsA0RpxE^yOEw8XvAa(Fk?O&v3lo`m^twtVu`4ahM}hAJg|vK`B@?hyHLa+e1at
zofr>L1}@x50W|T91Qu!kjY?2HY)pB-ocM7tM55lheDT?ijv%p2RtZ++JXl~PnTx<j
z!y+`}1d_3d-XJiiE#oaZh59=t%#>B~tUiK0vmd)y+T~|FVL}(Z+t~)4bk$;Hm*(U*
zYd~1w?m}F1GmDbE-DUa*Zx=xrvzVmLj=@DpPFnC2(#3{%MCcC6qv;jSGeT7-rIVa&
zUknP7htO{PvI)@dqY&jTVsh|llB_pz<O0UV&sWBhcBI*Cg!!A&E0zU3E=rcq10fV2
z%29c~PknleASE(VKVS~>@-DMI8F7qyJKKUw2#g@FK(HT=HbKzYLP<a1wel;;B{L>l
zGl45t5Ja_g;%Baa@-qu~I{}BLT&p_eF?VAj`Pql^0k>C8ZA%Y>dyj7IN~wOT^Iz33
z`GmzW8)5*o*O8}9YV%DP`~1Ndmq+Jb3dW5FE6mLogqCE|+NTFK|1^#?SJ#|erzX3}
zU6O-l+15NpyoCtFn|5OABcWb-p4MB+Cm{0RSw|cGez&A|0Xtz8U00qReWsZLp-xh`
z{blrR%Ayv;`|QDhTbCbhT@{?&w;t-RrSYHfUeMPm04_pHr?vQ@kz>HCQ>ZKBx5Gfk
zjrmy3uF%5KxKJ`#>6hYjpq+{pfA0V)bi+By@w6c5+a+h`X59~fef}u@(&DvSG!<Ti
zfsqo-Dv3Xy*dLN@WyRNrkx_JvLlcRrY(Ot(vfGIo#wd}z=^pko{Yt|hq5z##c>}QX
zu~t&Blv46^-2%-?_+P`0_oT+pd0b#bvf0x69WoAGha+-vjLFXd%~A&ra($X8$I^mp
zBqAxzufpUf4E2i$Nm!4P>N<ga5!l=$62}S={>?g5agfw(vbb-oKgxSMJ9!v5&F!`w
z#jpYGu&sxP*dM@WfzZ=gX65~=L+KV_7pjx1ZBw7*AnjLh*+ph6>d$3?KEHj8FZnq}
z{+#*A;ZsedamU*w%UZgLj3Ro6J^QYHQ3eYi<;f}G1r8v3X#+xH&QG`>-mz&gmFe>a
ziv7pv{2$0X?fnPc=)g{s;u!ksvGzeR(M(^b;5a+tEYT&&$lW%<p^O7Q_E`Bd6$Cg@
zkAz0Q8Sm}mz1l1`*>41c!<JNmtftBK31Tql|JMJVUupiRzF|!|!R@<XoH?erBG6iz
z=h<p-^B-P5jucmP1e_sIwaLBRr6OHvC31RDxN!w!aCV-IqLU{+n|E=nkGj{moBtL!
z_ceaLyxRM6S2}JZ;IWaUosyMh>EUtx@o(jQulP0d?S$GLv7DTo*l7>L?ZEIYj@qk>
ziKCm@FMD=(FJIrE#PCw#=^7br<$w8`e?8Ov`p=WU4Io!xp1MXp^i`x_y(YkJ`%z|-
z&G?DQEu-U%@_hMYn*|8p>e<H>BHV7DVPKkb9mM8V#|y~ZR+$dwc5NNlVEc?MP9Vym
zfZ#n1G}6>&Y+%^}XY^}FhfGVbX7k~qi+rp~t61@*CKW=E@pCd*rr3rs4`dd1hMD{c
zK>_<LWXaFDOW{cZyzYuK#zTBBWw=S9ZN=|jRwVfA3a#jAS7E-HQ!lL<qYnAjKd@<<
zy67<|ILaCh@I@+&>!({sbWXGMa^VJPqT~!hFvXRB;sjNCVoJfDh9_$Es;9bhyT!an
zr_HtrrDPa9lN2gl#Q?A%<L4b6R&4Q>@Pzmh4IYxd$7fc%at%$Q8-GU=AqqOSi9!pk
zao;Ee>p2Y5%XPSg##zgz*hWiIt{h|mu|l2SFMaC>A8hw^>e_O_rkQgZKmg|VhXC?G
zCK;W$%HqR>Qr#Q2o>fa=JkMrtTO9ob$Lm_{nBky=@+(8kqAlAf_GMU2Idd#_K#`zP
zfxDY5^kE;T5ta7kcu7k7PGWEGEJe73XNnmRWYh*>QT~X3^OJb`ANj^bXv1e|!>oQ_
zA_mA*l7iKL@I(FFfMN5AzG0z;0KDW#%>!73@BI3~SR8=5gvU8f<{Rl~#Fm(boib}D
zw`xD59iCHB8x@~D+$O{vvvVF8TM^c``-1Z<uKAeL$b$I(p71@UPKlLR)9C7O<JS6I
z)x(pena9fy7hesHmt`Oh{zRCENc~8Heq?w__zTfd#_3omOfv#kccMQwR#+~2a<JAD
z7?Yp6n5=_h{6l`bLHY(DWjh*BQJd>pkaoFC_<s%co(?Z0)mM%xFQM<XYeT}?q#xy3
zO7wUQwqf&P*mjIya=WBj^!{goNxCx12kCbr2KH?XOHwv1mDZ)t2^EF%gkU#`?ShaP
zA4@?moKb<=FxRm$Y%McmDzJ&DC3A##HkNf~lt?9R=4|VK|E~~BH+i&(oTX(^=-W{s
zc@bnA9f+mEUmSl+QB@8IUF0(%4d-?$&T+CT>;l-xWJV^5$ilt>xU#bhhwO;(omJ0D
z#}iC8r0k%ULoN=`2@r#@gr{SHciF>#M)GyXgr*C{bl^q0Y(VLFkok!ID=ysG5g6}=
z>~nv$NF)W4@(zIz)$@E5Aa^DwD+V|O=Ofi$O6U%K7wxPeYmWrwbw14koqxV9Wk>2G
zwGf520|pcc0sf$bmuUs$#g~>S-~e;bIDVb0b(MG-Y*=QRx+=q3o;Y5ZYU0LC;MoM^
zpMScjkxUi()w9%fCLv=g?pvN{3URZ$qGAniyR|xou9ebE4QIrV##YYOOm)cr<gu_x
z0z$x|YC5+c6h*_#hT_XomcM(#2PmJI9=-sCg0bb0#d<otU51#YrbY8ybC8asT5s}9
znl-E0NjcW+U+_Xv9swi8G=4sLk#h1D5C{Cz_a+5{^yu0B`gM<?Uf1hSjsHueW7jS@
zj*IaM1epm_Q>4|l3#;qB3AgKN_YhH0(c4wd`&F8&>AxFB_vZHlS=YR`^TT&kM_yZ^
z*E?DFRz4frK0$ibwxc_@i&Y<pWg50#9c7R2921G35J5RPD;gVB8V$GN<fyOFyf<n7
zD5bk|e>PRy9>AJ5>s6P2$1$jDr32DL3>TlH9j|-7fN6tB6u=vrdcK-y?_vj4O)DYi
zM}I8((}o2UA^~MFCE-SbYECw=jk1g9)iHo*`XZiB4kJ#z-|>0X(8`Qyi3yO;(i3O1
zYpG;6@2H^dR_P#JbG2yc%)<QGMHHjrhpflJDmR&uG?)!|9)-wnCTu`|`B6(b=^(rN
zaD>SHF`Q1B*W&Sp5lq?`%1-D5Ar=d!SFLw~gTzMi8%dR4>J=Zf&;2QxlG_uYz<LZ=
z*{rO2wC93~&z;QCm_ui&F@RhV8UXY$X)1c%Z1HXgHr^A$Lu$2JgKtLwDb0)!u;Os#
z3-*aDWCIMHjowO9c9jUOUSTSL`XbQGbl8`+!GC3A09e6iLkTK?nNMEJZB!$^?CK_4
z{~hfAl~shaE}2osc}`3c>0ec80)Z`o9hEX59|2o#{r6n9@MX9#-77hCW3Z{le>n{T
zCxUah>;n}T6nQq;Fn}@y{~0vl%~iOI3{rOr5DSqSV_$}cKgL_!rjfbT%w>Q(5R_rS
z>f5iPME=AqOtll_slRrr=r+1Pw({8tdIg9N9lF6dU@)caCa16e{S?5M@8>_O9ores
z7?<RnJzPzwtnzI=oL}*X-@n|tz8+pcQ;<wD6L`n;Wox>Gz$Ym`3JWTz{6BnPXJbDg
z;s3IHbMmqAUv>6yuwelFp<o7$qhc=rmjON&VMH*xhGaZei8+T<{}_#st+|2@-Iy|Z
zl0O$i0#KsLj!&^tV#PzUKk_gdun;f<_hk=}Y!r(O+>EU*yaap3`I))2Rk%NS)D%c^
z|52vSUHHzQ)0682+(SU01NPS-J1VnT+-x}*p9u#+&#OH}H@q`QZzi)kW)eg+$djBM
zYK20pWCCBE!35h&ur~6z>=g#<w+PK(+nyEz<M=~}ncFtzng&8skbj8dEf+ip9JgJD
z1DfqQw!2h=(rxb9*5&F%g)o#U(gjs`GPtGoNf5u{3jvxFIi7)TV)qWFQ$(SqoB<*8
zR8WZn>ib{CtJle9!U4K(f{*#8p&8-XT*vf)b|X*tUC;*9a6sGhcTxHn)5Y)3qvGuR
z_fSd`(HPqvRBLdxmTD|3>{$ep(m(pEtD*C1h;4_fagnFfDk0LO5eX-472{OgiAHtp
zrnWfI-Frz1j{_J7-V=sH*UmZ##KuFRRovEjMhZ}gGkoMHhWyD#ajV12qn`kH#jHjE
zG_f|(doGdan7;+A1Tm!o0R8v5n_8X%j47THx_kXl{X+Uv_!uCKvJFOs)6o<8!mMc3
zKN@2#T~1&TajtaehvgwUE8AHQ!iNCol9<9%RXTIy-v`NZ(R<Zz(%NntT#hJYq<vsV
z$CljH*;zEQ5+w#u9zR`BMw%uwxE&ct_ecTwa{7{K0*c!sym0;7E#pC+*q7$IN6!8B
zr29JL;h09VP|N-9Vp`ba6k>GMMt0Rob1P9*RdwIeJSDnxt-5#;?e;$j%a6=-yLNZk
z#7`N-+hQKBVjeOd?uPH?hfiQ2<aEBl<Loj}?f$=iLv_6L*n{klygvEzKuWA@m&`4k
zSK00n<LRP}6rhj#)h`nBgFoe}t=8S;OJAyLDzv|LA;S^42!-U;AruBdZW%UZ!pGY~
zlBDcRN-hZt*~2X{)X;gnSKRlX=0rX^3D8Oa)g9`pmPw^myqRz{e!k7Ge&{5Q;X;O3
zOJGf$ydAFoXssx@hkSV(j1qTh%f>>YPpThYLNf$XgcA-3Gr)XR88Q{m7a+zwKpPn}
zG%yBF=OXQI@r7dQeg=tClO=#B<(%L4RzrXfJjOaj145x1SeChDOg6YwuHCN*g&7UZ
z3pvI@D`;1K-zR1gfqLsAzDrVu?7bG;*JNDz_63!S7}#SP$7@zN0~Hf37jFfv1e{m^
zcssIo>}|9CDH(wx9$yws6@~R;E%NM4(qgGF95QXOwBf4#>ksEkI50HUdxykP668@N
z(+5C|_t?(h0nlOaaoM<U0kHt8RZ2Fu4>LDsEd65@4McYb9wKMNXX#GnQ{NEZ#?%=x
zA?Sn&Abjmvz^8fg4^DHoe}6AjdpGVQ7pgeP(XeZLUCL0<Py<>dKV%ju!i#Dhha!6-
zJ%rALZ_-~Kc|TmKK9+y~O_J?7=e3@aQZYS`spWp?-NEJOrLF5%=4pzm%<`HghxpM(
zh96x*COz?FNS`g<HFdxJ99blvp2_i3t1Rf_&#k!eAj78$ImtvG!(SG+mpe14v-IpL
z<}On*0vl*>QIm`Cid{mWM345n?S3iv+0{7(8$H~Tmg@Z|pQWhKubkbD1OZOs#FT$N
zQ|qjMdAL)8YC-Q0x~BlF=F52%Y+l%C(^MOOJBEH}khB&I*rqe*@8dC^DFxM*4?e8z
zsXBG1Zzngu+xrN@XAV5hyXYDnI({JgFBrbw3w3bwsj6-1eAn479~fo%PdvKDf$u#P
zQ*0PsaL!Zef}Z6h%!CxbEUt8CfWuQZ_@*5CR6aE)k2guMVCK05{d}o`AZ;Y?n`JG3
zp;qk98Lc=*Etz#3%hc_^X$a{QWsy$CghGE8ulJekdcWpll|;QgB|_=?f63Cl6r{~B
zPDgP3eCyUj@jM<15tT>>$%bhM+;ay6mHxer*yfx4ZFlKuE$O)ma=WLCzIrde=4PuU
z`kJz^#6x1~joW5-P%V*8HO74{WM$^MOx(v(AZ+0ovoeVaX&>s;(xqg072vP#U!vMl
z)!De}6Py%ahC0tqovcM3vb-lo>eR_w#51DQU*>9ki`>TgcrovgUA5>eGwvIC9j`J;
zV1aNa6jT4@Oxi~ZDz}UGJu7eIoh_7@F49|j)Mh@^<1~`XBn$Com?F(06k^HBN2$2a
zO`dzakcoRr6!(abQ4$})Yv#|AOA{zx$VD909$1}~wV4^j^SP0GIy(3&`%T0Cg?Bt(
z-ce^{d&+Elr{Y4Zmbri^^^fe4M8g#JvZYIkkg!KWN|G3S=iq)ZLB7{?#<8W2s`*kT
zZ%K8wtQfT9X_4k`;_0XyRllX^-2FPMcGRNQ0%$_M9L1Sa*U#o21m9NPY}ImUug7p4
z0C!dG(9RC(!>u0Y@ztP!-V47bO=58HT;2Y_%W9fkpF7MnrNaC^X#G+SVXi+*!)3si
z%KGk+O=Y+K{GP+GNva<ih(PEOJ@Pb%jWusOJ#_9BYd#{4kjeD^qdPS<)%4$^`T6<v
z^PTalrU!HH3@)6qPidFU4|iVu)n5H4-Agx%mt%VuSB*<Ijp7$>|ISqPLam>l(!k%B
zat%w*GM270Zf9cd$iq)}2S)32V*~uup5;r9$F;gnr#)NaFQEw^JTIJkacIN=E5}`x
z{^bk4MszD;UdZu!eX@N`zWS^p=i=4*V-DlWm60AwU`pF)d@Y#9p^M<bB0NQQ&Ca48
z)i_~14#PJPp;dsndcG#JIQ`YV0&&tnM2_%e&hDR9Rg8BeB$P3=AN)zVYue5+NZ~x$
zP8ESYj-U*YA{Jh=f97B-qL%V)AK$BimMN6SMEb;1&19oNUg)k(p(7d~ws{7Kp>IFp
zy%aRd-=agw&p(|w75Sw7!CWyqII&IkbI0Gf2+>-aV*Zdy*=UeI>X6@mP$d^19(B0!
ziwGVE0tPafxBoV3^st16#^b<~emXy9C0YpQ^V914u#S6Z$l|c&g~_R?V;8+0DH)!u
zNE{*dUR<$P@A)N^6mu<dQ9kfz46{d;M!aSfzvo(P6q03V*8@$4u*@c$-jt2!iBwh;
zCY$+>Lwn;U`Q9-NlGJ4O*MfLNIB@58`4LP-&^0;k8*h}_A0ilToTMiVC-GzO)QzI;
zA&fBlTh)z0&febs@<^k*d8WAq?qsGeMZE(8VVG&QiD>~3iJr|mO_O_%=wbTG+m9VA
zPj$bq3vJ(gpK=l_8(jy>zt>(Tp@Qip8ddy@59fGk8Qd|YZbH0CYY%rXw`x7M$6URu
z>P?#tG9TYx&={<w9<+Y-stOP8@M7$4zM8^@PNT<fO};bgf4n@HOJ{rBy-^b=zdf@P
zS?ekEzL_T0oE6?}R%NE{fV0YHzRdhx>m?X(ZeTcf5ojd-XGM?v!PUwT1>p1t1fri}
zGHJkR;{a6Ym<p)j^eWHT+w>E^QScBJ(JM2>qkD6v{(Q`P8vKaMO_sngolpi-`Me`O
zy=>ux8HV=>HA!xSAodL*Mb(0Xbng^MXlx4#;(^OiwKS-IivUk^#$n1{2uCnmg?<@x
zuEt|vDcHg(5(coe5BT0yb(t(8kT7s}sPIx7*^N{EO?=_kiABUMnOUksjH4ng_?O^7
z;|)T>aQfH;B{PdvSIbF6A>NS>XXeOr|6Gha`iF?tX_whA3WBcQGIS<|j!@Ta8ToQ$
z20IW%z?#CAp@FU&Ovbo>fvT*My&uX^Qo`2Fo#+FN5BQt=ND9U*M2ZPdK5i%HR0UUt
zmW)M{4a^mE3fVj1IdCS^!p#wuly&EzDoV$C-pvo&xvGyIl1sP}1Zy4Ea|&1Y(o=^_
z3$T7<f5E{P-@dMi3tfhpht-xxNq3EM<p|=%LXOe=dQ(evzA5<cb=zEe{WTmgnOv1-
z2r(@HN6Kk15&z${l%bjK7#_o!Lr2fLmKSF-!G;}^0d%nr4~ju#RX6%6?BQW5vq(X#
ztConHhI{Fx^z0Q3%u>B#M{m?ApSc$>DqN}O9;POl4jx+>yg0WhQB$D^Q!Z_uTK?3<
z^vod;+z{fOh2QYA|3$^>JK9LL(m(%mR>h;6kYfb2W-ZQ_8a1mn8a|Xas}@g!H5?L&
zY+7D}!C?B0o)^QnqLh@BUbkyBph5M(r@38zQtNTXBz_ejE>A4;0Kas2hK}A!6EXVm
z{mV^OHBX>b0FJ;r4o1DLl56_Uud>*mNL7h42p||1?aRAZlxt?G9M48bjHe|!hP{hq
z$n)nb)N?dOD4tNneEL<efitAMRqz`_N_=&8C?zAa+qS!1o$)kz$gSDdui57THJ5hd
zZrJ9e%{8n12RDdf|0qCq8PXF!P=AOpE?#q(AzNsZQx&M;9;H~vjWd|13Jz3g6Fov~
zG;LsMq&wcd?`gq5$xSTJI~}u)aY%`^c)#r*n1w8tFR+)o<ziyUP3A2Wj8qs4lVnn$
z!u0}gNrg_eM|;@wS8C2vZ9=(b#q-t}4K?hldI<W8cP>JLdg!cctFAFiJ{C_n39l14
z_hy|{kbA*7LaBixdpy1^oZkxN*~|DzU`Yx*JytxRun|QD?OE4vom|PLmFL0A_<4A2
zscRCB+MnetJM<{cd9x)v6#>edNYq=Q)=`jtQ43&dnu8|!Yffh$Li56KD+NaxNMA+?
zrhn%)`1w=C+~hT8P2F_+h0Is8+FdSKIeK_|%q-1S<Za+a(crFi)Lm1HQF?mc+3`D<
zcqRsJqbk<lMuy#@kuZ+K3knQ61w=NBzrQ#hQ5=gp7sX%HPi2eF8t|}6M0<A$mE*Le
zf)L^?*qwp_g|HkoA9?DvTx+1p=jgRPM)!!9F=Fyvw3@+36P3Dh4<+;FEpOe2w+3;n
zb0TBXOXo||S&dC&ub`o@=N<u`i*I1kTVj#0A!Rc6_8pNm+LgCT^2$~93{z<g#Gu1n
z`Q7?RhLH1ncrG@P1QR!oq-W-Gbzym%P3tPC5qKGiWc;SMzO99ujNrrYLYlZZ0^@vH
zl~4)BodXo6N_ytrnuTrlSXS_&LWpH@Od`U-C8E?iAtai>+|Q#r;Y7c|{4BlF_}Mts
z)22*hi^TAnch;Qr!>xd2KDOAex0{F8SWb-CdfM*q^ZHmq!9aUGb&z9FqsN+GjvRCn
z`W#32^5Zyg0^ME;|D#<QOb^%w<xzU)gjwd&?y4L`lq8;thIE-2_q89N!zuE-A(mFP
z%q1zHNzRUn%cwjxQ|J`eVCCnLLou&OvFPT6R`m1_xPx=j!C}fm<`env6Hbr>4s%N?
zduOE{ukd8o?<h@{*@0P1HdN6}2KtY}yl1yEJ9#XT&Uhgx+W5hsbPWH&3l~UDW&J65
zTub~1kV`K%V^c>6)8V?C(Y0cOe%sN8b7a}*?3GkY_C6Tmksc1!!v}wW?8MzCU#@kG
z74e(<NPvN%AH6W=z14RCDp})!k(dv-46dTo7yi8g#UypMEfs06ru4imB0#)Da7vLR
zKWF$^P<Yf{xBeofy+u`vo~mHY!b!~}m1a3VhGEBDKgsX4Sc@ehm4sSU<3z`^mW>QD
z<m|rtuZ7sXCm;-@w&6q?WF)unSS14ls0zoyMHyc!>g|E0>vbAF2N(R=LL;r>b<c)W
z(|xqnp)QL9*Rh9B{;-M6w|cSWg|%~MYK36YO{{K6@LlFpah7|=d}?nvR*PHXp22Me
zh$p=oT)I_x({X3>>C?$S+LYH#^_Ls%HNuC<BqVE;Y{0+3h&LL<hG&TBpY`?qfT8!W
z+x+g)J?L7mzDiHudh|~OQo?5@BhVe%BI+egPp<XocdbjzPR4Ar7D~ER-6Ej@Kdiun
zoKkg4w|LI$^j(6udtpFHPkL(a_@#+!9d$Wm##MBIJAihDKSIXvOZkvdvt8U23#c|R
zgY_(_U}Z8I**&Y~pKS$(gJ$WQF9tZG3v(WQyr5cRp(|x4wpkH-OA^*lXg{nOW>@zy
zn0cz6K0D_K_ERg4#DmMM=)HO`MWhsp6JGl>1;Pdq7_7#vR+%BcVhLaRd1!~jlLK(#
zI{33p<OHUdj8LI_xi-tpT`0t*aBO?s%>poX*WbAQCx$b5Tc9ODdGktRIGdFxErVd#
z#)aYucO9GRnz>&7R?z;1*EbwCVZTHLUX)@EQ4wZJk5i=LQq1gbbeUEVB6meawe;)d
zUvve<*)bvEj1Zg>Y8)<jHftiJm7Vx`6cq^1T_{m=gRaIw4UFqPaKZSP)4Y`VE$xpS
z!;-9{JzG1}&8E^gvgSUFQ%0OzIgfrz77R^@jHVp9WB|a=60`V?jskQZYGHUcE>Uza
z%KbXJRe_PA<=h2>;rvd;^qDk3Qh6wmY=;=XY1Qhbz_LwwziX0MId>*B*;6q~q1Lt$
z<gg`exb^!!R(5f@&oHYnot)^?ukHrdJV`fbJX}?EBi1~711BLc{N?deH1WUj4~7Jy
zc|X<56nL9j8cvs{$x2F{@o%m+&>W66Np?K~CS{|?qr@uABXYh=P$z%Hi^)u-kAw+`
zGJsC%mw|CP;Hcy_sSz``J<WNFuGs2p_)KlgQY>>}H`aVreN$P-zcQU}osu!3Ja>&;
z@^9HB0hbvhEc}PYFTMtrR4$GH1<CjksNFBQZ(*UCV$acYiXlo~SEqC&bMHNQx}dYq
zne$KARZ^iz@;rX|wLX07*9RPG->X^Fa%P7HiJ5fU{Ese&M7NGt1B*YS&C_m1&!WmG
zEhGOLhVC(c%M^P50?*rU9`}8tO7B=osSY2O$e{6?S9XpTBA}ZHR8aJBq=|>-ilaO_
za=q)-dA<7Iw8XSok?^PtRCm=^{1~-LlzPs9#*&^h^(DeA^!tLMs%4cKXvdguWZFi^
ztX?QEjpaKA_c^LKBk=w<#iHh6W;M8mBKzj~wKt>kchQaels<b;=`obNs3Cq4Wmm%$
zkjpMsKe^H6*Dw1KB4SRG%?}k7nfnxzCdV<2A2N^x9U3C8aVzUUV~0=01dg7({oGO6
zPA8*eTDci(Msbn8%>~8;9EZ3)J8U0uLnaYHWckhuq20E8ag5uMQuk9*5njtd-eY|2
zW9GqzJvN<^ebOBtazAMDNQd_uYQOLCG#&LQ`Kr9=Ue~4ObIue&l4d=$+?0nkw?L59
zIJ$bDhxNoeYpd#W;>(N;&uA`!uDW_M#oMo6@wPt`Ttbs;U1{;K^7Dy6Q{%>&>~U)m
zF}(QU<ZlAMMfjA%yZF6s=(G)Jot`mM#AMpqcPkG(;XFJGCW3>z9Ly%t*BzQg)&n>V
zjgkoLo0qm0j(h~w#P9gpS7K*pb@A}<+z&n)^*<56lAzk^A3wR3k!`W_98VkJi;D3%
z&+VqUX&1kh`qwKyLr1B*&zF<`GPUDjY`j7e<%_G*r9~p}j-i1IDUNl$+=-^iqb!qi
z)Y+{-^s$nP8F^M?D7_R@>%jHwdAU%w4W%*(F7gz^A=$zEOV3CQW<p|o-pPaKmawd`
z;^M@NO+DyREUNbix-d!PLX^>U1hFu9cy7G278X+~#yxnLvij3d7r*!D3p)!7CRmnk
zfoaszWiZM8b{?9tO2;piEfrfg*&2z!!b0E~#uPzHA|w!PHnV1xo68s{kW22|UGi;&
z^z0Zj4g}L6uBSXw?|}QWcZ>$@l~)~;Q%<sFnT;d30<7Fx?pwOmNkC;NBnp#GUIxc8
z_)3X%&0)#`(w9^QUNO~rrv4BD&eC^yx_;4=^lI;R&dhJr%&@Vwv!2Gr=!iR7#h|PI
z8F2RNM6_!RZ(V0lM`!=kcL5-qLm_%)LgX3K-Vg4tuAc)P{0qa%tX`ImQ$OB!4g7pF
zpbRGL#RG|-s0@fxrtvVo)71T9emv&Y+abV5tl%6k%CHjd%CvU=Zc#|!(ZL@tSqZUO
z=}1mdyid(XC;Jv5LqaBOcHnyClUdExM(_I1?Bl!`Igdekjb&NKdn(v3{<_@QE!Vn+
ztyad@uGgUrd1F14zvg{+{5yHm48R4_b@Zya|D$&C_X#SWL)czVdB;uGta+2!Nz9@v
zA~n7@zm$K+gTNAr?JS2Peo;m#ZDf+K;d>q4d)Z(cG@V)CVDWeIMgM)Z80x*^<xg=>
zt1P7(DSdJ2c!@Z(E{(`Ug^5!PW~AfyNhk1|JVkmAuw$<!f`(8KF9b)7ZVfmsicDk8
zP_6X<Y)om2d*;f1xni2^Ljm{*R8g+T090VuIX=9V8&29MfK6noKaFjL7s#+UMPd<=
zvzADw-y<w#T*2R#_04zR`IQ#DOAGSH^R;36$cxS^hseAlp*s6jtozJ1lSyOb`)7fY
zh>E$hr?6HQ{i1h7kod^g6bc{9rTcN!%~PLfv(4A4lm2ZE)AJyHO1eCV>{lY1P);{v
zDOW}y(G~i_5beLUGWiDI_+OXbJ$_OL83ZLX=sIkvaINa;F4Fxzu|)c(RIx9Q4I0ZM
z!d~UZ@#+@6133_`^*J|gMR5#v;b<3=Qe<I~sqqSkZ~GULo<8!O1hifX{(fXD-4sQT
z9L>`zds5ECPdu1)3@1+1e+kS8EB&K?6we@($8l#5u=t`%*k<ZN`$V^IoMq_@$$5<k
zmbN5knG~{jut;}__C6tOduDg2blJjr{Rk62M0w%;ZG~9Ln^&>a#AVg_bnJj+>|1&t
zLvXYQdT{5j74SsBL!kZA1n;V9c^sAyh!n_9cw64xYj%=|ADP3`8ELC=<G37$=T@~S
zlwcU>2UK9!vxLPLR6jI&!aVVmw84PHwps>_MFsZS=SSz>6q``+w)DqxuA+6Eq?@no
z-NVr|jAGTZ8}{2yuE;luW}unnJ!W|A(%4%VEHCPDeYx7N`-PFum$MX5|0^1<RwHUk
z`Wr77`}LOoPHErrQ+pjtP`fXq+%GiWxcP1$zvSRJl@dRf`hvt<r+HZWGEjQ=yR@vV
z%ukG9=wa^5jotOvhf|~S>gr2#gELM?qWWB4GPfs@?V-c^pUNlC0M@@M`<e39>#7oa
z>pEYi_M5DQjj6Z@nr>p1DB~2aBwis=J&3ZV0D{Atvu(^AeCT*(((Ib;qVw%8V|~E`
ztpCcMl^iRVi$F|NDR%>7zCtH^_rvYh)}@d}KOKW~88-!^QobM_<{COYNm$TeIj<3g
zkizaY@1U$xC?%7bLZ6W0&W-;IvOJ0ibB*<$_-%BO7q}{<fwzBHB$Ii<lY>jx5<f}Y
ztI1PD1|`d`Z(zy&qG8l>!eHJ-P4Tv}Ymx1jiHE2PID&H?^wEYhI~9vmappg@Ds_q@
z2>NW@Iu3?X{Kzz_Wa6~Ah^suQOK5lv_=E*v|Lt$%Fm+ZVdx&3>h^y9A9hJ|WjcA3{
zZ$?11UwE#i>=9|d9o1m|<e9pt^K>>J3!^%XI%~-$SH3NsOONLyH?ZBOg+*DXKu656
zuJuKQjFwF1cBq(Z9q7AF@%~aHa;Ul|DG-vfU4~1{q0$6K!<6YC)>GJZ!LVSVBm~(b
zDR^!kEs+q`bzZ3VVijUuVgu_3U{?rjzIT=%SvD)-8Vgb|T&F-NJ%e>w4mIQK_H7KU
z!qvyim&Zz3wVejojQ;Xm<RL^V;ILx9y~n<3xLh%R?R^m1M01jrHMa$(Uw2C8_np<5
zTfKTO>ehTSqPy#lGPcNQ>N~qRXuPe^LQ$Pa(jE@d&>xP>N6(8Z@AWtyHM=*C9c`4Q
zsd$>A<kPqlfU*T)vHmhch~zZdA>}`LLPj~uN{b3DmlO^JF*MF4&4Qv@kAdUxPwk{*
z;ij{y;3>M5tQ+wv2K||<bz8JvfzEc3BssXa^G0f^&pWS*H-d2*W4Q)d5YO3NxD^w9
z!H`GjDY*<c0!k<b*d*_loh<6!@H2nF^5&F)L8f-`>|Tim(J*@5BFN_$3zwb#baTr|
zfjNR|C<>1mp%oJjky&g7-;gSqP9b*VfsiuAV=%hvu>ja@BoJPxPMl6)quRmxDYP6#
zqDugi)A09B+4PyZxv}l+Zs_W?Kwz2+IRsM3f=X>z{8N3<n*b)T9CYwHL!)Ij6&3av
zCS()xTKmsm{aMPfGgK{^t_V$}BgQ2(DK^eN0DVrYX9w&5vGmpPboT%M+l)=jhNE+w
z>CU64I}g*H!*ny<-5ui$!;R^jO?NX)H`6uE@9p#b{d;@dk8#d*z2aG~Y?aRF!pMM5
zn_2Pr=dvX{F*GksRKA@32xC<Mj)Z~;g}&d<^WG;CBBNqsH&J7C5jN51^|KG<#pclg
zmyK~(H@m4VmtYKv&<|9#W-bf@Db&+<k=516e|eF3<X^4@{i9gEj+Yzz{cX#I5Am``
ztymieG5S8rgd?pNf{j+FIPl8dzIL#-17M(CuT4OAqwv4MlLIZStgeUkWjc#igNiJK
zhEe@A=@TwX%uWnt>u#PISRiJ0Z(246At}_SH!OP&AANt-FL8|M&aPJT==?Wn(=E~;
zXsxcOutDgKHJ5$tfLN~P$rNbN7R!ZYJB9QYmGcyDI0Gp7t<36=#xX~&|7L7J1VThF
zOo%JF2OUSqVLuEi<FU4L9b|Mh<$Lhc{Vo|v<e}T{7Ul5sq1V+_?t{l;8S^=z@l9B|
z9a>udJ5?&Lpm4mfbhj`#G<3Wse82X@#?H>Eht%z8@uD@A{C<UevpbsDd+Fjc`TxjI
zAXj@>A86WSc=QJ!$iG2*ntGw9%VJz;qf*JmJeW#5OEmpj!eXznjI8+wTO7b{;&5Tx
z$n!sxO&c@TuJ*}p?%2*-`ov1rQ|R%gV<iD(gWCMry0h#zfr+tISVEK+zN}@$;625%
ze=;`D>z<aHxwY10_UYHUAo$w0$x{ATsXd@19MZloU}b7BvE+wI8p$U>K|w6udWCr{
zB%xDaxx=VVV*DIF@Eey?J)akYx`8;lF~3V~j6%k!l;8L#gXHHXdPfN_{ji`;)hY8<
z)9NwjMo)_T9Xtj~8LBXooTY>6F`~R)Y?@u(gESWGZc66EFI!_osD@j8bZwh+J5e8-
zo|@@w-6sPbPGpD3dkcusM|KX0HvX2H>8%e=QH9dFX)swYoag?;oxt6tOv}#~-4b3R
z-QW{d-Ul2to5A~JxN;^12NZJbB3WS6uPZ;wC_t;wqEF=D8dtOV?e#f#5w_gl7Wbn(
zw7g$_OQA60+1@O%6ba^MhztO39rEKX@+h;KZv=A-iIp9kH)Zjf=zX<*HcKJJ#_9{L
z_P6;Z6gXJXkEp4v=f)s{{AmpTdop5sPMbl`6C(*xNPh95Bnt5yj-y?;#nU`rtwt>y
zwD_?s$D(4lQroS;ll6<~?pZe_!?oQ{qvVHkH-puEH~C|?hczeY$L{(CjvnF9)4V?a
z5_i{~$J%ZeLsPHv3_fS<4zZLyWz6`l#&k6DbK@D9w`Q(2B=SxS^d6ymUr)!xgtAIq
zT&jjv{0S12d0TXN64W2TkeF7Y*4e6%)d&bS-7(YR6hsP74BXh34AaBBl*a+VH%Sr*
z9k5gym0V+O8{aU&(dc3uRrmKNL}9yg;Owh<1-=b0y;y|N`xKF`u{7qiNpwoAUUcJm
z3UHIgl`o)x&|=g9$+56s-%Xw#HF{rlDldm)W<Pq?CYE-VuUY~}E1?lQ0Ch_}UX2ul
z!5A%J7+At7GT7%W!^OoUa#I56W={tdC!+A(Q06T$E4iNbU392HIq29u<c#<%RKL4;
zB9#5oPf?oaE0<2=sQL62QrtaD2c~|!Euo+2C4?|0doG%mn^q~;_I*TS*~5epU1sGn
zRQ>J+Ckgofjl0Gc*}83ey#0t~`-aqQ4J5v-Cv~VTe%%SNaS5_^%d<jAk6sv~h)6Gv
z5+KtFviy1-oM%kX*~^~wZ9~FBL4%buxWiv~EUGXu@MlDE2?>UGi2fk0Y=cv4FBMoy
z9s0NI>OEXbL&>R0&OU?m#6m{u*Ptl7g6kNKV_?hSm<T9tm^-G`CMsl3ggz^(5<a8b
zLA&XAB9%xeMQGP-LL`&me(jp+kMHy#yD?L&=jQC{Nfz_gJvgMC#LW@$yme>zV+n5}
z4YxMs(_+(Vewi!J({!TL3tf2y^Gf3hQmOeLnycT*mkE5M#&$fwdb@U_pQ|&3WnqC1
z0BtpqZR^~oDXhY*lA)o^F#auZ1jT^3_@1jp6h5_|^khHbDw%Ib@T|s0<HPDXKJydb
zzePTO2kxsMtI<hB>p!K3GbQrCFNdu+wm#rT5iiKS-}p(9c?4~2n1K>@LHck-lj9yh
zem6$GakZ6eVGBH&5BG<@>UzJ(f3}R+3D@_CEETP8w;YHkmzaE^KbncIqu3dE_R@QX
zIPV*MZC0P?Z2LJC5i(oC!u}0ol0xhLB63MgpRHXdsK@E_X?qu+ZAlwvLS%=qz9lAS
zD4^7!<oSt6p^9X&>9Nl2y^Iz~f`Y9;k@G|v94DJy9lpY5+530Y9<={2?D^Z}3O+va
zX78ZfuUREq4bG!m;!IL;=8Wo}zRDZd+O7rDHGhs@Y0z|n&GyEd0tHFP-fMwX7?|Q?
z-dhtaOVK5_bQeqJIHcKsR5Dbw^2kQ#@VEtmC8TVw;CcWY&U=u-upB~ogOqzu`(-O~
zZz#B`ORY^{(*ajL`o)KVtpB!V4z2(LfR@s&+SMJ2Q@;0V#xlbA1f2aSC{eJ0gfE}6
zqKNmuXR>>_2=5b3S>DkhEhlfBhmnjsB7vZFgdqrtRquJ9!hRCcosIqu%!JaQ8Yk+-
z`jf9*V}TYCG1m09T6KU;rv`N`=r`@J*KiUsL!#eA9z-7Td?${#^EP8&b9nC6I7f4J
zDLo{dUmDEsI5^n7PcdsTFCG+QP%rU7Qpe+K)hWXwRA%QYjG)d|={4cWq)q5l9045s
z$)99;1)i^#R`XVkdlyKpyR~w+{W$enC4VNJPy~g$$4vF-_fP9iCvLA3bp*bvZX8@u
z3g3rnzy#Ho4Sd)60KzRrrnk0m!E#hhZt#L#O?~r|MbTyJg`iB&&FRz8CoWQ(<z4i?
z4`YcH^iQrg&TDm__)I@spk~<m<%J*IIghQctHo4LNPug$G8vU~=i4_}GALBSg}bPi
zj=rl+u?ZLwcrPYA^^0;iK0@RHQ1eS^R2{3aLEFg^f)N12i94k*hd@o=538l;IFHEI
zx&HV01QE3Om155$j~3;HXtnuAb`Ll{3X?*2r1-Tcm&F%Z?92r$8Orx(>QYCFF6vS7
zktwbuOSFB)L1{c(;BxD3$hv%RRvs{;_MUgV)sJJx5Y2NyMysO@VN_`2R{P6KDh7`)
zi1ON$%gUf*Ju;375dsDDA1uq1n&zyBO#~aA)N~OVhP}yvhi@+T-y|tF|6@}G`TJZ_
zN3e_Q_SuZDge8=ek_rkd^EUnfG~zr->^QK^-QxC&%73Yl!p7;p)hM`vUjEx=cqBD`
zBWk3=O(dXsGXkpGeIAIVl~Sc#6<!^c!eIO-BharLU!}&`JQ5|OT(wHgM&aE6J5(#5
zAMej%(`nB|9zULp_=H5_#DvFh>R(s*Zk~JpdRiI;fj<P$EwxR7CSi9Lc{Y_iZOvdO
z2-)qyxCJ*tX~LD+avK*-o4!#EpuLlmNC|rD&_sLoXZkhD-i1IL>{Co1A@Kg^^|CKH
zEC~SpgfU6Mvr7SkuUrxCyNxISox5#AztPMX3=XCY$#?Xr`&IBm$V0AUT&*3Jm%}47
zk(7W~lc4+i{Hva;vyjLBOJVu0X%{3Q$t<MJS8ra6vupLjyNf)VQ^`6Cw^q>Z@S6$=
z6dp4XuF>{V&$(gI_g#5rZ)UcEe&0N6!%h}Umn`ciy#*u@l8CnM?rwv1Y~PQ&$%~x^
zm+K1P@pZ5&@hgwuwzVc0Cq!2J6s?kEU&@uwkH$OJds^OXQM~cWXTz!f<Llr%-?Ef7
zJHZ<1*|DpUk;5?;<maCBcJ}Kb7~NP-xQW3L9d`p}fQTKq0npX=XSqv6N57rmgQ@a9
zD;&gXq=EYfVZ+kf<p4R>D*m&s5FP3DxrKGJJ6?{&;y+Hu^;gMCdcyfRI412w<vlx+
zn0jy+4}95@fFyT#f^lt7{;61UW0kM~nf0XOb?4C5N-!Oe2o#vB)VSD^V9Z)ZgR3OT
z8L#Mc)3>$d)oc1jh9Ce@4)mdL|6tT^IPRa>N8DNb<~Cw7nzA%{{wo|$bpSOanY=)E
z@vwU8UmR5Pzn3rcrqO5juP=?F@}-kV-^K~4;CBPMg24+Zr)4T9%DJ<6S{d?zn3s7F
zzi*d5w8Kw(aXBX#dn!vL#jHl^+3?*-+GZs&6XuJjwOPYJ7UBWtKcH+Ok#e&w{M9dx
zGv$l~fSV9{K7ZeoIWF088Q2d$8Mp4YWqD5|Zos+k+0ApBQhEHcHDhPQj$<iT`zR!(
zl$!%Nvg{yjzzc~)EKfK_Q<y0K8~bZUjW{zxRIuEFJDRd7ZDI^&iw}Fu7DbVOseP=#
z;;qMO*+9t?`ty7go8{9*gXi|i{D)L94DbqPZu$D$5IU<r{;F63LfKOB%ei;=?^-TL
zd<&t5g{DUe_~w`0{96syd1v?@Nd{xr+MqFBaqRxbBUPg}R;#4yZ-}MnN~nipyBERq
zWQ;<x5rDsp1KU?vW#N(a!f1q-l@wPp?WCo};KBw6gi|8#5cS2CaTmChX<~(jpna#%
zd7ocgHTsMwKl4MKOMjhDRxuF`9g=V%2iojTOl-773Dn3^G<TjbI4h?=PpL^pg;BQ>
z|Kggm)|h@oUZz*}!(U#YLMeSAmJbt4;aG)emyhU=NNJcBMp7GPjT523ngU9|-Rig5
zmD1baor57SMdM#DUYNAQ3gUSheH#Y_Bj4ebVygF046$d3wH<o9Ru9x7-hQNCUAWb>
zq}BM#?KXrz21&m2m+Bh_z=C0&O<q4XP|ui8ul>3AOZ|7<pdVEt3m$27CuG3I!U!&P
zXUXNyHuwg(G82kV2hG84Q}p*^<MtTIo4+jTY^pVDcVemjd79Y3JJMiT6EaPs=wO4_
z7hhmwb#Zx?gcqN5-r{dU%enAUSoXUbvKAhukFb0k?#|4Su_Ze@_UXGsgi)_=V}p_f
zfpXuj+NbOeBig3in6~K6xP)SdEm17tep+67725WvfG>fwaxFT#Ml5+5g>nuwu<D4b
zK!ltfl(r}3@ZQ1S^|OP5OQy-}WVR$2bLKU%gCde-pC8{53%4bEwbSsFWgpXH-8s5`
zk!^@pI*lm<7;;5>x!;+~QujfbHg9eGfuO+LYcRNx^1P+ZPPjVI<Xg**S~P$}uTA35
zs#IOTAI)!$-V=Ju3A_S4UsT@6aDz^LG+Bme2h&=+PySnaGVgnQT@vNzyFcv{oqIQ%
zyOHX%lM4HM&G}@yTyO30Hv4DC|3XT0E?Tx2esq>L>F+4qE~7UK)GtVBybBSLPcRJ|
z?2F4BV$-75%>mt$i&0^tK6e3Fw79{!hl~w+yPqZz4oTpd4~@?~8nu#Oeb!_AAC@HQ
z>oRp-(3?SHiPhy@|I-$CRn<2}(HU#fy5%xJg7NIW11HlICrBEs;Fe|m^@|)zZ)h4s
z23IHytC<x2S;VB|QWOay6;<QVBp0U_9xs^P(zc;vJ1a;g=Ou7_eAU@)Mo7<roSjEr
zJ#l)BNJ}VC8HqeFw}v==tQhH4`nDdVZ(ni9abzM7LMHxx|6~6o9!8<yS5@Y3|7MFR
zI-(4Vno`!Tz_wb~VOwF4<5ka_tFJQs(NuXmk%>tqR83Ejg!A_!U@*O8x>-K(+TlbC
za7jgpSCC@3dsYwFg1MF^NF@SXBgaUVQsM!xFaS;fu*%Dky>)AYrGUG(beE&#1x1_5
zloAvA{3Dz=!r`Cxf{3Unyc50nP$KD5-UGma#ARGb@HI0V&N(HbJnp1#|I=r_1Sc>=
z#Zf(t+G!z%lZ-g+#~d8NCc1vRQZ3&nw!R#)=~ThEl4&exb?0#;5yb%_Lu}N)c7%9h
zbxU{yD};Hv+TO2=U39;#2O1Vf+la{K&-qZZ0^{nb^R!Y;zzE3W1`RX{qFws)WdE-m
zEqWHV311H36I`;{@#<{l`ko>Sd0+W~tFUv#$4nQ4r#B<}N5o2p*uLR<k|=on)WWwe
zMGbCww-HqF_mq#a)Z6q}0}BK%<uJP`kVd5K5u!Ro^!9Rb+2~OZKrz%M)!#v(Bx26{
zK+ws}_8}?|DD#$o*UtMnyirOJRaIpt@<_};U`dQ#h^cnUc~;s`4TyTKD6u=MYjtN?
zo~mhe|5mr+&GuaUy7#gg;z!(R<JN6c*W>&n4yH)po?n_`I!bbV)`L}4=x9nP;94yK
zv5-&3psyRYncdv$45}19Qiis%OyQ<G#My6rQF>FD)+D=xK67}}zM@k%L;jqy)D@eh
zNRi@9pP+J#%%OP)iw2cftZP)=G31gAkDj}zFlf<jn?_e~52u@MK?VA+TFqHqU5h#b
z-a=c*8t<x`h8wzERrXdUVP;N;`6Ei|Ym}N0;}dWS2O^I7In9(44W+`MG?4?6v$Qke
ziqxdvYqYfXg=KjJyrqZ?o*7#&)AzF(y@9GK4O}mtDlHs5F_!0z!CrkXuinjui#NcX
zK~E`AAWta^ufo8PFo)0#;wlUep?w@F`v+(Zdt2(*@;-0dAA)04Xy0p#(G(Em)SNwU
z-G<lG>9w7uut-^csX+i6b+v&1PqQQp>1?8&-}y({Y<`>S>}J=35CGtjbpS<r)W@I@
zQ8-93NR7jlc+u*bo`ZkfJaR~B_6AA^_Qwbdo|`1n=w=#8aMQN2X4--k^Eff-2A{#X
zTa@bND4fVA=_r06G%K6g-jlwkLa0`P$>L~}<CT!}jvpsZ`C@H`SOUO$krO!ARu}!Q
zs=$w*9E~fI!;>2vS=aU4d_4@ee_|sq|M^37==*TfxRv7tR9zzirx=Gm7tM_>yT3I&
z!!=xfjAvf*14&i!E>nMsBEs_M);x8;Jk+E_H;oBsBE`9iPIq>hm=Dmf?8e4DrWYo9
zs{+WWXF~eDXQ>sjeJ&dHuGhlS6C{R}A8W5?n>l^OX={(&8VXwp<DcaV#-uc+aIE9Q
zpWmh27|-~m0{sp>+=K0!En+eiu%;5@j|cae_1}cc_ht=;OU5|7Ilh_$V?n>@)XGXC
zSGP#}PzTm|ew%UEYU2>u4+m*MmiHZNtj0gMx?_$1dL()(T9ljO6wu+j-PpK@50~Xd
z3{AvJVEyjh9ZV*%ffw+^JYi$DZ5G;9crB5jSw3r=gi{2OPrgbBvqGmXn$ZRqvT<CS
zg`~vDSt7<vSN@{m_!sq6VTj>34h#Us^e?Y$oun7U@ssbbf;<0TWS49R<uq}S341H=
zKb`xZ3GwRNX-K*kUO%J!m#hVqUfU6>>^JAa^<$d==94D0u5DR%A*7EE5J~z<g82+g
zL~cw;-(8aP$boP&EFoz&AsE5X;7{-gFjJfZtZEW1;^4}ZJdHGg$mHw%=g{!k{65Da
z-rrsO{{(X8_OLrq-3D>B%6KvE(m)aVcOwLs@Q(wtqqgFf{hOn)ffw#7{R@e<h2ZNc
z>>mnztDki=eW<s*C0sq+S1jNn)e{L{g>H1OfGOM}=0eal(+f{dz&pQN_-X_nNCziw
zU;q3k5CJ{@Tw$<D?sNCiXO(KY4`3c*xj*l&)$dpAgnFFjkXk37%qSaH{yp;b-u3vf
z>rs-Za6F~^2xX1<tX?0>O~=Af*5VFeov+|u8L9dxWY{^?#nZS>s6E7xB`Bj*finKL
zdRys#=##3PJDYS>(g)koV&IRt(Tn}z?~Za(7!ni)&bpzBv7l%fbK9}1W2=~cOaXDp
zO^k^=UG55S=qfaGy)7c@sD*C+M^lNRCHq=9`o-o-TyTU~*O)Ri-TDO*pxQ*IjOmL`
z=UvICV*whm`lW<cjg-_WzcObyvp2%9sUY3HrpX5w6KX>I)6if<MMdcIR*^7yIUr7b
zK3-_}H?3(<e|oJT8yw$Bs27LKz@(3G;@gu<DZR8(x{U1TE71LF3RLP`PC_C~zJbNx
z(ZiQEe|z=0nEJO37puBH-7Zm$Aj#$w9s+l^IETB@jqA3!Ru)blWVsD-!4ZT75rxcu
zJbI2HvMGh$_pYfuwsq1~QT!I@lP7rkS*&nY2l<uvWbUXc*wqWiGk;@>H=du}d^vPt
zeHnx^|C^Ubha&4XOJ3a)%SFNnB8d?gR1ncoyhEZcwf14aq+%<c1J{!(Ue@FzeR`5g
zmCGO7CtBT-PVytMnv6P8?Lp-3YxKD-zt=6dD;LhgC{<_&q_Z-HIbIr%Lrva5L2w}R
zV6yvxYp8W7GmuzZMRdrIE%z__L4zmYyq>MnWW0jeG-234kFz30$l!v-4@1NLcVl8(
z-`y$GovAn-+dO9A=86WcAzw~W*@$hZQS}@O{gfjq^1j&pAl>H<G!tdRX;O;Q!M8>U
zH#h9tKBa=IAG|mn^7sRej_Cz13|d<0k-vPJ^ZpCZUA@0Kd<cjYKA(g?I>DoQ)+)}h
zEB<;G@Eki4Gud+Kk^P+A{JRoY+|mc6o)U1VEeYoAUi1zI(ya{Rj4W5rU8y>wo>`-@
z)B3fuBv$7BBmDBElsz`5oJx)z9XHdd8u^cHFJRV<w|k&k-NcHMjn<XDTXrshSnK(n
z=`PYg_N#-N^S-A(XZ4(BS9gK@=*&DcEAM2|GqiAb%u#H!l-mQ|b&ieFN>G|(wO+iX
z0Q4~66$Olihh^iSlFuzlQ^_NJ>Dbk=(3=Su4cWx_(W$$)EC**{tIy%;qnyD|1!OJ*
zxaGP+32!v(Ft`~JP{4*J3REtHp{UmCnT;h9zKk_KEOt3njFJrRbOPlehD<zZ6f~;u
zDXGL#p4)s#eiu<UY&EpfsYWK3IJ=9AH&s?B=LC7lGzK$<Kw*XLV0Bn@F9|8IvT^Al
zIYR=tj2lZ*Z#p=C?j9!9@kB0>m&D!sS~9V$PD)=x>%Te1q)%Xw$bZKz!qN}pNZguK
zBSVD|UC&!ZKi^+cj;%yaDUaiknEo{enWq41Y35=f!dj1bY2A2e8ePU<s>H1-XLxjT
z3(J?n3+9yGOd<fPo)-@Milj{=lF%`sw;QJ=*!w5IM-^Xq)Ry`W01r#<TNnx&@ADEl
zH(ae5gHAjGd+fNi%D0Z%(}Jsp^hZp5mrpl2;J1IbxIIP!<!ovtz5kjg`s1H^vFugq
z)?T$qDITR%zw)_H<cWtWLxSCw_<P*$R<~wIgA7j6`*-X;n|dGahxq}EEN~Fhn(@C;
z>#UZl&Wr3WT#`wK!&&`HnXWK^m}?gN7HQ2PsLM@J?;R)6QC{`DpD@*BF@FQyj_7QE
ze!*sdS7{Q#$j7(xJh5ubv5`$?mNt!&t#{lsfzhOX%W38UbuGzC(W2GV=r5X268ll7
zNFRG=j@kh6c1Zw#xXp9%6|<Mae}Bc!?neU!+rEl;LO`*oH8iUe?h(ilO~Nx5pr%~Q
zl|o(2!NRXUhltzBH;k67bXSxV31IZcrvc@rYT_IaxGfvbf#u|mYxY5fVevL0Vdd4F
z!JKl9Ex?5Y-|QDP;61{*m6nfHlc0>@z|@Ncw5w$TrWz-gYz(lz2vq2mMLds^8JuaO
z_`-up_(GH$$nS$>`emm!_iT=5ZmU4nY$*O~cZVXStU_2u`)3BT)}407HR<>D(^k9~
z3Y(|({N9(VZC`h3usW3ocC#{vL6}y_vvM%%UNmrC`J|MJI#vrGX$q56E}VXMNuSCH
zEDnG9gLY^XH$6C_>z5M26;^G_CGVdM**7#0`Dmzuhc}`_(Q8mvu*X_sAPQKKY=har
zUI(b{UTILbR3IEAfbQ`wq<*nJRA;FQWGlMD)I(74yBq{l@?=u}J)Jv9UvsK&vT^I_
zLCgYS*Vu#$P|fn{tymJ+Z@jns?;UjU^OC@jQ5b%wz>nE3Ziv2b{-X}!=hPZ0exZtb
zI?&vMZ@FOMX?CT(-OqbdQoVQ&pXQ!W=7VZ>nZ_pHM$wG6=#IA|hiC2s=_n<PPUnWd
ztf%iIte-rrdv5VB^&K2A|GbR+*$b2s=wSy-OLxt_sf~@iLP!3uy$e#^)u$PzM{JI6
zSPoVQ=$Bj%8OzB}*UK<p2$J^zlMZ`2Bb(0mZ!kXVXc8bcJI+sw)y3Y9+EkvIQ&Plt
zcQ5M)FG8L%V5Wan>j(-toW}ue>Q&=2-_te;UE!6ERLmpFHj^fgb;r@y?rS^ruPdfJ
zb{jwbj@1R5bZ$VwQTvD7=(@;4S}=Piyz4!ECkN)yNhiLP!jp^w_FO$X2T%HuN3wp0
zTvrrs+Ah8QlT5c*wU1ETp+7lLXvFp(P{B3PtJHCxh0nlbIQn_B(*wfG`=SCN@I-)~
zmWEPgAmuhD^c73Sh`gCNH%3cTN#J#`NN=()m%yRe78yZop{G_A(ycq6YeMzEt9{tz
zRpdI6#<t;<nzwBxX?**`=vUh)eoG~c+QM?aUUN0CmY$+=ffU_FbkXorfB`sC+=@@i
zfoG=GV4l0M7p4Z$u_yp4fFU-$H`kqY3{U1Q%pJkAbO19(@UGg^4vIwLB#ACg+B+l5
zB!Ig5Z+qjri@yOISGqrqvpBe!_;dpKbmq&rs~Awk^OA(Y>d>o^QA7sYp6>%J-&w-E
z+@@NsmQI`KMdF)tPMmOZnGE1RP`bTq#NJo&%j?~H<qdf5M>E@N1HLb&s84qH|6MjT
z=>9hYh$R4=_#BNT-WvG2Cr>k`UJ5@P5zXnvPM3YSG+1;<$-_Tx)FI7#uzndh6nlSN
zeb&M9qifP8JyePNGAC5yK^24Is8;XE{UA+wJP((7BxVRTqAvgmc4XYb8X}V)OWzlT
z2n#G9x$m5@=OjGtPmmv0{<e-Gn<6O?pZ`9jpmg0?{fK{ndIH(#7tq)OW!DW7xkLhi
zd%zS8((g`Fu!$_F&>EhW$a1k@)8s?DBC%IdZ`}$hvd$>&bUH2nJzkqX7LA8FThU9o
zkyOqSvYmCgWQgUK#e|mAGpaM?Lrs>0!7|`DWmK@KJVgApc)MmdA}UX&p*7CxWlm9W
z8MwGICO}5mx;w0ep}x}ZFKw1imjU_oDm<0TfKKD}^DURM9ALZHjuFj3_-(R&IYbo;
zrbZVuu^`El!FEU)cMVQyG%eoFV0XCP`nJ(|?@@KI91Ikui|eyxWFIrw;GFIQvvg$I
zrC8#qxecD396F>KVk`an#hdQy&bU3JVBW5k#FPmrvaOx9iFS}rF8du8{V%Ku!J$z{
z$Ir(L?-+xWlVdd2ESZp)WqsrS?>7mGNV^=r-tSmRf+kV@)IKm`$8YD$vTtO$<Kr?$
zrzeCz4`?JR0=}FC9(A9aoli0Iezl%4I|;!Bs1rSFyPud3FqCoQ-lZ|kx#zuqVC{=6
zhoRJKQ_;sLvZfmYL673d`c5<;&tGd6^Cu}p>)<2}>?%#t{^5&Y%<E%ED6f<IUVGx`
zC8Opixh--zWu=|PGlYU2^+2!W3mtZD!|!$CVWqbN*yP?jemvPF4$WMv;nCa`TC`3V
zs)TyUSCh3s1ar>Iul`=cjGi~!pWMl)yc1N7QjX1>F6`3{78bmN3h}2iAav)@YiS@<
z=<F@0f@R_=Z1a<y7cuxxH<p}420QQcr_LkLe~pwoabkSaL>c+ywi6?sG;1TnS~ZCJ
zrcs+U9jFD{$Xjbhz|uSZ-21v}ZGSiz`-R}KalLaeetq+!hS@0V<jziW#guBK>&H{2
zvikQS0p%wK%=?Hwv_lVje^}tpe;LT2IQi_Rf9eJJUOO9)u)}Hcr)<uFG#37-+b}RE
zo}NmOX_g<Ymb<5YDCPAz?^D1;A(N2XD%&DrO#YNJw{r~SNGZhmblA}?Y#((jo>$In
z<I{tE1Hprr)fiCVFms`47hUNh(GZfD)L|5s$H17(<X9LH6jj)-Mw<7Lcl!o3)KNmI
zP8rEzwc<Z~P1Ny)qH}<XcB)!g4yNPDT!kD+frZ5M2Ph~M81E$HHrjRLo3{p}6U5yT
zRN3J*nq~YDCWj%*gk}%ek%Y)3`RU9M_lwU4?}eRKk?TH;pxP+D0hI4oApIe#D^!d)
zA%Np-rp)Gh41Rr{ckIQBCQHHO9RF@lqTL43`!#(#X*ih?2+t`L0jk{Su#H_rXi<c`
zxMBaSXGqmwGVWdaX*KdDL%!UMD-w_zqgj?hCFLpDBkK9d4qKc;a)4rya@xI_qg&bR
z5l=Pn*~2THA_0pib(<k0CK34L!*d_b_`o{paYG8f-S+j&d02_(YTd9Prx{vaxEMbE
z=`+C#mV45W*3p0;r{%mk^OmPhxlQ2@?=UEyR3iso2-I{^Sv-dyvzUBl!uMM`c8Ya?
zovD}^xi+w}d{^VuYa3M`OP7-voVBDBVdL_zAV6+sopKd(Z8rIXSSH{xQ%tjOqBkO@
zPoc*tdIL3f)~OPICQW6*it!w$Hiu)|6Klx-{o6*IRQsft{r=mEvS141)r-o|6E=s4
zyQ@UxzwhHtW6@#!z-W&C5Bj#VNDz~Ewl^~93+Kk4mADRn>Aso6D4>;xh6PKiB)7dn
zE~~QHWz>*pml&Ju^|Mm!8Ha$yUjraZ^Kr=;9oBdQfK53Cb0}(JRDD*l0~i&Ds4fVM
z4O}=m)!#<Ik8Lix#Z|f&<ld!}2Gp>kndl0e*KdKgR6RxJRQitZrdlGhpoHnNutZ)F
zKD!rV87rznQq=zBqP*E0aMYEv$Cu!>1e*zAFe-3?O_%*nqa)T`BpoLU#<+_si=kk9
z`~%@@G3n1h*C)JY8?kk%bPg_v7y^R`PX@COwCM{n9ARqR`*L`jHEf3bl9PH+3P>tG
zYFNfFske>cHt9q8`q&v$B!(U5UZv4zWM%gULMz7{K@|bukQW5#&8iqp3JvYI(FQf9
zUtQkr1CySq*v~7Q=uL{RyIolsRgDg?eAACn2!c4e475zrSg;t82?}=gl#{V%gUH_5
zd8;PVU)of$1?Y+0B1`za6>8fiEc^d>iBQx4kcHQO>Cg|5n!Aw$G@Du7`22#Cwp2{Z
zb&TO*JmAe`9>fJG*H$3@YhZEhnZ#Xm&+p#26UuEJ6Ma^QRwy0ZpYE>GuD#$PagzFo
zTMg`9pYQgh|F9U&TA3_{8<er!m6U6E;s|@)g)O<8ht@GfVNASBJmq)?G%OThz;(7|
zfu5*4ou;&8xhj<}=nMP}aW6aD7tkg0A<pVi_c@w&Cxhm>z!vxVhez~6Im_T%M4STw
zoK@Wiu(Ai*v?qV2h-6ZZ0mcS|dB>$nwtzapmD>7#2zTJb41Fu7o*7XVPiaJpt+aS>
z)LYEw#`59iYMPRs9S}5l7&Z4K-#w+-#awc}V2oX`Io<n!ua}(eC|@W`T?LhVQbk`0
z<9+2SIR+)3^<!KsSa@qo!OaL)w~R&N)cT&}m?D{Cwq4Thtz~LwOVZfVFoED`Lj^Nh
zL>WTK(R-k7SX;2iYN(J44U_{gmvBm$F9+++6MtX$7xy(vThXFm)84TJkmWMt7z%jv
zmFWo{dv)jSrU~hUlt@01(@FDG>l-n(N89`tYcwt$gi1}=s!hN=*4hE^T>~TvpW3<m
zEw5O%x3KSE8S(Wqn>k{j;i9qwGAc<U%BHCBQGU-nLv-d3HfV*C6R&0tqCPfKHr!3B
zVWNh282J@SgZa>NqHA3JK!<*UHqCEa(r42%aX%(BPe!yTod+n`0t(nMD7j-KfrT~u
zH!Vd@L)07)M5haUW-+eVs#eA#KGYop{Ps0iv$Iz}$1bh%9~hWI#BVnz_`#SD5+3Y;
zh5P;=@dtpk|94y=?D$FwXUt*gAw>PY%eS$erkw2aEK1zNdFH|Lp3TaD(49}g|AFaB
z^UDAo0`|c})FNYWo(d!Qs%pqlU7*54uRw*fFOxokkqr}jo>e3sCREelBfk`D+04Rm
z#xC6!3GWtpo*>@Go<{dKEkiT`wWlnpAQ8hbogW{qJuVNm%zon4cEelSH*9BAO^C1`
zIXvk?y_6rnnj=cu7Zx4F_;PxwQ%%FO-kRcNP?Gmd4y7;w6)89xNo(2oGz18vUR~}<
zCIaPlS*d1*Y(~Tkw^(92(;vtuTsS;@B5F3j%W+5&4lO5d(w*O@8o+iN!~;HwSUsWD
z-g90-z2(JVY}+?$D5!9Wz?Qst4s|uHZB#pMiHtr#XAh6ysoV#hy;b0tr!cP;T<u7G
z?QUC2V|dc!_c~P}Iyob-kUp*CC5^k0TUdF--M`iXY_mD2JhpAUl#Po(h))p6&vxQd
z0XB>D^-saWtPt%{RPhIKsvf=50uxJbgyCp!W@0CG_<l^>?4?6M8J)X|qj%hSeo|Ki
z(|>n_B$xu2FDP+4f8(w~3xLqJC2l7{Vs@PRCl%j$fIpYN#x6Sv26*(-P<D9(qf_K8
zASb{=lhZVnmolLxOrqlIyy#s%EXBf8$tj|tS;99Db8d8H{WUIeJpjb8v01i9UL%L#
zA}KH2jJ^V4jRtDx-+RB%52ZcFpURAWjm8q7uw|BEod>jKJH4xG%7<TgjqF!{sORO<
z`?3#=9F6loRrB_?YeL>`nWGETj%|wJU!*h5!9v=Jm45HUh=TFVW6wMhC<3#x`+!hK
z<(|FP*LQ5|K62WC_s!V@>7x!I_i`KL)4eX<#5hP}GPXjig6083AA>egZi6FKJYO&t
z(1}3Q#;Q@(*ue8;2PGc?jN^{5PSxc=7E&;LR|sou)mp0y)sJ<@s-lJTcM1CYf`2u#
zeqqe8$q*R-0%$kDjy-SF1f-i}LRuS0EPFUzqAlJdqT#^)0k3ZYlKs=w7+U6^=ZnY*
z&`ks1OO#TVOqDo5T5@lhM@*vejo2w<eQ^nmC_`ViqUXq-rsJhVPXr>W1LXliz)G%t
za0)750HC-E?TKd*I*NxJm5eybzU)+<jkQ4Y!p3ZPUniRyeT-zAa%^MtRMxQ7f>E6%
z>Fyi=eGI>0wD^tKxi&74A*8V<Bmx@~hP=DXn=JDg*qi8q(3ulWbScejR-A^a$)fOa
zcpb5GyPB5_rIpbt(b`U!(fkC@7H2I6z^r;oAK~BUzv#&=TorLK%p>q37~=~G7aRbc
zNf^xLAtq>^*;V`|foH*xa_@s><J%HOu$0MENBzhNQ;bM}SWs^NLnMJ>Uc(8pJI}>Z
zCH1#j3;Nev3w8O4PIjyn{!t_Tr1ref*^hrx(OqmWt|Pj|p5v$QB4J)}2m;^igoTHE
zFXyQNT@9Z=cI6`@Bm{D*zG4#oY_|G8CinLUn#)<G|EiViF3*eVfQNPjw3YkCOhx-8
z)o_E`GnDisprc;8T4rCm8BnsX-uWzp6juJrimB(}{D5Dpc<WxCak#XyH3j`w?)A31
zZ9VdEpR$=MMTT>g4wVYzZ-l7X>URy?9rq*+7YuVb6!p0$GR6>8p3ez|ogZa=+uMj0
zeMUgxm6e=+W#92!4@=ntpoUzi;ZgpnafRtmw(%HK=(%Z#c+DfIJ1Fuvut(InJ%-Eo
z&ViMg@8CO=I-|2f+qp^8n!Tf?qB6}iej<K)aaM@QPFjZjhE>EMYgjRiPb&lT$JprK
zFV#b`^72aqI0)X+;m&<AEYWNd4G|Z#BcXX$8}bx*gTy}q_D3?ABWGdH0roPTK1|-x
z1Dljvr$+ojK80?U^6+wJj#RrNt{GnrMD>Z?aJ)bAAvhgl2o<QFHL11DqeBBigs;Ux
zh=H0OtO?|J$pmx4xokG6z$bDV>HWk&I>Es;i&H^@3C`)57V%Tdun!Mz!oZ7byz^Y8
zU2z1lgR+kCN+<Q@T9o#pVIw^&e*`0T7K#r5ZGb*Lbniki0A^P51z_cD>cL+4av*b3
za-d*8<XZ%%%-$PwDi+8KG=-FxI7as{{_LLdF0%Bm?O`>VbpYlWkaP;7mV-*m<71%s
z^!|!0<|Uer(X~<7X!ni>Z(LXQTLza<L^!c7h{Chx2Jdgrr0M-~*XPC0hXO_kl~be+
zP4O0h90|9UvW(~?R+@}A4F?iE7TXKUrl&?p?CP8_uI#Np{sZYd0!ntS(yLK78vR+h
zEXT#x0UJ85oP4(LM{sOi#NglfiFh(ihDSCVXGxAzoL)U9ggjZgIAWK$Nlst7mxcJr
zBpkI~{D_VHyk|l}N*Z5i(6HePVj3P~S&siVC{^rTc$|EAf%<JjJZgG07KE0?{4a9o
z+Nalhc4f%ouV$0%*!W}G4S3XnMzKGo+hSTL$e|tJN{ZYh2WfQSaV|WDWt*eK$O1q4
ze|<0meMq22+o(3)qIBQUY(yeKB4@8pGUp~$_SRR(@fC~gV^0DTYzF<YR4V4xX0rtt
zi=>iFZ;vvS>>Yf2S$xd{6XCqz)12Um0b&vL<fl&sQp_rgAv6NzoFET`S1{S4*cj;l
zdpUIU1cciwfy<7y6C}k5t~gGeQiA}3b%o%OMVr(45BMz7-|&OP)ZwO3R2#7Z8*31&
z0u<c)<vsg#(n`g=d<!bSEPl_;XFSQ-8lYxS5A|#_EZin>qv@|xL>L<GwmVD{hH{3s
zcdD{-L6*mTaz{SZAgA$JeAleAjP9bY(!`1S9$ZC$<t7zS@>8uA<j`c7OQp2wYFY9l
zpauJAiLzPdNuQ#%hhZ^Je`Bd!J=?js)mbS5fA}l4sR6>NCmR#B?T$m=l_O+T?UN;%
z6jJV8>xgsPdU-IVN(NX-8$CGJ|M*AZcg=GdY_?6I0pe{ZQvtZhadp<jsgxGqwA2O^
z=aoJqWjo3b$7B0yh_<`~+}k)r<T@QW>jV|#(U+~$6-OMo4Q;YaJ@rfcBeA#FPPa-1
zz6o~s={*jCLG=r4UBXu()i$n@0-sC8Mq*(qBTg5_;fFUHBV#1K#ZO>HPa3WB`h`7j
z2oEw??i*S6DmMR5GF45ycP`M4GpL;9mnXelBAyk&8A&0tjVTtA$4bjZUChl3FA$FS
z7%LV)i<^KyXV_$ENKE8F*`1}Df<;59C;XKoLK@zi8E{I2)I{ARBFI81KW_5ArT}fB
zic!w8N_=sgdOVtB0yz<5JDO@$jR5}W4_2sg0m7WVCaTM{s!qv?jh!JftqkP>VnEv1
z#sNkQs5GQh%<{}Wh;Q*Bb4nPOa+&|w!7iW!i+()*h)WW5UH+9EJnmf>_P2FsE1aCa
zqc~OJ4zy=KunAND-2f4+EKYfT9-7)@aj`(8Ix!3O#1kF3yLsur#@*Xe^-!MNGWbxs
z?_Y~8<m2bBxR<#i%}P81fwvbvfF8XOjle=EKS?sE*9N;5Aihq=5ByZ1CvD{VrH`-Y
z>hqi#%j3LMOqe5*Rm^bTj=+`$8Z4yW@Y(w7HB;m7(!c<MZkE|n;9yiI{h2-ZNVKiw
zZhM1jo}xX2Rf$PqWzWW;9df=MBW@+*M)D?zP$YvD&6p85de#5E5UP64pgvZ%K#ul%
z-EnAhXHE~?y<?5hEaj(-7~Bcz*s8y_sh1i|Ve&+T`I9@iQjaM1AXHejiurBod!i6Q
zIOimCT%mv<Tp8U0NhtMmQ~=xp%x}Nyl+j%;_?$8EHzTq#`axp-)!Zi`&#N#B8TWNH
zQaSj{Uer#ssHu!`lc@thoK>e*!8o#a(YbgBhBcs}KDoDWQ-Ow~qe}=mUOi?{f%Jey
zpB~P7ZDu8#nyaR2IHQ`4qrv9eO8_`sK6(RNfaPY5_~$WkdV2a5ocL1dwo~c<qvAip
z$(tRe=U%y=4JIvL?fnD11B<4u|2FIAg;|ukJT^TD%k^9t7Vyu!{!a~Xb2Qw4aLz;r
zGe;(2L?M7fPtMpb`QgpS!I_>yoWO&dQi=&OnVLysJ;yqKLqKW?D99{6=12e`>qf8s
z79sCMV87P8F$Q<3YXSnhjl}}?7k7R%+!+w2=$)26a{WIv5gf8H-BwIAy$>?o3+QGE
zj0HBtz?lP5=w@tU9|yIC)>vwpb%o@tVnDb;WF%AG;V}eA6!l@}HjS-!WeX90SFW??
z40cG$>s5Af?&md0_{mT?jU)X84o~YR1`>H8m;Er%akUaO%Xq=0s5%tBUEETfSInL)
zGiG5$X;QO+0`fDUI3D>b^y4DHQ8+<#ZCTQ9EYUFX$7C!Rc-sD(3idkQ@Qo}EXLDJK
zgrz|~otk+ixMGhJhSDa8yHs8MBw6@Z71*#7F_{3wO?H4q<dRx`n<3^<#l;ouK!8(3
zSHegogKehM5EqN`bd8nT?xPwhlCZ3JYYF)TA$s;?WBN0<q12fJPjA;BU9hl5i%J1h
z0{pb}^n3{_JBk5p096oPwX4h$qT=-bh%^I{DDh7L#sN|w##H;pqsq#AbcVSv3jpV_
zf58NXK;esGmqo|Iw%FSud}XunLyraQhoesxm*+G*gve|_lWzwT?lAZ2e7I_p!}wYA
zG_8d0_?6qx>A?+N;|<HxCwC{0Me1R>OHbQY8q$Sv7bNmVsR=jUa7<>I04*fnn5Z4X
zePjYk(|^2hi*lPv<;riH7=Yq+bpvuZ3Yj1oglVKGoh*B|5xwM3nE>*4UyGOTb}~A>
z`^;M@8`3~K1d#(Pzk1ZVdU5H*5}<EKHGu322%N<~c(nae!(HX1Qs4kuS)F|>K^E-G
zjDWVrSxUr_8Ipc4DF-qXRsRnuNcY^@l1qyaP5TN&=M?f`)fiG5GrL#1e|9Ixf;$}~
z5fp=Os-P-EguwEtpknWw41C9J<406jUaqD4?3mb$3KH-YS2^bmrw~B56QOi|Y}XbX
z0lSF=CF*GJj*UyV_rV8Z9Sx5=?d+)#CmTO6kLcSlAPt02tX@&7Bw*7J4seA=Z-ycK
zf0@6dOUoFv5|m=90G%_{!q$yYDIt5&xoQ~fiF^pZB_G$+yVlo!WdnoBHf^kf=rd=u
z(m`X^B&Zg3%uu8YrsQ;H?Vja>?BO&ZGzX%><iJO3QOT$8e91V9e`Wl$;vK7b&0#Up
zSQ4hgYHX%sT8bPF1mUaB0c}|6tuX~-PGVLd1gOolh{ycT*zC>=dHm<Unzw>Mv@G?&
zsztf9?vmT=L73<<lSL8XwWx`^Px<3l;Ip*FC5QVL8Aljfvy3w^Svy8sv4{f=gxiS$
zj06|}MN-5CO3Kj31M{RO(W&no#GSUQKLFQ|=x&pmQ$x#A(PkL)(!g;BvfLU-5dzUH
z<c%$$JzwE3xnBH#EhogURy*3@D%oI#^w7OW<GSG9POIXKy6|HNqnx)7VyMWjZe~Bq
zUOpNWQKJNV*FtH8;VOYK(nm}rZ83i$!b`RhI~VZSdgyGlcA@1q&Tg76tW3)Lv&y>v
zbxo}*u9_lEG-bu01c5g3Pl2`j8!z?i-2d2)D1l}bgxx81vALh_pp?arW_M{tAn23O
zQ2=qZ9JzWjuAh4kYp+xz`#C0_0R3Qv=^@q=3kQi;Zo~K=V<`^|isPv1Kxh`obAU=?
z%cdw!VFleok@T<{a{+azpLEn1%?&h+oLc6M^z5H9%iqwCB*}wsL2{Tvo*kHivmL~l
zir|&bb$ra^t=^p<q@vtL1~=Uow4<#QO^He)<j2OpZAlF0=S4g#;f0_)-Y&;zD{E>I
zUjEaj9OccG?x|R06N$AwPy|wex&UFTJJ$Zn1k5kU@T`BaA<u@`lX*a1=`hT2xkod@
z{YuS-RO@~%P|@y@x|SZ7Z+VFkz0f+!sBV)Dli9gpGt&`MhyJ7K1$XBF1m6v1n2B8N
zqlNLO;B<Zr*+pkk^Ma6BWG=#p9Ge9TEv;IPO>uiBXLmh-E7H3w|A<C0{?mohdJE+d
zUlpSQT8qx02FXDd1Bl`W`kQichiT*RxiU78WOTa>cDdKFOQIUMIZX>IvCO27`71|@
z)6A#YubGc4gq)G(5<kWow?4cYxSgl5J((NX$M)Glp{SO~{`s&iL#A8im>>UuF3{#%
z`RIAMe;}Kqh(d<K?9jaX2W$s_0FCj0S@rH%X4I~;7PVb57)oI{Q35j06mW~N-szaM
zLQM(a$cYX~-=~Nwr_Kp4^TkwI)tr&B4cBSTXtIA&$h-~AD*vwOp!pF7@W1^SC6Y)e
zDNEG0I~I~sXNegOX&z-@AuaIFrRU2C+8Gg}IilQA7+35nP(Pu+s8F+n@%h>x`!U>z
zDgOSHVud+kd$HZb227M7C^24zma?@)IjRF!#&}ywSD46g=5#wIIOS8hO@2Qm_V75=
zS)wonKO`er8&e_$=(0e3g2P=JYah^k*wjj$3C3fxsZTaGZ8_dqN~g4?rwhp8K2^6}
zSAY;QAD;Kc!9Wm5S$t6rPwSJ!Nmr!DncyB~+DQG+$UjWg!@PPxaT?^gp%o76FCh-Q
zAWTmyg!W=*cW%I?GTVd{RVQaK2C7M9P}sA#NPm~x^+*JrAh@8^rcMepLBIsjD)4HJ
zR4TiGU;~gJC<kwFSXXs%23MQJ@LylunoE5dv%(}K{48wTx@Tzpm~Eu<={3(A?2HI1
z(HE;~QCB~d->{<iL{W;lLDmS^eY19KkG9i^e%{OrsY_Xug2(EkcSkT`&f*?fQOcvs
zng|)FAWufMlXVhFu7O3&Nk{P7XAp@2*Jw~JSgnu;?@!-`49by3Ir(usgOvQnt3@B9
z3}AH@Swiq*f}lCDdx9tcFf%oSpWVV%6pDyJVDgB!akdwYz!ackfu=Y_M0`>C>KFRW
zo@<Zy5qZmwx(9G5KPFp&JQ$&?KBapelk3aXb6LApeY?R5;o2^vB^p@{z=+u&8i&Uk
zSUcJl_L4ezU|v5}NvK*nB|v6dgGuxoHuD$qVE_9kpD?rU*|OsdOng;Becx=>Uon$o
zq%ma7+r~E9HAWQ_-ws#v&o`@Wl5Q1UDk&X?!8Pf{@6pu(MuUoYE<6kn?!<F}1=K(5
z?cV;SH)0@sK_%<^)M<<qaZ|}rNFrC>1uJAMJ^`AOmUTv=SNL~({Z%wAj(b=^Nr!P)
z{+5L9Jn8r8Q&!IOsS<^dUc{8T8Fr4Xku^Pn6?l+et|fuKH#-iH$q|}z(yFD`teoM1
z+gkMTNy$J|&Ypw4OLEb$)rLG)>l){*&v3P6$}IgKFah7km4P@j2=g*{RMkUr$k#lv
zN+>z>AR=dAK@TVfMt*8c7E&Dz7|HXmS|Oik#T{IutctgyS!lmKX}^qZ{tRBla?^No
zLY?g6^>r2;C}9RA!0`z~*?<)oT1@b18ok>g=V*Y+8WqI+ea*o~$?RxVii5iQORFgZ
z2?BW@!>Wa&ZTb5*s46Zl*ltCo2A7UtM$*injhds%WyQl%OKuTaRCWgr4f(zd$PB$0
z%O=1lgfDwheXenM(in7Zqi^u9Rt&sie2-FaS}r~-f%p?Jc1MFUg9>EQ5BHUG??VcE
zy{E1J&q!=he+@M-c8m7|6o&9a%4t^C@I>#!PW*If?IW-R#8M!X5<CIn6A&E@@si*?
zl>y?xWcfTLbXD#@%%Dv**6=JFQsfxurLelBbK<-)k;4rqMV8AmB%X8%i4$K}Ao4&G
zO9oTQfAx}k7vj!eDXDxkIC^^tu8koC)TlA#Y&z_H=|pR+<|US@oLr_7?P~8@-me?i
zO9hw^bOrDLya-%&O+|(mWlW5!2UEm=ywriqQLAJp4rC#(>OKp*I-sKfhus~Uxi<gi
z&G;VbFso#2S8_^p$Ko)mn{P*UY?xbg0-a*k*FkxC_IGoBjJTc~sO}yGIS4~kv5ekO
zT%zTDnb1A^H>^4}n%1zI5TNh=J5G%LVMTXJGC6qhU<1n~%N)>xn**3~xS3Z~!TDjm
z!wff{LAI#U_<LtO229XDktuvCq?`!Y`O`@nRR#~9|D7k7N@8^Q2^?xVLu@`-Z+ZfL
zxC4weA3?swXfRLsg^t~AaDy&Gv|>LLmE{2KR)NWGdHb2IK@Dd<6lb>M^qqhDtmWz1
zKBhh++9zSzO3w7+KKGw%D2dLgVfn@E8g<Am!Vta*AQPbB`2&=eDy%Nx1L>0OS{<u~
zgLFHKVyolhQYo;+@Z?yr9+}|D;_>J(gP0>Zbz-r^#v1<Rc_2Av)QCaF0C2VU(W+26
zzLO)p)GQl8=y4^4#^NQoJ3CwdNvx1RA|^f7h>`QRAj@hz$15_ysbxtg@!>Znm&Gmi
zKQFqm4qL2KT8d998A5M|Ir#USd(xWPUkg=rQ&|)mBtOF28U;^oc^)>GzE@VZJ-Ub=
z3U=IpSPsD54`=J2K=mlzl3(<g#sIO|Teh_xh6U|5W5&%4ol^@Ay{=w^E3AekhNCWG
zbzUY!9xLC5bcjI4A{akInaNfO&_H^Z5Mf3;NTSn2m+;i@O#Yl&pKA?{1N}Tc<7VKJ
zdav{&?eFvaEOSC0OufIndQURd(|2#g{vv1pqg@RyV>Zgd?k22b=FnpYSkpScK1Yue
zKN8!?wikxL(`gJuQddSa;{dqL;w&OwdEO1uBG|H3!ixI<fs`OGo?RT#X{;6Z8^GU?
zh)|L2tna`}HZdLJe&U)MLmXe-iE<i)&M`@nT^h;^bTKso(E`x{{!kf%U&9<~d8G5Z
z9=gIL<3vHGFYhPP)1RGZbtm7wozr_QmRt-Y;&_({QZwcII$w2$Q4a+<!DI~U597_B
zX_nBj<4S>^ct1IV{2XCjL34jAKbFsJ#QVvu`&l|$c#joCp1!UdIFp77Bg<dDr&sMR
zO9|bLWJ`w1Y?llIiGRKttTVLLUgLE)7Qf(Yy=1MD8PWe@mi@0(WCeC$zn!}AP36fj
z?}V}R%&6oZjg=6>YkiQ51SaH-*ZT7a<BwUHI4tjFQtUay7ji~26?ra$l`2j@T`(#S
z?A%_ys5rU+R$o2gI|40~N8{(i_P#|!!^BqOKIWanpw+@--}0Vj5o!bfV-@y$MNYI^
z@+DWF1fIL1!;AkbK1%xh>tD!8wwhuB(gH<({dyg%SIr9-N$!el$-yx^hT(?hSn%*a
zZa+jJ(pBOR8u!G5CQdp{61crksx0@!I~*|$I&FHhfiI@TQTb_e`;`P&m=ZPGS*BmL
zA{fJAY;oVV#3?0aCpEwQ33BK*#BQULBaA7hWLOR*6f>|y3e^0%?M%ZSdliA4E}ml6
z!)8{YRc`;_7z|?FWA%8UL+l_8UWS^mOeMjwD&C|lUQFE-f|{Ua`Fx_!XPPirm!IJK
z(gocgO1x^TT_6G;qj;|Z9R`*UhFNLp%gXnnydnLnU4>i0Ee#8Z4M`y3U34yyOlX&?
zAUPR7c)vH{5h;%y#jj<IJyrZ&le;f&6cU`CKbpT=(hh<KfFM&`<K6GSyx^BeXHJ{9
zigo{3)FUe4x2IzpM&2dv0G#y4yil01&SN6N96CTgI+8!*OW+~T|39A2GA_zB-1@o&
zl<w{hkwzL+y1PRfq+#gp?v4SZC8fJVh6a(6&H<5bCEnZfJMa1E7dD%j=ee)@T5J86
zZ8{4ycLL%S00S|TV&cVY{#Q_&of*w6RlEiRx6p1L85MkhPXm$&xN@tjgT2_+`D0*F
zx05Ml`*Xg8@Ra`bcT#)S8_rz^TYot5^?{KY<F9C!GYr_Yv#J+;f1f}1P;74oQs@cl
zn@xTNU=~EmU*6|1<j|>FBD?-?g#$z_f_1EwKK>--(C5iwUrL4icfSfF1>GME6I@`e
zQpJ1T&y;4dB22R7%~{0M7lX^@aKpUJA|P&72H!@)CftkRZ||3ajQfq@ulcr5#_hdY
zNVpxyxW3B=6f+LlyVH^v{ruZb8=M%XZ}maIuiE8CElXNCz-zzrpAn$N|FGpeO4&D9
zuB4ybW5s#=KVPqh$&>rZ|DRuJT_gK<pxhe-G4e@>Zuu^K(3ZSmRpR>gejJz7KA3;A
z-#2QYdsw}t10;Hnq3oMO%yH0>xq;{Qd^-hcDBDbyUl}h6CxUJRbd$4Kh*Is+<bA}!
zPKd&&ngavaX7j_6C7~8yxImt8GilbXz33GP>@<sg&RYPWAs|rzpd}Zwo#E~qD^gZX
z2za~ClHD4Jkhx6Xe1V5r+II-;0k%O?=lhxOQR>gF+(Hn<lbCb_Wf=sOqv-&UHxn$2
zV{19&n17PYO1|JjyStow%YJ-))*q@jxp5qZ<0x!MNUCl-2ng)S)y@u5#_zT?e2<z*
z6N}l0<&pbtj^_9IBN~$UGTE1^ez0C6Jx}=Bau$*al#vQ+@Z)uI6OV7aDL_V*YlLI<
zSL~;b!bp2Og;LbL!L!>ka?b!R)K)v&@pM)+&3$gi7U|d!B5~X#HEw!TVo#H@VWkD!
zZ8)?~b-ipiTmc_V>o8tM<`pi;R2pTvOcUyU_Q!)vV}cQrb<WPHLu*J4-?>%=K|#<e
zQsH|BI^#3vWR#I0Qs8SHPY)Gu6T&geet0C(O((wN%%(X5gY`!*(-aVM0o?@yRC~r7
z06(_oRQL@%v>DBR!A;H9cjHb(uPCs9@IpRewKQ+`?igIuLK5;efI*-i69*$g)KOr$
zH$m+&W8FWf=Ej2P9S<N7O=5QK?tCQqUo64nxfLJLOHKiK$&4^meno6=;FQ^p3<-Kk
zjAy@z(2R~Kk3VDE%oaiiEQCNZ=F{JjJlaPA(1P_Fev)W5KO#mNaUJbU7CqJlvb|9T
zdVR*Pl{;@tcwU^1JGKnlHA?-y8vYNDyrlS3c=>2%>8RY$USAql1APRv8~_Xj$o_z8
z1li<iJgP8NqFL9l2LBC7fV^-Wd6$BRrfG>KF<i75F<<c|d62P&5-bKl0fnl0)Wm$I
z304Z<KeApLG@eD?{1ti;>wwAnFzWV5kcRn%5~0H66d^!YEnq}|jSSGlWAG#B@V{kb
z?2>@%WdWr#ApV+&f$zYn8@yBiQ02o`|Lq@=U^<nkT{klvpAW);M*nUL1WMav*Z7lL
zn_1+6Sn5(KU|P2Yl3y@j5Z6TlQ3)5GORpV`-1;KQoEeZ^26ztA32fPPHGqu~$n7k5
zaqF2X``Qw@HS)!?6lSYES}fDqI4i9>c%qycSz}~HR;LH?ymZaw30cikwxXNXZwKgX
z4_J;M4)M#7XT+IoBp}ZO0$eHkfSPdJO1*&b(le+Ry?WjS1Q70-BxmXFIkt}oOTl`u
z1-$&hg1P(go(Eqa4kjXyZx=o7rzk5WTf5lc0-M7GBY|Q*l&m+I!d-Lp^dMw#4l75J
zcY7(X?&;7V??J|;kAqaqA*(Cq(&o)U6&d!rQv@f>mVcy}efeKDEvs;dDpJR>eVjni
zs09>EO;DV4d!#cux`CtVAG0Igk`@r1Rj0B-Q#4Xf{5^O7@2+<8uy*nQ5qmm$dg%E-
z1uh|66wq+N%LFtu05UE3G+iYF)f_3>K!F7K$FV-^#iDl%80$NC6>?MJ$Xb>QeR$$=
zv{>H8m)9>*#H>hY;uHntP2Jsb-LPkP{<h4SxPGOG-B}fsu@S}@*V(5uMXRX(R*&Mu
z%Q#PY-Oi+(CBos|jEbjkQ3zr5nbi;{XeC)c>;2EXFw(Z2BENmD)HJI1{AyliFd@y(
zToUK#m>*yzcywt}ut=*}snR9Yg^;Mnk_R=6o=GMkQ6yBiEb^0JmoP&tasZ1Cz^W|4
zXy*#<fV6nT|I*`2Fz_y|HhJekMqw2l|Bwlu9v?$hvZ&e0Ff*o9yBG2)UJjKYH1?lf
zyH1@o9;?zk<i`b4uJhnbuXHZf+Q;KqeAcuky}Yb?82>|q;KI@%D|qPs*=@2k$a7gk
zy1Iw2La0@<Ri}u@wOMWA--B@TAFT*+?Z_lvNgP^g_3a;xPMpWCB7)m0y>bH*PK&~Q
zJCK1$x{!vHP$E8sR)2oFXmvD2lTvFgb)KHjGU!{ONAbhG37M~vXQi{LjbdNqcT(eM
z)KH=m|G{o&<<x8h#+-<Uz}@T#eg!Aj%~F5guA!&f^pyMGj0?;{B<l}9@%=CP^0s|G
zwv3D2t;TeM(L89EYyIk{Mfc`C&(oiQ1%MS@W7_edFao*mwY>}qeXC&idx-W>>Mbee
z8!%u(p!t8tp1<qq{C8>VxkF9B9cD(=v^pPWA*x{5dPN`Q+P^=mNz5L}PZCN~sF3ny
zHx#L5E0Xi8#h1Z4SYNj@zi|UMgDiJNJnRgEf$9=GkgG<$PRT*`6)5=9X5Dt*Vaq#j
z(t??xmvhjd^4AEe<L=tRwqO<@!$82*Oddq_0hV@y27m`JZ`=F)8s{j9!vXyP@JW))
zTfO#O_%BbzlT>#0X?%Rje^HGE+m^v_KwkSGxDT`v9I5tZ9sCzx<J55k_@6l6>*xlj
z&0$=4lHb1LWe+?P!^s`4eE9+cNi@E|1?-!IETjKznRs5cS``ce0Ht}dFLV(o4*2?J
z`@`VgW7c(0IEn*uNI-`i>_zo*=tv;^2F)U9#o>!2ju)R_rUmKJp$aAg9!V+H(2$>3
zX3c^Wvgzl&E@47&c^(u7Y4LnzBinkQ-5ZzqqSa8|_SfEm6?YIJXBuCk`6(?YZ~H0<
zXscloY;=ycum-bza5Y$sBdUw!`2#znftZWH91H^gkaYL2v%%78Jnf<2Ge4dJt6_l<
z&iUsI+`KJ15w`mlA2aW5l}`EOR9tnh_&*daLS`=E-}s=8r**Ftp-~4`HP*8vh`{Wy
zeQ%KT`tOSxh58L;MYz{k7V~%3?O8DI3^ow}gG^z$vvZGh0N&Z8&mPEfFf>lTA3vH0
z7nFV9u7i5LDOC>5c}9(BU{>Expq(7YV-gT_PIb96?iE3`XcnXeNR`NA{w<(XZF7L)
zO2BLZwnb--S41zd2`ytf?-W8XL6ZyZcE0~b?$^<OEe}1AXI{H}XKmGN5g~yKHCnf@
zH1SP}G^$8$K&}A;p_lybW$HkHuS*V-oY&vnS)ypJ&00wSq?a)VwG!0tsu#f_A?npM
z!Xa}D^n7Mr63QMInM?+%5#Tg!)bSuonza!1s1)zRD-FhtBV)I_Va`GsU;@-v%N{sw
zlA621q=FDra=-qRvRMU}el90DpLvl$vGFI|-{Sv+Oal{IzYId=VMT1H1hiD<-6Y<O
zg;C-LBeQ_vX(OKQp1aQ=u$K!N207_g2?Ec(bK$0nI~tJd+-*ZCQJJm}%7O1<#uYK2
zT~YG8qZJ+~iE`!vzK^WgSwxV(li_miui_W9UjX1qL$LL>&<6&CI)kY?12-5KP&W}2
zG~UO?GOR|W%z6=w%@fkO`JMsH5D3(tYm((=6?#E!y)IN<H7hF8LBWiqp^JB<yau@5
zwR$ZhIl)7`+V!Dq%1KKH-he@errG*akoLJAMFi&@bed=qPZ4~AueMS%+`Z8rbYf?0
z0tBgi@dg3EY`~|Erg!CoI=|({@Q!QWJ2J`Og4R`R$_%=H`}e2YHpNnTB#@eGo5NAn
zje|}Wke;8~#C^Q|Ex7J#l0KI`Aoe)Le%9kRE71MOC_2HU;JJ{?eLtO}4D_$3v#2**
zvhP}Y;G6i5@OAH?sTDhaU8KaU;BQYC<nUIXf_~F0_Lu*wq6OT8rT4viLN_~C^A)nU
zMZ56~>_Ec>2~l2BTVk!5>u$Z5aA;0xm{tz4{fjkn)r`%AN!^x8vm3DBPuX9A*u*fV
z8E8wfBh`S=+!cwSv~~*ZxLYXz4V#|00)c+`a}wVdxT}I89<IgT1m0HSX;+cF<zLQM
z(?ghH^IGb7Wa1_j#whAQ+ftj8Qb(*>y29O)V^}6nfVZ@}c4En=frXPlv{u7er*g@&
z?pZMqCi+xb=O~<~G5JN(;DZ|R%ST|ppJD;XODSS|{{<JohWAsG*aPQ;L=s3N*ovr3
zOY-0oWtC(Y;x)fJfM4z`om+k9O<WLV-4aPrISz#WJ*DTKdxTZ9H=3}?F>nMiPJl9&
zBS5XoV$OtSj}(oe0Lo`GsWZHjgKEvD&2<<XCDGD}|8(il2bfMeHzk1gdF)F)h@Kb%
z#&_p|d2Uwn6=Dx<i#{7bDD}tSOit`1C0Z-jFMXxN096g%lT?b__e=xmVP50k@lb%m
ztSONr+CzI80XW4)m8oQL2f?zocpG8mz2VPGS-A6v9C>uXTHbPCHtxT02DNxVo(E5F
z-8w!)J8-qM2$dcn-N(HoSUUL&8stCr1SD~cw7PQ`<I#ZyK!LH%{rv1QEP2plr6sUp
zw2Bh_uDZ-)--p_fUvv9U&<5>Vlg`%Rq|$ac;CydCO#M!@IK5eGlnWm`nzvTl$87Vd
z8w8@S7f*kTsg1yl_0sVjxCe;#%xsSKy*ov1hz+YBAXV1911}Z>_Wl@dz*q45C<9Sa
zgo8k)ayJN*5q|xu_nc^$Gyi^v@n2uFP|t<X2>jyr(UB?>9&Fd;W1M`a&A<Ip24WWg
zuorK=%G?KQz}^^HwN1Ul1aVRKrV-zz9k)u}V_jC@tIl8=0cyEjBaoF?U92ry!4bT)
zOumH%mc}Szcu|<)H0@f9BwmvWyKTN4=49cWwX5*o3fQtZdboy3(;8X&)tnJ8l#8|Y
z!77r86Cf)N?^ZetO6?8+4}?BaO>jH092#HBoXQ#k;x_}1!c7y8y;j|H-(u2hsAP$C
z88|U;6F=J>;IW47JWZ_ygv+Mq!Lv5}?gcjOy|uG;E88nNehvurV<U8tI?2qSOY5l>
zw+iX0au<eY8roQu$P>Z~`Cq7E6w9#fBV$M;{x|)OFYvmiqJ7Lt(;IgU%D?tIft@8N
z+&0r8k>vDYE>~Ma!qPks_vEH~3L=JF)grxOv&4fWnoL)xf!~@A@<=79YuaN3AC6q4
zth{JKVkIC7cgpGolr{;z&FvJ;l&YhVGKI4Zl7LPwRq=}YD$XybGLBk_D)VE+p+LMc
zK5U<5fpnm5Tan^xDgvW(vpFZP!&bLJN%_VWAP>;V?s#g|3ntTXo)4A#b};s@D=Wn;
zer{@>;64;__hySpLu0f&FD}odn@p|BP(bDTi6OJ`l2hL=VUM5vcCrnJ*su@7RLh1Y
z=HlRr&5v(m(L8AP58*cDpZLg&6btO9RKBJcgy%;F;|MU)EeT(tS=SDDFwAH}Mr7<`
zck8f5pJ_KM4-~LJzfEt8Se1I2SoO}@+WTKmNZ;->A!RhbWYyxnPY8IVX=k~(p)dfI
zTSrXFD|XJ#wp60UzXaC2Y@#V2JQ5NXWIo;rCI^E4%$`5p4P<#tD&xM8XVWc*`fbne
zhXXE!{rEm-b58R<7k5AxX;+dUV<g&64Gk`jU)={Wu>xajJT8(ZsM5?2>!s(~B;s$E
zF<xkIWN1j5&r#%IEm-!cwa#F1rFOs8>StCN-3Rlkg62Og)RBOlE18~q;C75eLT7=>
zn0~4PzB(a4{^m8I`ce83#LX&@a+oOlLv|qPP@<eUdfG_v<k<m@l~oL=ky-5?Hl+^<
z-B>qVZAtnb+F7z4ZW`w?Y>*Vy`F<TAt<zh}s>%MMe?HMgYZn;rjOb|2ws`2tNuO!x
z7DJh~zcg|v5_@5XGvG2tWtuii0a95@Zsy>3yu8(YPF0TEX55tVRPL8?pDSJoieoVv
ztdqJpcQ7{Yc%?Kz=()oCFLFdC9j^x;DvW~>CQJfLm^)8`5KE&84v(-GuEVQ5$gcme
zfkdHhFl~z0n!9PTnK~krYD;|~JC&Q{hO(l2$3KORnJ@b-@CY*#6=G*ta6WW2F2n0`
z$X(h;&ko3TV}p*sw;5w;-}AOEGjoko-?%*6qT_^9y%AgiK(bP5+hCVTRzNUoSOedC
z+aA`7u8M~&37j1&$7U&B^FL`Fc<kN^VMN&I)@S!qigt{UCj40X?VZY+=95v``6!bM
zQu^hU=unQmwIP&?G_+6I=XLTv2CFT1buS5+&0U*2Fapl*-?se3z+}_bwOz8i5>7wV
ztsoA$x{iE_sqGY!d6YNhC%@IErT_MI6Dv96AhG;2uI&D`-XhM^{SIssKx*It9j=Uy
zO<8PN=5BFulVIPkIzkFU2DvrE5`FKIhTs!VmwB^aN8O&nuA+GY4-XGJFLF=J&CFh`
z-?X_Ok&KOv-QQ0=33s1dfPnA@sb7?KQ};Zda|hhT-tnF7eo_3YF1>uw2_nDI;4G=(
z3e_f<|4sx&!b+KZw`@X|3SoZ%VW~tfzO%r|J^<!EWl>2Bkg0F9fa%5ItrW0@<*o0#
z__p^(9SA2F*hZ~PP!(k39$={^C-U-;3<TC30ec3Xq-A;bW|vi2YbT#+<#xTnnA#+1
ztU(Y?oh$Wsigr#%!ME=L3TEUr`lO?4R;5lVH&QmstJ6}3`WQ%fH03v*@aH*Q*hfKu
zo#fdtC--KFH&%(oIMJ*OZbq3@W2SGa%DG<f;gwmbNxhe_qB|Gc=CP<ScD7V{$CpcG
z*#VDla$xE3%gA|?(~Y}@z&T4Fvpp*S@NoVgt6UWgE!!CC75?UNn2zIBXT;5)^@D0K
z`32<RKOz^sk+rLCp6*4`SoCE1qv|8ST*>;Fmk+%D8RN2qcVm^N;xHP6N=p;tCATte
z!ad&)PRA@#dRs^LgX%|k`e<u`&gx`3tE=SJNE1dgrL#l9_;*iUw~p@`3UVzSYy#pI
z-O2B@2T?wCiO@u<p($?gm!kBW{~-EvX*i>z-5?0gs}D(Z7;_Neju~o2^>{@B>$v}9
z)EUl^Fb(ZbFC0Zi9Ys5%%USMsNYn31!~~{1pt+mt{(lJaJAdvBy5^nq_{8UXu5vwH
z&sHLsD9mDb4kQVN#eSQ5EGt`Vp*1@GRw(P_UDGS}6h8~idfl}Hf3W<~;_~7)K-e{{
zsYc$GX={0l^QI=9k4c~81z)kr-&WONXmlf%dJ3moU*{MRN-2vuZHi>dQgSZ_qUyN!
zaDfQQH_G7W4pa)L+=%K`FjH)S^y2<Rg3|w*4JLkpp*U|bc@X)sEAX1XhV4IBSswSW
zs&M8po%ZPR*Rl7Bd+&UEn3>yIisW$oq$Oq)L9>)JBLuuyu9=JrjX@N$wRIJq)r29U
z-Fo=I#NhY^3Vq`O{;IG9YqS?Ex*+23Y1t)DG~g?#Fy7BkPfRM$6&_qtz8Ez9-pUnG
zJ>{$)!7yngYs>%oUMeB!pKl{8MIF&-k^D0Z$B|!cdr{*#U##xmPN~!D>rJgOOAj(#
zk72u@0bGIWm(KyYRyleRX6Iq}RuF^H3dMJQ^+Eyv&37J)fQ>VNaRZ;pG)b#DTAa{c
zv~a6|MZIwp(P+&;c^1uUx1hQ<LnES#-`*KrK_KHouV7ffi*{O+8KZBQXLOjviw}xV
ztq4bT%)VUB<vx!)v+*r8H@0D6KJ8TtYuLVVuNK)lUrUUwacX7Tn*NS={f7ouSCy_l
zJM+iemu$M~^s=8*XKwe0MEtj|B4ETU0oUUMFW9$0t(P>z1yOczAxzlmi%ac{z*`WG
z!vhc5-C!Ehy4M2h>nk!L5zw2k#m09n-FYk%rfgt8(=3v1E1)#Bw{&!;kinIi9lJd|
zuF0x2i8m6*y9k`mSK$*)56??0zLT~8z7n}c`tOaqLtVeY$kv&@y~Eq^*!$aOLGI>U
zVp8o(sr5*XqDF9O3LS6Fn!oI<SG$zw=T{eEPZukYDG{hxK%PyL_vzKi<MGK}S-_vN
zk5L)7`kKJc-+%o~H{534xZc?5lk{hmW^cB#(u>Y(ncO><@VULnZra!v;$oJ~c4|Cs
z_e^_c88_wd&{MafL+AAZloZ5}9L0<VR7r5WWHKYy%zuQg6Gv2!HHP>1WzSj63sP9T
z1_3u7>E{son3E2xyu$MHO}NJ=AyH|@de$EjXv^D%8NmNyi-#q}Eao{y2Dk*C{zU^;
zj*a}nS!UhI78~K2Km}B-;?j5TO<NTO-2p6pNqX^)LV(qQBbK4F7EBklr<_C&Dnqg8
ze`6Q~bM&E`wmU?M7$uCSb9G)jr7a~tV?DVNF<K;6;lS(Xw`}4x#T(@Jmsm&t0tYuc
z{3N=pJu2f33DyM&H1Ex;M0c!$3KH<8W8XCQGZ~VV170*hN;3yRouL9f0^rs!%Dl)i
zWIMs&{Faqm!J5$EVhH7+DY9MZF`y>);763VYReN(4pp@%;@*nNvG;b5BpeF;G^)ls
zfVtMtt1)A~FJ1LkdIudl<K|Ki*fUyhRMM8%bq45UM~{2WzMiN~{DS7;vkm1;VA~#H
z|8p}m0wA7FcK_ov+|m~C*0YE@er9jS$|+)_`-yyQSiQ<*Cc$Hk12`&B6?Zz!fXWk2
zQJpj;q^!7!@UOT4c)9?i<{4-J7y#Zk=rk$v!WSBO*139LZ>Z119Fzq-6R&mEZcF(B
z)_=MAS1PaL3U`Fnht-U=`b*euKpHqdB!5ok)v3ugb86-Ah`Y}WA~E^GmCgVfGll#j
z;Cyt&&JbXl2RU8<+{KXsbuwG==)`da)+qSq8k#h+)QW0IM(-f|fJY0A;RQAXx=Ekn
zd`FkC#DM5^8cbx>Y162VJQSxw$9W)~oJbEN77Rxb;geDi6a8x8^BY6Boi<@4nAbA$
zo53<EhxX`s(N|s5mJ?_H$}yvLvkfF+m!>}T7Co{GxYu@);85%&cx<8$3Spd%aYT2V
zs1~S_=|j%|q=gj0oc9bOQ5Wf@pjMCo86c9^_PSa6)X%heZ<)Uy;1Uv52nw3%^IGRN
z!R!OgWp{B`$}9w<37+oOt!vOgihverKNSQt8SUm$Lo8xW+92UMN7xzeNZ^a(V$%%c
zwl#drNa%!t4!vXOR}NJUubV95?uT(S%#XI`)4TtMFfae2F%fx9j5!zZiO$-R&+7pr
zU;vx#$`3pvNUM|;U95XM&Bf0FO$9}Zf@I)8Oj7>n2x7-h18>u7xb(T&4j(1Givn#G
zb8I`rNwyU(#6)@~q(81~-@4|_a?|_o=LuxH_l8wJ2W1e2k+-8A1T3z(fz|Oc)L=7;
zzHC*Oyql6i3Yg`4O|N+$f47kz>#;(<WN-JoCnOmD96&aE%s&y1uYipwAfsfs1B0jH
zBwH%8CC}r<kRb>cK&I5f7HWRfb!GOn){fn*@ebcrQ{)@k&k+5&?-jWiPah1uE*SgE
zV-F&85(93_x@^zWOKa!w2XgckNr#nTq$;&MZ#ytRKh#8xGIP;Ch4856b$v8%n3~q`
zcZ;=iO~JsAQQndafpa!#(v*IwX2bDGQ*eQRM`18uRaI9zpYO<$4SUgdFKAbH&$Inu
zXFsl8BT=0o`F(BZ{<r^e)$PCThj?P}OSjbv^<EFjen&?5daNQ6K|1(yKLip(pVv0G
zInu=TRZMN*M_3YwaR6NTr5D&n&vHxF@J1vSdpeJ6Ete$E9sI>^CUkyMOJZF31Eh_U
zZb)_uG;?QLJ4e{pE5EAX{9BP5aBt0!05dST3Wo+zOWPpb67;rC4ZMZc%);J9fVO2`
zVUEM7QqR**XVC)xPJu*)Du2HkgD+P)x;D+IL}km8%9=|Rf3Qd`%hF4!3f)5Z&tDM5
zbW2C!3&ALHywYwKEXMf;p2G@C6xuN+f|M2Hir5*lBN$zvd2BSVaZFgJ-&{dAn0doR
zbqy^_&AaWw?beZ(4mvzd#wFgLjx^RemvNqYITVd~OZSRv|I$5Ak8C3GrZ!?y;2;(s
zTuEyXY}o#^e$-yOv^%??kJg1HuFhoTdThYt?9v`tGYWnTpw=^C&juc%lj^3odYnP{
zelFvGLtZw_0E0q>HJ2gBrvzU%hpPiY>IV9hv?yUO@sdw}w~c+U&+}r<)}T~^FsbXJ
z&?L&F(k(0icmUV%nfG#aU6G{dIh5UwLEvgaVQ7Dbx#pS)Z}ArPE$M5?&EArJd`NkA
z4|{QOAHqjY9gQn``Get8^5-vHVpr`J{y<i|g@xHe_jkW`MAKhzI|}yaBq+PMp69#_
z>^58<#qI66Imt8bGK<;gmi?eg&!a1WO~8t%sO21-SU@;HRL7%<XZtOP7Q@*Dniz{K
zcpfeTXO1-^4q>8~!$FfQO`a{cRE*xvl#$ntG;^-S7O2KH!rEj+4rfiFDYZ$aI~KnR
zjxUXbezb}`dzUPW-H5oLoLGz?zrcYX7DQPqCRIBqD@Vi#aGGz5W;|%Ko5kjT?lJ2j
zz2N&Sa?Y$Hc*CA1o7CDU^7+8?A34iCutrv4Pr=Bec(SfsIEOr=<~7gg8+{4c$y79k
za{}<+^V0Tsng3$diq|`SgCILv!K72Qd>3)qwTW}<yL<jCeo<x;PBDVgmWB&S@Ug)(
z%&8lql0)adMTQ^EP!2>SC=>E)jGdp(At_-s=!|eU(Nc8G0VvSqn^Y#9c7C!wMRQ_`
zw)tLU8lx48&*fZT5H#nBYd!)*MsOoDYEbhTd0<vlcW9Op?wa3RG=DVq&ZlE0Z1MKJ
zy~l7)uGay+d_?&fWYaDFzV8|UEYOuVrL^964G^;h!rvSwQWeg_Cr#)6L!Mt|D|$z!
z@ka?}1M$Ra(H<wc=euD8x1D#%px!rg54`PwNl5f|A22LOEvMjJ9tszw+qg{IJ`eWf
z@R@QlKgJzecuNv~mUt6cDa$O;t+1;_er(NYZvj_+`@B@L+9Yc#zuFYUvIZ7bO@aB*
z{%k`IQC_=zFH~X6(hL@nzrG0=BVfoh3TRbYM)q{>0DB4Z4aso=^t@#<71)>i-ZlaM
zqk_ZB+{A$tM1nB}LU*M5vV7kS&9>v7TB9Un<7B$QGs4&1EzQj^c+VZ*wRT5^E^zLP
zV36N954_@wy(>Er4I};Ld)szY_eei%GX}iFRT>Qj_Mps$TtjxRTO2ehRga_@=9tZT
zxoU1c?s9?1*otL>G?;lr)HlAbJ!6cV!E0?6qmd20sL12~V&w`xt}^RS_lvga`yJFh
z5Bo?t<T*>pL(9sF;P$_?A20JAkNuUGd-d3hhzgm=CMA(FkVAaKa2~vJe2a{W_@X!!
z&L;$hir8*BDJ%QVpHW+>LpJF=in@+Mc5Z)|a00Q~!s|9>WsAYuLY*5CLsN1vmsK7-
z{8GaOhoL&z5>wcOjg(QDU*gG=IL*O1HL6B0zJQOhU0xVP)2Fn)t+foczb^2xDh$VA
zt|sZ{pyaVGo4P1r_*$vABv%xikhiL%XXV1{UhvrI?`Z$^;l)Vil@~-`O=z5z0IxOf
z>7q>xql23&Js@6?eXMY@?3VCD<Bd;A)i--qQz7G;i=~;=`lPZOLn6i=yY!oX(SOiD
zV8FF+1I+4E29AK>qZF5A)PhTDE?a1f-1^&n!J=B^gZ9qZg2s6`!=t|T?T<MEMl00&
ze7FZw7K{8GD^QsW2%XHirqWiQBZ5xKPwG>osSSeQW_^}T%Pab)pN-1Ddzo!o+3=I`
z7aVaulXfPCj7s~z7EJ*yV?TH?l+Aly)joJ!{aU~jtL;7RE#XF6JnC@$*BFOe#_7lS
zEkQyqB6Qlu;Rak{KBO3opfY@BkeA?uq44`FK*$EtX^hWGV&jwMK<MDb!nbAooadTJ
z0I#<j%WpkbkK~X#UFYZiSJZx)HaaZyGbh>C8P>QGoAfNxbPm}ALF^ZNff6*8WlYhb
z@ZrnOXMAAn(haVvNiMFPanrSrpEgH$rIYSy*A!Ad`clwTG5vODjsPs}1~q~p7x?ZM
z82<kl`=%Z6^Rg*>bso)lrM)(*a@Oose)^)<{Ql!%@4=FbQ=MR9OJZ$9YG4mEbr%pK
zA2ZTmQAMkrmyQ}NnHFSkE^6e>Ps7h0@+-}BRY48Nr2*g(7o~Wy=sNq2CJ%zQ+3ToC
zKU)kDwiC40X1<-HGKT#{YUOGV0LwdRH;c|ekBC`*5wEZZC|>$+Na><>_2T2=2Mkdy
z00{0^-h_)`N%8_PuFr)}*vvB6T0?W#on2YL+|Zx~?BgS+V15?!a*tUcui-SrIb5HG
z&;#%Qfebl&VfLJWaF%$IxmM2GRaFUrvWs51v>;L!6ENK3s|LT(r!t{vV{5#06c8QB
ziJu3Dmo0ft9`1DA3vl#ZkXu=`HS>Y57M6G#nAha7zEw3#bvl%U7t28NLZlsdyu7={
zIwlpIEo;UL^Z@+}<$<o2E6J*h+jJ*_uI_``28Uri>=(*}>flu8r>v*TMZx^`@8;E<
zrUiBd%rq>Qv~Ujr!8EIO^UW}9ZTARpQ2}d4uJ&)>hh0kr(#aaE9H0Fh*s{8dZ{S|B
zTfU#g?CCw!s@PG8n}{Fy*j+oUF!KjR`}no~R~W+S%Yf6D%mgv_^NtS$cfNV3glsBb
z7w&+uN{N7g;E_Ed$P2vX?y5z>h80aXzq9-T;%Kz_m7TA3hUo0UV@hp|^nU0<h%66Q
z2-^zfyA&TK4yiAJG;waU#j%A~oDSTuJ3Xm^w63+Y#k%EA*l%<ZGBK$XTyLdukV*p9
zCZS0sXMdk}MJFZ)C~44d3Y;N?^mwjmwsFsKa2&;WqYBDngNY=ABfh@1PN3<Aub^>D
zqUt8y?(F($>C^5-6QB!i{5D7eH+fhum}B8W$C(U77q@c6apH4ir3{cRPV=Tuxb*#P
zB}|d5#<Dny`Q@D4mZxdMNuOGH?Z6<(BYV(6y>M;cZpLTCf*k(?)Q!b2(a7L5Q<Br6
zyt>J<-m;X>AA{bjrP1a4ypc>dGTkR?rPjWQFjylt{)^2`p8`R(un`DZ4U(S1KK#TU
zNz{tn&cH45sOzBo*G>#Tinr?K=AZXCS)lP7X2H75Sz2d-1zYG{vcxZ7Dy{m_pnyM~
z@!2XokHeiVilcflNJfX)n_8M7?721G;P0!|kf>iwuMPxf-hYSLE#cVlia*diY)B1n
zo*xGSH^%*AivD;itJ1{Iu!>`bo{@OExsn3mEoOfANRE;M_|vC7EuV-DeDxh>_WvU$
zGJq=d!l*-Sx8oSP`xjL2wNU5|p_+Ls^@hCW)%vCDza=KzXSo2c<ah>Y&D;Fw|D+3)
zfu60Ci@6HNUF3}u(PoClv~~q72G`ZVJlsvB{@c%^--+mHw!=T9ED$4iea4L(L|04;
zC8jg`EcfAW4yT4o`v<InY1C8NpmP;`$xNAl;aqCPsM%Ehuv7%e^sMn5@avE2#!<yC
zks>t?9}uc0Pm|q(uDalIr!VQ?94O#GD&&h06<1bh)zqblH2ta6t2JB#N1}0m^A%Eb
z6zSO^Q;awVVK`Ea1cT7OHZYoXPmc}fB2GXt(zqa@Wv4ERIhbw=k1L1$xDYHYs8^eE
zF3v0BMX_Ff{+=wM=ruiHNFNmT73jtD%eJ-7!7axaqH(d&7d$>KJ`ZPr6z6U4DCBiA
z(@f9*d}~V{!J!~oAvgIsoiCG3Pk|tG^Fxwk)z4{6UHjPD2bESG_&<ltUilSUhhIW{
z35~3imxs(Zm9{69Fq`?$hX|%Df_SWpL&-ze?}%_Hgf(`h%;*v=!=r2DnCfMG!TOji
z3&yR>0Z!`t?)m;{b1>JrJM}hqC4&2vRS~x-@Z3-SN@X?4UODCm73?@=$T4$OGPNkU
zR2KWgG-C_MVDreTXjuaVLB-qgD*`Z$bsj9zMKD5yeVah1>AqR61!O;okH^TTRcZ>g
zjaMqJI{y8*fckEi$E0xuOLY?FcHAo&VV5r1K0KMCHaR4|*rgMns<%?Cb8j4Yq=Lgp
z2k)mvzz@BpQ%ydfiRowzXInPR=SHPZx7Agez-w*i+X2@q(@!WlhA;VI0nzoVh(U=^
z^eu3CyGr(Bs@I1H6AmQ~Dr7+J9mRZ)|93P(a<a9J@#x$pFa9)ob-3^kRNm_UAP8c4
z|LUruq8!(h{vMWpy;8jYeZ@f8(U@?A%CtKw5f6Xsqp*kWzp;i0iU7rY0jnjQ1vqOX
zw?b=hxm2-O3!0BEW(aS-Si1S&Ei2>5zv}2>KYfn*(J{&unQzB(?7g-Hg#nxTNDlPl
z$yPXW3Vn#)JlN5JD@K@Jcy;rD&DdU?G{dVcJjv&VNo`BDOZ)uy<2hO&{88!i7Su5I
z3l{*nCig^yZ_ITGr<|eEeuZOi-#kyf0THMhV3jabEh;WnY;2RtC0JVeSq3aH!)rJU
zvGf|%q=c*;0F|e5yO!@1qn185Nm;}(k&ownOGb$^Pqu98tNJI`ncjp@HZ0NMjlXiT
z+<tuKEhiMcrDf3!*)-hek_}#9)8C8EZ0QOx4YfKGD7Vh*AU+4G8OlyLox}k;t|U8C
z31((BZ?wJ(vNCgfG92|TNYbS)WVx)bwuj)nhH(YM>b8b5prSo;FurieZp3?J+d({g
zNJGM+?67VX0<0d4R_BAr;0SY-qR~@HmKg1qSTioj1|9)l-D1V$;@@$!vEX%^L|HfI
zW{vKb&RCy_($`TiBJ5T<rrdRmHRa&$wIi3(^RzdxVz+dE)}VjCW1DeXv+juo#@%j6
z!+i@TtBYwV?5p1vOExR_Ij#Tc8e+|FjNAODEgC~c|0M)0#bBTG(fe(W+}qL@pB)$e
zZN4c*At~Q_X_elk(<NQlej>_NaKx(zzkv|QKYf<Psi^oYkCR?tYD3d)+rVvFbi{G_
z!CZI4m{pOO&%K{Ghnq)1li=7`mIG;<M^c-PD_m+4>sVJ;4tX1oJr>!FbEeW_NKPfC
z;4qxuKe!;;XTsqhc$j8n%lQi0N@(+ozvr#uG#;^G5^`Z?*}|U{>m`LfUQ1Ml4zr0`
z%IM-~Xh{rGpNcN4ekz248!-gbY5XKBgwc4{iJ*YBFRj?#ueE3DX#!ll^G`HR2K4G?
zW$aw`vg+SCg`8@jo2QGR_46doMW4W5<GIs`kAdSNT5h6u-%gOY1%@&Tzt7WKSQos2
z35T0L<RDm*m%V;&fT`~Xo~1m&VJ`MFz!^F4p<V|G{Cc{JOYaCeZ0rmlzq)x*uB`2G
z1F+kQyj=Os@9W+qDX3%&uKS&vFa$?Yt?j!B(J6cPC>+isVJ=Xp68fV!c}6!2mdp@r
z?BpU<73R2J$tOhn_-Dw^#J|c?fGdkM=f{6w1r_H!OTfrXO71a!-tc9Qmn0+*RmktB
zIHp?Wn=ID3i_YC!1R(!)h-4fj5BbUC*BoKq&06v4R_*b>O)E>r*RRea{XBwmNFEWB
z^UoGir_<~?hMe@NbJY!iZx0HGQf$c?AcDvn(#UEoGdI3S4uc}{Y0fI*(WDGu@K7E3
zB@C$?m}s8x$jddWFn4kD2xJ{ISp3C#`N6#%V|EA)(4mTKAvW!yl#Ddmpn~tm9D_^t
z*3g#Yb^OwQJ%749x>*(;2<Kl%2i;9ahz^FK5`wZR2z>GkkX10p{_MqZ_pJTVTOaK4
z^v|(c5K841G)Z;=hm$4r)Y&Zn_R(-<6Cf{h>@g+7=V9=D2@2WZrg>0B@2*o0zs-X{
zWN?1p7#gh*$L78hKYPD=Gi{#LLQ!2WX4LUVRgP!~>{5km4J-hiOKN%3ewgu~RW#tw
zZwxpA%8v#D)F!l5RhW|UzFqg<Nn&7vjuG;^=h*w1H({`}(Pae(@}7nyL*;8=;THvJ
z?ET;&Ca2SXc}x2<<Z%r&V+@Gu;uMGefE2;y$e~-yq>~@QES&@c5=bnOcYmAh-x-}L
zvyDHTu*6vnI^F&s5-UG6`sgYv!P@js&Ge)K{=`=z%suIR{j)vvG{EkfRB?XMXVE@G
z^u5Q=?5nvB#q6cWU(q6}!t}YQlvI!mER>kO&ZwZVbli~+&gYf-(rBNAYcR4;mV-0f
zt+rRfw*AmOyxsEi8CfoDGaPxhYSMQ}Fp#h8lbqRpf>#tyJ>BEFJFT29xq8Z{<>$0w
zAfJ0)`e6p=L=upPptU^)e%;RIG**dn%o8TPE^h`-gqtV99b?7xu2<6+P)&N;Tc^O(
z9JcP~i_RE%_tH^t{ORUvjnkLB60?{00grOMc~5I0G50r)f(q;CcSU7yWwEJNC<7*_
z{Vz5T!8CP<)^%y2NR^JV?r}r59&kcqwI|1Zv9Dd$okrUGbly)<Sq+$d46evnL!2h5
z!KtRdSn%ZsriX&u%1*FDIm0WW-E6V2%M$s*I*m7wR_p|M>@bm|T%;6y`uPhFhp=9_
zM6=2ke+sT)vaSd%U;#Y2K8-NZ2@OsSVl1z@BTiHCBnz(WX(l9-52*~PZw$%of-`Gy
zX{goW(KJ``W$4wIL0OkG)i^1r?A|_TVj~tWF&@1n3;Okn=4?~HUL)I`WB#W_CYx0#
zZF*|iyie<U*5kj<G^SI*iDN=<Wtf4&UB2V`c9|``pjKsMBZ3ns`gB38oNwo|Ryegy
zX}$snW7S(b!6Ga%oZA%Jzi$uiXJF)6;}Gxmz)}{Pg%$L+PN80O$z(7o7aL)1*x8er
zl_x}@;+&o~5lA!C3%9D7Rg2(6D9X~v;m&8j-Hq)VHP53>5nwoHP)$7@)QhO>5XU2S
zXtvuTT61a~ReK|CyOUB%!{WFz4ED+Z8}^+D2}pN%#84_4+07%yGK|Gw=j8Ko?})Mm
zL69mHzhbq5%Y#Ui&A?urW>U7B4FH`?Q4W-v)R>;NrwKY|pWjZr_2b*gE8(iA+eG55
zXLa_ic*k6Hc59Q<Rv(7TVAn`q8iOR0966^DhU8mgXIg<m>1C23FXWdFI&F2ofL4Gh
zhs!VUE9cmP#2+hm@%$qC#@)~o5ggmQtL6c%g#vh}>q=moOE@1{dEK0|Y?9}2IWn+E
zzF$YzwQU<uMHp~#q-qVsH2maZ+WwBdsO~2TXV&c9WgI~?Lc}j!H9Y@A=@YMiI$KrG
zi}x2zJyH3com2nT@GyzI6MiAdF@nL%VaFzco~3P|4j<Jd(3K_1kmXL!07flnS?zjd
zm?qGov-c>jIqB%`wQrNsS#Y|glz-^cPvivrw&0=<@R!JmwJbF0ApL&UD;Ji~T|4Hz
zyxnn<y*YXu81a=)poJ>00UpW|&X!E)P<!mzd5nOan>4l7yWKGfUVJYJz`TZpsQ#*z
zvi;|0-b?U*_=)62Ld>$?<aovxOCCDe$Obpf6$M7rZ$Sq%6klCEd_GuhBh3UIy#GKK
zj!^fkN(oQFoVf%pwS!h5o&zIA6UG)Vea*#9e-T4cIrY_i++>B+lg6*uSZ(Yd&MPk5
z&#8WLA7!qy&s0F3w``Uek;gtCN-f&>wjE>H&}6p9tQltBHyC4j{jy(UrrA$V{3%c~
zjaNfS!S5`aPf4xW2LsOcR|W@~&Bk^5fvdgWl=t<2C&LNL2W<xHlw2r-FQ=w)tE%*j
z7>g=BO=7w78bO`BjZ#@UH%8v}YfUi6#Oi*;^$PK{L&{={&g_^go%s0ir$tSwMS<xi
z0%ykyO!YKoD>opx`v0BL2eR=ly7b|F$hu#8z@Ovo`-Yy@!2zK9WW71f<L17<1Uf%^
z#&zh-NK9s>`E7Rx{BCL8CmVdG&OO-(Sn_)To(ljXP*bNd)_pJLiT?|@kA4e0H)Jff
z_{$SvO0y%0><p!mWtL^Z&=zMOOc+dEQ=#jp!(YrKV$n?lVh1KJz>PcEIRvLx*T_Tm
zx22JF4NWSnL6VREeKfAr%j3r+LfkTBS(NY!V<r?`@ppk2>61I2qZVI2*z!4rE={81
zl8F^>0|p6;0#mvUJ1L7fdK|O{i(vx^+aWWtm>JhAWtQ8^p;;QtaWSkfkX;S(d4dWk
z3r27N<F<wnyy|wVq<9@&yi0rQ%E_i2F<P4)i$)TQKMa?#a7{U06SZTPS)#7^eEG0|
zu}BPe)^C?{q^qxUTo+<MVSw+A0YgfsjCH`xOA3EFt)F}(;($Q$K9$3eD~nxUran+P
zr($s88n85#XK?PC#;(XfX%4{M4wm;iDdGe7$@S7v2lx=o>aCqDw9$y^5WbJ2(<&+8
zm+zvbGr9nF&+3%CpbeTZo83*;I57~|qBTp)>($0ie@o>z<El{D@<?*;L%`u3RxJvu
zm3>a2mfgvc>gM4NFmY|tm+jxS)g|LW1jix@xKX5Qnn95;sCH6Xx7vweb(Pl^c$+1j
zh8QiQy>Y)62<in^J|qFYo38k_TZApFo@b8|!}{lQA3sdKSfCQS>Jm9&%y9w5ZkU~q
zi2r#CV#*CPS2B4MH-i+q{@bL2E1ac4=oA#Umll{Zadr5;WXDR%wc0}&Pj4X08NJw=
z#A1(bog)KM?lfMM-rDCA1cOBij*oA3Z4+J<h1&Yi+6cs{H_-oVmKJgR?(Qq<Q!Ux2
za(gT0zg+3E8er08ioP*dseO0#@yU(tV>eoF6fWQx=0E7hQ;I#x2|rZoeLT#$Jsa}@
z{6`;xN-(8^umI!Nx5RhYDTwO#&Q@+bnxU>@ElbiV^vOJymGFx{v~J*U4GcE}Ry8d$
z$*F`le_}f`N5J}qUK}dZXrEAGgQ1yTizW&6P4wFFj;X-~*>>Dk|M6`;eb$Ac#lNF#
ziRHnvQrfQyL#I}(xw80XG@}j`bAV@{=VpT|f}!3QU}L*U>vcxX`<3#K^2SCEgw-Wu
zKnCZV0t2>}#u)O4fO(aut<Y}L?`|LdILJ(4XOqaMfpIa9RCQusCmdj%jK2iSFE+ja
zocbY~aI+Ldb7FVzwdX&)DvbR{$kvv5F9#(+F^c_$d7nh`hUF|6sp4=W?Ph}u;+nYj
z_%nXs&S(h4#gn@feWI0jva9bc@J^>wT%x>nl@j&OoUZqSzEfab<zRTf1>B7hHtZG`
zZSTjmJO-=N#Zky|$2rf<%w+|i@%xx4Sm8POH;jS6qhZn)b94no21gr!`vGeEK{x|L
z#Z{+(Wuc#&b?TJkJ=lL*uOzO=hjoZ#J#-B(ybA`dmrf>N)&c7HeTMws`;{Ebi{IhB
zj9`E<DB!)k^T_WSV()%n5<3kVUOMvWgvWncgScmV<mK)C+wEYQ0v@7B3_b`5oS8j6
zI&0<724^fk{ntPr5^(k4;o@6yT3_`|yYk;~GpwagS3aGi1rf|<1-cW^S(gnX=9!52
zLbj&MM!1|=Fjx&4L9=X-DKgD_bPmlumSDt14>PGCC*+i1%3-tL$ujp$x-=qRQfEd;
zSSkuRNW`;mUT*ngNg&tL6pffxXxH<g5a*5rCw42~NmPWg%JZARl!iC8q_I$g`RNrA
zLo|4Z5nVM2?%%VQ9wLA#8`OmE$I{gBBEde%Eub$QqN=z|p}#6jjS3DzL6DUj`L2SW
z<-NAkC$8DgvlnJRqu@pcdJTZlyxE5)r!}1$2qMMMIcqfE6^=d?y5CgcJoJvALY?IV
zDj{GeDei`yNo7vT=1gb&K-lcPLL_a&rlqhqx>P5xM^iZle22h3)KvAJzVa-up+N(J
zWcw!PT4uim(mqPdr>DUcEDL{$tuK<mp;WsX*)$mmB$dr5vh*4kPa5geVnFr&EQ90B
z@@`(=u<y+q3%8%{y=uH6C9f}s(T4I!X{hUTCO9;CfQB>$9<_2fx{?y=(uBlXvCz5K
zt#=BYZ267^$&A4gjhxz~<CTv6BG?OkCkb<vrp`vIjKd}t-(0mSZL8;_t7L{knVq^!
zAfK$>i8yYAeei=8f56e0wCQvZe(*il%Ft6{q~7f=VM(h}T0p)Xi|MFp4rhXazQR6P
zRSOMs;9&u35#C@_kTjY2yo#rHmpEq;(Dd=8#^MJIMmBz}$(Yv>qJb5^09Vp>LnV27
zB}2m6?`g6Zh3Zs_I9d$n074U6bG}6S1_X6n!uyM!E|31t=NNF#7|@UTFpcRvC317R
z5rX;O>FsvH#SeAmvjR-N$o~n<NqyYXtp0kVxN_Vk?A0&<dQY3GzI7j&RItP}vHDXn
z2rQ7k9mCcANlR^=$E*vE*z-%N2<6V!1kxe2mB=Wllr5ghL=H5Qt#-!!i@63@>9-Ff
z{X7~#E;PY2v*Bv62s*v}Z4*<3s|z}Qm`^9wQ#xz8i+L{>_uDnNNFFoCu8+CZSr^Xt
zqsP;_z!|)AJcI&-94@E-js-Nn{m^($5RW@Kp2jb#`A#!by#<b7wpoW2k<cH>qeK+m
z|NP5qu31SFH^R6Qjv!E@x+TqQ1&<!r=HiKX?4xsNzfC)P^FFn9fzY6RIbmx5gE+*^
zF_K(DgNvqiWg3q&i-+erqJLU-(mHe-%TXv&Kx4uQ42{qa5Kp0CwjpFwz2=Al7EsO7
z;g^rs>QO#!cUOXKVZUru#yBme3gt4y!BrYgDBpA4k{E1slNJRgR(~Q05T;f#FDXMg
z9f-COIElqW$CbgBxoV~-TP`{EC{NyOUw!Ns2()*B-}8sdH+c%%zt}x@rz{s1F~A}o
z?l2fvI`I%O>boVELK$#G{djwz2z((^UAr8Vw?EJo3{47+&l`H4$exLv{gm|I38DwY
z*`Bj<Qdl5iiSJaKKzsbu(*tInMGIwj)g$J{pn3e$&CN<*R@--NyFVW%{%AQoV?RRM
z?*KPE0-dxvCHr=B)iGpR)dghTli0;$L075maEt=t@qHFzLkW`RZ!>WdK~udkGnyuy
z6}zr@k|sNdXa-3%8vn=}mK}#Pbs~x8_3aAt(=dIKH0LcDtZ$9d&jGvs8fJwjMJJ1b
z0~l;G-(f^3yRlNdR^YtJV$Iq?52KP>B5iH7%Wq0I{+h-uD9zY_$Bh(C2QqDdu%H|K
z7ef<}>AHf9f4E6eRKCez!ciYJp;DpGW-X||fs~-NvtrBRL;$C-Ztg>Yaxw!?_*UX;
zS*+p2V$PDP_cI#hd+5nTf=Yp`ebc!5fi_w%WEqOeS3JIqM`h7u(<~P08@>cBxiUZx
z8o+H41o>UHp{-MiASV29&^9E{GiQc32##d#3(0Qyq*o`?W>~Ihb-`5bg=7t%o%>%3
z;ADP%#oFPADHaVJHBIRg*3fL#S{|UU*^QVs34BQBU!<#kd-USb+H9kMxzO2WQ546?
zlP2Ow`gg+m)Is5h)`CgHHi!E7=L)WDR<K6-jV=sAB`)EU8Z#S@yDVPnX8Yk_5gf%T
z3vOmx-;c30{}b2t+;<6SiXV`B1K1tGB_8hGgQF4{%a}{n9a|nHG6t)pF8-Cn67Q!;
z9h#&S2wkC6CN*}T>lzimFKlH$76^&oo_9+wnCBUuaB4Su?#sO}e7>B(Y{9z*8AeN}
zX$E52FoR2H>oWYl)DuK#hSFB4z0}{U<4wLv|MwIJ7XYCsE(jCnx2x3UN~Vw{@Z2*0
z_X0F0gNDeL0rxo}L0*z1y=iIqYd76DXOA~MPxK$3;7|AY7acu+JF04G{EGk@cYfi5
z!J(1d|0IexI$es09=O=SGM-uv+ZKlrpvs_G__C-u<Pou@K!n(&C3J6FYJBd@ox_?9
z_DU}9&4U_4g}i9x=7^s?>R3!;f@MCx-QqX#*nqg#3^wGDm!g6d_4f_l*TAV&FWmjJ
zz?|96!|z^)C&pt^+w*9X#E|c^{a<s}w5ZO$CXt+OAOVgRuvAuKm$Ec0ISX*+0#HcZ
z!z<17VOc7aX3AMI?^s{#Q;I?4h7z<2G<}f_XiYuX(4m^u950)rhxrh&e2K=wmv;6J
zRJ(J*t)Y_z202is_$);-B)X(w2GZIt1Z)ZlJ8q-I)bnpLX<}S=hQWv3SXW?H^l$r|
z1&Sh|lSQLkip#^uruFSy;f?LBCjDUhOfo!j5Z(U}!Sdrg-?p%j-=yUPn6-4W6^4;9
zNQMgtISuPUQ{rr;g-=zJ0l0NVW&I~h!-4aHokkWx+^HnpqQbnpD(_^hORR;o$3}~C
z8zP$V^8U_Eh{mDvuy*!J=ZhZR@Hc8kJW8jrD^@V#t^1h9b?lA$E|)X=U*F_!!q;6w
z1O8^|-*3+9kEgxqzF8p)cv5_A=SKc`-)5goC#i1Xf7wS5{_*B9LD{J6VvgutARpP+
ziv~x5b3V&wL5@dAPCH>hP?}_-iZm_e?EkbAb7hwujh9zmn280pT0xAW206_hy*rTp
zO#UD!_0~Ea^7S8z1(LV+LR7Li(dkNS8j1AC%b3LNrWK;P98B3L<cy!~V~}Ql45IK`
z?K4F`6SA#W_z-*~*?0t0vG3%JvBfg|P*av2^a5H|J(jD^5skig$q0p(K;rE>Dy*cT
z{*uo9P1Ov?=LP0+f&izwF_+aw_=qQZUj0w98KPD+1I{!Jd}@1e`v{W@9|@L4(Wu|T
z`bmaV{7FtKV~G;!H>N?jLukQkvT`ffz96x9RbY;LAmPxh{~%NBNCGK>VH&`5_OHrY
zplO;Xo?HuOH!Evtx6U=*Rm%JPb`{xs%}{2E!fzr+Uqt;HCt=-s8v&VKemCMcAHMP-
zY?5iCflp7T{hv|zSNH9G*6SQRRzs&J(;5EtMM^Hs@zY*8L_P&TwSb!MhXK#WHx)a1
zBnOE%>}dQ#ejdJypsuE!D0^IuKBi^_IAQWgalUct8*&nN`4mV52s44UH=vu)mBai*
zg+6Z8gmGolKq`5+&gR|@mz{(DlXyyqe}go0QK1YDSW_l*;~i%bdUXnjlRhb_^*NF`
zZe|Yzf+bKh<=LVQ>T(P}sBAkP;)<{lh*#X*7sR$u-vtRHFfzzDp&4N{%scDUShmfR
zMf<vKycuTX=hoi3q4)^L&A8#<jr)s{W-UaWYtmc3Bn<36IE<f}a2QK3cp^j($uqbc
zByn)qjxZKa1nq7<ROp7a$)(3N%ggn1_W<h1Zu?WeH+az~FbsfwF~9HaseHHp!ymWF
zJdwWvpc)`-g01Ii2gzeH&C^oc8qp}Aaa2p2hz~dlnmYSzF}1ziyX40(fJ_yaK43XD
zbnVM(Sd=#>YYv(e#6`-u^+!0>xj_9mT@)^0l{Dx~OcNtJ7-aX`%<d@;|Ds`c$8)j>
zfEl9F*SE{8Mka%(Y67_g2<&#=-6EuTrARw8f>FHBbVJJbGRzS83~T4kJdNA>t+e!6
zQ@I+Y{vT0q8C7N5wQVaQf^?_gq7ms1QM#3q5TrvvS~?V^yHmIT0cq(D=}rNWZjkPV
zcX~f#eBXZ^%3__zdCWbw`4loMB~eluba9(VaWyjz8^U*I_Is-0#u8bRwvO8g--GX(
z$}Ne#IAo5nv3ErD4u#{4<8rqbhuD;OM&?*t4!iVc!kWN;yKBfSCmDk74U=+-Xb|Ra
z&on|-5_&I{R6l0zy2!{ov+b8LW8<iBJAAO>hKa_fu#@#<NG0;bzkBAdMDVIZicEh*
z5?!Uz#ZxVML-3I`lXX!LtAhv2izhjw<;Xv)?Zx44@xV0rap)sUrILrK?B)Vov8LGx
z=XpmS`gzut_bDXO?4=2&D<zljdV7Ug<I-C>QNH=TGx`DL^E*WGk}sz};{l3Yi;iVm
zVhzIk<O;j#Y&=t>Y<PuT|M!ad><@8VsjJO!y|uLZO#!iM;laJdfsg#u`G@061Xo?3
z6+Dhw;M?RW{k%nSc+r3N^}P4hXz<l>7M0(Prr{lxc}9Kj02?vk&cKbPnvVaq$bQA%
zvnHkHp!#Q56Ih8M9vU0mx%Z}Ph#=!8r0C@!#jg%Dswo2h=16LE+$*MU?=J-m6^GS3
zM;l{aEBS9XN|4eMa+61Ee?<u_OdYD1LMK%8l~<OLr<e9pjG$wRQWT_38cNB=ZB>{*
z_(VN&Dq$s-z2^7ZS~4b^De6nQoC!~CeHKQQ$1FnGn7lQjp)r_M0q(i3Y7%kuj@lV(
z(wJ8f8$OUmzo!&w=kq1x(b5H{=E*%PN8J{>pQ?2C8ms4^=@&f4Dy5bc#Sg}5Wi8aZ
z!hM*_10eO(pDIDg*r&m*NK$}IY&3d;GyO}M<{J%}#`6jFe^aJk=7&;0`J`+|>AS*9
z^#@F0=Uu*fY*xp)%RD>U1@r!}rG#J#D*k0MFcv#<4dTo-HOq?*f|v`{8)y+4rre(~
zkK)oV`EvI@+z%o9bu=Aui4!X*r2XGpG{PsbjT9wX-9Kg=)<d;Jh6SWmV@(-4mwl75
zl8*-hbY79FTW2;;Ka*@Kte{jLB(Aop2?~*Phlj<cIcxiA5=#gxPLVnUZ9~$#MC|fg
z%|OcyI2+zDlC~IZc-H1!3oW@2{zN+h&+l=fzu5vA9J)+q4azq6+WN7b*@<f=9huF(
z5d8k7{)(ZLRf8J{c~HJWW)A_E3|8neC1Bv=L-v3Ao!~UD^5#Wvusf?dm*O&(n{|L9
zBodb@9uxU_DLP}WghzjAmyc7yJzRem97!_lKwFv=LKVdHlZWUaLFT608x9udrpp<R
z+f#2bv705ao2cU3&f7EU|M^q;x5xV4|96Sm`EP0k^xW4}{#iP4EAQgswv8`2r$qfh
znC4dDs4nL=spDTf2)n;j$&=`li{Iw6Daq1@=@el*Xp~rXNPlzaOU0_UQt|i$<hb`2
zozhYtH+LDvs1E*f7fj2ZREx)k{A}=--#YRHqk35iRx=?#)_~7-L-^B2gVZEC^9`$V
z88|NC(1e+4xyUtANVW7!-NZgEq>oEJCv?~Uw4Te{L@Do#)PFq2>P1;LOl&*e{?3e&
zh-khOy`n%_Fn`pc1=;0Lv!&`wiPnD5Xkm}KO3DG-b`WNFcfEmyl$`w?qQcd-L;m-$
z)ZDJs!o%Jt>z$u`$A7!NviLBS$L=egjD&rB$F9z`G+GNo0&JFrx7)H(Y&uDxTqkLr
zfrWTe017QTBq>o;deol_^pSgpH!_y0ugNnY^DlzwIZ%C9j9cHNDCXQRV_=cav~F{x
z+*6(mt1~=;F1@VA^RvQe4Cuk5n|e;}4`Fk2+qZzQyfrN8=Y5^*+YaC3+8@x4|G_pl
z3l{Mz>%V-cwb(jnQ!?vxCNt+0GUs_k4FAf{SQfE^-PE>V{8{HYkzGOIR!H1<9Y$(g
zadcoA!3^$do7M5cdyg65-V0LoR>)d>L^t|#o@mhEOJKh4S3&wG(IhKTFb1G!N?P4j
z9a2YEP#e-fm5RG>JWSbzZxIy2`XJ0~4oeOnD4X`u)aF$|Z>uRf)`JvD+qX_V$LkOu
z9+~H@vtXvc=EM8KvR%n6{*0x+N`+uzg#6RJqOKi{6+#W-qH!i=n0o57(6O|CN=ExW
zqL(A^D>|du7Dha9S@(17cuW6=ZbY_+EVSF_r<EMf{q0B#6fBL52;Qzu43(s)Hs<U7
zca^F90_xpHNf-?(l0Q`wSrT>U$iVwS6AOWLX6cNIU`uAk5#(<rs`?odxc&*VEx%`|
zI6ERc5)(<nwTEC;%8D85vS3lDV2v&4IEJS3C7)iWggmPAiwTo~0V3DajkkHm)%7GD
zRv+~4y$ML~xoSUupQ^QO>&7>@5ZLm(Y^K7KZ3{C(7EJ23slh&|ee8h|zQfc^$yiox
zDDsqv{4BD08&WO0f3~B!d1gI3dGZjWrJ`3%rdg>{;)`y^0VMYwU-%@D$FqKo=0=rO
zW3$Goo+LIn_eqb8!k0vULm133ie3Y*1cXp@257<v{_!z%6xwukYFiR(@Ci^s!*O*h
z{LHGLRSQhNR*JA+)$X}S0SjF#U+=yljppVBNk#j+35R-QV&kx#1-6L1cBrp>ToHk4
z&vy~m{9dU1rgxP~-_<-t`^Nn1;NI8t!Q<@EB3Z@dCkNT`b3}hOn!F5g_~p-Rc3#gr
zKh)28?@pPDtKFJi^kdYH6crKnQW~jkVR>XWB8B1TKw%8#2K_zW%$Ry?8zIj2;@iEM
z|4*f%K5IrhThYHU^_oZX`iS-rw{EkO+B-$;B2#Qp(`!W&DDQ$UTd)#x*5+vC&Hg_x
zT&9TktS!qE;7^c;=KDhn9rLV^fzSSrwUT<E;Plk9qP@Q|^h;YwHzumY+GOEV=K*4i
zaHJFer(n>{mX)P^eX?(vRxLgL@;s5B)%Y(<zEbknA_+nBAG*B<T^08>-}_&G8r1D>
z`z7ge;xtDu*PfS|Fy?X%vy)83)n}%K)GUnWthPAXdRKRgtEj-287VQg$+N(3<0R8-
zS4WLo;pW$|$m=r_l;CBw9;qZLQSVnRSNCiDsCcR3la8KVZGjZJ@3^DzqIN{RTJ7`#
zLF<rbfS<#xdFr=yBAO@o)i_!_zgn!m@Ci?VTFILrBfK*&_X~kYew;p>KM9eV-G9HR
zv^=Quw636(kLz$8rU?ONK<;nHpKU{J{kxToa*Eul#E;2{^T_U5cj;Jy+=&AJZjWq~
zKrVf5LM1)jl7}+r<bId&k6B-p>PM&UDvYi1d!Z_{9~-Mf50_y?ahhiJRqyI#=tE&q
zf6zIneTLqX8|S1`w}XROn|UW<NF6|q!GdPFGc_+WAmKecnoOC@vbQ6geJ<mDMk(fY
ziv&TZeM{=5=dU!+_9xi(=bl!}{`_?|C-wmu7pn#G63<oWyjH8yOzPGM=<ByIT+*!$
zioN%gAw$?rUMx#GJfShygJpnAzi7b!MnXPdlS|IGV{piXP|iYRiH`nZ&LvZ+<oHVt
zjo6r&MiMUDFFR_nzF|;&%%GzW!6%f;h`|e~>aXlAtW+8djTj;HE;!t*L#pF?-Sb1e
zLRQrfdy2}<ar+KqNMMH_p$3v1?0>@GUcr34#QY-V+cWtjj6xlA!jMOucVEBv_oJfR
zlYUJ0wLVLw_b<GOoL(rRFk&#EN=Rt5Ejv`POWuNdWjreQ%*?s|Z@qI#AG~mj*>*-g
zAYsIb&xMjzKXR3j{QNdaesTobfte{|HI4k=gfV3CTp{Dlpkaqpi_jg=mRJZ&1ctzV
zpIdmv$$kESz-U4-WdWhcwaeYarS+IID#pNPvhPoI7X=oosO#Eqn%7{Ji^ABnKxE+O
zOWENOSjRnGr3)MIppb4&iq{F$tl;ZUFUAQYzX?iwOaAv>EoS$-2Mxq-=<78s1qd{B
zBk?Cy8szpM6h!4!A=3|~qKcGH)a|-+Rgg{9mbiM)BrA?K@uB9&n9Ky-&Kss^s^Xu#
zQj7b^RJBV@gEjJg#%hA(5@`a&$o{khf<a_op|>ueBq*x3Je0sRROt!5SzF^d>8SaE
zb1uf51BeZdI7`6AyX5*K)WgbjdvQZKIh9^_C1+#YOV*38&s{u2GIlC64Q76uWNSJn
z9XSt?SUpCXfaW3JZ_Oj>tG&dK=RsA=O5z4juFJk>v`PqeEq4r=XQi5!t>H^Dzs1jF
zf-slL;O5nuzRT$U*-!|q`FP}Eui;?1k@%31%jBYeX}(c&zR3+&%GAES^-yfD6xhz*
zJe(_nEK{JFHo2bhxP_<^qR|y^%Vr#0TO(QcldzR>xiz^2`SfUt^RI@>rdTIRuUe+U
z_F4uYZVjPiP7R*#X1q7^8&7m3!s$LuQ0JD6p28-WlgpcGW?1bPyJjoe2?%0ALA_=y
zW3yi3;8<3rDcBD&T!9Q#BJS6v0+5m#9^*SCN<W0^6SVhf<RmjIbO#|ES@nk&qN<Zd
zegaC7xQu0u1oM9AwJed`lH#3f#fYXK(q3)Wb6LAVa;bT;Xm|?r3ec<jWG^kOwGl*7
z+H*6Mvy0LdCE1B?fh7tBy$l0`@8Zwi3{s$xe}mk5`}R|7Qf|%Cu0)m@PK6SKw;+t*
ziV@~`sHy#?zMr$F*bTXKJD-HH?9inLUxQfDd87n0Igb#0ozFJ2p+d&HqhDnIE_)A)
z-;;*Csjw(uruQ#~H4W0TxjX|cyHVlNk3T=4ZN;Mf6%xx~N~<bQ2Nf4&I5(GmAuAEx
zPYcP$rAH^Jx(D6eSTPs-X#K0oOp&_DC1}i?=S~y@ocFfZyRnVE;SvRK!0i9x9?nk)
zT`#+tZ|e99;QO_DTmM4}<9lfz+<=z3Cn&Cpy<L6dsJt}LX`Ba$&dlfRf*c;KXf9u0
zxczTTYAL-A2%`!S>x;7E3h%OhVvf>57eD}qWiuAet9IBNH@E?wG~s$E(#0FusBEUv
zkldaw{5JIOOJ+YT-nQBGm`yCaQm~@CXZ_5akn4yxiYLn=a5-3BTDSNO6Yh)6LOf4>
zMH*hzeezmyUQ3SG_<>@zh|hH4>AyJZM8gY<WHNV4XPnV9V=}Th)XF&=eM7Qgc`%kR
zyvi}+{<}}8ge4XP$kzXagA$6h>Q{cFydiXOwUPa)?mDsYHiK>jC+&|>rm<Q-X-G@N
zOsaogk7A$u9Lb%#ckakZK3882LFTwI85X*8g(LEFV^iP}*UNzot-PSzUYOJdu|RPv
z_+@{T0!cXt#G4Uzl>0q(Ra=?K$@%w*!XD5h481+CIpc=+hdG1RncZBBW*C%-eOyQQ
zes7w=Vie4W+Qll1_Fq@S0Hy>MsQQjdacaQ}*1nd-tI74V<)s9e<Zz#|5f<z&$weA0
zI)!QKAHN~y8<21yKY$S@KZR}E5&)$Mm;?y5jpltJ_tQfbDKkxue*TdaUFjOU@vM*z
z=wZNd^rda5*|d;1!vG`t+fxR4buJ|Q*>Yq6VJ2n#m~h?@fVF1xC_Wu8dn_=gBMOUp
zl?P30ElSh}e@q?U$4|Gh3VSpPKm!j!35<d&N{qOkW~^@Q#gE{9r!~&br$5G0m&=G5
zvpDjEpJ~?BEhuigZ1Tg9N#JHbUa?-$%Wibc7o#UymZ4yhONeC~ymjors>yB!KfcH!
zit0t+?c&<(nNrP;ehKTmO_mV1e#ROPTu&IlCi}U<0`Z}}F@_ffAJ)S?loip<h6kG0
zCb1#~&ybx5_xq%Al;T`_OIR{#7Dommik83|jGUlKT<)$YviKzpI|fUJom|#?T<{Zk
zU0O@X>!|k8+>Ph5lSUJty#95_a9lr-Q^YG!rDeG8oAR19nx7ZML&ls3kYm2x4g9d$
zqYb5y{lrbB3<206eDl{A5;TOQ=jd?^3^v*lhUpm)q4+3foysF+LFs&16yTq4p4dmB
zKi*B&fevH$>JWecxN3c<J4C$lm3|HhJpH;h+g|#t?f1>6$#(QF!B$mkTn7^566u)*
z%Z?CW$}`>`S+f*Ltslede<y96Z4#PXpV0W2^kop8A)1klpj!El>wu0SMw{=Hd8Xa#
zEQTqz5Q8X$sy;>7q4NWapuVha*wOtV0D&j!=)p4RJgadL#I{<T1>$f1V{0QaB`9QD
zv5Tvjf>!|5O%!VxO#u+{_DJ^bKQHhbpUV8~7@d*D%^w!v?vxq4DU*<-Xui(<x-vlZ
z(Vv0!7Q<`!Lc=ni9}NNd?&3^d-{YWpn%;#!ku1TZ-4RRxgN42Hzg7@Un7-5TgGIlx
zm;D(%2XhvOr_RSCJ!C)>Hce+K*Jvo&GM@Lda?%@7C-utH{=B4y3d!>A;Oe%9PrcHe
z?GJqOJ@X7+c9YY+m3Jn*kM)T8Icw2qf0%+i7{{(NkwV&fKhaUW9U9)3%@FM!!vXw`
z_IczpHB;D*vRmyCcV61<G`NymQPlO5>a5)VA?$7s$_}tDSY>6^lk~_f?Smxi7R~ZS
zqmubDAo~G7oZcNJtI=~`s~8b%=2Br8Jkj^pNVTwA*CyobG_6L8Gjo62JR4Z8O_y!=
z@>ut0>#`G{XkYDfzt;V|Rl*f15ou7jTQcdutRwV1WrWI=ko9v)E~g`kP?p8pGdpoy
z)U(JS76sM`F4@wy!`GrdOsidz4*?5bB9>wG{Xf~O6<@uZ<-{jY*l?Mcd^^w|4@hAU
zlGIRxp>qVGy?;flUsj)D-A)YtGFHAoD*RxAP?*B*$Yn;qlt_zx7uQI_Y5!%?tjMH9
zb#Nc-+^24K;W-nv;OA+#kp63M!>!Q8R@-10I_+S2VF>Xj^TEMmBl0RIiMQG^z0Z2i
zv|;H77?@UEgUwXQ{6WMM6cTLPnBP}XNGg(eU2JP$H1O0F>xiI#*f7gr%%VsHemfaF
z7@7HJc6~!c-j=?0v_y<Q)QPNG-HChQr|!hQZ&c!U;*+2%ZN4s3{V8FqpHS%)gT0lb
z%&!M6yw>knV9Rpp%<=gN_h<pHw60~C)_2oDW4z~#kMa9I$&6xa7n%OHs5`ojh}$k5
zb_!TZIAMBZ^Oh;f@y_Jj)<sGZ9z*+fWAB2_PO{TX(;|(<;30?7;R^Lx)%V6!x`-FL
zv#`ZLJI{>F4UZpoN*;-*GMhG9*_8=qEcs1~>eDIH!(~YG#4m;$a`Hj{7-KrEtEuhy
zc?gx>L}+Fi=csk1J*`t}Fj#cM8?-qR5TeDTVz<W9dXKrk6q6+%2lNZzkMlpQn}lCy
zT#bF)c~E<EFbA*iN|Kh$4Ru;QTm*qb5xT}#3s|BZJ(f{2a}{FNPzZ+HEZ4lBbtw-+
zD4^AZhf&*kl{x=+t5smy?U{Hy^S<QZ{mmO`O^Zn%jo4xJFr_u^A~Vu6FgD`_<PzDE
z)`LD;pHMB3TK+-Iaa-P}{JAaCHKp_-DZ8LT8f~-@Ob6?bPqKIek`hEBl6FWW<ySMQ
zGPsOB=P@KmD+0tt1rP=FVKOX9GBiaf7|%P{S$-sd-4UE`MV`i*H%7U(@6n@$&dL5~
zescNHi}k<SD7y3>{3L|R569KVE$^><`=d!ZrY(M3eW2v3J)<GHW4A9FH0R&DGIRGq
zEG7UXU6+Q1V717?T4h49$@xH?8E|uIn(KBGOq}Nr!*9<J_<M6rhKUfQTKHuZIp^j=
z#25fNIK;MQ-Hf7|ZrxuT^c7-f&aX@~#Ub6UPlX+ftBn@?YI5bTW!LcLzkBufFJEuz
z`^zKUPft}%7R=K6f^(De0<?an5|qw-n51yQYygZ4@M0k?ODLDUh7qG4^$iW3@;M}a
z4+O+SJ|iS1(BEQ<j_Y7U<UaJ%U>6Q4D!N@%#F$2yB9VZwCY?DG#P&rh#!VZxIc$+e
z><E59?ND_vxWbZ9t&|76vg3wFb;$ZOW>*0m@*Vy8XK&~Y);oWD1;mJRW=3-Jz(uW*
z0@_0XV81$lZ|RYYTjO&Rc;50_`IjKD3)Lo)JWiw^yiZ{neu(kiBjy<|V{^X)Q7~AR
zWh*08s4-kDsA{y9LA%2bj0v?7=6Oe*n->QO$5rFR_sA90eu(}w5#`gE|C*CsZCoTv
z`ao$9FdslngW%F4-NKG97k4aX6*+CCdl*LtSzbVHzPa{$p=Um#(mF(#p@z6obohQO
zc91iG%sMEy$P;VRMzK{?$@96Pur;9>xmSZ#Z(%Z&?M(t42tOwA2z}sy-Vs)pj`J;Y
zC=_i$H^P3gjBRmWf_)hT%$#vy(!4(VC!I{B4yqb~(B*}|Df6uDH|3ab0H8=#!uycK
z&kLEe1ZHnG(H?XJJTGSRe&0(&sHLZ^9;oHzc5UH^W(N}sO8n>h7d3bZuwcU2heW}Z
zPxzD}ji~khPyLN;EaSKvk3X9rpy*W1g71aTT4l=Oj}?%10lcOBg3a5@FkC=RRxLei
zOar2!1myvrUN{ii->N2_yZJ!|FybRHn()QWFIx2vbUd<(FEzj24OvRizaG<HL8d;t
zb9s5GTjvzIvSNBODRybj`sB%ewAgv{+`2-2-f=}^AhnxO=yiI4ulD;%9yE}E7Nn4#
z2^~nSAG!}J@2~f><spnWFi<6=qEw4^{v2HgoTetOw$D5=RUnA-R!SG#b2)n#4a-1T
z@Z(SJK?<hiWd%?GOfr`uzM+HYT)(i60#2Ie91wFLf9BxG<xhf+ab^D`I9t}rNf%%p
zPl|SM&SxDQFgwruL3M!B>7QN+9#TEWLT&}NLM-O7Pq*ZooXsyObAQ2QcMfx)ZAwZ1
ziH>k`Lv-~2Lg{>0x0%7r)yZ%OwF|r)q`qk~RJ6NmqKqcaGz|COPvvmqGcIN6_<<ty
z*L8ulP7i*$Ew9o-PXY?kRjFyIMhH!`5yDR7?keKE`D(p^X-g;NHrXar7K6~~&ziYA
zae`(Mq-Cz*a4%Ujz9H2=y+@H)5@iZ{tv$WVc-~Kl+R<U|q*$UbbSE@;+wxX>kRl=o
zCCdJld8LWxa5+w!{~-)ful4-9>$`_@uRd*nw0eS;n!xlIsdc|kCw}5#?q25$o5Ke-
z!aau+dN+$L*xtLm{va6NUa`NK_jc*zyy=(VKWjCw)G9E8=0@37n<dVpyZ^9Ubb*pl
z+4YBE%z0#A({8@zz&8qIxR?JUk;1C}Z8K1q;u`}r&iad*Ebla($dAT+A_<J4eEr|E
zk^LU0TKM1HeZL-vzW8b)Yb}6zRDJkc9QT(E)__jzZbC*vhDs*9!7BA$*k-e5eAj$#
zOF+C!?1<Fmk*7=-clrL^i<i8uR|#)^jF-B4m38)y11CU0x7W_8MDoqVdJSGJi%S2-
zmLt`xWNbYdD^ec;hA<jSjHhcR!{2uAft%-qF#nE5E+~h8$5qY;)Plr}Zl{U%aMgd*
zU?*kU)2Hs4^x5iuL;HMw50!Bt1yB>xGwc=yz#3~FlqvOC(>(3ans@Nmq*Cb4RD2NV
z{zoUxQt@;kZMi9Tt%xQwh3s`&_JB&d^Sa2+gC(4rp}u-_DeyH_l7v2K*GFy#`UA31
zic5(r+5g;1{d&@Cr6%#SOX9YdP!Z@-vGf@{^kJxRmPIiO-26$???PoqV;}@C@e{VZ
z%e;X)P3j5LJbaCOgM71?b*J(CH-TPglE>7iuZ(M#X&61M5to-HtllZt)4u(1u*ew{
z=2$mkG5PMlK82b;+T7f#)}ml{IT_Uy$y?RXPqP1&bX@F!_pjqD{q)83yRw%9^Iy<`
zxv#<F>Qts6t;VcH9+$VbxbqP$KEE&KiHRHsckvFEcCPu^*$M-w|FaTAic$+T{cP0k
zcHuN~K<$99e5RT!SjC_yNg~ph?jx&Aa%p`}FcMfs!L0wX6iw_pYEU6)0wDUC)4nAc
z=G*!HTO!_%yWY2#=i75P?`LmsP7g=Lu1D+lS{|^nvmgFzWncX%z9tVBJ^9UlfnWU2
zv(RP{fIm>a5_NKC^CFB^A>zAWtHnGVV@+apuwv<KozMaPnnSSZ{{T5@9AU#mfvp6t
z)_hbD2!bd8S8Wi%7l_q@g8(M@YR{WAomaGS^cQcsbXuh+h=9&>Xk4tlxOnn0$NEFH
zrOG$;Z_HNZl=UT+1K1$Qn~)b(Sb6PFg_Vk7t_wZI#|H$?8crIdFIVycE|^JSc7!^|
z08pm>%tk<LNAH2YS3g<1k?9a=Kt(x~B&_2Gj`N;LTqc1lk5snlC5Wtq+8>`stCLdJ
zh7D1DkiIL2@wkE#=G=dUXr8317a@qxTp39OI}#dEUjlSuhtx~n@bRYaKF42NqftgN
za0sEb^$FNC79H*iM_(dUP|G!GXZdd}=;NX4E9zphg5uzb)0NrcRX_~Pe%vmSND{}*
zQ6$2GOYAUFWEVu-M~(Z3;@=6pXJ1YX1&1>Yv@^Io|L9Qr+~HDc_#G+4v+nJ6+lr@;
zY=L#v<->n{Ivl;(wVM8iw`X5F&)M0x4z{Gjsa<U&)2^@AT<px7g_V^L<`-E09V*r3
zqie<bzv-3K2&ESML7|7f_U*>tz$odxwxzC#Vs-y_WH&3PUsw8cH#R#K>Q#@OvF7>T
z7_#=!g?{S18R~q1&#Gru83UnyRSta&e&_3e{-RzpO&+}@gj%;n43|(T9UeXNPpvfJ
zBWuf*Hc{i;I75W_kOOVWLu(AG0t!ilQs)D7f+16M&y>9ADn}<b+Em?5PDhFC0|rJj
zd)8m}D3Kql;<;FqX*k$J)5E`nWXobU1kez7<&2imX9wMjcSgbJ7$NG3Q%=a1jI<ph
zR*Y3uw!o$E#F2}FM$G)nCo&Vqppb<xJ-C-k%>V^~^v4Vi_#7&~YTVNSU;`F=`w0Rg
zuET@ARUEV{{z#LE2;0nv$HIKjI<@L<QI+D3nwAcXc>*_qc2SqokK2AC9u3_UGmq@Z
zM?WCAYWof(A4=hWTuKE?#3F_@^!(jD<7tb+Dyk?t)Aith7ZNJ6c+Vk1xC&A~L~p>4
z5>{3!8FPrFkDTXvU0VHP%y?a9T2RxdkbALbCnDA@3{5i+&EFTqg;}SBC)$ccVpY$}
zuLB;u78P=OMxW1_$UnIy1Ux2#JS3MV|1FEZG1<U~%l?mrE8gkih!U?GSO|k%;jW`4
zLY35<YNZ_NiRR;VCXT*f4)$j&&7r%tmI!&p1pOk`H?UmoC0ih`<S~0p?8(@MWdwu2
zUXnz3pFnK3!Nzy|dicL_SXXWeepdI{@Oeb~^G%oZRA$`h_chFMp$r_C-LGXv4-zI^
zJ2%K0lP+El-+B1dBVq2Ue{Q+Z#`7_N8V=V6%j)a<VjpX|?EOUBaH{+NuNfXk#NmaN
z-*pywY9RV$UeX7w6YD`*Wuqc~X}B5Kof;*E>st>|_$m^%G57A%La8t#cfUwD8+H|@
zQtT}{Q40sSY?h!yi7=AR=XESEVbK<r7L#`UN*m!Caj!`2pKN}NYWI?mqHd21x8OJ`
zTSr7^;yHfs80J{PfP^Nw?jN$f)g^*&0nuXsDiAd{bGb50W>WsW;ab11^_9`OqQ;@9
zt)fJEl`1FayGfIad~`N5Tq1IIxekn_y=2|#&J_JSBOT+48oz964w6B>SbEO9K{&%0
z?TG5r3++5pLDAPH@XA*y8znohuNDP^{=S;h&u=E^TarlYO96R~m~n+Bj!Bi?9NWoM
z{5H?&vfsn>R{b0$iY`e?FG~nJ=9Kp55sMBIgOJ`;&#<GuQr6p^=1_}p4<TB-gjwg$
zcFJXa3*92U6|A28-%KC8S&UCU`Bbg)v|;?`iO~QXp>{5Mj=u5wa4E97wh67bVN421
z$JM06>y%XH{onn>yH7=>g4&?am2S1+aVz@ZYS4Qh>GX*P1W<2hc109$8KI|1?0UT*
zWb^cCH*kf_3Qd<f5dk;;yZZBDzy-a%;m$9YK^i!-CuqKm@41S0PCKJOeKz;+dZB-A
z?-MXSM6E!M6k)(+Q~I*v`-)-=`|pP#R{__bI!{R7BINJ|^&VNW0FIeQp}8xZ%w^%Z
z$)aMnQEt+s{Ks1-*PNMyc%yBkrG6YwKg<Mh1Zgw9K5B?CA9u&Bm)%1>B=-M{2=-W)
zq6tL)<7U~W(}}8@r%lBlVlf5GclF-AqUcc`>8B8wTD-@3oT41lc2uvmp504)>?1`A
z=^H(D?ZqyWuyc8{f<mfAKSaJReC(c3yFn<3mt}YTi#PM;AGTBnGMk5dcK`ly;K7p8
z_{-`>o5P{8dk*w?o~{j?l96<TQmb2}>0BCwyA#TTD-hV~WO!)XimRu=lqkea6dm^D
zRX>FPN|`Vn!Kt9Tt3V(S``8^cB!tcVo%Hzyh-?xLJHUsYz@{y(M#^a7h^i6gVpi5E
z4;zqk_SfH`i3PyAh#q-Se^vz`Z`^}7&|?BYj5L17-y7)X(9#5b49Z9lKRFI-4cuQi
zJ+#c+o|i$CodNH!fDNuY1eGRlA!jUy^nfVXV9U0t#3r)`xUp0q%GV3vNsCJ$Gn-~b
zW$kvxD8|mY+ruXav#i}!0Kup$pqK3p%fcywNZTM2)Dw_vhnt0vw<G+a)%!yW&@|HF
zZqiF<iYg|hYnXu7Zn~NCaNyu5?$R=LcfSm#8KUWXWO$}Q*XkX35juB<b+giR6%QzN
zyOS+#uj>i#ua`S@y9!TFK2u-KlW^Y0YHdEOFW#)sFM^z`A7Ypi6|$$)4c>Pd<$wPb
zn6?UiX_5#|iLnLO?b^AXkAPWvQu~OBHD5dR#MeV2;yi|AJsbt1|Au*doxi58?kB&$
z;xL05?j=1ra(o>L3m}$5d)<N1{x<C&*3Ez`$cSy{sD#JY`zp*>y~DDCLm2={3{^VV
zr|g>O8H5tilVze0b#q5ZU391amb~oW^ULI2a2i2Hr9w$!uxvaUY{fEix<rx`v2C^6
zp=F&Yg8|qrf}mdP=<OC3-0wi0fIMTPm`VO)$LB$;oO*mUHZ?8cf|pW{O(Be98CqS+
zZ?Yb!misKZ>nz=Z6XxGAHC743>;&2Xg33Asj&rSr{hDqd;ZMKXN`4bUNz$M%)H!D=
zh$979GaZ875wTv+T+T7Ko-~0jY4l0!an?7v9_6{rwF?u+P>2C8sk!3u%#wKIK{>s~
zw=dpfhE5K!Qef^$M1nyF)&&KGVo0hk0k|*u6Wp^!g{KDBOxeO1V8C2h2V^lXKOy1|
zNn|#4feXF$#8P2rHfil*hqsl`Ae;t|0#I`bEi72)2AH;8zXi&g;U=*Kf})>Mg1)tq
zB!-Gdkkugmr@JY?$&>yl2dOb7ggdv&oo(>_5umD#q<=`5D<s<e_3yD4?X(w_;Ytky
zbc1p~I>JPo<35&Bb_OXSH#!&6tE<FKHo2rIZX(d&z&l!mrKib8YD49sm{MkTh+gQN
z2U6?iF8bnuD+GF*qZN*2ic6_TWR}%iF!^<*C|4(@XzQ5_8;uOV;|qJH-EOB_LO~ZK
znSUs@aNXYve{OIkB4t%wtR0?7mqA9#&Vx%P$l&kH?<!C}f%XXioy{j`!U4HNpj?yv
zp&cIj);uO^r>8{BIxqlw!*5u>%b~_MtRC23G<2a%q$E%F)%$Fm7U+oW1pLC&y!vMO
zn<GC&;bQZ0kPOO(KBCG8XGOy1uyCli!jnR2IAD=*3a7eyL9Oe1{etXQbAER;(&%)O
zC~o!Dd@xU<FXLEre{UTA4&@U@4u=GpJ4rrea^K-uAz!>R8f3i>o6|8?Ql8JMPwu$>
zyGhiomCiogU(Ngg@<O;SiRE$~ahoWlB>^H_`lmdSJ}}k|TaO3}z#Sgl*Ck2ZK-G^w
zy~p~BOT9DAb4oPw6|Dk(VR{5PGgn`18B|sdrlm>LqT%nQQwLt1Mh-%}3LbZ4c7vSF
zp<~2>U&KK9duk1zob3pmH)|3vtRt$_8!wAo!MDU`zFlN59y&$Tb3Br3{$;vI*D<05
z2x@qupgg;b6<C_J-!kygO&RkE#L<&Q{nv_*AozBqRQE<d^4{KZceKaFXquUw-9F^w
zAz@~`Q@dGIe13j2k8|@;|EgVo`Jbk7I_E9cbYj`mAw0W}UG%1*r>(6m{)*pM{8PSW
z@x!az{0$C5{wQ(^s_;7EAl(_NA7Xpev!o>=k)}m-l-Zh2R98%bAGO}hklY>bb)cVS
zlGfCKr6!fs_K;q!X~(B-Di3G<dEWmp2oqA$y1bo6V~Y!xPX-1^es5pa@7RWdP9ah?
zED=TLwfMmwJ6qZs_wcNr7$QqHVuIZu83b2ghOy5NaP7Ijc)wTM%FC)(@Dv)$G)go!
zdG(uNiJ#YkI!Cj%i+XQ{jOPQA4?to9X$m7wAnTa&c{tAEmu{cxh~h0Q7mtd5+bxl=
zeaYb7o0TMq;%>|qnOq4$8U$WwCI*#8c1E?fG{htY?AYU7u4<;<G~>CbYNokyKSu~5
z{N;VU_)wIQ#{8WzlBZ%uw;~BEiy#$az>fymwnY%RAM4*H@dw}w(Ux?aAR;M(V(>eW
z`f(Y7fAG_4(x}t|w_O}fd%!|}q}#o&p2Mu-(U)l{g{2QNd6Jwwek~fJDjWj5YA%7+
z(iTj}eoMy^NkPSHn6c#srL*ONx+GW*Qzwyhf#>pmz}dp!Y+`jVEW>FI%oVB4=1?nL
z7ryI7h?j#-!>!-LvDe1J78F(7bGIPh)~rZ&*v(XT)LYZLP9cuI$C%@4V-n#-$JJ_6
zAuZB0ETLusmssqvG9y%01sFK8LToQ&?8GO~H-zq1GKR?HvA*gFLS)JsT~N0pC71sx
zhkcZ=Q?`~CehY^AOM4p>N+$D1ik}()jTH!AJ=uhBa5FTx<aGCpzy2iqS?jx*0GRu^
zsm8TdCH%J9HlaTnObTpzODAy7GR^tUKj-lOi?QD>ZOgdYnRglUPc366@_bqw7lxJz
z2lpz48CEDnD=ae=o|xO4aCLT(Xyssa(%P`pJ7xKibbtoIh>h>2EbG1=$;lh0WD5HQ
zbf@ObOOh6DB#3m7A2m(xgux5>U#TihtSOYF9R&SFo!0avl&h70X2+VKuqFd4agR2`
zAp_tlk7n+RkxT3>1&FzOCD+Bv_Yoh8)N#UG51c6;tF#qP*&E>FUB>6ZN4FBKnGASK
zVgkX>RcR?kIlGW>Umk_lU@6!H@~{oO?Zmj$uNlMhKqng1vc}B;vZo)ix+OGE1cf4(
z*3~~1q=S_&3loK5&I?g<4$YgzZCsjw`wsL)JAnZ&rF=krkIU0u<S||MXI1BB)dWH$
zCa}Wqf;I<b=O}(&sJ+fK0xOs(RmZYy{NesqZI^l@0A`0(wcm%XhZD#GmAGV#I4h)w
zBK%cg^cQnJp_Dn?x2(`DyrYKtH6=@B;P2KEWoqqY`tFwma;Bi3ijD}K0qDko{%7y;
zS0{M&Wt%^nMw6@Um8J>m)dG%p0`+AH^-{doM!d`I<=kNLRZp?)@m~$EHD@pXaF1nL
zkEtrpsfi<xe`m_+tUEhw7rvr+rij0`wsu~Vb}4w%Eq2bRe{_c9wfFg|h5ZVt;mz*m
z#ZA*V=U=nV1&J5$5B{uRP(N)_NP?!-(+8%}1-Y3lL*y5_mLVO=YN3r3(G2LBH2>@X
z=aiHdY=m7@!!RODm9LhzMZmCd`(^L<0!3oGs{~*PF_pG;Cn5m<78^+U4~ILqwZA1D
zPwi<Cx#gle>nsj5-s;Ixh#4|JY#b-&S7Ek8l=7v`tlNE0`uR%(;+p5Z!NLG6{3{4m
zZlZpDLPlLnOn6@%QF2xYtM}U5*Sa)c9A+joG-`=rSjP_(#`QzaRFwal#rIzrh;HM?
zl=G;NXX*a7aEbZ0_C?*E+<((Zh;Of5XaB@hzZbo^e5upwRJTPyu#0GRqYhYeP&h92
z3rd$S-@3z&Z#_p#;0h+Xw9Loy$J$?4+~NYSNQ$%G6j>W{S@Y}EBBFT-qv+aCP}ToD
zXqTKAGI5j=DELa_rTu>HJ=$^Mc)-_G8D2)jk@TlK1>>EQ$+axh_YZ_KKi;^ICl3?0
z&^-!}O40PBRN}>G^Ph%3yvOO+GRV#{7H?&C;fo#2<KJ$|u$QU#+q-ae&fU(Kzs2<Q
zI-3>T(fPiId^+G@nBU3BwvEi3hdA^+?HeV(xN7lFHO*Q37X<;m)Z2TmAtmK;qe;Ud
zzlQhg@$6#B>xr{6%;}F6m95h5ZA<}BhXqZb(aJuIqyx>(P)wdfBa#py7Z75=Z!j;7
z_R?8F4Mr(I6hu-V<UW3~$g{An9yY)M{%Pc}X@<axnI+E~Ulux9Nfq%&nRd|0GUDwG
z_3Mwle!BU@Kb1nYXB`A=E=O8x-uKXthSu(`2rpz6)i$Y4(Aba}zpVv*wt&cp3J0uH
zK*yJhis1f0+{2*9zx<~$blczL1D=X@rQ^Wv(eJ<2>r1B+ujtHz87?KlL&G{9yO;ZA
z7y?EJ#qhU&ZDdF#eY@%|dkrC6=FHN`4|@{NWn8W!$st@1p!a#-X)DaP*vQjN3;{+M
z*$j@;L46eA<76&);3$PrM{!F^G`%bVC0bCuH|pCS5Ky2iKT<*)XGym6hP7<$!|!eP
zElgcA1+rPMcTtH(exYGnhfY>$jU)%F#vh>T3Kv4?jYGu~CK(kXJ78hBZyh%v6U@fv
z@hF}D&JqQ0)^+as5FG0Dnn%!D#jFo_>~8eOX-*Z@(__9dlrfo=6dWviITUozkQKvn
z$*plm?ZmMsq$JLd+~QNx?FYrRtz32Ss$cRD0Rxy$zFAexj72a^ae0Wyx6_($a_b&b
zv(cDjq$&K6HUS?Rb6L(*#0eP%p36#4oX@|Qy&cn&@TTDTcqi@VD982Rv2lL5R`P%J
zrQW8i-umlpo7<#R#`{QIiJ0E7pr_B-bHbU*<GRypt8@Jr7{aa8HyPfCKXJSw(k^n-
zjGUa}=;Q>PcPN`K@|${c)p8n_EzcUmcQ4Uy#Yiy|($a48=Gs7lnB1+o+HiOct?Du{
z+*QAG`Fq+?)wNUR4V(7x(`&?$p5Y1W`2*SkfK0=@zBJ0fHiH@iP84SghmrTXT7P0+
zj3P$%fT(}Zbc4EK@$PVmCtx7q9SqZ=^2Exr)hBpHpEV~})xvrjX;|41nyj=i#G-}N
z==Rq6*^~UyQ^cw~G3?|3JZb9$d0#BI$Dm$WLTyi1ICQXED;gekW@Eq|nKEQzlqbtH
zm=Nb<V7;<C-Qp_5*TM40HsR9dNMuI0>LuaRdQP`EXU*de+?;0x_soNcop1qs6GVp>
zt;$h>EU>ZNRSk<E>m}GB;qnm*AuxZi#tz7SX^$@xf~X3sM%kH4pTnl;g|*9&i1!n8
z!nZ6aLsY^7TJG^&K7s4^@Q_~wWGcjclsuib`|F1Y{lKwtH+ipD=@J~^Do9a%06M%M
zOEkASp|(($B`>e3hXlrYm}H~f>@N=WPxGWMc-eWBJf2Z)KX>J${xit)SO(|iR>|Mh
z3)B1Nr#nf)WAn4Ln=6Fa@r4AO#fJ*sCrG6zw|j;DuG{io-ElO|4%aHRY(fJ@V&09<
z|5H<6q|GeyKedZ(crM%1Tl_Ix+2nj|SFp9l_kf=;^i+P#J5*_O`$Brx(0fXLSyvc>
z-di}Hib`g=aggsOy2xi&p%oL7y$a|YSm@JwFiv@4b*$@DTcnk>+D^_EVEnLlx*5yp
z6@vqcVV~QS6oNhgl}E_<dkmKkEr~R<9m~!OkiT=*?s=TJecrs}D^vl*qX4JIv*?A9
zoroQ+&FiFA4YMHPcxV5uW8EaDXU=rkIk5dBC!g@iSEE<0FWZy$UkV~)h%`;YcI@;J
z{wIKGgfSeJm^I@d)HM`I6|}!Hg=z0r$0&V<k(<=KC!2N59*_tWU-971JgC@BbDsdJ
zB!gH?<{xIAJd8qOhid@5=+@eSj)KFU`2c%kc4!UgX593_4e2sAgd=3PlvU^dIkUgw
zsNkGCHcJNK5-7-kZqe2wavdssB+9t>bA=p;t5OMcfIRA04ae)u!GKw4e`t(hLxAdC
z33NGW#v2V&e#}0E{lxZu@=q9>Z%bK+sZe3=rAG}@GJ_ji74+9j)qC+7jV=$*(+uyx
zI-js7Ty7J>*l-p4hgzl<%gJJPdES!ApZv?=-jrwj4b%LVcp!lq+-z_xusvAfg3{<b
zu1203Vg_pjL;bOX12)RQdY|80wd<-Fr9OaLbo5Ou^`(9kA*Cw*^OuSVnM5q4Ip^aW
zuN6+&^tHSn!*NpVeZ6`9yYuM@me`B@F+`X#10$pU!EtxDT#;_g0sU=%iE-x{&h3Jj
z?P?#{#r4zMzxvmY-y9}Q>@5bfbMf<^N`>E&+%D?xJ#IR9tbcm?o6N;yJy&VgzF;7@
z4tbGhrAzY7Ua=dtQ^el5F#$lC?2z!Novi+GrbeB_Uol{|s5DEPe_?6O(xZ?fCa8$5
zwRJR849)fF7Jl&p!mN`!HR{|#+G<&wC=ue!`F1V^gSC_8u<aXiNDT@^7e!iM`v_4f
zcEMV=5z$$L$F`=Fl+&i13UXc9CL8sqTbS8JBcyJ|8SZ^k%8GX1*)l#nrQTQVzVCMK
zki6)q(;Y~#(7k~aTW~{9@pUS632soY6cmXVA22`xN`+2t-AnrExBv2u)g{MCbF2g2
z&N&X7G@AUw*#Qp{QzP5porz;&HkAPy9%D{0d<~*`HO61lg2dxZiK=eYcr|8z;FT+#
zZ}QdzQ;K6)`Yx*4bO7iGe04}Fj|S8r|5C1?rf>KMZO9pxs3Yz-3S*!zefjIM!l{ir
z+Fe9I$x+6tlpcl00DUeF(oKiIf}de>mW%r0S1OvZ`Q?|rkW7v}X<mzdQ+4ltE|M1=
zWt<%b;2;KzPk%VZ78JoQ036&sbEQZ@^}(oYgf)wNyLw3sp3>So<B!xi`ePR)21N*i
z5pRm5rrST86b=5@?;qQJ9N^N&J55d}Dg8sbzpsFG#blOe0g;TxHl{dUb#Nv<=nwJB
z#+nH>@N*!{W0uwPW^gI60z0j`zEg5`8H$&XZq3YpBBacb%9jO&XR}^>ub3ZGhtQ>g
z&F)bxPBv|5)D^(Ygx}ec`#}HU<QU(Jf=W2gKeNceW-7ZitACiATMw`Bl=EBjU9n8M
z-oZoVJ?CqzU8Re@yn>#C?0ipT3mz+En)}0YpR;5mlhVuUmiG$2(f04WlL{T^+hWzs
zB*XU4=O)aGlw`z5S&;R)bN`3oA7{<@)7aQq-&4Em8JyeilA3jY8J_<)`S(=qYe82?
zH9xIQ{i$c#+UfU{1o{2hYt&a15^?W5buvFxZu*zAl2km&Xl3AHP{5b{!Wx4w!Vzy=
zGrrX0<gbwYzGGD7;cF&blt(1vLROCcnJzL;+>YLdlNvjT$_~4_aTLOOs(4>Fbibsz
zy7;ZdCFxp!AXP|SuT@-c-EH{Mabq8Ky3@-Y8Cc2@H(&D_i!WKCHts=soBH_jg>wA6
zL5Mw(t3}GKW9W4*wXZO&EJuW9(=W0+a6Z&yvNvMsud?}7T|bO?_g&FEbJ>VtL$*sT
zn0KI%|2!9MO+bL5JVH)wMLMf9^eY>+psrbtrGs^0AVt7j>ukmy!NKr*RHCm*L#?+0
z)?Z4*I?wc`iIug_dQhsx?qA=LAb7kOPk=U_j6D7qc{~X@+d-vNC3LX9+Ap11%d0E0
zgszlTW%HPcQ3*aNNA1=0bCI0vQr$oAAT`36Gb&Q0yH?YF&-#u`j~(`lY}<NV_Ad&_
zR6-JH8h+n%)dDPe{2qFdmX7}*Qy~AA!7>Gia9$HW>NOc<#p%fPBTc<I2izn5N<M!6
z#Ds+TjhwXAbpMBct}kE+y8U9~x%pb;blSoBnyhHkesBKy)9Z~FhjZ7?w=?-Sx9skx
zSMBlRZ*}<ACWOlUvHd$#r}F^0^|Y{20fCs7M(e=ONE|iMk%*tvrId)*E<P1(7|F1y
zE2v^s&GspJOG%~x+b~p=Ol_D4{ITO~MbgAX8Xi#=Whxrn)jrl@qL-E8DY6PKsj%3|
zSkF_eFdrrth>61U3md#qP07-s7Pphh&qva*eVgAq=V=n=A--lNb%(cC?sGbYtp}G<
zp-nKQ1iHz*TUM|A`)AYYlbX1eC~73XW2Yiz*o&w3ZGy)hPQ9nl@IyBJ<oXs<mvvC$
zBUWSGqK3&v*js%`O=<5R5_IUQ*|x{ymu#{>s&Lo#w4xjpJCBjsY@Dh~$I>G>s?8ac
z7~Gs^Q#R3fWI|_#h(cenWfSlc?*ANd*lUfjQfExVjk%Z-tcxvbzI16E`WBMa$DeX?
z{=Cbn(<sL{RLc|5du|}ut;iO%860RkfxkDRsHi(_#s1%)GBN=pB{fCzy8K^LSSPRC
zA%7K;L+f0ia7D8`(bUito|Z)DTp6t4r`$Z)XdZUEO)(Le%yQaqcFvgA(7aNJohBgE
zk(<aPcG*SQ{|N|{bv`TS7POy7TePC|&$dZwevE&6i|YLKCbxOc-7$>1#yrR~t4+_F
zdM@0}8R={y(}%{O@KMpD*i<D&&4k|0$>?xA&#!WgYiQ+uY_VaXBq)kzk}cib!MsJ*
z9|Dsx{CngU@nR1t-dE}n*WiVjeIPs(utmBElbL?Wl#_90_;K@|R;oXun=(<+VNTJn
zT_>$r?~h3dS2H%!YANSxFlXGRXXTVpl3;YOofVgo`Mj-lrcwoMA7i)5_|_1CZgWXY
zx7HuYSgD8ZPm7G}?(#d;vW}r8B=t3_dXV_rJLH6~)Cjx@DC#<*1kz>rL$+w8bt^up
zr>>6-Ic!B6U9;pZnJkds;4YpouI`8)x?+bv_!hKsp*y!>pE;5x>?kLtK26j1K=R~0
zTTzJ?)qg!NH~y##@j29xUUo)_)Ut;Qj;PvH_=(sRw)D%#e>O(XsFg0XDO254+YP>7
z^MaB}LbF}zxTb=5qOy)MP2SFhwR^Pc@0Wl<--M#lDyH(Kp8+Pr<h68IzUJ0rlNLId
zy3dCM1X7I(1^=^j4arhm4c?y_%u0@BV`mLK8?Vv*A_kQlr>Vu%4MG`4RS_<Ig9hH^
z#CGx*?Bh&~if)zR+4AuG>`mJTUuU5te(#|EI<@>q8BP3bFC8&jBcNF&^ii@!{j4MQ
z{k<?-q}?ZyqykSxr#c@jJQGyT--Re(1l_7Vo+{_pcb%*InXt~|b}RW}>g-eDNA5)a
zw53T4p4p?p>)?Wq`uSdqj9y3@8s{%=H^gR+OKgtmysk(1ylvgEIv*s+lIa#6$;t4J
zlY3_P;huOO8IX;>tgf8N`cpE*i#qW@Hi<mB@M(WzXTMH(0z(-`oRL2BfNDjjBty@B
zH|Aw8IwC3Ik0WdR=M}=vY}Eo@6ybprq?34j-9J{`yz)*Q8i`iZG?)^Nb`>q%-GuJJ
z=q!5s&mUzZf1%(Z7(L5$czv}OXm>(HpgV0bTB5reZ088AwOPjdZEfZ`w5lTNQbsn9
zq#5#Pdl9q;6-M*c!k;wX`S8t^c`}UP5zC|3za%AkM*6{s_LVwI-tfcff~tEiUQ9`=
zpL*Y^30xZiHXeh*^L~sP16r?K53RXYnT>;8S*w|4*Q|I-CINHE#M@?TP0}D~tGf5;
z%!_PXB;4@a>rBk9#H=j&-FK#@TH8*UkvEs^Jh(3oN4VKxiQfG~OO3gcFA{~8fo8Ba
zisqDr;Cqmw1C7t38eS+I8D>m8_ge6n&pd_NzuEa~x7+(ZEago)3C%*tMx9*@9hb~}
zTwv;67?DnHx=a-uD5^N?@m2CVTSjp|u)8@YVc+N9P0kSfao%)wLRL?+e0#C}Fii2=
z*;$F*Kmp(RNkvfCtp>f<@-ks6_1<BVLsFJb_$P-`W&=rI?_&bSYG*bX&hryljEl|b
zJB_8Ya-H^7??OkYUy)1HM`F7);^riHycaFrJ}=<+FZX9NW{!@@wKFShALX}1dEWJ@
z1ZCx0xAdq{&7@460qfhH#kX#i!P~)-nTb|0G#@`LE&1ajR6~@Nvs6r77Ii|`;{Q6b
zergori!{htQ{B)4fOgL>L6Jxr)4w8~`D{j`_~Z_gq}#(NqxQ)1a?yb?fw{pH^VPW-
z;lID|vxu@zFcOS*OMg(*SUqrx|2mF4ffGoMp*LmBYItxK$>j-vlp0DhDV-vIe-YPA
zzl@KM6eMZMkHbP&P(7|>6UF7*%ucA%-c2l?zy9E-{EV+?xIz@ZOMN6pm@^;Yi29yq
zn?Ku%BhJ8Wr0QzRN{=7s69JFt3*Ew(yTU0c@&uzAul{<bzA1<dBD-2fR-^nXi;Q%7
z&{n}xxpo+6Y|L|kUBxmNYEq{A=tE++REb_%4)JE639X6oJ(&(QbvM3v{`P)vwxXrW
zNBy798pJBJmdvtxwOGXG(dx^`7uO}D+_THtWSF%yorSK3Yn0dusuH@TBTW*+so3m3
z)U{5qX(o0jAblDqD;=W*1~9dt>SE(0>)OO0lt;_PLQVV2h&Y1(x;SEkhc?`Q8_NC}
zO6Zf+vc!rs{lqEEm&lOD^ZVWFWdym`{idT<nb}`+EjU)FSO12My4#|O&8ySfYwtlB
zv5VoE+!^jckyg#73yuG!N^LgQ{?-nFu5s4E-&itr=ic*X{>FVQuLTmX1*L0DSSD+`
zfJA+J);Txg&@^f9kycVJ=e}O@i!>t8f_37eZ)M5j>NlyKM7puN$cM79Jo#UB#A@fH
zINRO(0Q-79_UQH3yWBOTyZTJ?7mprS+*O+t*qus~iKBDWSj~zR;&;06IFb#72XH)D
z;+fI9TSyK<_qRpaOJ!Trvd`3CmKX1N;qHazYlOYXb}@Jn=CHY~^1Y(d&6uX8Qo+va
zww)I{JXp4jB%T7ruxnoqPvLas!RxL5&gyFB@hR>}CY<eld_*J<Z^!6rK4qC9)6Ebp
z&8r_G{S<%`6K!HlUXj}S>UF)B8p+2+Bk?=wRa3;l6%X+jjr2Bt3wV7@@9N@0X_P&d
zC1pWdjZ!p25pUT?di58%XxEXf_9+}B4`-8`@^hD*mZ~O8ndq^?!gAJsKdeYq%9@%S
zOMnYzvC&2$-@O993!Y21S{4Tbo(lJdxB9qWe!20%ouwf%H%kmDZ?=*9xFzCSI!CH9
zev7qLDm4c7<ZR1VWy9Y6NedjU!UvveLK&I)2VI=ZT;G(Go~aO#T|K)W60V&6LpKti
z`$+tR7+-hh?|*stlL>7250KI|Ly}n>G8KE^W+fMQzcnJEC^GP!$E8l}z|hZA3Bu;O
z-Z{k)^=?Mmx}tsO&6$hfS3-FXgNkwPW?!6~3YmYf&()~V8)EeJsq@u7DCSr9ZSXzY
zS-$SMZ+ja|J>6;FFvu^>ag=Bk;+{PUN5%2ihjAqKR$LNucGcQsLnV|&*3O*jzMkzU
zr6{4FdX*<uDKsCOWC%SlmYc1LSnHAYx0(BX&~_}8&0=SZ2;2>1zE*hr#8Js)@~KeT
zT2=)EJ1f6jBK+~i594mw<`Ywh8YY5A8IRj|P?a=(>&w$9;KznZnXeDW(h!U#Dkkbq
z?Y!J}m;LhVlCSCT3IjG9%+|4Y9?RosL<)$+eM?DNzWRSWeRWinUDU4%($dl)HH36G
zN_Pkd(lB&)hteTAbV`?ifV6aXcQ;5k2rt~j{qA?y;xAoGnP;A}&;Hfkk`WOHHY6fI
z)`U1>wRl@ZsEO~Mb(;K+?b@L|PTN@coMe{t_l3?1rfJirs)eFL;k-2lJugByyw#O(
z>&>6;N5=>Za>h@JQeWt3Q~JAkn-;c;yNLqrf26PNiu{VM&|TTkX!V80mkRf{eDYUX
z6&Vh?!fU->Vvy=p75vT9+a1X*{p#b}apRGSntHu`M@xRVT($Vf@$1dlX1>N+sT#A%
zLq2e0J<L?giMyo0<>LlJs#lzDwQm%12e2FJ+ePWy>ul9-J5t^d>4sFQ1y(x8-+8Bt
zG}H@iJapIp9yc-fvUXuYk%6gJe<53f?G_Vr8516744jL2-(k0q$3$7Arw*KM4Ku_s
zvSnh2R@QER3O$(eM2Fqcg;Q3}5Gc#;EzRw7#NItFzw=Z2+$M4j73qNjw`d8%(X8*$
z?CrnV))D-c)1bP`3tuorB&+qp?Ge58@t}3&znlNjW=>N^y@P>@`-vgpc0cRU0s?{D
zl7lHDSeP=@7{a&jE-o&9Ycc&_XNmVZs{Y#L{v-*%^IrMr&F7{)5x%;1JCQp7Blo^=
ze}?uBP5%HWy=QJ%Oo1$?=Z4}{mveA58N<cFsuuV%iA7HNgcH_5?}ujv$2glC&+2Zk
z6a|2PQiOv;?r=~fUZE)mz?_8>ShkBs69x@4Qi(L^(X<T>lhERo-48P2Crd@n^bj;*
zNB8f;iJ|&`wa8}Ox@O6f2hTwH0DFmg@ldRzHb&4RY38l9ry99^dbgLf(o{JJtJo+M
znF@zy(`~<$k9s<_IVzZsJiIetd~2da5x$Kubsr@i6+=_S2Ceh3enZ06@VCkR<GZ5!
zOcsNmA43^r#L<F^3XlQ^%=76ZFa5z%qJ*d*YB>@*S^~d-(8k59fgToO&{HB$25BA<
zRixNiHwep3gWt{bH&cj_U`N<KZhqcf<D~H5RfGCbSdD<u_R66E5TE4r%v0D^NL6w6
zNkR2#-hC(YsvE}I6~1##J$e>cUB6HRve}NJ)bZ4mL|6*#%G=AE82&Mqqoz$ZktDGc
zanzN{il0vA)AejYv;2^?Dl*#o-w0(b{&_o`#G{CQ!vIf(x`(+^lobB?Lk6N5n@f=2
zeTKGk2w5Y9EjH?$>PG(DYPEg@e1q~e2b3Z9JPW(AXYfxOs#3gMu;xLkR=+F7)YN8J
z)HaLON6ZNV1;jc@NR8vf)~Wbkb-&rG%&+PPB}2~K18vv0{}OD<t#71>7(h$eGYOfk
z^ZLp8?|VMlE0HbBd?~LzOoSgLD2TESXbRaYPJ4@FGOBw{RE?tiCF3OZf^O9VWs>B*
zs3{{TMfQuX{o;YfJ<7P(Z#S)LLj+rwK4=&NRBqHbF!rYChLKtZVliNx11Vt6?HA?7
zA(T<$l#onpJL~RxS};xO8T$wm|Cv!-oSr=Js{+?7*F=Sc?S{c7VmI;cPaecBv><0S
z+8la-U4|k#eXcU$9Q)iSw&t~6^t5a?5Ge{u>_@GG#h@~bv<401<ZMQyBpULRCn+PR
zrz?W{u_XB`)gFIMBTE-;<kh#w+`|KlPrIwrUJM!M{(1Tk+u%(if^<8Hs{NA_;bH8I
ztwUK9W<ms)mcQ`zW@NWsWj|XMdzX7*FZ{!cw>Su!gt67$wz<P82@o-j7fq!4fdnLJ
zve~jAoj-|-jnI1YfHJydx#3CHkCnt?Yd&?4mr0Hp%pP^(WK%lq=+vj#VQj}X(i<Ih
zQ8OrzN68S<gZ*%>QND&4!~Irqc!#%j{v$eEaQO?_`|Q5p<3%!XgF~xLs&x(~iTKe-
zvqCA^G9%B(9|sXvr@S6*3}4;-)o2}Hc{*ol{d*8C@_1`yq<#F^@4)Uqrjluf)=Rn8
zuIp=VTE+CW_xg7O<!r_-E^og)9xqhRTc10L+&Ed=*}?jad0qbEDvA7enP&O;NO`qU
zCA}^!>5PJ$-P!2I9dhux3oU!`wlj&FKq*C#J<E&%e^dWqJYR)Bs6(jQ13ezb$b&!^
z*EN2I%ck0W-qp36qV0k!^P&)(@kt_HYd1#p1lYs;G-*#9M2Z$f{r>4J?>qiVUz5z2
zLOOg8(WIlb|L?;GrV!?%Y$<vRdvn2VCpi~u5d+G%hD}W8-<=!+tOSI1&e=dG4vykQ
zdw|4*iq9<C>G^p;jOFak=Q62gD~?NV@T#q!QLE@?kGp-bL6ba_?&wa7+;{($eBG>R
z#qJEx-2KL1S3HtIE`DRDyu2`(mA+8^GIr1<U--vF&99{wU@$*_N<lp5F@uD=$)?+n
zcmo&l^sEs{eGM4oH!CT#jVr<2OH?Sbu=_9Rhc}oI?J^2{?uA2VMx)?Ry1$EL1qG+y
z;Ak8jz<@B%PE=wh2HlW0jIb1j;?Q79Ikb!yn1@|?#guXeP5zdx%8D^jE!-&CLb7Ir
z{KBb}>xeJ0bPd`6(}j;7yS}{bYsapo{kllbua?7SU;m5Z*6%)o4ZA?HOS#yUsoj8@
z8p)6yr~T~JnJ-TBw(jQYiycQc=j}+1^OKpTt0a1lt>o4Ww&iNDW84QwR%wA0F&cFo
z=HNl<EMb1w`S*ea>2@#H!S}X)2*y(i&gxpw_SZYnbaFnn5hI(nhATWT@b58=Tphj4
z$nM-8X3mC3&F59!hZoR|I1nJEZ0u;Sq{dR%S@(H;eC4FGdhF;^*sJ9v&5)BVgKhg_
zlRicJN;r!6G_%=byzR&t`CDFj3_BN^0t6|0p%Mw}Zml>}vKp-1cxj6{ao8kZpx|hc
z5no(AzUS0DkvrqUsvcc!#l&pq;y$`Ul1XmwTGlVmJwO{B?c`H9=u}4pKom6<(M30b
z&-*p71i54Uj2f}$lg`4SqY;1S`WY?@BUk<J<(*L$jrzvvZP0r}Ddf!`qs?1>v91mN
zjmSL8f?T`OUWp5MpT6!lzI$`LAw8;s)tZ|RS~%(ncd@Wd@NZR(lbj~VY0H}$4dHM6
zvCHmjI|D6&!Z{o17ysIAwuG?LLi{><$Qv|?oyi*|*mrYyDz~$`3ibJYP4LE>Ua$!}
z)_2M35?n>eQ^&*-jEcKk)U@NwGFaC##x;s|OeM=l5HwGFM$e4Brj9G<iX0|m8r<Ac
zLs<-{{N8@@=U0bj$A?6*$HEFdF;U9$L1S#JOuZ7&pmZlI@|9JO|4wF+D0bD|se<gz
zHc|c2j{6#OJZ9EBZMN2u^8W8lK?^NTTS^L{?-{;dtlwqy<2b4QlLk;q&Psu}`t#?{
z2a~Lgir*`i`2V%~{KtUL-3W>p)JvFLlg`nix^wJG1vhL`PJv8B%QT;xz~Q(nLxtYr
z{B(unBi(PcjBLQ`b6Q1GU<_@~9V1Ci$A(cSaSM2#e!!B5ZD^U?o)Zr*#6K_(?+Wm8
z1P`Odr@2R-xx;laCq0}sGklDQpCnD*2_UhnBW7gP&KtOMWVl6C&62SCOdOa6y94Pg
zPI7X;bD#)+Yd_$IL_971<BK`cqL&?(j=l|omRQm%bo(Dci9B-wYHrxUk>wnk`cIw#
zdK`sVi))+mqsuu9?c2bjwzjt~cMBjKu*ldjiyQ|P6F^>o!jTkKH%-PxVHir;uEs4%
z;ot`CeNWsPm|ex)F*3B3Q;}vT=q!mpIE)UmIYyA-^~$Y-#_E5yz559W;6qAFe>X$>
zRG{d!X1IqPRNs*Hap_Qiw`I&<F)wJ6$C57B{_cdks>4NSX&QT3IH`hh@GWCcPg}d%
zTWWgaHTW$_+ry-$Rlvq}N8;)8Hua1Acl!}g)V{lSZy%4ToIqD_zUqDa3S`VNg#Se6
z;#jQIwYEhJz5#F<E0`jc@}hqe2gEH$pGG1Wk1x%?Jf_LMxa*BR+1-?)O&*arIo~_c
zZ-y0^p5#IA^fWL7!CjoLZwGA<mg<2WGC}kbQk}GEH5F6ysE9!E0x)vm^P(vO@%pCJ
zfLbHB((2FLsIQX|34rzLKJ$KHHTI)uH|CbjrR638yI33{!{Y26lG^4jJW>_})0v=_
zb~(S6P)Y8W3S|HyfK-B^IVx2I<Hrxu1mU-RXTQUU;iveyRmAc!X9@pNK#>D_K2R6O
z;~<<BLE09lg-at}M$xiQIaMN6h7RQ3J6>9UEKVMh1hZ3j?WV!V9LOaF`7B*Nm6`yz
zFM;EzGm%&<VdKmD0SE3hyh05k#?SZg6eM&sw`n(u6IFtMR3P$6OA`=!z`Eze$5-vW
zHv^iy|IuB#o`s`FHDE_$!;zV`<|FY6C}H8@xRf@RCpDf~k%`^T%(A&wRPT)>G9nJB
zI5LJmDU7jh1dEsLI*iJr(8p`qqQUo$O_^ip<^+eIAnME?8h8k&6oqe!>1crFc;~Nc
z(f|^ga3cHd%oRf-zP&kPf7@9<En4QEFSI5l^7&xrg)`vkF~i)&p$2?;x%_v5G7LGe
z$o($BYubg?PhTZ=8}oN)N{Wvw6i_-dM$3+F+=_=QWx8`UZ`J&@X@B`=c>NwIqCIdJ
zNMF~TMumDcTF&44Zu0mBWZk7?-3<zN>svHW!`y9aB%lUo8vsF((47xk3Cv|GiK|Mr
zDMel!kZQq&ghHHL3VXdNdf>b+KxlUJOTch&w<zDo`~&+F>tF!vzji?pQFK=RViPZa
z)LS^T|3R|5;2N#9y3>g$_gm8V##vlywq=o(3lSag42cI9+ZWo0^j%<wtK=$#oac+(
z8&Gg}x{QP6m)ev~04L=Bu`OxD_PLbKeFF(uBpBh@M;ot4+pgud>v7zaJ?YLudiKa~
z0ow?4;o`o(Hx!bIqIR92g}srM(b-2h<J$rRs$QtbJsdlIgxf0e1o3e>{gpadWwjEy
zCzu?#f+AfQq6Jmsv#-Ck{OM4pwy~~}*u<ZFjijz>Z0a~ZtMFisL7Le_!)$(6Khj)H
zXj<haIenK~W9?+%WLncb4;JJ0LK$#Ovg1F@n()I#7i9y%0<7WW7sFq~C+ZRHc=y($
zMw{qZXW26Rk5!aMZZCb6_>l7cBD3#{&HdTl5x~mmESNi$&g`~6t_l0zr<jxor74jR
z?>@Po<F(#&@Plhp#P@_cnw$*EK)hYeQeHVvM)}e~Z0{Sh-S4}?ERoUAiye;sO@E(q
zQoEi`o}4jg&8jK{njbz+2^3p+(vlA9o45os@6IlQHh5?o3JNEd$NbgWHu`_^YK53Q
znpT${=cbWX0=iXf8RWwJ=-7a39~gT}0E^s-`L@a7>}|bDes+)j+a_sx;cNDRlM(xH
zvlAh=7B|m((poLb#;O&kQz?h=g7PX_{QQ{kF~UYPn{O9!CdjHP_WobXkJb@i;lrYd
zmVVwi2xTRg$9(@gv~BdWjtT9HDZAA~rQk?KU46@P?s(l8ZqVH)G@8IO`R}kvLM^e7
zGqremdHk`%1NB5JIA?VtU5>XzLQ?YFy)frsvdw*@V20{(lJgG7jF~ZT1pvkHnS9i!
zYXTp1su8dBiXxT4!I?6WJP^EL4R+iS1p`xF+^a2m`QuLJ=+=-bmCAuyA24kV6fYnK
z8Hk%{mcI$zLqXl{vK1_r{@~c(?GlHTux$ypZ1Cz1<}TDABUdhc9WDsk_EK9A`%Dan
zB(_lyOpYSFDn79G@HqtJ_NLxKIKX|nKv2^4=;&`bYsieh{verT^r`nk)b_JieTx#N
zsE(?5j>6D8L-X(XL42oE&Z(mr^LZ=P$$dSy{fMUZRrC2~P2!v>_=#y?9E&Jld!}f5
zFZv(Xjrr1*QM(WXO!3Z}=POvJ!SVIi07Q)Qh}PSPrZ?-04J&_TmTWO@Z<NaH9r?R8
zn2#}eZ~+CKJRo^D1E|sW25HD*+n)KoEuYe%Fv7TkuST^YWI^>&+4SPhDNa_TJ*p6L
zA)Sr$nZ7Zg)|%#z>Ie@w`g{e??|sj>aSG5zWxJG9Mtfy?ot-6qh=K=Dz>X#CX9_E}
z4iwvzX>lnVaOMu*MFfoyEin(TvSL*$322hP6kpx#Tru>AqEHA5W);SBg7e2ws=%O2
zteX(FkD{V$K1niK$mfhCCM_ZHNkZ86oy@NV0(*;U0F4bS*=r9b$+>(`x8X0kc@v3$
zO;(2PQJOOHg<E;?^-v2rzDA^w-_5aO5I#(Z%*r0H{>_-v*x`JA6-jh3{1WtWMXojj
z7O+mJ40b_Vxn_CvcDTBW+aAh_zE9}p>qB};4`bm`XYL;Rc>SLY*z~HoP!BZSENAiD
z7*;*E1tg;dH{uunNlvybzSkSP>hDH-mp6ldQXMw5oO=k6<9XLKDcOA5?v{M8-`#X+
zdisE}Y|2BcuwQKHqOP!uQTiqhz3)ye-;2j(Se^NsC9JrPwO&PqbMh>`omC&RSlQX0
zn-kj?gmRV-6;Ivtio>R>j(-RAO`2h;1#T0S-Xa4RaDbny|3hPhr=(!k*J@OH>X@;n
zIgq#F<Zf6zG9Ss^$at~dDbb^f1s;EotC_UChB-P(`z`Up<<ZxD>v$@dOf2^&tewic
zPwBnU@eU7P@e_#w2RCsj;LMl}hbpQ}4W>E2-9RJF#<Im&UM6l~MJvA1`4=^~;++n}
zgR~mqqiiwm9JY)|Lm~#!!TcCjcXy@CZp-Kjow==`1c>e;EvZThOV@`uUUuf0%_>H=
zST;7?psSu3JKYN9lL&kNVr%$S3<A%W^P!4h0dJoXr3W!-bv+ftzUq~Tqmh5WCOvqP
zRm-)Y=Tf`phRVPKpfX^5TaY8ze`5iR$_uVZK(Cm~936WTx*skLhYWUk8Y!`T^jiBQ
zOC&C7h=Tn~?F481ZB!IZ@ihteqF7w8ES-$b^05J*@Kp=oIs3)mfYHZ%_Qfs|&Sv~M
zYgDn^>gCohdkbW+c>*?tbs+B*l(Qy=kJXXj4LIoO?8ihLP@|=Q8_%;gwBx%)OeArh
z5x<2Pv;ea(p0$t0IkR4CdrdR$-i#fzKZ?v06X|+7CDoHPDJ<ReJqmaO$oyi_S@m{*
z6`G@0x!<hm?|PQfo|WR)@zw$9m3!@0El$W|ed1wZ+Xf)AYBT8F%jIvq;@ksa`<yo;
z#&_3|C=Es51U5dYH{sP;`rygR2s;;bU1cz1gRL{S6b<!?`B09G$X}CKeBYlQi)q4T
zV1T*U$PIS1qZqlnN;KznYjpqEcaI<b&>y6blZx5xyJ`7a!atgX-oaQ<A+(RUrE=0l
z&}M?9K0SGaH7%lYoTy6eV7yweip@|BEVsR;yY5kbz2?qnVpPOllQ|6VTWRFu-c<-V
z8b#Um?t=Z_KE)~fM;4gt$;vq4J9-mCC&(JNjQEkoOq{6KGKYKhCgaZrJUuow<O$46
zx=aTt-;O5}Ibi7@cmd(e{aiAkd|O2pMa7QU6m#69Jq<++w8y4i(lt;Hdmna{NUs8R
z`E4V*Dqh+;!kA!&S3T8i8AhT=?fMns*OkMry`mb%5cb$vYe)T-4TV&q;!s(uwTT*`
z0TzL1x6*H$4mPEOW=a)jPq()pxwXZ7FH)X`R<y+<3cr&&#m$gY3=zRmWB#7#%v9qh
zo3n;Vt5+Y&-Y;>$*4u^n9`BNvPZLnJIsaw$5w#3Dt@6M&p6<dr!H$HbD;HmpgN!N8
zXKw6XH?8a?OWE&VLRq;miyhQT0nQzvMW{<24vCxu0WTR#87E)e47E5=hR{&AdCfjA
z0@3>=#YX~6N|ghC=U`^21q{&FoDxQr3uc8-71?C(jXho*LKsIW6X^R)nwaK()J5vY
zQGaK=k3beBtkuRN9h68ip$w)pW`_%rTm>L0uHQPox|GS9=SMPoO27CoIKz%bp43L&
zmA|LpLkT*SMjZvktBR84mZD%B4`cD*%J$RE*F)MS512BWr`xXJ3aYVC!jKXVsAM`~
z%lhJ^4?h^K)r0nQjF?MTb)Jq~Knr461iD`p3yl3aW-(d=W^yRwTOc2iFh+Rmi6A|b
zVhR6VEM7ST#1t{(jN?9F0v2%F7#<l*eC^*$#9mn7OUBP$zk|`EXnz?3@Sma+c=azv
zIZl(Vmm=`JHt7tJ7#I|q%nI5*8hjk0hi{vJo$(m?`P<#KScc8;m$Kp*0dJmj<VBiI
z9FS?nw<kX1FJ9RDgc73MB`flq#&P0OHij1rbUCMho^DB*k2Mj3Gp%WN?cvbykpe4!
zrLIrHMRCNGk27)Y!i;-PKm5XNwfaV(*=-dM67#e)fGXx0H3|$H4fW={j}Hsz7OGp#
zwr-^0;&)uwf<ufDtfK9;6K)#N5tj7YV^52&nyjp^oZMW~^v57&sGKLFlsjw;wToVg
zRw$_dGf$Dlddsug0$O6W0SC4HMPYDnU1bgf;aZ+~VC?FdmF2KfAQB2G?nlyZ{-dy@
ze7n{TG8L16R4|)Kb!JLx@cyps@2yji{`~e(7QDCS4jI}qqKgx6HlvaOc(HP^2<X4~
zJPS)DR|IpyQ3uew4tOVL%I<HDaiPo`i?(WBT~gw=$=N8Jva6=0G@p}k_L1(ADb69J
zvBBq(!;?pixn~i?@u~=mHfi2EyK{ik^%m_#oil&qZd=^m{pOoyy&ZRFWii@aePtS#
z=lzJEf!C|pAK@+!IutF7dDD&DyEEEoBw3CV#{BOMN`wFPRNQl;o?b89_qm^|S6n2#
zUvZpxS0V)@PEfOrAp<QuY*?hcuMCz+c;P922XoKYe_Wd4ZEaDo42*85pYO&15G`zf
zq5RsZVWk&UlvADPzswwGo9L?^Oi`rHrrjt1(DBAa=ksOHI$ve3%qS3PoT$-xt|JZ&
z{VV=2c%+<z>a!iq1wBF(Pk{`3Gh_@3?bd{8h@3WQ0f&mP!|$QbUVSMwcp$o|Xen<(
z1t6{W-2)^ju)V(+4YeEIYWO9nquSJz4CMTpJLG;_FG`1n^&9Gkr;m=!@7JoAE2ZwX
zH|(m@x#z64DDf1MKqFw3+5x1rB<`j^W`1M}alCaaIWu*HO38Up{)hy!!7JgDWsSAg
z|A?hL&OQgMdb@^!jWYa+gF4&;b>&9}MGW9*bU7#BOj;<)o6Tz|o)1Ox<8y1PEpV$g
zAgPEIT6q+)GtQIU<;lpaY7)pz5X^01^DmZBJgxVekj*D;e&FlZ5$=&IRRDlhb~eYx
z0~d!st&9SQ4<tlNr@-QeenZmY(M&n>7x(~P8yoT0ln-CXy`h%bxHV2S5;ZHesh%eL
zr9+N6sUD6DI~u`MK)E}=vjKd2K{Xq2>Dj7z{tkK7Nh$Mdf*_6)xIgiH7R{6>ddM?Z
z@@AZP*%P<6tb?cjy%39SsGkmEZM*Q+)7;-c{4QK-f*maHwKw7P%26e<(mEcln>w;Z
z0b+W38_NbfBF$U`S7@L~7kJo=KJGmeYw>jnp|w+^r+LGmyL*4V;^+3NPx~aN)#Le(
zXBzC7<uC)&b?vZc)oKOs!1t6=q&P8kJJs)?sAzH<FnQ4R=Cy8uTk$Sv_7G(Ov04N+
zf!7MfM%kOzNU-O`v2p>kiKS2)Fi0hq_!Lw@RObmm-6%Ujx^E^v?RFymW2gTOD6gtG
zJ>OS7ToYg%^uT2%lgdzN>O_(yUpYfkrMaKYfH@8JBX3RrEg4&RyMeI!M~?<)8dM-Q
zvE;WX4jC<*)#JFYb~4JPsT}@`5f|T{a}TG&q2DN|jc`AEh;baq!{muE3CpG`@F5+f
zA?E-FNl^a5J`3R#&j7BY!GrBceF|Gl%9-B$JO@N{2%6Y}Sd#4Q&W>VIzKY;Nrq1^p
zCR_p~34$m{jF8;e<?@Fm7aV^^8F|y3@A<8z3OYY8L5Ha03k*xZt5(ra4l?xK3uZ+|
z3rXIQ38#Y^e}V!(LP}JR7N+h;n78g~3R3C@F*;P>4#wOBa~efm;Um=Frs&^a8~|{5
zv6d<LbTC`N#xX(vAB&%B2595XSA6)$mi5<qk5eddChz{k|2%K|r3i>1#Sgd5Fg1R$
z$Museld|zbtN(GfwcZkk+=05bJ+%n;!l!9PMztDrvkyN8Z{W(U-5Nlde5>AI#_tuf
z&xj;iE6Lv;Uz}Zx&BDy{rThjU0xSi<P17KWDOPY6U%}f@7&83iMo{BNFD~k-?8k)`
zBQ25w5?RPuVk5iRgDi8D3QoW2&+sJA0Ve`?IqfZ&jvE`Xj~j=8tZLG9w1y~LP=JZB
z_D}fWhyq!4GrNH?A!^ulRx<ImfiON-C=tVa@(3b&BA9v)oN*gGS$c*|HnB=X-wV-n
zl&pUM|Euwroe%U;)N;I;qp;ahfLtM;Aa~+{p$rk;Mz6pt&5n~VbV*QeB=#((+gt&-
zb=#NC9tW>&6H#5N@ai4@WSBYS42xmWoNwE=Se-qvm2dL>n}>I84q|8uJ7ezH5{X~z
z(CQl}yVXnp#iBY%h^AoJu#kJ;uI7cF#V-u*@F_1mC8inBRLH0n3SCMQY?z5VC&=qS
z`Ej=UyNM!O2BG&6;M|{Y$T|h`IuXn<gGQ`$^T&H^Tg<p9%fn%*hA*V=UKSK6E2caw
zdEEdy#W8ng6Q68RZuT6|%{udz7eht?H!NG@Z&QF$((?(j+Nc6G_wfGP#Tynn&>OAr
zr}*Y?)BClPkIy8LXa2WM1aSZC%?lTxj52YZi7A_akL~f$N6<?<>AgL{T@y82xsMCf
zz`62hMmghLevIK=HnkMskd0C(xc%F0gDbIdsu{ko6Mx;dK8RlzW{y%q$(9zOqDW+2
z%M|s!&>W3qq^!;rI+v;Y^I~j;<Y{NIjxH+MRcv+_4lx`Uw{myKYIJiNWie5>m*lxS
z>dI&9-MGc7ckNA@<RB$ZmfT`yd&<(ufRY3%OWF~|B2z^6E-Ydnc!j`pG%j`@c<Y)F
zps5#!niEkqo~QN%8;j@jr&cwOr(P#qo%gViPo7DgTC5jvxjNZf38k5?=Fe0EFxi+G
zK`nFG-`(KGEVeY|iiMSU0Ag{J?f8I7n!@dwe5K1_@6Di_cGdG$Yt)pHbbU`ed@fB>
z;l_G>>uBDn`$4OizE0!MRST~jlpbdk<BJtKQb)|0^H%vjWnH27yRW|*>K{vH&6<Rt
zHlo4Ic0u2-|Jv5UYq?{Wp4@i!>T$&`qkzl#u7WJZ5JJ9NNR~>2B>EU#Xghu=ke9=s
zQZxcAs#|hqJh#&o`esc~!-jQT;<kF`vjY(q5n8heyC(Xz|3?fh7Q4=N`nHXS;d-H-
z2~gwL)xA!o#2Po&)3T_Xuca$={veq^1Ed;2G#?C8+y#DQE~uU$_%%(24Q7}f;%;yT
zGi%iH!@mT?|5=}Rrs&YxkYz-L{=8uWv%z^S{^8%22pJrO#NwL_m5KOp#t4N8$wgBL
zvB{|Uh1!9DG(17-fHik`4joB`yn`h~%x&+SJX3Ecdemku_+__dpqkoiYM=$<!VprD
z39P|u!Ei*eJG1zq&stb=VoGw&gRmDPxnF`Je$E;l>05($pqVY#-0_^CZ~f3&G1?-P
z*|lY^5_?W%D`xMtn6rAvm@&2Zp)Q-@>E&>vd3`QXDkbMW1!d*-#^(^J-3y%sY*u>-
z1ksfLw&AgVjT8Nv!Gz$7B}mFJlWA}_F@C2>9_T4@VFLBb979l1zVHiOPlMKe)hFT~
zvcFdT;!JZQluHTQemc-HHbQ{S)73Euv9m>nS%0?{&rQevBR+`g@nDb4f|)vbaR3Hr
ztkE847%5SF+q`?0ZYgB|BKgx6RD;)gb%+9*?j3yLUr!3nBC9qpS`n{4D{b7ld{mB3
zLz#{Zpch7guoTL#)K&Bqr=S7v&$Its3M1imqu~v9L61BCz}OiGR)gofedrkx^vNH{
zZF)KuXT><1xEsOJ^+nE=QVx2cErq4fy^nk;)%#j(%+&`WYP2#8H%p|c9Y|sN{B?Ym
z|6~D;_Qx-~FH$x-v;Z?OCaMI?!3Cu?`T)?o%CZEE)8YufO6ZmZU^{>YFhm+|>AcE7
z&&ZZyHUu{C?yzFC{!mxM0vduZ|JeyB-{<b^ZucOlC%RTL0xDIU|GXr4ujVhf`M66q
z5nck|#EwHjJZ<r2dAS%rcUI?UF}XxuT!RiG4@{|DXaFx(0Rg*42v>%(q?8T?E#s@R
z8fSSObb&uH;v_=g1Q<uS$(s(sXE8(>bg@`&9ASg+$-#|x=-d_V#%K~3ZhrW^C6QAd
zGBG}@6X50h0m`{e-00t$91($VbY=9bL>DXC*|w}(#P{KapR;Sgpti2>;S@Yrnf0(a
znkl79{g~E^!I}|y8ruoA=0tor<HHh54mT$uQ#$W!dG&OmQE@t-G*f&<lK0}UNk^z)
z65;8uUB*^R!;-Gw=<10lEtdiiR6GPe&pProEf`e`fiS>kpL<;L*>dYa(gE0^vI9IT
zva?2heCXNY%`BmqbtOKZ{|d!=WpZlTeE}EXUii?8s6X5`uO6nOVRPis-B1$tl1LJi
z8{Z_CYb&huEwgQHaV-;q?biZn<6MS@`mk(SPRX7IV{s`8gHkAmGaR_LHdBxialkzV
z2A?#l;2Q@a-s~g~cWahh-(|J9CICE!CPDd@W6XpQxUu+Sb7IE8I>PQT+>x>x-sVrc
z6Otin@@5P+F;RrZlnA<mG{loB1Cp90=yS(fpsrF<oE74~PC7Fq%cLRZakt2C=cxP3
z>)Yo|F(NN4+IvnR&N<U)uK7lx-PH$>q1JUlcvXD%^BSQJzRkmJR~`mRiP@WM`qE$F
zM;T`b8!7FU9X`UcXR2A9BN1;*Q{*K)3>!vdo1>3H9*z*r`ouW_t0o7)P%2~P3K1p`
zr%|x%5@wI6ZY+Jif~yu@fAWZr>KAO4$a$Ub!2CUD`KG}RA~rOo`SZG=cG-{;XB$XX
zjPW?uHB2qu4r!wDyOK3h;;?>k7jSYj$~#&98(ZhOYVodn`9L<MEffZN`m3m64Q3n1
z<_Ritj%1duA`D$&_a9wwPJEfo>Si_byDKSZAY;;REcrv-TZB62J@|gM2R<i(m($$q
zVI{)qLE|K)b4=XVp1Z)wGru34RKZwNW_*)M2(}35xSOJCSkb<eU(H*;2_hRkyIY5D
zqUmH6l`9|db-BC-w#UCmi3)T<1SO*4FpqP|Nbiod${<S6XPo_(X+Ama_!Qt|rNHhG
zaBn&RY%nolVY}XZ1>j2}luJoYJugw;b=TT^ZRb?)%3u5AcOjMCK>aDW=t!sS|9i3c
zRx4(4+zFe~^rw*X7c$+iW>TfbBDj)qYT~)jhNY5Ab95avbv%#XhSFfj6nTTKRW5|x
zqGT_;zBG!MVX|RQm$((2(fKOSq;%)W+;I&NR_L51#zD%-f31*A)CmpH!#;I*tLMnf
zV!t`IfoMWXV9~@CrLf;RScSAkcfn}7<RV|ipvZs&y<2VdfT@XZf{Es<9=1WNUWSq2
zFCKTARRMo&=11SI%47NG_r$vwpnb{>=_Qa$d4P%V_FfEfmFVKr|7n7|t^Cj9(T-dX
zFm3;sqaPk;|M$XcAW^Chu{*q!PGyjSsCK+yOC4*~p)Hcd#^+i5jlfJ>_-QY-!ze%G
zAjG-`;b&d2F%^G){Kqo0Y8~QtgQAp<?u_5c`*jxfCWsn`c5<GrMs);cd+vg*I=hFK
zY5A6J@bB^3Uv}q(71OC9qK8vshZ-Z#9Kac?L&zuR^<~G1>VHe|i~#etXvf+IzHXsv
z%K!FD@wx9JPnL2-L@UgzUfoj52js$JARKJRS%biYyE~iay{nY~+z@xkVI{I#&<!Fk
zXP@I$7<BLU&#$F-IsM+}bG&}OWgQWN3LukpQAeJ>C=r{nXdcH6b`0gw<x_IrquahC
z$U|1mA9s^ZHL89KFIsgmHO${oB)K&8#ReT%VF7_8WFJElmQhZ?OOV>3xCrj1_VqAo
z5OBcP(=w|z$$lQyZ<m@;-VNg^5oH12XpJB{n}T(KPS#Ev4Vr{ezpGgvxO+m-@z~T7
zZVm@8z{^-FKzf|jU#WTI;83kVVR*VOW}>KA*igB=&6A~TSi@$ZL?KCUJ5~!%d`?TA
zzO#}d7!H5NLzr?0XZ6DGQp)dALFC#g>wZapi@wF~YKFLt?77YwdQosR(0N6$G21x1
zJ7>2+QE4)@OC}Ems3lp)<BLDZO3rEE-Zhd7YUWH-B6nbZ31-P!-3UOHkQd8eaRaKv
z>5vgHa_JtY4Lbry=z?N&icWmOzpp;|yllGLhguP$IymoF^S(j)8NbucYlJ^{#*Py|
zIGj_4dg!{|Bw7j~l(VqYyiK_vX1c}?+N)itw!jF)xp87Cn<U{SA&@&4>+Ol9j)n8`
zN_a1a4RorMqP6)e>u+rP0&eB}LLvsaG&3DQ{Ij*yE7e=MrJ-UF8q5{44<p!cp_L>`
z1QMG^wOI2BD=e}~Ick#3v%?$~5gWD-3f*k204E}Iap-^=ceqe^Q3RnV5A^<<fiDdj
z35!7u+dEr2c{8Zn^wQDy-yc+TmrdTa{L$OrAeBqW?a@g#-~>>`Yd3!E3XaH!ZM#}9
zD=<4Qf;f+A;UKfM<w#|Jp@~hi>{|rez-{w??W`$bAQ$j3!|@&%++M1x-nBlhdw>%4
zu-ESZTKK!m|0cnUVy34Ft>$av#C4WfmpVx?SaW{}SQY7<P>lZ9Mitm3`r~QEYxnjN
zj$^3#6~m)zk=v)qOGW;~R`j><6bEH5leqbL0VZ;cn}2Y#!__axx%~6hYLOJd7L2dq
z?9OYi!J@J=bEeApZjOrsAzd^qW`7qi>c|$*xmi43P#~8NWy=h+FG@FPp`J+t`5%x)
zaqz6^Hnj?63%4RsO1t$15>Ft-yP$eKx*Wc5Ah2^+A4MJrfQvYSySN!Q{%+<x8sdPA
z_v)^`NM#=^f-5K0|L51;pxZ!PBM2xE!0HL2W}T*kAgI9gOqgEyoIAkLBFXHDP9}vB
z3f3yG4avev6UfMQ=vXr%&ICr8CJFyNM8a^q*9>BZXJSD4N|0SNVz8=GR*a1%77f_F
zg*TKYprqVoL<H$ozLfSiT`t$B!0ci-;~@E<HHgvQt@inEhq1>Gut64w!jjJWt^*J7
zCu$lV?sl1F*I(t*bSq8WeAM{hWhlxA!PJZ*X7FE<WJ-f$R$WK14TmbBqrb7`oC+8Z
zez6dre0Tt>!Mu68=sj%Ie0Y2wevoqmVveIVWw&J~=tH{5ZCn6ci6&tH-q6!WxO|=~
z;fHSxUsa2z^e?ra7G{%1ZL%5IZ%+EZAMY>%tDazEYai<1o5n;r%y9J|403st{1~IV
zxBs?LfOY{Ko`B*gc`w*Bw+0lF@&hOlLocl=<Mh2?v*xZCq8tg4oTEvT7=I3+PJs(O
zsG#rX)O}yz!(fFT@Gb>AF67O-c>sU#s;_Gh5H(74(}4JzAJ`9U!a&QRuw8pip|=JC
z2j_h&f>GqZW(e4Uk18Qz5g<ZtQ{WN67MnG53MA|6g%?xSFnRdEX;J6li?iQnd$IZN
zWnBrSq!~NudglJ6?&iW;Q{<|tm2{{kjCmF)nJU*wIL1)^+Fd8$N{)CBd|DY)I2-8R
z&3wiU!p8i>z%vXyJLeh1>Hwto%R6gqx?5BSu%YlN5%`arS#~5lRZ}hvg!IJO&E2{%
zB}FEN>b$2>0B7X{*av6A-5|7U1!+tU?JQky&?9NbaXn^#PV5Q%MnErNghd_ziF#!W
z7euLCwZxbse<29<Qkj$t?LY<VY2}2OjZpMgIQ5o~zXxt1B$Rg#TVi~@i?i1wPW9hK
zbY?<aya^j*Ko+C0u+-7C8z-vgr7}NSP8Ct`r`qi2>9+T-TPbT*q#4AXoF@Q3k(tB!
z!{{KX_B~$#I4q2fL^s%Dqm)nP2~rK<^;Z7c8w#!U8HeAS(tatq;;UXT&KF(-4#^JO
zC}{7+7Ynr8>njvM#el93iiir5r*}QmD8LEC7!S*RDW1ML`6W~PCvcEtbjxb|y3aw_
zFAC0)P6n68B|W6;>S@3g5v(_7OEL7L2mQ~NpVx%EHh87F$63P;#tg$E4|}i37a7ig
zb=;_whMUdz-FaB6r$t{v89K05-wc`FRjR6z5Gb7;SqbQ@9L9tZ&b34rmx?=wjW+)t
zSusivxi-EX*d~oD7*G-s9CR!j-5vRPF%`-1)D1WJAVMb6=4Lx32}w@<K_j0xr~Q4i
z@#QGXtV}Y)y)5BC7*d2fR-8czyWi~uLiIe7&thSG_}ZN?Pl-<t2PCBT6J6Dit5qA$
ztmW?zz#9xaiUtBPRgE975uj#j_5zt276kmGZ*l`_bNh`;t+bufWy}{ryyX1K9oKNz
zxLpRlH$EMhBR@GEe6qyy;jzS^O;87iW(7!r(~}Om1eidwoO^OBUqE%pLMtuB)#C0N
zkn|bE^p;IjR!cqIUJn6cRO7%0%d%OM%FuiY5xw>F(PN~%E|43=kh;Ep95_$!V!a>|
zOOe|EPg7mKE$J|-krFOovxg%~8O51?esKZIM=bb<wzu;h1tfQ<<YFQTK_?>}C+@X(
z+Phl|5`>0=IP00koJn;}52uw5!fvhsDb~0OiL9JK;3WP}1dD`J38bEoRcS0NoA!xz
zXSJJ5aaoW~|9mxK7p0(>5C-`jTmR1YX<125dyKPYWo$pd$1)MW98D_*{cXurL+xbu
z*(>W=TH;wgcIGNyJ2A-$4Nm}YvMZ^*&MC8@)l8SJb&>BPTRH*5?O8ke&C%aTboZZY
z^z_bG#{pw~Hbn}G&n@NThuoIPs!;y>f2O#j7_MBNY`R~Q^!w?&%idNLAcZV=I!R*&
z%jbdI7k@d+C9Ed1nfOgo^g)H?oSsJ}co@V9WWW~#ES=KWApQerh|1Pv_s(=9FoH}Z
zQ%jWSWouTw*1G<G^DeS70a<?5EMBM%WGO7abWSd6A{Lz`9VGANkb%9f&(7NlY@u)s
z)0*wMShM}*59k&<-+#WKg=?WN+r9b!45&*P5GR2-NelX6N*MqMLcf%eyIMv2X8N3Y
zFd*B24>Vo4;8O#hF@4$mWemHTvw~!zajn$dN+TtQc4&1gZtg@4=%A=@CV|prKuHRS
z=!Re8JK*|js9)(pHRQ)>9}C=PF=PN`xh4~iVy-1Q^{8JeWuT|O$JzGt>zTnTp(-|*
z)D1o_YbLc}k@xx#zmiU=QA112mb;NEo4q$E34)win=4-&Ht#YH-e%T+&L78rZ#szM
z9&fBhi?lXnojN(Fpjh!6NO&GkqTfz*DylSzf#YO2dX*zANBUuWo{5bHRn!@{^JyGZ
z38zDgP{98%Rt*T-95W9PM824?#2Gx>*CfmM*7>06FSUb8?F_IdKw*Z&gI9c<5MRaI
zlL@i~N?T4O3vJ<yY%xBI3Aa}$$Qgi9V>hW<sRFc*$HN}MM*?h)z?Y6|o;u5in9sp6
z(`u|Lhk%-%ud<`22B>*T!iGQkp949-lLns0fr5i~3EDHovRO>mdYkQI_gCnq9O2YU
z8Dwi+nUeIMb~bF2Z=*9w%&Mun{^SSB5Co@dTj6l1`JT5ETF>q(F<U?LC^~a{?_MU9
z*QZ&&(Ghmd`_4FsW*z->I)&X2$hTFFcAzu?jP5Qd!42uT0;uuXh2p~0SPN*HB#Sq?
zl&K%kmRuLmb+hKZ<3-#ZBJ+yR)&po##q5$aZYC^Yn=lKYb?~?~PI;jNIT&2zNb5k>
z3Zd=#J$?ZP{wkpy=!S$!tMd=Q&Yu7R6gXS=JUBPoh!l}6OdI}0Vn)HDY8oJNm^|G3
zJe0E09K~i9h&LnOKKb{B`s?qb+$`H%4k}7*eL~1y8=;Tu8n(2HUrjdfqYtB$<JJ36
zW2Snr&=)68VF;WM6%mg6n#{z*6t(!uvP8HzZkG3YR^7xLQ{#co()m`=j(!RxN~ahC
zV)zQN{XK!uo5;(cV{;lGi`?#n3^JiX8t|&JXKvl9rG8>}oOSYP0wWiEZyMenr;Gq&
z#RH3e(}<hxjQi8$E6}n7#Wh^sILG_mGpO3i<{Ou?iLiZ9mkx7}$o<Ca%=#*^nuB*v
zZg73xY#l65a{w$FuU=X`_|E*V)$PXdzpvoy*Pd>v-yalD?)SmB^b^p#7;sIB+));H
zOxz$zk*1P6?V3gbJ}??;YMk+yx5yIMJaKTnh@w3cD(KNPguG-)b&Vj`MYU;hoT9LK
zN3r_n^Ez=--IsFIk!5fCsFQ%r`@9??L!{;Gi0t==&el4!yKC-om1^+nv6??GRo)g|
zk)TKA*ZXk~W}o4LHdkM;dPLUe9i~CST4OoaJ=$^%6h{V3)o>WK#tVyK$y@}JF%&2p
z#XH6b>V>xh!-@yv<5ciuRRAjyGSdc<<6$Mz&pFxxY41~mlwOP^-q!J*OW6SHskRF-
z*G9an%+4~cXd_YD9Ax8pL_Cy-@jE6o^&AdoLf)5!J5|q-ThG;XR5Koroq|5Dtl{3!
zzYIZvBU;eV&HKP_UBGuifB<qNKD1l6;&A{O;YwIH#~8H4{D_}mQ%`UfFJyPb?0u6U
zb+dc+3Yi7TLCRBk@S?xhMP4HRp#G4BhEFilpp2~z^&{-H6%SH&JSXTk?}|fALB8b3
zX7Yvab#>M@N7fx$X=y3%NJjOyBt6f4uQOaflV@ggS5FH-2x>B{pTb`ffe~=KON~eb
zjjwtl2iowgRbX<dwE#X2gclE;Njz(TU=f<u5w9Yj3Lrrg{%H!nhQ+n+<DitOhi`ru
zbvxi<<P`vC^raXB7-l)%E2P3DVUREa;r?7|w{}kBd-Zf}@<?}|xLWzirMqE%e+W8m
zG7AK-8|dF@Sk;35(S0^&e7i2U%MJF=f$sTPSrX?T0FKvYCqAIxg4MvAi19E;cc4SB
zbBS5R*a>T3iq7Sm&5;kBBatm))hEga0!9<~ljCYB)@tU{e|<tqwRM4>XEd=krg<JE
zV+7cQ1%jA}{f@V~Cl?d4IA8m%;U(hs7KFnA7D1Nn?<>ctopDm$R++?~)OTaJ8VGiy
zByE5zJ=c1_Vtv?jyn6+FnN*+DaAk71oE`rAYAg=OC~3*5S>3S3Or_jid*CJg`#IzY
zWYyY%Cm0C%_Yn>`cCS8OXkB$BkY?&zo{IJM+6RD%X>OyFHG+?iUc1Rd5_gLjL+@m;
zpY|KnUHL%;`5^ny*H2F^`Yi3O|5Z$qUBv23ZxUPuCbHWgjKkfr+r?@MkN-z&7OG)}
znr?rt5!o2E0B(h*S(R)ml5@}l%Jx#2_6==q0AVqf6}QDHw`aEEkoA`jc0qW~fH^jr
zPAuQy-OI4h5g<ZDk=ujLTb!Yz7%R`ShHVYCaMr`LSh8%F6&9sDBH3NjhuaecKGxA|
za};YgYj!NQGZx4jOCiL?3jjOdz}ssa4&xvML6&Dxxm25!ffRM|jkJVIJnPnkE8-#s
zz!iznB+vUYIssd8BGlRgk^tfZKTbkr7LB<@AtfA%+=TqhJn!M;71ovc+G~!O3pKO>
z`3$_!_CMm+^0_J{V90=t9ZE8?=C-*`*q}yBSoN_5M<f^O6=>K7MGSvFp_A_wTAX}B
zoUE9F_4+`~4U-Qv>R5a5gnh#?FsGcHBUT_%m)M~SkrnhS{kV9ZEKSUbNS4LR9gGhs
zohi{o`1jYI`BL<TzX}O*g)RYOF-FKgXZu0894FA;usQ$@SPZJDaX4_J6sXVOka<p1
zMtbPJ|Gr>nP4so*xwea-j`_b^iLxFi{RT!KhDJ}jS$x5kiuo)P2ZvaEJ@;=}Pj^{Q
zdV<}HC5j3rz(nE{M6jGaC^@)M86Y9AzJ8F@-S=;IRYYk@84#s3C8dCLSR`HVOCJ&k
zjr$H53}n>NbphAnY#WfhdDag#Y4OOFWj3{6wRXM1&`zMimD%`QBL!WvU)4zjmznQ>
zixD971uf>LCZ6u=<rKcyAaYRMEF??sIfFkr*=8Q9*@=P6=Y2X$F#+bW6g2HzuD@Z$
zm&gCK)TvD`#x>3s!TAAR0|`9qj#=miU+&wF&2~_)mkzq^Z$wE%K&}``G*VQ@{4~+h
zsEm+D?K<gLC*uV`R&reJ5&VLQU_eJwg<fM&_P`-i7d?%FHA~wWT$ExLX#1zjhv#eX
zN*F2q09Up$=Cec=Zn1h4eA8w8FDm$E`~8a1#2>Lv4Itco+CxZglt*%AiuPUt7v))_
zgR@UBdBd6Y3=v$dmY9~<x?1L7xBHW5Y`^tZIrCOi!T}<P1A+)=6?LWNKH$sH)C0?U
z-x%&WfL$2}hD|T`i$1XCQ4%Hb0IL~uIc>b3+vd|t5{8JkgH2bq|Ltl5D`jFhE4O2H
z`P9A6inWs}jGmTC-jouws^G^O3`J?y+&ni;oA%5KlaGhOELN&3+UBPr;JPv?<q(IN
zuqkkG#fpDyQ{}M&FR`ksMIsU0<5mYk63?RdmUN2W?fC1=F9T<HlLzVgC%>jEd3h6m
zjH8AUQAj~Pp^+qY%VsSh6<<A+@pu33Z=U=ToaxUvxMjqS&#lEcjc~}+<8E#j6|hac
zT3;1#aQBlFv0krlyqL2Dess0u;;;;XTu^!IjLPJ3T<ZFw!!ZC18Hyqg2Bi~Z=^KBp
zAyPKlfT<QX4LTK<GJwjDxQBQc<`0AMPUY6F^=8a+tFFaYf~Hk$o}b%B1qC}V2H!jb
zqD`2L?Po;b_PRNKpH0sW!jcS0JX$()hd-oGflQ30Yhp*ylf+_;Gfq8$75RQqcj+Cq
zNl{=Mm6sEETxPjkAdk`mZc8dCAqd1&O*#O3Sj;-`I^SP!1F|oos4=e{N23%>iU@x-
z>neWUCk_#%V3rpGsXZd0B;b+<NQxto{e%Fi{&9I~dbR0aru+*G65ckEr`@M?CBJ`@
z)Y!-)83MhTLSDD|`u}K%A`zZPXkRzGmpxy2O?!iBq{e((i?gL&&XJ(7#6=!>Uk7mk
zJ>=og>l!Fb?7xLt>|J4<FyxU6mrzR-RI1_U`Br!J9~pAaXz^>!uY(Msg-{Ot?@q>o
z-C9l`oYvSg*@27<LIwJ*e>yuaJfai}?s^hPdti{rZxP!b-)R(p$sY9o#~Fhg?i9e^
zPI|C#-M0NtF{DC&g@SPMw*THNhWfU1l6Z*+w)5Gi4TR;q>CYG<wX08lEnvD?a5QKX
zy25{y?sWtTSL2M^Z2EkWB=p_yfIv)aY%Vn&9s}^gi5fx3O}O#B)T`%fmQ*^zM1{*Q
z9zF#`9FPJ7HsXyjJ*`3s6{aJbAtt=?$)b6<Jex&K1fkvBC7)iKVsFkuwaoXKMHV}A
zeY<W{BQ3Ym31|b`hXmsZwn*<Kg4aQ<o&&px6uN`U60Dl#O1US$CJ85;0tLJv<WN8Q
z%2M)&!I`nMpw6FErwG*z1UU0_&+kT-hXYvUPA}}UiW|}cBgq=DFJ0IitvCcsyfF@c
z^*jMNujU5FXvO)#YNJez8OLw9r6$1&NPFw!BCUsc%Wc7Vx*z8MrlB)*;Ci7oY(Exi
zUP`AB;iU8(BV<q1xc`-mw47aeJwcu&3uEOfw{zobTG{&Nn$&Ux{fk}l^8*3)tLE&%
z6(fInc^oS_nGKMR2bbuXf6Oy4tu$&^rFjC2Mt<Sfd(NjE1+uv+7K>_4cr`6d_J*P9
z%fJp|tM)IC^3{)3qWV|f=~3V{7koyobxeQWt^IRFl!+sdld{*1F2pAW8}9+cxa%5c
zJ4X&+NuS>;PE!we7^g1dW^{B&2-!sigDtAi3QDlCb6&vlteMt?WTU+JUC{SEpDI0F
z{_MVRcHUY`-c)|X?zeO@KW)J$BdZw~_v_D*(>pJ0wqBrr)6v;o^7PvB$NOcc9!E#*
zEVu>@83gt$CzI;VZ$|#53Fuy^GVCe|-*8UHyrVvk?ldGoPF6MAPz~6jpf53N+B|Lf
zkMFY;9PpDNtFDJcDX?bZWvK#+a?Tn#(U`)P)!tiC(M(!WEjUV(uoo@i&gem*Ico*)
z{hB%F09bdEoRX5Q$#`z&cWp)@;+sy)V|%(5fbpqf_NhwgOhfA##|1M#T?^FB!J^OB
zR!H*m`T+st<WfBtf{{?~rxeACcK0`*)c+k|v_^rz0e~Qo!V?diY?_`EiQxPU%R<}Q
z9sRfeef2V{#>GL1m=@l&z>$ap{z%Yl(UyPa<x7y9#Z~)aRLy0)7W%LtbjKUa0Ai=x
zQ}V&09mXw~tx*5zJpu2l4LpGTKxje4;e_)=mOyjxK?s<uu^r!EslmDkh|Q~=uYvTA
zO!~r+MEt<0F1-Rq7^#V~!r}^U$5t)1f&6~^4U@DqF7W8C-d9?fd2OH~!I{lo&{qAv
z-uE46!hsE-FmhoQWGgT+4U?HvvD(F%aw}O46JY4mk@pi@`o8_I3Ck*+S_O<Vb8tEB
zdB0+i*ZHSv+Hma_TqB45y?s5A8eG=3v;|&(wHm#Sp+T1M_gposrpFe=4j!<k=<)s-
z`s=C6G;GnMC$c4A^R>8hP<%1?6+k71^`hf~k?&9cL^9St*Nm?Iar1}bSoq#{!S{#u
z;Ap%DgQT#6F~{Em-<W|ueSho2<a>ZiK<Xo64DOz}g36}v;lyv(mn<1jB#5^dIV&;>
zlLGsik4|*6_~{KtBQdP-5_n$@>!7snbt86dr;A-}jFDxFkw8dZoGS<<>$PS#fZb%<
zlU(erHj+#t8?Hknz?nvcW~%#|$t4}VQIQTO@`*s$Vad?mJxF5U3_+Lb!M+3w6%sr~
zy)!ZAjx{kuIdi6W0Jd?r&Xyihvq0B{OhCZcn??rUWF9C+s&1ge=Uz0X<hP0xma7pi
zVW{RIH)(=C$K#5!Ps4IIz0YrXqjoPAN6UV#U=m(G8_Rluj)m)WfsA+H)AI?7GWB~w
zA4oMUG6BjQ*fK!cEK}e1T@WUZF&MGO83o2<OuBe;r@?`;QLj#n8En5QpX$xtAr9$)
z@mPYX1wd0>*xU4?oESbAy$0%Ey;K%hP2CyWk9!sx%se=>B|lQ}99DOCw@duQFPeJ*
zI>gR<d=wHwr89tykUL_cNeOq??*!i=5CTU5$*NSWv8d?B(mesex+#Kg8wN0OU_SZG
z2P`h6t=rjlOp(lO9~wfNoSYQAKdgQ}WBK_sD`8|XFbfg{=#NwSKglP1A9yBgYeffa
z${eTR_}>sG0A(s@eKP=2?p^z1+-*Md^uM9EM$dU@;7t+pas)!<R_q-JVplC=O<g85
zhK}!}3)L<F(jDusPg2M(%zIZdA#w&x)v2N-H}m5g!5^`<KlyIJyF4=-WZ0+JYg#Ro
zXXf#f3o$*t|8!0i>ub_yBH%(f?T6yH8`_8Hl}^-X+M)N0zqXYiMm8+f{XZ<7by$?!
z_x2SK>F$n^?rx;JL0Y=II}~Z@1_6Ns3J4O?0@4iv!q7d0G)N5j+kD^mzvFeCbLMj9
zdG=oGUibZ3dGcnj)97=hIXFJqH<5o{6v~>BFRX@vCznr7O5d8#)HaM+iP@oJ+vtzy
zi|x|chST@qe-)ej9!)ZVW5(UvHLD_ou2ou?d8spJms~IeaiC<X)&YycTd(YdjCCDq
ze2PU;<nvQ&FTunG7d*m;-OX<LtC#kerbdX9!47e%EXB0+u}G;m&j!#3<@tf$pED|Z
zyzp(M%6Z}oKhA>k$OX4<%A;sTEJ!W<6$ieLD6ZZ=gcy27;_l6Qhln+u*{@SxrPAw;
z_%3#PAf{?zqQ<I@n(tnNui!c~oH|@=SUI#LR}o$4da(f>1oVAsYlpZ|XNi7q0p(Js
z`gyXNWm2@^@FJd^gcFeWJ@~;6GB#7Y^TF^XhRFYh%5&K89n!o<U+Y7jQcO!#$aPzr
zfq>iU5m2eZQVce=051<m*!6~p#>&@I{4ZtKEr9Iq$S?xsbUPn&(gWbHlMjk1MqobX
z#xe3{#1bzSSor`+ArD#ss3}*YJD_o(C!bc#V<BfJmr0v*#s>aXRMesV7mYA@DKfP2
zwpTR+L}XHd<wPPVFacOEpKy2&4<+XyeQ?spORk6qKwq97B_k@8E}?e?9*u1rGp@f3
z7R<xAyPO5&g&oeil_L%l`~0T>d!gkmt|shu$CGX5hG3&x(W7Ll2YY=oQcQu72Nb2?
z)qBs?dx^l3PuqyEsYo{Wc2L%+IioFeU^80TWZ?*pTh~#%OZ~ur1324d`64^-*Z6=q
zUpqygzSbEy!DiN}y&hXB`qM2eMT3f*5qDBHG`J!9L$9>(yK=B4F^Vrx02o$=`uP&o
zOLrF~6c~HArm#E0#HX+Rdw%!prNpnSj(@Mz3&e&d9u6kXuphc4*Xj%J**H6{H;fVF
zAAQQ5Rb|~0zNn3V<8P^(q;>s!_j{5XqEx)laD8Z}Pokl4J-5zNRWty4<WvZFhVN*s
z%NI81jWI#=Y;!{%Ovp$wi9zmk;Ey#x9x(S}a=69_nk?c?7vI|Q2Hy{3?I%3vg_E>=
z9bb<{0{p-TL@OnbmOS*l5istKG*Tt$^~nH?owMHs6zDLB1<Duw+~Cb&EZ~K=6d?vP
zb6-YHP_s+ncn0%~{lfjcGph-41Cn}L&S2$_uqN_O;4t$Q5h$QeP55hV)h#ZXDPTqz
z?|XX-?l`Y~_6mCFx6PN?6{97$=s+6)5r;3XvvbbIVC(>3OP%P&q3wT-I5MiaeuA&w
zvfEsQyuD;S5EkcWn|?D%!LI}mIncA;da)oC)~}Od`rOZH4ODM*JRB&oq%3<e{iD0Z
zq4iUM9*bNDL0HmI`@;N%&d=q|>Q@uV&%%Lk-$^D9^?KWSalfAoSaQUkB*0})PqZ)j
zq`BS!M&LUjv1~tuhnui8g7=nfV&D|<pH@#noV_z&v~DXF4&H|o+oL@!)&Eu-D^~%0
zE3hpcEJM*@fTb(dbGmEEK=Bh^XsYK1Uu(vKM-jjX#@>xk7cN4Gp}@Hw!3L<Cist}f
z(cx>8H$75iO8kxb0|+*N3=GYOs#(yb_hYJ$|Myg{mGN)`Kf>vN-E6@2n>6tjl(C|o
zu;9Q*Y)YJkO`{10!WsB_P(P6Pw^@msK$%#=fB~>OgN)v-GSt?nix+cy3;)rRu_n-7
zjcvOJ7uvbb2zSmh5JTa4U&{O~T84JsF0Acm`T7IV{Bf857XBAT3E);a!*am%u^Jkh
zP_CMObv)q_(55kQHq^^k4Uik}aYk?xGKvOO6HI11sOvgOc7SWc)`j5vRT{Rsma=@a
z;hvX6N#s7*+ACcu{@;pU&bS;<o3*uV#0X)4<qIxqu#ITSPYiS;IYaV7^$QFZOu$pG
zqnz_;(SG`G5W$8>bU%LZrZS-*%5`lJ12aSoxPtrDvx@e4vyu+UATAmt{+W)5X!JL|
z(#)}kaKO~ij3Ec1k`C>S+p~w;f^XpwG&Cp(x_TxjVaHdYNnZ*>b3<+~Pp=*yRu!Pf
zi7((%v&NJqt)Z`RKA0^0tuZ?G0#C}Zn3w{s1+aLYxpmeL0=?-ab_US@0OG@!yfU7g
z4(A3G<xigDP#kwaC1FfV;LVBsT`T7LCUDeR&i~S2*t~r9i(3E)RWwr9HBXnO^UNE^
zq{Z0>-=ATDdaqDKxE9^#$bbU`Z(6d=W!Ry&5Y+_BZk(KE8SX<eA&EjUiE*GUfx@$u
z&da7!Wl6K<XL!dUZl?;K97$ZmE=ZL(6v+L-nb3(L2c@~+H+?sK92Ay&-=IMd$2|-H
zc4}J7NN%$jzzhNK7~snA4;_4R@uJm2U2eL${^R}k^&^h3hfHZUj>t$>n8p*rj;>BU
zvo{W^X+sv0wf&10MFyqoOjhE=TK-$RvCK}Qk1J1cN}@f4o(v4o%NGFd;q$!TZ+_f1
zzKjHI<=pRqm~M@CS|#%oLf-7ydYMWJeYxTltiY2pQ>_NQj7ivdkln5u`*j*a%96Q1
zP1OuhCDco~rheFBnOLegFN$~}5nR(-KF@aEhXnO&1^0dAo7K`}K2nXG;9bp}{~_5r
z0weNQl6_a(DsPF`%!1Eos3irYXiv8dk|kpT8LQM*l?KiVgD@$Q8lqyN|LBn}o-5^&
zm8S6ewYf}^wgT)_FJ?mY<u-6r7aZbGTT6Ha=97z3S{q7BnXKZKDKit|PT%#1Yo=bW
zNx#GAou8C19kDZ80=GxAoyb_AS4<Ie@WCL-1#WHE1tJ44?!$FAQU1Guh#;O`dNM%%
zDjxJN$S#9fB3CxhwsL5xW`D{Z0x6Te(gdPJVy?U-u;G43aDnfZ3fFXdQ$Yn3VCS%S
zW58l)2L3&xZ~W|3OL?zs?Eg_vgUn3pbeY==KrHCi4e&NgL9{~c@V3q8G^@b*AmL#O
ztBpi375(hc(p(+QHuTRqYy4N~{xw^=l|f?65Zf!v%zXrf*}rWV6iG=G8~|IS&M6pR
zZT7CkBt1W|=a=tZn?w>j!Ee1B&VuFtvbkEqI8=1?4PE{jx?u&ZJ|LjmNUiKHO5g6?
z(9O}{Pq+M-@7Dxs*MI>_cU3NlkXW5y6M0);Q=qZ^v&U8_@I7X-w)&W6UYDHb-JP6z
zU?qG3S*ML5a{ejYjFL3cJFcxkSx<dlXoJaH6UUP(a{d+GLKz(!U?M;BI?kteeZp_G
zt&Q3d?W<bTfyS12c2U*F3OvKaO|n0hWP;8?J`f}$Pdh>aK*=q>*mH_dR!6tC=otcN
zO#s%xtmBTr$fLD==FyrtQXlUF0v~A_B;RIV0_&#90>R1O@!zj39Mg5S_X}PIc~gPZ
zfV)Yds)-#NsRa-D^4QDJfR@nL?K_%>3(cY7ut0!2*CEW6m&*j{erq71yYJ3ZRj@I*
zMfN&%CeomQ`5b6b-fgDE#ds${bNMcvlg2VZ>PH@#0+)FUZ(467s@oh1CxINTtn7DF
zi1Wm%=}uV*nTO0Pl%kb-kY54ca?O&tAEiVA(YL-%MFe@AkJ?hmz!x1)r}7^?9VGZF
z;8viaIwn>{fe&=)+0|&?Y6?ghqM^S4#{pX!%_BlpRUzdUx=iSSnWWn4DI9)v{Nz<-
zR97o1=Lz!-P>5+p<@N#H{5h~E!J%QH5fcS(zs%OxIzLuFvsa4>JD<Kf8w_5@#McY`
zeWC7{3s|yeuCc9E_}{Jocu8>=RmQ?{*N6kW=jL#5LgUO3O1OW1ZMU1$0p0N{FtxBs
zuh5Ojvo0os{fMn*YL@z?U8*5c_CIJ=S%2Wj#)f>Fq(mp?uMW$PVAKZJejgHvL}{!_
z7>^q9Zu4|TMbtj7J9Y>HZ&nXkTMvAWIPfJKCvZC<lZzjiSw=u-1LZiX0l`UDyr6TB
zcNZ1d09~Gn^rU|mF#-R|JBSpCd;J1qI^Ymt0A7xeYfQEB0V}$D`!W13tkXB&c9nr#
z5A{`K$tEhqu>X&B_cL$;&b<xa3A%m`5)g<AYd*H;9yeek&*d}HxjkVqzAUSH-p_9E
zVCNt6Z9nJbQz4gG8=%JWC}@5Q147N!(F906efsU&#5SyCAtu%~%YXs`$<bwl2h6g}
z40I6HjIY%Ife$h<rK0~7<@o@yQUeF=?S4V`;u4bQ!(Bikg)q=YKS_o?U{VJi%++=d
z>IAR;tt#kMhDy_Itx7g@r<_8LjhW7X^{|K@si2PShzZP_j@=t1JZ$nf&E~e-Z_mLc
zA-}rb)h^thiu(lR<R=9@F^~bZr~rdtVDfDVPH%)wQd@vM({|Uh2`m;QXV>d^JF_An
zoNyGLAaIM7C9nu~PJ*N<^{2N_F&S_MO?rue1PwkCEw#AzHn1WCoVRxSgb7C!4F1n0
z3BFM|@H-TzZklLq1Qu9GgEG{QaY<k~x^4NNS#4&!OSAnyIW9j2mNFYb-#~C&4eOFN
zOGL3)fjx6-U8vHa5EUVR^+@Ibk}J^})sdSaf&*6#?tImfo^ap86kAO$;-&>65bQ}w
zTzd(!iAY4i8V023uhk%x^dH@VE4`>U$jZdSj0yu+yRZ?8TU(t9m>I^JeqLg{Pp*bd
zk_i1!V7=6prRU|DQly_&t=a)Lr4L00U8{*T`vx2!XxUJZP^S2gy>Nvo*2avP!yTz)
zZyyk!1X;x&CaFKs7c<{YF=7`p^GJp|qJ@K5P2^an97i=S(#a71COCs0@8XH!vlnsh
z9LRnzU)#~4bv{gAc|4jNvFFDh`~80m!%t(sH#jIK_h3*etD|YXYb1fIX1jYM%bzj(
zdppM%Gi9)rk(J_WRJX0yuVaZ$`6?$1N0!L{)M;Hcm1Fgsf_63c;s)!1GpTkdxe<1f
z??ebRfRwnefPA!w-eSj)upbT7P*r_q2p>2#J%Bw<R!5j8g@#%XoJIuF$Zx}|S@;wV
zpung4#ES>qWS|~{nZZ!Zu>UaN>?a+<4H}b(KZvzz0-LJG&u38VuNRcQ3q&3|v|u!-
zfT^vO0roV259LlFBCmM2S>lA2^EHZYBs;DfklJYJ9O!W;H6bKImUvk)G4^c$&e5-p
z4D~|>C3<rS#pe8b8l7)^Nbe8Q2Ku$EmBL$Pu8TG|doeRI`opD{fC)t^6@c?Wh7EMc
zfEJBdiX=1f#V!bB9QV~|Cx-jHDI`B<a}5q82rgeQS5q4xGL7&oE-n(p%UK)Br~U`!
z3sy}1aNJ4iIEP&g7$)GSf4u(3^A_i1_km8Z(sXzTMcf$W9y>D)%$kqPKbx+)o{yNd
z^uNT;6au(Cw*aQNn1+_RWg8a!B4Kxc4Fx{u!H;i(O-F#jorJIB;LQ0t69V3-XZ}F<
z0lCVecHufsa?<a!FP$4d+%h{dvAwwgMLlpao<<~Ry~$fz+B=K2+T5)e?Zfo217<_;
z@UmV!RNJ0jp02!w9(g+p<RUstlEBu=vWI1l-M@oN!Vy|_!P6rrPyvR!05>P6s~@w^
zGE<;7iZkEP(%ccR5Kz@WagOLHj+?*zg0>}#U+TLaNX*y<#q30<r!siRyw(uS%zoIb
z=(TK^0QR6IO;A(nsOCb{e69>!H8y=ki1w>!Lq^E^9%GoqC;yRtmxlh-iY)pK<GYm}
zAVyHqp%nho*=Bj1G;ZNVgD?>(3a*hn0JEB*s>Dp8(^SsBwuoU-8HK&j{FkwB3&jST
z6xPAP)Vnj`><PR(8WLP%{w&>KcyQA=Lf!GBOWx}ja2eAn-OZIV=ORXj-L%x3EdH%o
z_DFeEbNs%%)lXz7?6%yBBF8^yC$jVZ@k)Lge>bao*8_|NkVsIBCBlI3Urh49WG)c7
z0G|h|MSM|Krv)Paaj5ygITw6XiW%6b;0vJzknUJgva*VrS+X@aEs#hbEGY)C%deat
z?*3vmRfwnJVv;g_QSG0`B|}VVan@b?i2!E%?2RHa#TCOt^B8rI^=mK=luLaTl{iqE
zAdtBOmK|hL3kwf<sbG#*3)bd`*Zzp58(RAMLrKBMQCuoR$UyzKUkJEu9S0B~Q3V>p
z4(?uGCf)@7l4hobh?~BgK+hgiVET_PeG~b~9`zNkC0l7p*Psd;bXOsL5Jfy(y^Vvw
zXiw!GAVOMpL&&E=-6>7i4fOZx`jMwpFnCqpdy!K*n#MO5ops#5s4$JrF5=;x!gEcK
zpg;FWc~Ql}labgb@ANbx-u`V<=>qUnAirDOayb79tZ9rQCDWPT&6|Dp@XO?vii!#W
zTI6!r<Bl%iEzdb(gQ~6v{HL`;d=~{cC+cPNz$@y<7dA2@@pzcUaeiv`O2YH&s4{qv
zbb;GLzp)Kqwp;Y}J#6Y_tkzCyORT^}ajxRydf*M8)JX$~H4w=*aDDBxc2=J>ipQ%r
zMW6N}W^ZZc5NkH({*AIckZ%Cq0QOGYlFX}Q%Emlkl*ke^3wF@lf6hrS6&}NYfL^OK
zdDnn!i6HEKm%`PeF>wmGoI?Eh^9jLi`C387@kD~UXdWr17@(^f!PjTmjr&+M1V+ny
zUk$`ToOPx@eCFl}S<}whh8P&&O#WKJ0+3w8OquFnHz1=!7832PYb8Klvg7aE&}YQw
z$~0F!0Y;k$LZ<5Oj|cg}4UIcG4w&U1=fcNPf}I`F0AHh4cMFqz0$Q>1b=?Vz;*JJl
z;Rx!Qw4B`|Zw`ze{XIAuYJ_F?XQ!YibG<@JNGnS{Bej)l*hl37sZ0wN@^-63N+eoD
zeN|7lwzPV!{ApJ9XIPn=A1nq7EHT0}ZkUp^aRt5L&F`v*zs=rmSW+l{puKk6@I5{8
z|I7K~AruI6e)|cmlru^S7pD2<INjC^aj?K|{Lx?Iz@&ySz@e&7M{pmcY)LGbxbu?E
zM<jjIIrHZ?Dzln)4G|<HGZVo;kUK45k}~keODvGI9g>Y?-q2xW083r`1Ix$XjPzeb
z|7jJ7X5?L}xYDx8dAU>f>C+h3a%`>1{^{r#Afxmb#X=|-UsZtaI^l|%N&8R;(EL0=
z_z1``;DeC>!NlO3H#c7-zK)^5f{gXNm775Sn6kz$dGv~6;f|&{Ih)=s@S1{@^f{}V
zw;>(rq;Lo0NNp@1<>KbK-E6D_3%+5Rw8>6+A95T7M0oLWc!;J0TuYUfH2od)vf0)2
zx=v-N=8e}j72@wYJA?CIZu3y&D4+kc%*Tl>SN&L<F6Bf67J33_4EJY#Z;(~`^r@Bi
zHMsrkjX6LfaVWO@59Q<cK{OCp1b>}`{FS54PO}SQ*JpONw!V+;b#>i6AU)}X&|+G!
zbNX*uWA${-Px>P2XmGv+3`wm5=x()AAVmGCZQWN~7y*}W@>j1f*}>b<3xcxifD$op
zk`5dj4DuK)Ux+K0zW&Qrl)Jt?UX<}#J88)hsA&(UmXJ^{**Ss5FIpF~yu~?FDHMt+
zixoUPRpPd(H7Oi1ro3q!06RlK4VRvV+YuTNeY?%WR43cO`l*H7CBD<=7$TZa3jzvv
zyW(j8b31=Q^l|_X`~%Qft9PcGg2!Edph)|hbBv^%66CTRlQFYUl8@jov;nK@>Q~g$
zJK*|vDKK?3XX5eDS)Mr`gINK*kE7LVySdJi)1X%+-;=p~J=nhMeYe?VV(=C)CW5k<
z^2yN{0QOCB)UY)7cnGhd?nT>R?xGR}kkcm4P7t2I(Nn-`*Pa50^NKgwz^_xFfaHd}
zgkr~#TB-%xvjr&B@4ckWs5a18eEWJzH+D(B0mJmo-}~h|%b;|1Pfr##D3Mx~<IpIq
z<hm%E<c)~*OVSSrGk7z7YVsT$OA|mz1)0TX;Kn`;W=i;xQJEkiGeBR)+U)wvUtSj1
z6cDKVf}J^Ow#HtCgAnDBrOnf;vV60m4|-K5Mqi)zri(iORk6P7^~8<m?1kUuV4iVD
zBvDhA%=-BsOTQaP%DBF8TizIc15BeJF_*&$krzJTo+X!g29l8A(3_rTDik5LaQHH@
zH-ju{nrG+w2JkHb5FVZuZw%s!9{zN(Wy7jMRA9O@!6pDUJSZyounjO$6N}9#5ckd=
zfLJTYH-E}j_nP?lYX$(Nv=vE~B-7TH<jHk(vN`*oP;KG_id?h-9!d(pH6_5vs67`t
zs4LJN;w22yN!RC<gn+45nhr3b!#idRG_Eo7e7KXKX1MOpKug>Q)W3l%M6?zrojs;w
zNlhB!88{G3$dZmG@VWUK7i79Mi?azl3&mfg2zHtrTeWploDW~P03zo{MGA07K3Pp0
z@~OzHXu&-}nWXj+g#a+hF92-g@}7G<R6__85aCITjg1N7`$RidT$=(8Y6n&>S=XA3
z0IVC2GC^jO*&6k&f#dJp#U;TkE6enC(8NFd>n5_(&l9U*99jM8mJ1U5m&BXFhi6y=
z45>m=g8;^8-trv#c+BhAIrBoka5W?}19*JiYpPC+S2r><dV8C$f1%2jg6aWZ3;?R?
zr8s>27h2UpqVTvogi0W1Tf%R?8wh+jwc>@~#FwAgerUnS0EvVItfV8>&&)iahXGV=
z#TfbDx;7m?GOEK8-Zj0>98uI!f6N-A#^im#6pHS~kG!M<b_&pc_4_Ws>t7K3F{TVC
z;9rutPTnUQH%R19R`>xJ)~{28>WkIn$*bYO(ju!O$XTNPJ|c3pAOc-mc5Ziv<$G{m
z+Ls_7`EKg%ivUx9HFYA=ENvZ|PRxG#Gy@zOF53B~ohVqCe8!9y;RqHev52Wjh^m&7
z0^~Zd%N;<7b@jh~9OcoOXTsI90JQp{zJ<Y>;wivQc9hTpah0#2w&OMpG3^~KG^=_B
z4)bI#0*u2phgZPl{R@?efp6*c>((AF>vrK^7r{}XF~iH`uhrgripX!6B8@)TfHI&i
z1#2IuEy`yHP7zi_&A^ujqK}q5^l#E0FNcgl1kwEwfLOMZAJ4>t?0sDxt{92;j5zF|
zNV<Vx%;UgNKhvSI^hZ%a_bDi#T1H|=NB%Ug-6QO}UbLTI;>`fZu6gak=46PRNMroR
zUR{PKsx~m)rNn9<`(+RH0nSPEu+lJUv~dF!CtK4T@SQk{^i%~Vk&A|aIv508-0zcl
zo!G&XUJobHZ(Ps~dVU+H1y;;~8AFW;KITmMJO;SB{2$)rgKWKrj*SHMj$fe}Xaxu3
zM#Xs@5Fif+o{{~#WoD^sAh;ynhcxhI$~&6l3BYr@M-Www$4ijuL?^c1tBY)yX;6)t
z_}81`hNWkH1lM`fAq_Y~+o1C$Y0Lms&Ee80tf{_oN}3f<KKJ(ZL@4j4$YZq-jp`{2
zk`9KA;<w-N<NG>@HBK8VWMxaTc>vYW&Fnk)x%#JYu@YaQ$~(=0#07AUDKJcfl>{*0
zdmK{z0EPrxRw7VVap#GF-vmH^;swjH0!zD|u5O_G@3``(Kz<X(W*2Fy%KQ6ED9{xj
zSIcXtm$uYSgY@r99X${<@v^^b!|zfdr&L$9NruEaEt)<&9^Ct-mqdrG#D}fCUm1Be
zee3+4O+^bLr(B#nL&4a%(L5NpNzN!)n1|oTSew5D85gciy{0mQ3ngakJBjwGI}fD9
zvl(V84&gD^rnN1tlo3CR!1~#K|KQKvLwc0jwq&ZE{&B(#g6w~aAbP&h)g-3DL1f-i
zO&4EGy@QXbsTyfm!Q3muleq5@SQX0_#|G5{%Sptz@~rBeJOisDY225cjiDFW-+cot
z%B9Lnxl`g#kHis$e%W&(PWp;hyRvwDn_Uy)b;Ryyn}(<*b?Dc7715KW2YJyWsPIRs
zz_2RT%|)=*!+Za|{`At;YO%^_*^_BT&(XBA?{5H3GLkd2=Mc5Pa##bmMT=~}DTnix
zDiQhywJl1m1puo5{ASMzXT!l9mX*QVvrpEAyk=vvxZ38__}-WddFMaT4O`TA4ByF1
zO0tWIwx)CYZ9VTHUSNE3|A>Y0Dn&Ns^lf}ci6PKXVM0j%aiwl?7JGZ=C6p!Z1k5Ey
zFd+zLKaTX?p>H2%8(-~B?<C+R%U%-TGewP+dPW;4Zl|;(?q>D?WKz^LP-A?B8y2=E
z@8x(VX()TPA8!VwI(!1sI1A@rbF+7ZGz@#1ZvJvljp$KQuWFp<#ToE4?GQJ!;v%)y
z;2z(4oer#^Z%tg!5>)i$wJo-h)9z+U(RdL(E-@K%Ru4SwA#%nI249)CHff0*(ZI$(
zOBM|=ytV3BHhoq~j!aQq5#)h>?F06*W=+6pk&Z}aI~(6)b7IeTGpWGCiuq{M{q2tr
zRrXlDi4N&VRL-zMz{V5<AZ09xrgUAJvrunh7npVRyduK#u;fu`aPsW<W6@q-yvR`$
zKDeF&c^NgOx>Yi&5B9q0Mro(4I2R2`>%o9}BrEeg^DoQf=U(Ua1*i7S8?)?6)3g+C
z>#UQ-A!`Zws{PxQBWl$nmd_Yp=$Tg85!^?%3~`X2wwAru+|5{N*cAGCcQ*3#JpcG-
zJ8v*~9sjMbZ#zuCM=Z(jnEkizy#cY-GbIA1vYwB2mWvGI#L6zSWGpQPG47hOuh%-5
zSMP80vIGpPSyf>rCLGMy)Yn=CruJO=id_01c~j2&{z>au@KLVUp4+euSh116!pG#v
zxe7Z9paLp9c&O?3P%1Gqtc3vj+9ygoG?E7&6Aydn4`EuPD{eYOAooF4iH^;)d<h0%
z!<BlG?>-zKbeu27?FXidih%#aHvv9~lsO&o`iUaivM=fQ^&6(lO-~=4qN<$5iIzXc
zKbar^7+|&tb&J>M^rj$FuU(n)x9w2Z;m3KAVxVXlNYgL7vif1=49UdB{tqGIvgh*&
z;>2>4#?9hvtde$%C1|C9YYwL%0eauRv8;OWb69@ZPh)3N;|YC278lCzSE1OidaO4d
ze7_$CGDS(2_}Nws1x`0PlPb$)4=U7BCPg`>#vuvu<aQiIos(PCR-Bv}p@+9epsGS`
zTCZ;FboFaYu?-2N!cG3H2EYBvtV@W9^y|ej1BK}&21ApY#_hko^I1IfAqq+rokOxu
zMlRf+u+C4Y%uqUN{;X-Yj1g<NCC4JeJ9^}#>v|v7`TP>>;Av@2CLDfD4Sbn@(XvJ;
zSK7aaLr_Nb^;YWL!bf0OtF1;>V6j8N5JeE0Qh`~8M`%gUO)1(fa+*n%?{BuOb#Ihq
zPi~j2wt;#!K>|n0w3Ls91sU-#gt#+o=%~MBdo~;Go#+z_&P%3SSL4`;98Wxx5I18a
z#5`M5jbhbHh;sBWXOds62`Ulh$r?7@228{8t#Sv3G&suREnLrg#I_ka#?6)olh0z!
z5WqErK)O|F$Gg_~;Ic9CFb=huLNHJ&*!Wx4DHdSKqM}zjNaubcvZ>g(c~i(+F|@mn
z(wV=t(c#qBl5okdlGGGb6-K`_%Uh})+Awr3h8K&JqG)|<*JYmU_j8l-?vKX(pTqq1
z#q=t+egnmgUjIz=SCZ83Jj5{N(g+%qCH_pCq6Imk7qet!T?HOCY;3f#bRe{zThRwk
zAPaPxc3>li#g6D56%L=xMe_7n$AI5%I}Q~|>a)|_w?BGsO?ZSP3c4?lv?qW0?vtK0
z^`3O9*=MeLV#NKPay}}I<xY<~+3<{*e@8OGF{|aB*JzuT_#%6Fz;H=dSfXINV{fV2
z%zEHLu;&dq**-hw#CgmMn@em4lwrK;9rPb2VjmpZOj!Kp>~Erqr3`yt#EEVrnI+|;
zQ|8!*?*)IVw(xLm?{SpbBc*#eFPLu4NBN2pbFG=mr_Z{)?DrI#^MDM_o#U)MgE9|Y
z{rstML(d*pP+aW(n+oTF(sDR&a0hXbBE$9$jhWk~pA)KWOLd*Djuv~gYVYcH?kEiB
zL{s%;gIzZZ3PK&zXA*&2n5JTskVq~qQ~aW_^}Io{l~vB-<)jl6Lp*<EWd=tp47|G-
zc69z!y+Mz_)=jwV858{qW6?$krjNW6L@Yb9H}<&P3mG^~@eX_bpGM&^7rYs)Kv?(<
zC?rFw65|nGh)d!rus@DzS(#*aVXG$AWjLF+SlC6`Skz*Cz3pkF9Ud*i@-pknr9IMW
zu<BaAV^K}(jxs#d;o#jhn$?2u+-8S$eq3GASeV@Us%LkW`SCC;TEkyIOaw#$+zN*J
zLKb|{J$p>><{tUQ*O|XA#2)^Y@+W>QC7I|&QKEw-c13zd&qVbv<CB6naF9{@+aCh;
z>PDpx*77{Kb+tI9jp{u=hAt$`jtj?Yb)h7xk=G;jHU!gjR2az#a@WRwNhdAe0iQrV
z`cyTh!jd;V*4>^@1p@K#ZC4XoGY%)zu~^6*)0L2CAcN4TP*BI4o*P<Rd@z&YCmoZ`
z9$YrKK7>%5q{(q)2BX&vnfzl<8vn`u{93g0Ty$}k^U0ztFZR$c=G<Xi>m1!FX3RjG
z@%&vwS=3?R5^u`-mU2Xy<GEmB(n+V8QJ;@Ecv2G+``}X}<P@Z8xM<fA4*+-im6ahV
zUWo$wiYA^7%Ac568gM5asreerHsTeF71<!uv+NI7=U};5=)sbuX=yTswrYoE>NgSr
z%$W%e3iU|g`bk@GqF)slf&pqYiXz<v=d7WdgIK450+F2jlm*oS?`wf=^5C~~_U9{z
za!69Yd`%qVT3q%m=Qv_(dKE?99|kJEcSHT*BH<z;)}(1y3IGSu*@Jy=r%R4!oY}pd
ze|SSC&2|;#?%ez|CZNw&5s!Z><`#Kp2eW-}71b5(7~CThDA>VGe|2_>)AH83urRS$
z+jak=ZLuoM3M%HNb8g&9c~5u_>q<Y~_Y+si$h6gR*k2@mriC-ukDzL8g+kX%Wss<#
z&=eDC?OBUsWf)uST0i^8T|BA$odM^~fK5Zu)T~RCPKs0Gplf8KYi&(A4yG~%h5G3Z
zr`xj{Hzq0B)W$yQ*E;pPy8<OSC`O#*_-l?0@fI!iW#}to?BS!Bj5i218ih^Ii+And
zSY_R#`fcC%bZyZ+R=D#we=Eji>63{y&8<N@d~;-Em{)>6D>BiXbwxy^=Zb)!^Sr&I
zSU_p_NTr&Dw|z{Aj1s8fMhQ1R*reDIlbMt=f!pA;>4<c*6oza!Yh+U4trUsq?CVW6
zufO$BP60{^`?4N?4(WCm^98DT|K!r#;m8Dzt1)X}BvJZcBOr>LWX8lRievSt7-yi!
zP>6CXd>D(vsBfH|*`rF^uWY)xYoE%y?}9c&GcgXGv;T|nr5kiG=i@mPj}IzYUwwTp
z2`$+Kbk9y{z^yTeXHxD>s(Mp|J{>7E`B|2QY;DjW_ucgcd&Aue`S{v{hC)U+BqahN
zu20qtMV_`*V;x1kYGe7U%&wxSy*{Sh>s)pd1vaj(N9y@d<6-iPser%A8I^i3v`%&~
zr0c*s>?Q5|$T5ypKkJx@&gLlF*1O&NP5Y>xtS9%OHPL{NVTD7e4ft4a9PG@B`Xd~$
zLar-SY~K8BCn@NWTWIZh=t4r*%Rnc0L8$w6i;dwL^m5xbJ<$}px#gHvq)XCw!u!%^
zar^X~5IN1_C}$Mh*amY4bt~E_a$vsTP<DRm(si1kLWsO!aY?Xg!sgfLqRBR}{9x|p
zNJNK|{;I$Jf%-sRWoOhVzR?w2elRl{*C`g89ccv#A_zgUfs;Mk3<2K=1L<X7HZ?1;
zEH_1kX+*1|rNILn?%^YKcB<c7W;qAyItu72&Z6V8Tf4qq-{&PBcd?{@m7`frCt{s}
zO%2Q@e{B>U%hU`qNfe_<VxpB~tu5Q%t=ns^%xDWZOS&~LTvX=t?4hiOckZ=KEe=zT
z9Y}N<%kjHx3*H~DSGnx(5bxi0?`YJoOu<k6RNW`6b7klq%3f~p5u|krwRc&XTKLjL
z9d@q@NyI_W?)D``$C^0Y>gM@9*swQ|CnyQl{T_}`sj?}X522@TPTIHfuiF0^-TV~x
z_KvU2LmR3YmF0+n+`KWH;G78vXpl3m7ZUokML8`zn@iSEt)$=N@7U2VFH^eEKr$kC
z-P^u7m^x?+j^GzYrfDO2rU<EnAv(eft$pz-KA^ldU8aMmj1u78{cgA7(~H-sRD(!;
zu3Yplh>xzJlt@mB2G`v$57WPKN-Z{@4Y3%@>M%;Xzs>3uymd~-0C!OL4_(QIy}d{g
z(CnViaOm}cMYrRc48ztE!^G;Y<_iL~;e*Q4G=x|+c3eldq%1^x_TdSc;}1nhr;6dX
z3|@<yx}w}JX4fubwL3>OrYb58&2@YFdTs)mYognT&5=xTqhn)Ql`|bqFV@Ik5#TU0
zNO!o$Am@<BoP)=8q!cS!Ui+>u2e0h#>@nyH;|tV??av)H6ulHH3ikBWj$CB3_X@Qv
zb@yzpw><B-K9hGF)ZOmY%}lh#$5Q;XxN9y93jf}qOG2TxNm~Z$RT%$nGu4+jd2@*a
zO&t0|{qOns@dqMvf<Glukuy*vGd1xQ<bLNijyZoPBhwZ-=?I=MOOI9EMDWd7YyG+~
zr@`$#H*YULAEq$%4=V91DaB2Fm6&ArR#3#1(HNaFBT>pA6Xm6>^`zMwhdlu)C36~5
zOTK*CCYgWi&b}Vv$+C)a?^Ug3wgQV5jDP6H!&>8*>R2gX`1dW=Ks{G^n-}$e`<k#&
zh6K&a^1JAcboq)Qq`u}RT|bH?kr_g%{q5wo6j-pDxMIKOU*PsR5Zb(IROS8gkaR5P
z-L=mS$|_SUtVZ)9%h}O1Esa&J;O*Dj`RVFMN)5comDz?7gNbdM^6E8zDk&u-Sstxe
z#ovOeRX;Fg#n=X9(wi+%-Dz-lCmt{M&IMlKV!1oBsg;&^d6o{@Ge~ojZcpAvfXe^$
zu#$h4oa}Kk$7UN1^%Aq+@T)`c^$AyLW`CM~h4T-LhvV-iTSuoD$C5(LE;>H`kTKhV
zPSA?4b+tT~_tFAN;heqGCu6r$6*JxIxfEKk9&B*F0qYf5dS3L08~8sRhLPFFw)Ws;
zyolVX>(6KnWt+qb%)a8M$~jgHIn%0ZX4W`oDeG@8>Jw5GkgpBfzI+aZn`F5ZYsC#?
zeMi~%2vL}o?D@$MO+~oZYH6bs%ZBk!ApKae;lr-|Z*sY*dwdP5%Rws!AoR}bV|(6Z
z#m})4Ack&#-GsFMzG<O1{AG6CO?-|0xGs4@-g!#?v<p)G)6=P{H<@d3ZpZoE>(M(A
z;C}74oy0s!hu7B$qBe`k1h9`u*I*4qjr)_@YJcPU)+ZsnkwO>MeeNQoFi+w|p8$K;
z@Y39FHwkVH4yp}PuK?A7k|F~NDkW__RaqK*bnV031Ll5I#SLH2D94+6Yt!Xz&}SP#
z1u3zV5GZ>rc&M#3xVtR%K+*7aW6&TjiGsAFZ0aJz%cksHt0FG;(0>%GSL!57onnBV
z;>38eY}0K#w=se(Ru#K<Xyre*$yw=W+Z%W*6;<xu?@1TkX$G=X(ur#o_P_}qIc{e$
z8Va1DzB?8(7QE&%KC{CyM)`<P8_!CF^2^cBc@=Ujc;-xmE7f?jt;pak$Y@l8)+8gn
z(}jRbH~g8DR&g>0cQVRydNa3k^hKFuTXw>WCf*eG0h{)6t;1XbJld#0B8euI2CEZd
zI{~oIp$n#*1>*io`uB24|It%WYw)IHc)3Gj_q(u{_|bk9=9`??3(Lg}@<hw)P*6{t
z^uo{z@BhB4pv+ImbjvVbdi|^;?D1LW$$A9z{_@UuE2!M@@_$Duyjh%jHp-DlCngQ!
z<%w9*NCwBms6IC(_E=5M;B6!OzFI^`pU~>hpF`yt$gs?JOlV?tO#s3K2CE+BOHxe~
zBDs;aEf|pZvIuzb{22{p@(r725ns@_L$`W`>L6a_OZ|<M)w8<p7RTlvRS0cFYaMuq
z`)W+Yx@4Nfr0c?Mg@5qJ!<C6gF20Krv!cbhPVrNu*3xY5ku>g4{ZR#;EpGPJZ6_wm
z^*MuFSeZOGE*+B!1-Xirs(kfn-VFPyh#D^?gXg5^R8h-x@IdjJZ|ab>rXJy9BPw!4
zTc*^rzZW-$5`B5Y<RSQGMy@#6L!e*_zA%YRV`ZoMG*Xn)@$%DVvq?46*5<?PKG@$x
zEdl%t;rjuK@55bN>>GR5QwXYe!22hnK}WE*3_NLqccTONX@+ld$<pSTKt74fMe;}?
z6RG>TgJ`7yMCFr-gQqoLU!QigwlO^$U?4sJTQLN#nSz4vwF{cKFRs_|@u4waGFpdK
zMlk85VijL+thuM6!?4(A5blVa7>FR0s8jrRk|(jw|H?2}n}|dNa>)u_*0JGamL9Tv
zJ`4nvk@Y4mXr`)jvooV<uFFZ`&4p>#M&wPf#=^9mBlaL98{Fy3B-2B(&^QGaRTjKo
z=bx8mTMYizOwKCFS(kZ+668~dStB?%rMZ<Q90r8yuE^`WG#5G~FF<==b2#_i5UXeJ
z#K%r~1x0OUYd%s$V+`tlrkueF<M{r7NTnJj%#*nx9)$kWN<=}YzFGG#4!SbRdj$@v
z74;~1yIrP%%U9KH%dcf)x7b+Rsfj0x8sq2D6IH)8!OmOs#kL|ujK>4S`nyAX9k-m=
zy~DG1NNz{gTb05ltRCNHqf_xiTQqLzx<~j$@Wg+HslR;1qI7U)a|1aon1`v`Yr0{c
zc~OpO2`>ZuC!SqRY&$R!ft#*5F@CT0gEgwwNfMA-?nsY8d(z!@g>`Zi^5PBo14c+m
zB{-bNRB7c!$hQ6&kxzK#3=JsBBk=j4kGb2kh~$|{W}eAJzZ9-tDVuNjS|QI#EHUaW
zA>UGUHV0Qj67ycPp`kLTD}*qiezc?i_}_iNA?V64iP9FADV9AAzQGIP@aLDy-?|BQ
zY_gk&PXMxk-v-wzKLv@hwxh${X}r@9zBIq_EiTF=RlY%Z0#UqPMiM2tJi&_&-=c18
z3hL4#BS<$7<rww9s~V>Nq_Y!8sKx3(?sQpm3@gig$wr}7LYUwbWP)_x!RGObnNEJ2
z6ugtY&ZV7<FHNL2KWsK(Ylw|T1Rv+=TcXB7v8Dmw31v8pWdF|5&DdW%55VRl$tah%
z{W*HPCHlSMeKN^eqN8-cyU_cdr&29rtlC4T+s&zH^RI1IrY$a-B0=DL$M*7C?W-yW
z(1o6SE?ESXj<pg5A2XF2B{iX=k#SYMeoUIKcXE;sAr!^0rHY<LQ}w1Pul?_$7rM{i
z_Uq3;uE!pSF<%(9XC#{ChzfoBxi`Ett_Cw<Yp%>U(N}##&XUUXZyVxvNH?V!>{Mh&
z<R~U?wj*6WR@qsnwp|Qb4&I!k)A2oufdSs;f&tK*0Dcs2PW)-DRSzh1FlC>Yorj{A
zEX*`<j@QxPq`sQM?Zne`+{;dkcWp7|;1aav$=YQM2YQyNtYRi;2L7uu{$zZGFWX<Q
zb(A;m;Vv}PRQ>0xFD7VDfVdIPbatTv9FJcI+pY+&@6ap13GhE0Y>@$51z2qJE{T$I
z1Kx+M&uZ#C@QNri_Jy48G%ZoOQZ%BjUc`6PuP0NUX-QP1wp^P#IbnBmT$~*I0j{-O
z_d28%6@BxVVJ%h=!lG|E>1_-^vIdl$fWMBI)4uk*$h2D4T!ibC)1*X|EMXpH@7Lx@
z5?|BN>8Q#J3#HTGrIIMg892e;zSn8rFS5Pqp70f8ID?Xjg>iZ+AgwV<!iPjXvX9BY
zDPyT7ha*;Zii1E!IkwRqh(6{C-%}_v{N})>Cn8p77d}?4BnFwDVgn-h9@*vP(u`R?
zr-x&eQe)HI=Ny477E>(DFaGWV8Zz95Xo|uAc29hoA{<S__#}z3Py4c<wc<%vN9-q4
z$`O$j(u#Obj^T5?$RgeAR^bQ$vhfdS=+lr8QsZMfo$;~(CmS{e1%w>E&P2!1!M+_H
z4J8hB8Qan>A2Cv)hs3uJ^fEQwu>o!vQ*RGA(hL%*dx%gKfs(VeS>%+-^>%E0=5}58
zgH;LhLhD$c_RAjM;F7=62v3@*MwNgTMl^Vjy4E3mx=8g+K0$$1eVQ!YlaqO?trudz
za%EjoN{fcjvp0Bo)3JzO3M3S8As?G>lx=P{e)YjDF#QVT^87D5P9{t~)h!eL0>csY
z!k`c1Nr{)%IKV2U+s!QOt(*Nbuk#T&X6V6*+T&Y8Njwl4-2EtM@~uKBQ9%*!6iG4C
z#*#M22>N+y@|RbwGqKWE)OnZ%YTfDD{tfdbm>nD{@ldQ?`$u^i@04aRSk0qE%4TCz
zkXg5vpN>jXF8c275=AR$e{1k|q>S$jSrt0ZLo*ZCFP@r$-<?+ip<<I>LE$<NC_J!w
z!V1bX_t&75NzJH;b6ognQw(|kT-haJHN-X+u}5h|IPHtm0^o0XxwCbpWjP8J4A9>q
z3)N3;gH+Xf8m^pp?#zU<F_k=USATk|cLIp2N>s~|gYhc-p^tDqjFE0nTKtR_Kc1f0
zA+V1z1x>|-jWu}pHw}G`J!h<P--Q^1XVZhteOrpCgu~Z;kury)*26c-Y5&_D{<I@l
zhb2QO@E_BYUeq6p&**yHy!v>*x)U9C9Sy%8kQ^_(8!G&{zSB<1wpd!8`-fol+M~SM
zLYO0>oQXG*1lAb4*(z&IS|Z3uoGy1sFb&2ndksXV$)j27>@^1abg^_{JH5S^p4|gU
zSZNv`<_}1o4hwqm+f1kN9uLEB<rJh!9-b{-jxw7AM}=#_Ej^&b9wpiU^)?Q%vZ=L_
zH4k<BHVMGd-3_TY-BDv}Omnk={^jgs=^0Q#``nqDL{+UzM%&f<G@2+lG8avl{ZeG#
z%}c+(I^vwtR#UIy18-~ssNh71UK7b7b!l{yDBcW_3r^dDDnv$un^K+^Y|OzSXU}ga
z+h)v#$b2o6sd92Jo$B%w!?VbV=wU|23;oa7^mIo`b`o2&FbHvm1za6Yj(mDfi_+#J
zB@;E2KqzD%v*aS(o+HOe&!aRu5f?Yx%!wq+3VgRFnln5wr}Octn#QIiqa(zu(bfl|
z6n*=D8Hujv;D^<1Lk36sl3<2RrfjG9GCwdOydp(iYj#FIn2C~CXi^+Nu+Qw#9e-r_
zsjXk`rv_9qPw}lhgJ-0wQHt%DcGfrl_Povs(n3?r%gbt3oOyV0tb{1nOBAA>s6yak
zNOZNDukqtjFSnslp_2@JB+xLcIy-YrgSGa%g`TtA#-WDpu6P*b_6c&jcsjO>DKc~`
zBw?Ljpc6tSEqs3!3Y!6mLs!4WjYox+6qM7~zt?GrSkS;hsT2iPml&@pm&#}ofnOFx
zM1x@_k^YKLC4CYV)q3TQO7KIU9>KD7rYso9CaPJ32UUfGs<(!6+{GI{H))ZY^2`ou
z9D2o09W+A3xuU+5qW_Hv?eU=Sp7-Gd`!U(5%6DvH)t=~t`mT%mWKSb3cOvA-Ysqgl
zB=`wJgh+k;jF|K7PX;%C#`yU7yMu`bHf-^KzsW-}-9BWrc7vec3cvB^(p&2Ef&S@H
zw;T9Ra<$bndIEy`S@+STe}N7yL%}m<Jz<We!=T8}<A?tdWn^l)&s+zxaEitUrUq}0
z^o1qfoR(@h*UzaN@uo?je4ng2C!&`MB;~VuIkWZLVi6dYJAFW;%Y`K;b)G#xY^LI5
z7rdU$P@`JOvkTSf@(_QRH7yx+W3kU1(KHGiOJMBrY;K*0Y0q<>Phj-Ie;3hGOl<a*
zB5S-|ktINiE616RY%VOA%&Q!cb?eanF+E(GjRtz~%zYMGLkQjpuX|pFQfZY{+e7i7
zGp!Wu-YV4BM8F6XLlNE(_p#aI{q+DU@CR)b1p*}bDO_7$q)<Qa_>9)5!&abtwuN3A
z=`M(9aQ+Ff?da!4_o>#cCJYXWM+JAQr@c_$Yq~*eiE;ng$Ff%4^TiXZ944>6XPA9!
zwO)ikbD6B=11;d|gB;ZfH?YNn-vmle1k`~4ObN}$egHXO8cq`Dpfh*&xOgz7=|Zt6
zb@HntR&upxGX!y0BBX;jS@zxDJE6d1GR|0pX}D0d!aa9dmcFsKcUqZav4H@X7ApSo
zdza11aJ>M>YpCvP(T<+D>k9zG&~w2fvh|ie5p=!Pq>Fd+6xHZqp@9+huWi{lC$-o+
zd`gGi8M+@V-y1=#eQCxuMY>_ii%DXbiXUxCy!rDu^EtRvV|j98&uU9#?zZvI#X2=*
znp`Y#b5eBEQ$SemfRig5qh4D-=R3>KAhUP5mfHGi3|_TOd;yz>{geT)$A5(nIvo!e
z-_M`g<-570dy@a1p$iLSk8sIHyNBh@TR*pJ|IRbk&TcSD(l{;G2tf;fdR&7~<IT33
z!fu=3Erqu&`x5;g6ro!Rp&!id`pmXs84EtoJFkqZ1e$IXJqx?a*9qSA%<$LT7_?7~
ztQhX`dICZ+H}C<a>lXPwJ<eK4QG^V<=?GN&YWaT0tS5hh6MA}>Tcdik)z;t5|2z(S
ztMA|dFl3|FM>pgdutvAcpSt=(@(4fp@@C4$!M_yp5}9;FKE%qZhYpNbvjbm=zTtL$
zyK6(JfWT9B$eoVnVa*q5(em@>>_(Az09x#FCK%V$ENHgq9u<=a;LnpBFPqU&B>M-H
zb{$L9%PQ0}LGIHNWUZav(O*Z6{55{V@%~v#xGc2Buo}+mgYdAFyDe;x`11X0ZbAz#
z(v`OaT56K`I^51+HZ7=|F|FFU0G(w&tNmT=xSGji;`a9gVyNCl1Cq$=@=~{nV{u+H
zs}5qij<k>n)F(-ta@tT`6_)(!U8Gth7|YTTmsA;kGIYX4ivbEi46+T2OFn18%0VBS
z4atanP@`BVB-K`A%Zhdz?ey?PuB3QN|MDj3=c`-#>-t<NrQTH8smUs9oEOV=S}xeV
zdHCUeNjCoz<d12)pRdYxlp1|9bW2H!BpH+S-%(8j^B(Djc!#!c4Sa9@lD&*d1y2`2
z8tq#nNANpPE_%R+=P|*Rb3HeSTXxwO;9EP6pXG#6N*{do0GtF(*RwFS;-Y7D5c4@K
z4U?6vHT$@-zoNEa<Wj%zC+5W~!08ZfS+WkKQD-2tel-7RcD>2F8x_eU%|>F)x4Gzj
z-;0<=kcL~(YE(KknKRf|L+N%aas$m5J;MrKM>hlOshS@9>665|zerp#Qw8qQ+S=LW
zWM{8kg&tpBr%=OF@+2}Ha(|w5qu<?4Jl<Wp6f>RemL#41(a~t%-h3GAgbNa&y$C&H
z3`=D!@Zaz$?BT+|*d#Pw^lb~7-g(KDn{XxxGNTaKW2&A=*<*ZY^~ZP>od)>gRU+?9
zB}z;60I))?wrX1ePtu@0*~n$5>HUfLCfls;B^CpiWoEA)Yy?tGes+`e6rZfS7-M6K
z9VOVfJ0^aUA{&DYp_&&BcWjl7z#~?sOiV0$vS`fNlL580KSP<g$8+w0tXCTEWt0x6
z4}_`*^e3_az-@sjjB>enX%>_c-adK3VuP_mY(QZKjl5)FXyiqvItP(83$c~Ih_nPQ
z9WnZ{_D918CvXSXQV>f$>vP9>@y`7<gb)?jE_O3jpf$0P)(mIsPCqfAxn!wKT)D9;
zKbYoor0G;b8{F(Jyed--_DHriyY_9R5!wTC$R<W_U6%WDTlaU~Kf8SyQa^KBR{tq`
z)RprOD1@YE1U(8^dmZ*3<V{8J%pa3ErJ;FbN%@@gH0~20j@z0IbihdPl}M`^H|fTS
zV%$4_dnix*VKYf%TmG}UnsW1t-oj`}L;~y~rYN;`rwwQJa3Pa=n>$A&bH2sS&m9>{
zNBrM33iDnOV!8*jsg(;-3#G*KWM0&-eAjIS3I@2+TM~$9P73Z{fbrLrx-EP&&1uxl
z*WoMC<1s!um}qZ_-$ofYGJ)|lRhy=T8TZg?7!<=yys0ar?)^@{mL=^rE826Vf9RC_
z8T2L0KnnrWG4RAAK~`<nJH;AC&FOx%Zd#IK;lpu&twb_(l}esh>tPp8nA<8D_IOgz
zO?hg1|8_mmq;~&81RyBFC6AkIZqESW<$tMpf0==bH<dz(GPMRvK0&nxTiuc9(2&tI
z4T&MQ`&l=Sv#p^#u*@%boA+*m^I?Nr<a>Vbx#x?C^JJTM?J}sDPAq$0%_8HG-Y<#b
zE)s=hB#r~&@GdY9)YZs37GNLnLtkORc2|yA<4#H<@|<Aynr9!e514!qt5BAim6Tq0
zYa@manSA6dT=`s@bGI9U|E=NQ;h*?(a!M>LMu)EqIrcc%ARJj%b$bXbdLK9aKV^>q
z<=je`15`hQvWTNZcn?h-2}BjH_-HWzprs^-@QshJBiJ_rRv|CM#A6)e@%YV*T~z_0
ziNK=K2Ig5hRaCVoR2|P22fRVH$anSyjYkSg??%meeme1oeGwYzfA3}wFLs(sH(F>s
z{&Vs3bK~}5)OY-U?F*U#whtNX%di{QE=DwWccDJ83*)IBw&V#)s0w!#1_efQ!J`{g
zlWjap1f#6`;eFiN&P7M_*5bdX;*y_Iy-EOHsX|1*#|69}B#I`hmbR=}Ey@0+_iZav
z_D)1O{y1Ylat>|Oz6<gTw2>7<%tbi5)kYs9Z3g>~zOjZDLU(+>TZG*F#6^+<qpJyW
z&RJbxhNzDe0uu~S#m4JJg8PWaZ11eI(AefUnOgV?A6vEFNawnT_)U#zJaMM+r?+mP
zFYBS8Q1%3Xp_~Tfyk(Rt^=<GPKUpVc|DwHr2A-pwx2-cR4FWD>(m1lZ)AExpcqz{m
z<WqrFrQ8P-hyXm3IFj(+e4rnHyl!CK+K{69M7h>$ec4Da&<Sm(M{o}e%#DnE(LXRS
z-{eLO8kw*`(AfaFyW_P;eh0qBX|A^jW$3{`=*QK&u~kUn#Ofcb(1)F{f+yP7;^V4o
z6Dc`)+un{CcMj~GVXQQrYI&(g!xWf;gW2ZP6x%kCq37~t2b1{8N90JABXdBta`D}i
zffViJ*TJ(dZ|<cQVXg8kWabdcbuj76fo8V{1(YhO)Mt~%oNmF?TBF9WSH$J@xJlLF
z@$^oZ3G>e-;|`_Ed;A5ufx5^yZBbGMC<BOh@DcI@YIPocRr3X~Oww2}^;bt%y4!NG
z(I5f^iISE+#*<S`+w_HoIS)0=ZJBr2Ucl2>k9y#uS$hvG4?&#yG1M=ux@`p-B-!_%
zno??T0&+{-&~JW=`3i^nh2NTi8wa+$-h@4?IIfOFwVyee?9kif40ZSoUF@GNp8FQ;
z)B?vvU<Y){^sCFN9lBO{4?XkIeuMXM&S*zjez-#$+bn_ihy@d2S&<Q&u3pr%1%s!<
zGY<XMPe!hI*th4|rH$i)9Io1`QO!<bGPD8rTZGUH&^!UY@6?P-w^YieRnE$U3C2iS
z`N9Dy#uZESg9nilT0=D^m49`#Kfauar0w_BhY;45Wha9|iKh2#vEcBxxBZU(Djuj{
zKr^q=Jmg79s&!-P66Dl_S1}TUJ#|u8Nu|tX-C~`8*2-D+S~L5?f*fPPdrsZz_pvgs
zL+k&Tdds*dyEf`u6h&HTDTxu0Qjm~FK|w;1kOpa_yBj4XrI8Xr0U1h$E@=^v9AIdq
zL|VF@W3Kysp5OcR`rxJKoaf%hUhBWs{`R_-Le+4p{&SOx-Yi@Yp@P|ORaXj;ublr(
zTg}LTnO*PVikdf1ygO7w^qJ+|f|1yRw2p5KzUO16B#{S?lWK4C`E0o{tcPAW8Tu^I
zmaqER{a`i(JKYIO9Q8+U-#nl9Yc1pQUC{7dAZFmq?1PIw7aTp%Yf#72+}v!n?zj6#
zor;?Ja9LKe#wTs{2x(fz<+L^T*bD*?;%pQI*9j&)+l&zNeG4mzJsZ$P4Mt2fJ;TR*
zZHo7XJPD#8t4%%nVAqGf`a|Wp48r}dbr|@C=Zo`r)oLrY8kON)_u5R7^DN0?T32lm
zdXs+vCW{#JepeA3jKBwhj-0q_83p7Z%hi~Ow{nk9%e-YPt2E1CZp3-|+gLXE;HXV$
zp#K7RrGQM1?F!D1T_se13*Ie!ot_H0MPqP??{L%&Rxy-tbaxjGUUD6^UiCh<j07K|
z+C8Qg&RUdYxSkljLgL1^I?IO#YbPy};hH*DmYp9v;U!h|RVsL`YD>qj_TB=R_^bC5
zqSwgck*hCO4)O+I_x>pbft{JVeTkkrH7+BmwU;vy1mev}3*Zs2r*ux(PxY_A4Xfn$
z%NZ3a+M^a#>6fxI$?DmOa0x7v<D+jX(W5Qs0tcYY<RCR-?4{Y3rSyV)8?_IpI4O_b
z1;|(=DI3_Wbe3HW4B^7>j4N4l^vfN1TuRT?(YDjnfPbl}w&Qrm&rYDyi+FW#$u$)B
z7D?nM3wAb8tcm{B2<Wve%I9`_?NM74vKfe|vXI3e{Ay-q6jwGuU0VuM0NgKVH?Q5i
z!962nf3ez}r9**J3Cv1X<2wXh_qyc#7gB#DIHo1qGVi_Heb1Z{k{2<G>>k#JZWSfR
zt$TPOs-JQmKH+<jXhnb*{QeW`!>CA8q$qN5i3GpN?<D$sYtK*eXgNo>%02+w{W&(K
z1!?8&;3!I>E7iyAwH_y19fxNd=e|wU&e4)*4<vW)OFH$O?yq4wRt7)2P43-2YirCJ
zT-n6lId3;!ITvNNZDr8--CeMqG?S*IOs^M-rh!F`F_fH8UaV5{b9OSkF}&i}mQoHm
zqSK2vTW+`CgM1{0iH?*nE-EG>WK&F1XaAS|pVXyVSa(Jb3GhP5n2w$@%~!LRu>&TS
z@>5p0u|F^7>irK<uS9vGGw+fU%fB&+ZT@t#%*S%LFIb!}6>8T1dSb7J@N;o^A7y)A
z_*&b5E>BVof(2(vAjy1MIj+-2?TD)*cS_)a!5@3GqRUNNTSPZS_(QqE#9J&HFKf%`
zPro}HwwX$gxKaGpRyZCbBG885Qh=Qht#-Y#*RE)`b|?+i{YYuH8oZG{^qq=bz5Rs8
z0jxRsQ$z+)@J$QZ+?=WLy<A#A(;Db__fG31qnsdDNO5K{Ry>8X27-b420m7iOAi0*
zS&v-pAK|7gcwAvzGA2+$pJc?>Dr&x6OxIJXtLk7qjXSb?^6r<3X1jxK0mG$k$7Na)
zMxommqTiVvU-{*8bRvKt-=U^Q>xh6CmGbkkuwbHDQT{B`rk@ypPk99K+2$c7+m5e`
zEz|WmkOl~cd6e3I>m;p;xm2`66B-#|sQ0ma0vx5iuFO|u$RZFbwzNYvS4y*56`3D$
zaDk-Old=mu+v2iC>MN-^;|^K&3KtgU6txGP&9OsDWA9E~7EriZt2HOvchC8tea|l#
z4hH-68w&=fg???Vt+?dm<dxnN*e%g}`+er`46pUs)-ZCc?DF_U{94h>&GW6B<|Za@
zu;baAmRYOqkjn91yO%s$mU^>|kVDnw?FEzbi4w2-wqt^H%N>B^;3)x(4~_&$WFAdV
ztGh#fZ+-$5-V-u<+NW44>}pd$_RyZ<sHcKCQ-hbs_Z=|B3!H!h(q)G2^q-`)_;id2
zRXm(tt4jKrf%>4F`{ojHo5yZon0qnY!Q_RB7~L#{IQ`jULUB8vT=HI(PXLtj`SYTA
z4@OCQ?|6BgrTt6;u*k!`ZaWF3Zyj5X*E}v-)%f#+JH5t|FP>;8X#d>{xi<j5B?+B!
zmwAi*OWorV%^3<-nWaN{xO8U^2!6R!{4J`3uYXk{FF&=eYGS7PE!Har%_raao!&ma
z+(lX0L5u>O+w(dUcc0a)#}vxkP?@ZG2c>fxxl}r(;&Eh|-ll)NKh?z*lz&fpr4v=f
zZ<8c1`#D$JnYdOXeP<_`+^{h<ZMCg$GITfZiR%5X$3d_ToAegHNgStD8g6ZAv)4s5
z)ijslzCzHg__eGLE*0N-)W=eoe6>_rUDNIlCOxEkfd?kDj`jhpYkGb<$KmuAH=2>%
zKAU3Uu#J{A<!TbA-q~`SdRgp+==GbQ@>9W@;oHC%WF=o^{l&&k7#i5r-;eNfwM>WJ
zcK*TH`Hm}FxUpake0LG2m}H|S<<kh=7bO>(LE-*)>`e4KEj|~BmN;UMm3xD&E6!EP
z-^4VYVk%u!G7nmBk4{YdHaP3{dz!B{H(F}!eb{X<rB9@#um7YjW%iR$7$7wvB&65z
z<>4^TOmGjbMliouI;p}fAdpAs5Z-llNx_C<u-#Q`J-yShbRut}|FF4MwvzKUYF@>1
zWkMih%9nA_6^`Fzfitm0$K5v$-D4CMc#Ko7>b3OMxzI<a8f|7vOY0Undw3KL@0k7|
zi+c>)7f*c6J5yHgRmTK>#5T2gH^2oYdeLK#KKQiZ0*0&g^^TZW(&re0M&>PZVIG&Z
z%(2yF!J36EnmttQ5gsYAkuXm#>h7Deey02N!fhJm2}QLdzply&+3wKpB`)8OeH>uK
z_p|@!?%R*HW8U*d2^~D~5WQ$VS+vzj%tAO?i}i@+og}VrZT+;diBum4aqBWEP~Kf_
zL9LJYT-tsg=89<i(Boi8-?EM;y}6aX;#A6%y6{Vl8_qt#Yu93ARANDE$;`GM<!O2C
zXFU@cdbh*D>-;6vaqy$s#d8p*(-<yY$LVmrZp(s$U~}p=AOpF0ryhWbpOoZa_6~<X
zdcXu6>_eS$&9C`&8#7EGE0tSuH*5}E!IOP|vgXz9@%PbNVO$cDyn{XO$iBvJ-`#nB
zoHbJzG%7Qb6K*Y^T?8yYz&&Q#+c6-wK)$2oFbt1v^sn}Z<W8*I2KJh+Sf>TsQR+^=
zkN(5WI*r>AQ({g|PXCFue&-wKJ{K6h{_;wk%#!B39(PmAs(&$MkyU0Z;=S@S`0i0S
zK2M^^Uk1e)Psjm?e93#e(Dc{eATlBZ(cp-LLE%I7M?1{6@kTPD(U@}wq-elp(f9up
zIXg2-8W_R_*KK9rL89&I2!fc^?<VCq;4#2gol%aH6s>Re+9R@eta!WH42LOj6-qxR
zGf~E3nKT}%-$r(V4r1hhNwYk?*ph$sGddMA1d9X?&X`+V3iN)%sMW$Pv1pX7>MHFw
z{>Ep&J<p6FnF7?+s~^vgEG((Gd)O%P?<Es7nsqyyrwLIL@_mS^GfQ(TWO=!E^szXG
zC^lMeTCk&p9{Ue5_;6?I@1sb|{e(}nFZ{jO${?4xir?^$;q?tz){{6GoCj|)CWeI|
zbg3K5*sj7=FF@ykqJk<651+;I+vr<l@p7_(m~Kk2G9XJpgh%r%#LnZ;tQtMlpls{p
z%q6WVaM&4ZIMC13c>Is$eD#o$qQxwC(d*tHp=<tlA#*27Hye}se^%VNZ|7z!BzL;j
zx<6nA@5)0xUac=2u#ZxZpY4^{3WKt+C{+7Z050cyf6P`6vkUsJTIa5}y@9SJ)6;ti
zaJSi8z?Z~<KdJimREd{eh*-j5gPpqtd%f-P2JLLhTaj6RIp*tT=f4c18v`(>wT(pt
zGuzvI)B8Ttm8a*hh<vfTRoc+Iyu7?de`M2h=;&2Dsk5>SRy$-i@+TgXWlvQk-#Lh2
zff2(%TnDTdwwt8efZW1;S3$L{nU>>U@>;B!->g-A(A#HqBO!*Hj|%I&HZEKWyS9^d
zmiOMU8lpq_-tVXr(>}G}FxHlIVKHtJHun?VFn<Jyox;^d_C_4CQrp}Tx}+D&I;Ec9
zJMHRH=O0{<4!M%=Oe|8<t2dv5{zg7lu$>a6z``E1+}9?Il$9a_8poRJ4Htnk5#7U0
zYGeI+(^~Sj%}Nh}Zo>zZ4egeugxwDL&O`+aEe2#y4>H`)3Y4!`sr)?hS|Osb`S%gS
zu=nFG(7-lQ>1KTJ!t2Zp3^5@a&wl+$5k%Ks$@Z7yqG4hWYn8@<xA?G|_=in8APrM~
z_Sd?%zTBY-UW*O8LL+}nDc-UB#TncS5fEzvXFYc!?Bd%}!`D3DTq4U@B_q9pDlA+8
zgb=<c;_p2-q^3nv)+)8IRuVDL`ilHbV8{XIrCLpcqxIXdx<!ud5yzY;=LfxwUjvY}
zwoV50ZjL$SX0=`XZ60)$+Do~|&UxDoJPEZlQPTOwo3pHs`8!x|Bp%MXz1x{EV!(-3
zd(bpxG$(pG&}db~jbnikoZVdKvtqi|RA$RB{BN>n*3iJY3fUpAhs~Tg<|=TpJHXCw
zA^6>1M4p9fXK0usZw+``t*;z#oTaCkDjF(&NdcTR@QY4#HgE_Pqiefl=K10Q_wXYD
z2B1|jnCzO@TSbNHd!)C(WO+z2xCnSf&d>nlQ>4i*PYDS5Gc_46$#V^<t!MT++9TdB
zxSwowsZ+GfZhWATt<*!R)qrw>LHc`S$m*2mb6i=QqH++V`jmWCRh3vIPl=5TLH3XM
z!{g@6{!OvYHfcDS7T?hn>&qDz(6T$SnJTsZU4to4oG%;WJsJ7LbT6jrEbi(-xs9I(
zsx|4E9i%-5D~Xn4qeI-M11>^5;dlD|Ls1EuiGHk`(e9`R%PU`cG;d;998otk?Cp5<
zab>5F$UYgQ7u7^C5XVK~K7F&WR*<{o0o?0uV(@}H5NIUz?N;^b@K)qm-n&Mct;GX!
z*GKkWYKpW40U;q~B2^?4t!H?k*!XeZ=ZDaclQF0{Ty&=QD9y49KxY6Wh&J`HLb)fV
z^(jkPs}PSJi^4_Ho<&&v)4kZrCq!>;0w<uaZ$(D@<ExO%RAZx*GD#K<RpVxM%ZIxa
zU{FH7jwI^Sk?UPMe@u;|S4PM4lh)^d%UFN8q-UP3WU^6t-TMA_C-eMdO{@nKRYdxG
zq2>`kQ`v;_(_oWc>u;T{m*^Rk$YBuBehp8qvn^;D6=W&w?d<1}#9rV+Y;qldi*UpF
z2WNHDY7*^6N()WO^caWgvay9ui*Tf2`P2|QLiz;#IZ0R|(TtDUI5+#KR5;}x1X{vM
z>bRRS(u=oFRNj*^F#d+wxG+r*^!hN|TN2*xG+JaHXO0PB8MVhFs6S?CCI;ddQ2rHz
z*(**zy;DOpz2Z~MHMXxFcu3`A_)_t$K{;{l+l4A?{!{BIpkvbQOUW@?+%_BxewW;s
z9Eu*^KWOxWJqre~l)IUeGbc~X(S7Cs_Pb?u!j3epOfk_jSH;Ll{Ps4_OIT9ijCnxU
z@#N}dnZ#9E1CjdPw<;luT%E_H6r)1%HMXlS1#H0OQyK=>=3aVq;~m@G#I-q39)Y&S
zR~(NNDsL$6E}o~-4oAHW4KelVWYH8-_Gj_E*tz!d7B(AJuOvk_rmC1lgEHX4@A^xp
zc<}V;#cxCr(KqR`2Y3(0-x5GP;UiPYz^{V<V`<$<H4oeTUQu($Hzp~T=$WLmc0<8&
z5py&KtGduX+fC}9Vye7#0Wg;5`c9bWv*qZiUgk!td)6V4?zt3pZ!4y?%@GFit<H>#
zagZ8T1?f%~os+BmX&#5oX=RO_XY=yUt^~HQ5=9Z(Pe@uKHdcC?XRx*#nlS})g?(5(
z4a7slcJw7xSM@JqG?cJwGcYcKRC7kvEkC8=&8buA>7Mf@>_>ij&Lea%-bS#I+FkxR
zd8UgK`vJE)leCrK;~^4P3E;HeJSGM+iX#Rq$o;a9lka%dFn$Z6K(uQn{H(a*LB#;#
zCaI(Qzwy!1)odC;PGZW*?Iddb$we2>7uE@=fK`v*<ji}ddYo|nT!6!)fw0D6;nI&<
z??YT1kR$%dOf#t-t}d}plhiEy`-xKTEpq#1la`Zn1B!&p#{}FxvwNSYelAJ_gX+-y
z=7irQ)y_-N+JO=LE;4v2?-$3OIDXcA{RQ_iK54X+)T3WEeB$SgZJtRrMl1vuQL9t;
z&WM>9sk%Cj37LG3{#6SDyOLx48j=Box$t%0!+G&4JFiT6H7@GP;Y!G7xKp`VRl^dN
z0?Ps=LB_d|b%uHQd}(#w!$Z=w4tw1Ohu7~@;M^F|<=cwkw%571>E^rSwm#(Hjo$j_
zovx|Lh9LYfz04E_*EjJmR}QK*R#dC0vLjmF<*{-NQIM>A<?dRI^MF|m@uQH_KD%$#
zR$@Jxf+OUYSwNDN?}&D@dd+4as<CKA%}LLwwA~Q&xV?-Qg0}GX0L20_?v1WDDL;;@
z8^8iH@h$JZbYF{a>)LZ&WL)GGj}GT3r&SyHxl0(5Zr{}aV4yp{f-0!~hFHgL+lr<j
zmPB5)iM(BGZ8XY=7fc2zu=;G$>0UuQWvUxJe{FX}kg`Qq$}2RZaDDrFM^*h%6}$m-
z)Ix`Us>uv$l|g^UND_9jK*Y=AJfL}v=!a<6NSQ)K<=>BKf=NFZAae!b=4LgrZm1gI
zWDkZ5s&tq>B-68Q6^vDJldneV!)<jdA<(%JywwBQ>sPx}in<-FjB%C3SxJQ)1*XRT
zh_b4~R8tv3_fOBU!*+02hqOt~Lzu<s!h2z0tucZV3q&cTK}u!p+uV(y)W;I8-7x2=
zx6N%xJa{Zf^?T=T5%2rx`{~ho<i%r4i_4@N6Bji&7~te}bLYO7YX6Mu(7Vjq@r55|
zA&>aLKKOa4l19o%JC1a>(F(#v1jKOJmrg8B9uw5J;)Nsxx5G_}5xIh(HQeZXqGfkD
z9m6>BehFyMc+3^}(FU!!0XtYoy!&yAHzPY&3wt0kkugJF`#dzjYdxx=crP}(Urh<<
zFhH4}>+9ZP3WGV~{-v!o!4=r2u(+7U!D6mbSt--Thb}w2JB^H@n7TuEdyV4}7x98{
zgOzf<XkD&O-7Rg>CV;96>p7-XHLQ|*IZr!bUG>_hJ^<C0F_o@re)@zST?yszirT68
zS1rV6KaY{zW-EVZ^3j=}5VO1dhia8Oy;Lz?c8}y5;aUUD9cYlaYxjVTm`|sw;j|}B
zsoLh(aw^5#L$$%YWFqFUi#4xRH)xmhf<7K^lzWk1^51{9V8`(`_j#s-kA5UO!MmlC
z!=^(^YRpyyp++G=-%oT3i1Yzd0%s9-i+3Q8xzE6b1%op)CSLKA8{VT_$J9?;@xgH?
z1OgKhQ()0s^FHJtW1@<K!vM4+kI)jp2MKZ4d?c36j=BP<IGn|2{bjfirBl$`5CnAA
zwKBwjfOQ6p$1Ggsx*B&XUCi)|sxl<)C}VChPN`<)YrQjN`JitFW;~?LIy(T<W+^0&
zgkNy*5-)55J86>{!*Z_2Is-;ZxyeLZ3FWt7Ie0GqpuUj6bexCN1<255UFK%sc&6iD
z0dB<ULOJMqW4wRS#cOTSQK#dZo@41R0hZ1kF__q}3PEuOr;}wx#p)<5DB$n^27IW>
zi^Z#>(5xZgTd$>bj@4*rpp+QbCD^YZd+LtPz%hA|WhF8iB@IUZ*8)Z2Rsh~2$;P)l
zPuzvlEzwM5@ig*}_1(yidfH3qWgcebr4pH|exrDw?TqVRk(e35^)w4ZlHnzNEO~(>
z8ga~Xj9uHf1Is!anh^|>km`K|&Mh!i6Hn4A^4ja)YhfuR2~U$`d*wMRY7PV|835D_
z(@P^OgiShG6I+!=8<}0D{R21FePVdjD1nqsoZspl!vQ;ixNQ!i{_@=EyNT`A%u$9S
z-djT=8&d|O`g<0uRN;pA<F?E<AkY6%ha|-!)E{AAIy9RR#szR`52{y1&y=0gtRlPg
zGtH7w7wHd_UJ&GIdyu`3(#)Axw`V4rTy<Qs3re&#@Vi*x$d0$c(D+1AkL%FSJ^JsB
zR|i+lY{5nu6M%Fat|?Z?cYjWVrhM;+X)0kTE47NuwCPxQn;UGfB5GikpSLcumtph)
z^7Nah9kSq9R?T-`Nu1|1cmZWy&1K>PFH&Zm-;5d2_mouYzF4Md7$uQsiDQooyZi-S
zUT#<TR5vTv8JT|->enFcUg=q`xVsavwV+LI_}zxRIVq72*AxdQI);lSlqKbhL64r5
zg_<Z^8Ph9a#*akE)1!`f|JaR38;ufR^NBxwa|w`r)s^M1Ljo{E=erU$c;~f>lizq!
z{sNY)tW@v?G^f`WNPbcKS2jU(4lm?=UETQH@|R4E&Khmqw+X;RAiVTKF|DoOJF6Go
zgYpqJ#zbMkRIGe+o{_xNJVuXNePOxc>ubmH6Xpdtk$tBAQ(v)nuT3q~vO}tz9>5<5
z$J{V8VVFC<3b+6RJ7)130&oDE2O{NEcZeU{?`MLp9Rz#CH~t9t-vHnueoO#3KZqI*
zks)bDquBW5df3m^aLH~5t&JNVwiO<xFS4?=KNMH?bA?#*cdALC3Lh7y3QpIlEx^9+
zz4j<2_$b^|S(!{&EOie4ju12d{x9ttMSnjxz&KhqU*M67-FweUi9hGg2>#t*A-83C
zKc+qExV^iq6|Ne7hUAAx+Z_4+I~&5<C|h_d;Tt?0wdxqJPtc4q2&mKDOA-wpFbs^7
z<44R_G4}g^v*D}PTmy4a7ai?JVQUo2LmAuIw*I%y6c(x@3bKZ@R>w)997&PHIjG_W
zGZ2zhDUUu}YU$cIpkCd$d~|V>>gr|0vtkrH@Y9hZ+pW=RKI1X=8QzSD?bcjcn8(b}
zCde6`YAlJ~$30T*CDt<ezIyHO4t&jqUDVNpX#@;CESXoqN()=40-rqru;`BiyqaG)
zfM_V|9~cv*Spu9eGOzs=hn#dN4Ce8oY~<DK()sKtLbz8A*CfFE`+hp+FO9QO+Hsx}
zx()?>8jz=1+?4xMQ+x{A{Jb$13qHvUG&<7BC9huQseRuQ)jYp+>d|n!SdI7Z<c{~g
zqDhrGUpF2J74R)n@wr#~cGxRjU`$(~r7C}mdsBF>66Ml8`@<}$t>960&U*s9s?XHi
zpR}M%ZR9mO8=Q{7tdpg<OoYjvxc7zQ3ZSUU9Nv02TPov<l>Sr%KgrEu#b*I@6(Dzf
zIhM@hLQWW!)LNKtl9I*9=%RBcfF_e{=-X+k4LfY~?KOkvKgA{U0xVYJP7<ezkM{;6
zxRqot`P3$f${Adkx@e^Ybg2}?H9*L5uW1FixnL#c`5Ops*xTm7ZG^<v@k8PC23;P&
zzU?!x+oti`fjkew2z&^O|FqxBAew+#x!~t(nO@(93krpwrz0G-_=YInmny!A$mPjd
zxq6pt<VKea?A`D!IU}Gl{7Z0;N{>)V6z(lpx9(*`IyzQ^<MmI#PsV&C6^0A!EU>#~
zS=duMHH43MB3L}{9xr<KQx|w@1o@chRqxz^HExG^Y?{8eskvYA&GBa^P~o?tR_Kk#
zI=N-Rn1=}2*8*KGm#aT6VnHXfk5%>PZ`zRZUFD~I->jCrA<8eqv#z1n=$jpAG>Q#3
zRTuj&9VSev=*UXFx<|P<YGp#ws9uGX)!W7Na`$YLYd_-KS2XKt3&hy$6j3B?S-%Ok
ze4%(~OqdY$RHGaRXSnP-uD$J}3k}9>vMdvg-v0o_?qNkJz{y}>gquo`a*s3*&XsO4
zXOt&Y(AR;BxIXE40Ge;d6V-3dbU2dFR`)aYW@CLWnHVu&{Mh8Fy5Evd&TN$3ggO}H
z4pHS)f?YCXpTgZY{~3aPt2agL0>DbQuE&Byg{6dN)o6wtLvay;qL6!c@ZCr;W!pJp
zOW?wXJT<&I>`V^cT+hDV{sDTdiQJ-Pt^yy+{SV{y!s;u_Vm=k^;0RXU1U3X#gj6HG
zYoxpxk;<#7P|6?|e5tlTKWcUB#K_0X$*rN_ugF<!EXBr$9tV5&1&!N-*SEK3(#%*U
zRJZ{*wf@ErxDM&h#ZH-^A$7<&_N*1@YEjTUm34PE2(XaL8yYy+NkTj5MuTSsKMSmT
zP+gF^0#VmdA@?GWU7fHc5BQHN7(3Cy?pM7X(&*6Fla4@mbQAkE@G*tC2TCEAl&Z#+
z$Nv723k-Sfb#}2+Dft2;&0%k!AQpcw#_O{8&oN%v`1jhouOAk7DJilSL}`%pc=w4F
z%4=4YB9XHl7`S%24>B;O*#-8ds_^|_cSxWZ6qy5-cH;y^eDl_0ODUEFwQ+C7lB@bI
z*cp~9=<$L^iS2ALyq}2m0#{6ENTVgaJkmix_7+o=U^FVvaWc%2=B1pnRS&#N1CU)A
zJ5%^QjnR7LewdgHzCykNwX{xznf+%Op(`fC!KJEBq$r!0OmH7f?o+d1t#Js^oJSQ_
z>+aFG>zQ3Uof%?y!78{W)BNuDD18SP=XKG>Ik^rELk|V4*>-p(?a+Igpz@0w<m4R3
zx*ocRLXJB;>>iw12b4Xwyi>9e_I)ym&ip_lV#h%fO*4M-(<s>t7v~5^ozF!W2+$!{
znVF#Og60qYQIxIfYffipx^9P3J<<*r_*$*GLKi$`JQ?3#=oU!uM9auDIn(1_pP>7C
zL7_<o)p%jxGmN6sJ^wDxK!uJH03GrQ<#6=zL_^+J{5TxTfYiAe#Hto@^ADOK7!1K;
zK6w%XOFsn2a|dAfBPR@_LHV22yfF<6`<Qw1LPHVS2X83Jj#|)FW#DGqd?<8(#*Wwk
z=}hzN$vyfIIjK)WBfMN8(m8h^{eb+VSZxEpL#+Qhu6m<0LSL&*AXYrlzwIUj%6yKU
z2#K3o9ChJ&ls`83H%$5deD?&KbXZJF66$yR{r1*(zKt=Bnfl{!_(XpR&<DnK+Kbe!
z3UBZOF{-)7XP^!&J_L$jk^)B@Wa8yNf^mqL$jibHJ{YmDMJV?;W8K{R3Fhl@9<SJ<
z=RGld+|v2?U~6#W|Dz~iGu`-HAPl;oaKf^F8;wV0-sTE}jnA_3E${$_SCLE1SE1Af
zV9*{^vlPoavl^3+0c!+i`9=K#3|xJe@Ni%F)BX1Ul*6}T@n4Gw5Xfo&8y82m#H#E#
zpnjg1uk^yk_<iVkuK(G;6+eAO0V;v17<F!u&I6=flyxl+Of^HbZ02YYv2W%VFfQQf
zsPX%oiGWW9TqykY@4?MvaLY3(BRvacEF{q$@zMv!8TNjy1pVnF{5~;{6AK6mR!ZzP
z6L#JEcB*t&AkIjG>ORZ<jm%7+wir}X0$wHT?M3kT;81h{>dj>nWvB`X{6@)rAR_PJ
zwj2a4O`>*g)5?K8Q2}NmGJ$;Wt4E1h=WIi?(KE0o%XctQ;@i9PhnYP9XJByPMBadn
z_Eq1jK#1aK$~~#1hqV*b%w$wIcsF5PL=2OL71=6s!NqB!Coku`$Y_Tsg6yR}#$3hW
z*-@wL+kNZG&SW^wGplb}VLZyOPS{cw`WAtBku}1rIjROY1h`Jsn%ES))c4|Tlt4tD
zrJ)u{_`=M^+F${LprDr1%P}yu{5!P$eD{J~fLv2wn~y_&SAQ%;Y}QX8M3sgS40~%b
zAQXm+hiorV|B-=urxbP&o`-9GkZYXu=s%?pHn4Q^=86ObU(3!=lcdHJEZiOS1h^#<
zJU1weDH0kS9^irlw7oaWr6y34n6*sU8OGI1wQVc_1ntr#>ezKX+|)xlJLpn(1YHWk
zfBDDqE=mbbaSP`jf5{Albrv0SYgcz~(#UT&TzWj^SJJ7={H)pr-25*tJwo;BmZCgh
zr2x?q3}rTNl7F^AJ%gA>^%LdoxBp;aW6fvJx45g9i*&3Wcr0KI|Cp%A+<u#oL^ER%
z#`lNIX{b=g`_`tHG(@ie)F#8v<WNyy@0k0P_d2W~%oA3pi5L+$mi0Kpqg9DzV(t>k
zN4gRl98i%18u+N*a+*A;5I7&H@8|Wp`GwFp^#Zr$AM1?mqjQJzf9cN+0Oo*|!;Tw!
z7M1(xj|U!TcD5D~0wJ*<sSbt<P1F+fyTwLc?pF-?yi|ybh=us87^skG5Mn(hJ?N&o
zt&xhMtOeN7d9c}4m*-^-{1NRf#Y4Zz9`f+G+_O#!*(Xt6rKMl@xb{-y%oS-{U0Y@k
zT^NupCGB0*vazI}=9Q@(8B$gX$ur|kiFEX=?~xMgOz7e9b1D69yu^tylA^i{j3Trp
zKgQxQh&8)7q>w<boihL?!h39VjGVzeMoNbDD`{M^ruZ#8kFAxdfT09|kuN?xCSq6Y
zG?liGXUJkz@U2NGD{H>~&$Oo<C0OL{MyOwhc3sqJ>DZPr&T4p=zn$MN0qgP9$ZxD1
zPld3FoV+!t7y?|1gy=I&Flh|$;Y<q1z`yb7VqYcj2r8@}iJqK8<-{)XEmv4F(ru8g
zDdfqZ+gZ=@|DNm}?2|PydMH@SmNy?4-gz>0A8L5IXFM-@g*zBMOo91Ri5E1yjpW{J
zEfQ+^v#(>&qQ=8$lwF{#7#w!R6k6u`xgt0dNRXhn$35V|MKsd>>$v;X!C9=Q)(mf%
zMlf-nS%C$)de<Fx5!LF+(g(LSXo+%LoQcx5tethT1k~U6paRUatzC&6`<Z~RG&neU
z4VDfj+7Hg(XN!8Qkm{D=n5zsC(#CQmT{4&KJoDBPPk$Pz=VsH;L;<OeDe40e`ISiX
zD~_L1&kt<B2{bt5Kh=ZrJ@7A`JW~YAFcL=5>1}6{j@^6Kc$1F(+@H-mCr$J71_T($
zm{XcflD%2pEP23!1+`1|Z0{D_`6&o7E;r<5;RJy2eA1hyH%+cWc6uvjwVn;*kngGl
zxlH7eA%<r~?$z<Mbp^72Kxwn7%tmIm{knbGg`6?p6BX>C#=CM5J6dRfa{cB(^*927
zU$0em6;@kV2Z5M^<O*N^g)RFxDmtGBct=%3o$?134m48}7iAlj7E~d^LZ=j&1pZwP
z4t#qiPqU^3;K1s=_MOddC*wuC*UdW4TM6RzG#?;E#YQW|`(f>?!8tuQPhu_H3Uf`P
zuz{NCV8>KLh*G<t%Tn67I(xe?TF<OL_mVHq{B)hwo5tfN3a7SK*QkyBj8U3rvdP}>
zsaX{<&u?fU3x`CmC{2Ff6$4STS~ErlA;rA@`BUddx!jzcHBNzo`hQW?MjxbE$%c_*
z3LN;Qa1!k((s2#1O=s!*rdL;#sY4RH_DjWPyyh)?lZ-_Bbd6QKF3TIk+OL9Ysi}|o
znIHsNuwb3J-;gcSS29+t%Yz;V;9ZKvc~YxQBSK<YSu6l?RhPZu##(M|GD=!c-~~5c
zx2wIL=f&+^ucoo&iOlm=>`T0XgFP3@y;-nPT`bPnCzpzfGEJUVy9PaTe=2FwX=gw6
z5ljxw{B0=}f=+D@7#m(}12zZh3nXR<D|os}T4(zy%wJ+axRBBO2#5C4<J)q363(^3
zkzC<#yc8~|w=pph)EW)yqUS+)hcKgYLxF9CWXCSKEK+S8_!}QIl^qA!Ovgs%%=?WG
zptH|bF@lZ>r$DKbBcu6ekZXPUhX6{!5>mqry@+y8nl{|!*dUh@PjD;|s+Yb*O$P&K
ze^%#TslXQFr+>m9S6G1yNYnSEQk;02U5L!>LLAdIf>4-#vbe6r$uwb*zqL?7D^*tm
zn*w4HtPmwVgyRr(A}tgaC`%!z2%vw`H^EN!(a*%`Oh|?pWv{-}E53cC&+X31z|gS^
z(PkB(vKV?~>>u9OyLQquOXpyMcwA<8Ve-F8ZTDaC@5-iOrMEDAL&1+l0ozSszTcQS
zqE^(}GktBO;W(q};e!;-n8?H@VqDP903DC_Q+7Cc;e5w1yuGIKgwJ<$7c?$FJqE7>
z#PS4x{h-MJnXmW{7<_f8>shAV<H>f)9%5z+YYikzoCl+AG^%^4!6Bj8VB&Y%Li?Wg
zG-yOnoV>^?8<MhI5u=l8r+y#Me;s$HOO<-)=M>eNwho3l3Pn~ba&JaOlM#$t)5Y*1
zn>>(`77#8U`8WjL1IznE8VDW&KxK&rViM5Zyq(fCq%udaQm!%NQ-WQ59H<FGLAic}
z6AOaTRbL8Sg>nlkL6_lu$#<7;U5xW*Ke+#K%ssSO&0Tc6uT1Y*u|RcAJ`?^$pl+bh
z!sKJK-dqh9Wf!rTWC0W+vHE4_c$9(?Z~iv}oqsEd($z((s){Brm`2F^FBG~ApKmka
z`{I5eoKmVY-S}|BT~(+SoU6gleLB+EgEqXNx0C2V*|GN)3tDa7D%CJ6fuae>=t<aH
z#KinH?@81a0pJN%OPPmEjB$s>)A~$^2%0<venec@6-ai#(n*x-?Uj%o>rk`ss+OlT
z=AF^OlSY>|!wnJ)mcSb5hr!e&9|^~@<_+=kRy$v(3`lqQG|n5pKHECpvj-&@)(RHt
zmK`xF&zF3%Po^j89#1+}>8=BB9W<Qxgo}w!PtJbdJy`CSfg>Pj-(1qbVuW2L4x8l#
zHw=delRF`(S#azWZ)bwEvqIVD<derghCV5l=1o+Umq65Nz0votW24C0dUqJnXsEu|
z-;kb0nF8Z5sne>glm^X=kY7+6G_z!aBPSJ#vqp3}T1DYAGPO1EOn2txg6oopT5bDk
zcterR9L$ScM-DD9faj(z9$(!<xk0cc5*dc485P15U_5<LFFJw|+4yhAo1Vm;LP}KB
z5BG*?;RLw<m(#ZLH+Wv(pQJ6+{J<kQn(G6>I3A1RKsKP;B_Uro{(O7wG8Aj(*xdU-
zfTic4dA9A5I`fTKa98F80zo+Col<xOdB<DSq}eHZdpoO%Eh_5FjEVVo*dPkCHQ0E3
zF-_bm)*rFy6B2yTz$y)%JGPn!N_k){DtPTwIs~%&)bO*0!LLGU8CqYUoTP0>7XVqt
zvG=#HlVG1wXoxaVNN&yd@(6*=lfN?|O9bxj&<(d(_=CG=S1U?VXSol3xfnCDCq_E_
zuh5IT=TRj+;sw)nk{F_HqR*_j2nQVtZtJ7+SfHkJZunW%W@C@-P_71FhqYMeX}K^~
zE?=5YdzJ>?O7riE_CcMGcIC!Upjhp`nToC#_^f41=MMh6AqmN?{QX6-7M$2v#WzJQ
zHAvaqqJCZm-M*GG28ix9G@lqqy=m4Dw#ROP{|?vluoH6BYXIL)eu*1n=Bx^BXbI&Z
z5k~%JvOxvi*%2fY#0O(R=f|eZ3~gg5nt>({-V})Xi9zbz{G|>+F*hZ|oF^lDqii9Y
zqN<_zy-cpU$jQC6GH{H~Wqp~%&%q2XkJgs0JUwhMky#8`XPgJH1fX7clFSUmSksk-
z+O+)kJrH@P<BO({wp$ZM>F*1b6uZ~7PzWdyyCMGsJauTdu+&=^k|*!pg89v(LoLc4
z4wHfqSgv2FZUC861r3BJ=+m|B2G_N8x+%GPK3W}2MU=)gF2l>ewq1+S3c!$Q1AMTY
z2qA07nz8Q=dR!eXj^gc3FP`wl>h-(zqBM7sY=K{ge0?oPebnlURa!OI(1vOwZ?prT
z71`RboWM91arph$+SoKuND$r*34UM2$c0X@IZm=fXBCP*h>}Dv<HXs|A2#a#-*cb&
zGS~!6v@uXdiog_tq^%CkodManOM)U8?lGj{pt2Ga99HYDmuRZPlEM=&>V49A4~sA(
z9Bp%2w573x#B<(f5A4?@yFFFS5e&^cC%PCT78!OX8;z0%d&_iS<JyIIY;61Xdx%=y
z&jTL%&IRC)q>6Mr83Zx{a^QfKrY@B94?TTi0`a!>WwT;45#54s9<JFBrFr2$UYR#^
z#P`cLhCln#NE+>Q-M6sUwctIk)Aq%L<9=}_!bX6{geHAioqT>-?YI(+=rqXA$=Fuf
z4=G9><`^F+@W7nB^6D?`>{<<@%kZb6fl|F_$ix9pk9-Yp#)uAWX_g{C`w^xl!;VAz
z8tK{(BP1tVE-Gr_paa~73H3J2og-An((p&ptfJ$O#a5#xA0OszH^>zM5&NUzZJ)i!
zcNsx>=p1&fk7U0JZjMcKCrD#TJ4+V@x`lil?7m+(LY8v%H{dw`H{(6WdIR7aIk6<I
zR(Oln)09zYaHQKrP@s3K)Yk1^v3HGCP(|ypLk)@LpOcmla}>1%%R^9IUXAe>S1<3e
zv_G53*udz`zaG&MIn<0%!OVbzf48Bo8M`44S{1Dw-@K`?6Fns=3$a+xXu*lGyB9Ei
zTtbS6$j}=_Tqh~<V=t|D@vODTgI{qrH-QS|s~9dg9DLe-k+;peNZ`Bbn3KCR<V`&3
zEGK&jRM5ESw-$e>77`!k7d6gSH*hOrc8Q}XooX+lD!$ElKoYP%ADqRU6-Y)~865Yn
z=?pu=I1Pjx<n$`6R5TP}E`&WWb)jtJm%VOuzqN_i?4^?e`~2LTWWh0un3RA?x;i1w
zBeg*Tg4dSVtgbSLA`AWKv&R62LGI%IA&uT3R63W=9o-K^TR9b1@Y6DoKir(Yi^j2?
zv+=d)YjWeKuEu(pM-#%>5s#)z=5;>Rl3>kGsuNwc;cqXLLdF6OOd-r{l7j;v<f3D>
zG(Y<b4eqmM*1>mXIo>P(t^gR+pQTu2Wt}E%VAoPMz{`&Cwn|h{xJ`TQurPx1tnd7g
z%u0!Pb5I0Km@TlQ4jNjd!FqwhvSDPG@(;fZ@IBxGHaMbWcqtc=r9W-xSgi6|o+tpm
z1uI3{78EE<_d=P7gF6C%%1;CR9<Ou5j7AaParUU@DdJ|Q=3Rc+%%e_@AgO36-R2<v
zM1MiO%4zQ|#wwvf+XSbW%X3*UEJhK*G+U#rsz`*fD~~e$2o)JmR6%G|0DtiO7w53p
z@V7CG6+ocWWXv8gOx-Fm;68FJU5$`v+1v52B$EKlH_uc37t#=Bswb5bQ)@3c7=j2F
zdabnE{(hPmc>U*Wv25i*DA}lxqkrE27337KTOpLvz($Hx8Y@iD_?c*>7X%8jBIXv^
zruZ>EdOl=>0m$H;gF^z)>`P-5<)B-Z<>m7#-Dt2Gp<f3w1_yK9t0!9utC*r%P35`c
z=;+N-R_RY4n=`WbuV40CY5Asih0rjyAk0uytX?@MnaHV6W-rm8(M%*BLkqQB@*Pq_
z41x8nWDl<;bG8D#^J|EUsz6U)zXDE|Y>qGB+@srPun4wV19zhY8h}3{L2cU7n4jQi
zarRXf>3rQYI|rL-o%XqbYV&|1bX$GSP7U?v3JVRKrj5kSU8CYW;67DI0C0uxgXe-f
zsCK$LHtHzHk8Y^1rw77*uA93unQK+nduBui>6^1P#oxY3mo@TEQQC~8j>7}gjYit`
z51j5bE(NZ+hkgw82_mY>>U8~oyW4;IkN1QfzAdcdZ@e-bG&uCIOC=aVh!?-1gci=G
zVC|~(UjTj{|1vHdt3j_K!?o@(wPf<0wA8Ifm!0(3_3k8@D`3jC;iHQF%P<3Yp1CDc
zZ}qNV_zE1gwO)ULPC4P)DJUDG?c0gWCJeg<AtJqbySliX6cUJq%IoD8X)U87Olw0w
zN|nG7hNM0ARxHdkP&R^liG@6cU7GxQV0~`F4&9ng#wrHJ0~pz0`-<5te5aEz$ZgOJ
z-vw11_ylfYk2WzGXM05fYwiz_vdUCpvL?{>3uj&QiGFU5olg5ZGnU%%;xq<hFTw-O
z#e~F)X%yNmtYKAs5tXR?!{)RG|0X<V{XvU=9N}A8GMJL<0;@GId4-IM^7~)f;8Lq~
zueT9TJ-Fb`bbWkyYW@E|nw2ZnE~F<t$2}Yw=1A(+ThH8c3V2?`f1_6>nPIa~91wvL
zE|v?`8BaIM2?Y>OKgEpNTnEGqw5~agn~imxnQrIGfj1vW{D8ZH*z;P>W+`Bj#09U$
ztf9c(1kqI;NdnqN;~yadV<tuxvdzo=Jt&{bcW;DJX&G9Ycf=GsphFKA2RINTK1CkJ
zk5d9wntuR~W?g}EyDH)lpL*LXXuN2tg?WHGRa}R>F9@cf6b4s;nuWVAFbe|>AKU*A
z*)yL0^3IILv&RpauE!>Ku21c~UDauq3zSml)hx)bSBt!c=k}B4h_Z_=qM_(UOJrh{
zt)Ap;PImHFfzv;x%2cs3I4!Qwf0A;i)y{*;u41-N_KFP7<Uh|Pr}>WbWVz`r_Im7i
zwX&0*1M}4-usB~=Kdmr^#d|<nf|2D0{`Oy^bV%&H;s{&{;KHXMXT}beMj{H#*Ym~&
zJoCDXzp>Q@_uQ1I{J!Ds6_8LPIZYjd%b42jWW?7YYVP!xo9#jQ)MDosyN3e<ih*S^
z`9<Y^?9JM$x2Z|N#1$0v)p2~l(C%8D|CY3GM-y2Hfk%M3;$|vIh%wAmad1lmB%7=|
za`LsI6cq>_aQ<LMN12uP4<rkvV|r&@^rGQw9U6HWyrB2!wN<TaQ9h^wIVqjU-(-C|
zeResy#-nG|?>@DFD^Ap53MEOj4tti}$@9k7wNrZ*9Nms-L@|}H$)L_`+}OP9YIf_@
z)wwEMbkWj()HQY-{DL%UdOVaiJE$%77@qD8D#~muE}k=v3AtnX{yXXVOi5tZK$Vqc
zNEU<I@*f_kY)L0p=H~x-(FzXS&RlO0R~4z}-XLNnC92z@w@rB&TuD1u=)eXN2#nG5
zqqWCpOZU5-i6rBgu`n?CJoK<BE|GWg!GsIRp0>YIAN&yzi*9!jr$?LeQMZ(d|0#eY
z%TvlWldF&2M*pVL-&C01bXiGkd8>kl-ua~%0)79)ouJG4JyYC#&kA}8u=?)IN)l9$
zDo+s+ZK;p=ym#)uE_P$op{#G#U$9rWpcU2qMAf|9O4lvFg<6}|W|EBoa$M2yWhe+F
zAPw(pOmP1~n%|iBJp0}wFFh&I{*u$fim`ZN#E$*QmTgv=*ZSoj4j7V{4p@93>Kb!(
zy1yW8p^9HU2)QydYd9tth`X+Y@P`(s>C&D5sfW`Z*lih-uE{7!^tZO~{fAb%o7iP6
zCH=9;+vukvmY*2Ld^p)oA%;i`bg)G*Dh+=E4eQCplb`4Av5gN@zpoG<D3P`v{62jT
zzq7v-8x@wP=xl?gJ{bD*zchO2uF<+@bO0dDM>a#3BnZ!aF&JA72U;N1R{BVX(V>Vr
zZPMXf%F8%cN=Qj&7-qzxr~{{XDro3nqr%p}-z1&GSwQvX4=)O&6;#SgZ(-^Jn*=E9
zVV)n~oIcjL>$3dP&gvzvcXk57eVwbIvMvfy$~=5!#?(+aY$eLGD$cA0@_28@mpg3K
z-3)Wg=QK|B;VaMj&4qn)_bX(kdmo_C@&}`2=}=WBQuCOF3%cNVAr^7RY@!L}m$l|T
zij0^KM1)yPyuy*$c8dJpnSY-IE5E{r2`mNhQEZO9k8r|L0%-1X3I5_rE8*h%5Dv<J
z(Cb>UT1{i7>t=bk&0tS%6tBRd)qhhryj}@*Oncd<9_N*LpBlYY<PV*1hL>$PZ}E84
zMJejt(aKE?p;{1ywyfXHIw!;9hJ4?mX5W-)E)rs~&)*=m-i(GpLm!75{*Q}Cr1Gp6
zE+nNtm3gS<bjjfiO~`rSn@hC4l-UDv!v_r6F;`h5&gdr&BYXFBrwf<Xrr1H6?jL{+
zStYg!Nif-~G>xra7Tm2g+g&RfOOK!lH-y*!E&EkP(!WzAD0;Ne#t7FHa%O@sreWkm
zS<pn^x_JoOx)O0PFbujHpp6!W?=Yhk4lYbjEm-DugT`LQ4sf`mGrvnm9gRUWh@ADS
z=D3_SLGuCLK?g$q*(S*ct0!kz{RDJCi6f0&ovZ;~xRHTRh{hZlGxHD17*P-9s7m!l
zML*1pYF?RiL_O{-E#(JV(oekcyPbG-ljVHEVFMKC&ECHG!D#wbU_$VZ>PW#9ix@Co
z5$vUE8X&{NP&NBYj_l#)Whh@@CX0u1`h!1*SoEaB(?%cT0KtpaATs0nlbO|0N<*Ra
z=<9K?B$Ri9N2A8yBu+kN(>$ocy;!KHHfT5_xO9YqFFYYQ+-Q_2mxUfv$~0H$f66tP
zp4O1DL|CM<n(+TUymDprYq-ekzOa+3>|>e$BUUCG`<n{Y&b5DE7U(%8Eqhf2>`=iI
zb1&M@ofv_dJ0(?YbH2NU;BygLZ-vN&ZktF;^rOU0rDkzdVQ+DVk80%x^%z#<^O@4G
zv$;G|e125C)7^N|9WQ??ic$PW@#%Q6!(40F>eN2q>tc=n6dGF-I5kSRBC4x}&$=WJ
zZ)0nLM$;T^UD{sNT$@L3alsWDV2^{CSw5vE%>Mz>n6zMdR?Z$zRPaZfFBK*o-0b5S
z>^}`Cl#yEmf%G>BrCBlme2(DWSiTP=_hUBk&V)p-<jcLYmom!F<4cIP>8nZCvjUg*
z83ndZ9ViI;HT5f4$}MlLY(wCu*QuPafj7eqCBtzNzk?HkTvWS^U^`rw09FW`DV2I^
z;P9x^S6fkqJ;J0cBfZCJjLJ}jm;d&4SGYu|sApS^qSLEYl?)D=E51L!&j5dNxFLBA
zt}dy}!%<ioCv>WPjQjB#EO;tUT`dJ^O+;=snnln6{sNX~l{Ta&Ate6^XoUt=BTFm7
znp_=!|Am^+iw}fKaZb*Hq@jds_rI62GO;rE!by<i!6k{)mFGLveLhhTUyjC>aC*;N
zx;Zn~76DgeTCHdHm>n*ipJbNXO-lqeGx<SNLoc@9hT!eRn+HvJ>j#IC0CCp9jaAQh
z1V%;Uvx;;ei2>wscIp|+Tf?!Fv<@Dqbgo#F?}l1VJ@Ol8!?GlC;aBe2BmrqGb3N)^
z9xRp+Oa;ht@R9?#HUR7<2Pz0s;bs-3c&}O}1hwSr6NHuP-N2n*S_i@b-deP-BGjt^
zVtPN}dA|xb%(MyJl+!|+v*MR^N#Tkg$f#6`qw)TRb9?R11us;^drEf87`(=T^U)*q
z?CexTOR-GWnu)@J2P2SewYUY6=jhyAZBUoo?W$Sm2<JS0DJ#K3ON6P;UdG($*wbFR
z*zil+mi>dCRSpe_7v8-0d8uKb-GJ_NkBTZ0#bk4d-ewa&gAC5kw9yL^{HeZ!eZ~JB
z5bG-HxZzzs`BA*7D!Hrb{YN$P;*3+~F}=ab9QEPa^qGzC`jYQ@>apZ@Cu{B5`OVWD
zzgBFRGvRio_qNh?e^3G!Qom=`eQOuZD@U~8$UOW#s+ov1+>wNP9e~7ynuSr{pQnSu
zXH(LufCYDzD8d~gBSU~~nO&K~sFy&7l?7o~%?N?T#9Nkn#<dlMn~-H8C+w=T>bn5N
z0$sY!XsW+syCbf?YlX1b!|Z*0g&cG*7~RpmNy63R&?J3HtnRZ^eG)Bp$-{hq4tp{k
zZR*0kcmgNyzX^gHBPt<=$6d4?9Rm+C(Dz>p;>*t=Q|i*K!=%wW$v-<nH?H||56?Zy
zSCSTZ`uIfbHw>^+WENkjplXP}o0*PyUH-^jKy-feJ5E^h@7*L_tk#qa?dl~{YB%KG
zRNILtNSviB^xEr9Tb1M`pOHn|_6sWF_DNSTKt3!k)H4V-CI)xKVv}2B^?#G+%PkbS
zf8Oiu?}T?ZcO=huFqL5+J9DBXeyBbdcWLQ>TnxHlPk=|iuCwuc|NKlzxMMf?$N6!O
z$5eT7Q@|_&ZdiLDk+|6C?G+0e>sy<0!>1Gn2mOX;>p`lZQGh4zcQQZ=jD9DW-@vWR
z8`sWv$=fjCj0!D#o#HoTUpEN7fojp{wf?2XC2@Kt&aW2XkBGyRX6Mrg!fihd=LgG3
zClo^xz(ceC-^8sdRWz{oDVS6LsdMuS08C2rwS`O&8C={@d2)8(<mnsuREi9;*Qi2M
z<SJoZWX8e|O+k5nrWfi&Kt&q31zglOIDic6=a)JVEH&Y2KfUh}XPA<>_v6F}>a1AY
zY(zSH_g5<BM)j3sE1{ko(gb3C#>nKEhLyjr;T<}WMx%fR)fgFxkYp^SiQyp{ho{uI
zO|)8vxO9QWyBsqs<iPnq)x&$q9fRLwSrI<}04_t+tE7&2%%fYinHp_Xq5baNyH_+<
z;t*SzXM=uoJ{w-sE*n-fSAq?#&;4pAeWjNUzGJa&_HAI@{5DSrme3gNZ3lLJ`?*jy
zP}|tl*Wh=zMReOIQ|Bu=X%~b(K^HAPd{yLn>LmZ<;*dIFkMpR|?=(sc5fL~(CH|q6
zbnj!2TC(e9^ilU(r%U3P5KWk@2>qqPjkI=!EfTtifDMP-DI|@c6uc`3FI=qbbJK!6
zdOuTPN_)Zbv*h$k&-Pxq@}dJ&fO}|=G(N?k!H-`>baYQv6Yl~L501kmReZ)0ufdw5
zW$t5TDq*n?M*MWSg_wAZh=$M_o!Nr^-6$L2lb@l=f+`cAf^!PCbZkU@1tG&s<q=)M
z(;};nI)b*(H<_gnOz$J`;93CaT^{~+r;^{K+3&76weD>D#m(&MleF_kM|Fm_`?Mq2
zozPAc$Gs3ihW{)>>}e%tVXNcj(qQHpKU6uPzlGpP?4<N<Ml>EoRG#R91`TM3pyZZl
z-_~CvAGG)Ma|!;Q*nWzfl}_Z32}ixzE=~d8?SQ9O^rc@F4XJZsOyO#nFi!O+gV)RY
zncylC{}9_TqW!wPXGIkc?!`~2r)b-Jse;w<v@NVCSwzwX@mPw?K`$DDhdfE5AN1_t
zcB{rUFSz^6mFXGBQTv))jX3N4>fw&f<5V2r)QH$xzyf$yd`!q!tPy$1;44(saAA!3
z&cin`+9l9`4g2fPl+R_{7g-bAKKSqJJsQ#F8e`)G>okIg=u-EwA0$)SC!?O+ut{2>
zqNj;mQ7)o2Z!xM~8B3poK~{5gklgTCJ>3_JPPBA@8^Y1f*DegAuHP!LH9Fe}3OOQV
zvXlM4&oOXmQju&~C**aiOb!%eMW2OkoS$t>H=MrDJb~K--b<biO4jGkpVT+*?`vvl
z&0?Ebix0<k8^og{Bl|5Y<k$=vcRj!ox$Cnc21G=qov2vc+_`rj7r^?Ttd5M9XG$7l
z3sZgHjh?^-xJH33tShUGpa9i<8+wII&dSk(uEwb7&stv!6c;~o4XY#LMeD3KE->^I
zTdsW_8iU^I#^uT+^4JThEB4t#X6%$>8JD4z0NVe*3aBp@azN4sTZ_m^`Sy}@mk`2C
zKRtstHO#is*p0^<9QZ494^s=2lz4#+3XY4VBk19tH6P~zhq254Rc*xyw2fM#p~yEd
zOff$7PhZuoM|^E+$-jxkc3o4IqD?=d#i`9h@Ujf6N0AE3WH<hVjaFK|cQe;PyNKg9
z7jIX)!?prT9I$kVtL%A7=#nvOu!Yasfog@j1Ou10be`+CeG><54EKeMzyzJ9R7=D;
zI>*F#I*skunroUGhl{G;zMYWm(toe@1~L3Dd#J(-SArg{HyqnC`kidq;4Ihh{J*}G
z!43yKH}qx4+u6#Y%F82?Lr^`;O0YR~dKq$p$uJT>p*WTQnINA8Hv|Gun4R})*aif-
zUYl<)<$3lP_u9FF7zm71V~rZzIF|~j6nqLG*_d1>GraNinu9p<^&s5L28uQ7b^Mb4
z0k9nQh#Oe`s0{*JCz#K9#ih*%b!|QX@(TrjTwONJU|NBFREhWW<$9E$9b_(SzAPt#
zTww&^gmUAolqgAsr0O%S$!Cy+&4){3NXV;c^5rgoH>t7?QK05ac{{xDHw5E7&qzg4
zm#%ov($$;*4;dDOsd=s^aHqoAy5n)23|Gf3+BljXgzMOFot9>lg5FAniw$1BMwgQ~
z=|hT8yj$NN;y=8mOq`Zm@I9r1O8f2<H_ZH@A;~v3lhkX1BcU3RpF-xvlS*Q1>hB-z
zrMIXPnJHK9`qwf1ufYj&1#7rWO`j!CUrRuC(6LW^-+rzYU*dSI?<4z<y_cKO=LgaE
z`S`w@OYGcevJ07T>J$I@^Cv9m21f%)2OH<7jsF#Jx}8h_ZiUUeVH0BDN;EWUu#5mM
z3JxAz(WD6%C(1zG*3MdC8oft)q;`+9rkzEdF<amK5~qv_bb86~-Y~1qBO`n&c7Gab
z@CdkuhpqU0I2jtn51}uw_iIJym0MA3nmn%y*dR3qmz2;`^TjglKVO1Wb=130wgsuf
zBit`nX;G^X>$jFotHv6AkjclFzX-PS+QACLrLbQ};U=_nB|X?SsVfJM1QEch>TCW{
zSuy=h73Yg8V<-4M-(b@nxw!61D^JSSQu{Nn(xRXtrQA~F({~&&*bSxGxbDZ-__VB|
z=Z9_dUO?dkPEXkiZG)DRGS$^?va8DLG*uZP^PT``K)%{SP+p3{ySNp#?uTPR2v;F2
zQ~`^6J-TEl?R-b73U<(b=iewYeGtaLRmUV@=kM0r63zdIrmGB#s_WXKhzub}NGUn=
zNSCCfbPpZUDcv9l(ji?#Ntd+5P}1EXAt4P)H%Rzx-sk$}hkx`M&zybsT6e9pH#h12
z9UI=s$>!tretx+&Xn1#gZrle3@^dZ+LaK8ZZaV!gakTaIY4O4;0SNcKaQ91A@>#{V
zAGrE(oDmO3>4m!giFp4bglYT!au3K_Q}A+bVLWL3nL~yx_W=5a)satN%Chz+t@uf0
zRSlI8%n;h5tQrUnk2C)CEJy68a->^8lM`oNLk>eYDmq@Edlb*chaYFZ7mxP8<IwN8
zX}Z^nqTn`?ZQ-k^z%R@*1%KQ59*p{Yuh^VIa~x>rie~0iPc582S;r8OohAS4b2-OL
zQSf~!bpl4R@LOPEm;c64#Pgxr;GUtmQM!Gu29{e7ibXOr4tfehe?ujwQHzn|UYlp<
zi(gA!YoitCRjwqGi4Gl+I288J@>eSrZq1u{g3C;cUcCR#UqRHx=xve}OJwq!qmlX<
zl%#GXZDc|ylC7WF*EhlG8(a(yJ?|0LTnH_uct%G;sAf`ZU9HY-HTzRQr!K{VeVzt~
zikvH<HJXfWSLtDasGXBOH}{16av-GCHDXduq#^$)t1L0!vq!vL(Hs-7&Ir|XN;*Pi
z_LI6R*&on;R2js}pLOc;gbGR09HOI;hVhJNv`AvTqHP!)#*@bJf!Fj;cY;DxcX)eQ
zrg)qvL1Eu|Vo&|r<kr$QQd=2(4)2!rc^_SNE;sGv(p#Ok1sCB4jOy<z^T0E;yxC^X
z26_K-$&p64gw2LeM*g}w=OF`$18>f|31=HFSu}%$^>O=BZr=+5;8>^>np@%65<Y7F
z#=ZKpJ+fB=oOx3$=l~81;%s$mQfe~&!i=Gt+q~GMQ@`YpK7-%In0!tL(~#G&-du4!
zV0A@x4Cl9}(ZEXFsMl`r5HgCZ<~}^kNjM9Z<TLyh8W*__gGO@l=<?7%m*uj(eF<kY
zWY!uC{u@ji#U}GxUhJ=3Xc;ps0XyJo4}3>$@L3WjV{|wL?QYZux3?%JA_!LdH*_hi
z_S@f>(FoOOLr;d6C}mX=n!^hRIEr!QKk@x*gsIf3C)qqNh2sw@%S=QzJxaED!r1|}
zZ`W-MASyssn4}*Iu<kmz`iZ5R=P^pypN#x#`FGVR=$v=<!B24MRdce%1H;kl->n1$
z1kIdF?wlVF9vd)M(-{oa?_(FIFi{9Ed3~{74}@eCkA_6^{yNn%Y+a%s;aF8aJT;G!
z>ZerXEZ-gF`>wOGk-^T){YgHQH?(eEC<N0Rf23l{W0jlTK@7ds*}g0P7Cawhx{1i?
z498~u%}cQ>HkI|^zV2TMC#Br-3RE(jZc^mnxr7e#Dh>t@IB_fexU^$3Q;DnMc5&=T
zFqfg}N3HjIYLxEChRvp)O3(YpmS@+!+431JNs~!dHd&PxyN6SY?`&i9_o4Arw2)Tc
zLOHW@I+TlG^1&a1PcE;=|6r}2R>sX_Y<PCtCA=%bfUC8u6aFN8q!zDY7REJbvQZ#9
zG#0_e9aJ5g7O1I^$S{n)hUNdow#b&}{hzwvV#d6a>7(hsHN{_MtG!>RV4qnNLe=~!
z@*`WC1AbO=!-lD)Rac?nYJ>sndzv4WSO~>(zI^$zd@D=!fFhP}xM9v!dLlAB`?x)K
zm=`@exvnQ&-?t%4FV0+VF6($;KQPPNIJL~b7vJVgJi0X=(<~5I??C=_t=7tr&ClTn
z?p{xSH4CguG@ZVYl}>`RI<yS7b1h4A8s_vddx9g&Y=qyX->`K+iTt)Sm`ueS%<Kp5
zrrst&t&@7pC;5k%?(7;^Z+Z+Oi<wR^-pR-0$=?QUAo@pzTE+T(wpjbMachLK<n&5)
zL&B)U(9sn8?x2*Z*MBo~i`mXi&U;WA0Uxg9yG56KDTaQ-*A2mMX6uf^Tji?XKFW>c
z8{Qlg<fkW+lmC>GiO+t7|JN!K_n~%DQ3}&uczb^k2cevJY&*7Euw5WUVPD)AeICo%
zzmEd!-FG=L@opH}u_qVgn~bV7>Yp-)3(=dd_i)MazumeTXSItu%ng+INP7Z^XkCvx
zeEbyy_;}c>ptrsK!izaa>Xoj=03Haw+ZgK~vk}Ce&qgV{P0s9hB^Ul68$^fH$hiM1
z_IYvIYcB1MSSDjz*Y#u>9;mW&jg&VtQLk~W{kcn6j2-3NaV@EH^O#qOJ2qt5jn509
z!A^7QUhdc@R)Aj)nGE7|bqhs;Lfas)0geTyRJ@vq65gvGDelIIfUmu;_{X-&wPyFr
z%szd-Tq5jEB0ltW#mIXu+s~OyEo&y<r+s@S5}s`06^ahELofc={QSGJ@M~#yt7HEZ
zTX)c25&>CYAo5g9S(ZeOSEFh*24ruNbvPncq!_Coz+(?{n{~Wg4ZjrQVe49C+g6V8
zXnoxbP>h4-KuvZfv06gQ@pG7+SY~xRxfXXGXQ8SqB*O5!WLo=FxVf<udU({E8CE*Z
z17pc*^mw%@<{#+&Oj`|tRHu8>%La5lKh>?8tNb;7qVm7Ox|E0eG5Srn`c<2GMsZe*
z*_em4+}Y}Wpvm6EBeFd0q(Tjeu;m-spntbP`XjK!SSa7my4NYFrVllCo{Q73OZBcy
zZY4eauorCCmC+he+@G#r%;sRC$hwu-UQBV4WjL2_Tm+kb#goRv`ab%wBl7#-6&icp
zs))(R2yc8v*2L7u?0#Ogy-LD$ePSEqYNB>-FYR2+k%{5u6D&JivKXT2<1A!fdB3{b
zir^O_Z{>e^j`?o-axL^l6cPW)zB=iP0-FpMTn*0XVUT9M!y-d_!~`+}gCQ>PJwt*^
zVt4q{jxPtny9z%0r!Ob4<X;gv*(E&dg3yYa<eML^Sv<E)z+x(JyKHzX(=!|0u;l2F
zi>e=MmEgjYhWz-UY}vU(@9@X_D7U$HwdPLt%TA8&vI*uX&f|sBco%3a?tfVb)z*GT
zkq*ZxjzW+#7cpjGxn+2mgr7>U@o)IVGqVQgk2trIm(DpJ-+vDDF;FhiQxIZF5WDji
znR<R}PH)|xFH13wl6$(IOhNF=-Oazwyn0?r_p4E$YwGxgp-cefNb<8xy$Y(p<p|!t
zUwL7lYvn>Z*cBhs6Ui>T^vFwnSCd*VctJ1h=Tp@wJDg`-oUR2sS#5dNV%1jx4px~D
z{|+5)s1#VS95lCBcdXetu8C$9QJ-ynX_S7vdY%%|gpUQ^lpfsFDwRMx%?L#lpUe&j
z)Dt;&2JXeaCH{`!h#NWyyjBb>?}N?Af3CBQlXqZ)fE!gb|1-^wrcir{dVfSLPF}g5
z%T~Xgg&7^?&L^qjAaLRIMLMOhUb8Y3qMw-hm;FnnYW=LG4lIFUXeeqj(w~gu+<1Xa
zRL%5HU60hH*PFw=s88b8$&F<y^;|l%*sLS7B#l0IvLRtvW8b(HS-=b8w4_yq|9R$f
z=G$5M?3!Xyz1!aG)h!|xtkTFm2-aIj8_Nn};0SaKC=p9&zFoUhQ~KK4Y0(Vo@t5Tm
z>HC6Qmffp!8$}4CLgnRuTFQ+nr->_8$vHqoK%Tbs4`4F){m++S&X|$6DK&#;<>Y)J
z2LAQloB$a8o9z9(5)7Hy>~PFBJ_I)nlxn-P7m6^A24A){=K8-c-v3N#?3mkNVzrH$
zeR*~BBl?A?XBvhs0>ZMZXgSC4{+Ar@a5uKc{nee6T^~`oEAqz)-$eE05Rx{j1o_Nm
z2BKY_Dp+%LE;V*R(dkcCC2%}WkQ!AWDWmyYfQrby5FJ$x9#21Q%~0J13q|t(^40f@
zp9jsq6@NVKdb@Uje~>#qHm&|oZ)+g6c-%UtFXKb~*kWDMYE*nyqxGJz)33nAY84}J
z1nV3T<7q*W&D+X{KVVU{jgtL+kAKZZIJIx}23iG|%UV`P7D#s$zgzFdS`1yRdTT*C
zQOPar>E%DH5k@JC98Wz8veEDH9g_XAeq)`;h5tfcyqQf@(+mo&3nAxSO~7iG{jqwg
z>M2}-(a5ycZEuJo*sAO~KM24EfE(_NsLM3Vk`R26P0Oa!;DbGdhoF#J&eMgemA@Ra
zOmfNe8qVp_wBML!t1Hhb#h4Lz3NWu_xO{>sco>qEGPS{G#!A-=TFYo1lpS7wN9yoZ
zf0eBMiJnl3lUG$_!9cop`C&)Pvw87oLod7A_gjA^JXtFHPz}<nbvRQ|kLGo{7U&f@
zSPqmff9_HFCLCv*DERL8gX5-*E!Lek-VNw1dQp;HmU+An)GN(%c0&db<=f(X=z!+y
z^YGQAlW~5ZKJK1|^B~U2p6lg6a}NSfGX$S)`p$bUOSxE3!PW9j8TzUv>`8IENUEVL
z^bkjC+&86pC$qAN-6dBaXGb5K_4P$LbfCgVn0)|xKGp-LpP|eMaC#|96Ic|5OLR{$
z{KhDY`~1Er!R0YLc3WPI<v;-vmw+E(ChU<HMn=2vr_pCcC%TG)M$REmd$lT&Ni=;_
z1I;nKyst?1PA^0zm|6~$05uNv1va`+j#WMy{D+bn9!(D#s##i0D0&0#KNoaC4}4Gy
zd^?Ksd8MNBQ{UMvgMHV`VCKH-U|uZto4dnwT~cp0^_ZFBbw?Q}?3=F}FIZ1%b#I#`
zQ#P$|E=mI?DeY;?rv_fg`?Suv6E{5D(kLma9tsQMV6~3|9pYwTjamH{T?g|Zs^NDb
zN?FxRtqKz>u%MioIu+md&i0?`v+PICkv`^vMP>Q=#F~_004yzZn(`O|z`F^~vp$^8
zXw|y6<ogSOS5wBR`>^E+-=$Tby_^t={Ds|-6f{W0iN=;q{nt)tKJQ>q4NL~6j`jJQ
zZLnQtoXP}zpa5e%QCmb&C2zr0`$2F?<@nd34_5|@0WH6H$UUFcP>ROV{z)1!?tX_f
z-%JvT|0w5;U*Oy70&W8U*GgqG3}mE91<mG0^tl3<9-iJq(_`YXC$DV86Oq*-8{K8k
z#6A}HC15AU%bAq~ckH$H&e-9Mn}Rn&4j1QR)D^Hn@ND(7jjbDuG9f<y4pv~BL03oB
zyL+X6Z*>oLJ(>RpY7SHZsCkDxKkN?O%YKV6Lk)ZP9Eub5@_IM_QRduN_6&QsB&@_B
z+05T_sh3FCA=w}k%zSZ_I5SV(tFT(qWH)-fpINFL>WZ&40EVq??Cx%X(1@&F<tijC
zt>5g&LrJn)CbF%IeQu^6;vF831fG=s2&uCS>MMPpd{MBg<fATQoNazAl4)6^#eS5V
zQ2Im-K#UqIXQ!i#TAYv*o(fo}*j$^knOm<gtq^aLFtX_7lCQnNUOa7BRPcvOp}yM-
zubLQtvg{ExHAR-B-vU<?!TF3+utY?0wZPMcc_nsr67SYw-FjCYvy6f=itv8T$A&|&
zP??}MWm6Y;b{;dI5&x*yXBkfa_eC&GY+;XYSIYJhRH$kqD{l5ZwVylB$2y}_b!el`
zf)#$Ko}4=;85Ls$BkUits=JuJ^IoQc^3|x{LlCfOo=p0DW>>;daqLXs8}2^1>nh~L
zo+G<r;Mm5nERnc;iucC%F3cDu?|%qgk?O&Br7o+UvL(RSRxKzD3+zN}IX&dl%>`RE
z))F%yd)2<PQi^ereiy^<e-dUWW$Ai-DRC{`s+{}wd*tEWx@^da4R=zBXHbcq)xz#Z
zT0R+eTD+Y7tJVVFE(=z758HgN6h#0RI>Le`$bS0bv#$PWkKBt5$!FxUjZ%{%{LJ<K
zITM;)LSYa(O4_Sy2i>c(K8xw&-PP)HWp+57uU&ex9_O4#RK8-udfN9#yomb1(h%Jr
zz6lc<kFgff^%J$frle~#O}ZcF#a+Kfy2_c@uhs}1j}fK@RV!*tgsQcR&GE^<?&oM9
zi<-?sJGJl>&{BiS3)xTp2^Tb62$kma&u4sLlUS6@#T4wta2dyC!gGP{5|DCRJUZQ$
zD-aPHhl(<#AmsOyncps*-9_w%&je{bL2c;6NXBF$V!qx3)4&3ilMrzdn354YIu$)m
zK$7YMZI4p5dm)b$s$T$2-%Q-3*6vF&lbT5Pn7X;$P2K3YQU>euHyGfvbU^U%@cIql
zH6AU4Q%aXdoR~O-eu}s!pue6l!`*sO(D~})FxlwkFBgfx!LTGZU2{*mfAvLeN_8^+
z%c@vhD~=+KApF!F9)>%XiiRp9tF1|q!s#O<hQ5EQ5Bjdhon^$Z#{P+0{ef!DH{SG<
zjMmQjgU@_TdMm}X{F7`zK>-XB)Dh5M8JyE%C+1V+#0~Q2Nj&A7iN)2F%^)HpMOvg=
zWH<=3=$6ooSiclsCsVt_-CziypRq(nRv6&F&kvXRg+2H3y6A+*dU1L-3w1Z@5tp`c
zzMXLiMhzR$yB*gEDH)4DZ<{N*jct}uJPF?y8#`GS2Nd^1i`yV-=AtYJ1nrvgVUSEP
z8J&D{N=uazu7!k5I6hSVCUhUxfqi_f!{LOP?uP8Ba=Y^w4j>kh)vf?%3zj(KZdGy~
z7pGuULpd?=6=4b5X}TL&=`qRmZ{;_2yC3o9Yz6LO>D5VHMdm?erd~zi*z$PnYsFIz
zY%)?&Rp;e0a@f=VgDy*7N3CK12Txf6JQZ3Egb=y)N=w_;$--uq8s5a31+Xq{JATv7
z(dPk86XoCHqz!{pYn5jUXXliuA}k4KzCA`W-tIzQU3(SBU=Fl%^D67=@o&W5kdpN4
zu`mK*hvuV#X{HO_utm0&PjAyg<j5b<cSYfid}k@V{Mnum;jNarRK$+0`n;5sK6xOr
zIYrZrE(HOaQ;6=@=FDs35lEciyhc(Hf)ubX{~dZO^2hFeuh+W7NaQqA9<$ePsaPQC
zv)XkQ^A6T~6Z01`^UbYYEKr_IK9EGz`x6bII?DcvQYK8r$E3}XaHtVj+lLduV_&zZ
zuMlM!h{qjk)x{>;V=z}sbocEU4FrRp4`{7F*+`a}LoPAvkVCHC&|_!677gt!B#kRn
zhOVDO!y4&8?#Z5E^(L;HYuBgga$}GE;jHb?7pW1)8Zy1gW`FJa{94#HHD-*qHwNh(
z<EW^lIr?xxvy0BObacrcn%r@PLR#Hy+O}9+mCe9hX1nUJSn6Z+*w*(CrDc*S5PE9=
zDPK8&poQWF>}8M6MptqN1?P)=JAUF|<`-Vlmw**SRNw2fE<q*}py}2#iwFQ`Wg9X$
z=Usi#KC1mh(xTng76lmjTU}bkZ-cv$LeB@#9`Tw!r}J5sUduL?M{l1VGuiFl4j(dZ
za$&f;Vc1w2^X{DJ4O(R96?j@mG`L;T!AS;wN|)Tc2~$B2GV#x&&1oE(LDURwhlYIX
zKNy&V@f(P%U<G}8d4PL8R61mbSAU=J*RUD6@3LdZEzC9=JLOq%IvAE=O}~#SDV}hr
z@mhQ!Gf?jn&6>atPi>KD+n=?lGOpIBna#Sq`F)ol-xixRJSu|pvRnl8B$T0ljPS8S
zV$4VRbxZ-gk5F{#c39+5P69)JrAttFwkn5dNs{CbRqd|`hPbXf=!17<c2t8%rM3JP
zT>V59Q}CtD%|ZpH46}`57NnpYi}fqV(l(6ue|Xwd6GJIGP7tQizbsuUflHqU`c_n2
zu`;(SsLPM&m)%=y2B)yClCY=$O#2i|O2MH}ZEBfc!YJ!5P0|k-K?A>3`2yCa&1O#@
zaIyumh6a^^RI@eg<h_xHZau;~*bz%yUyMs;8#}re839|<n;XsjMWF5V<!`;f`DIUr
zFsP3C2>aAb-C_reWv8$3oP$8DE!{Spw3d^cK%nUX8b~WHs!1@tb!1>g*WkpqvEX#(
zU$Roak2RNd`65k|ly*5k1EQeo*IBpMi5{QO^dVZ{!xQZ|t^2b_i$T$Q68Js|;&t-X
zKs-^qG|jZ*i@;zq&oM@J|52SRs9NNnA;H6ha%S7>VGQU+$;Y0J8Zs8!k|D9qNd=U0
zcr*NJaq45R-#(Wfm>>Pgj>?+EbD&u#HD}EV%z<yvL#tsG+ue+*xWpx5HF?f?`kfO5
z=As>K*f<tAJ4sR&&p~x!98LD(kN;S)^k+?qQJX2WMv%wuIUR9$B$Hf&N{7zk`tK<I
z60GaVk3<JivuT0EC0*l5!>Jn>M%1$;o~GKt#RC$mQcS4e0BE+etJNqLd{(dySa74~
zoBxu`!WiU_@JaNOAiqQ(_8Iz^-@a%4Z_g|hf_3?bPnVv4{k5f<v%uVH3i=y6ElV`r
zsb;NGk9(HlNrTb;OFjo`P<EUx45)12#|;(Nrx_pr?WN7z*4Zk)N<@C+UE{PMLM~o_
zhGDe4eqv|?{G&gg$i81h6+W@+0%Xq{p8}qQV{!MOoLv4j$0&KLUq;2ky}mU~`uy#@
z8Je9$667B2zPIk3jqotD5VL<?sa|{87K^qQcN848N~s9?Mr2D$Q1yuHu<5qc!Y*>E
zY(ac<njqMq>wJtxmcu$pYEGHN^VtO&xpY8Dv7=XUhgdfyen5+3+weXv2}B`Ly2b}n
zR}%_^Nl-*$$s8T+#Ss4G(prdse8!J&GJzv0K{<#=D{fY6u2O4|aj`RU+)W6m;T&n2
zLRuAtqmld2Y(z1F3H@@U<=G^qR5C40Ki_DlFIeKaxfVjIbNYc53*U($ZmX|%T3L1o
z-C(*N(@{SsW)TNqML>~^MwQ09I+^`8XS@0EWVGTeC73;8W8tYqjhWF7sbkU#>Q#aB
zoVrz#OF4ynSX_O@1K-V?<9UOE|7f`qpymE<uPdga7QpLii5terbc)47J$E$4)Jz*|
z9;n4@+su9&1iLYh7gOsiLc&VQ#&_sb)#&%zp66+=(rO0bUW$El`4|EwS%4fzHZQE^
zT3f&H%NSjH)HAumy^not8a+d*VUHsu+vw%tErU4fCKH<AGRZb}F!$B1E2u^uZtk_g
z27#Q>@J&QA-i%zg>^K&`EQBr>Yf%tI9Nqu@dWX%j6qjuV%`5O$F=9{ld2t*OMR06z
zo+F})^BDJyjOOcBW#cMqpe~#6t_FG21lxt?7k?YzA1ZaH6_-?lHiN3#Mw52^J8{z>
zpc=Ir_6gT?oimJ^`Csn9vLvGr{qn|p`m8y*$93iJte8@9^%f$2D~QLuQG@=iiLr{I
zcl`R}y&_A3RT5W+#F)ESGiBD|q`K&jp=iezPBV+`SQMr5AD7=htnwvgJDT}*{J14C
z5$RE;{ai2iZ4+%eo^rL67tLqi>4}<}$G+vKUX7CNlCIg&;eHFzHdS!s_AnEz_~xIx
z$bU{oSz$y+ZjawUw91HHE#I?5ipTeOz`=dFT2Lv}<3r-E1}m@UG+~%X%XF@h)y=xe
zpt?M;HIS3;DyE*9`!Rmo1_wizH8%R-Y;lWwIw*l#$hFrz{QAtK4vhst?P!<(_f5B>
z{qAHc8sO>gc9?V@MMDS457)`Op8-kWsMhB<O{iL{PYupPgxROwq|${yP2g5C<7k_s
z5c5-2>?#6)^}_1s&8dl80>wnY=gG)h$2S&9qn(U#flzjP$*oL3d=Yc+?KLR~oRe4f
zbG#t71Ueg6>acm(hE8S;89m7iKPM2`Jby|ybve}zz@!~=1*1SiyQP#!YEw5GXbgVy
zr4Na##3{;zN+=HGDK{x$AoMvhg~laV#l+;15YiAOMiAo51OSeajt|vBh}C@MJTW1y
z8=+TZ1!6=`iUsPg?bmuDDbHx<e5hzedWMw1Me`Zi&7BZs>U2*ZLsLfee{&p}$SV|R
z4-A)bOPa4sE^H57=E_}P8fD@gi+tE+2xnSDz0z<=H~%*1N$!+xY8&NaFiSd~fd(`<
z-_@aq-ZaU6CsBL#K0=z2yse)x{|BhE0HA&b$jP6&k#;UV0`&`$#%D%xKxZp|DX(2j
zDp))I#n>91s-aONH?LHyv7;-xScjB#fy~L(FT6lnuvI4;P#LlY^;c2!(>8oY9mAQA
z;<`*qYddlNP%0vqQh~gjxztB91B4#YIH$kdnIAzxE!{ev^71XOq1@c54|%g6;tqf+
z;B9{doiA%q9S8O@JAZs4K~SGu>Le;<8hBVX&vS($XWV}eQ*%k?*p`6ZjhX1P<4(Gs
z*s7hv>aiUz@wLyT5QW6k+b!R&N{^e7#M<ETi@$nHS<6GoX1=@Y5928f(SN^wR^s&Q
z6KnU_dS|b;8k7r8l`*<JcQRYMm)$#DE-s_VH}Q>KJ<XqtT88xPJn|q5w&7@*iW|8Q
z$`)2U_j6@Ni6w#WC(}v~BY5s6mJ*g=d7MYms6d*XQwKDjuiqheAz-urj30>p`?Fg%
zyid^IJQ$7}@#^e|!sDORJH2w0jWaK*?uxhJ)}j4G+21hd96BLb+F-Opj0nieTOjKy
zl3hCgV4gdovuJFO=iq5S5OEP~e~l1s(wqH$FNWO|DFgbye1~>8o}>_Gj`L0EK=T56
z{QS_XGskY{1^7LSWF8@#y>Y0_-0`4;*6^OAfL>&Vv>Q)F9qI_odi`BK#Z*rqg-Q;O
zp2(oq5U4Id=T>BeKdra9=<$@?+jTlSOs|HCB-acQP9X~y0E(M{ndBNd8)pw~d}8dK
z&+k+)r$Pl5+|J98SfVN&&a3|3B=XR=O-cxe+goz5S2{eFWDSZdm3hYCB=qTNL{Hdh
z6-3p{TqFw))}4dH9|Ea6A0hK&>ZwBJcUSj52fK&23$b)6Nlv+I$7ceMU<y4`aOg2D
zwssuu`a@MsXPobB!$sB@;V59|r@aiRE_MJt&3tlmbt!@Qn)CCk1P|jf#eVgwtwEZ3
znrP;0HJ}lTbxfH8EL2FujOr*Aled5n8J(xe^8S!a*>5>ZT3YMF4wdCRe=A(vCs`L`
za0_0oZB|kzMQo9$eRrc6;G;V~VY&)6TkTH3N_Cr)kTPDSXTX5L<#h-@A@WD#EjXZU
z(Rk;DDCnt8lQ9)byX;Q*0P)_<de6_Cu?p56OCN;X7sG%H@_WtfoX*qcz5A|+vKhV@
zSX5a=zTQH{y^7dlFl0R}1~;WMh{UJsi|f)^Ts0rPhr!3j2-*z3e!dnhBk%QCNW1e}
z1T=MdVu?T;ZmH>FT_=@bs&_C{>AK<HfczxWHXHG}{`ehf-`PGm`AV{18DNBk-68bV
z1kORiX)>PV^MZ5(CW>$YR#^`-@1eek*EM=8YlT1UQ}#YJ=leL?4=nW+!lz%?tD0FY
z;a%j+r0|qt1Si#ZzV($z@`uTB$&wDj;}QAn*m#YTF{R)JQ%X|uO|UpxR}Br9cM{p3
zvTeoMVG=%qkLV6Od>T#BLt=zb(IB*aS%)Uf%P-DUFz`E{W7KvEIqnKD13R3#B5v{G
z{m|O}LxTnZQ~NkpZ%bKWZ!M0Kl6P}cF%R?;(eWF;2^d*q@;f<YcbS(;B;+mZBNz6N
z>O+DPifRNfp^E93@*nKnUoNm^b4C#Jp_EWh!mNm^6W%tN1#k|U2|HP%#6o&}<>!=N
zPqU1d6Fax>)$krYOKpo*s|zL<P2e~Z4>k}>$m?R1rluhT0!_oT>#t+i+ut>k52Zy#
zMcbn>?pIC+tio`L$I+uZ@8pZZ7xE9WUhxtqZKbsXq|P_2>;6`UXVI#!b|mudrt=Oc
zj-?qSB93nU-r~_FuhPJu<}zD}8SI3d7a_XE+JIfoi)X3HZ9LgEY@Gbqitprm);`y^
zSQTBc?ncYo^mNgRJnMXS9IY&_J->x9x=g*wWuvkhDNynJ_HwS9_Y(83vv(?){uA-x
zWD#MXuVE#9sE%I5i{>R9jZ7vC95rm_tC&!k9OKQL3Og%cNqn+u+iQ)WmI;&rLhn=>
zX^VCr?*Y~rLF%TDwnVIH%(23Bb#_kaaW+cL`=TjWkhgg9&J#|iB(Pc-qvXY$4m9O2
zhCCc?hN|4c3_Yv~om)|aTxiPom5`hckX!<I;#lIs`G)0uPrAji;teXVeCT8(*%Qw5
zLaIUTPfA((5|j<RhnK(wZDa!eAy|l5Q%_%qQEbnZ&6rOzK*kB+t6*Z8kspC+Is_2V
z%QBL2WC=Hcu!DTS4M$<pzzeIy?U9EpdkR-Q8O@2h90X8Q_iOU&cwF8{$Z7&~kitUW
zA0S-2{C0Z9`m0+acmKPz)&jG4EAKh-{%da{_w5bnoVr3GCMSD3RH#Jm^b5O<KBp29
zBHzc?SOt+Cns%q2J}tp8IlZL}C*$m{J%jyNcTe{Yr^WTKAEd0d(O0u!Ji479PfFq7
z$$6{G17y%C7&g89y(l!XgesosX`Xxcv0K*u`9x}~%8kNJia%J{{DJO8h-i8}x`=Oe
zR57F8&X9*sS$OuQ`4o_|VfyW31S1ujh9L`FUheD8Wt`m+sz40Aicl&uw!Rm&Mf$r7
z%llFg0qpkIcn{(Ng_I79v`J}L&HeByg^r6;nSyhc!vYzGTQ2tH^{eH_7=EW)CaY|o
ze29apaKqJUi9XmR<dk|Q9Xf0~@4b8TV~VdvZ&UnjntTAT=z6mZcC!2MfJ+kcx>K93
zVSZG%zr`vvyMO9^h6!`N_{nA!5$72`p3ueoNp4EAdtrlU$B}1k1+dJP(kS~m8XxYs
z;){#`AKgUc7r_$Qb)V`|!8E}bS5TC>-NbGXV+~ub$^_qcjn2%`AoBNf+la1~)=^L<
zwwo)#y#W0?i3#jRPtW;WUhG)t7y*fRymqPX)T@kqM7Q%>GVmyhv`lRLyF!xrT#qcj
zGMt2~G2;DC0rU-I0B#U&7o*t6vQmH*z)k)@ziL5lc&{RDO<EAmfzyByLry<TE!I^s
zx4xg7KXXTW3XCuiAF(^r15etf1_yT>Z1i?-NNk{>V=PMBK1FXf1JoJ$VP6L{`{h8O
zag!UC2~lF?Lf6b^Ru*J~Fum1GQGp+q<!KuR!(Z5cjYQVK9PO6@OiC)HdouYE5Q#kJ
zSQq<kjw3u$+NiB)9xF|(RW4qq{b4Zz2w3eYf^@cDJ&AdGP*!y;rr?3Th`BUWmjbG?
z>JjtF5oW(f$!v;8xdcEwOhtar%<eZOVK}ihyf;B1HI&&#JD?3M!36buUjyZurWfJi
z$O7KxVjoFS2;`itXl|>-V`NV(+0n##D7aPv=rk!E*uN%Zo-kl1=5XbHE#}&77V`vN
zblDniH=`UJzsY*cSxm|R3w=WUiw#~JVw5KdJ1V0pd447jycRkqfxz-78d7+)d)KSe
zyvq0sGZH78$ZTO({^~f2+&4~@lRIgEt}IS;214R)o+mbzCJB`pP#yjP=dsk{uclU+
zP){mB)5I~q=@e=1*zfCxsKuZT%)>5(&l{`!|Lcsr_w^Mx0mDfO?ubI42S721Pp+eT
z@qQj?Diuqlhq3st(%FobC0^Lf-P{d&1Bg`21SE=8RLD>=@dykKd~=%&NHsKj0Ix+m
zw4g7tC9sfp$+Axn@itee>io;)eY5o&Fz?(PxWm%-XMXZj<l>UUAi;?Z8$^IUQxxiL
z4{4PnzD_B)1z*8_fr^{jm*NEhl^y40jf!blEB^6-qOWfOi%AiOT>`ELvOT?^@7o}M
zQUTlfiR>@w;h<%F8*V_j<gW39?NG&V3m+-~g`GG-flwloP_MrB>h^Q$CHonY<h0FG
z`SioCL<AqOJ|7w^Wl(ZRS4V~yJNcG{=QCn$vAbzr6atv(_G5W2ia+1;S}zW&u|wBL
z&3W$hy<;bXM1Uq)SBOYoW!CcNH(CDSxX;A{Kx|}twjCb$eqDFOAjJv1kD|q2F3F#$
z+m!kuQn1EUjfb15AiTKWrI3n-H|IcNVS8lE^r|v^hJ`G>mRrhhs4x0AKT1;h7s`fl
zcgSP{*m*khQfh>zB(`csToIS6wNIOJaYYnkAHai3^52y(72{%k;>|+22`Zd!{ca!k
zzeRNCzF$NQYdYY$f*P;PAuYN8j2A-VySk0%{)B~A96Y+s--OI_jgIC9Us7K5di#_G
z7xx2Te%z%KrFJ5TM=!rUEC#+H!%7<1C`f~BZ7-9a!>$M<@97fd-2)Kqu+=xRS<hz1
zyMLNG;!95#V~Vp5FRc?nch$yqwkz(4D(~EDO&e{t!sz|U54z~+%lE5N5MbA*;(B^(
zB|79L^o`<DU(DDfoHol6bpglq2l8z2$Stgx!WldtE!iQs4`tRP$jYi?3r(fJlwu55
z2UE*nEv96XjaPCH1RsCg7HH91$)%q^35tpV`%C>bQ6{?Rj|t0JAX=_y%=GKdm_)Pn
zYn32g=rXABs87JelR!jaV0vurdou^MBBv=p9YE!B8eS&K)XTU1t&U}8_i~LIyehQ2
zxEl}gkaaKe>Sp69jfhw*Z8bqj-9S8<t40Z$U5Yp6(k>sk;1cSp@o%qNl#Pqcspaq<
z(GM`^)qGrxjV>MlIvT<Zs|5(p#ta-v_S4$s>s251>5U5y9GZ669B6XXzc9YD;dm$=
zB%wX3_U5an>v`GbQBB0J5O*m6b41&{gTx*PKm0&%oNu1Pgjes)G6?@frk~|J2<MVq
z?HEa*5UFVcvT!h+vz2`ZU^SqJyciPGAfx+*6xbYc+EBhhX>Q=|m?i+tv61@ZD1I+w
z#0>Pn`P=58>%S%MEJ(-P#cdTN6PL|rHh^+&VgZs<Bs<t(ulLS2hBFP`%$<?b>aPGh
zt!Ip&$Dpe$vs3lQvW|t+Twlw{4PPh381>3nu3K4!5W*`@;^|q^E7dtdM0}G1q8WRc
zgRs!8=dS_~`Qt1k+d9I`vrIeE>ko?rutAWaj%v~`WBa-<OoJt1Ar$a}3HdRm9pE(X
zb4jS7BB(hrfjO9B>Y1R2*wF54l<ikWU)!zx<CQC51d@Bau*}7;H6qw^6jxt)vB4wi
zs%x?=<+G<jAtL*_R(*4F>v`BjB(Oj47VGyNE3z=f6q3fd!=?qv`1?`8{oD@pP?2Q<
zIjBU#zsvjxm}BAAeSD8~krbDfv7i8`dYh?zSD`5A!HQA|c`!Iu9D%`7=HctY&aMdb
zc-wu~lu<odYP07BsEl6|U3rLXvEY(%c?-)wF_(_H7fT2*DewIk5!ri$qg9@om%5z4
z#}42L(EWnxYQ`&(0)2ZTH+*LHt7`1(tu@@M79$n+jPj&6ZDE%W_}(KGt=5z++^4f7
z_d(dRxRH?0wd7v&fR#_*UbW1Gb!r1&O~b(PbS>dFY}C-lDRur45j+V!K9cp<?j($4
z>c>LK#*kAGJ|+^y!TxzQ=czJbn67?uTrrTS@bE&L;blY7YrTxGI^WgWWe;$O^RDi<
znfa}9x=;srl1)s(f0|{47pw9BF8q?}3-FzyAN@%R&L0Gls;)j@dBE}VA!g=1G68*i
zfgu|{F=wj-(o4<FoIUD<0GPXVf3e!J#TGJHAT=rSWUe(W0#fik-#h#C>m%%RbhUU-
z+x+nCfrZ~qKDQ$=1!c}&uZh+`p<2{#i!~2ZO1I{8IDWBGvGf9H&JF>nx@rvCAnXYY
zqOjV=VVG=cgUX#LdLJ~bkT(qm$1bNBx^y@EWlRd>n<3f7_0%Z-Bs$JxV<09DEA~xA
z&t;)SHWR~>l+7GEQCTBJ2eRz7lg0>N6E#&)Mw)0?KlfYt%|obb;=RmqYW$G3z|<<m
znx*7dWPv{H<A2QlKCQS8;>ERIEK*DbvWpX-!4d68T^?3FH*6f=0&%P3y959&NW7le
zD8!F&0h29LXo62L3BT)RDFVcl4GL{beBRX@jm;17x;h_Scp0`b$Xvw_>W`GEKegEM
zJ-@*qI;R5`;jK&`pwA}sL`4WDz}HI(Mm86|c-ikd-@cQYfoPSaym+4sqS;~}VUMIK
zMQgrJGHn>Re4t<d>;H*}J~BP@lXhmBjWq`*{F4Tw{Yl**8psDZiviV(Tds4OFj%o`
zuN7?A*GR3+&ffWuX%>Op1@TJj5}1YyUF#4qHtZ4{n&GQ7;P1=_S?!~*CeDP1t9!-^
zqM+iq7WWXdk7lI<3LvDXS541Y7;?^8dIC!y&bP$063qT<jug0y;i+Wue%DJe2MaIp
z%o!JDq~jpK@y~q~AvO1|${4SZ<W}2`)5Ma)k(z;o$wHH9fy~3_{8-pGL8YD>>}7hQ
zK~u3x?o*J}1c6gZ5E`DOq?L7Qe}&5@Dock-nO(xf!9gpYV{YA`BAWzJv`x@1O<f@I
z@HrX5l^L*LNCq~1dZoa5%3pRNdf&;J<9M61gWoOp?wj)AQtrLQ(++(TPy4h01#HRr
z+y&gfOKp;xrmZfyqpAuPeycTl?3LJ2RYWR0>hcOJQ#l{B9DfU-e828sQ7WHiOe0k)
zpCKo$+LVN7;yBGe*Z-h=Lu=JT(cC<W{DHBkElp4eLX(Hytb1_~;9#zTy4pPfuSQm#
z-ZuZ@8@`;*pqhDV4pdMPPrFnozTU!13Y}=!&?~(gzI<LxBwHrs@Tir=I=e%TIOkeW
zc)!jf@M*!u27#G<!HJP6Q#8e{#a6j~<=0hOt5<6ZSa2#u&c+&t*2}nv^rW&H?qd%t
zZ}}$M!)%kQ=N#*pY<^y4a-algacdR*$s_T6JTVv*pGNzVRhE&2!%RI#Vk{L^!Z8^4
zYTDlYxr32K`sf2$Jn4t56JrGJUBFX%Ok4t7z_tmvyneQk4cHj(Ug>A+PakzGtb4E0
zf-GuGcqGD1nVYKw4bWPm{m&^ujEK0=*z)P2&V;A*ebUva+HqkuSdg&rViRj;5lJ|e
zG#>xj$FQtwRZh!UG#U|eJg6;?EeObrr*`WtW*u(maxO4P<}pg+&FMC0`rY+Y{}=Lj
z8+G3fL8)I4`Z|4L7PW?nq*x{@cN-^!Oe%+|X?-Q`N^Mg<uKT@aLrdo?pPLtN768a4
zQT^L=9he#qA=ql<h=!H7jGE^bI36d9^zy30-o>4OK-)Zd3Q8*o1}E3|f;b2L^bQPu
za#hYXc#EOAsarwh)&s)GQ0<;xrMmmO4w4_>du!z2K;LrcOvuD{YC6d2_gv8Nd+tzw
zfJa9HXU>`wSYoMB9w}8^wDP^uJKW5%x7jw@!RRF#5cK<+y0vQ_&`LIu$_LdB&K|G~
zm%!a{4st|9%mBiGb8rFBiNJ+apkufhLs6UrZb2Tk$twN-;!s?{c9CTHvtD+i)R6Xl
zhA#&aeQ}h80k@8AW7sHR)v1_GRb)p4BMH!9AedhE9WAVTWgCOp4=~a&!bK^~u|T=H
zYS7mPLC5!J67N?oCc&T`lv1ITJ1N@*X1Zb~$jw;!thr@@guuz_G_geHbhaf%9Iynx
zSQoMJK6?TIpitn-c*yVHF6mxe{~Z@)xY`8Pd4VuUwl7DsZm~OyNn2tRg$Ka4q^9R|
z8QP+=Z2c&f2>;6?(B8*oLE1$eUdi>?|BLk|#h(xKdDd38-{iPp8Q`8eGsDYW&3#tY
z)5!Fb3wi$l)w=ffUeYZ-;|Kc88m5`;ZIsem+(H0ToqTKm2CF~imA|}6eCnHiT;?sF
z#6&0ZuZzR5_4T`A>*FEwwMrf0LK2WvW+zC^h*g?X#)+5D6sZFy&*ehm(`x7m2n?s9
zL0AvupB|VH6fTf^?kb5B$QajHoy=M%CL;P_Lz0H%+DY~pIE5gBh$E*4;?h^k<8;%4
zA52!e7hX0*E}&}<lo@_QhPfn|Qp&t|pQu%=%}_0YFbBl-uw!vW@PIcuvBvc=hd(|o
zGAk2t{YYc;y@}lDaCyb9g3RM@n{!RPO`OeqW*~{WFpd@ub2ReqkekQ1)!%qT#;s%a
zh)M>gsW8weYeDK@<~}eMLHq;)1YSM`5ZkfgjyTf@OGJovkAc*L_vA-D#jy3rSfZ2K
zG&SFgpqJ@vqVcq{N#SJ!d3ag~H_H#X%ud2hMg4g)QYq-*UhOEv?-{j$3{C1|h*1J=
zUI3d6c!>rgtaG3^o6%I8JR+G655bh(d>3>^`n+WNVyEoCkU6<1o>(Zwxy(&vs$$`k
zH8)*6<cIaydsd1<k`mYv$fZq2nC)Zh-|Ykmbx0geMzfj}VR+MImE^1Nz_H+Tp`1L3
z=@lP?q;`P1fleog)L82igs;~I%49P7*ySI8<a6m1{JhsOf(nl<nj-}Y!H>p%qVluL
z%dqoDy*rAJT0L@mK{8=VHMOz$Xp7g&vU=#PV%_$r7|0%t*1RR-lDp!Nlo^~9r8~)w
zw~DFMsqrw+Z68nSHf8w>qRj}*CLqj0L#teZXcW?ZG`e~$86;Hl9s>hCkD52yF7en?
zOjmRE<+X>(fc2tjwk@R$-e-!F{In0{yGo2?7-z_QnQmn;(!w`fq=OuHA-4D?$6p)Q
zGcvZ-T}EXh(odv}l?ZUz9hVZbr&EhmQ;xHz`@oM8EbTZ?pR$3F==!mBA0#O2#7HW`
zhNTs^QTothALGD4GbURu`Ds#GL{bk-9%OT?g^nFN0-=K+o1-YY8Aldgc-I<N>(m33
zAsSb$%u_ta)L5ci<Km)CEq#o!HquA`NM6#I^gedPH5_9KfhWn-_}D2^`@slhIwCkb
z_-GZ?IH;)lVsN1(9`+m$XRA#OE(JucN0t~4I9um0ZktjsCO?2EgolMs#ldyX!%K8a
zSpzE$?$jOqkSa2#+?^x8f9nvAH%V<#c)|{OAZ`NuSqYgQfc1@kbWJ;}c(6f0j$FLk
z5NO~F8BT7miTsS>CQ<`W7jfyOV?O?9e2m9si>)0Dkli7sQn_js7l;eL!viP}P)%W1
zuUNP4Hi(JuzeFKLRt+GFt)#gJV=aVIh%~*iAmyOWu`bekodS%~d|OLF_q<p#&9}(y
zrU}Q&Z_WAw$J9@Igo6tz4eC8Mr9V|?lP`s&WsvZCKQ)RQhy8wGpKPUhjJ+5}Q6n_u
z9Kwr0`n%y4Hdk{3pX&*5^CUie)fZDU`;*ke6-wg8UO*Qw5wOISXKV$v>H$&EZF%Rl
z^$lwl>g$>qFyUjpxNb4ZI+=miTrJcSrpB|w<F}?pZ<(ZEccBO5V+I*dD*;{?MGny)
z_wJLt={91(Yca=s0P8leVu={ozNio7lC8ART;*s(ISWLAKJk^uc7!GIL0HqHtXPVw
z=}Ie27gJPU>g~u!`=`g}x3E4oZU{+OEXA`tl18Ws4}pth2R6qI426nn5U*JJC})93
zD$$GswFEWnqr9}GwQw%q(Fe38GQSz`P3e;-7gF~D7Y2`+`-CdxASytXSnz@z#+wlP
zgF_MFCdCAZh7BEf7S{?Xbg3U<abKY5eMJb?se&l(9^XJPP+~_e<wncDuUtAJKq#T3
zzni0ma>aZGq;8*=qWw_M2QyYULaMPU)9?aOgQcr(i)n{W9|)Dq)~aOeRq@Fnh<|}$
zC#4FPtMy_Yu>g@)mo~BPWz(i1pz?j)_SxAd=8a52h4}q`M3&Mh=KkENW#?T9M>vfA
zoW`q$K%xJ{BAG|N<is=z+TY;Dwr0+2D5h1GD%hf3sL!EHNJ7;mh(g0#D!jz<%mS?2
za>Pv20@?ceH4x|fF#r<#gdk6OZ&N3{BAk&S=3}&?BSx}kXuqLDEsGmb9X4I1-;H|6
zRM!K37lPWkOkk}hS%6EH9}HzVI%P|^KR>3|^v22v^(UH0lj^bpDGsf#q*^CQ-Yt1G
zT0GlL>=$THlRpuNCt4=C3OCR8wVY32i3}X(5X{HSS!(x5N8Sc{i)ZomS7JX_(D<$R
z$nN4r2bZL2Kx)=9>NBwcX77joWu;(@kU-LWH%R^_5Q@KE7fsE(B9)zE6KF+FK|I78
zY9=G5f{K8G*`^U#TNJs$_P^h~Nv#J5(f`RkBKL9+%f12_w?bHfA*dG)UPy@_)7oXJ
z&D8=RrANj!hz63|%XYsuQiu5qlbKG_W3*!&jH^Jr_2psMBql?J?%&Y(1lGk}kTnO9
zOAwoiF4IuS6CBUAEds61<HR&Xef0fX&^U<P=+zaX*7i755F5V!JKQdko5xFma!p9l
zvyr~Stw8Wg<Q3X^2*9_LlXrcSB~>Xo<cxkb=1lWd>-AU5r;7yeBp}JaG|M{mj74M{
zKmuE;mnu|^2l1TZ^-_*D&(i}<;8;qL1L<j4RhwZ4AWKa*`crdJyKP{3y~qBLNdpUf
z=SPZ{Pa?zA@MDzI6Ru5nKf5PMslpC#W-$8Gdq3luBroOLH5S&;yk+tS1AzSvGi=Fz
znmylx?w4MQT|;uXs!N~$&C!D(BL6LtGIBDKi_rE=2n;MGV1r+W7r~7w8B95Z9xs^N
za3D1l5Xkk3VYorSB^DLTcrcBrL$n9>tUCB;o{^};JsY*S09`uQv{LT}L=E%~Fh~JL
z2_KZe#^8_DG}%odtOF^R&mL*Ukad4#@-M;QKyk!K?c)_8uU7Z%>s2Cz{HGa&H`_d3
z7WudJFWv80=??Pb$rHh|DYcs#zw=R#yQ?h<uQ|W7iaWyZxAWhJyCaBDJkIAVu0sAD
zQndcw_^$c++b)v+R|1>D)w0y5pz!uP=ojsY*Hy}!!0DR7De&3$h~ONA110B<?_T`?
zX~S=_$gtxNbiPCTefKfx=!$7Rq`I))-_R{NTwpBD3x(_<wNJb{O_#gW(~uQQ<~5&T
za$uCi`%yAb*3nT6k#iQSy@GW*8~bDVoRDH|(ndObS`sixwG-8xu`5Zkl~j7V5W}YE
z7=WrB3@%BPR;dq{TI3g|fXP0V-l_!f_b%5{dAN7lXB!#%H#=D0XquUVcBR_Ex3S-5
zVzZtrqs;DVD_#Jf2%8Z6GaQXq7c8t#-!*MIV4f+wDhavsd8DEC#i|4h4w>!JgP?|!
z`!cmmrD_ddmDWh^&T3@ZEs1w7;2UJ{WQC~SnG#bmwo~fX5B6XP#j|lhj|+~S7(Yme
z07FXhsc7u=jcVzXX1m>BxuL^-Q2%U(gh;`iIS5AwtaE}efGv|g<|8$;Wl12d4Hp?H
zvugyD@C6^rqD<e&7!~2wd1c8aCPz3l)*KfVLc>bce>|xMt#l;;qtY@mc+#9PC<>4D
zRgOJHImGBc6cNfd=|KOJ7#NPr2|oT?5xIHdd#34^=y%2XkCx)^Sjgq@uNwv5zlDAX
zo0;;ehRdGtZlc!1$9DsUSIj=AxfI=83chQ+zMr!G7H6$a<jZ|6`|wYE-`@3Q03S%&
zgJG;HHxdy*^)TRaV13!^ub+ZV4U)3$5PXpE&D1M({E%6CJ!?%UGBC*xOj#LCl52L6
zlS>ZQYxpz5`)TO{Ao2r4`*LI&;CA3}H%lL;I?K)Wlq#73!pH^zg8j%;wCBoFD4znz
zZDO^=fb_v|utL*UHIC!Q;lP_~^|EOboBN`tit|bfT#@-?f%XI>9;S3X>-Kx(cq2ML
zygfR@c9Ma_-Xk)da`eHzqWC~Cuq^Vdc0lvDUwp~tMDr7#p&`>bef$mR-ar{J^21|0
zGJ!&X)scpc&ngdUYJk=|YdlP|PxzQ-E($AB?mb%-r+pm$Wgy}PQ>)?PpZr+nKRAZ{
zTb`!JH!xD7$=V?=QFj3tzvl~aQn9qa)b9@O@6X3G)6YNXIZgj~TQxeRMBK<eNl6##
zL}d;Rh=9lc^f|8(z$EE;hT%y#f{AadP+-0zQAIhR3e%@jcnlH_J3=8)!WjRUH>s?i
zme%1hW)f(kh|R0k7~NsOqoAP^kqpDqHG(90@bNLTFlqD!HmXogj#z3rLG_zj*j09&
z_<u~vZz)l??SJdP{6wVB{&w0PCB}?|jEw9i>iZGzrLNzOeaqh2Bv<~p<)QBPGu@kG
z@V1tDgS)21JTOMa`s<#(#T4?ArE3+9!FGJE_hSJ1=7ih1ybuO@Whq!yFc-1$4sUtr
zmBdd%6FW9rRc<)w9&;%XWt}G&#^4`<#F@{ARCL8J60k)9ZtXGtvA%A+;eYZIq#PGQ
zod*f`p9+<`7hD>FA*<EIMF@D?QUBz7jU65kr<X@3O&ZD4`|Z%IPbZCM`Z|kBph~&G
zwCL2F%7nEe;`rQraazkp>zf?f@s3j{7~^%DSvhVepjtjvO{FQr1G9`7W$c<y3y7=b
zTM{&%l_p8$<~cX*I9S<>#pDr(%Xd!z?MNnW+*ORuF0y>!>y-RwR`nt+CK}dt5Pr+Z
zj)rX=`J`c!QN8nCXNMRAC#V%4INY`Nx#mWHFicoc80}l6!X)MrK-|K-Ma6hGSZ}8h
zsGI=|9ntL(tH_jrdOL2~XeV758Hnbz$b2N~?kawRhgS@!FbFQlaQO2QnZ&`K6<RG#
z55GZICWuMYJMH&eB`^LcD5r><A0V+#D5AYS_%b|BcSQ>k!1Rfz7uC_=R!+hIMcowl
zP0~ikk<rZ%lIP#CrtoQ9#l;2lLE-S+|HJR*gSwBo468vz2Fv#U_P&3(<kn8Q8nO`X
z1h4C=b$7uUJqcY)F+X`yv*J;+<xqM6JZ1pafGT(ON%NeqUT@`8HAtk{qivGRl|XwH
z)b+yTfrUT(3)}pZ%VB4vAV6-a%`d+RRsbjT`w}djK0^AY!mvdgVFpGq)1u775c$B>
zpx`Sk^ADjY1QXgv$9ApVe}dUDIDRz<ti~743nDYUaSeo0s^-3>UOw~lx%zn47=+)W
z{t8<^_VrJeDuHtlbU{Y}Awi$)A=Mu~;@y4N_+B+1=RYL-;{$wqc{)cMv#jCX<LBiW
z9ZRM%VFm{Dp-@`|z)=jEd2@CGr`|_3=0D_r=ubww``Ks0Gw~>F^1aRp(|Swx=*+>r
z^+Kq>GNl?;UCfS&$Fes}d`<O;w*wpB1#VPp4z#61y!NRGnf_${z`pq!E4o>-CpUL;
z-V)B77LqD&LIs20@2hFma5DvqHJQd>Y&TIQ<GkRA9fi=z1Q#$RU<biIRV7z$c!eZ(
zUz_o1n_7kk=6!r46)Jc}qrgc`CH)u*fu8hgKID2$<l4w$B8}Taa(5Gri=%O|uZb~P
zAAXl|vbGieIOZOlQNWj6A6EW4h%aE2lf&`ld-dv-<Jq4d?k<tvuXz8_m6n$JY;tCK
zZeMh=L9seJ|0xwt9{hQMtxY$KI!a+r)qe<5t{{Q%x?X3^+&wsdqXW#Sq7fs8f-)dA
zY?~k~1e{Bd81@&WS_Lw6K0@@G8xtQyAr}6;?!()S^Z`|Vg`R|&Tj2jn`^%`RxAzMi
z-E<>@NGc$Wv>-@#r-X!nw1jj>2!eEn(ujbxv~+iahyn`IA<`fvb)S8{zwsaU#l82<
zb)0dAV{F*&y*_I_F`qf-%Ifj?H9I*{_J%>S<kR6h=x3=_!bm{I`&kLoe?OI9sSzfX
zee4P^eQO3iYvo|aeB&U{iDymcjU+|G+1RTsv6@3|bzHzd!~EMHNnU@GgsYR_@JAJd
z-1i1ZX{$-wgJ=FL%ROG_n(z9xq;CZ~le^{$OJhrxSX8h}$62U!tvE&2niY!FXj;GG
zbU9><f@cP9A$pC2<GMJ~?-Y6KXFR9V(f?X6n!Y<YHHc0q<Nj;bs5UGmThQ_hsMz_E
zGS<YrAf6&Cxow?Xa(LDqC8&H;-*~*)<Q<<o2FjKi>V?-q7)hRtO#~aQEH~Z*?`op(
zr+k#fK{<s%&N~Dtqsq=z>~W0Y>_aq7e55OKoQJz;a%GYO=nVQ-EwsM%#fby5W@)Qt
zd#!lceB$}tU54%A9yxjlukwSID}1ET@8uLu`|dmczp+<0udn8>eOmW|xt^unpmDlB
zGP!oYnwvlOiS?x#ZT^~s+tD9gXh^yK#EwC+HyOw*9@V3}EuV+>gL#>MS|TOmqCXe@
z1#}JcfD%fo=YQz<xp@}UWz*iRD%^O)V$Q`qfXerg-Ea8d6RmOUUR^GaPH+I#LBkLh
zJoQn|CthZyz^DhcCPNW&$^>8uaICIQmr&3I2CXj)9@?<!6{mXXjA+C&gpVY>Vo~a1
zuw0PA&j{1Twl|^+dwfgZsx&(h2TH|fz^{lE>6MHi3e>jH#wXBvve)(2@D-WiI~xNb
zBj(sdty?ae30`75;XS;y#TCN)XQRJ>=4-Li6ZMOHs6TgQl8PyT!PUlk_+3mFdlai3
z2vr`!=b;@<3C2lacWt(Oun#q#h@7!X#u{lIsBcK9@U#gs=h#0yARuqN>6byw)ZkQV
z;DzauVKqxpUoR<_#>k*4#~_g;8E;8U!KdFj$q>}<>e<3#PKie=ohYfp5X7$VIE<Jm
z?A(W`*$<N*t+PSz?Msc}fw%&1g$BW9u2bW3f*!k=bE^k;@s7+;R$2y6{yq4FlnjCa
z4Y};i+82kyZHK}aOI%k=i&OrqPyB-SUk>~KQ@`F&EkB3<i?1T>p2o5+2D0|R2dZUb
z_3rV^VQ+7*@J_BH!BgXUfhFcqP97&|KE`rvzeVq44@K)z)3MUWE`}+9sKQt?H3YcP
zv_T7i8aHNgNo>Wu3KX=w-OI<3Sz4(+H2DZ{+6g-&&~roF7X54#&`j*<<sV4K*zDvP
zmGx0Ct+Qa19YGwz=K9IV%s(L#$&wEgTs$$E>x$L7Ygn`@7El~T)F3fVnhNclM_wS?
z5Om^l#z9Cz1SU-W#R<vJc^cWW`*}?>oE#UN3w9S}dD0T<B<+%7*pyWQeIKg5I~ZI%
z7L-V|1f{;wJJ-q|?Bi@3FwF;%(iZ~f&Xm-w1l$hL@yN&4RN|*5d-g&>5e75v;=xao
zDlvQ<+B$Sf2WLXkS~syhFz>!%9wg!aCL-ieX{#YXAoL&$THow+u4a}MsvbtI0tdUo
zs%Kky_?KZUO8vZH>v{ar?<%x0=x~<s!RbLcqt(DUabWw&((ckPcaxQtE4!26CE?$H
zTFnjehw=PR<R74{i0!U6H0&N(>mi5aStq~nKa_2XoNrb8UL9tM-+HZ|w0}0{|Gw>N
z>iS>0@h;8vvj4F7Dc1FC@vCKVKjnQaaPkV-?;ZK>sl<k!B~;zt4z#zqus+?A2G-AI
z!2G~e0EInZefp*8z=$WSI~bCjTRK|41xkT5wi5TmbR#UIq#x;+8OH1r74BE3Us0$=
zh}A!RXKAsTn2IF6x-f`sC8-qOca0U6UPY6q0uA9qO0VKx;IUXruYyJfDClkHnRjgS
zdzsZZ7UWrr?~yQ3fP$l?GPZQpgDq1(g{Hv9`J+w&OoY6-dvFd*tqqfZPy%f_o%i_Q
zWesey?8l&yWCE@V%T2)hgkL4IjOdl50_&Szh1X3d$7b_Xa}Y%JLx#(~uKdcH1J*lY
z-XbP?4b1|Jv0U4(CA&)@tUCAG%2k!6bx5uJRQzwl=uU9txA0x(BvZSF(x_sE>Bd2Z
zu!bM*sTk%tw-`XZqNex+<Y!2djajW6XEisY^5M9s{$Njb-hGC9^R1b|IL`m}Fe00G
z`qGo<=7P>iTc^hUzAA6a7FE&n+tZ6j92l?pTvF&FCvjyYuMutkO4<hhbA^SYwSRFq
zc9`CFmR?^rWO8ojPw?90YSqN*#fxYY2&`BO0?_B2)|j00Uv0Nt)=`UkzkTxt4KcXN
zyXtaw?&k-RaQ^a`%Wr3vkj2(HJq2qu?y2xXfCSbVfKO&Upg944yyV+`>nBaX=hdQe
z2l<?2!a?*5nuWepqUC$N5>s2;tfq_tHyF`){l*eZJ3lx2kp&18!ve6^6fr&tsDP;x
znDYQG9Q+FhSdiNI?|@-^D4$Gt`g{L&-|eA>9^A7}8AXFI(*#;?q9szL4T`jlBTC@N
z_-=U907Uczf6u(tt!1p8fHfT#k1;VgAtcKJJASN12R}I7D0K4<mVk;)vO55*gi_p2
zF4Y!Bqx2*H9$*P5>?PjPiPtNp_#~k8yI93>?Qo<?I%HG~O#r^BZZXps%?l0QLxp;f
zC6h3`Re8rZQkNh9!Gh&us!UP`yHXgjWJ0iljS~Ao|L6Qn!%CH|x0O^2%z?CCM~}qp
zn<9Ma=vC-^CSLuNm!>;8i0mmr-S#OxP%J-py-q17dz~d@_<!SYmP6=OksYd~cgx+j
zW4pHAjt&%c@k>7czw-WW_2gAGHNos{jnV(Ou6o*NBkuWaB>QayEMH`{9eSQN`Je3i
z&5t?OIDp_bschw4?JpeXG-mA#{bFt>DF7mt?f7#TR%bjzKi|jnJm-X3brgQ8eU5!W
zc?t}JS>ka8e^x;<<1*ScE#!sc8^+#o`~_6~mF~en`9OBH!ao;MM=)VvlwZp0W)6%&
z7wZKe=cqe=Q+&|j=zqjn_mA_sZ(RU}cL6{oa*L^X0TXhy6ekcn{5m@tEuZ|29QbA$
z-O`WCcdA)moV}yPLwftZP$X$)D;3WH{p~@f^(18^ePP2pxhU%^OA&l9Q#&?#@imXJ
zJlkSEWg%y_-=l%eHS9y|I_zp3$tl4+bxoY}ty9d39nof(YBEd<-^^I&4UWMRL2Tk#
z(ipMs!OMmK61p|Ja=(nlkcP}?wt?r5Ll|6w8SU6!NLy{m*~j|MYKYKo>_jr1xgWZV
zvpn80Y!wkIzj0P?BI&zrO5;Hy?op-pe_F=1|HokskM5<j&oJ=eU~z$C%&i@)o~h5E
z^0@5oFi0kxxpAh`Px)SCrj^olEMo4hwy*-c2W<e*7?2YdK2qN!T^g9O4L8UNReP<L
zSS2t~T>rE4EF!18E?5@;CPWk34V@$xDpJFB;AB1@%GY`LTjkYN4;fPLaW?!zb|=Oc
z`uQKLO7Ej*QY#Ir53p-!7rUi!^b<~(jySQjXaxEVzin4Z{?KsFO&R5@dt&nrABMFl
zm^c0X%-cu5<l3*e2IlAID4gAsQGB9bywQ*9a4h;yEi8NOu6KGgaTR{Ky*g6;sA&^J
z^T5p<W@p_w(dcegUM`_3ccq{3lRDgZD(Apgr#cWe0Uf+m2`Cn67=sk$(gKQxfE_u<
z_w05UmP{t<@4=EBuJXU4l(>#N4JW@2{^{<Wq*GllSo^Q2ewZJ6@Vb2c(#-^^cBj~F
zqG&%sg;qDOIGhOS)zu$;raZ=RlabZmuT(zw;{zwDhv~QdGQP5JfIOhvEEQ(|oi&_N
z?dqbn{opTQejn2K>So>fR+I?H7AIWI9;Q7Exp&C?M93G)h%A5KR2t!y{tmup1?zOm
z6}okxYzG$<<|(`95yQ!4C#RQBJkA;phse_gOIwE;-ceGiMz$zQIIfLzZbFR#P%4}p
zq-`6xz{L{*0R7#T^FpXPPFR#|NCP~{q$~m*e^Kq{rHfkNfN_RgSC0L2_9@k2uUW$+
z93YIDv~54L!QE6%uf9*3BU2&#_q4j?4sgD{S-LSgI?5|Xl$K-|uq+o)_#`4^6ABF|
z(+MKuX<TxJdtx#4lWyjrDktiH`lIbm%B;BG8AThT$$+DnL_X}xbcvN-$NK@T3Re>K
z>g2;uV_}n%OM9ZWcc+PGb5>|nLgEKY<;Vg}J`*|a2R^J$5-=Ku=epO?(%k`<0IwG>
ztLo8x4+{OH>AT(%t$d(4k+S}CxSEazQzO07bYLR==OPqzG2SEE>(>buX~P3OxW)He
zE&T_xjJ6koS+8T&v#c5%D*5-O(?O&(9IqP=q{Z)PJy4p1XF++JZ|DyD3YU1(=njL$
z$k(%#R9R@GzZ&n5CRtMGV<Pnu0&ZuqTw8ILXD;Ezfq8?oJLk_My3O<pY}tUsuzxTa
zhF_@9mQH}D8vm$48;H8#xgclU{VROzOZXS<;F0{EjaA%4+QnzYUs9t|&T!P;(vBH^
zdZXVu@Z6S{V1{Yn@hn4+HmOo$xWWTMCb8rbrXXiIw*HbZ{=}Y$7_z*2&E9}?+VChr
z)s3c`55yV0Gb4BzzwqsuW22Wha`@j6iNHR0DE1k=IH-DC;QMjD$9+C(UF~yQA3c)N
zbgOv^Y2EXImDdn085k1%3WX6-WCa(~MHoqT4Rr=!X*ZWu^{Bj}rCHR<XJvnUnFGB-
z;8?#%A80nO@q|fq?QV;o$V7qxT}Su&kidDpq*U(5!}~Nwz%l{=9eR^Appa!(iXM2`
zGtNu2Ry<e}Z>#D9n<0?g9~a!D5&n^n9Dv>gRk8(ZEVD2$(h0Qu&h0!Yd~O1#Quu&1
zoh>!4!i`k+HVhx0PAgYxr-qa|e>NI5pIY+^1V<4tD;sQ;CT9i{CGUx~6*CHEm5pJJ
z=RrhK&CN9>EUuBa=(3B6b(FDO%q{c>=|_1{CTPum%S62i;v4#60dw;nrC+oI=zedV
zxudZY-U_Nx2;bBaXY|dyr-h<CV%QY$n%4V#{lwI;dRO~%iyzajDbw%`;TRF|C(`M7
zEw^$S7L>jEabDf<{o}@S{RCOl<bn+eO@Qzv#r3{$6O>}hXz2twDzu7Uvro~G;u=fo
zJe$pCCr6nCOWzBN^c_A6&n+z;9U+peQNERiTvZSc0W|^KVrIEJ9+MWkM}tj?6e@UM
zz;raGR;gaoxY66T1!lviv#V%GmGU@eB5sFw*V=_G!D~==>=1|oz6$%Qm|;JZ>=-7<
zrZ&7twvmz$@T}rG?l+O_5QIYf2=@s8-YCU}PH<l!`k{6zNVTpOR$mdvn-=2&f|_O-
ze~*0BtZY<jiLVRCT+UdEi195xsbMm=HmP*t{_<Pe!TzdSMmqgL?Kt$Eg1Nggr<#Gq
zD}DiX{6O0P<C*Ma?j!SA?p}g>vwB2~HzL?5zO3uln_^b)R((C;E<N%d&yPF5_!Vo`
z<exc(*D5gaqI~}A$ZPDdrQMyUG{Ne(P%Hk60Pm3z0F8NxOROP+4BV84JHD;&*F3@E
z_LRR1(g9H6JW2_HQwN+uu3x`_3r-{WRf0H>>*+31pJ)>YxUuovU03g-;f`D2qIf8u
zj=eB*5(37Fxi8WciQom0RU@(Of7^5p<XQsm>_&^w%>hjTyc0_5&Oa6{6h#ZUD=*&{
zIQKo=A&SR!_-h5ekyi&;Qw`n=duR7W|5d*>P+X_)GoV$({$RQZ)>iX4XM-bR#&6B+
zaJM4@N?CuWNxfwN!KalUFn3O!<<cw*R~p(>4RdX6EcD;6QmJSmRdu%lwHL5XFpQ*_
z+P_+YKU(1or9RW%QdEm%d-q6iHjTB+234I=jaMosX5j7wW{(~7s&V=!=QcLxF)mmC
ztKH1Gt&^_#T}}NmL-CJ>oc~RG>;4lYHso@^fdqFKWMyPfPN(Q43T?$%XU$zZcM`)V
zAar!0|GmuvvSW~qjb~x_UfIUpxZ?<}TRe(saNg3W9{K>|7Ftc>f<T~`2E6vr$AXGh
z$Xt9axQQNo_?iXPi%z>6WI@4!s0T-}rO6Z+vj9=?rI+<^mhlGf5xtzT%*4afy!A;h
z8@o>N^NDnHXRwJ|DffHn>Oe|j{(J!gv?3O3%>BIIRHCgE3+xL466%;Xw{uL(dV?^E
zufQkONs<^W$*QGut=l~~s0|JJ0Bp3z2r5SOaM}nGD`#A(e9f==#T12Afh65qQK*;>
zuW2m1ot4s7UyM)TG6z42PF|*I|G8zCDEj#BEuXzNzrI~cYK;FML>CA<jTpqWcrNGQ
zW~%e;4C~L;^m^o!54piz=VP$2u@*?j{}l&j9}_Gapl8i$`?&}<isFsarZo-+`&yBJ
z9RuC8GMJe8_TZ@|vH<}n7gFVA@S^yk4LzK>^-86v+lKH>`58vP#Ug$Xp5T#)`PW32
zDP4Rsj|3jwLDT$?JmI4-;xGKbfHn9ajCXeJ_X>|oMR1(+uu_X7U$S!~M*bdgGdDd+
zpx_Yr19&UuIZ*HoQJ&LhInLi%3c+bC2HE-$e`I4Gj}v`@%9k=**y?tftN^)dkPbGg
zq$s|04k;bNru#i+zV`6X+3kIHrD+;@l}jRK6~(W<`iui}z9!4(HchDIwN|@zmb*{>
z?BKQRmXrRy@qTKZK)gx+nl4lEKW=(cIPz>}DoPL)!0=C0Wuv}IVDY{HmR<+|KcEs>
zk9k<{z2Ib(rvx&;l+%m#<P%WyG=F}ux-lZe{F8BqP{cytti_@FM-><%_`B`DhAGfv
zq}5z1SW|%G8@<4il_!uf&Vmc3kRs9~9L{aEmtM>#OwAx%gFYkrYFPdWDb#vz<clp(
zum0L5XOlUD5|&w=3pN}W9JGH^USp+Xi(r4#ug8|oo<z`C9Y<Ds@bGi#fVRrJJcHoq
zMhxsSPwL}bk(|;85G^woF>#X0H0l5|&#`#^wkl}bS;ji)v49Rh9J31CZqmncV%6Mq
zN>xvtEN?JbT+qHO74$CGIhcz%57d1pp$q;@XZwdk;-6jF{yTps$nz(Wzz|cb4C*aV
z<AczGkO?JAWCDC-02iQyA-dh2&_mbwm)p+<cU$U*NJ{wGRaXc2BddzQ6LtLitc+KU
z1LQ)8_k`)C#IfbN(87Y+5lAVS@4JA0W4d{9lqIM%GI&1%Cb0#=czazz*C)T|6w^T6
z4fI5tWH%Vjmv~|#A>umaixc&7o;|6ny?1~R0YkF#CwIJMD$9U|1vmdY$m*?Yz2s>s
z)gP;M>!$6eL0bw+k>uk7@(6rl`=dK@#!>7;z|n`%cmliTR|_GvcEFPFNh&6fTLjY%
zOjBW;I)m4a#Y%d4(P_tMUrTN0iJ!AXa{qjPafT07vCB>m3tb!o1qy?r6sqrm_DvLI
z*3>|*l3A$(?}6FXmThuoe4km*3R3+?V=TBJ-IoTEv0nj0pvz8Bfq;7lT%E`I)II~v
z-_&{~5f@yq>P<V9epytSx!8CPBSBX0>QQjxfdPMB3yx+2(&i*VdvDn44UKq<g!S!%
zER(tArBQVKz=R@G8#XG{bXv9e*J-8Bxx(gl<lrU(JQoM<zfxCSb#*uiQPCkiMR~lM
z_u?#5xfSCWI+SN%JZ#lsx)0#=YVm;7NcRo%T6q*e+fSXr5Cpyi?+05p-~m?PW9TGc
zPeip;j`1nD@Z95ltk@grQZ6R06wWJ|CZZQ`E~&j_Y`1yP^|bZWIKJQdt`3C-g@t0E
zS710FnQztGx~W8!xOlp%D3#`)jmCCA#k+m}Tdh5ZY$S)dCA$x_9{{gL=&b%V3wE5N
zZ~+EW99Lk5QJGW0u8Wsr=PDE;q1Kbg`M<KjInvxN(h>+tF`vQ8I$FHx=jz}USQh&R
z2Nw^#do!EAYCON{Fy8~}$1)w(P2B6vshui;{cx?0@L*z6W$P_j7`f=P73wuqTpUE>
z<TD*El5pe*uHG>D!92*T#y-ui%dsT$><?RYyY)h^J%P~$w|TDR{%kzveg5>c(yy+l
zWm{^HZ4UFIxPhke&|)&3DI_PSn7Eh$1~@R(_sh6f1;SMxcL?>V7!b4=2M(07?!7W;
zPMLpu-nrYta%v$Fe>h@=6K4@_|K5gOkr=zb|C$TkA{8cZZ}DXlxcM-oCs=K8Eu}lc
ziD$Vv{^v}F`x+VCEv1(M6A-4)GA|_-Xaq@=`RL^~TSzac6rqVT6pg(;OYAE$5V}5E
z5`$HRBme?i!AgzZueYu)1?u2QULS_WeDS0Qr_fG@MOU(OL1(FvZ$IqyfhjQ4LV+e=
zm2L0hQrSHUM2hX3UoWKFEEK@cZwZqXYA-(aXy~?xnFC4i9x9VYYGVVZYNWEO=NFMc
z1de&f<DP|q3HM1#-8q-Lf<T6U3pN-jD@&47&l-}fuL6V0UiYKVpk;v>x#9zh55`}M
z(kQEG^LK+}4v))*OI}Q`V0{{%yWhEcWo|aC>>C=%?3#^Fn|IHDmH`n^Hu$?ZqKHio
zL^N?=z9c*9niBuf>%Jgaf-NoV;_Xo96Qtb)?p!cZuEIir3-o9ZRRAC|MEWN3oP=Gq
z11>`_f<48Jr*+k<o<Tc`?(en%WWF9S&4FM98*ZTwr|t54Q-(3;;fi-AjEn<`o{FUa
z!WhEl{L!rXah+fay2>TGYCvq}a~H?ohi4^540PbgrkwaA{Ux;u{awSdiq8(<Fd*G5
z2nNlmtHrVjtDTFmKX>>#zc6IbW?R@^Y@OXM86kdh!sL5Ah*Bp+D476y+}boQXKfV%
z)pv>y;`+%O>4cY&!Z>uk7w__FqnG>KKoMah(_3QfRh;4Y_o^@lkq><!9o<dqA84_m
zaR-0iv%R^eT$m7%51!2%?D;t#Hh}<^%ZfWF@<>*Pflszq!xe>v^~2EN&;W6MKCR;3
zY-R6_?N+k?dZ;F(;1%Fo9Uu-egL~hhV{1bX&>JiYPn`#kMsLB;w$1ltW-E#F(X&C9
z?<?OP2M?WVjh5>+)QYbHg1=gH1a9^RrNYweMr*obbR)pB<{J{yyaztC;Ivl>j)7EI
zzQ>P~X%&gUHGz~EJ((~c428v9n!7z~d!J6HJ2~VH8gzBy5YMtK&b3?KW@3NKx5O^}
zk@va_Yf`^31@;ok_1QZ@X#tsc9U2VUs<Kn0l#B%7PubNjUSBGPQRUNqFjLAfVI`4a
zZCZXE=|f0Sp87=VU^NDP&cCGAF%UC_lBGDPpCLgzF0A=&GIsyhW;SbZUgsxsEoL$;
z;SE8$+=LWAv&Buy|G@fz_3WBJRWb=i2`WatU$yO}d$mEV*87#zO-Jrs-r3V<oJ?Sl
zlB-WH@!txrAh#TLf(Xh-{l{n9<@4sH4#*Tf9SNFq9VBm1`RT48HeDe(n|#{C`BveN
z`d`ncOj@NqZ||8i&%l;$f-&aIE2hg?OwhV78P8k$6YpJ&Gx=FKt+r<i;6dMHq5p-`
zko~P&nY%UZf_77P&HKKin64kIxFC^<CvoLI2J*8r({p`FFbnW_yD#qh3^vo*W%3DV
zuPmfAcyHMoF~w-PsTMe}%INpbL2L_sX&OTG!OrP(Hhl@34}C0N(`a_B7K+gT8A(pm
z$$RHH&yGvoTHb4aInJ%P18gBiOjQ)=MAj1CG=|*bXMZU3*`QOYEqnY>hb<%`n{^sI
zN;Ftge~O5y!^Y88B~{iM=f`m8{H-l3m3Y{;du^Cxk|p=_f3aT<$)E>9UGlLNa3<rx
zBh=-)Orm8c13kDYyX1^OG$Sxv{v#<y^Ri^!rTN9-h_Ezhxo+4`4xj_j)avX5E<xQI
zTAg|z_2?ZFuMA=zx|qT{2i8tD30x;({iNixJ<{#EwS_;&p$LCAr6qhUyocDmX|NrK
zVpK|7mQD#nM>tDymPQ#QpF>e^QvA<HUI3#8iYkIs8|0bCv&?k)zVM9F#k3D1N8G{j
zi=hM_x9EU$Cm-_on<!_I{V8K*!K@6s=Hs4O=Nq(DzSoX8I;c822Qs%*^X>mY^=LKl
ztjG#-vHmB@Z|s>sU8&<6%m;O9;Ysjpv&;XghP^jPuXBJ0!Htpc#U0gmf^`O9<jP|%
zo?~i*w3;5t9~l8+a*xbA(`+;FiVMseqD9A=fLgaO&1-1K%2O@__r2(2>q^_bh9M(|
zBdgaR95#K|e0Sz+0quH|?vs)x#e{Cy1(U3<3a8~;@i=O=_W90(!|GtULodCzxU5RZ
zM0dSrI0KB(?_XG`a7o0=3Gdn@D<-~8{{^;9bR!3UbQH04;+=D&&;+FG>EfvHsRn&h
z=`pTVW#;WT^yW%t$$TO*@Jx+#D%N!x-Zl1A>N+gE`l!FwEfwBV99~4`SQ_X_CWW2A
zK&R@D?kDYYBZmdIxcGJ(;`W6=?&(`^^uK5JlYC*1etz1cF`4d&)B@3ae*VF{XboIR
zA`|9o=jupZA+Qglc^F&3!d527F|e96@#kBup4BtkGM$8oUAHA4e<2MAOQq`jIv_|F
z#sNzQ6dR@rzVLXeubfnXY$sxas)2n{rSU<QuymHNXebM*{X>-y<4&Mh9q4oGFUSYR
zkW^{wp0#^t0>Q)a*AB_3s6IPK9Ssf%IV0^Ffw2y%?7;PC)y4S@s!gAk-MJfyySlH7
zS30L2-YQO#to{f{F&L^WZYjMR@d)pUP#n;>q4f-M(zc9d_tlly`}NBT^}Gd+iWlkc
z-jTW)U7Kbpb;Kf&lvglVaF;3$8H}7E0i&d%hz_uua~D2YX=)Z=j3NOBDhzj?Scth>
z1<dh*gNx-_&sNl{;v}UE95p`GR(H?t?$}4*2v_4k_j)^EsIIPe_#!MJe}S2zwhX8%
zfn|drPslamsRw(M<4hDo>DVF~&~Hmr2;(SL1IKqAkHcv3`OH0Kbl{J0ZmrW`r{RZW
z^p!U_bEYvIL?qu^mC{AcLGP^d^7Tu}*=@}myrk?i%#1|+#-1uS@X@yT6F)l$<^`)C
zuRgzDzVf3nqU`pZ7dXpTu`iEi3bRCX{`te!SVpFk?@1wrox_qdwduS*dPgl&D43m0
z+!NQY;(sF3v@Laf;G1HD<ZNyKh%)WoL^{TW<NO?*(qns7%LXfAYj6;6m~$r|l>*J?
zS9WAEl*Hs(y}55a<qa8pZ`u|^TBm;Eq`p}^TO5z)+w7R?cMCN>_O#j#dG*g<Zb2_U
zh-WBWi+LAKK<XW%y>!P{iwy3Z_3qyhDl@t~+R>%)V^fY8A6Lvau69$MF?kwZ@{$rM
z^e0GQ=G@Am@nGHMs5c2!_Bxj}R}S&kw+4X@D8*y53|8LsrGaTj4YwlaOB!O?oBBzW
zm6ZW6$P~GEb=ilcCir%KHOb$SP02s&63Yubd4zoE4mYTQ(xsoW7r)G*OLg1FJPSDA
zcvC*G!;>r*8FAn-1sDG}6(Yc1ii)Ln)a12j_3#M`>Ne>5s8eKG#0x)Bg7u&=5n+TO
zty#&Rx(a76n3zVi6?_JL&*xrmYqopLXRpijS9MG2cTMWl%5%=3-B3;}`i=WQd@YUS
z4kcH~Yu5D~E;9$IPb-o5QrP_mZ^IiV@b-4nR-4+q(nqeFAu^+XQ%%+Vv9KZ)9Gj)T
zkXQh>F;DmO+IPg>l9k=~rM^#1dQpcAV}Y;Z$@VWQnLxBI*O#sT)yvc_D4R`CvUBu{
zybc4H^whk<k_{n=B-ORUG->1kYcrS`R6jTlC+X-IUiBC=-RkQ{wKtNTV2G)^53Kn0
zff~vt`YiGKioUsib0AAP-2<4vHl&{yt{w7w`;3KGKZ)lr?-KhX`}H-|#o0uz0g2>;
zNk^uG)K&)4(40SFyGerl7=-FftEKcS_!I!<Mu&`qs((iEw(c61kn%kLkEn>@icCmB
zioIY<V~M<H2rMA3%f;IUG6R&mzK-jJr&-d<$Yof3no{l51~{%({%)4guO;s(81Tu1
z2-}l#ftl4g<DMcU#Jv@{Hvl|U%&|bZ)OmfdIzz1@72*I6O5i75I#if2+`-52-lzN{
zs+F_H^ioS(M_PQ{OC5>yu|vxhrjb2cQq_A|XH7Jc$6dGl8b!!H{kXW8Te2wjHfDU~
z-O{Tr*4il_TgQ_6(?9pYf47kQ-!&u)ATFz7=Q*PHyUsJZ`n@D3nX!KnkGag9ZL>)p
z0k&$`6K_)!D>Qgvv;ZQ&w)06sY_OA4Oyb$6&%p<$xIVSomsZsQQQZPN_-7`qzUiqn
zBn5$$WQi6ONlaQ_b!J)UR7tuoi}lhX8_$I{YDzT;;w%)_$@-05Q_nSDoRp_u#c5|r
zIn>;fX^kRo>5<4{>X4FM{`UA?mSA1K`g=pGJO|RQ?f)(KF0z|J1*)mzEQM3kI<JSc
zHqIz}a`;V=27~3LtDsZ}Puf386GKW}n%4KBVNPI&@U@?ta>d>qjWLoi7<<RY#XdWV
zs{3LWt`(}=U6RMqPva%v<wdtt%{HYHl#C+j4XUKQeS<4Xam9lp0e`8Ld)W9_+^cPO
zzZ9Goyk1lW7uPdQhlTk&fBhGB85kt66{nw|c;<oV56#xy;3O}{<IF1sRc(7%yR@`^
zOi!a=r>3UQ+^&W+Q}_P5aD-X`hfZY_2k&zmO}U%S1Z<J@@k9!pBkLzv-gbknOgv+o
z85z<$>zwMG1BuKhb7Sjbjd$bin8>ZeiA0~~k^dM&OO*2E?w3conK{WA<VGB`OiP6Q
z+d#yO<hRir;rB+NZ_aYFei~@sFAZc)_9eoyh;{feskuJA?$nGzJYhwh&CodhfO*}|
z+A*L^>hx~%2CGX$Gj1|lF&!~3&pw6vF!qtO>YdQ*^CEPUm0~B|BZSU(??OBL!bxYR
z#PCIq>d^NOa+;MweQM+ikMRG=MTiSuc)y_0dnMVkICZzGjpJ&#gYc-pv!^gt@ucw7
z&R&eB0bL8zYjIg0Ea#_qPS>_I^>|pxr7w&MPw7`!ljGw9Zqmz16lr&kj1Xw$t8~DH
z&Ikk#50750ed@r4=gw?^9(yk=1`0wblQ5EuD<mt6CL<$*h?F$=(<gSnD^Z_g`0XW=
z>wjc{Z{C<XI^KAv6#T$?^x4*QrIV*;s8n`)p=R#h;h1>3{U%&#OwMD5PEAcsL`@yF
z+!rfNOSG`J*mWVo$jJEeY&AJsn{DyWp9q_Y{KbDqn?_aT*f#ftP^SsW97Ma@#^&TI
z%Bjm`PM6bNSL<_&IwWW?{&@>78(8ue+x|T@u_3jEn?kl_Uu(Qxim=t*-2C8l%iK*}
z06Q2pgG*;Qbm(0nKe^_UCs>sl*a1XeT;&uPQGfqIXDRvG^<_AoM_8nhF?B(+qt8z?
z`u_7JY*Wu_pKk%Vgt0<SeEZ%OW+?x>2f6(!ZZL(OcFxNx6}aPUSVZxBBX7Kpp^Y)q
zOcxvb{{6dxl9H5y0yZBdK|*38&#QlKE9>j@oScN|4_|bK5Kx`E=jY{hOihu|%g3N2
zhO@+J$T*+22cTgsY;H#W{p+TZ!i@rF?e|O_7eb=syV-JY^y%LT^~JR{Om%hjr!C&3
zI5;@dwJ)g1&8Mp@@t9Q8f|i!d1cim)EDLt`_THnVMKm-tXq0@{`}{m!aSQVe3M{g`
zn;W0s=_1zjj~|P_Jm-6Ya7dhXW*<>*rKV)Q=CQ&#3eeZl(UHv7SXk@|jT>6!81-M0
zqWD(hF6gs~QueDTq^=WX^w^{D+i#7sGFiA+P3Og(ZYc=_-Ol&l8o2^eY&T66ZeURN
zm=*5W?tQ2wOfw)9iYaNM3d<z&EVk7cqR4*mnx;07#{;n?a}=D_@uiVxuQ6RuW4^0C
zube_W7=7&csD0S<3VUM_HT=#m(>HD{wSGGs`mW`9wY9gW%8aDP-iw}9u73U}gwE_g
zwe4)I_u%eV!da%usw(2Uckcq=`8qk7%=gtRxSbrQbs;i5yrX(dERulgp+w%}>?h^M
z8b#WBM@OquJXXWxGIDZezZ%`AtF3QXjlGY)xbVfN;QctTO&Kl6@=j@MzKsTsfux)q
zX1-cxL_z}bOCKLJL=261Or?3>JyFp+P$8TyMRM)!fbhDaB92Sr$XgUHNGGHbi*jD>
z<)2nFzrHV(_H~?EP98C{=ogK$(KZ_tk5cxS=cC8HTTco%cS{WW7X4K3Oy{cImFxZR
zxvpzBaHal<R_I~g_rmc_&63iN>q&Ji#PClK>g)UE0YAT7j0=S$oQUu=t@$SB<KL9D
zweBj)z2Z<3dwCP_#e7km?yZ|6v6oA;gft0PuO>P^j>b|Cx4#J476BRSuhaz_Z6hOs
zRFz4KCb?V`gx}q0IW%VG0;{RR8Lu=u`pNS8dh&sRf&P&ZSxZaCsNR2B4Q%gl-(V7S
z);ljF?>mg5engCX3TkSEVrMHP2nk6^6u}tjHor@i$)bskg|)R_@ni)CY#bb()Pm+_
z>ibX1WOQ|Jv$3(YJ5zr);Lv6xMSLhM#1{4byAUak2#bp1)9?FHZS8z^U<q3{U8TIV
zbXDD)WOMTIl?^MFn69xgAs1IU)6|4cA;LbiD%60R{!MZ<nuGYarP+~Cr>eR)3jG>7
zT3F!<{Wm)9RaJMPm7*|n>PWGmZ^=r%9AF^-W*BfSYA_%BgoaPQFdXCcuq1Yjd0~-x
zp>3JnCyeLCuaC48xOXVjFWUNP{~kEy^PFMx(92h1lTAhm40+6${rOgejgQ}5X4C}d
z5-)c=O!5t;WO;d;tsop_yxZ*UwJ&BW{IC7+?~0JW!McH|aV^V693GdPj1}MS)}-#o
zDlGJnT9lTK4#jhKt|yenFRZ;?VtQtVUPvfy-ofrUA0<B3L;UmOZM!dPi;Lrbj)=Y@
zzbaM?&IgE4PqPDNHG@#*TxRPF0va0$iLf>DzDqiOQfKk#6zadz?c`n++VsgUs{O%}
z-Q|ow#?>=U#Aw9v*x$VgPwd#6e%nQ%w^z_&5Qk#jYm6mZiE(Seqf3&aC35?%ZSPs{
zbHW)B5m8|`#d&pW7cP8#&O7$mW?~O+B*meipqOs+7hhUlKCAumMZ<|YG$I1m?`(2#
zbhKBV=2ehlA`?pC?y#o5{y^Qy?mP_yZ6<k#(TqptRWTJ6+#eyd!Oh5jxOiyQnTJLZ
zx0P&}g9@tAJ^3)-KH13T{U%W&@S%1TI|hw<J2+1GXUC5@LS8H$nP*2Q#~c%UO~(7V
zb4{pp8s>=N5vO<$`FddlmmKEeY8c|LlY36CZ%~|vtFv=JdODTI&MXcM4Najo8@oyC
zti@esW@d<fOPQW?lBT9~RwHRWbw`_1H=LcF!(wCmk~j>BC@CX^J$GmyJRqT?qe~I-
za!H)ofgme(ec>7u8k$>NjH{f)9*~<$KQHF7rT2H|7ZozlezqY2jk#{wQyj#_cz1oh
zp#5Ak#YDbpR~o<F0>~e!1RNH=WJ~Kcy0KndUJAQ<-u!Cym*5xS)Q;}5SrMhJD#2jM
zXO{A>l6Ksk9ct*}WLSS)JVOdp0+m!b4je?<{T)SM%91^B6pm2nIYOhgd-UPwGp{>B
zX~c8VA!AEjPMntPeg^y$L(HPm1ut>v?l9fLbzW#kSy^2*{qjLluilC7q2t`_vsU<M
z8KOS7A*;wpNW4}ww6$e}r&#%X{N7)9P7tyLUDs6JocV1+T<p3$Ke>7PcE}TxSER6)
zUl$ZG&vO!oJDu*C!p4-5k$DDB^mJza3KQYGPEe#Bk;tr$Oe_zqhT9=k?f<kZ&#tXa
zioqt1lz<Q?sg>LH%U0ADUv4@QVeNyGafh*j1lPGuE_Tz?E}dgR<|UunJUuny#l%sz
z>`k;&X}25ALL%6rPc5{6-;`<`PT?_kTI|F~v}wCjZ?@DOPLMCkbUU)r`^ff_aZBU)
zad)>20~6E3l)IFaKNsHaL~^x0Kk#jTjlxv$vAVh%1@Y?N-^BtiA0PAGxt3^$bRySW
z)QUT0Pi(V&v@gQ9WeUyZW*f^(Xy(f(9`n#zyv`>aiP^NaY5v%LFfg{z5Fn<Zusw2z
zkS-|2$lF`Uf|ugo@pj};GS_lx-CQS>CgV+>j;^6{3MG0gi|N}lb@VJO*o1_HVG$9X
zaEWYs#>vImVag*{CeT_yDOlmYImsDYZZIZEi}(}O-eo|jBQ$BT&x}IruWe;VaMY8P
zmlzrFXx#5~XZt3LvGC1?0jLu6h)Df|hz`OGN_(5V4!NJTBj6j~yLS&&K8BjdQ!nUq
zTF)I66=h}V$B&6Uwx$CL3t8-cHIgDzyQXIR06%%GO^w0dKi~1VTK*E>=6pKM!Se?n
z^D`f1<efjeFSD^e)a~VPNuXSK{Y6CGT|jdTrk9bTOpGOxN+Y46as0ZOLS$#|jInUO
z+ZH=z;(w>wdJGc*<r|ub|0Tc2?%b~rSmX%U7qsw+5R@-=aXm{6cQWk;_)Su>(5^&^
z#dExUBqZN(lSpCHJ?zWDm1r@JZEhC%q*oc31sfKbbXQhZQh0c^c*BVd?o*_zV?R^z
z$G8&lMg(5Ib^5$l*@jIQMD+EDy_mA_(<k&k=Zp5VST8THMW`1uYwu*LVP&fIiI*1_
z+fE#i-hO(an00zSTkNbcn3<lDOk|e%GqLM;?QT-+mHbCxeQlI!qselU`IQR9WbgO8
zDZ<4!m_q(_$<xWIM&GWw@n`0F*lnrzYA*&%%__+yE|0#d6I!XT$Lf|~>}_60#g$oT
zVti<75|+7lb8u)#H9t8yIepfxDOdNCUckqX3|C${I;7rze~;U2?kb5Iq3iMa5rv!_
z1tyfu(|0q~FouT9+=DVgTU#4ZNb@1(AjHx%O0wj8PEHQ0QeIU@OC-()?rj}}5(6<Z
z1Ei*=hDAqXAjqTTEUc`|t}f2t>-zZkOwSnh#W8hVWK=xwl)&F!#4SR*WMc^WEE+0$
zg>qSs&5)#6r0wyigHS5XToZ#9!NAA}VGU(*aS@Vh_hhM|rjgOe32sbmY$%lLuo^c$
zv@Z8A$1g#x;_$067>b^L3*N+=KT3PvK6}!<Ka2Q#m?Uj4--Ze?hM5R^VR;!1>P>Qd
zU6&a;D*D@z8~rpJy0@PoS`P;A*SoG0LUF69s~eIn9eR(R{>?6V&f_B5TUXvMX5fU}
zq@)}a_up+f*_-p(HT_no**`e=XJ>C_^|aG^d}=;ZZHP#!@vQ<w&b>E8sO_kD5{MLW
zf4{^6Uw^}IQ2Xkuj<I4PBANb5+P_46M5UkC?;3koHUJw$O5vokA9c@WkqPDM;Sog?
z)e2T>?#7W1l-*Vl5k#>2YGZKQ)o!t1>5gD`?ajT93)|qsL;fN=3F+%o_q`G5o?m?X
z9{c^97M;xC=WOiF4J#KTvlpuTwsvVIeTnjjlH%faXO4N}GX6%}mY~l%%JL2V=U#Sy
zE`QGYG}7;$^v%>IDJ!H+Oj65XZ`3CpsN0SGE&Be)YHH$b%o;a}4`r1J)~{x-3EN&2
zh9I{xC?8MH&(npY2(zi>!nSQ433E}2j?4+>T3=C(j*dQ@r&!*9T?J_rs_r$yKDKeO
zEo!3{8VYl$;GtmUe-!8#O*v%xSInN~r{lKtdFi%L<Csud*Y8{0h-{7LNAd`v^rxH!
zH=MEl`^^R#Qe4^OPjyZF$z8u+O^bb={~3C0IOA@S<<1bQe==qaO9bc4N*h6p{M$4+
z8zLMBw%f;#AN59B5;$&dca2J(SDHV+uDP*Ycbt7W>9ICv{S60J5Sy;M_sdO*|K5N~
z8iN+Os$tRDsB!;$NdrXzfdG}TtcnW$$H%O1l`L})Av>$Q7om7ubnE^5_j*-;0sZ;Y
zD4=m4##iV863-5%D@@-sbXC2asdJRq(13=J8rf7%6sV^QGn9;{jU779C!z)*un4Jt
z&q;9LuXc<|J}P$~&Ri};Fdz^RW0H-%aU7fX<#M;F`%Dj32EJ?_3A+Eeb-eQ{61q~_
zM~@oP>1)*NCO=8q*sxx|JRA`ixsxEP{J%K*`0-<8Q)=EHM@KI1Cj!Ol@SQZB=NpF?
zdbYO31LwRCAJ+Pye%5%4C46JGc}-s?#ym)bj-US>>#YdPiMF#fDhO@*`uY^&+sm=y
zRxS-08*YsbQD6R{V`5g_5Ib$3HPw69->(4qS>)<bpY!n4RC1|F+gy~UepbHNSQHQD
z>%4*jPdkyxz=LOf8l!fpj4Ogds0s#*%*<F&Q-(lCBmUyzbic5;xby6AeS|#K9aao&
zc6Jt7%#5T9Ot9u!{XX(W)&h`3FZi-G!g$PkX_=Y(qT)qgt&OB(d3t&xyPcDrU(fj1
z^r~5I)ZSff&evCp$@M`=)MdSY|AC@6;&k28(sB)Zz?NOF0%2-u>g4X;Q>0Z;Vb+8D
zHYsW8@?>7X^ZD%}Xtk*O`2YDez1+`9&GG*CNxvEsF0ZGCryX-wG@`+ET`G6{Sw|qo
z8sUhU=3V<cgI*)mZ>i<w5m{!CP96VB?K0E+4gM+i;zymLq9VDdH=mT0)LwGi1%_d*
zeFqfOA0{@2FWAEPH@3zOg?F3ZKr>3u$9HQeiK7d|<&|ECYn;9qYm#0St{S60|GwO8
zFPP4ejf_i4ncfm<fpXz{V{=Oj)n5%ZF0S<%dsX@47uwplV9z;S{&R;as0}Zt+F@sB
zhbUa;;(W96;J_u_w_J2GHEX?BMpkxl*Z*4BP3V`FyD-<_o6I*Z#_N^v=`j|xhP%4Z
zpvlavD*OHW`6r{M+oYtVdw>6;BdCQv@Ld|HJ32Zts|f7A#=lC>%DR!Q(Z4HirgSg<
z89c21;o%jZJZLQcFSxG}j7A9^9?qsf{H?FA7t!gVk?^VjtFBFvq=f&r^`E}lM$g^5
z&@n-S<>x1c_@JI87B^r)F5o~7@XG$5l5#VsKUwbI?;3n}pT}{oxnKt`_rdDQ3ax+u
zS=;sH?&8)~R7M6h`_p<N|I;p-1%L;3_Skiya)3WXckjl)o<tT85HwRneQFz|t8qDA
zy>I*X=l<K|<Ymvc^E+@5owlZVJ{i^_@t%i{3s!Po70WE^H8HX)^$wekD(nVQQmANO
z7WQFp93DPx^UGAt5R88$B*Y6Ft`=%eD{k^(vd^UbLqq18nwl!v8r<L|@z7~OqSoP;
zkglPj-Od;GIe6lF5)z0+1qOCr-h|7OwrhS*1>M}dvBSD~zo8JqJA+V+xw^QRNr&K@
z3VUolqwY%;aC(p{GVT2K&|qx#ky<<X(K$0RHdQiy!%O*p0ueICnVCb7`0pj2|F>X=
zgN$1C_@u9quh2aV!TtXb&-7~khiIj}IG-vv36-E-aHNpMrh6gzC@KZ-7hX!7u9AvM
zHylu8z3gWYdskG$zyKG%_#uWIHeIR2qbQN^7zA;;H8=UsmQM-{fpv8PTU)>DZh=h<
zOKAM@2h=z>JwO-aqR3J59^(iI2&km;Vmv8(x;O(*GhDygs^j~2QZ+R->>D@a9UM64
zuf9Jqp@Yf`icY}O`u=WAoIT7`3k3s?uCcKj@Pkl_O6ltAR_{|kazRHPU?(Sua^sc?
zI1=B!f8XDR7)s-R!DIY#5kv6NBh&4lH4B1-<t@jQ-p5-w^78V?4uQt+l&)>KSf`}I
zbJrO8KJZKMAH=ul_hKMuyLtoL1{&C|zCJ8C%>qBJevju?KI=ebH)_09R#vumeB2Aa
zZBSEwecC5(Hu#p6X#yV;O+-|5|A%!}jqRNvBCWSwja;GIKdM9LuM%S?In{-3(tmt&
z1#eyjt62s6^mvG?tpY+%RmU5w2QkYj2BlVTuFasSu@Jx5L%Yw(X=P($v-Pb=D~hN0
z>T=irLA*jYR3blu@yPVxuo%`k+-ci+<hoXol0xoru%hSz07I?Up_P@R;~1w=V>h`u
z`+or(*Y65?8s3JdqsLo6y2_q5bV3LlgbIEdzQd^(f*Wd|TAvdKkJUj^kL6y1{HaZS
z7Z+Y(*ERem?<2NzfPYf7*(~A1zz2Z_(Fc&rAAqXqL7utK!C@(s8DrGs5smV!?cxB>
zJgPEXz>&8|dvI+ua|jm?PapaWS5HrCR#w)FWBF*xo)wozhBdaNO>P^+GfkdcB-rR%
zu(}*On_FA<OvK^SP*a`Rgc9B{gNLnBs#|yFFOQu~J(>APrR4j;3>W-^CVpp(&<xpq
zulne!iJ~C%D^u8$906Llb|hf4wzW09GuJZnyLo(koSl=iY`hj4S8iB5s0A%~FVEJq
ztncf7dKSg6@_tPGx>@nSGO7~(l5uu=x;KVW0q3!)X9|Li$V2ENue{*&1Du&QAkwD#
z-v4sbMF0BgGIjj<%KDGiSHg({7E#NBy?U<>sCmqS2E3t-0i<%sAS#@v7b4h?-e{_`
zb*%z5%kc2<QiCL-ue^h%M(JI5n|#lBUESR+;hk+KTt2#fQS|?6+SAjcpr$4V$AInr
zeN#S46a*ksXJ=>F1O&k)C3d+(YwPRFM^h$_pG!Y{Kv4H(o3!~cO}Go)fh<XBw6^u5
zx*wp6pEY)8Ht-^Vi3X*&p&mg%9GrXh1J>0XawC9)g+)aH#^rs`${YG#s<W}P%K&7G
zj*24w;&gBE%X7YZuR|tOR8#~4s%)kqSe6e^(Cf^kJg_DNC?Y&hIXMS#uX(?74z&!y
z@q9rXy0#qZ<C2|zQ^elhqBQb^!-nfkU{wBji++Pz8<9YmF67RJiiTDRILzt4f1S_y
z%%Gu%3dYIJ&EDh3+E0E;0%$=F4i0FE!&QD-@aiWHK|f3hP5ARKYYPkQ*V`*A7GFI8
z3}82ICWjO%2_HElGZXTpRpO?vP@*3DLoZGu>{G7Hr0%{x^Rt81{(*rGI9^|>t<?e2
ztGeIG(9NVlXx9|8!X2}`H+a8Xf-~8CeKgbP=3N^MHiTm6Y>N+>tgP&GjUAau(8cL;
ztgjDX+hUi0wPu?<%g0T=z;Zy2>xOlv`Wze*61Xe~jnC)Cb{srBX-&<W{~%c9PsL|u
zM#F&|&J<40I$n`oK+wi3v^zho-lPz8q9wd5g567gjVB~5jI4<FPEH)PR8*Z^U8U{q
z*&l0Z#V00C{BETW`b@aWt@SIu3%c8|z(CXs0L7VPIW~+T@VkXYM&iK(uW(&gd)n$t
z!K{`c0qJgD%x;n$z+3u<4{!hY@uMCNF8pI*kQ)&aN=mq`7YE8v_kO6TAcCf2y55-{
z@d0)ZyqC1EuL$8Cp&%${7vT%&wfTv*Uj5sYTUJz4>lMG+zkz%X>tk8+05uU(3%-Vl
z$-{FYBy5o7_YY_O)yW-*H9?+pURc(n89b@X02sicBG&^pG1VP!)1BD{7FO1dbjQ|k
zNU%%&@$_KOBMBclsB`qZekAn)me0wKEWA5pY@rt|LDQ6jrumIRp?UfFOS!RPoiTTW
z0cy~I{;R5r_c>o0VR`!J_y#U1hRNOON^|>#_JHBJx%a_AK@~8e`LE1;&`aLrcPW$(
zI7Q;nstVh?s=Ym@r1bP5F8{p+$%k&gHJIWRQlRTi=P}2eKd!engo2zTvS+xjFH|*4
zEEB>>4A0~qs%<Zb?faj>cxZThJbqx+<u(Py0Gxn<I&q2_^O~9(b{?L1)?2vw+H60l
z9Gg3mxJ*Vv@F_aSv@|t4dSmVkK}tt!Z@s=c$0XBu(5nY1ab#H7Gve?d0C1`;hmv^S
zbhOvSQu0T%98DTB&15}%L8+{u@cc+jfHIn$n>$W5jj!lC^pBMCk3OBd=>1<EV!>tp
zRTMpn$0}(sk?5-fWi+OrpPz!N>WjYybdcyjK5`?VrPJ+x&X)sKEF6`9$3@+ET+PpR
z;-FU0qie*BBf_f!pu69qN{(hs<Paa~ujb}vB*MEj*OKlqFN0ml!m-M2_6Qm;y4y0P
zy1DT3_`^!QsP^`DhnO%TDT7#-lq1OP|Eog2us-|$(JB4^SD^fVt8=btI_4!f7*8+;
Q5b#G)R!yc%>RI6b3+^p~8UO$Q

literal 307238
zcmc$_^;euv&^;I@cyNc{!Gk*lcL=T-f-}J2HfVyoySqDsTY|eY1a}644DPa%&-?Cv
z_v}BgKQMF7(B0>$>gu|;>h=@*Rapidg&5_<ix=o}vXW{qUcCD8;>FAL_wNwDa8E67
zBfj{ZrF5Lt?aZCsj2z8gC>c51+t@kV0F6JonmIZF?QA(&xmY=wKUz9F+dB!cv4Q@d
z0jze87Hp$gk`stUknLr4onE}aH2U}VvP85P_~PY@7jlwf8txf~k6?(|g6>M_6I8EF
za0NtV%SDtLTgjm=UP><aUV$U?18q)BWVlC;_F(%@A|IZ9a=%wZpQIC!g1XT@A^#No
zoPZjmy71QhPc^f@X20~ufDhu|81Lyg@;e@ZZcR`RUoUI8({bcrXShZdzjSZMHRS%#
zX?)emdoAN!ebsXUTY<T@_kTXvi1r9PUMJ}ohtmGfhr3TL&5!@r|Jj+5*Z<FW1vG_M
z&Hp=9f%z1>C+-F7|Gkt#dvDGEFH=8sOp@4h|KcBLae@(|z57%M*c;PD7?g$N4kwC~
z&Jb`xsWe=+-q>Nc38e*ixEBr7(7t>#o+&u6KT~YDH}=nXzzd<XL23iH6&JP(b0KQx
zXd{Gq(|t3-K`h80KNF(8SO{feVq!ai{`1@fQcw?Nu}ZdW*E5^>{|@cp{BJ0PB|b4R
zap$)aVoBp0N6S>!|Fcp@b0FLQ@vMT7+5gA2Blsu(d0GKYf6H&t0}^)?tMwgX<vTB*
zn<(DU6`Xz7x5jFSc-S`zuWoR^R~D9)5qNMWxuem@o#jQJBg}xkz6Oxii!${XlDAv=
z^aw-`Yzeuq8I&z%qcgN0(mQdq$kI@=^s=j=nBYn?QB|_11+MMz+)bMNWfEGAOF?X9
zy+7?sFYJ?5r;YBnS65enJtf^uhn$ApqI6p-xi|t3^(39mGHEjwsU7>(t(UOJS3d=1
zkA-q}^b#<*CP!RQKFjAPAkn5#k%fGdVCc}+fqlvhmd_gD#}I1zp!7X<0NnD+b8WRh
zcp%=BJg3koeTTbbWYGBB12^_vK6k8^q`C4J>dKJFZcFbU5j>Jkwc=Kg)=d}0UKn@B
zy%0g>+J@hpZaKHD><~OMO5y}~K0Sb|x}NTKJR!Z<(f;#m_HW?AdZk|gUm9o>k>$?S
z-r{yQo_Kz=Z*ZC&H>!NF%^Ya$TN7?L*MZsNSJ;(>A%$jNpIhbkAZ;R{H&{2R_#<oJ
z7A~gme|as8J~d;xqbgrD<q<o`km06t@QLMyqn!tKI_QGlkf0Fr{92-+iZbu~$e-VC
zPVX&tMYq#PWBM17#5~&cSm@6mDEko-E|5~ozk&BGaf;`B+s=rPL`iycg8aP_G_*FA
zFN26Zgm>)^|F9#1QRdrxufv15-8i;_K?7ftpo!ibC~>TTwwZsw6&oMbMO?#D&J=t9
zi*#O6Z%UC8Nk}BWb9N9ZOQS;v!P%b=%5ajiz0+JrOM)^#7FBf9de^V|^VpM%9u;o0
zP;~)%Q@afxr&oU$2Q-G+xH@+<Nx11P$3(u!NpZ}|s4h~J<ndHTCVgkhILC!@^b9!(
z5&*30qlHl~XLeH;<t?4sZnXkHB9drhl$#_(0nYF0h-6TY+!|y>5aZl2F;{m-s0Dzb
zh}aS5O6#(1x_@Oigsl=#bZK4r4f_{uk~?Vvl!FOmxy6`)G<T&g{0A)9P-)4Maq&%J
z+G^rqGbnEKz(ha;2R0Ac5k8q-l^P2j&)bEs=Z`z2xU*{gj)3;Yc9)dUW))H9Z(J9l
z%p3R+1|wVE>tV*Kz+SPtI2mi{&ua-Yk(A{?o90xSzHscgVE|cxsNv!F@>q<)@)P-j
z36{Kx@8H{yf;ow^R;zkQWs9mF!G)C5%pZJ9JImhIEohi(j)={*@A#CwYSC%1an6oa
z?UlwMAuG#33f(qR1E++r4(&QcBaDpG@p)hE!@QP{u*{LqyFf;_jlV_rBiHoq1!@;>
ze+++`Ytb31WF94sl`Velcrd(B^y@<0k~;+dY#~K9fOP1{#qyk0#7_S$>UP-zK0L67
zC&01Rs_1K_P7N6EsEC}313euSqAX#_O3kKJsK~S_E6S1wCLAZDNoH9<Grg!?P3D)w
zGd7&hy*w6<lA=<$S8@II;NtmK)#Qy@4#}K_!b7)@=8yR7GOUnsx9W$BuNqOn{l$r~
zc94iK#gZWRJem84#cQ$s0lGB3P}_RS^w24!@}zM199~(PQsPzNqPXF|eM97wWI}=3
z-1HQBCFLx%G_Aj4YG;I|bPql$znGtn9(E?CG76L|9e57K-ue$y1%E{vG%&$S)vm#C
z>ChK%d2cOjR<hU<)c2-tvC)Nu9CwrfiBGYlgb^cdw5ab%BgFh1&?n!7IOm&J8nTKK
zJn4_$+HUWS2nb$ZO&%RScz^Q}C1}WHJ$#$bO=rpb_!H%PQ?T~LhZ~N{vXkM9@$l1}
zZ=50H>=S&bd{UaeJ$Q`OaxLUU`Zf?|xB4rPSAj(q4`Uery`(x)9z*3hy#yD58KyS%
zl)?XF+c<oJkl^wq?=cm=it1m6nD`#e8XC2U{ZP35oXykyj_K~YS)>@P#?W#ig&|O?
z_o1OI>7q1=iwkri<z3MAss;FdcvHmg*rc+!yW!1nAd`Mp-B)&fuc{=;Tky!Ey=VIZ
zUB0HZ0M>K~ZIZ=CM`Q$@zU92g%3Hr{q-)=!<)0`{3RIXK^sXCceOE3{_bsk7P4|?s
zyjFm3wrlOT7+LW?Erw2s%!<3x^xSJh>q1&Xm#aOv3S?z9POI*+(>VoZPwv?+E^obe
z^HhPfob3;3bH(iw*%(Q0?EgXfCT$Q&)K%66h$M39=WtAVc8xxVOG#n7&QzP4%hBfx
zlMnLw6xm)p$$e8!$w+@<;(F%TJVPJ`(vJ9y=Fv?^5UpN4why3|_DU{5QGiXAot5#~
zo)WiRtR(n7Jk~^yXGu~26`o_kB0hL26`gjpf3~iqseRgWra66#^3}qJF`Bfs-PBH%
z@VVU{`0p*OQI8Fet}htr-Ybv=OiN>KmIXVm<VvkZ+U=}<X7c8(cE5dm#)7ObTxNz%
zXUH;3DA&;q#9)kn#6qaN20HxGglMq67uyg&#uDc;BbzxZ#FRg{*}a@eJCeLQZ9&7P
zN>Znn(vv?omWQHJ{i-%&YT_M+_9#aDAyaHRQ*6e7xT&-LOB=m@wS{lXDo9V3$U}+m
zHI}fe4D?+ZYyq)>eK8sS)%B&z4XV7sqZAG$lB~7|Mt%!1!~HgKL5(-{fqmCopIZzo
z%*p8z!&%PEDY)9|Z<pBGRpjL&RA&mM#qiTDj$DMW5P5!tl;Uwu=5{FiL9X*8knMZK
z28$gSmEvI?sbwcg|K@M?lhx`~W78?8=sD-n5ic(<zsp}-iu*qU?)OV;%Br{zf1)1=
zpZ^*@G9i$-c`HirCN=w(sB_o4^WBf<N8jhQTTfp&{M7e>;u#6<`+VO5%kpf7ci#V5
zUf!1Ry^y&bwGffe)NH?MsNV6tUwh&OJm1YXO}NZ|b35%=yYF~d$bKZ^UGw7XIzjFF
zCAfNu;RAE?33ltcU-^#>J*;CseeAmS><WgaHFX?;yiWWnw);u!yvVON5k%$GVC$Aj
z_mc8#c5lrJ#$e^P&J*n-ZYU{XygQ==i5YsUwxBElj<_mAp}A<`qvXY_?DLZL3>nph
zm0}pc^yn`VbAJ4>RP@^UYdC{1YJ&}Y0xZLVht$o;q;$zXegokN;0|mXJ%|@Niq$uL
zVUA1_jGfUjl%^s;_JN!T5Wl)Jz$)k=9__o%1ultF=wSaUehC+@AMX4(u@X9_K0U2S
zhqHR!d42j<yDRO~eN^S4SLH<eY1nXs?P-(EeZq6KNAynSIpq1$_t9`^;9*ckq`OqT
z*nMef_1&*)-{)&5Sf<;Ve%GZwf<T6?-Da+Bq<8M8-#oxXCn}p(QGPiK>{$5hTHGWW
z-r_}(3lE=e4S?5wA~>h#+DNd-1!ep7!}7!<Dcobet@(P@lL+mdjy3qYZEfsw-f)6m
z^j1zfA4~McU)25kQhC{CCftkO4m_fN(<AyIk@snn-uI0Dw#e>|PUN~}f`#4(!8>Pq
zCG>I578_}{f9QDs%pmRb%3{CX4BxYqgVq~-cL)zS{SfMi5w^&ok2VPSC>SkMllDlK
z%5P*-=B{0)C8Fi2`Ig#$YGM^tRI$j~ZSpgfZ`(<epa#F!!+b_xkgSY$ML)4o66zFJ
zuFFA7h4_MvzOEo*HHLC<VqK}#9eC-w?nu{ifNfWgPC8yPtylKS%tlc^vb{q7mJlwd
zuIWo2?ycR6plRr%j3e3)y}n*zS@+|<C(5lu@xDu)k4s&Tzt+0$%DaYZO$M6J#|3Yr
z;5Qggt84e3UJqLocF1eNq7O!$=Ym~0g}4^u85{FHNArA6yJ*GAnXlG^efENH_GCnd
zs~;DuPt`>))qi|w@O_AWX0Ha1SuetFTdVI|Ur*dmJl~31fIyMYXNFIxYiBHL-SFop
zxYP6F^*NI72~vg-82g9s4dyc?+<icB@!o5jn&LPAK=s3LHH1FR@V>#&1Tp#FzO0jU
zoswi<`1~smc2g3$>xZ!q-nR6MXaB9q-S`10bLhNd((c92CD;_e$WWx{5njl$yqpPF
z15@TSP=S-0j^tSe2g<cXROgOskWC$M0TNnNPW)gnvk-ylqM>vSrWW$ZVna||58mj&
z9Vt(Cr4O;8R-u7cWh7I(?u5H<&C1e4MMYsvRN0lH{Kz;2Wof+zJ=Hl&0CEPSM(we9
zj)`(%@1zUm0v7BSViiLLCRU@oNf9+gGXHVMHNa<+-p5bH=Pcd^1lsWXbupUCx$H1N
z=C#G^L)~?o-Ziwedc{NTbs{Em8me+K@w6cNdi^~6SqQ3U+ji9WI$9UE?`P>JxwsR=
z`MLWAVg<AhMnB&~|K~)5R%c{umYMwoPK)su_SY+JPInjkNe?gE_m%7fP6)Ff4=q{{
zi3m}QA+b)JXWYZPo{5fA5oBq9^TB7lP?goF4^@@qe4j8fW~GUyia&HrCZ1(<a*k0$
zOfnshSCiy5Rv4Ev>umQn+IxE0!o`HBK!w;JfBogXFW*QWya1xnFZIguInSUdMqn&B
zXjd!WF4!Wo#g!N7X$KD*m<fY*n|$Z5PPqe<8{{!@$|RD77al7TBrAl{HQC+X`nAh`
z4qsGlyJKkRjMOTqElN&%ee~2f&y1SBEAlrCWOvXW;G{1@<f9yxCTEoV=EO>xo@e<M
zbM@9w_*(sUlg~kuV-Gs-$w=43$iLzeM(VcgfMllo>LlyBJ(%L=C&ju1#oY&rr;}jt
zTsfex9B|3_qJN9MpHSp+U&UnfH|*&8{v7nznbrDj$8`tX%+oU+)91v0*}ez80646_
zdX{tr&PGGbuygMrxfxq~7#pb)bUpkN@00Nl(m!?{<I!i$9(h@6CqY$N9?DrXw3PK4
z;=^R;9&8J#I4YU@X4RmM6iw^0nKGGgWax}G2Xl4M^1lS5j2R~_QcW4q5EIWS_x=D&
zSsWLE0Q236FZCQ^XvnD|l^S;YLM-A9bm`$P5_$Lo$J_x{B3L`MOwLGpi+To=m9<+j
zQ~BWN?6U3k?=DxVT{luks~fIeE?H#)8j0So2yTqayOO0VoM+WV21IcX!8r3n?Jq-v
zk>MvBJ3A*tq@4fr8YujFi{km_=aSb_lQ+rQwfEY9jOd-J$ioKa`O4bk%8^U8=gL71
z)Ny~h@ajng9|2~*y`oP8&na#l2j70A@HaL#E*{u*J-U*Ex3T_(W+;v$$K7nPh|4<r
zKa@L<jrI!I8i=m1t2;}`@<lKqcbjE5U~hA?!9csaxBw|^DutLTUyeFZcM=G^yil3`
zC|O@drumUTn<{P0ox8<A5w*c3A<>$xenWB@azY%L?4mpvFbxW%gN8AO4jM}l%YAMn
z#AKENsrGFiSVrI?scdOeMaFh!eDUEITRpc3R@4kXG7x;$FAA9z|D0)E9;fX%$0J!+
z`d2zyju8fboaw8tToGLUEBbKjQRQ%1pF*3(1D&7e_qh1<{E+?ZRMoOeJf6YdbHDcd
z^tx&P(xd%;1-|Qhm;FTFc~QuAnu&m<BbTNXR}wgQ!t>@B{$%TO4D*Sa?>Kg`v9&#`
zjrNW}oZLCDb6($b6On5KW(`+A-&LO@@G224SOVeAQ{RG_)Z~V>(<nFl5tgPC1Y9Qj
zUPM1(<xhoBJfBh^?Ur9(x5I=V4kw<N*RDWo{h~KW&p!^T+M<Ffd{Wk)j@EXAeQ$zW
zj$041##S17vDt{)4{z4Pl0^YO;08!L4EO<$!7Fvq>27YHG(4u3?PC<J;s!|vAHpF%
zi~Ts(Wc9-wEG#p1=6bs3h1Lx*J-WXuAmWUf*5~&T0))JAQfXW~rnTlV{RO2!naM<`
znqHt<BRVsywcCOSkQ6fXN=(SW#j92FuJ27H-hTZT8cYqEBXMq4NMrex9c7$0Rji)F
z7bNid^@m&>wwPfHh3Gqj)~;-{D*0D83SAcp{~1QWYOu~-4MwN9n+#^13Fa;-3(J1m
z&)zkB8u1-ksJCc&yjr^bk9qAQ2mpBXZ#kx(H6jQ2g#UNcqXduFua@j!1U?%Ci|>=)
z{o4)q`L@-*rL`y5f4zy<^9aCtBo@7m7ahWUTE}#}I@IOi;khZRzGLj%(eK=G$i6px
z2>1PsI0o0Jck^9YGsP<B{{Ze`67%t^X9s%_;z+FRcp|76c#C9haEy0tbmHl7;;eG*
zuF_QZ-Iu1Novx?(=i%sw!RQn1=f8$#T*MAX4fAK0hR>HHVTk^n5K{9)z{O!>KI=$>
z{^%X5uePh0J4K!-e(2to(Nt}_rK>E}Z7OY$7``Y2dQQW1e^wIrO!vEK%T&!IU0o-0
z4l}hye>_GNYg;+-Las>k%s;Z>Y~h30O4u~q@>AI3wbg}-x45z<q!}d>JnUfhjfE0f
zRxN5KLAwJeG+rC&GWM(`0nrALefyJ(oF+P)<*1PZNWNxz?j?51benDX<sS@hv~}IL
z-4;JzWS{NwcBNu^t-pRa@O`Xqxj$~>_j$N6YkH2QxW)28Aeg`5(~jZp4x2Ck^Qq{Q
z%zr#@!1rOG{TPus)F${zF?7FuGxPO*<}i+h@neeIeB@nmR##9=a<@JCM~&MrEA4hV
z?~k{V^libz*M^<i3PN~7tJhC`L2SkIXNvZGge^cOp1JJlabMA8TWjU@yld2LtBva*
zXW`Y*wmLP<jHD=CV5-(np(O6DJae_#?S}jcmIj<Sxk(nfSkt+}Sh{dDf95z}pg4VM
z6Pp=dVBVDz&TkwdUVJ74y|Hp#B#x6TKUDY7ky^`>l2@g6`pd3W$u^NZoeu9QoL8==
zLz2nhqR<)AU!0GthAwYT#_W7J?Yy=~Zu<-$q+Vn=Wr%pEb)CCCjhh`Q{7^4eJ~a`&
zGI==adTd%6`A4-fm5Y1q9{TNsN@0j<m*8C&f>ud%t-+xB|6@4(X{BrA?(WV^ppQO$
z$MbSp8lXFVJl1?_375-k@63>$QTJwvE`BK_6jwPW;fbBhrtIuqi}fDvUCy51`xMrD
z`gzD|P*XkA8aW@@n|qd$!lVsL8&(G_eg#1k!_#ziIfq%c>WtNN&dMk~s%TNA#juH{
zb>5p01M76+1H(+{LuIK~wBGM<kLP=3xPfBoP__Bo)lAPg&YE-D+SF?5z{n6tHab`C
zipo{nxni`DwkB#bv%)_JJ->c_TC-!D(K|<9J3;r_#K~^n^bX5wJ~Un1cP3Ak#FY|L
z#XVZ<**GDxy~S2R`j^gbV<|Qe6v)QL=H%a`f7E;#wL@2-eXxd?g}+|d%tDW2l*yKF
zdgqK%!w~SVNjxy~f~P;!+-HANGDzzjCN$nQ!fFS<rde#TZ=thir#`aJ7}*R6?vXs-
zw|Y&Gn=o3glvixNXen!8=FdPIGoY&*m~{NR(a67t-#%8g!jxUcbLb~Nd}CfSc!14{
z#5t0?zMH+$>TI%THB)1|ArYiE$;GObws>l9wOVGwm<c3x4vl@VX%zEIN&jKC)F?v0
zNtwQvm88Ql@rP5MEaVJa;nm&4?~UGKLmy#3xv|pPxpeLLl+~Fs8Q>~v*j)C09SX&V
zE<*7Z*Ly>8mJx-YcKd|j+O4wB6U;ZOsq67**>Qx$>(GQ^tI7AisrIUt-sRy!<(cbw
z<oT>isC=ojSAD?E_u7rZo{X)wH$KoQ3$-Uh#<loRg-IH$L_bA(L}$1=t8!vM*0ZZl
zot53;UKp!0rnuIp+tqOXXWGyed~o9uEnEhwZ6F~j9qtSF6@545QD-@dwy!Rf&eI=I
zw5YT6xi;?vi7_VFP`w~sKEFN47P~fXOusj{CyKyu-kAJ5K=)Q#Fl;(@p4;pwMU(ZR
z;u9lKw{e@bU4?P`OHr(Cu?ZngMzcJY%$XapY>}Fkkt(|{%Xsi!Lfw>K&=K0iUgkjg
zZcB?-1t}^}B)lhhspbs);e2!ZL(5wC#g;^>_n1^L^3EnaICSs#8Yb5CiI<O&>W;9^
zZhv`y%c=sJU5lAN?6o%DOD}<&btKWjwHp#&|KzTRolk|$|LXWG_$|55rMB-SJN&=E
zgn7d~4ezh*`L-|iXC?-A{|O`$FO<@+562^n0IJZaMB7n${90j#)NQ;~3hzxMkJznL
z@*a`<K~}JtTm+DbL&>Fbs^uHQ-u21a1f=`SE239mc@f0uJd>AZL9_^4D_u$<s4KqN
zBq7rUl4%}^KdmMXj0tBkwUu^q!_7p)F~=D(693Re+c{L#UT3wt{`{gBChx;$C8Nq;
zWOmbm&(Wge{bupwDcAIG=6ru$Izm?rO=&Yvyx{QRu`LhLY^@o!-LHukouyyz{MAAD
z9Q`DDd-zr7q-eHT6O6X3@kw4dn6r`}hMo-Fqa-L{YH!hJ#1dkpG1Us+-l@3Io}$@n
zV*QGYM2}!MXm>yTwL`!L5c<q+j5QENw)HQZhR!pNjzG$p0(}T_<73--yF1{O#ib-S
zbHM;w(xgi?*hJCklril5ldYApkZSLsQ6pL&k`Pj8eej0}7IBv5$r>d`HRgNHDD%wF
zcnYv_7Z|9K^4feNV?CTWvC2g+#_hMHLf0jYl}h7-7WBvTtBuJ~NnjIEvCyB~pxlF}
z>`%O~h6Z|OfJcB(t|TZs>)*z)duQ3L`dDN_A|+GzVMKPzvdn@D9mCcOA_lDC<vf#o
zw&?WQ#9hz`KzT)Bqj*l(@H?DH4cd|6SpAk!y_!`r!ufaSj0XPFjUN+-zkHCGYc9oY
zGapv+yfx`aG7r$_d@VSLr--A*S}Pg8wL;ucsC5n$P#WoE589@L*8!tZvCtfU$l@Hp
z;D1~Y?xT@IaR1d`10q-KtB=QSuZ>^b{uvPI|JRrhJ@&3=-deP-E-Yhws(wCr-W2TI
z63j{jg>4c3*mw5$IA9xEjyaYP+oP_K<nGYTKT)XqTVe_4ijH=xjxL<5<4=uX`6tlt
z)J&r2!_j7k{E>891K4ykjgN>F7n@qI^wdON<0BD+6fTg99EMleFlseY0MSWS*DbLQ
zrq$qG{2OXlE~x~OCk9SShM3?7qo`EKQQztZ55~y`ct-0M3MkT0*s1l_gSgxU3LI2X
zbprS4r9eak(PZ$v?C*DjPW*MpEh#Wm(vM-m1J$~4hGB5x*$D*ggWnP_ara%(&_1VL
z1I_UXtT6KuYw*wJX8sfwk8YfH8uEjND6bg{97I%>p;TC3a`Y45tw8ezBL%v_!s$Dv
zX%yF5_Wt9q`_*faRm8{HKdSq$Cf$5Zaq}X6C;)z&1%D~Y>-vuy&4@nEyzn}U&)z;<
zsQ<4=hWLJ%cs_rQaEhXM+Mw{!Y8)W!SzIJeX8rbxCjGdNj@g?5p)&67(yiDge`@_J
z0#y)ti`%escpMSzR)khOU3AdPv1_fzU)H*<0F%~Obs8weQ?7R56J@?xUhqEd;@W(^
z>_3pp9&nInlYUss793fHh?7WFpgX(?ou;X!3#V?~FvX({l0~MJqn0us{W&Z<!+}4T
zVR&HG&pFb`_8Qr;8w}+BDq;!YvZ|UuJ^IDWV;b75Cs?aNRxjJg6e)iB6{z99NN?pP
zGF1@LvjCb1!#|P3x5(5dt%oh4{F&5pA~>VG4pA+m4U3Fox6*@ba0|^0F;Bcfa=OjR
zgd64G+Aumy^jXJ!wsVtGqX!+{f=u#Fb8(jES8NrDQ(s3N5WKEx<=jCuAD^%PRT(&<
zr#J@9_CM!4uNu5A@+m-(l!(ra7+Zquxe)9hBm379+q=GRKlldUocDd4*VWhegf^A}
zMR{rLL(6gWtp8DDS~1^P;}mTn2>-@|eLr8d{=oMgp;p%nirr;5hf(fw20QSAs%t;q
z(pJW~CqJeE>FN4-g09&DMRLV;9Kqd=!CZ;3MlONYX3keJ9Z|bvexYH1ZSoyl0IE?q
zb)=REB$8Wo)^t8JsvH@hjK~CwdUEy&baW_*ZR?)`C*z#f@y0#-rTSJYQ0^Fk3Ov$p
zn!aCsb=UalFgC=eAgP|**BU9lAHZlz&m$R24-IRzT`u8vDbWzd4;9+e$ix{8Lf|%C
zw$-m{XxJgc-xma9)EpMNXk_7ujRR>tp6(;~<==zRX~Kzr_5Ia7hu=!(<9Q2MhJ4Fc
zVXn-7@6lJ6m2R`k@*o{wvQLi;ym|2A53(_I;s3$*Cd(!%sv)VYENg&8f3x(XT-@t$
zf7&V^cQ()}63>8*iu_qwe@>}ilnlc}2Z79?vK7Gl5oNx@8SL&k`~+E#T_L&#5Ifqk
z6*p$heySThZ(Dm#=5Ji|?uPj=z+QCT(LebrmP^BEL?vXj#6uIsNhqBoWksXVKK82Q
z?rx6?WW;?nfhUH*u=2meGnJg2-N^3P)_RmcsI-3mU=Dd+YILBP`wb2FpqC#o(C;aP
za=<L|03y?2104*@O|(|7GiRKVOhdP?krD0>HQjfKr&FJZk(R%RHWvZ+aPqzgiN7Ac
z1xMoW1DdPZIilT9xrMWFgsTc(ItLGK8F4H;I0XZ?HVwn`@2ey~>wjDKtKHt$VCuLA
z6&`;P<QWYTQ9Q(G`cCqCI6U=_x9-jhz|rS^=1Vg~bDBA)g3Q9LSg3~E$X9D4V{gdn
zfSq0OJWOyRa?dP?cf~$oSa%!_^ZJ_$B3~wTdWNPPsUm-ZwqZ%>)(h50$p0ZvRtbC2
znZVB6!xmlpki9vj^F8bGT~4ehah@d?2b;Dh6;5|Zd<dXGmRf=on2~tM7v1Zj^GlcB
zqh|3{nBV&9#WX;xEOh_Gk{0h1o?d_5!Zbz#Po0?cnpGDSi$}lYL8C`}e;VH?q4L(P
zJ1ZWef*z_v?Kpe6sa3EHou>}lHZ;&GO{TGr|Lbsmx`UH67La}PB2K3)CNwvlSb(xd
zmnkIS_r|2M`a5;;qcWx_V*TE&hPrI!f-$GUm(It-{c+L+I)4USnR|Yh5+VV*L`n!W
zYra-;7Q`AA;xwTqnbGfs{_kCn^B`5g={hu^zKyeADiW1m5Yx1FbMz{~>Hcj~=HYo|
zHGf}z8z=c{mc6;f@MAjlxXlD1K3=UVDFqFh(@3F1=-H$S`ax!jH8g1(p4v@KWEO0c
z&h)tX7JId*L_%60yx`7+p37L^cl<TJH%z(Yq!1`+9*{GXS<tOB&zkbqZ%CZbc<{h7
zB+pLil^IJcM;ZRK<ylb!%hJB8J5m*;EoGd9r-K`D<&t9CEa8--be@7D6`3rZ<!%uA
z+~G@lHjVyzIica*lJ{G`mVc)9M@r0rDHzSw&?EA}_pRyj*3ATrF}gOBi#b@Em5ElN
zGmx0nUNdj7>$+2>zdNBGQ=COh8B~KQ8AmGX+yC6kIL+wxoY-w%?+iC#jdU8GR^6&?
z$j2e6a9alZ1YDnJ@=|Ot*+qaYJ%=sG7dAS>amfFWI{4_sXmhX7?B2}<D8z>$ycerg
z;X(yf>)9<ZHyC=D(@Y}%1tSS>pQLDQr<5%`|LTd0!TnmrW#LF)rZ^A1Fp64<Uc<v9
z6gky_PQ&9@uD!C}BvgNBn5hHe+ku(;D}cXQ0P>c_(*HCEU8lI)87G<v&$BbmtHr(r
zvvRjcAk+;zJ)h=|HZ0()yc8o<=q|=2bL8iMzxh+3OSKLC*D~kB9p@U#n<`gKBj!5{
znR`%q&&jVdhenmSx;A_LLALl)D%9_Qy*W7bal}A=>%b}Tg77cI*)|xPn>O1t)0GGJ
zw^Sxd4UVA}1pjjpNQl0q0s@A&aT&R38D|z(Fv~L_=C{ZV!N!lBtQvZF*BsE>>f=!F
z`Q5TswyID~$QM%UxLHd&ELg%9_UH^qwK5olLZJrrX?3e|d~MlI4r&gNbo=vDkw-3{
zT*Euf=suJc)Spsy?u7?mxV)hLj#)quk+FWaJE6ex#EaNi-_TH5lQDVx`sq*|IIXFe
zlO>mXb%~XNuBW_3DH(2JJdg9|c1hwqS;!CIY=kn-N{aUS>wTMy@%+U)PCQ{fvcdcT
z>MiDiR|^n_8OthR#mGDy8Pg%5Ct9AOv7iGfEGFFm<e&(e<!aDtCa??%>;&SRu{P1-
z>m&|X0E=YNNBH36E8}xVSxbm=JI~0PBg+VA5T(0btJ6nLNV#8OPq6-v6zi_V&R})C
zM(SklIb2fGCuHXRVcPl_;W!PA_KiE2Rp%`%@;_-Oef9Sl3z_!ww^*<%3p5M79KOZy
z&nP_fcx~z`#MV`YBYTBdf-_<D_xv)R?Bm53YGc%boWqM9lWpr|1L41K-I~*GbQbK;
z6`PfSxpkj&`zO{PYJ@(=p?>6v=?$|b9qQuB!(sB!6{p8Y<Dj1F90m^$BOWY6+t}l!
z#s?XmxmvF23RA39xgl&NUUtHKS!c<Lmn@R()nxT~##-RBMKx%Ws_jAsu0~?<897Cf
zI2Wg@i;2z@F$lL5OF|x>Y5EhDEyTY6gG`vRaR{}9XgJ?=_<k?;ddJ>sVxv`xz5)OE
zapJU^^zsIy<k2TfCZBMzRW+=HD`)yCXDXZskuzKF@xQgchorqQCLzyICKI_A$p{I?
z0u3&#zZHj|{>`5=Vzs04AfMirXNGj>_E}J@Sa65mrY%MP48Cuod)W6ETy}C8>`3mn
z!EiEfpoWIwFfnKz{YllD)!H>#t;xb32zmtejTdarTFLJSwauhWLron<r49RaSx$GN
znHp2GX{=BaFC)%TROZILHNTLG1yCsk&R$IO@%^A~E!7ywdyic?r8g3Qf}{x=QW$Au
zTfzeAy#&-9C>Ds18x0C7qWN>>3Civl#dV}ey#w{8<RuJjI4G-XB<Y%^;&Rt>JOH4h
z@&oo-U*FYf>V;PdCw<SEa*QLGCXDSZ8`D&jD6Ec`x-bqM*{2H<qmKjP=BmMZAQ1wi
z<&_oG)hyt*U-OH{5+c4YtA+*D&50fSk=$pF>t{~#&P{iOV4m_cn4m?IWMxV^<^6f;
z%mEo_5oMG9ZxAk)RBj>oY8zYBpHzI@!om{a6-mU=aYyh4@8tgWL-%aB$JS*A$?kF+
zzqy6R$*t49P@S;3o0_VC6t9*+TN*n96Kj-wMni=uGJUieA5wf>(OJ|%^wnoQ2cxc>
zMubY{V8u;49VC`iBXq^)Kp3qp#f3T#&Z)@l<t%4e(#f&0qj=f>#rPvM%ocw}j#`a6
ztrtLCyBNAh_pVaHIr2=0Ehv&vqvw2)eUv3}jFyz-7g$mMW&CWYEQ2D=T%RP#6doNG
zQzp$eQ;@1zE=RJ|JCm5J3jb-rq82kf=;4@re|;Ko=TG*WV$fSDvr=oIk|6xl7?i}*
zig5(2l*48(x|j(O!=ITzW74Hc?Ei2;J+y4>$dJwmJlLh<D>R$Uammt6A&96h$)LwP
zWl=gDr2<Ao7R5_C<FIdY1x$IW5Se2oLjhB7>iIyl!jkwT*tf<}eqcI;0{u^&VKC8y
zBBX{<e|;S3`ZV2d`9f%QgoP~EAmz_`r%CD2_BJsN@Iv-%FZ=X*PtHzz*X-~fqWrvm
z-8~DegIHrd!Erc&j?h77h>5>Q!?mJ<0z!L&Oyyr?$=Q}x(pEEOXlEG6@wY7QB+gga
z-^oTR*9ldyoNk}!*uvhWfz4`$6rC~5F^X2}laoune7N*8SB8Wbqp|nLSSsN&Ow;5M
zpr#HP_sA2+(*2gq!zE{N;QWc3)DR$)94JA=RQVl;%ZsvHm4TLqC#JH$f+~@d0umwa
z8dd2k8$l@T0hDBuoaNHwf|=35OKK2pAlG8rr5V2sh;hRFp`v`)w5mX`Q<OAl<6(J+
zr9eKk@pF{U5W>-;AGt3Cu`)je{Irggd7rJ;nn{*HJSNYZ1mp^(<W$3pOt2I`pXR7Q
zF)vh9VvxvC^p{mYZ6^tLYE2J<Leio$NMJ64ms|2x`$05x;fZ~LWn0OpMh4^z|55{e
z<dMVg<Z(>?CSDUgd9&Pf$%-PZ^sP9Vak!Hg?bb>)zm$>i^*2%ncpv-qH)4nDKrH=u
zzxJz7a2+{ns{B-~s4-yRSy=vKz@@DsP-?PsF_f^vTyZ$<v7HY3nTd%maqI3<q`qvk
zvWa|)g_mq6cZDe1*Ezbn<%my#WN=Huikpp$tN5U}%v-1?y2>6FL^_-h2WHayc_E4#
zu8MEcEu<gtQCf9&mo#mCa!!xYSM|ML6vWEK8%Q<Bm!I|>#GcGG!Ss7ntf{y_0O+?*
z#pAS+RVtjQiqO%=n_mRH7%AblhlaP%@<rg}?-#QVAdzX+Az~Q07Nyc2EF6~Cc&NnW
zN=J*ZYNNu~S4&R8OXFnkI3X_%4slU7m>X;iV~|4H8eiU7QSj^h96>kx8$F9yDNPdO
zYBHmyEMXdrY?O#m+I*}O_!0m2Nj+b+HkD%m#~)C3^%RG^v=IQI&Kzc&zZx-C-ZJVY
zltGdq0gMGW{7+&YkOGleT$_9CKg5iw{w{#<9q1g<Rk#?FucoSc4v@`PoJCdh{q7Ou
ze0msV?P=c(hg~69Q}=2?Ufl<&uzWKAODwS2!h`}ujr6Ff!queWb;9nf-W&2O$C0Os
zJRIxB^SDJYz`^m04H_79Yi-wzkLh(W9&UqC{IJoX?=?VD1?xBIAn<5nalqS<?`0Fe
z@KI`&1?4(M8)CIjZ;7bBRi%p~wUH>Vv#rK_2cUG5Mq@1L2UUbZVkF7@Dl|g|ROJVi
zjk#25s*G?(sr1G|KQw>-^~)rtT_O+?Y0W!9i94u*R|b0X2#|Ph4OjJ176InDP)(in
z#pV)N-|6-aB}M=k>mrz;9Q;y&3@nuOH6_fRVA^j1b;brzMTG$fB;rZ|<)Flhf^+~Q
z_!YrbJpOcIUmAdTN+lrYsy?N`R*^JSD}1lLoer}ymX<w8HaO%XjA#e1M;A|T<9kQI
zk9$|w1`Yan_pHo6k<<dA_yXj9W9NETni8YdFbe<AWr402uYP)kZN14x{+~48Ol4lU
zlchk7XgL^L=*1<o797c-71LtcL=qKX`QAeFD_n4nJNb6ImOF-SHzL@wR$vdS5e~Xr
zs;1KHvgle2bb*DA+<HKsJE|Y^@>=WjRW5%raSezy5vh}UB_+6t&IUY@ZYwmbL?V3L
z$%Qx=HYWtp$LIUOZVXlfP(B#j_@To==*#(vBDRaFw!aP#Yr<VE<^Fa)gWnT8Rsktw
zL3?wOFbi>??>hxK*Ba=%(JoG8SrSa*eJnQd68K{733cn05Lb!txu$CzD-+yLuiuq&
zgd$T*{&~}piLy2CJh}N&2d-aP;iRUbfr1<g<>)JnC%g-#RG@m78RBCj2~Q!6P{VDo
zP}5Lknp(&E{i{~Ev@|$xsN8C?^dsQs89<Rr>BOZZ#*FgNqcYvmF())4cAsrfqz+Kj
zMVJuqaxO`CCv_JOA?@htjz>F^FS>3UjoY0YlMp!JTMx-87G8?%^;fvzyaOuIC&D5k
z`aM&9sp0;IoP~I+f%Dsivj@wYpM)vHifY9?7gOiQ4sd3|HF!FbAGK#O&TcLGJXP?C
zYz6t<ttcSHh-<S3h&v`2-#rcNVYigghl+^CX9Bc-F;o{$%o>H#IKV$&<>r158c4-W
zYGcD#@O92c;-z19GV0ysD=rb17NbvM=Ctn6!050{fz~GgtUgQ2uDe)&8@GwP-Mpow
zzhjuMc56Q94nh58*t+wny45xl(Zn~c-Dt#l{dB&PeyD;9^<bE<F!kE(RDx&;wWik#
zR`RvXz7%1Nsx*~UrETSbtLk;fg7%E`*g*A+id**MNHWDSiEOZ|TABBTE`bCKDbk~q
zj&*;YygBasr$2Ns<%7v918#^aC*2d+jf4R9dqE2Wnu7|sY(WNw`~-~ZKh&@pwmLMP
z6BG|E73m{mC4#-u^BNTuMri4YJ2*x)Y|UT#Ydk@cT%9D=-=)Ts;glnnXC{adV+F4t
zEcy>@bc|Y#r!RCsRW-Q5JAWn=!lu=?PJcf#2`C0~h}+O&%M5~&WJkx?GlvMfxv0>O
zYzw!~tvW39WD^}yv=3q2FzL5_;~E?vE13fdxDE&u>GPsVopH2gLgti5<>`y!i9H-7
z(}*Znew;JLh!KHg2ftNh5I7nLx)yWbS3@GrRWnw{_9RSjW1*z`bpKlm`PPh|BhoHg
z7MY{1?=S=+DbG1*9Jc5P)c9cC$D~quQitP>D4lP5AN3C>td99bw|`c$d)8wgY=40&
zNLJI}pn0WlX9wEUDPubv>Gg9n5oHUu4P*<|^ferOvlWK!D8OX+a1EW7*a5zXYGEa{
zPLPGKA<Y4Hi>2Q~glVgCQ^osj!d~l~P4x5w*sKMdPoEg(4A$|ak{m-eWbm>F1gW+g
zlYc$q<oVD?yn{FVX;xgeZeU~<tzDzaMt%R3kL3N7_+~wdFYjD`&~tlAS#;J{elOB9
ze5}PtBSY?Ll7cu3uyNcUp^4vyF+Y0VawYcah++Q>rFOO?IFN$LH@CSNvrgs04>L{7
zS++h1=pc3fiILiuEaUm5o6&F7?+96caYY<%;+f)PYbOpAh7S9j0W$igk1>wsKY($h
z!Zgx(s1fqAe^l>mZ)|hPlCI}))1Ax#ZC4E>%ILVnMoP8qq@6XudE(n(WN1DKi82PZ
zfQ)AW<yFyA#DaQQw}5$znFXUjzmX9UjGN%v>1-NBSCU*m*s2^^ZwyeeS0raID}z>t
z0D5uJa2iR2p~HQxu6<zugI7u-GiU6|d75#J5TZrjt3ZhG-~DDt+XUtUaq@PeSqw3>
zmEChDOkvCK(sp!y#U}zG<0Owq-*m76Og%Hs3!Nkor7cH!uN1up=@(!rF(nnK@S_5k
zKbLu7y8TG4h&5RcVkbukCL)_RWP_JfXqLgY0UT*vUI*+?)0G3wpBZnaTPVvj2@iy_
zo9*IPbo*+3hT8xx=)Py`7`}JwJj4?qdL%{ErN;gwcP|9>>IX-jXUQIDQ5W_g%Tsc?
zG)#Y@HqE1aHx8CJ(9d<KLS}KbB;)v#V5e`PhNW5w9p(x6SW{5MxRa>i{qe(8QHpp<
z<`SQBORL-#m0=t9jQXV241Tlj{rxKoA921MXKW45+Ar*IpbTX?8+#ZRpD|KsQ8q0l
z=4Xvv=F0vmSEUggu-yx5vx|(&sJuA&b*^f{JNdeo1pOy+*Lth!mB{r;-0vt9*^F~O
zp-fbiP5A%{U#Hd{*5tMlhuD<V0gt8@A#<WQdV|wd_~ITkkxL)pvgIwDBvP85C<ipy
z^vS<&5X%-^lDA73^`>B-y_wOepHgWC=PPvK#7n0s;~G%`Gd&LowYl?wV?F?macft4
zJlHFv1RavwsQ0>(bnTXSgT>~wG5uorPLOOz$n`^H?SD7F-I0ym)J|^e?-4w`+g>X#
zAupS_?vuG_b&6)x0kp0CZ!4ZOc#L2IeIY$bs8w*|7!`*Vqj<J~bh}<>`3Jt#5p(S+
zO>9vGiyU)he5BYgQ<d&Z_-#ItE76#|?do1F;wcgBfdX#3j!QyA&wGcWhOMn@(9A^d
zquX^~?w;Fph&5EeYvV9RPSt6m59`H%w!n+!_E-n6LJ4dZxpf?IruMjBMqHL~)xD3>
z>PIQ4;WE0>piL>S6ssVsr5r|%z0V;^?|+Ch1iN9t8X*1RI*#hxxUH1YGkiL#;WNUx
z{hz+FLLL)huzTPCbjnTpw4|X=ADH(=BeNKE@J(#@+HdppeU>@7uoGX~=XTCt%fJVB
z$yy8*4S|W~vQ&7eu7Y@Ekd6et42NqKJHMq=oMl~3_yLU*nZ7X>9Ym)R?@Vis8;W>W
zgpg%i57sMbl$l>e24xy9qr(JEmAOtAVW6$)1^4~b=A$)L;}3R}rD(7F1X2ayY=e!k
zrul?SDSyVDkdQ%QKHi4ujScM5#^$DfYLy%H;LS^yJhIszB&6D@om2+3aIu1&MU%m!
ziTU;b+4eKb>P5wVE*{eTLT4N)w$N@qQ3+G=5Z9xjgD@b&8+QXeqM!V6kFoPnbGD;`
z0PvXlt*kKUzjj``<a!1yHj^RS+nG({xI&S7&{+O}kN|uCY;opbxo&Fb_YuSM#JBW_
zvp6Ik`ek3>=y63aKilsnpd~{fj(d{i?D$t`mU0E3L%~3UwJ-o&Ht;licIw$kn2adt
zT=*AM+)eF*ros@nOcJU)y!}6g-T1e$<t64NOAECNZ=^L~*tI%jiniJX1C0Lxv9htd
zyW8+;#pVwE9gRb8xpL(Rf;-)~Wv^}P8#=L6w>%A^cC=Mnd98rx5hJ^qtBL00g0-My
zp&66Ia~~FeGrl3_J)RwuefKS16egT|SyV6#!{vcIVpoztn{!f8SYSD*<yP}?+$4@G
zUMb)7gTvI8(li=r<@V?RQp4X=N#dzm?D8mQawL6eK|EMD^V+#dr}9fPqx_xNx9Y)9
z4e&l3A(bsaqcUSi{99f{6F=a4W8A@b$aj~)*_n;I&+~jKB(0&wF?2a(as{WBR@N^g
zE3GteV+B<3L+|-e^aMj)V^aqs+y3auL{e3=)OT}~GS}x7y;s*VD4`SNrGFHECp`QU
zvr_9zqnMcH&*?M=)mrW(%rB94Ze`~==@*iCgUwpMCn+ldsfo<-nsplSR)L}0ZE89b
zs}<XuorP!xjc16&4p*JD>yM>F9bX#^nMK@%4b{rf@qALW`45o|%|BVwJG@CWNl^H4
z@VVOE(z?4)_2udeIu7dDuqrgE!FE_XLoYX*_+J&i_82aFWsZla6CKShMj;8Wf%rgY
z+&q>F`bb>eydkEPgeU;E!%A&LI!|l+!!`;gDEihdyGvG8uS#%wqgl-+ni1kBSf#sj
z002icD0e?Q!Dz3}!<g9(mM6fsLwKvFcmuav_ii_vEn!?*2o>AGHdXVjV7{OBx2Ewi
znYPitX=@MLL_zh$n7H8E8$nCtqC_c%F)Ow$9U4ya!p+@ME8w+eB7OCNtCo%CT*7J-
z=!>EJ+(58;rB<5kEsr)!m~q6kCGC-@-=|f9(s#PBg3AN$QitX*gpv5zAF2Baw)co5
zHTd@RVWoe+%b)x%Fn%;}BTPv^f}{zG|7`4A0O~J^vsT83X^qXz8T&9-nsSO)rv5ZA
z+>j1$IN5a|K5Hg)>~JRuCj73^{0r%~NyMQ7-trXj5bK+0XxLE>|AnI8xYD2oPD@;0
zm@V-PYFe9eUPz)ePzfmPsXm0x%p|ix#`?iEP?3&ZuOC)viY!+k+fBON8$#94n1@<X
znFyX#=C}{j!zVOrUR^BWC_N$0E4m}i=#(BR%X3h+-uXH{;@AD#eV0fe9Q}Arnzz)?
z<WDfPP4)c6KEPhx&6u)9Af@W7<62(T!UWk&w}VfLXb0x-%&-4mQVoBfn3&?<?id$B
zQCCivSXg57QP$|DHSsn_q#}ZQQo*Io7;0%Nvbu#we~;Lm38A`?tCIC;K3FHJ2`h#L
zZekCi6z|k;WJZ=~<vh83+4sAnGYaTzf~zn*dZY}4gA)V;c$!cVmN#OdM<N>fXs*Tw
z2_lJhxousFb~27R4Ml1M+gtl-ebqzY_`CMMvy5)v2v1vNP?W0*ysl!sJ#=a4a11$z
ztpT1r0!`V3CriFQNh`qL^2YB;6Wb=>XGKfOojLFsO-x#92px2@9mmg1LLWM#B^cp1
zNC-9YXGf0JBBZ5X9!6Z2dE1l^F&fd6eyQa$1}Ui(3&C{s!$YcsxZ#+DP_z6VK)g|^
zSZSG2*$_Wb({T2u-ye!mi8c<F^?ynEl^E77x9pkWG7C_?YCxq+4D!<BchF9JFrbl^
z9~<c4HZ8-r>nShQ+e?)hV75&z8hVe`I>e;GPoa?Yhc6muhA=?9aQ0m>XC0e{lESfb
z2{$8cLF+=1*7Lq3U7k{87DLXDbq~-3*`{pLUv1)|0%zk?xs2W*4PcxOnyJ9F4s38F
zt^MjyiH2#CC+<g&_xU7w6hdMfde4t1Uj`UbcK$i2Ilh(AClHDLW<r2HiEC>=qnakX
zrc^$7ep(|l($+FZf=I=d1zbp|Y}QzJ4(PON)I|%l@;uMd2}xTBn0WnPkz9q%wOv)R
z11tDY$B6-ou8Z>rEI<x3tW}(+1M$9T;ahWYU<~o|Q}SWfm(f9kl#5!PWTQ!VTZJIp
zJEC42uX=z(EgeMHhG{i|OeLB0Q)zQi%|F8Rx)pd^vBtY=N<RG!^TBo)Iel5(>H4KB
z{RD=2bP{|GQimp8Dvy>bFVgK%7|MZ{5X{fEqeqxv;9)hBAXPI8V0AN0620CIn_4lD
zT&dkX4p%q+hUghihoWB_5azZcqOHv)8rA1=mBk~}s51U!5HNK-V8XOQaIEANyB?J9
zE#Sb+!bpFvZbntX=dC~=9-QYmyUM{J-q6E8wZdPDpGKF#`hGcna55B6o}TiJt{z|<
zIt$qRQ8w8be}g{McK*ZhO>u2TVIYk=6-btf+IC*HNu339#z(Np+iu!3M`_yZYWxZ7
zDAy(jLz9ez%Rkb8VdNtq<6J|lj6E>%56SypSuyMoe;&u&N(tLF)H4hKQI!Aq!8`4;
z3$jk~?S|Q;7$n@q-|y;jGQ+tkV<tB`-Ux2_`DiGFt1A-!{1QZ8-a$aM&}f_f7m$xp
zv-hRw<3Mn8a*%A;sxsqTEsep5UtVeBTZs@;3VFOmceW~hJn17KnR8SbI#f$OB}ev^
zIBQY7mGP0V=1!>rXSU8L)haO66x*N;s?fLmHqRzAU!0xEY%_F^21J7QCc64{W!3=Z
z?0-q({6`TU<(%Lx@31vpAiBipH$mo3OMSdEH=?)`)wr|l{-8gf1hBiMDx@q8-%z#>
z;2^~|S{C3s6d<&xWgo{cZ0ufFQ^HRdKS7Ap&nbvQzMA%#VJ8O~Dm9`$o8$o8FACaP
z2>?=VWle@PD33B#>;G%xstQgNQ!H*Q1C#-RccB{>9?~{k@zc{vTz2jIguG<?4i0K(
z`o4Fc)APd+#IgOzjO1{Pk$hzWJcEAlY5n=B8!QmJamic7&J?isH1w!XGt61Vk}bm7
zy|u7MpU?n%cy;^wv$lHaE@R&}Sg-*xo@IWh7$Zw<o(Y;mo~+lh)ul~Jwm^0uQ@7^<
z&302`oG%?cE2TcU44z(XV!aM7RK8)-lik@FqRU6=kx>L9^P}dFhIp5yLN^%7(o&Nr
zoXHFZ=@JNO6_g@S>8$#5#CSs)m<9^s;7iRF>>?1nL|d8Kxp8%^sA*}jc}7;XjF|1Z
zq<I}M0}?TMKctW*O&czraz2M!^XML60+8r<;321^Nz61Xw^tkDR`#nyR%PR=Wx~(r
zmectlsTbQ9yL|wZu<3O+j?j3_ASgbBl$nj&n+=gX8l=oDr6|#vk!5rd@BFf2NfX8d
z%(++Vfl7Ne+W8VlBu4X)OHT#4*&tJ2yO;Tye=b@9D!CX3iTxxgQ}Ve`irbk1;_p5h
z8L`%<(W({uVBd|xaQ@>G6CWDzuA#=4tZXd#c_{VtnHtL7p^%F7s|pB#d12o>KMU~^
zEByXfE+&yY(>^R#Q-YxYhK!hwJn?P??Ys~^Y<hiq_6lcXEkT5$=m)j;@$JtRY3>Pp
zB<cb7z1BAwhmQ?f2Ww4+QzxK!upwFNDzb4NDp|2cqEJ=x)yADI9z@$Og^D1w)abSP
zA!tF)Cl)yy_uC<nj{l3KvkZ%BeY^NM77~JVBi-F4NDU07G&8^;%`kK~D&0B6P%?CP
zgC4qLh@k`oq(nekBz(93_ak4pxVZLy_Wi8&TkB@EMCe+|cAX5e08d~z9CFr~AC!jE
zV*ACvYs%d3?ny=j8zty*$bR~>(}o?&KW9O6YiAd=l9T20gu7QB4n`i{KBK^&aRUII
z!a1fRRFrIfPD`{#8SS#&u#R{6PXTe)F&>|ch7=SpR!^<Y#mt&Mu;;hkZ?8ns>rl|H
zEvgL~;-juBI9`dnX;`yB5J|@gpIK787vvs`@OI%_k6w<sY2)OxVPxo8UqY#&1;J?C
zsd<Mr2MV9ELf07Z{!*5H0lJ9Fleq?RC^$@TCyH*#yMv4b!y5CAl%caHOmb}LLCpS2
z8r#YOdmnNX$GK-(8p8}5aqP0yb*AMWV6k)B!}|2+5d!DZHp%Do?7^``Z(ekm6-rc5
zW@^qh&ux5OmbNWu;P%&M8kg25wqmLN!}-BD`=^}x`6yI?MT3={E#8v~!GRx&_LglB
zDIir<5Z~l<fV`V)ev~1zLQs+FC|`)Kqk8_#f&%UjR!$!|UR?|`Rbnc~D%fmO$?@4J
zta)0O6w_^0s#BD-xJfTqUb<?_^^rJ{F7Fym2!D0|-z7wrkjC9e-}}BXg|{3{F`<{b
zx2~^Iu^jO&aHz<2oExL)C%{qT>|p)Tzlz#A!v14<F6kzD1*@Nv&0YB`E>*G|t2<8O
zTuok<B|{k@9*8CdEjNOiJpHbmz%s<sZj{oNtq`K6BQ<C<FbQxA8l-?XgKSn_u3L-d
zs<JU$P=<PR<&&t?NmNr4@xiUz;Duv|*R>v7$qaT^cDF+Uu(y(xnT83A_OKxT++Z3v
z=a2|5a&tHLaHD)!aRvRK4gB5qgWIST?ekxg5_vm6V3YS%JM^`(U>%d-6RWF}y|p->
zzyUXy*C7pMq*OAN47nY<dv!gF<HN}_w0{woop9BDb{i#aovc+`mm}xR8W1ecDxlbP
zI`?iuhk54Rj8LUxWn}}lMdS^<|Ncpn5fz^;kyVj2x|dgkw8>A)q%BLJvWr7_ra`MW
zQ>c*H%~YE6W&g5{j21_FN*w&rbwP}!l1ixq_a;`);PeiKB|PAMaG->6<I>Osmu)^i
zEoicZ`o_%Ei;<X7J}v*j8j*Szr@+A;OQueHT45_%X}zvbL+KfT8jNG+I{}Il<_p3Z
zxK(;yRz6o!(HZeJ;kXfX?rq8~tjmMg4E*<%&gSk`Kbmkh1$4RijaPg@KFbN>=;ZL9
zj2nr>nUBKS@n}HI&oY$Yj<%8Tg>o#0MOV1CZQ@@7Z_`3@ooS32)y{cprZx$899Z)U
z64E_!Qcp^YR%=~FcBSF{EUQ|4fhChMMB77s+K^W>Q(m*WJr5Q0%JJ6m4^RV>+fJ*M
z8AoObfONdchQAwTKQS6UYQ{#mTw=rHMs{|y$fzlxK33%oxL6R2m7);Rr!WB|obtd-
ziW-X3sGe8pJu~V`m{##G`I8PM`S>03E>YWRjz_hrR8?MzKW)ha8e;&|`-VTBno`Wq
z*K|E6%~UViQ*$(?yL`KMEe)(=)y)wv%Ci~<2D}?YR$ph{X1ArCuB9CS0!ifWouCd3
zrP@($e+G#E9G&gj+}r7-szlUw>~nT2pS_7fwr{9qNPO`PtMI-pl@V~?Z4Yu@-a9$0
zl^@tCLSOz|@7|DRw;7PX4!vpnGn2WH1KaNW^xfZ^om_<t+9M#_G4%QRyb!?ODwf@b
zym)e;2d5)At@V?$(M^vYoByj4+|2gbEa=ylLM3mG;^*vz(P{cNhB6VNnIhILyuV&)
z#`qtVjgK*QmAd^KYUu6!>*Mv(c-Aveyer77Acz)D&#i=7tb2<`e3&m(U_+;7k0z);
z`DLsjHu>J*jN1SrgLxoclHMZ5#nj&mw|Sr>g3Bg0vp&W%wo~-fO-PYub~uSGf|Li0
zV0K-L6=Jlfa<cz-p`gMXk~yzCpKsXWBy~xm#DRQDoYn&!b7falv<l;ccRgPmwyGwH
z&mv9HofB)+C6JhUtY8{AUV<+5b+2}8{UvgXiJQ?IS1M;=L-m5x7oV3)>yPDl>Eo3Z
zA-_<8L@md`%%O(5Ufq&8cow2e{=*V{zU809YwhZr53>I#4z$5B6|PQiXw<^7eA2!7
z@=(f3j!zSf;eQ{YF$%2T>1YIbmlH0`Kn4#{?y$aN`(FQn1CPNcbW0rmSW}(mSjR+W
zOjAh)jX!)m*?AnWy@rQ-G?pKwL!1DTJ-`Y|4=W1$=A}2GPL|3RG0$sTz4Ba^T2YP*
zaalqRD5urn9wiylm{H-%usLGGk6Ht7q1ZSNaKVf1^sb8X?{z4rM+;@x?WJh8I!|5b
zrN<bfNDfqs(*d}K_4-XE2lw*PnmK>A%|*_+UKJDP2m2av2ZtBuQ-8W7=TM8=>8Zx;
z)RSyq2Cyt+j#}P7+ecbQ9ip4Q@*BCp83liN(4FSuKi?d8mO^w}8sPiZqj&7}<@KMg
zL-qzqOI_%1c>lmg9H7v|zW&l0j$6|!qh0M@MR^u|(|Y3TEWNx9L?3tz|NMjPI61qd
z4yfM*<~9JC%j~MtJU}(Lk2^$WceFT=!dZ}wuhNU0aPGAGE7FDD>XbG4mDa848HyoG
zCoT{bw(oL0l>(aSc^342lSYJ)K*o?tSKTDyDOIGd<%8t!D0bn~+!^RNE{KO9hSi{q
zH}Al%`ddG8a<f~Xytsv-g*X$*V5Tow+iNeD6|Z5*iO_K33n2BQhvELK8K)-Nz#69{
zf(oYF=<8Y@?RcvQA16wfftD0o`RMAk;~mI8nSyfIc4th-vs7iM&renX$t7QoR8f+{
z=)y4A&cNJTGOaZ~n9QKxfGdS(#?%ACkY-`N!kw-eMfU@a7j=gUEtPzaCu<QrC69Jb
zdxQST(p&vbP*CHL@?TPuZh_SUP?n&Pk*2hEY<ZcNx&-9?Mo)HI_HOo<m(;Q*#~Z0V
zz?6N0R+-JqMdIMX=uh*xW&W^-+rRFLghNsnQj#UOxJ!!had2MKJzZ1h*!jlDCSw^D
zhjvHLlN#07<NyIFQ`&#uD#o^?QYCm=%}81`lm4~I4p^=0bGmJjP4?m6kbq)aNKhR)
z+A%$nV(~*|KXYCzpb-l42_H3q-8>IqJfUb4yh*wK&!>qQbRGs~t?q!sM5=j_vsIVY
z|6NPgwhMAktGgZgM`kB@x9=~$TjV_Iklfk4kIM!3RJl8zlYY7<=Wv~L`%?^DxPc%4
ze()L_b`oDKZWK%i!K<z-XtDb?Laq768)6H!La6lH+vD|#RZ39T^`IB<ZAHd^B5?K5
zK99aKrE}k1Vxo)?d9s=bfw={;DItdJrsCg>o*4j?0nzc2^)E@CkUq<-hrg74C%h`Z
z1Y7>c7`vSO*3V32gfCz=WfeZMQ)De#oe8A%6f3?pe)MynY(~n^k5fL0Vh+`b(Iz2=
zzow&(Gqez|bKw#68a_XJZ@@*q@srBN1{FEOIymm-VDT?j_LV_^dJ+AB;^WEisSwg_
zG%6jfW9>%Ajna)5;L$jA`Zf6r8@Q<$;9~pox+x`>obDg?Hf_>kSKkt^0ZyVrJ2s?>
ztZ>cv@CRWm_aj!?Cxdmw?xdAJgRWpgT-@T3Y<@2bE9ljEX4KrO@?9OoY>KndU6RK>
zgXA}jDPS(X8|H7z)WuTuG_&!{Ikj2y2>?$5eJL~F`{jK<0#7COoil6uN1dv29)Z~J
zw)X-~1zO%rMp#+#RzwsvzC4kp3g+y3%xl}EyWHaqR+mqoH>qpL_Y)=FiQ<JwjHhc)
z>_Om$X!wH}8-j5Xjd3d-TR{CWH5u`MDR(q`zP9MW6zhyvpdeg_SREZ9LUw!z7q@?w
z{<?OxmofOJKN(&2)^Op+6!&H6>DgxR+N4^1)3*XIt3Id%bG)vhv=AaUv$^I`>H`IR
zewwbkuFV-iqMc{!9erIH_V2sgHSx-S9cuFowmTwVSc?E9?JOToR`;F&%t9EZB4vzX
zBye1^ZQg+gl)CJPx0|``j@FNPUXBI^yT$-1cv1VEbv`Y&BgYt-?BHWYQ!kbS)@boI
z4JjJSc99!xD{H^n%<TVf@RjZFo#L&=#3MInml!YCF;L<xZ4e-VzXJqL8z0@%p7qb!
z0(rEjfFkv?+od~s6+W@gViq%c^4|_%=a=NHX4jd|jaNX~Q^TUk$|EfJZ#D$hTY6OG
zLy}KOd6O^B78U*K(Pi|BKePuvTJH)`*5m#zXWb;{5C;5q7zsuzEN;SKsHqHaKw3E1
z06=jAI9l|-A??jIh*EnuyTiR(Tt!MWCU6T;oqX|^RJGHCALH?6DQQY;_I@|%az2z!
zQlRCIpA5jGO~9Ru<<|&5HlaOaCighlIW!~vG7OE*Zm;4af9K)!wE0Ze##4KT4spoU
z%s!LADO@zs!hO9YJ6ZK;m(MFxYZo<4TI7IIK?Tz!Qvv+pz7NTNTPDv?ypo?igEe8t
z8hPjlXWrP^jA|RYeOAxer$j%rHTa)-v<yB@Rb2kA`h<E2sc%dB%vA@Bf1O2<<?yk_
z0xDw>skS01toT;n7xqFjpb|H%r9pt&_tZF6xgNe7N!4?CKOsIu7{!R~`Da&2&SDmN
ziSM-j@r6@sC=cmoU%}#3e@``KF!#%l@roq_BqDkZ`UFm=ft>S~hL3};t_yAc-ZJqc
zMkm2$unxn-T_$_;u%_pR6Z~^$W78$M={UKN_*C)OfsIzcT{<yNRG?{=^6i66To%1e
zGZCLUj2#oKI<n&9_zZX-^^!J8Dg_@XQ~TYc>Lx`d!@oYW6C={nIK>A}{r-R#nf3-U
zF&(!0b3e4WTR0vAzojW9#>xQ4$W_wV?|2tZvJqvGTSbGUc$p4L4GXQ@OC&1>aMyCB
z3k9e#ICDK~P#26*7phi|&|}KE<K3xrmK!~pk3;gJ8}eiX;ufjN9TPlqjJ~)Xa@Tfv
z{|BUxD5r?_l~JUXT(sR`%YH}QqP8wjVQya4PFx?;&cVvpweEmJN_pua_cQ-!1K;k-
zZ@%2%{%B!+bVu2i@WqQFfOW{nX({=kg>PHTCLRHSxLPK;1I#{TpURpTPSGec+V1vp
zYdv_UH`O6gvvM=q<(y;)B+n#VF%c&MO{SXfa2sPpeHI)q?~9vZp99eO=JB}hefxag
z<YO(%e&j)SDd$U>*$)$LBW-0STq?xSE(R%uQJ-a<j3}Z0NU0tzwv~W1Vf=ad>)WA-
zr0WtJ1}|&l30+83cvkJEwc#9-NMnT=<6A{=wTDHrajFra4cV&TDZg+B8*>WvURf@>
zE&;-2_5`6J6zqGQK(jT1Zp-z>B(g}CE3l9x_eKmj@-1&|RH5;*Jf2~|an@KiP)=<!
zhM@O{*)A*AT9R*(no_^VrYgy54a%_-!Sd3uLkKV0%^@MV5M5G1s-IbylObl3ngZYF
zJR~(5Ln-N<1ym{StoAzb^6t6cy}#+r>2KBpJ@2KmB-bhMx1+aE)F9=VrDF--L&88E
z)Uxm@L9TwRUX1mdC@bHD5~tV3l7N@sXhp^zSI!eEvU1I#`g?hyPhpdjJ)NDbV`PG$
z_948$E}0pF>Q!)lSvVua%Pr+iyO=xYNSNOblGXT&L*|L3HRI2Bf+9J{UvjNV7}Wa(
zhAJ1UF$Wk(VP?p^5Hp+zw|Ft>u6ga=c-of<M8YX*de9+O$`6fV{F0yZJ^t)S#zm#P
z{~C>{dgR{LfU&C1s9TFyV_~VBsXJwuv<_Qd>dpZSJIUjfif~<vFVNxBl`Q=qns}GC
zVq0C7(B+Rkse@@(DH1!~zaG8ZEFx35?Gm=^<>bY!dTeL<v?W}+|4m%C{X5oRfG;8t
zK4xdSAxKopaq;8RdD7{}^dR^^68qYB<#1?!FoJdF8xjC8B7ekbOcLSGsGR}EpWAFb
zp<Rnz;L#7cP~IJs0@fBk@5RnI9$d?=egokmGfC@gynC+GJ{2kFEs!tYImI^=aC=xA
zL=ACVWS~?AZK@u1OA<a>B0*?5$p-ih@D2UG9eHRXxw{Xrpt)gp@R6#f+oisn*HDjh
znUH%<mym0%DNO0_v4;~rZ@9?ENz=O=xhdRf;tHe4x0sNrv;}6%{ht>F=?hx&>Z=*;
z8)BfAZ(i9-6E$QmMmh6|(-apArx~S#VIthUS6vmm+Do9$`SZ9w!_HS<P`EE_%ru%;
zWV#;L(mHL((zVedQWmG<3BXmw%;l_%ZP}xASZA90lo6j1sWEJktzAx|Or`Z`^wmeE
zdEo-|1RCG}GaMqqhZtRMkoN*EXZ>MchLy-xjVpo<Rx_PRl0gb4<=b{Jnb8@c8g7tE
z6jQl_<LV=B{M4V*2}}rf-8lF1e=6SS7`!I~;_uDc{rk;?h;Ax7`WrvsltX713lB+P
z^ozH9@8hc7xUA*~g_uJ2u12R8D&Ctn&o}oSt_@5cS-RabTI47(SwH)5AZ}Vlp7}~z
zOd-_x`Yjdi)CQY~FyW11#c;*5ph*o(M|${>-^!!`jI}MYb^GV`vlEe5>Gqpy6B8;}
zD`xuh`eaSPXqK+CxIRBk#6Yz6xPydcx`txkQI&Pd0w)V{Vi*7KnLP^<aS<mv!gi$_
zR;T?4QhS#|-g$+lt?0!q#7mHp`YG{fs+n(rcjkO1(<(GA9yaPP&|^QWF(g>Kl$#DP
zvqznhJB^7ybRb#t_G#=105Gz-$Y|3n^0u2&qmwPA{ILq(l*~9!3rK3$JE$|og*#|;
zOd-ezS}#%!U}8VC*>r7m!Jo`i9csmVdH9|lvprhq6b$4`y9j*SVJVrW19fv1hW3-g
zb2*6S|H&DmQgPHA?pHTHRJKl5rll(8<SsXji$P5QV_4fL&Ic~?6qti#{W7jUnxGb`
z%&ecT>J6kcc1-DzK!APyz5}uE!+g{(2>i)M9!2@C6Y7x=8w^c3vvK*3{^VH(tfj|j
zy`iUkJ}y8Y{<`>2;0vETT6asf?b^@V?Alc}%FvTtJ8!Dp^c#txcO0<<?LGif=%!PM
zv;+|TSP>aOtzK@9gY377@-JqkDSs~)v%24Vk7>LG|2ggZ{O5;M%OIOu(cXY!F!nbZ
zU&|`13j;2GVq4ws^ELe4cFzW$ZoQs*c(Rjw7B(p}ikCvmv39}(pHH5H54=mxZ_w}7
z50!wA#6Hub$f1-{BBn4QnPpNKpVaVDt(+FtSBzmLdb5?A)C=FBq7+-171kU(ArY(|
z_?80F_0sU@ThnLXn=25`r=FOgy~9?K524&(jZ!Iw2S-&UoIF)}PJy~8Why@UcRDbk
zN>k(i@CsJ4$efDQNT{Wg`?t;;jVavJNvZpws)f?gO=ZMev6wdt8&ouwIHwJRQd6R8
z)~NJPt?5~*+($cVZW*+(Se>z`&&_jn%UU|>LL-RNQ4-aN1Bu_BmP$I&6rL{77Q)q%
z@a0>p4ZAlQ4E7=X!H8!bx{H$U8g0Mj{Ay$)I>no6{uTBNf>2{rv*C*tdm>Qq6*Q~G
zoVAjj5$6Gsl}MFD-H5HUsu{8Xn$nk7A0k=P>&{^$$uw-&H9ZB@$gnfjWI{e?rW-iA
z+~Jz8_R0#@jjBftIKIYX_=0kUyF5`(&M>rPBl;F_Re>;Kxaz!dv;?@|HfH6b?}YZp
zsD}j4!)DKr8IR1RiS10}7k1`=?_gNDV}&K`oDo9eR+Q~}=$~<s*948%1!M;8=jdZi
z^-0IqHeOCaCQPv`6Am=>^(pQerBnXt(-Q>ILXsbLip#Q)Vi^%qyQmxr<+s&Iq5D|<
z<;H;26pJXMtS7=e+8`!ZU7@6>#xF!)YNe@|;iN_*xWhKTq%LLZ9yZoc+aE6nSt&hc
z7m8sOud*dR=+4xpNu?QSJ@sYxBmdGPmKq3hHshg4Y-W#b)>lfrlw5oLH(CT|W$9Kd
zR`#1()2v|YOw9qpaxLuS9EQU8pIth$%9{gEemuo7ZAIyJM6H5H)H!<msU{uLKDZsV
zZ7myo+ThW4bx%4OI8(Ymw3~FlDrTyLkDCYGHGgq!n%ckkOL!561lvYd>qHy6y7O&)
zf$`s{|HX?hi;A+TK6!fQ=Yj_&Xj4R}3aB&4hy01{u68Y^$|E*$e;WwMC(DklM+nFR
zdAxJdXi1${I#Ph2;9J*lc)Mr&L6alh9?hEm&#D4NI9@}KeKZ2|q+&VMh$I#G7P!Ic
zW%&C2UX;v<rfX+S6W3$orIsd)4+aSujMWm(#y^}`(<DeuxT)Dbi=4LSNamSre??dP
ziQk4c^dgV9g>Ij%;n&Qm8Gm0`c7yv|v(+a3>{r?J1%5)4!nYodYBZ&dlnO>P2y9~}
zZQ>>5P?_^FfNN4`n9@=HcwO9_tsDVB7!DLdd;Pw<OJf>9L_MJW*-$ogTri-$sinDT
zK_;ste^|(^J3oa=#YSbTs!DaEZVjWws#m{&>ld#M#TI_4&yQ3o@KRR)Cd#hQ=Tjn<
z!7c_bqGlyly}`ZXMOSV6id;-<vG_wYow5xZkSDfJqa8fmW0h;m%fvGHZ1Kexq+ZFu
z6S)JQGG`=$1OjWA$=xZci3}{5+=|0VABK9MbaXJd>?}!;O_75#$BB~KVuNa?*aA{o
zZjSt^h^PKx{}j6|@(bh_3<~A2Dt_}XS191@R+R}CM<H-<ISfU_{iF4Zt3COuJOCER
z1_6Cf<YuteD8&zibZImL#p7C2nH6Ztf38K-z4f@g%~}MiBia5r-mPgI9sRch5A=St
z11spdr!B0C(J&c~tbgstHlNQ-z}wVQ4~-aVR8T7F!)wh4<rweRVYQ48M*p1v9?>d5
zjBSvp_O-Sf&EQx2@&4Hgrp>`<CJLa{{9C6OWA_G<^(S+zoy2StC*0oChLajmaVJDJ
zq*EMTwpj=0*2(iN<r~a6!(pwY4;G$Ab{koM%JiQcKJc4^DrT`%hZ6TGRO!ei^|1R7
zNW}=Z(S=661dE1B+B)%BOvevfX8FZ(I40Tz=(@uZy+zM^4|+XHfgUFn)5f)pIfqEA
zQ2k}B8k^zZ0JyHFtu789l7&`$>ae}YunYJvRxXBaWAzLnX9(CJOMr;n{a3<VHIG@$
z-}C^co4PU=D0|vMTk^#(q0&t|*ERy~XS4w0>Q`*z&F!S!<mq&J+xyR#En8iib9fTY
z=Swu(x0@lK5<NOaI?6YAYN&sHy-bVk`v7zJ33GE`+sM<f(k9fs|M`gzxC;TdnciSR
zhAMYc5jEeprS^6Wm&|Rt{pS5WYB30ZquILRN>V&uPXRlU9ZeH~$<f6qL|3-`o6YIT
zO4KN~W_c!=*a{1y^<LKV@q+^TKWxswqJn9f$zK(EtxQ!qOqEpuTz^J{yX82~wVXI~
zS=$l%?8rf4=bOWzItL#0d-q<K1Ur?(O!D+z2#{xzryL4|;ceHg6`GQ2Jd751Jhzq(
zJDaH<uC{?na@%2^;-;!64f#(rRg@#Z;n_&RoHU>?1MC}r#mUFF0eCpGpbUk~Pa-^s
zh}jrbpVDs@7ujsY65e{)4w66mGE~|;OkwSku|wfDk`sQOJ`qZwXzFSp&nz!e<w*iW
zOkgy)a6%oCN!-pOc2|dPr6Ox4d9LkDK|YO=aAbICmn_I;09#CQ{1RM5T@(?_Q9?Sf
z_}ud4-mp{rDKJ#rnbK{c=Lw1V@7>S(t%69i4eQm2?c=!`)gPd>ruU=yO^Ej&A=o;;
z&kVbdffRLQF5Oq;5=us#!-ka%l=eSz?wD9U@6E8++W*!iS^c44a`@mFXXKqy^GrTz
zXk50$Z_~yZ1?t6o{WuMw-%hX9#Z?ts;%Cl9v?yYaWCc{t%${VQK?O?kJ&pGVzmmK0
znpHQX3uWgDGJd@V+QdIS#nAmKTHe_SF`@3B8-M+18GveH9zQaIrDANVzZpoqcv3F#
zoJbK-Ol`gsWM%GyQ}LR^sL!F9r+T!@+v%~33)F2YoOr9jmgYm0fpIl^vMpPzBzCIp
z&{tCc`RIxj%jd?WNdDGGw_bzI+?!kt#`v*wHU`KlX$zVVD47&CY~G(uV9dF48YM-s
zGHLmxzcn%p&lTWBYBOcIRW(hHL@?03XhJV{mG-uHit@VTg`qU@9dAiSzw)x4)KpVi
z%wDm@q&-xm(rK~Q{lQ@giX}IUHsQR@n{Ufr9?AWdiRUMC(&dC3Q>8hnFs3(dlvz_f
zlgywDJ@;dqt<QA%0zgxcFK#Lhu5Xuuf_FuBjgJSqj+PB3pLqaRbacjOR59A!v^5<V
zHBX=?FU^BU7i#PRcanqdS9d2pf?d>gps)b{_=P;!R{GXMhl6W*d9MihE}zt3nzQ*<
zItvfelWPx_gj&`&y$&2bR0H9w)RkEq^6MVaxnM9_g~gCZYyH2XQ`tJRruhh75)xT>
z10#j2v5<#<&|k@OuS5uVeXSKEF-62q#(*-LNhW;FUQ55m^zE6zAPa2*altkPTurIi
zI!UGKduBz0qk*Al4NNcm#C~!NEG?9Bu%Pj5$Vjgvy1+^lOcy=dat_1;y-V1Dl6;Zz
zE6&~@LgD?i7s9w2=Ev#@YyJ`zg{4Gos@ZVH#xB59Zy>jROtP)7!J)*ZS!BFV<quYL
z3`^6}7;7uzMDRFuYqK)n4}WadFFl2-8PO*^F8rwA9Xx|~Q#_9_J}!P&SSH)pVe3=c
zk7Xd`l1{c@i(RC$&U_=z^I^;1*!UF9T>ElXJ;CXweL#ki+g^Cck4;03v|gX;ada|w
zs!HpiWHsri`e3Qn81%yr9*wHpI1{}*^r_}(j<E*9pKaC}Og8+<vU#^nJkz118UHjh
zW@g#2c8{{c(P-u&|NW69mF(fuWC3mc@B##X;d<}ZTBjI`!P!u>Qkmx=q4!goY8aYL
z@QWWOPgxp^xS*O?t|T*IK@3)-q1Rtv1NB}laj@71Vfr@!8u)#K-jT7pIb*Sor8_fQ
z!k><|C=IzaBWV|4sw*?R3EXR<;1)t4hO0UCQLORh0-#ADSU(fZWU|iqw$A<Y$(oUs
zS%3SxpN(_!qL?LO#X&xhN|6uAbMUlgdgmi$L^6(^iU*{K_983rozryxFt;W{sk>rk
z&Xt98Av!w-WY^6D&Afmq3W&$yjZG>;n*r%tWF-~rweumf7IXGYT8P6C5MqP;qmu)r
zt)R0isrwS+2#_BiPi<XA3s#7ax6ri0S7w&`+X?On!jhNuUl=T-YP_$4>KYO*7e=Sq
zHU1^OIA}iZl(V~(kd;NiCb(4cpF3HSK~8C%eh1aYHS*S<A5)W5Qq<Pz%6!<_h)e;j
z$Hp4fEEiX-OCXKQffd#!(5*2!p3z`-KUmB+h_jtJeJRhqNMm%~SeDD*FoUFkZqv=0
zqOI5sfLXrNF(=lwt0g6%Z6>;X!#;l@7q*?CS9yC9+Ic(r;$=T@eQN#U4nQ32ftO$8
z;Cnrs#l9EKh#!<<osBUr9b2pdGIC$_0HCbClfU8R<WfGwSsDvO6JKm&Ngt+tSpR(X
z;KkjopZmeDu~vGPXoA1PC2eRJDG#pScH6!^QR3$9eVRuA=_g<B{-%ba<V`gJi(=YF
z5&lJ#p{B<|9#w;K<66Ar=MX<c(TmD4ON+drt=(7(A}6gj!ST?Wg(OhJ$L}0MJh_=H
zclzv$zMSezp^YE=9>Po6lj-!*J~W)h#iGtS*{+bOmsB4}9_n)#T1WSfi4t>$@}gT}
zFjbb4e3T`MY8UhZjnO!e{u5-Jw&Px@`<Jq6%}p-up|z^Nc3$dCOGAvYl0fW!X#`7m
z8u0ARs`_u7Er%J7)qkIPlXDE5rW5sX!HqDAw$(*fhiVOulyE~GsmoK;eo=|1c5kUw
zpT7j2kS01fbt~D%(3dXdWruv_bPpSfW&%^A`AbM$B+J0@HBX6+wi7)_0reCvgH06T
z;aVG0PXAl(r%1@Nt5i~87m#9Qi*mMRHI@_a2MObL`xtZdw-7w&aHhd%5VRw;Rmdr3
z^#mICx*|8KtX-arj@{%A=%`V=A!vxQatm8}zB28~zM6gq?`b3R2<<;rTbUPHnCF_O
zD}9_uJ~3mDlkN5{mb1&$B`a0#S-%>xmXJCCJ9eQM^IlM1I+mP={@T&B6}z2qVCc79
zZjiG0b8j+D^{NCY2Y_1x{~9=#kZL`Kj@FuPUo{Q63#7Y6I!S?|fh`^2a<f3cFQ~>m
zd6WTsjuD0;3Fjmcegma1mT+WctRSXE$vYhv{H=L=jligQJ<`*7Ar_*@c}#SS&0qjl
zoK#j;{y08vY&boCoEh(2gUjT+Tsg6>!WLSb-Qee>;$vAov}sf3I~~ncyW>Gdl-y=J
z$U08+Up$MZaua_hFU2O=2XQSP6~|uvm{<J`#>$Pki@+c*U0+2NWIunT-TzO$4A)XN
zxT=pePM%b!Cd2SG5CqpMx*(-)P)=1Y3A*!Q?1(QmUj{Y21TY(<V(?{&VjjE7Kv7CP
zf>70|z?;JEB{IXi(mf?ufSq(?p%FV(8J0{~3l0w|?)%40)l3R+*V?y@@-*OWC$o|O
zCgoxwnmX)DHTwcjEpI?0svAt46GyJ%%NwpN0{pK4<$s~d7O0{?H``J?nL5^1rzYQ4
ziTZ?%r}ktPX4#l+Kzq4=<>E_)mIQyZczoRT3`K#ts`4hOkzd}r+zN7*St`wD1>j7E
zFAQhXmuZo$rC)rD#-7od{S~|=dHUo+NqMM!0OGuij_6q1Z&pjy_oAF!?&2pu@MdMD
ziJ26p#0QU?xV#+XsWlt09|&o`A^9A#t0J!Y`0tlmM!WX3UPizoZu1;EIG{6fqA-eq
zsZ${2?tT)!_-vPYhc2n%m7W1OM~iMiQ=1IyhJ;tMqIx+cg`<jB&F6nmX1u5{)TLrj
z54!&9le4n|`}2MGFd#(iYx$@2kdu>}B|3_4UK66Nn!R5GAb~N4lx$lYBzz^@<@U6{
z;hHqmwn*hNQIGoWP*z>f%}fQ=+1P>SLe;zhq~$wW_&oUO+;ljhJL)*}Mx^Fx$`dE7
zwtTT8%vyBRrb#xOPa8e9&9{DODTZs83kod8kXh+Q-y3kL$djJ7uYrvZ283ILfVYd=
zw9xeYKVbh2(DP(c_~8&v&B~uAifv##QnrT2wdR50)RXGAOxujOjftJ*OoWZ8$P~xa
z<!i$!Nj|fcdT{(6uhC2<Qj;QVtIHZ1Wwqvy!7oi_7e>tupMP|Nh!6~Ik3K%71E^zW
zEeBSfIs5Vi+j_d}&Sbiq=5&k$GE|4(Nekn=pc(rdUq8cFbEO&FSN7j%48D{G0U|dn
zKKKI$WdUn=S+khGLxvrz7w=jSkNfViA%e9Zsm|?;r4>C~W#?wsRlKEeAr<nWlGxgY
z`ZPd%399!?W14G*?Pk<OtC#J$5yg)gXn%v1C9uNHY#D)Lyg;g@h>pI0Zn9B~cQG%G
z3=5Rs>LJXuYt(5}V9=JK$CMF|Tj>sfi9TPAXY$9Sucr}n4L^a0p$IkE&x=~w9xd9S
z`=5DQ0BZphgUJ7U)+*1je$!`6@+$AQc2`@zhqns|HWJfVcly|=;ehq4JA-!D-dlBM
zNHR5_4uiC{p;%}Q?F`4J5hWBK$IMM!SxA)l0bG>BU2&6OwQeS%S+oARMN5<3nfFn(
zQ*M_ERn3|{tI9Zi8coknOi~x}dG>U4eASa?2amL!>mU;9jc$7>=ITsx-R6}kmE#5x
zt^qd6mOdOS6?k|Wc|q}EH{E&Va>0X>%3dcMO=L;N-sP*jfhV<$F6?1*n+es~>M(or
zeZf<%SQbMW_Xfyu(Ydfv6*(d1vnC#7iSP{iBqs8TzE+UNKe{O}vD%vMxb$2Ee!bId
zXQx{yQ=N&VI4H<gOFCx+?-ip&6BIQj!Z`^ct3-BUv%3A?qy=p$<z7R<U4s}{Z@lV_
zX=A!&oFtjRk8RV%{Ce{_vf{#22kmn`?Hcb_0Y}|}7HJFPqcO_?FmK2I=t~Vx@PJ_T
z#_vM<ztiv3%Z90N3vC;E0H&0#-#Bcg-)df{C5BJrfB8IcyzZ9_d=sXlua{R6&q|4$
zJuHlW#-#GZRiG-|fn6O6D-Ab4^JRz4C<i@B`;J68;+X-~1OdLeGqEChma1-e+Un@(
zE4ueY;;Ir0Uwm6wxP&rWBe@ShmYfLbWPO(yd;HrjWPGv-ooe4vpZSlWj0N<`1X{OZ
zV|FyApdv-86B}*a&k2bego+IRExZU*J;45B7;cq`ZP=r;){&uB!f|t}{>c4O#7Hl#
zd6--sj!DoM<Nbz%Z4t^Ve;&$8OnbLDeE>8n2!)O(u@fh!Xeu7`cf+uR$MnfUPUG=Z
zAWz9Bp&0bYVdNglnI?IFx%#<Nqn(ScyG7dB@MIc5k=;SS5Wlnt7)kw_mht{@eQR9;
zW?a`;O5!P#j;EEIT4{0<Lwrj3=RHP*ZLyUuB!{$uCn7EBnLTRDAytYymF+itT-~<J
zVTz@blwoOtT1&$rkT%{jE@@9~<c64QgobRB`l${@JqqM0BUEqr4T=1EY$}>9lZv{7
zOjb2X)~b>A`45Bn5JQ};`pw0p=oW24VY;xS?<sId^!1>$HDPSr?=!WkB&>pYd|}N?
z13vPgW1^N6A}fq935&8|LN9B&C<)7J*IkFg-gc$dj#<gfM1YW~N`DZi=|uug#fk7^
zTjhJ<&e7T6>zXIt7DB`R(x!aFUP6&a2;vl(_N1`*4M2BTStXrx94wryokU(m5khT2
zs)4_5KK+)uUNdi52SlIp<9_)N-$<O+k`x5F<1Hw*Zcg;yJSg;EB<ZGZ#EF^KPdoa}
zH})0&1q};T?;01UpB;4$k8hEpo+8<4n27vbY#4PXVxaEmBGHm&!i2*1w=0i_a%M<;
zQB)9v7&%o^+GxEIOO~zRIOmR{jkhJQ#31@jI3?gR!UXqd$WnfKN++4{C)AMrmq3H<
zOoL7Pvw~4Opm@>1-Te)W`w@ioTT8y07;}766Q4Q<G+j4m)P@RntjIRaS&ru^(uZ;G
z9aQgiph#6|sNt=y^7iR_ln95fI8+nS=CDP;csVi)d*In1v=z!Aue;oy$j%^MwZ*HL
z^s!Ijw+Jx#oE{FxPs!RoV-8ga<c)A949Yu`f}Dy%=a2N*7t_vaFjp)$%9FZF@!8b4
z(I65cLY*bVj1SrcM*$y&42fUUM-sj{zff~JdSIxd?q(aHZe4T{uwE?GnMz_sX4k?2
z*b|o~i6`kM;y@8gd1TA=l#iKZ=2(A1wuCfL(_r_vvfzXPKpGnZ^qgPHF=A0j?Qt`J
zV7<zhnYypC>?Q@+ictrSziB=d8>vsIlMx&rA<4mFj`**asSkl~yiFl(y?o{3rw14l
zt$Fgw4}Ugp*8k(Fk-4r4Ij*&ponJU5`g2DKG|O3@mZ{i@^Ep6<rLw$M`rE3!Ej#M1
zQN$I8%^3+17w&;$2)N29n*6_J)Ycr;%wiIaqJ}QJ1uLl4k7;o4`950k2{RkVkHz5!
zK5Uw-I^1!3K>u6<PfN+;o75PvU?_QW<{Z>2W9HnlyuaGhvcrvci1N7Xr0qZcI75<`
z$Ps4uB#`ppSL9w7>IX)F9h7EUaP&dQImFN%V8IImMEIX+fc#XPy8C~y3IQccFz)w1
zROrV&u?mb<^;l147Wc%(q<-BFRK&-don|pzO7agUT^<S^9F=`)Jb*?aL3el(lll;7
zRNS46Z#(V<TAIL4p-UXXOh`t_{7{HixYZkL(ZS~lRx|S>mRdG87B6|#K_dGE@pk0l
z<W9R#Bk19pObaNb;tk|Q(<ZY`Jz+$9=KY|G3IsbhEj^V>u2?O*H(lek7AbY8HE3mT
z!_qH^UwB2Mwd-$GF*mq(>EV#%vBzjXKQ@|TGFJ8cI@!>*D8u~@Ls$vvoh*7A!mfg!
zSeh|t+L`VanDhWm93NLw!Iv?pD|XZ$v7&uRT_jZOWlFRCCBTjK@Mc+vmzVY~O>G3x
z*=AlRQTbC+m!)~!ndBrtcn^`EdVFVLihT4lBX~dK7qC$W!f)F49|CqpBX?dvbW$SG
zy)WeQ=RHy2yO7h@_qaegriGPN=;zCYdm5mCR9-9H=JogMyzkn9Z#~BQR#ICYV^a7~
zp8nU&gC$u56+-U?{RH4QmTnnK(OUK{IuR802usjYXxAGR{CR$<S13Vp#T~bbk9;RE
zP~<$mHU19xj}D{1P8s9(x)>KmDNs|j6(!-X%^D=~JL}#tNM*v_P=t4MdM5c&Oo8zL
zAa>HZzp%zxH3l<0sCGR(;Gnx@0UBmd+9Zu@q%TtxE|=P>vc8?V(5azp<Q(N@P|j#-
zPZB^bO&}yOY<ybCV%yx5KNxQ1!p+mpoKqM0lv5Bo8$Hm|He>iNfqJaRv^0j3@Qj-v
zD<id!G0eZFh!{c#4i}a@t-2sk8t~8^AAwWQJ}D0y_1;utCm>V&7BtOK<7wSuhl;X=
zr!6plZ6Pew&-6~20gHJZF6Gbvm{c!O;VZymg(@9D?KTvO8SB1V(&las&#Eo#n76u|
zWoK+WOa=M)a@GD$;1;kGpr~&fVIKUT`s)<K@&*9u*@i$HuwY4XwV$JBc0@;i!`c`9
z@6K+d?&e&9!aCk>$#>UT^l~1yiCs;=dkOd`#s|f8=HPONAuj|yFsw7oLKKb)D1rBS
zX6s8{liF#=PH$a@<Hfp|U=Ts<#^*mP<|Fi_uVoL`Zz`!@cR%>0>X?v?`Yy|AcK*&C
zp6US!Q>i419a*&hU2c&yV#yjS3PGQ!>&$=3eWerry}OyBa>szA^?CHSUhroo|6JoJ
zL;0p417kHAN^6=8A>~l2al<T6wggX>E?qJUH!&^*Xcl$WM-6BqcjfkO2ASu{R6^$c
z$?0$4h>@}OXl<7HW|(7<FgE$vha*{~o{C4Aa#e#x^o_JCOkKi8SKl#S1E`X45B~ht
zg*{o)cir2mBbq6OEbQ!oc0zr;bt$RZzQ2(se?7(`uPtsP8|h9UqK?rb^J6>QL#J5(
zQyiAI8tm|2<?HG?shVG3SeOqo4bq1>pFlJ@uw>23-^dKPyC?C97eJml>bN7{!@ZA_
zgL!G3ILK?N3p@vbaTVtG8~&H=1#VHrRdhs$+Dp{2D{a|of<Cz8iWF60$3`1DG$f&l
z?|Z_}5_R7!wP&~1^dR6b);)O(%Tzj7!OUx}ebRtZ*nV%HxKZ_0V|PflOKkJYJwf6o
z3h3rZuS-!r?_~ZL@InT&a&fo5kRS#=lC~=BI5Cw_1i6z>`oN1I6PDOcE<nYmWOiFb
zef#lR<q9&Nw`dQf`Q5z*><p#o62|S_A`6#^{Un?`0bl!y_WxgeFZQHFV0S^+anSKr
z{{3voX!iQ8t=!M|L_a5eKBaaZf64~7keGYl5QY5e>-uwdybJV%8(sfh{dLs7%W(hO
z{LcYr@vrOrKi9j~v_XPuQ6=LA@SJu*E^fD(7FKQJOpF7}TSCEv)RP%f1UY8+G~E6;
z)AV}m>v^?eEKsxPGHiPoX#6U`vWT2i1^o=n!ncvuqyB;#F2(_ZGGHI)S{nC=&x!*b
z{GijWtyjJ@K}nj&Z@z{m_xlankkfg1F%D>dFixh*+G(e;iq?r?#d?xrUeUcMVr6HJ
zwhqMDvrqZC3GA+Ekbdt-&-{mDV;KBYQm&nQvNDyY{I5pNjvwWG@@WFMQ4valWvO%`
z0~aar$P{P9ltDf?WP|Ah38}3E8ikehqD&l{X@S#rCVBujbD@T97;tld3<+m#CIv+-
zi;ysSzh6Bvm|Z#tU@&vAfVe>!$p+S+#jj377UIIuTs*jFhIDfNjF@qMfr3u%;C2gG
zqX=Ky&E^}z%;v&N5?sA7A&*?v&3?wMgahFkvxAuord&xJpzsR=R%Ukf4gtwWOUh@v
zEx;xZYRq@dYNewMQLpotHUJ$>c#=mN+>fX+DRfVe53;A+-k6Yh!6ZPxlKR0!;>izk
zHUL3k>D2QEVpPGl8E9|7^)#iTpfd-K23A&ReSNyyR54KG@?p7Gy+mq8Ybndn(+WPP
z6+1Ta!nOUv=igA?pvzA`%uDv{gi_Vn9bSoWDeNpvR;dPH`Y?kHY9PBRVzeE7MuZ<N
z7h4$k69-8q&yxrkMKTDNT~e~tU_!K(>$_koHq!BRab2^OBjVC72Av0v#fA#cBpfKC
z4JIw4m+B_#CHR(d$qqO5fWEas(3rS2aqTB(ByS3ytwF%$S&GKZ<BY+_eb3uxl149J
znjE?xiw{l0r+)$6oQXe%-6W#Pa_0oqSnC<_vT`|>CXu`)`P5Iq;{cQN#Xpd3{GP(H
z<t0~pll<09o3&kz2=&r*OgfmwjWzsH7p0RM8;E?@=EcEt&n`@C`=N4vVRstD45*Uk
z@-wT$yasZJ%*~ld$3zW4a?F=+55bWIs=i=etK5Kp&QD<XB>>)A7$61efll)xs*4JB
z7Zez*bvQ{xZJd^}D1J<-3ps=@=56uN1BY%w*~^;(gL}Y9mkY8@1nMAYS-W<Y|3%!T
zVXAg+1&0h0;@{InHmXH6V`6gQBM&noa$Znqj(z-nZs|$KniM&(?mzX$>gS~JIQ{zX
z@2RqFVq5R4iSzwD6Ww&&;AtN2Nhu&yVJLwWK|tyUU~-{AsF$|g+@{@iF^ff*Bwj!c
z*yC)|Cr$_sUeZFwKh9Dq6gjMD6>-+dr#|+v-~;+XvBaG%j1Q#Y{q~0LMH#M58SWPP
zK!n%*E}4I2D+S0w>;sHu@L=BxX&m;8d(~AJJx)yr-q*iJ$J(z!iPPB6zhd7>3ck*l
zn^?b=35mS={O2xX_5SAbpPTEMj8FYQDSZ89!;9OV7vECzZ&`s&N<;Z`9&!d0cjY*(
zes2+qG+Ex;+#oz9in0D^N&S_y42nEDO;+)b4QKn*jmvltV_(`L8?N6J)Z@<3Rn7Fw
z^yz^DgOx`X;z`6d!rJzdJWCAG9^PM?NEH_=S&ydpE<w?ZZMXvI6fJv=CB-)K`t`O3
zeDK3h)IK{VPB4)b3Y7zXI%^?dNV>2st||`XXPG;F?rN&{7+t31r3>Ovq1!LXG1dZL
zx)Dnym<&h+o_{n6bea=?$F4ozeVdxa=2Qsd<||`u%6dguSAE8}_wew6g4w<@Ag%$O
z^4ng5dz>$9@hHW%Aer@W`ugOSK9IoH9%BJ1z{E6)B!}Jx!@xhwKfq`_nmF{df5Ta&
zi@WfayJ)Iac;=ra6E>=&*)B`xWmDgbxccL!l~j|_9oPZrK0x7A#IfJ<=a+X1#I=d?
zs-dqlYXF~AwoVE#iXj`HgASWnlXqUfSlq$iy|e*UeAwdQlAzl^>CBe`+UYj6BpBG@
zm2a}BmdZ)8Hu~x1_<wEPv<N;)xBv=cG_d}-^*TVgNX9<fR!*e&P@Maezbbbvgru2;
zp~P=u31hLnPP&yUe`}~s2MOw3=}zN$w|?65XLPM8C95NO?+aZ+8-n{BTK?uPouX`f
z38|7sWRT@Iq|`wM9v(_FRbT1Yr8WA@>&qrwk}<s3%t}$-=L&6n$=AR7!d>Vku#A>@
zK?!H!-S`30%7VtU7X19uuAWKPUFIPCW7ANP8gzvxOA{yMIEq*H%1*a_g<<!b%`mGd
zO4QI=XYPA>te{>XMb1?`yec1bIG5ziO=w!p+5$9VC0ywxoCD2gafX&_nPJ$Q;(F_Q
zv&DRHiRpe>F7jz=f0DyA2>MmaP~B1Abu`w*zZod>6~Zn3B+BdS=@N0}I^i&g3I1?Z
zCzEVn<CZjTPo{bUI5@&_7$94}k$i~Txcs=(MN>ei?A241QhTNFNsDREndf8-BiEc7
zK(%T~_OP9WWMj^1maDDL;9Ajt(!;tFAc>9T3qqZumFFjFW1l>SPrm>9s6-JR@pIBc
z>GBMfo!|*MC^e;_uz&1h+StqSVr^ry?jtGf{i)NCpRe8c!)&WjXv19xkiJglR}H|$
z^R@OF%v$9#*dEsHd#36SL~Gxj#zeMqXd;F%&i&;Vgl>xheO)`ifB|#@^e1lsvOCa%
z>bo+Q(13XjkZhf|@;>I9<418f0`Xv$0pQx0p!e`i4A9#~XFdXyr5=aZ2Ojqd1p;l2
z@7X2i{!Fa<A9$R0%g0s5{F812239()-)4vOb*|qXU7iAKEW3Xk9-g@TIsDxHL+<Zp
zxoeY<1DBJ(JE8u!;NU$D!`EzJt>5*F`^Z1<U)*6{1nExQ==*GPQYyUyq<RT_LTQYb
zFE>wRSf6dYwy}^msSsRL{MPWy?P+{vv)eSqd)}i6?CNCGTLERXrV<ZOpH4Fmz1Z#-
z*b0wJ(b3Lh=U4HJ_T!?7f^gBaz=yMWx`)E!I&3fY0~r<HDOuX27-KY;6V5FYF26A@
zXfVs>Z8UB6uElAP;=av${IB3}G%KNY70lLpdpRC*U{}<m)hxlY^h`=`#*hRjOR*Qq
z=4m)M=w-#Hhuf5-`(~2?|AFUj5RbY*Xo3zqEma3tlZ3F$i<IVo_1rUap;}TBn_|+2
zc`YUWe51>&PKIwWhwR5|<=KV6%8o4CLXMpz1&V?+OZnQxY(q^AVlvv|;W&0JK>S<V
zf<!e<qYP71`-u(s{+-f&oE%sq8W*9#oQ3c&%C4J9)>PS*rBoI6-^+_t>Tv6io{yX7
z-MkgttvO?HmR#Ph{d6QPT@FkQN3K%?1z~z7suBslv+=6G7R@OChFyJEFDijRV)7sg
zu4C1-c?;79g4}UD+O>9p=lGO<Zv!+P$t36x0X<f|iX2RGGI+2c)tTQYH7@m89(X2g
z#YjspKeBLS-%anVQzO6OKdIPx)%@!Q`EytPcQYMiU+(|!6&KC7zg$9qjL73HMT!U&
zL9nl4h9Yqh``5HVY`U<0(#xg<EGMsGw&;fv#o`A#2=*(rqkN9|zHd+DU;Hr$x6%~Z
zD9u*}2AQfW<`GcrvspAy@>?fG$U+?(gFa=FMtpjg;JN6%I?iDD(G$q;mI#mH3nagm
zlhpB==JqDll8BFv&xtx~AcU@V%o|Ib7&ic_`+Qf+H>91R`f!(ZL`+LPsc6)y+VW6C
zS(?f|Ahe2AQZKwXbe~WM(WvS-TL(L9;WI4LofV(4Ff8XaNKH~xQXDIPt_}?=1+)EB
zZys!Ou9Etc;^xpCDjbtrKRyTBAN=B4GzMJ|<de;|!X0v9_MBOCWu=qBJCOP;VUxC!
z<cE_hsmP<U-Crh7i2pW6HG8m_4wp%yWIl=EP~(Fl3<@;<|8Jv=yd4z6T)bz;T)59p
z>?P>~qH(m34-jae+TKXMqc*>4U<v$<A(lz~(Ce>VZ4IOu0Dguy9_ZQ=;t!jv1uEU8
zaYu)T!NvMA3N{{@BK#FtPd|6Dt!)k?-)*sn>i9BS4{G}ZrE-u2XE756_C@9)ebVrv
z>&+z#9^7mLv1SiIucmSmOb*1J&ba?p%TS=Gm#_965PNv^iFHVg@;p_=?2IyzP-<=J
z*Tv@(gQ=0Ld%>Ctil2Y?IYG5^B+J&N+%_Nita>ewAD@}6Uzh<)K2NZ%pTwT>A?|m8
zlOs{%J_Zc{G)u(57A!8O-S9hLZ<8Hhog+U3HXvl;No-&v%>zQsO?x?eV8IJseeCNo
z6lPqQYrsaD|GRj)0$NXYty5gKUC3V4;6L7~K*X-eYMvKKu&E3uUaS|J-*#3yvS_FE
z&<36e<><$hPj3zsPS0)l(OL~VNNm`f6GCab(<Em!0kqhn2k00DaA4}(axO>u2Yf+M
zc1Z+vRuA1GtHXYDIadKuC!ogrzUb?b2C1!Gq%i+8wWgF%dusqfJWtgmd=w?)SRYQV
z5);gf<MM7anyZ+AJx7~)C0|8XtWNHcasV64)}UBlIo%%lw3gTq4s|R2apRhXTAVGw
zZ~&HeZn|TFJKQPK7$_yG$4s?VXUrTc+kbOw9ehl5tjX1<7RxJ6GV$(tmT>=;ax<C|
zAZV<93O#CY{-84?ow%=j8mx?BSq02<YTz67B=>2!B!h(NWAl!>sU1N(s+r%N(Hq#U
zhgfMLN>!;p7kOIZPpCs~5axp)Nl(gw<KV4_A_wOA+LgqbYnYqPiQb<Le(fsOTZvj}
zuzY-U)8;0posTKW5Ht3;pEz{<gf`?<aBFn5rP?-UJsaI@*`4ot6a6pHW<0SG2jl_5
zEwlu5sxEX>7F};n<jf^or<$5JI|z7RmxxR?_U73SKw<FS+YIWweIZTLZmnjiowD~4
z&;=&-BEqGtmdjstQ4`E=3nj@(pO%?og?T)xoDPXYI4t;U9nQW?PuO&DSw@@aQA?LU
zQn<2I;m`##{K<zY5JAm3ZA~@^?_dS6Ox*Zb8DQDfdWu>K=uI0X_>`}pkip`IE<pO&
zgMXachm#Gu=gfPWjnpi0r>bV1FE~vD(yMr=dCd+PkRB8xH?}y8LZ`H-{_RpIrB7V)
zd>x}6_oNZspP>9_RZ|25P2RJ`);{V101}qFFsr+`R35JNU!$p1eBz(b?!aYW@`onS
z<uy8Rmy`Q{Ouc1X)9w2|tSG3ImvpOi=V%m>kVcw~jUGK31OyRKkVaxO(#`0R(%mp%
zbT`rs3is#xdvgCDZx8m`d7ekS<2aJ6ljA?V4*L&(517b5xIp=VJwUdeExu<&x}hhh
zS`GqC46L^-4aAx8!-NK-^H1(<w!4#S)JgDP@NN^q8>5J{9HeFeN6|cl_h9RfIzFK2
zRn39Zec;rI60Z1JzPC3e>9ao&h0$OHrk)>2wDd4Zif>&KkbT#&mJCr$jAL|922GJP
z#wXD2vHcC^L!1ojr>~DN%)=c|?yU|-0ud^hs<%hAn>JBazphgyH;2US6en|HbYkAU
zdY$orHh8j{mTa2%szZ)ct!{ypTZZ&;uPGz{3$$~JH7RI0Z-FR?_*HbYvd-&5Wnul{
zB+*09gQUx*%jBNBwSO1GB_*$m@yU4%ujMV|6{db~{rP=iwB+4#>6^Y3>#`--Ici`v
zp3<s5l56t~BMP>x_6-J=dEU!JEu<6B*w@;RS5lmY+X;jSu9mD0JhvLuHhrJxRf_K(
zoMnGRmsVnZ#b*5GbsrbbhhAb`B7{Du5T>y_+F(<zh=0>8pPJBgWUby)bM@!9z?-G+
zad!pHey1c7D{^$<>EK<<?~ZN9Vw|JT;9qxV`PapM=Vd3N_nS-W<}YulJ{`=j{BPm#
zfbrdcUlM?Q{JPKc`@aR6S1Nuxe7++sr@O6IjljByhWKqS3O2i(kn-8}fhreAo88XF
zeiuX>fy&pR)mDhWLUsK4(u6Y)KOt=#VsW%{>iI`Zbeyof<_oTWlOT#<UC7jWjf-bt
z!|;TA+t7B_Gmx-z3w4ma1Ck@fY<b$vHx?!@TlU|@*fIBtJ@V+CzkUkjF{U_We!`vl
ze3my2xgOy`M}VC$uplRZ>xd<>?nM?dTeG*-Nc_9CwFisfyx=J!-_f4kMCgB84dPur
zDJYoomaoRsaPFL?r_`pz8fxl;;-&r$Z)tf_#@AVi8j1w_Mn(7<lCTMeBoI`^+^+q=
z@*%AwrsY*-@q0_rN^|6?1h%Kt!MK+nUMszZsSo@I(lDXEBBt8j_6{^p4y|E7EVgZ4
zOoAcJE(+oa^696cJJ)~CH!>e23ns)txwQ|7RkE6*)#Z(F+9l`qu9a2G1c<fY`hHe4
zq5b<O3TjonaguV`O{YQHklssVsX1JlCi3J>!u>J!(6w3h8m8eeehEVnN*RfR72}LF
z$BtDAVOvwkk)_)qreVux^5ElX%XbT{7D-vzuA{EjpR)>+^2JS)UHnqllAT6R*=?^v
z@<pyU^n-K0ho`GkY(Q?Ua1Q0(l|3k5vKQ|b^}yWMqVLo$`FjM%YEtoL0z;KMMAnU}
zg(H*`9zW-#9(ZP0iew}nV`s~yYn3zRiqz~tI-n7p79mp~k+ay$+^Oz*NQ^{QgGIWc
zRjC%!wt^kMo_z8y<GD|)j0!`!{EU;tNN@i(;<|cVX8xdVejemx{|Fnc)VU$JW<eno
zvC*i)6g)64tQ3|U2G3%p7v%Nep*OC;foX$!&^85TI>fpET1~Sz%xL>xK$Y0a3&{2}
z<9ytt!tneRx8Qx8gWZ*Gbg-TY87!ggw?pw~uEU$p;BR741$megxz9cbre&v&HMaCc
zI8*sCw|FlxB{ko#;p^S8Z&hIBxP-j}TbPRd<_>pmiMhS3L?(~Qnwy>rD8%}|<qOlL
zttTjljIkej&PhMxSWx4zy<}Fr)|{L?#{L{kvVM%Ok6mMgIlj?`Te={zwm|Nfi7n0A
z$F)U%on&ne&&x*MU!G>MNQ9GM?7^~FKlOJ(89T0PZH%I~QsOhwi<dsfR7>MeG0G$w
z)p#SqPcfmVa0M+rdpzPsSb|dGTH<n|#HSCh!QkVGFTreOwMzZTHtdnc0~1NX)Y<Y1
z*o3Anb}lMjEh7n~8T&>zDS{!4*mh{pwa?F!S?~`8)gL!RQFo3#jrwNHQx{*Z40c+E
zNVB`%C(r8$g%-t9gz$w9h|Dv(`E$6w)a<!{p53d}1|NF<EBo^JZ-|CZ!2h8cW&dhg
z(07i0_lNhscN4#Ew7;4DK>3-Y7DXlB7}vXONMawnrE<MxB-=Eu!B0T)?HTqlIQmBo
z6d9RAty=|zRWYp}Gt55aYH0Ijkjtv17PWUJcg3{+x7q$_06Q!tUPx^_xMR@%?SwmX
zn!tCdbxQuj^hK_TjW`8O0hR;S+C;)OW&N|Nx3j)M5!aw0^rV^x76D$bF+{#I^PS59
z&<DI6FbnJV0+XK<f_a5p-0t{px_^EiQsD^U3V!bcM_AH#t?k}Ypm?%f+6E^EF#({2
zMA42>Cqn8#Z`<r%u8K4Lc#Ft9dHJ|W5U3gvo6eUeF(SfV6l9ZHrKoaJl*PV2QhFqi
zJpLGx7hb_FYq;jQn#HU$yZ}U9H8s?C^Kq?;w_Mw3!79y3A{(R9#;BliAF|v|Sxwo2
zs_FeY=*!1NRrc&}@}VO|$Vl@^5I0!^>ma8Lxxu*25Z8`JqJ-*d>WZK&26Q)idpUFX
z*=ObM+o@N723&iRs0Z_H0X1+-Eb+IBQvJF;kR!^m*3-7OL;IG8A(C`ajI=|^B*inb
zZ?d^(Z_D+|#-s|=kd13SCpDfC)XO<-A0lp5wec?dcrsWvl2sg?TQTQRIC{0@YaQix
zGoCBO4w4fpJmN(MWhgCr+?TeNEfqcZJO1llK%}ws)9<KztQ5|j=t9R`Y|A~oTT>ZZ
zHS?4kXCUJ}Q$O>aeR?muidh#;t}sQK(;|D1Hw!&#(#rf5lPQRU%9VTp?Q%oI6&K4x
z^Ug47r60mHtg`yPyTC@1GwFvatKK`ekXX!=HLEQ4h&~W)S2a^P?GSP|1#+u|5+{U}
zyyC<49rxhz_E%bRwXO~V$2kkC)jt)YsIZQ92l;$}rooKAUcgH^SPSR3@-j=)dSF-*
zxA0jNlHAhRXX+tZFS82mvhB<q`c<xlOgVu*=ypkGbl>Pzu!<oTI}!`$<VSGibWu&Y
z5Q-S`ii$q^vTPg0_2^egLH!%3%NC;r)ushl2BMU#rK$f7+$G3sut{<`=CVGz;6c=-
z7Ux{<NPYImkOe*QEs3AHxS_2w=aDK8uIboBwa$20DgL`2lIt+5op0c_l!O;-Rf%E=
z#1-b-S9#n21ci^C)vnXtj4%(XDEWrEghC{Wz?zvHP?baqsb=y~7q!K@@I=4>U}KkP
zh5H7(x~Uh&Mf=w3G){Cl;m;*Z1Va2~WHsaP=Z4E@x?EB@V=+xfoP&K_-&toBk$1D2
z<|McFGvf3f7Kw7KyO?oK9a2e8FIBY*WX7dEsyFv~*siEywO`?#(NuVQc;1Ws`YrZO
zHKIoP-~sq>eEPusQPoni_sPY1OW})h#GKEz?8{Z{UpEsC4@&luHGFrq|0YSFS~c%1
zg{)oChDyo0KWNrP?1>Rap8l4^jcu~1Q^mCgm(7<x7}s@;ckUQh6ONfwJdKgt`GW+@
zR}M}DvC*pNA=ke%!%D)*9G|4$AeDNAXZF2;q@T_uA7q!Z=d8R?5E;vnZ#);h&!){m
zYgHBZBs&@$Q#@?(dupPuhJ>rZbk|_kFiqsq&RI2dY_<m4_kD068O#Z8>A}c_lhveF
za?3ub(!b$FmAWy78Yd=XA%NMTvjORjRT%UU%u8(4+9G~?8EK;w2T$dE@*QDaO&^Hb
zD>)WUX6;TWLe(4e2HNRnctAPws_|AU83*KRy+9FTW1ZsD))V-$YmlIZ*SVx~>@PmY
z0Ft!#Z^(3;;t@{Mr_<J8H<&=Ubq1ZEuqNl)mRRvrxwbZTq8C<9iq6Eis5u~>@n-Hp
zKs1Bq92W->j9%<@K{vl?qBaG9Se~dfwTmyG`8sX9y2PLpcTKEg!kuBs;W&2WV0BKy
z_zV9Q*9UF%EO9U`!)4HmX82L#ui4e_AB2Ev?O4Jeag-hP+;hqF<%r-_0-y1Z?6jb?
z$N{~9*F3g&31zAL_}bp-tsmVD7w0nI0xF!|$YH_Kx|+%ETK6yK=dKKDu6;b`-2Hhg
zQEyv%tQJ#8kC56vnW573JCGK7D+{iN%c4Q;ziYnS(%H25?Jc=Ix2wm~bP$Yi9j2s)
z>z19?NTS-lU`>nH#&pL6vTPH(II;^Y7$~W{1B>^pkz|YkaWRL+so0FkX=K~nr_BcB
zbPqpI+?1ZpOteF$Ro`!?(er0e8PrFWPku?Ew8ENe%jBEvWp-K2R!xeYnaRZks&P6B
zS!*7Ryiq_A0S#m>6HDXm?Z+SZkACU-*&CK){vVYAVe;L_+?y)~XQsM5x9WWSGl<4N
zm%iMOvPw3`xG56sC%<yW(W^v%yoE(Yyz_K>u*Sb8*q<bi3)juu8!sgzi^yZg6RrZS
zxgklcUch-+@kRu7YX_~05f~SqYJ~#Qj)LW|=d6EE>?o~4ZoVmdkACG}N=P+gUk@rr
z=4pAB_K(%S@6FL|vra1J2q61|n6nr~+kD0g!mG+@1-Su72a|sHXtVzt<(AgAcqmvR
z%vJmWC;cijZvp2@qc<?u1ND6S35BMflOvU@l$u;<hRlG7O{0W6j}o<R_ShwN`XXXw
z8;|WY4g)Uv+D{YwCW=x$L8@wQHDm9jQ!gwQ+-@|&($CrK0y3w{qdSxOa({aXT#1B{
z1$s?=;jOA~qI^jE8oI9cWTC+})?!3rNCn-nBo{Ifx(IU-LlO5!+3Xzx!a)P6qI?Nw
zzF{c7CxOqQJLV}gmmdU7AwM+bgOPTwf@)Oq)Iq$5x|qke+M+Zg`?V}-MYhxXzZ<_?
z*V_I=b3S-@H~;I__3xjNJMRC(M7KXk_b*)m>zRT1+p&P&0_TYBuJhOO^72>XtMA4S
zmjB0=01M^s-P8T<uFs!dZTa18U9YoA*YiI1E5<T%JdM?=x!ExNRUnpA`DJec|GYtL
zQl}?h?}a1RMZ9C!f1U`!3UbqV3d)&Xu%4(bs3i3hV<&Q`st7l)NzsLuf_kkMH(jNZ
z=xex+n_B(qRLT1^B;&Z`rXib~HZ=jE8dDi(dyYlSI+R=JBui7>kmCdbT;`sx2_nwq
zvM@1-CAk-FGuNMjNG1<&Y)skW?q=9&C|t|wt%jnCzXu9`Cz6X;$_<g`2o;`{U;{t-
z$h42n*VJtY&wVFT80Nm(Kbt%ME|34Y7r0d8BC=3L&zff~B|eqgNfkvTMSQ~`i5M#(
z&&`13IiteGHYDoh-?~jjBGh6rHJKejb|LQ?x1J3)kWfxLB?%C_Qp1XKqE-Yq4*1u7
z(*Lu}n%S+x)4zQ?he)U+)Bc!<uOrN3JB#WQx}{HF-23?G`G*(AlF2XN<c5e8!#Tvk
zbJ_9A&A?z-d)9i(ze&jq;rV%dPX{1w-tyc{+A#&0Dj3Dv<ME2t;<T5n+VyB37zJ5_
zs_Fepi91qhAN@*qJho2#s7r3Dd)gMr(11GL`@}1ro+pNs>3N~Qa`?qIB{<e?!E4Hm
zYUorra7ggmbqz$q1a&u8%W<M7fAe|IkPjx$C}K;to3u$o=u<x^p2iZ#=6a@3jUf5O
zD!#07SW8`PFNJxBVJZGh*S`eL!Bt}dHx$UqW?OJ^tpx`yTiB{~)pJC<6`bV1R43Vm
z(ESS<Gf5GAxZOa}NdMs~75yi|ESa;9a|~a^_7{Rikzh<)I=OTHt#N^;3{el#EtNLq
zEjlySQn-xg{dm^<?98=+Hz$FjxEDrb!U>xFJIotqK<T`r<?@v%0%cw?=PJJKedZHZ
zBtv}R)z_7^?ny)N_cz6SK(5^xPT0Ojn7<XqTKKfutcm-_X=+x{PMydj<8fbAdDX-Z
z;sZ^q4pb*M-&K`Vy9Dg(1sT39nLfdS6hmW9<-7TQ`>4-=@W)omcYtG1)rmaLK#ZOG
z^wZ1H8Yx>H*{3@<D|+&W6xfc{J6=)H`Uf5=CX{VKq<+tJ2@uiH>`buWuCX(z5{zr*
zD5*2eS}&1=YiR0A(G5^7ej@yGwD}duG#=Z?^w6dU9D=f|rlT)KD^6_9f!cXX_mZ22
zL+Ao8={>wap;wO%mw?{I8L@nNGJ~@;USGN50#d(n5nmA;oY|R{19xTHaE}E=_Av=x
zPw!oGb{K`W$U~M5fuE#evd0}0)7bHJVANWYCcB{X6QRf>6++2Ow}S@^>LgNzJyi(u
z54QY~F2MW3RP`1GjR|&C5RQ|e9O{~!(d#ZrnR^nk{I}@K1ILXUg#LrO!I8g&9V^i-
zN0fIsTU%Q`M}aTbDgBNqvE9+x+1X`#SO4Zm{?R4p9lJ_jx~8V4me<sDu}NPm-Srz^
zb1l_kJ?neL((=WSRQyufY)*8z?@-vI^@fMqEOvDloO!A<o4S7^;oIZB+D9XKV~V;C
z8-fGOKeOWakNJ5NUHo4KCt3fleBWSGtXZB-NZHI1LLLAc<tLJj_^w6>X8>2=vCyFO
z=xYNs<Fwfy2S%^3JToueAFurW(OAzVum`MUBtn>tGg4xG8^CB&j}ux$RyR9(E^Nf8
zs-lp&K0py)=Pz%bUG+lDAWs3NBC4e?p(q<W-=`Jp{4+cx2Y-~TJx@kth8*rI++J!M
zQNsRANVxI?C@wG4)G#ZUvuHB$*(|Ef2)nl1x}`kQgLZrPj_c&#Kj$3Oz2Q$M>j)o8
zx<LQ^47*-~Y$TEAs7EKYrS&-dpfWGTjs#X)gd%nT-reEc*~J*I8TOjvY1AQl#zuS<
zYG?`N7Uazi&qG=un;lQ*S6H!6-fHq@{}^c!XXbk>@%zOz-G!QZe#|o;*JAk*q3aLE
z;-+-{-7^GqU#>R42FL#JyUHb1(zXEsw~j(H`E<39E;&eaB*E=XicZLVa9DVhv`DKU
z;iA|tfk;=Jfn{^7s6h_Pryv3bLMB34tL2n+-Hzb0i=8LLW$8R}tRF*~^9$sctacnX
z<jatbb~#?BdM~bc^lq0jgf1lEu(=~D_6P@<s46Bv9?C8^!D-y6Db*iplj<s>u^}{O
zYp+J^j4CZ_x#XEMs|EJMI<9a{C>c!-uEm%pR;c2)XXUWl#Z($up=lbrUzL}uJ~5=s
zQCa}WdC8_egUM8SX*(Eal<yPe@wSLM5#FSCD7=y{%6#Tpqw{2qpGVi~W3!qwQs@~$
z^OcEm*4G{+FKnlyIjqnoPPgk?Dt>}Vjd=IgH_D1w!V@X8@RVqE476&2baABmNyZZ7
zFnUf%-1zqMn}jbas)k@qZPD{tG9gG_>JVuqbhQIS<TGbcY>shR2L^_!CCA$>pZ71b
ziBZV2^>PkVa^zQiyEQkDh;mU4v`g{ttdAGQu&;T>6f7~_%IJ!w4wAI1He?zif*qDD
z8`qBQ7g&Z=HkH~o7)_}t2s>EhITXGF(oU@t|1ZnsbI#&_8l~#7-8GMi-$b}DM7{^!
z>+lwJ^5QZz9?wY5)~uUvSY=@n3kw)?G_4o*(#z-FNG%vxzQHh3FBERTOr1=O=&>l?
z;MZ_1Dq-YX84ANyGCHpH`}Z2Q1f?Ax-V{$}4fcP6iK~K#I;PCe0*+9lgu$wq-zeE+
ztZH^@6m)6Xg2MGyFzYah&W*kwFOv0HPnBGZEPmhJ{tboxEdv@}Ia4_*oR*-zIZw|k
zK7Azi=O(7g-dc*Pd(YwA0{oFIUUiBrX*D8@#mk>r{F-%y{_L)F|4~aR=<EGGWNDx7
z->(+We*LJtRaIGSaBV}VM$G$x3cZgiu&%0*ZXRVWOHPFki9T5`xXIVW%WUC<KLc{~
zSc!A16foz)0U2x5p7_S^nF98*>_7h116kxCDO~!8$!GjFxP{#-rX=&i!-(XsYbP+G
zh7qa*-uvN`{%9)_F?9bu$fh3M`VJAIfccLon#w#!5*bIXM%1v#Zj>Nq`p|yB!0M{u
zeKJcQfn}lnyUh0opom=vWEMh)C=_eforma6>^VT8Kd1vN$=%t?RP@uHz$n!*M%a~E
z%hvej;7mL2i9?t`6|!*q&*N4e2>XZE`rat{xmec;oeBzi?Rob%@ukTWF}<4DfGK{E
zZ`zI_c3gn|Lp{y+ZB7Xl?XS9;O*~|~`I-+7N{V*8D54IlXZGp_V*&=``<eSW_ADdr
z!4FYZN^O1Imc{!nh>k7oGT2f7PCyU(k$3aZ7p!>|{In*UdQB0O@v<2$Ga`CD9!TxG
zWJYJ%<RfTRw)Mh_+*$9xcwxUI-^M|VmebShAN#0|=g>J568K<yDo~8ZAj=of57Ih_
zerYuOMHzc<1bD@qn2?#L;B~+=8Od=4;*djFn|mls;_FId#>1tK)~4If*V6w2%bcV%
zz-oB*+Rb)6*qlS9U^T38jpGDo+Xv!MQ`Y+OfgEOT<s#(^8o=9UPdYN;#4WxK;WVGB
zVs5V3O4>UxFX?WJ6*H$H@AOiGAUV%vwJ*`{3LbrM-=a-Z_*`6Wm>2)b^dwY^yfis)
zEqNKNWp9t?a^*tb_Y}!q9{|mSxWT5sXhH<FVq>-3_a61cF!t06xTl5E5D*QJxM_~m
zY+t>9I%@snBnJ31vz6M}`;|g%(`KudPQ|wC>^(ZuXcDq8@;kxwDsC%YnxyerzlZT<
zWRe=JJLzFPXb?B%W2@iK5-6__LpU;ZMChc5)k#$ZCpgm1jE3>`&S&Cgrdjl<e7h9K
z)EY!P_PwU;gOfSLOiNMTYdwW-Ff;E8IHn4k8CXOkD_YKd#1mGE9Fkdr=dX=<u;96l
zx;XKqVZiMwJ(VAFbXodTcD>eSC3xmC3O<hK(9DqpDq^b)@^JZnaXWDbHkBOnq?OSP
z)iqF;Z_G7)spu5M)k5r%9E#(++QZD#<8|ESmp(nE3qv^ygb{em5j%5q2Jn}9iiK=@
zQvmVJi`gGXe!*nV$$iv}sXXK)CRT|%Sx?ryilxl4cB}srdvP7qf=<m`Rjux4#5nd7
zEY=3<gjnu-$jCKwMz5f$ywIz7>xNvdm+-rk{(E(~(d@V7UG`wTvE|1vueDVD5hPWA
zzwzYw>WFg1EuX8Gffc-F6CHA4H@Slu6BdZ-eWY%n6+?`ZZ22RRoO)8H#w}TIwxZVM
z3s_chqo%~oM{x70{lnMKaST&yB44*{>JD-9Pz5XN02`OTvM-srJX`F}*#O7<-lG)@
zDxduZFDSNVeHLci&2p+r!cX&<4>M7~n_u3Aqpy%cZ~P^QI0xRv&HK%cibF^NTv`Sm
z5MT+`5+HkIcoq@KG1yc%8;7)!fr;)pB->DBbZ??LJKx}bU9gv=yYRu|Ny*YH-(<HU
zfrn?OCgC4YET*qN{{-_lcn(=@0y+&KYcujVtI9legYGbQ7X>!%8sTw%5c<e;S7eje
zBMv|wD^#EwuMoqzigo7aSqRm*4)J1J5kulAM_izCRpOQ}TDR@L`V`N%KZggO<kxUf
z?Jz0ZD_`*N+PCkva){cwr5G{Z7tm04AcQc(`?%O#fu0GA_wZ%o*c$dA{GsQaa)RH{
zcR(Zs>6jquHpQp;`{ZSP|9)n8VifC(Ti;*#9&;1IayD%(-fIE;TsxjXX{Z*_{p)Tl
z#b@(^1;fAAg~5%?6wv5*-*m-*ayGCjQV=xi%#7&Nk?q0yZ=54ZFFwD+s^>{^97Mvp
zr4@q~+>}-<rmPMNO;FPqxyTpOK{s?f=viJA%jZ^RUNZZ)=5=Ix5WU`|NpglR?Y>v(
zh#yJH3qZjZlQ4u-8>!_Nj$luCv*MEEQa?z=o3q?otlafhnK`Y{Zn`0;5Y*G2&&A#!
z#Lr{EOs=U}ddMY<)#=P?vz&wH?^nJsm=01;#B#@`R%`j`kv0uUl_vJ!4wU5B+7hbi
z2Z_3-;pr4{1Z`2B`KBQ4#!nGAbFcK}ew@npcS)JXkTKU=$HC$keq)cPcc<N;hUC!P
zztk|pCD-Eje0}6oMUWL|xfTR%J6jD;U}cebTAoz0f*e?BS$THca)7lglhe)zYtd0_
zL71cu0M!k)1z)pBAhM|mb$M;*_i8hF3JdR2b)np(!&cPU>@2D(-#P6TDU@F+ai<`{
z0v%dC<^T+)660QnUWrX02eP%-%1=76E^ez5@sRY;kFWxTKHJ%k!bcMhVWNER(Kumz
z3`0Qwrv?^na%osTE*1HJ`4-r09iOimlxg#Kc0mAwDW7w%uY9pQ$7<}!Z`!iw)ZBZ_
zT{rG15F*5~XKT<YPQRqr{Wj1`Yt)$sbpahlQfpoyU*l$!E^lgTj$hi34Y9^ZR82a2
z@#9~Al6F&s#!1n!RKo%j)a9aXv_}pEq6E90+c1rYyQ}95SIzU6htNavuRfQQ$M@&o
zgMAJbtxKPi2G7lPZY7%?>*5%R2;(w|Y~=k%Y$L1r9zw=l<xM0)e(Jw%p-R89#gR}B
zSH8=&+MY(M4^-@U#*o&p7!pE=5hI-<;pgnfZDQ*_ycKVh^t3U{oL6U@c%_=am^(Xw
zPm@6L>@eZ${0_y%$HaW}*TiUrj*BI{;Ezastfo!pVu(nPQBlv3AX$DJK>+Mb>|%sd
zY_%5@!?XA5C{p}~EC(|83_Z~ynPO(eYR*Y<;aTBuPIG3=gfme+YoyCdNS=a8(li<s
z1Nl~5v+yD{!fsa;u+w_=|I0ME>L`i(Mr;V}d9#E})xhR#ffIweq3}>%5*Zq?yiV!(
z`0cm!y30h6m>FuZJmnw(_M+qpvxme9N#T1x58tI?fv@WgtYq*JR>(8vC4;%W>;9#W
z#5aF{VN-{NkcDCoaYJ4;u=Rida4E2rrR*`_=Qypapf;hQY4Q5h2{fWIVC>(cwUq5-
zGML68+H>Eh4_LbZ?OA5~wYYK~#sXl^FMWLgd}&bndY41vpjzBG=Z^<nB&z`Z(rT3N
z73VKMet@dBOdTO5e)4eSPfdke8eHpWpv>q$OWdGL*Xd6jb`gG;*S-pxZNFDy%IMU-
zKf^vP8xxS9eX-#p#<}D%a*g*G)yK%uw<y1Q**@~`@xoN&*4KM=-UogMFI(7$oj$@u
z9_eq~pb`;T)njDFYU9=v{jce)7~4kOtY=(C2yr-0-pu9LbpEaGT~B`rwB9<Y6*Vpu
zdkt;{8*eo+q=v+wd3tiz6d626s&p%Z)Z-|CEUTb7Y>+AWxJp+x+kCs74^gnrAWZ01
zYD%r*mc`7dyDl%E?F7kkj;qevN3QEMqGfWzp<qpdvX7J;eL9&XS=DhmVYMt_26?Yi
zU08&j2_w=S$DTtaP$5<+IuUMsY|h;qQdUU`WSp9v6{-+)1#jgO{|NZ*`=15L!U^uW
zRvg4<-rhQKG4At~YJ?rHxa1yPn~(@axVWjAS8X_n9c_zIz**$yGI$y~icsQ)mN60&
z6IyhvF0IvsfoYK)5kp^pH-A{FIku0mfgat=1c>s<s^>jBl?!zrGq;W7j5Q!s7*CR#
zKB}w1Y|1`)b0O+;01NWjvF`t8s+g+JDJ5gvm}wYkO=Ro|GG>!g%X&X5qT6?iI!lzt
z%|HESb<x=5fzCkCs*=jUXBxW4xVuMbW>H<H(yyEdn=h=}s-E!RKBS!bvp?{3M9|Z+
zwrhVNdacmu#zQiFJL5FQ(2$2=4u5p?DqerYz>*wsPVec_5{95|@~XEW*$78cCk%u~
z!46$36`cvw_K|^G_*A#S475jFWkxPiVKWYrfA22$Pi_}K9r@NDF4Zq`J-9yLTV47W
zaqF|vCO54R<aGXxD8J%8dxDz0ZYUn>V83v*7}ruLT@@M2TPHOLT7Fsa#cL+b5dKrd
z!(R?)Iy8is2g}UGX=&5u^wYUsQ7k<-JkH<5T}zO>)`qEkdhg23M?0&Lb1YikizC1X
zYsOXXOmF_CfgGoKn_llK`J?+~Wo2HG9Y=|jwvDuph|d@;t9-6{BL_yzoQ#57$X~<G
zx(+?8KDW^y)Z@{UyJa}4c;R3)WuYsd7n<V5Z9ZR8vZSn4p$3V3T|TLxActZ(9I92V
z3E{l)d8I_}tSJD^Y5LN<e#dS>Lz&pnt=<DK+Z%swFDw|}l5eNnA{Yt0JB!?R+C;KX
zPXdgbrfnqwp4gv8y2Y0V6Hw*V4>b3JJYaXLQ!Q2T6hyFH;Mb0H6(eCfHM0X~5Mcs`
zJ5eA~4O`365BW9%GH6f-59F1U39G!-7xm$mwIa)#h$uf7!AW_t7VUS}PR&QhSFRdy
z`LepcW^!f7w6mr;%zWqXfPRn#ysyW>VR@$V_1l`cO>@zPzt|R+1m{E--~agWT~S0n
zM&huePKT7G^DBy9zeP6*p0jeVrlz5KK9pTrgVOkiuXFWBz<Gj|i~G3tHQXp-92COS
z=kD&CzIs9&yi2cVrA<+!*$WbF8DlelL~+BypSeFr1{7GRW|lj6&El1cbjX@P%f{tD
z$F1{)=7e8&Wzs`Z9OaYK&XWiTizf&)-JtkFY9ME0+wBeIlF)X6dV1mBx^5RU0QR$B
z#7GN&u;vv;nwwF_d@$lVo#gC+&bPDJKwfaLf6;s9nWW(Pt$3{3T<M8Gxt;<Ub)Scn
zN?X##)=;cg-b4IRsLnafQl-{|S!G(NFp()z2(qr46;M~u8>$7?0d+C!mY<_LGWJQt
zkj!j_Vjz{BVoZ@`6WN4ToKizn_5(~HGQwZMXuz{htMGX<k7@_XE}NFdU6+(nxJvPh
zLLz}ixx-@r>^_Hydry_MxFEP9-bM|ZQPO#(HA!Q`q0TWdx^o><pfy1R)LxG1d}4;1
zI&j$$(U}7Yw>0uU)IL2XWou&>WUO5M@?j_^j%>J$p5C5-D>l|(fXg=uWmh@Kc|c4#
z|FQ>KP)+@xnvbRu!wG&+sf`J3mTy!2<Z68&vG0gh6L{wVTm85qBnb3;{QX6{^3mH5
z7z25`hVbn7qqgWa=KdL-;OnBoI8}vrm`KENh+)b+-PPaUBhu|v!U_YaNb?hat~cXZ
zn>AI}5aXLnyElU032!f@eZJ>kG^DDV-A$D(#hN@ESiZOp*h}%}oS;3)T5S{S76iR@
zsp+AN+7z60ArkRPsq#`--nPgkJ5kd751zGk*YfZ9xu5aD?I+YF{=4Seuq$c5m50XO
zVzIpp-<=D|6fzaM2wod|syR9ll7}CFs}X@lCSe;edE;{*>M=B(fHfI>2!#CaaB|1%
zCz)J*=+j~yx@1;ffhB<*2^aiN+<gfJG<|wd=e*>+?kZJpo|H($%Iu9q<-tDNK>Q14
zUgeYbXRqlsDjCZhuISF7K0UAcIV1C*<yXZi#g!T_aYhbT5zeM}nT~AjwwjmineEY-
z=awo8QX}#>`)zdT!enWL1uL0#yGDCe#^h`pj+$=LzawbO+D8`|7cPCBjUqb2MTaSy
zT+qvA>Hc`Zu!KOOU-Y<IXu`BFCYXsDGLw!URp9qVx}JGm8Ssj^)9JlB`%lcnL!g$Q
zT+dD(B!8oR-YvjZ^-YWq`&<t+X3yXTLlH46?3?{O-t3COj!?Fj34+i#!A2&UqR|P$
zFb$E<FzgSl9VriJSYUHgaH|LqV^X<<LagD+MxLM*(L?p5E^(~LqqoO(@AM8M%y;ns
z7U~$7gyIBJs~`X5b*F5+W9Rnv`p(zrZQB0;poiEe47!+Q2rHszd69in7W;XguKMav
z+6$l<Rq7A;9XayB(f*@|;KjcU{b8yemONI|?U(h7`7-9UYfgT2&X<B_43q#Q<nWy7
zrdbK{UX!8|H<8G~QRu6^cUCP8uj}~kLhQTdzgvE(GM5XFCjHnbzuw=roP!kx?9&uA
zr-Tc>5zA5r#^usxi0Z)AF3`-`c<?N(>sFY?1q$sLas&&6Lr8PQ>)+qj<tZ(RvaGa?
zj-3TD7sqqtfKxzj>!t>}Cz7#H&<}ZzE>kr%?JD)8XhzjlIu@5f_^1{MsTwg<Q?ufS
z%1QiQ&d<L07*Ie*5oh1kki~>#q$|={CNUJR#6hkn*Y3tJ?e1L;Gfz*uj@cKG(#bRX
zw1N_`UwtxSI^aWp^U+HlG=vwO8_UV8Ycv_-X&Q#Chl8o|ShF=)MnZ(fK3L}VtLQK9
zqwU7^$wE+Uu^C9KLZ$chL1?S*IhpnaWNa!JFR=)(m9e-FvEQ;vY`j=+>4X2J2b=z2
z?iv+F&+!6SU}a%_>XzynbccO|I-Zg|l7&Mh9Rb`*TAtv2q?%?2Gl|;zh9Eo%X_p|K
z%(^WI3(25S;{RDuGz8R-OSJ-IE>?8#$LQhTnXeF8C7mMffGE*E-6?l#+;JwAPw=hN
zbdR8WQBl^~%!J~}WWJ+u<qv|f23i`8KOSSkN|)aErU}_fs*|llM8Eb7ANHJn>qw7s
zOtJIs$yOR<jK91i)i`{L^CRegP|$DhzHA#+@Du<=0l0<b+=>Nq#W0x)8WXbyR)y%f
z*$37Yj0vaxOCLD&@(4zmA3c%w+_-gFYH|Io|KMc7xbwcpael2hYLhD2jYDsETvwsk
z`hd<uV!Zu1-3xscqr5syry!p(<HKv^;Brj~_MXvEBJmk`m+dJz(>Ii;L0)RByIMm?
zR!g*I8-AjV^|-Jr&vn&_%s?GhBzI0<=R8@uj)v*LU5Yx;;C-`nI%_JQR*;HPw}F4@
zUIWG2kHWE2Zq(T$;v3$;*4Je;xj#wOO|lfavcPi3mEpp&_VMqt9(nO(%rw}<S@xLa
zubJ#f$S*{bwT~D^7{XMpV2(!Ld^Qvj*1AIu*p-TXkw*GYL?-**F$^vu5njg@ZP_*f
zs5YR^2hjhDAqq)Y=I@rO;>+P0RRZ<=XSI9@&_N-NQK{Hg0Q=XCyGC=sCaRYg`Bzt%
zSqGJU47W12s#XlCIMW^SdDd^Tio}yxx#MLWYu75V=W$=fw=U#+HyFEg0jmkS*P2-y
z7n_Q$?PBfHR)5gIhSn&_9OA{-Uqga#{Tu_#cYU^R@N_Xwf+5$LhAC3bq(@6WyJ|*~
zVV?J<VK&^;i6wDACxcPv{6uLZODRX2T_n{(|4<`BpAo7ZuXyplT*s>nZ$0zJS-Z>J
zdY;IJtEpG9<nFcO*&pbvx+F>+gbN~?qYtj~Qbvk~T3zcsDTlwS-H+Z7*3^skdBWjn
zd^O^iy1)1tce&P3`aa5d*_w`&G7ILAY*{M$D59u{ohG6IB`{ea;-u!Iof^UQsa&rA
znVVNU3_*>P>?oP_%UvzDsl>pNMd@|7$W8z?z~mE+Ah+d+oGn<zLvF5hVP%r?T&!E6
zn3)Nj3O<@%T{=UaW5!)!m93Jslt0yP&=U+<d&xKI{Zw;$td*s$-TXwmMT~=70BVxW
z52vzJ?Et0(#gZ3ALw9P)jrs*F!WT|?AgJ#MDls%*0Ntsjn?=jS#HS8xMOty3a_1Me
zR;h^?veNp8WmonlfR!{w4U&V(&N`amAKB|XpgJE1WF$Fk8Ql}ngnbkH*JkQ-`}NSd
zr`w(s&YJj)-2Dcuj9!3PPG3~YLxML}Y|ZY2tC3`ykcJAktodl=%;D1?%?b^5fyg1!
zIRrrSJHM8nu}E^I&P!D%-2`)Z?WCIwf?6x|yBQ<kT7ratqC)^vm&0w_4>@*|5zt^U
z$gt`3Sx>|ZM9cSxB(jo!rjG*N;F=^LKQj+4T1_ziy=?bEXf|CQ{4B^me=Ry*>+Caf
zz1*>e$Ke;B(kW#88|bPvo?#8HZZt`E=KE8U8CT=L<H~pB&(S9Px;?UgYj0*SD>d7u
zvvgm?Yx1Z$kQnH;nfU<8jTa~cs41v*7e+nYlNHv=4_(cLmRFd&N0;r9PKf`a^@R0r
z7(k<T&$y1<(odHDR=oWG0R%qYWm2Ozu>ZtZ4G|n?4e7smP5L26?6CzVd7+I!dL-82
zXa&*1I<(jmn@Rv?V-1iGZjf<`PK^q~zw%eO^Q#ffMgPptbP~%QUoCamO=sh|M_x`@
z3l6Jn+Q=m8SinQ_LKrIcAqIKd<(5hn#BZWbLmMd4!ch;plUcir<^*Q;l&t`gzr6$?
zItB5&hS(a!W(<7eH^<I-HsWHny)p;6Cx~oV&ir7pME1q-syTN7H??CT0zw#RX*Op;
zana@?JJ43M%{M;{IgODqIh}9_=6XaOVhLBvkn0-ASw63dZp>z}0Rs=4<i>XtUd7UN
z&{$eoME=CS?Vplr9L3&c{+<M!B`GE=4J1IM#+aD*EUH^T2`u!K8|8>+UjuV~pMqb$
z3&t;?7mDa4YQ#9tJhgX&DXg<$z75ESn$rO4fNflnY<PdZQR8y%ns{EOtDR)VnsVfk
z19oE*se##+O8w@uvX}^lxY$*7#Y!R;>4N^aLBHI)Hu_v!x8GmcT?!CIu+<uV^HN);
z;-o2FnbJ$~iaCwX<^YmGI@dEo?2{9T)sl6LV|bl3bld~i^f>sJG;qETBo3wRk8_`B
z8%cmiU5NU%njJL`t37?GGKGnhGzsedB4EuOs6e=NcssP5VWalWM{n=^e($i$egD%x
z*Opj<*88hNpL83YuX&|wPvJBl%W9m{nU15qnSRFF)Q3{om8&WZOxsJk!3gFidhwNt
z6IpWQ?pVleIEFnQzOg&=bxxqk%M0}u7!c-SN-bQ0P&5a;FB>$gk$7qaFLW86{3PeB
zTMS`1gD^dr%lU6cO>Z(LRY!M1S<<R-mEVcGL0O0g(hyWaMug5541Vs7B9lSM8!2mg
zI#QjO&;i2RO=opv0$B~Qn*4WF%Xf)^$k#S3Ts*bK<H8q>41q``%_*tmJgL-9?zMoT
z7ndgrCPLOp(?PuuLE+Z|AZu~FW^>aWnh<5aqLiSfxK(l0+*hJ**WmPBVxxtpp7-&@
z^dpuemQu-`P;RZ5tj<r|9uVHoZv;S1UN25|lM>s%x<eF4YO!7yJlNHnpq}%6ie|t=
zKmcgZucbCDko+11bQqJ`z0u2kVatJ9Hz0KP%XfpImV+_xS(y6(lRpc6$GGq;YJ2RV
z$1L$FaLIYC6Xms5XC!<_So;Zp=X$(P@rjReB3WVN(+^C57i3hQ@U3icaCTyYjT3hg
z<G-U)9tJf=iS$18p8S34=>`QY5$OJHH8j>aQ2<})#=pa3ZpG}UR$PFlopoaI@M?z=
zGpP)0%`6>Xs%76S1y5ri&*!);e2ciD>0UVs;D2v@LQpn<Kw|Y*i`%e9i(|sl<^0|M
zOC3D#e*HZ7MaBC_xujHl{b)04xV3eV(?45bqf8q`T^hxhFuXj6WZ`cHirYvWE9Kxp
zK_VX&u>jQT2V3%5m5R#am7O(u`G^^Iv#!BpmhW<Q!h-6E;%&wyim%7EP_8kfhShDE
zqJmp@rq!cJuKcYpD&tZVN_{g@D=x1WT8m<&v0B743#deCoXK+<9w7?AwV<O+`;+4g
zK!xyS7<q-L>7TjUPnOOdb#3D~NrYX`j-@83ZvlL1q}}nALm2Rm#E?o*o#}qE7;>Dy
zO`oU*R9v{Vw`LVuZ;2ZLneiUc#=_Kw)W!xzyexH?`nz}#Z}#Aq<)`hrMY;*&IueRr
ziK#<o2T>%Il@MAfoF20MbS_j+q@7?0K_r_%2R|Dhmyhs`2svZVgnLXD^AH!FzwCgJ
z7k}v0NrwO*17BjdI6w?Y-%Nj%>Ctx5N@*(bI=658ZQ2PMUSG+y^I4r@9Z;ReNT2`V
z1)#zEyQI%PJ1cSd7Z3o0#jsQ6XLC{K8S<puuCbJ*2Qho~IrO|P!!>EZ9RH{c9a*G4
zxtWPU+Kqya{?@$pdL~-y1BvXT-68-4ZC%1ZTh;bkv;RvEW0~pCb-AQO8$u$N8m4+0
zH(vsf_;MWiq5FfDt{aD?f&1f^KJme^(UrF4ZV^btD3QI7iYmT52XXgzoW(8oz^MrX
zPGU`2=xo>ANfV+YgYs!HB70-}T-BN;3rbuYD`qS^a9XJ#>L#V4(A!)Jv^-u5ALA+=
zlY*L?fAPpld;<1Cm6H3vD(j?fR<2`DvHXqQww^}j5rW4}<HWjR*ePdQi(IV>nw*wE
zWWw(Z?T$hrs!GVhs+AdgcRCMAV%puewBM;OKfH)yeb({uY_fiQRY|{6tL$TmaRcii
z;=3)N>4J@{WU#@Mcavkiy04|olPE#%+2z_IwEiIoYY-J#`;v_8*HL=m1x5U}SpZ;)
z>yv)3II2S-<OU5(NqKuH{RgP2>Iq!|k*N}Dlm?F{fy?(fW>k!(r^y>_Lbs{o9^n`^
zSHGV$QS+(@1QLOWL^;+>?|a!jIlN}vSYcmvN$Ojiajzezd{CQL8dK#qUTIl9D3-<+
zo=|BlT+@kA&yIof0ybvM%ubyfRDkE-Iz|$p5Z28_MeBU)*MW+l?Ou$v#o0=(r>b`V
z4`cn~G&G%0{*M89Yli6_k<tvHP8K(|Jc9-1Klkng7nUm$OVS5>X#qTNKwiayGZ5fz
zVK0e1|M~P?Gp?8=jZY8-q66)c=I77GKoOFZFZ4yE)VH3AiW>g?^BlH?sSR$Q<+cOV
zboVFmuJ6|J(-$^+(&z8~Up~RO$){ic<#2}PN3C2#A)Vd8LWQ<f!Y^Dnh%Qk<A~hDq
zL^1UGG+tz3#@H%^bSmZOb~EW|N>~wv+>id0T$?zrtSklf&-D7xw&qf8opM0`{~j)c
zrVDGf(qd8w1_&T-j?w4?)1}(&D^@&-Q_z+VuX$i~0rukWeq;y0o9A=s-L{H6bzPBz
zvA?d)aF!E_;|!iFQk4304cv>dUHOxb>_s<w>PZcF9sC=UYq)%=!}Y2826^2PGOd_K
zdWmKpHa_77Al`$aM^1Fk^@D{)$ua|*JHLtlr{%E~w@$uLlhzQwRvBIcV)!t;H_f!y
zBFUe4E;OxHuzn&Uyy^0D7SS0ZFfT~5gQYB$*MJ!pc%~V9r3(|jOP0kvY2990PM3n~
zi~P)kUH=*;U6?mD5Ub8c$0lR~8gdh<al#$-4~+RJs3UVv8`I}kukC!hizJcp!oXby
z<fS<USn5vTqX~y6OCp}<_I*{ih%f6?6u%lpSN(2=m?&(9HKgig6B7XJ-9v)d(Eo!G
zLH3{ASZd0FaJ2c@r5XUp%K-8w%Xe`U03P@wfr00-kOdWR<N&8QauC+Mxv?f*mOjj}
z8x+_)P@MmoM}O;hU*vzPp?mPe_Pf7moI_=7H@_Q&jZC_}+(zTx{e$MezwX5XFY#L)
z8}F*_FZJu6vT%NULQ!b#slCzamNsc*Htw3{zHX7YAtx`JoRIVV>m+R{Sh`nFVxzLC
zP<J)hiuTMmxcU9V8Yt<|SMJjK@SNgGMS_@;wow|X4~%WIkw@oR@YG^jQ;Osuw_~V^
z8CDgRoyp2&QS_=STOq^E5DpX4z|ASk=18dUKTO>2U3{_0f+;si+RT!beRYRIEGOZi
z;ecrftf|Sq?57)O)%W@$W9xXrA?06sn9Vkw28oean)4y!x&@LXuYI1>swBLqxZ$;9
zhE6xXvjPF$VAW%hCJcqQOyD0!j3U0iHnxIWiP5H!;#D}7CIZFx&aQ_)v2(RqA@<nW
zNKXv;@f85Ixp1()!IA_Y4h?-BLdzz`LkC?efi4~b-D#|bg{|4N(-uv0*d0y0{!(mi
zzK1uUnxlqu=^^%~-rj@GXTXFeVgBWJIyJ`7BGZmfbrl<crtJi<1DE_~$+89??&CJ4
zgZF=5h_*a|VT1+NzJvR2pMR!1gSA&@Q5yVRRoSA`FhkjUrVatVIHvSjq5|M|+PTy#
z%3$gmRs`??{+R=gi^|Q+JT;Gce)YGbFlPHlvFOJ?a@trxlBjjGD*?z^#YjoxrZPa0
zxXHYbkz(Jd-_0ZHhIv#iU#vPoyjLABr`HbHH+L39D8t-J&40VruYZ!R>FQ7#1O~Hw
zrb_g)+iU#&gO5Q2XoQuG+1RjJxc*V?;RF2U)RdhY_$P&Bu-`mAI&HgI3Mkr~06CGG
zRsnu8BW5nHcOAm+AlSS!_?#kPphiy<A*{3*&3=$!MUdk;>O%FH!b{gxM8)%zaovq|
znzUn-`Oh3BDlJiiGp6Y^4;M&{@Cl#~3ry#u=jwVu*5LR0&^@UX^GGjysU(`Qc5CL(
zIh5F&<b|Fnf0%1`XjQFp!S*eh<`h2vo6zgpZMYAV>#(kKapsUx8rjJ3#(FpAyBCW=
zQvJ~x6tlb=G@|{Po@2VX&!AQZ^+)$?cJ~~}_uZySYO+}2df=QEE=JumuOyMfM*4Ik
z#m>i4eO{+#eRfuWQcT7q<&Qi4Y<hT+sNpzEud;vOB0X=tbZB2_=ob^$(b%!CnI>bp
zFs_p?MOo3E?qVvPnm+Z4Gxqg=uSkKi1Qkut&Y#M%3cBDS?V;dJZi@OUuXW-?vX&^(
zayauQv|-clX8uvLr(eGjcH_gZ+4cBXmVxT!`hUFiBaA$D!<ImOF){|4-tXO$O3MW<
z3(64<<P5$A`jg4E`sGjdVLfO_9delTGd<vnI2|z(+wi0iAR-{X<^9CV1R`qdAJp!y
z_|Y4t+#pf>tn68}RN6jzz>``#TS%?B>PPLR*w@25qqlyaI`_6z7}U%a2B3M*^1|7R
z8#aB?1=W}l1t{E0Gg`d#LIG<@`x@M@dmQ|Zi3un44$R%olk#DLA<9r@maztYxrUIu
zqe&ccHVz3Mfd0nhSR^S5DrE~BG|<2$Bv4G|hT2$bP)P8EJAKcnlN#yBR0WTJ`v|GH
zlwHHJ#>Fge1TT0_C0=PBMC4u?6Q}y{DuqkPNQ<AywLL5ABXu>(rhp^jNI_*XqGZA!
z%fG(3aD+fPc>XzGe^sYSa1inS0H%;SD3n$G-=iNWj!%LXa9KT_&hN@J?@f+_SS5BM
z%U%un%WA8Hc0(gHzL#2?*f#7!lm&c$A686W7+B((>aiPTB|fH?O?b3av~Bp*QU&v>
zT(!ZUIl6>dwb{<4W~&P8UPs76aOs28(lJ{;RgE1&tGe3~SypWO$;8A!P+bZG?E-GB
zro;c72L-vT#aTbwl7OCCES(`@a|*=L<>=H%0aNO8g%?kJU&0dpY>x?U1Y&Rg&hEed
z40}m`)t~)LQ}6i6Vl6>(w?2ca*!C^JdO<r<)fV36Z+aqQHgJP%#+}_Syx;npM_Uj{
zpLDWb^2+iiPzU=`g2wo1B%r)5qd7vfg?o7bkm=uHgv~9Y#q;m#aUgzL70S*dW^5_t
zS2IF+prIoeFE--0eDhUw7-?TCUsX4WzSfrZ*!zD`Lw)s4K)1H$^FKx~sW=B;cBcic
zQ!(Mf4M9=y>wPur0g>5_WbHvJ1uSrZdW|v^8{93S+sU8P)xvo;o&{+<CY04q$#0E8
zXgmA`2z>z9>;qX+B(%?*GJkMw*%JJC=JJqapstSH4&Y`!z$Z|H<gaepz}(;a*yDn`
zc7@|O-w##cMz~C0PVad$(&^zYadLgyXaG(fs$xZDDMcp_0081iEbF=DaSn)DvF&Rr
zyd1CG=^rE`WJRP{so_G~?Q0Q0#^>piKcsMU-cVLuV331TNDoZ-a{@=l2?lTu1Qiul
z0IE^iw^4^b>cqM1C~QIs8E6113a%Kp02;0r#jp2e5Das*$CxxC0I|l*eqA-4D?O>O
z3ch+iS6q0wI<Ys<lLk7bw3~wDw>&|i^-r}M@Le_IgaH;;vf~GTUwy^o>Xb(qKto<j
z>DE0VQ|^7`l$%HynZjSCkO{q9h#Yu=I-A}R`@3tG09Y9{>Y)~Y{V(@TQdo_}lgG|g
z(_3bHbm(jO7HM4p_E%rtHZB23eZ;p6PyT()LV2#N#4R;Rw@m=JHu3Enzlg1?LaR^L
zJ;rY~BTjL{a&k6k<dT?qpmRs!e?9|UVO*Y#K^}P?>p~0pun#8i%Yaong@xC{h;FI-
zVRTWW-$uT`Vpr?w@A=D?H0-Baqqm=}*q=#6DuHDaDdu0w2fR&Yn;-%(fOb)$R31PX
zYi6K=58o_!Ou3dM5s1!ire)yiVH{Rr)oixhwozn*q&7uw1^9B<g`^A=hc<mxNHF^l
zs!0!vt*}WXS7jP=>Ef&+Z>?OvUGiAMvzl^CR?rIO!Br%-k1Zjgu!%{yLf0?)dC@vf
zjLTVS>uCMhfXiYg>d@O)ZlymJdka2F7kdv(Thdh9)6emjs|jEVUNPG1ilKc|P;8H%
z5+#1X?f9th3&WofVE9NS){U^|zarT<eQu$j%AH`DnGAyp34gnCtj7^qiP&K0gKdKt
zw2ltZ_t2ask)s(Kg6KEHcKr<&NjheKM_FR3_GS%`d4jX>R0pc|hElt!?w3mf)BB3V
zE_$J0Cz_2Sz@cU^2b#85VQL>I`u~_Z>!>KWs9!4zN-9XF(w##ODAFyR10oFFjf5i7
zp>%`P00Rs?q@*ws(nxogbQ|})cdh%~?{Aj@4(IH%_w#$U{bcDN@@?$kv08Rb3iM2j
z!p3DOmC^OFAuSOWQ477D(Q#3@ju|vZ@S!7p42jwE)cGaA5{FdM^4Qt_v@=aCSq8m_
ztzViY!EH8Ukg>bGWf?gRZk8=Q2hl9=*LiGMhUvUZkzti|>VAKXbNR`dlD;?x<Vfz<
zOdVf6$bHw58i%lIay0$gv67yBXPp1)s)FaHt>^jQU0gT)@keSovks@0nvSO+TVE*u
zquAw<UjWzVAxA<{9PPr{u%UO?hk?ug`!M*owZ(AM-<JjG3g=;)mV?LB<Y9RgPq+h2
zD1&l%-Ye^P`E`|Bk*oUOzIjIKHJ;=aYWdKBi|2lRMyYdZo`Wx^w{5-OcRM4z#_+WW
z#c{b<te(iXBj<jOY~XIz#MHa;gcxO=+H$swa&b!J^vDc@o@S(88*#Q=Ry$cVB-%p|
zTZE@f;YESOnN(&rYFiG6$0nPnRSCY9TUPBpoLa5<nueg;)ND^|Ga#L0iww<saT$`M
zxvVVuZ0No<vUEH!Nq^6fW#IGoEH-ZmT4&$AF7829(9?ml=daqabI_GR%?yGoA2`t{
z$~ZU75&6R&=}b|ECA?mq!!h9@=Gj9&k&FX4I3u`b!YtxS?ZCr?EY(U?RFI1--8`y&
zRFJ9czj#qa1^JZdJrR>Q#~Zt2^aqLV26b>iAmVJ@cxCC;7#(3-Uk<f}T?NT(Y8|NW
zqnb4ALBtavLtVPP4R#!~@>39B$Ce%)e<x4%wNC!k+<=3$siBlQ+K}fLqh$f-qcNhq
zFQq*zFv&shoos2<nSzct@S&<}n@>CH)ya+e3f#Vd_f*T?n{6C(fBx0l%igrGq1a$e
zzQ^5tecT(*ucn0;#Y~PKwY$gsoG8@$;r;q_%<g-l<XymsUrp>$km_{{a@f9aeH(rI
zplYSBd)GHbdrEjfSETRI>gx?f5XsWJPxr14ezu=3{dHh<0|D!OPndU0oFEe}q6lHp
zI=G0goalFRMvX`6Yb?k!R=aXt<RAt<z7cYrK9~i)4zz{#q{~Li90EP0T8RuZzGu=D
zYq$O?UGJrk;Z8hEl}McjlN2JWU4A*LoJCyZ5*wwi(;UoW(%ryAApe>_>DQnjgc<@h
zVXNXMYd~0-NHkl&XPs5jVTm)0I?JsfT2axlDa&n93zZRzm7~c{7M7Sbx!#W|-Y&`i
zY0%q-wkdbhRA&l5@aSo8F}41PR4w5!(weGwRhwxa%eT&1@3%GlOx7j7-^>bILLj-u
zw66-9QH=b0-mQ6fYwB?19;eW2ZC!35Wf^-(i5#9`TJ`ZSqu=`zYV49MTM}5m3+mAE
zxe-o5pp+Q3L@eqfM~e;nus9pMv?U*_+=aa+%H|=)A5Cck^t3kEZd&6L!hl@fP2qZ1
zT*@NpL%Vqv;P!iwjy1`vC-w*Vb6<~;j_Rv$b7|IrqPwm|hwkr4Jf_c$tV6eXrDzCa
z(Kesg$-4PvU8q6jam~1P(_zC@b-n+U`1!?^$he_l7(?qUXbPTm6;O>c+oUIf_wslP
z>?ZgJbm)5gc@REk;<_hwc;|r#92`akR3JYJc#TK2p4VqIOx*Y-!p`Qr;1}nd!bb(y
zn*I>|>tV~i{oic^#~gmGmwrnrqaTasTc2hZvnQH%_;C`n1H@abKfRd6?-`F9VIMBe
z3^SGi#C?uu=Cdf)19K;D_u3*TW%xWp-!y9(r!l&3GI)5+#3^Ump?XDfyIH|ZH)~J_
zG~FohWs)54cfF^+FOm^u8g;7u^{4y6cdlLIrq9Ct&0V$s#nT@Ia!kykntqPUr;`SZ
zixnejBBQjA#n;%kR+v{kS^F52pBbm39ct#>wsFXf3#xUk$z)Rne2w)7C+ylw^C+h0
z=Ls4`j&Nxc;rfE6R;Jw8@1>$W60{XY5YJw02fYS>y6*i1S#4d=D~3SOvm#+Ap*X=+
zQwSI#EAxPUOI-;iuaJhdfMQ}=tq~_mv`DK?9fq|<X06yv#2;tH<Sfbh>A%wCHm#X;
zHrudwvFtcVdxAt3PCdAm1pwxQQ<j2PFNzR29vho`I9PjVHF33lFDWwn#iOp{tW%>8
z@Q3aN4Ab+oc-yx`pi9d~J1m|fM?fJC8X;?<qJHj`KZ?!Ut<sbKD;dU~{u<P559+L3
zpgaB>5?7cvjk8u15av(Uhll3L=9|wv>3Lx3sIwNCEfTo2S#{{PCAexM{WiHn@H*7|
zE!orvB@e!{BX+>KZGX_zhU;T~Tp<ls0XR=~Y;#!t-FsAb8yu$(-^%%^n~^5P_A-+B
z?ziyV^S$Q#%bas*b{8BA&jVLJN~S)0gdw&Wk6|C?aCm4Ma*$#h@W4s2CtI1q8AZgC
z=RUQo^k7!Rz_0zKP@^GUraSHCH<j7fNPiYrwblk@W4XIbM{rql7L$iHeUSTWW8MU6
zvX5W+Y3cmT?<R}Qq1<Zbgm>$%EsXnhxt5YBCA{<u-sG%={eoqL<!jqg=`uzzJ;zip
zcIjNF$cTarHEN$#FLsO`xqphtLv&6QMpVRWU9b;P@n~DSDQ8iTl5J-^ZK0KY--8nT
zU7_~=Yll_Y!RRTAv3u*FSPrWx_l6YDcmh(}87{2v#m!NZl756=pmt%5nW#=yXewF1
z56^tP-P^opXOA`KO+YAXPRCCOauT0ofAbFT+qGza_KQmsbCi8$z%#JSIOuzA)%l9f
zK6l2YcB)#O`o9U5(Iz&l1YzvhkVDihdp3qEFtr<;?Yn>>08fR~fh<4n+e*uR>AktG
zzzIn9sNj}+`Wq^0g9mq02a*8D7H4@fsVH>*mhPgNDA&(pFuc;M$<WFLn;Og|J3tA@
zqGd*6lf2%)N6paq6LY660J3o@)uO0B+3GI1C8(x^&vEqpt>t@%z1AS=e3~PsPfqT{
z)}FM92WU!~c5moL!IUWIa{Ka|x%*I#x~qW-!(CdHUp*uQ)m_;p-6cdQ{njJo-|8dj
z+y4b4z8Q}o=g}^1GfJqeCo<x&qjED7BoeZAlhMPEhdo>-3PJyHR$TrLjuv#;%Lbv*
zppQm<J=`0P@q%%J*73gyFk_(3hPExN-spEcb>%lmOD$iwrN-y9;)s0AMN#3fQy6NC
zXHY4iXPK&C_t;C=+gTFJ;h@a6=^*aZ3)))uIQ18E!q96~Dq4M+De!3o+)rW2&!(Pc
z#OV2b?6ov%Ei)b6H=K5JtZ*Hn==(Z}sriKSc9M=<K}&)>;<;>Us{wy*2>(>LHG5Kj
z`#fq8$*j&q)xH&-lyAR%+jXAVEAGcpcYY8T^F~I(w;SGjZ7Mh9!<z~=Ky(Ps95e~y
zIYoZ)`*I?ETC+I{+4RkLZA%S2N<UI$iwHPE-5*}+UK}fVu}G;FmoWd`2kl6jEg(}N
ziTBgfngVZ+ZZk|zHQcjq>C!KtMSV#KU<Z9$hnn99!WQCWF^R+WOjTmJ#jTzaHg$v8
zKjICtc(lPdKvTN5dk3;nd#MPr6Vk_5tZrJh<K9;cK?n3(rym*G>*Zy@D+R!gj}!$=
zoOV_LN4fPY4m2Z!D;%I`o_rrL!^W^Mv{ZKrJ%IhaHUHYpo^p8Lj9D3l<k>BQcF~jw
zG-%_3MXyL8POzkK+r(1peQ@RJw0>rRn_7hfMDGXx4bc^!(%q|7^Q*7-|AJPMOT*vV
zcUQvTfhsj+t!N<usQ_au4`?Kn{Hn?&cgWYA0WjDW#km`-!f1?OvCw~9G&x6<fR){#
zraUrz0K%X{uFtSR1q~?+b*FET>ML++5{WC-(XHqFF1Ixg$)^}Rc?;veij$cW*<qh{
zGyYj9&#^YB=N`uDFFPmvjX588>&!95jS4aN<SsNM?Cu_M6~7jXoi%o|8Qs~Pk&X7I
zuo&1JYQ1nTEaoj?Pag3X!o7gk2Az4M5Rl8IRi1K;C*?NK!P~&9_S*5hL8G)k72azg
zo_4IiPoX!;Z@Ov9{jrx{Z)Ddv-JDjnxx|PUHk+^Aw24+D`tq!gX16)Pt?B2}RzA0o
z2HKU{-7r=KTAi2oKAy5?zp!2ZXTV<F9E^<5_f<`&r;lO$MP!-Z`YK$GVpuRuuBVw2
zTV_6~p7OXRA4}b`;ozvVOe+2L7Y`rOlij|o@U@3ig1)DMg@<zi%POgtbe>l&xsr@S
zDvqiIVmTGH#Ep&(2VR3a@HD|Jo4UXdF_*?!pS!UNA&=<2uO_Dy?xTB|Q-)di9O&ol
z;@Kb8!)*bwvx~H7*gFjo6^)3F65|^2%+8Jrx7is?`~dE@;)~gJ`R;o4TpE+R?Lq!>
z#t}`GC7Hr)+e}knszL7#jXk1^{*!|-|Chei&Ug7oa>^h3>Czwh0Kj1R7f`CE#xnA9
zA++)&EU=qdMQ$Bv?uw$Z<h4`Pd3hX!tt)h)Ssl6?tSMOw$kJqQc|pfYq5zS5sc0_B
zcmy%eZ1%usaH~+9CLzsa5?R_Cv@%mD#h37Q(CWmz5BfKzyq!P|Kx92NWwBV(kF=&9
z@Q+Ti^m;7j2}?9Gc7?=&Ed9D(h-JC$uSx2)RTEY}Sp24Thw~LVx_=Gkwal$Pc8^<_
zJgO$D<nfmKLd2I=c;*>a=8qIo=geT$0pTTbX_U0;#yVkb0(wlIm!$K+@xF_=8~EY&
zrS^D730iq>E+uuE9TrP#Oj9?|mL<#LJ$Ag--#PqT;?$UV2K>EJhUqhCCc=*N!2HTW
zN>gM=ISJ?@m`@E^o|tkAww>7a?{LSvW$wOk#1?6hGH7b$L7f$iPaSiUD|gx)#{D-+
z>}c?FYzZyOf3W42*Z7fz2lKk@Kv2`&OxkGVoaet;0!>>_d%!vPXrwKjPKeCG3|Jm_
zyt}OY1)vJ)w_g>rrhL(j8}xBZIKj*GC&-n_aVnp1;<toTKC_^>h;Vnq9Gsg<dk1zD
zn@Mg(KVp+Bb?n7)MGs!B(9qn^=`r}UStwuK8_gW*8AOx(hbFCR{Jc;$+nTgK4lx;%
zfB60%mY78AP3qG4owm7~b5TByejqFx`fBlAjm$bMZ-pw1laYG$C4t)7E202IY4x}}
zVDDE!A(F`x&R<3uT1E*OJ?aU4bvNI1L=70ex?|tx*a+b1m_0Xfbs_a=Qc~{6yGMwb
z&?P_wm(71|z6(hxDd19cF#`*Y9jT#nU$%^PUu><+QL7gaZ*Y=mum~%2ej;_|Lci8c
zib=sz;3dtP9U<f{c-FsrKJk(<8v>^+l_B%3=Zv83?(p$!&WW|5){VlmK(_1Xk}Ng8
zob!}Q<dN3~nZ5MpUd(A)ZF<ancR6F9YKB5Lq*FjmB1gvHfG4Cw|KyP$?_lkRDGkv0
z53{Gf*Ty5@+H75=q-w3Kk))qgZ+o9dr@$GqtKc<7lTND_d&dE53UPos+dF~Q5N5ZM
z*IE>zqVw#Ks+@`MWd`z)_x-GCBr<uT;R&ZFI|&cg40BwXWfmfa#sodz2FJSWs2nbg
zK&Vd}U&WckhU8$L%<*j64XI*0a@xn%ta7U6+<nDVcO-EB-TWXXbPXKG2oNHX;1bF^
z%>3rpZE141(wuu{&u`y{*4g}{?NjPL#0sDhKCud*^FY1=slZCAc~Ix~C!kZ1YuaAi
zDwI<40S^!$MIr2<_2c;pmg!LiZyUIXQ-NiN;or7;&`vUb22QV5m?il2g-szBqGN9S
zkPHe<ikB0v>QTtecYnv9|NfuKz)WQLS#JAt;=8Wnj~iLT15+{R*?HRghAA>x(a~J}
zaxd`Qb@bG`8F>tQ@rLsm*J)KL9>V7=M-qsUc>`(Qq=>!zHBZj44`eT{?3Zi`;J98e
zr_Xwm>>XuU-=TAcH1hiY(v~*w8nG`QvkwYD)X5;^7G^=P&Z%=XZgr-z`?XbL(jrPp
z>8%Q>D4QxJF5b<Z=FfxpqgNBj+gjYRA4a6+zU&!$>ct@_?d4>zD)@OLsXuA1(PADy
zFgK3is@RJ3wj!s1Qucre4<g=^TK1eu9MS3VfG+5FpVNi^qUdjWW!Xr-Q${Q54c|B`
zakNx)9M^Uh%BG^y&S(-$b-?_FkYu$@$=ysU8;IcwZS7~#Y4Xsy*X<zBKnqgD=jyqc
znLJpEA|K|q@TGlRx)?cp-&S^8zxF^YJro*k82K)|v?J!r$-3k4$@(g;^U(rr)r`CE
zj@6KZ4SoP-{n5ME<~@g#1QfXca{1^^JGrDTZs`H9d(p$VGRmyKBP@}Dqqr6!+5H)b
zY?62g_%q&ac(6_M<ABP;>InCs^H}__ftXfM=Qqz&$wQkKy3lH-I%_`$SZbBv^(h}P
z<@TT_`o*bx+#jZ4jf+F5HS-wG#WUVg8Ru{`nftHrzky=y-*$7!mEnJUd=|T7>FsSh
zr)Os23vpa;em5F!ZO(>?+1oK=K>vg`mT8EhEXLqm2uFJOmyhq8VxII;EsZRLv@ivI
z{R9(~;CHEkWdOH1E!Qr%EA>Y`C`j|~{;ZXzlSEd}`30ixwUeY!86R0_g}vwQ0uPl#
zCVQ#tI5-5^%{}Sr=%COE$?I#Y8pPX}bt28xe8azIOrj8l%q!u9cm`lz$f}WsVp#wy
zI3AZg3wzny*igHP;QB)oM;x@8l%t1~iZ}Nt_wbY8t>J`(nv`Q}5G8OH_Y9=z{_MQc
z^HR;c8{Wm8l3!H<cM&4PE8a79tfNIef;v0<xT<F>j;FPx5*M%+vC|l%I~7!?J(Zsk
z?bVTR$A)>T<A;veQTv5)3yyv$kXcshim+7rZ8ks*^UFsCCg7S>zk0Bc!X)^gLTMj)
z4>t2+IMKR$hI?%k@+Plt^|WwVaIgd!mSZ#O9y({p@i<UFV$rpI>ts+S+`L5!K|(<?
zOzYVc*i^V?D28mZ;E*~EjY8cnTZE7M4KZ8?2LhXoEZSJak0k(uv(qL*NYEW~oBAxS
z$TA%#*zU-dc-(`htGriWxp=8Tb9CbOZW#r*(bks!Z9fB<xF{P3;5`{&^RSq;1`4*7
zQSSULamNN2xlw0w4knXAb<t-U^dYtm1AaZr$rIRaHXYOs+WyJ<?)cHy;=8{a&~0<k
zzPACLuIZzCYm?eWyD}To#vgD?IAwg*bk|L^Cif^IR2ES|y<_qe@f^r2L5@NtwW1f*
z>R@EmJsFIV!C=sY<SJL0sVDxz>OtVUM%Ra2zSSV%eSiM_OwH@bZCfbHj&_(t!v+^V
zsr1wNwH7I%O>~l+I9k7%ZjQ9DBhEdx4Md4-X+f-4-Y<1j6Vt%VJKV5yH5NRtYe&h|
z-QQBtYD!kh_=2}AjGaptQBw64;WotGSL!fIcM~U*kQ&nsu-#&xxq$T>aAIGF<(lgX
zD<(!z(En&MMTmtBP0DoC%Y1n{^i$MRwLmTZJ)*y=M#)_R!Qo=YOUdhTlcPRd%Wju+
zwYc(S>-0(^x{T&73O)iM1do(TsSNi+LBtxB>B2+GP^xbJ)Q3C+9n}Q&U_;#P%{8dR
zk~&iMVd^2ECkV6_ptA_-dYL7P@27&G)KCoAjD6OAneGIEB?johEE@;KGJ9<8xzSW&
z^U}E<?-E{-lC?<Xelk52ip7UOi2QkNCBPXxJWqeJ+4DhJ*VS?y@NE{h62tpG(+?fN
z>BAi<s;pr&DTm#NQ;2M`p{|-k<I-k|4fk=y##tV1^8byydUvPADcbyB`|li=^qJom
z=aThzemk_EN&H)AZ@yLk_k3<-dgtL~4aM|y|MX99_0by+bu&!h-uoL{61R7h9@T1*
zdY%@ThHqr^=f}xmF10L{gXQJ_jb8s~edFffhfE*Qp(S`ToiY(Alzl6LmN}$fLJ9=y
zW$?JEcRKnWYF;Gk4O*k}@yE6C()C&q7VqJdANt(YOyT{m+=LZIu?l6vlJ!U$NYUOn
zf!%#=)r%XMrGi5fIlI1V!43ELS~dJo4<vh3SRILUS<zk{ZIkl{#MsYL^%lBtE}Hx@
zqSL5ayDmdzm>R5j4>UVX(I&j;(a)rW<AuweN@tniY2{~}P||8~SHZ;5$CQ4~v34dQ
z*B=rGJ_`;x4w!m?>lo<~7rRq7s2w=5(l)JO4RoMcly?Gh64r%NSf*ThVi%kx9oY%J
z^n`&9mp<lxRI8FoKT><-ow7dFeGtq`>HXlMo$6ZZ`!0uNVJ+IPpOHt8sEH#t9(CM*
z{4z>ucoj5?1L9YWByxQNG9Q^80I6}Y#|rOFN~9AFZR|@RqGNbhK1{(d7*QMLr+sv0
z7F}m6wvAuEs;a2tp8P?csppuTdm{oUuVx(uR8`^y$P%FSbH~&6Ex*q^{JF<IP<+~=
zJ(K!-9bHV>99HB2WW4BRA|o?kM$#(q0CcNZ|MmU?F}14~6A+E2N&Nm^;q<qF;p??Y
z(9*lwbF;hOeclldZ{;%`CwEUZsP`i`JyTpEv6HddxK$JU{18IiXyWV$n*rsK32Z{0
zseJ5*r%`n!Ki%F)nZIl^$oQ(q-sJS~&@MCsmo}tBq&@GcUqkXH8QW=^=&<b2#MO9h
zjBpKN&p68NKKMx=2=)A{2K&~6a;lm;pLR4d<t^+!-{fcP$GclU`@t4;(fp;v@16%{
z2EDGt*SnSUeHG$(p=EYro1kQBoRvsk@6IAx=lEn`A~{{o+=oY%jLJvwiSwF}KOLyB
zivOsqVJZLD3OQLV5Z{VU0@|<rpl}mOCy{{LOz)?e2%|Me4ZyB5vkj?&Bj`~tu7s<D
zWCxm{*Ft7V)X%$pz#|++^!D$~A5GwD(A)%(Pv#-6W|VSBFLS>DVW1PG0EBQV;kd6)
zSHZTaU+JsBQD0ex1$*dJeK6+J{A@y(bG)EwzwB=8srV%Z_xbs}NOYrVEqig^dqirw
z)tyVzXoDg0tRF?wOZH6xXT3OXMo(B-9t0Xld<pIW<9#!Uoa5KD9!}kCeOw6T&QFWG
zLdqYR&=Ea)8Czs_*lpNacXRG)VYlIG9(bkogo=Uv?G<r99JWS%erGQ_r~lsjcgE(r
z;G_BWI@|b8Tvtr{Z%y-fHTz_%y1|dC7ZoWe!_W6Z8+ZLg{&ZN?!pFUxBBSmL-~13u
zkCfr+59OGxk-xV;(<MHs=l%^eIlKtS-8v!tMkN7evZ<vJ6{=gd^jq9xDvd}^I=PO%
zglEn)W6CJ|oo`glF(!mhIk{HV3PrpK!+3AF#GCSa2j0lZP>p^oDzn^^u+S2~Eo!Ww
z9&OW0R`YO_wQ1iv`lb>7r;Tl`n}5xglO3g%R;pEJ+buK5O2Z?tp5E-b{%wPoYAbNb
z%-q3b%Fv4CN&OO4vj#?B=AaoU0^hcUBJ2hez<&}sh}wUCu=bHZ+3mNAsJN@zjT1fw
z5*d{t#*<wAvS4l2vEJ##IOC25qvaG@<=hAXNa?J}IGeEv;R1FWdVIcD)bw4A=T0GL
z8@HmZ>h=}c*tE9IYQdK&-{eCV-#e-Ps;r=?gXWp~#%X9p#VE<LKNcja1peIBU#lNF
zM3RCvr{kZWO_Og$11WQ}x81+9T1kj4Q2#Uu5%eV&e%Eqy@|&z$nPY6hLf^|wv8i6F
zzh3=AWu!N5pr(r%yBoW7zRvn@n~RMAEAx`!%y7MHxK;WgbHU=of5+@3VJ;1gw7F!q
zN73PNsK4c6Bf&<KUAifQ@0cEnntsVXn5dv?xi9-YNIn0)N|gLBT`FqwpNQZQ_OjZQ
zEL&}o1v|6b+{E@<);9PX%7kcKLKM$h7s`gXq}sWZ;Le(o8lyYY?o|q@Kve!mRCGc8
zo{3AW3+!}HjcUnRufEWAa5YO=m;Gtk$1lS=!Q~f$(R(4}Q#frkbD0prd8)8hPpBp7
z<_o4)4#oRj&DQ+t+KbRffRIcA>CV)bU-pcy(7$gjhHV<S<hw-2_r7`v$9ih7wK5N}
zEJKG>Ho-y9ZIQB`A32&wVR$AL9NOP!a3&ye^1elqAe{x<$sno`Xgco4)&G?;GOAn{
z282IoHjEd(1~#AIZu$mqW3kPKUviu9ADOADZlOFLyT<hVpAauY+9i&f>Q_8kA3IId
zK7a-7N2(LucHT%7`t>+$;hRY&o_#i!dc_&qIeo<Pz)+!<8bqwsxZ;Y5)$av6xmOR$
zNfyng-3C(3Jq`q>Z9JVXPY~E>i=!rgCff1k>oTc66(0FYaa?>rf}bEacf(#JxtM<h
z?K>^A(Ni92_X?0$jY$y)Nj8uYjj@T2qWEOf>MK~VXT0z2NUx#o?A&n2*S)2ua)}D?
zkY7r1pU5<eg{}XkRw|!XP_=2G`@AmhRS=&_XrU!|t8udQSah=@c!2L;?z>gVo6~k3
z{1@-}<{3QRk$Q-0>Slbt6FN+R^`2;48=hvc4*%mMSrzP5HAx;b2vBtUrL`OD=mk`_
z?uH3NtR;X?we&4)l_cV9)+EDrp?z&SOd3mdr05JV5+3$oKCHKsYY-!V^WEDqR-sPx
ztb)!342YM<un;P_tx_fle{hTxT2+hNbXUdea!7_?QOh$Gyb<?JH^d<=qWwwX#_R8=
zRiW4m8`f4+3obgmgwCRrv-IZa!T_EC65YI;xZ`zSII-DkXWSb>Z1<RFv{HRFE77l6
zgjj-;!?MTUiIbgL`87yipc$$5y`{Y8MBKDaXLmjr3|rI27L}!0<*zu-0NhCj@)(Zo
zW~MVCoWlDs9ngRS0=Rmk<#rj&a(?6%oXQMO=@wE7gp3PK6%sr_bkJ$(O?Ry;_r3AQ
zsRsbP!Lg2ie`!<w)I7gjVO$OeUe@@$f*6qxx8)Sn)Gi!Z2}p65P~Ev^eRn1s&Ycj&
zCza(YsJ?wmZZtY_P6+DgjVWi%t4_=f_gIwZFa@w40ZFd$E|cgDkPj519H`aVs<7|0
zjk{5XVZ&j^Cw*8t<I~P_o&n*6OUt2a*r4HDR}K>Dn@u6SiR@dzSJNosI&*M;9j-i3
zFGjEU&94tyDwf^@t)F_v%Q=f4GWbO9X&RjJAg*UJSW|oA_n;yQB=HX0$NvA&&n0tf
zhnrX@jFmD`E^@_yHIxVYmS-=+PaFAu-;map^K&y?wJ_6?YVYS)(kL7!OjrGxOR0ZI
z11Qac4w;A2IY^a_zxQsY)!3au0RY(7kEaT_#b(Dc(x}O(ngohC8{ngSjKFpqeO;<H
zTZ2sd#-gk>Ho<QzzVFi$@k}HrvAYeU$8z@9d*OTC=O?goyjHpDwihBEFT~&J07WNE
z6xi7g(Cg6|&C4G8nHFXsDH0(9mal@Bb)qC_ss)ZfM&EEo|HEr3a3G{Ojj8o}IUwNo
zwjC%Vem2lOH}LB2cKbF(|D#cofFp_;=8(q1l4_xV5e83FLNIP80V%QP*ZaJ~2R>6%
z1xR~m^RR%8m5**oYw0yR!bln|jt5pS(MI(9+DsgcVu91b!abu2uWl|IkA6HN5h&^r
zmb2r{swo$BfMOQrR!Y+}Oui4Sau9MMXsTZz`I?X7&@d<<kyjy5@0c~x^N@&6|0sUS
zTud?D`maiHpuW-c;p+0HcXTT$+ch8)&E7&-IP9hW8s=6)vqp~{ogMMwR+4L3+%g1I
z9YBa?=HV_pT?R9r#zu(#4}>nY7w&g~I1`wH%Kn0G#r%#MA$Xg;Z$D*wenIEe{eSbq
zlkL8QOCnbt7M~VSirRW=WMZ^(E~&}F;IZ2WS@|orz9MEpLRQ#cZ!Ah;bPX<inWfb5
ze?Yha7(_=OtNT#XLTCmnm~A5L?yDRap9N$dq;xxDa6E@nRy|=<g7)9;Z)E_YB99CL
z0lma1?PPUYc#XkMhE;h3G%?PBvo=1BKmyhC^l^Al7JiL`*OA&u6f*_ti={1P*fNZJ
z@=qB}5lnxnmYGW#Y0OM+C$BeU+l)98WrYn}wZLoJiCw3Gne=wLxi_+(`rRMTzGH}W
zJgF|QIt+p2Y_<BR{;$h>3Hbm)_03LRK_Km;5D0eRlNvdlm`=1!HL;=NjYu<!>%Dg;
zd%t*Q(O$@(4j-}kDmzGs;+hhRJeh8+QZkn6%lk;2;CEBnN|(TZN@D%mDEit-s4CAn
ztrvShjl@(U=ZE!cj#mvo!0ieyv$%@N0@2|utj{@YyVv>igVfaWEW=@H0&HIDYfarm
zk1)7W!|Pze!k5D<mfq{wV3qZE6Ru=B0Y&e4K$NepEbndFb%ICtYU?UM#javoEa~UY
zK%D_(w7))C!kbG&FGjq9w<$HQ_&(~#7biCn)ySv>l*6fi&f=<Xf`F|FxGCgWHl>$-
z&e-EW{xd39vs~Nyrt<K_HUp$PfMIYhxKGtam9~4iy5hZ?;Sod7Y8dr!X28|W|A&C2
zn1i#T=)SJa?t9Z+NG#5N6^dqgbm^PGkEi3eJjot`d}#L|7~IGx!{xLOBv<5*%3lU6
zJ_$8WV`Bm3Fvog(rJ-gHd_-scERL<fa$rO`Itn<~AiUtj1F+51Sn@_2YrpC*Q|{}o
z(T}HgD&r3eP@uM|<3<{PTgqYKcow+S>GLtCo$BGAci8GdxBUY*6<w+uA*iUz$%p>f
zZoHGeKIx<}-C4@q#t}hJokd|g{sAT@iDfu|$Gm>IXn^Q_0^*Y>mp9vE&tO!qu}se1
zq4GSSs=gbZG6@DyG+hA9<|wm5i~>41fDtl#ZeF9N;tBiwU_f^|9Q?+ElplhX>;qw2
zM3MpEQjy5{qLhJH8Iu1_0Z7es78O}rL3KRqc&O*N_80K)W%?555>E!unIQ845#^Jp
z2Vi6)a<yUtbo#UJv(eolZr`|y@kOvWnn6gcn}4hR6#G*zab*tg%wf~MH>@Tfv1bJO
z7$-0Bq?QCY%J9_3zn1~mzfhz{uZ6<%D!(@^KfEz;YMVQ>7l9&kL|e7G@DN^KNbL2u
z3iUQ`D5jPmXYruwJDx~jIE6+r00zSJg*2Hjeh56{>(sMc?)ZrDmL!GCo8i9t{Sz_l
zGN1c(?@yyK8t2{ib>QE!cZVUM#5=6{yIJh)VPJO@g+v?}f(#lIb>cW`b=II(G_YA>
z+fai{m!krM&-t%0VkQr-8q4n=nF=QmC8fB~?Y7OhtHZK7rBYM4<V}r!CcUZY8ToQx
zxPNBq?2H4gT|#Ua5_AK?f${e+#HN*-GK}N2FzKZ>VDw?>wfD93^vd$(@M7{4h^tC*
zuV!9VZJ{_bW>H0j%wtAiOBZ_|l>&CO+$+(phxg|Mzxg5vIN=pff66OsL*D3)cp__9
zc{%m=?V}hRkaACV77w_XrWkqr_~ZQ60y=jfEZkPy&U<f2TWeNMzj<eth1{7Mn=(4p
zgC&v+7ZmY7@7A6F0ePLF9iv~*e}y&oVhR#AkTg)VJ$?_f462=3>lgaE^Yp!7t=F66
zai<LT7cSJX<)pj8mQ;)ecJ{)Ys8Qi`jWh*WST#<ge+m$8*)>V{)1?+)0JZ3NuK-~W
zOr0`}YM}13jXWWmMTrmg<mlXc%A%EfB-p$~s1QYyGm~0=87)7=@aO1RWfr&Zitmuw
z@G}inv*(X?9*h|R_}o^Ke0aa%y1$#sj^k}dMM|`cEaJ11o!0S0PAeJ|uuV4}nvM-k
z)l^5W)U^xf^p7lvJS|R&YcP?~k#w>ywY8{px}lAI0#b%xwDgDZ+sKzaL(j?6o5|YH
z{tzE31#u&@lcj%=p920zDk4hs1w4tKh^uE0X%A80Ci1iKz&*B0^lvPdEXy{6V5W0(
z1~AnN=X{NvGUpsIh92jok=Gm$*{Q@)<>m7HZ*A48hDpMp%>}xn$P^7D*wn|Ssyo<-
zS=S~YbdHj70JN^M+4wn6&Z2`GRCyEvZe140smjL*sbS+!5Brh&uvU91G%PZlN7f{y
zCUDMhFGRnfh>T`bz$?FQ+4EVm^4y4>KjVmV3v1vKiFSbY(Q8@(Tp{b~7+$G4s}`nX
zsEN}9^I+?xAa!K3MdP6Ffnz6dJXEfo%v+?!F4M#^lnq-_Z-I^|IJhFIFcbZa7TeT9
z8>V}=AbG8fRZS35>}@hXYd_$_MAq6n6=$6_w6Wf><b^9VGhf@9uRHTqlFsk@!Zh9y
zccNX^Dfrtz;YYsp@a!GP%6Gn+KEKP*`m^95ZJt3fVCQerj}##P+#5Z00-^tQZ~yO^
zsR_owvIo5Rh?1uqoai|bkVL{A&!V1wNXx3Fkd^{S6<>t%P!AM7QqzPUU7hWK$K&ag
zZy7<`#;s{Lpp>C|dtl3vkrf$E)uh+Rh@W$NPq7&T@fOyqRA!Mbp)aeIuqRzeY5%*H
z>Yed($1n?(0m0I`Fl9qms?DY2z<KxOPXWuH+pqezs@m=`jUKLJ9Wex6N1QhC3#!W+
zY^Nlg`Y9MEO(Uv7UI0CrztIn2;;|QC^JPUmv%I-2_&lXyJx{OjreC`^8d=(>>ei2%
z&yT0(3MwmlM#qXFg(RW+lz!@5w}77ojL#g@XBy7sniPj5_e^jCjP+vcL$G4RwfcLy
zd{a!K!x1bpX3nm~bm6FER4kJv=A{5^eJ=?g2IgS$!#>+@-ymw$1lq6Sl+E;G^o%iL
zNni!6UQy;wsp&>D(r{A{Taa)@cDMQ25KW5FGhD9zq)|8Z*|x|c8cHs@e5Y4Gg`*3X
z15(_GS=G8ez*BTm#yIdz6GL(a66ffMCRVtsdR~dBs*u}hq$QKW3j{Y+lk2&Asx0w-
zTBAlxL1lN>7}5J>ubL{2yEY?Ig?cQBNW6*=PyW5qNrQU=ohsh9x)M@yB}H9TVK|Q%
zgX+{M;i&6{RCKX-3mDmlgQjmY$_*0fCq~Y#feI?-H_7yx$g$fsRns|9@Z7QEP2U2|
z9o{wmt<y_Zz=sk&kcL!E9UB&k`-D9c6~U7J=h}4*CC({#7(4$q;dd!$)qV&Xia=2P
zs8^sem#=Z`($!I%*^ruZprIG-nKSQn-1ERA=t$ucNRvJx6NXJ$#gqER*>X6+)&dP~
z!14p>ckc$lfZyh_xz$r}t-6A|GMf(!&I7iMbN`(c=lmBKTykv(Mv<ULGP{?(0oN;5
z2fxpqcBElDxBC~&eqFVTT`5*@`3$a>*4<5bsFejGv@zD+Yt>T?7C%ImshO^}D2=Sx
zwT5w{{u;>#Xa9Jz+M0i0cy^CwDdh>OOrv>#*}O&e9>%JD?{f+vY<M-?#fd)_Kj?Jv
zV?{~<R$ILnv4QA`*=|_9we0raUWxp^2`kSec2PZJeS4i$0VNFZy@db?Y+3Oe{;FhI
z*Or43ZL0L~l4Zat#zuKnHCZahjKo&Fr^&qg)#dIjtfzr4)^dovYt|n?bjb}nhl_1v
z`EO`91fwGG6{3?wNl}GsD)Y0DA@NKWTWWv|J2pslsfqN{WOlt(i%Nils9^?wtYC9a
z&t%(!4$!XNfr3RorUT12h~WoB{!sJ+h>c|pk+hw;_Kz}yjdRY<*x*kAaV7EeqJaHg
z9KhPhOm>Vilb1(3og8P<SiwQ3%>w}UT|JCN><&<qt$u|L{W*)rVvatJ8pRx7ru(cJ
z(YHMy0`MQe4PYtuzK(`$S!Gm@SMS^sfoLm#9DaXIp0oXbk%3x5>GNHdZhm;3nlj#9
z?MJCpb959_KW7fE2q>t06!801nv3kr;tfXJAqlN{Yt(A1c$oJ#L@lIx@(H4a@_nZU
z{=cJB`TEMNvmgC-b|{9QOqhkYpTE0$7<Zb3yFORJCq?T4B~~}nn*vV1gA8g~o%~_B
z;_}|ua*_a$)Z&F{zHTpj^JU8fO!E@-NCbSPw*qpsxNMMi+kzsf!h0-{vhJ0vMm>Gd
zm4KXzc%vmpW#&xM$B3xdy<2igsql)9o(94q3djyNRzw3NyRkftO}tn~Sw4e>Wk46w
z5X>aL49p%9Io&UN#LResH9$sNe`2?4muuS%Q~OZ&;#y&O<k;%N6VTW+6Euz*a*9tb
z^iwv9RyCRuMj`=vN&rq0(nPmKenKJE>UHkDW`!?xYne{z<X%vRY>zAhvVq3WR|8^O
zjyWFEpNy<n5P%+Y@}A?Jqr1EAQb#)*XIr$V@|i6K6+#3z@<2cCj(f%FNkEwJ5|UR2
zJw_s@a5~G&jaZw$q=OA+f<@idc;gUmPWXEKFb~Ii4k{BzRkX@;gn%vsZ>k2DJGM2Y
zH?$132bRSYb5Wll!`6(xEeu4+8y^jt;S+{*w<&5*)X`=q+6@9TZRuVV8yWB3#yE?c
z{+L5L6g^|z&UUof{)gtW2?Y8cRJyn}4f-Bi^)gnts>kMk__Q$pnsoE`bYdou;RIbz
zdWhFSj;6iw)yJ>5&*=+ZA(+vnZcW@*>5mvr_{`gxfAo84z4LD@EN9;GJzoGJ(`t$P
zEbt#5{!Sypyqi?$JivW*<X%Uzwfzt5GxyLJ<e<0x?F;S8@mtQE<jw9mERrg1Rp~n&
z=pEN_9tD8=diX?JZp7|c=-Td*fu%`aN%UXqjX3q_C=_48eaz$3k~GPp^XIWbz?1hF
zPWH(O8&ty9kt1s^PT1g5HV^!h&{{Hmcnbj83x8fzQGS%JSc_AXe-kP`mOVN4O?uMF
z9lMTt&-2UmLVYgvxjtQOks%7KVcI&mQ%kCT%0vSdR~{x0)hU*?zKLo5zNfHZa*6i1
z4=w$&CC=2=ZbM3WOh}xIW?q2sQ{|Goq^qn3Bu8l1yjsxT;SU4Pf56TfFBQnxUih&<
zE&(aM<111thG{5@@ZbrC<nm~P-jm<Lh!)X=k*;g?XH!j`Dc?5@KQJ!NM}t8Wm<jO=
z=78(YBvIZJU9TTcTAZZazkbTw3pfTl&E)$ZRGwXJ#}Wd2`8!rEAU2+MmK<D7s7*h5
zrf3nCr;IqO4Tt5-?2Wjq)o7o+dc;QD?xslc0*5CWM=YBaN~yT-Z3>W+gefR3X5ZV%
zbZ$M|nLsRq1;}_Ui;6%``}a77Mp_THvOaK$ovu5Fw{)`rV}V4~s5@mHZP0O%j>wz{
z=!mQc7TbS0c%l7rSUCL~iyq-EEOQU|D%GEW92q-zV?Etb>6GHqu~hk8&Hek+VV%#f
zHG{6Ch?ZN*HNOTC#nKKUjT0Q!+e!sKb1&=tYOth)OcaLls~Jd0s!rMZ<^QoSjWHf3
z5&CI8(5VG&=95lyHQ%Yhnk<?ME+=36h!7SF+o#B>NHib><~YDIfz>^REZYGsa2o9h
z6f2)3T_%jI*gAdes|kyjmAq`4@{|iTJZ9|!J71{ZZpOY>ponqTGht!FLQZ*R6-qu2
zLA^p_2{FX8NCi}n^Ax=Ouy)0$5&i6emlrt368KBgib7Yr$@f+zT?^muHjAM9E$B<d
zz8W#EdBGA5Tftte(S$D7$J34dCDz;dD?tF?1;ln@;TJpVYbZ22&Sq5{hYr&vOvrH+
zt+HaPcGDJO>0srjuAM}gp`jlc%&$d~qu1-U0NM3O8vVmK)I@+^(^@#n>@f4H7>z?X
zJa$ZbPjzRbTTM;KRszl1^QBbS^IkaGY3%8Eeg}H`z|heP3WoIvi-#Sp_P&4Z|JyV3
z%{#jNwh06~gRXwQUuqNrCL9~6|H#JW^M(P?y4tw`U`?)3)^UH|2Rm+~^Wk5(EB^_f
z!OZHU?*rk|n|OTye2JMOKu>@Hx_`12bfO$|oa_O%UE$*dclQYDQEO)^bL&L?=rCFm
z1zp5c&D7o3S*M54cO4)$|7U=OFPEpWn)r1ZHK<NzUwEFv7l1N*`k)E;2EK*|U;p1B
zzZjm-(ALh|`O=ENie4w2WT$V~2cun7CBiG+`SCY~)RW`*I2PS!H&h0yZ(i-<sMX`n
zrP!+PBEbHPv;+nEMy`%c6H9|cB<i|<GzNs6s?|eG=I&BwwEl&EsK}f~^n$idddlyP
zdU}Rtbf?VlKb(90_7FPj)yLcpRM_>VJNj*j;?s~1e|?XAhf=`{J%`m7-B)oy4i?%;
z<S1+5F6&G!a;}~rwb!gcgZ@MLM*nnSx%~-}_*BNILI)2(FmS-5CTQ7HNu;23ex6Hg
z4hcfIyYH}=R%TFBleI|?vu6Sdz~lp+cq}E4j(0OJn9;oF3e0sg<&U&#Wy#h{tm6iR
zH$iHb286Kha*%Tr7G~5_Pnf&(=Zpx@E~AjX_@HCkUWYB!nwm`@M#Z=giEaBqGn5TH
z<4K`tvZMW=R1cPv)M2O&dq{?|F2BwVrfqA)8yn@{`0TnnourU2A>lzmL>;$$_s$Hk
zcy_yeKNyI>h}`2X$jUDPk6q9XHVo@Kd~2ULH7uL6-~iHIh=A2dbzs9XZxvlxWdAm@
z5_|d?aGM3n+`lZxfzsB|OA4sW)aC@h@>n?|jU?r*;yvn*M)GsQlo+yR9ERq_0#2F~
zJXU%!R!r^3M86BVrEXRLr#b~-Ru2CvGTg@f9Xi~`I;sbo)2xh_Gv46Q+!sOQ3274p
zZ;$e9JgN*8Uv+g-6OXcBS)0l)?{f&9<|i~Y=@y3D1KD&R@nI7HlLy%F0jR#>9v^x9
zL#W!zqBO;;woOwC2>QX^k>#wB<xa8H3%~CLc7q@(dc~P=V#1>{mGrECBl185c$#PS
zZXIO9*Q1I}d9^icsJ5cl|6vy2VW!cDt-xiMjXIAKU!+F1ugC6Opk#HX46|asRTTRE
zL(A`l``5Dch93kdI|@1p9xA5=<B$_PW>RG$?s}2)I+GED#^Pz_h;j@Y3T$y%@*HYn
zn=CbzJpAgO^?XrpQfqf^0y1iO@c=^Ipi%VIzS8CE(yjFHA@Rl2{KP#8)`hTFw9Ps4
z-OXu*-#@;oQ!`?wGSZLNtMSQ;$ZX%raP*-=dD6|ILiK&_$WOsYHpPwf1qpQ!ZE2=z
z>KXScBqVg!CCSFwE%z}osUOn&RjKqO5hF|zIh~6gmIXZ5TPdg`*Y(pLj82YP!HrAl
zh#L=7wa5-Pt_(g<Xun&0KV7UY6AhAUR&X!e^eM8?_+0l*#rs-A&_P31j-UYIy;9{~
z{;uQb#1AU+B+uFR6cm*8^i$VJ2-Yxn?%uhhEGMHo2d#YgBv{cLqAZt*tB(1yH$$0m
z2k8Vm675b=PqM0$ihfF*cA+WNqHIi>HOMS2tWxN4ed7G-)6%}s#gkvzydW|^4(~b-
zevOPX5O{afq<&#@xBYJ~E}@AHy|t&5j+&fuufp8gH5bp%Od<sD@KD32)Fi#WX-y8U
z;7_8oA#927>k&N;Qxfx1gvW_7+^MC?W_vnIk|wYd_He@JVv&qikJj}*$hs!9B`m)L
zOfi4&P&E8kq%@!y$lnJXQXIG9JT_m(X8vL<_mu)QR3mKxi?Of8=dxT#Z!v_!AiG{{
zs;=~1eVnE21Asz^C5iN$(|ujLRwe3cH`6OhX0OMhF}c*{KeXoNTWV?4ldq-qz8>9E
zOQRmQC2$yv<?T=O{O<%c;q9LFL*@1et3e}6{fwno`>;YYD|zKK(=;jfzE9b5^O7dS
zi2tT{2A6N44Wm!Lb1DRJYjoqCsEB-G=5fILYBl0b`(*C-CKiMXZ>*eA4d2?L3-bL=
z<T{6{vL(3A9lf$tzZO~U3X7_|n0GT<&Y*6_JIsCnzWk1#!XlF7w`-r-Y9r7Er(?;y
z!%p?P<>q})f9i#nNm0y=`0^XO9S*V$sZgU?9c3|A^dOIgZP?teh`PMqj+Kd3kM!HJ
z>RWSrkw$(d6-|Mv^Rq5cPE%%HxE8cdUQf%27-jSN>#;QH2aWU<BieNiac$7mlb5N}
z=b2i6yHr&2xcrsTSap$*-{HJXg<Z|~sgg&ZTaYVyb+IWTdH7EfWWWzfNc9#c3U!*;
zfYa)=durK$^ZT?x!`G`4qcOGdn&c=sL_Dta_+sTlT$x*2|B+hu*Wl<}a?m`~x9F8D
zR-KLO++kex%=vOBr{eL}2-=I>O(RRslRMd<^3s<dk5WX9Ri+o;gy(Ou&k$eF)<2wr
zosAk{U+<QlaMZ>1(!U*>C5*xEV_;Fg9>1izEwwFd{_sW2yvGXO(;+ltJ6A_1OG2H3
zE0ZG2G4x6{k#S*!Y_;w+F~z!kWWH{V3r8}koUnA!g_V}xvN5#@T`HGTd|Inp<Lcr}
zf*7COj07F#DOcwwA!JQgy;met`K9>X9ehSPosCl>>x5<x;vR7`7B9h97j)(=P9+2m
zQJod%IDGq_P13GCgNW~oK7yaFSN|sYXzG(BrVjW{2<T>{D9pv9ighvvUOiP)>EdP@
z;w<@A->ihEFWlBYR`itqC!PB+WY=*K|97nG)uxt;7DE*oGh;10onUYOq)~c4)?l^Y
z499)gM!0ddp5UhmYv3?q4L3kqP3TwdE$xpt&tz=z=oH^yjCD>NJIa>eSPVgwz7OD;
zj8_rnkwWM~vQe?6RNQNIasRkVR0KYLwCgO{WZ|rLW#alqiG_#CGxKil(v9Il@M`R~
z<}Ba0;~j1>eL4&FM{+G2<EXG4w#3G%Cv45rLZ-LH=B3W<3=>|QpDE~!>1^VLAe6Or
zvv?V2<AQV7=S>}s7H6VTy(Q+w$+B(7e}a!*4+Xut8$og_yYH#5e0R#wN_{XJYcrxJ
zHFwbO)U^|$bj+RH$SQMX*K^($g0ib8mt<9Pzfq6YT%>(G!FwwDh@#cIf=#fPf6ab)
z<Hs7=u!~5O&#%)7g%Ez8>!m-X4EA^aegAp+me@w-tr2C>x!tdPni9^JNCVE$C?xM{
ziDkIeOjyLnqK<eh{u`WECHmcxd}<HM#GW9si)v@wY$pUPEVLh63wD|IP+e-I1RVIv
z=yff2bPJN|>TMV=bqj>PXrDy;U3&H;+P%OCTz`+Hr{UQ;sb}%){L9R&f{(g@&Bn=?
z>0XsGCvrRdQOhD*XDRfPN!rNbdoq-Qt87-aSmS43AU`*{ar#6i#xU%~>u7Cy(kM=i
zR(&V#V}j&z<adwVJx<FrW<L8Y<zSnga#5bXD0TO@gci;){#&aomgDDTVwjN$2?SwI
z4AbA`&WI2r!;pL)axFaec}ibs?WCN{MqylX0n)Y!r#r0gFZ=4kI0)p7czB{JMU7l$
z3jI1mJf4lY_|QuG`72E&S9gqZuiDZRGmHPAM*cZdNCcLeX3_@-(1h`K3DKWn20e~X
zHn714H095Nj_GfhYdj%jGJK@i^0;xOmV2CZ29NQ>5Uz&-JqtPvCI(F+XZ*s+dP%hr
zI=`|jd=Ce=(tnVTKV|9n$ccH`j1IHQ3^=jQ_0bIcyM9XZJF$}9qyF${&^mKm{#!xy
zi#CQBELt5KG8-$SzGwr9EM7}zA}N(+p-087t&_{OpSG^tUcw|tYQ4Z(+i~qQY-~Sl
zhUTB8l8<rK71@S3-=aDXpQlCf>aPS^yq1XT(kSj6Zfos%cXhn?Tn+1*cQ5tbWpKxb
zy%4xekHYxoPvd(HzdAHRNM&_qb|eTg$n|=PG|=2-Hyz038Nn01c=ihAKrys^Uj&lf
zYpP9dinPDkP`QRgU1!>Mrd*h(i+Sr;)9^KC=PSJ%!r<HoUU5!(TE)shR-LU%K+c{@
zoF?eZ!~9j2_+WAcx|*HH@@Oa|!{YLw%&0rrg+MB2&iQ!W8rThvE?e2tvHLc}-qt1P
z=CJo^3Cj~%XM8~QubAtt|1l$F@nh+eBGfURC@XwTu0ops;X{$~-{UFxixWSiXLQG6
zTpv$MRfDeOMIFobfZ3#K#tkeZzGce2)ClZtBlR7N{20fB2t{fz2soT{$&aa|QEW$j
zaOkIOvCBM@%3cV@7<YGRx-UweiW{+hu;Ju%lm+{HP3e3cA>3nl8e7lU2GaP+a=zq6
z>bnv5wZ&q|YdtPBFE*H#*lvzPu#@U0B;`9WH+_CcuEnW9!kMH}x-f*T=~5SQLPF^4
z*3j*Z+2GQ&o(}8%Bvs$5Dcv$u8MO1>{N36|8xd;S!8fcyF_yb$8QS0J_+(cze)u|#
zEH@p97~B%I|M?XGz6uI7bm;d7hrn><v}%9>(unX2UmZwpW%0aJ2vRVb2u}*@+=1pH
zRT5g?l>H|{a5~zzj{bN2!!o})?_}3GJkBPboW-U#Vhujn{%?g&p^erQd@`>Fa$AQK
zU5NfmIQ_qR!i_qB?j_zt!NAGbNFL`01~&rn(v<p(!#5tBdm33zwy8MLE}oSs@E)<x
z0v-v`dEJ7#`E|?{r!;(WeMVX**WpjV_s~up$6xk=iRiW;>#(CpUvw1P_~YRTx&6K3
za$$8G!pzKW95VOSY()+q4TG3Ja@@a>Z*g&d+A}G2rz{=6oD;+)+0i8{iZK+)sB=t&
zXH8<qzJD>UcE)7HQQXGg#|LxHVWE|Q>Dd#0Hf=yA{O~7+dy(eEqJVhIGm|dUf^G$^
z;*JR7RElmb$0+T>5y#o%ET7n(H6xEE!qRwqgY+?G$`_kZQuVmJQo%(ehFr1s;qxCx
z4q#R>xoW>;NoE7h`4|%r{MwwMd7e_F53dtFxU16&?|<mf!j)cp%D(p1gW-AKsIL4;
zL_rq+uokDdkWG;%b(FTki~~>R*EZE|!7Fg%h1e+>={tJC%ocWB1E)I#bQi^8Zd+FJ
zS&^<wLVL_qszrOowozTr|II!Nul-#12S-c4DoQYJvrFcTJNwbT_jInp?{DL>;D94V
zMm;hID>Nbi^BP`2##T!yA3wDBXEchU^6dL*0;0g8ggU(Tv8sXCrA`XiZYtRGv*&KA
z9;5RkAsGc<oPd)Z!e7~y+cOf%TBFs3M4QhZ^^RV8J~H3S5KHOHeQS~UZyn?E=kdK$
z*w3ZEu|7r$98vk&y5_@=_@eOSOATFSp<idfvO)e6%Q$iYEo~*U^N}7QmYGdYszfVn
z<G{4!Ux}0}`P^Hs<o}1KuMUf<i}tnvDS1J<M5H?g1VoU|0f_;I?nb&xLAr--C8fJd
zLb^eQZWuyZQoy^t-+jLS^%*^9pS{-l)%uxHOVF2aCB8Mulw_bJl2b|(d8W-+Y?6*B
z6wzh?XzUdKju#^P?YSDm`4hu3D@X6P0gVN*sh-d{5~ttKrVvso+%c+@YC`Ar<xb6^
zSBtr125&N)>l11b_Ig;Vw)zwv!W~Q7b#2t(Xu+p16=Orqx!C39HsSensdUm1Hpi~*
zCd;`TMcr0M2678M*s<+HyxeaIW@7qStNIn`7~!<X?#OAmNdyc*kVn8lQ|S@cLs&Oz
z->|Y8*g{oa#wVa7>2qM%*)8|IN7-~e=ex9;;S|*Z4YT}iGuq@pdXRYKIMLSW7eDUQ
zJx>N5s`Pf7IAS!5lFtM3ShcOOl6Zw8pZVFSU{Vk6;~GB_5e>E)XXQN!d%~bo9H0^7
zo?H5jUWAtUzx4o`DR6#V3;l*6gIZ4zYHdHOe!MJat+k8auPeWs5ZILO6IUT>cdJ}A
z#jinLdhlSw?X_*)Y=JrM<BGR^&wdOFn7Y1NoOayyQ!`=xJsNvUJM1)HKa%a^)?av-
zQL|hN<WMfcfmX3`BxqJ}N3JMqCA!(>PX!uaJ6~A#7aJ_jquT4$I*}OZhl~+V7sjt@
zI1P-53w#bcKU>ix1k+VC)*V(&xu)JQyN_wyE_GaW9u4{awL3ZkqsPi0Dd8-c$h#+n
z99p&#T#Qo8cE>a$a%ZR2mDkPbmkU&FG|e5@%ZvAPhUD@M_OrDdO9mt>-TaVSfenQi
zi`zr*SGRPipb`ECesZR8TJg<jiImM&*VgYbQVy-5Jiui_OZ*yPW1Va?8*$^}>s;39
zhWu(r;Zd(25rh`OgKhfid?Uk*jj>bX01JX!#@`m2@`f3Up*?mP_CQKNj$Oy1WdrmX
zQh0EC2#ErOs+vqNxL12B$BZK552nRXig{*d5lw1cUqq-`hDze3C;NW%I|RvaK*B~-
zWcorh_%+b)f-ZcBfA&C7D_A;qJ9#d-=6(2F?SH{W3)~)lEhBZGhI*9evV%9xq!6el
zj-6f$Mm5q|lBqk!@4uMseoIqrxkwmNkz0w!JvgCD5NM&vKagiAFtN0)w<DBVv&DDL
z-=DNki46SQ@Jy<)^=DO6-SQ?l)^^CviCNSc9%BakbMho%uI|#!nM$z^Bcna?DVpu2
zc%ocb*R=A^YehmSGlHHBHe*}dV7uh~$1zs1C!dy{$<(CuO?^~e-UIt~thC8v`W}Lp
z2lx-Kp>^=wf4SJXHA9me2fc=j$xy<nAj6re{nWHqSTTRR?D29gNd6~GQc+6iW?jHd
zEB`UTW87v`70OfJ%x~aRYe-;z<b7N6fYZQA<&o(5FoXY=^5QKZ(nS0By?xm`u2%e`
zFiZpq`Rt;i?F^@zwt3+H=~WjMmDaIkEm%AIuBxaz2*jk65;*p_>jCMR9Na-Txr&5{
zZ&-8jZb1*DZ=3^9X=7e?qH>cOsh%(0^R#j9+so;$gZO}#BiM*Ez8$#GU2vGr(VN1O
z&+ynUEG!FpAtd$ly1QXvHbfSio&!zUT$lvatqI>0V4XFRCa*D#53?~A1@%rm*p_aY
z{#>|c6@Rbf5D;~#8^X9~o^2+5pr>X`j#q$xIhIU0OuJs%QwNzLXV`~QjPYBqzw5cC
zHLI0Bw&6F-e;<Nd@=e_}vcOO9usn8O4tI5i4^2T$Zol<<LY`)(7>fd$Zt)uYQ|7?d
z9Lq!=T!I3>K{h2!V=+s+TDF-q2HXWnB#n~7C^0_cPy)@+#K&NE47mk$l~|AQEQ`r{
z+*^nRK|#hSJ-ckaI1(0~LjcXDRg~@Am%Xe~5q;dvX0kq{E~lc76E#HlpG4xJ9ic^O
zH318nnx^b&07mXd;R?%nmbYWZn~XT*IX{(71J29_2UyG-P)^>t106p5Pe3}#E{fj+
zg2_b~<gS#wDFt8$FH_F}?^oGT1tVSy=Cg!pox-ED`+Qsw&t(~_PWu(AHEr^WLl*%Z
zdZuFWxPdkV(eZDMc;uyp<vWH`=hJzv=k$V{2el_`fo5RguJcSfu@vmYlwgfq2+Tk!
z=1k$eD}N#?{D9!XUQd+m!Syody0D_40e!O43btFK&2FPB$9h9=XyUmXiZ%{F0*KZ|
zmsNUCu`#|m3XjsdcEPW&<gv+^l_d22c`W=1Drc%a&aE+6@R!_lpXXBJEx*y>*8VXY
z9D9TTeY`L}4ogHeEC3-Ul<28thNkX8lS`Yuba4;7b7|^leuSW`b5b0kUs5?{V;$jt
z-z=;Hocr?vXB<HsM?MqU)8^&s_E&7=hC~(tx}iL}AW*;70VmNj;nag}fVr8W@cWb$
zTQ1FBD3gSk3SH}Bxyp0DG+urBO(_@(RUE0w?8*GVR1MibsLIcad<Xl+$I4^P?<#lJ
zST|#cZzfEs@XT&%#>|wWK=+{K*ZJG<YGeGZCq++!NfHg&(QzY!i@2Yu4GJ;5U*FYW
z%gJEloZFKR=u=wRk_Rdp`Rm1Q%m>}UH2mwWhWOh~Hy;3uPI+3retyRs)63J_K3NZ^
z7yJ_+`$6IuuX>j;Iin#s|9OEp>PUb=xj^K(sFdcKJTiBs@>_BELA1ma_tld3D-oyM
zz8>W!L0h0HIlj4%luD^ysPo+Kk;&hbKw>BcCM4NS&krYjvT<^+HKjHm1|b&j-zSF4
zPdF$B2~=pIqZK?8JsB3d9P>>}(*LWM_vSSJj?7BMh$*%$W@+Go{+!K58v5VS2%Xy|
zyhEQ$&h;4ghgaEuBiub9&)yRXeguqqx%d#zi7oEX_<N#>gwcRV7(xuqhCPC$e(U>O
zEgOSa*R?9v$KWzh`@oxkr|?{;89k!iST|>1H?W5zz7Se0isDInYrP<-7a}kt&6D!|
zQoL7-8$^T894Ro49bTMgwEz3<#B^%fyM!3?mo;47JP<IN>bJ~nd3A{8_y68L5dft%
z_uKnK-2#90#C|+b`<YI|S6W23OE8?Xrc>MJ+{bBmio3v~s_pTO8L#+c03De%3{_gi
zZ6hS^FHSPN3wh=>2rcK(ZJ_bz^??@Q%+_Q>FTL@HY$B!tV2Gd3Uol_VN)#SgrxYt`
zKMmv<VAWOI+D_{`H;RWLP7hB&V;tN1x!lAdpE2Z+iJkNsQ;cmai>susCiJk;$civA
zva+O@wnEZ8VaqI2Q61G&!op-+-ZbwFM$DaN^jsk&Cn5{2H2JW1wNtvAEMf?5JT-=|
zdCygtcdUbQ2C1`VLb=sQ_CDt@k&=@(hnF+cSp}7K>-4cIkzi73ZghM5j^2GXWBYM*
z{<Y71$oyAd2FLUA^S9J7z9K%p-`jYzF;F5#X(wy@zU1`-Z2vu8TT;v3mq`A5yn&qi
zHG4br#zI)&@?YHo4@`-;N|vmeQcshNAMA%Dm$Vw~WXs9fG$b$(aW220I?O0d2_h?J
z_s9|EgwqMC6De1`_Oxo~=HUPs9W^Q<MY+ve9@h1DJ;?PwFy?j~?N+9~9T`FT^L2jc
zNk`K+IqPrEv3^v7s(#&CvwVEI60b(F3qDnT<JYdYv{c`6`t4aSOM1;c#I)f<<otJ-
zaZob3VaDkx!P>w%#k%ZIB7uy)j568_FWmX^$a8>dhSa2-C%K^4h`)bPBCuh-l8f?v
zqhMbfetpoA+Z@H2i%q8URA>JYg#71!mV5Y)`oyKi4+GrvUhXC5hjyOjwBG*3zRY?t
z#MEdsG~2*Lx8qHam)9il$d!mqqJqzx<ef&{HWt>>TY0C$m`BeFzCOsor4q$2Gw$P?
zJ@67piY{lVWy^%IjA{)u5gPr`U)g9XV=&SNq1drR^fu54rU6fsOHjVT$)X#Mv|krn
zEJyIpTchzxQV4WgtoI&5$C@|w-;CZ!F^dXO;!;*OM)`%74Lnj4)1H#uLm2Blk%v`b
z_rW<sbMfg-JhPm>e)P>l0O&GRBHd5+Yd<n262^@rMVaiXlYBblv-M7Vx_aGI#&06J
zU6zaH$9D6R!Un3S*WOW7(%L$;f$VVWs{16N-)W=S;&W~|+O6^(Ode&?-8_S{*`Hgz
z>LR`tv2EZW6C$Ulq~xbmJ?|W%U!WRYdo;Qyztr=gS$)dZSk!k0TciZDq;S#@@7J|i
zBDhej7})O?KVJWt&D4GTp^Wd}F`EC<X>h+81rV%`p_{H;Oc>0mS+UW(8I-l2qGL(U
zXX3^;1n;T}<nz{dRdGm2sP1%Zs5xyffh(~&TtK@!qnHhDGlxUYV1^dKRPtPcMLoZZ
z|H<jHz1BIBrB2fL$DDYI3(yKx92uvwh^gvQZz@8iR7}^MAkRteS(v$86Qk|M?Xg&-
zB#a2?&L5M{kn_Jr3V*P#C55nsS8sYy*F^HRlH2~ZBs_Ik=fze<&uWCcp673V1O{23
z`-#uqsD~DH1DjoKH&H{AFn(=guIv7D5z)I}LM`W?9QTIK+r8V@Es4YM#pACk3UW=R
z1y4PhgKEkjCGKR7X#^x-A=Xs`tthHxh^3^8m>}$dBhlRlbs(wy#FF6Hlpu!*Yabes
z**^5W?s)3fX+FVB^}k*catWyPth}#)K7GLTpwk3L5ldsR(hxqcVMgK?^;%mpB70X`
z9Ox@WthD-34Wt-GG%<dSEaj9G46#L2Lt{j3R%$IAosFd~RFt{RvZ^JNbnj|n7Mql3
zr4&nylg`jwKEcJOKWnwo7jSX1AEwY@6_29-QxjSkjN9(Rh014z&PUT5o<#q!2iBY;
z(=ZF`+p8&x+bdohFhz}ac<d)gWvFZm=F*^cM;NqWj{cNYb3EKoTpc8-0J?#oHiAfy
z3b~HbpW8OYcSOzhajzbFe$Vzbi-`9eroiCcHj1{`1%ZLt9$ryNHm4N!S??38Mg3N-
zWByKc{OJN)AO6mqcRz|fx}!}pb4@j&n8vpwd1WA>#*y2nt0ojcB9-&P0yMfg6V(uH
z*UfRM=p&ZVu*ueFGF8E>5MlBVfr3G29lN=_!G9<EXO==Y|2Sa~z~q<T!X2HLa*xu$
zPQ$h=g;-)x6agV)S$tZ-e#;U0T&5;;OqkLcWhytsy(7c`fQO|lg_W9jpTanA_$6Rf
z?1^lN_*FUKsS?HqpGCjv(P~L5H%4v=$#F9ptvNDWSH>_toLb%2&aAi5t$r|qt=gUo
zGDqc9c=UfPzCCMlZw5TZPPhZnRO=J88{9RtUU*QkID+`fN=!e=wpJ3`Ie&xOk~~tw
zIQlp~Ba4;W4g<t`m`M&1L)8$;uKeGGQxu`}74no4wDZ(az5$XSe$^!bI}jKcs&9<u
zopC>a*yX8!Hx{e~?X5FOWoM^K$u^H;&R|`IN14vep=A7QURhyZL7V|*!B>PieN68|
zzqBEyhGOM4y%0E?-gnQG@Fw&`zbUW@03E^}W#D}m`lb)t5qq0{%&uN8JUl9CfQ9*L
zz+&cl`c2^`7(f=bo;Tdw-g}$iK)14e&r>q9NbTBdMNM+;&L-VLy5SC~X1+WzaGO*g
zkO7;+1f<x}<*F)1d#W_%o&#)5@A<AK79^C2xr4E6?2=Ddx@r`uBYmEsxUs%GS}7mB
zxv2XIv20>^(J_INujur@dWHPQli{7-=o<|EL)*IHp*Gf~b#!E}MpzCrI8xD>jgq+W
z@@z89=$p&e$;qLbg?8zAQlo7B0vXb5&v%y%cAEfvU%6aa$(~tE01wF~Y)q0~iGPw?
z%7nSs4OTKSH6M`I+CA<|=*}v~dIiOKiZ2D9I8*$9BP^`Yr|dkM^hg6qL?l7-^xSh~
zOd+Cz7lV^dNQFaN=74tMn^DzV-EK33l?r|#+W)5C81Ayzzu>&u>%6UgXPcsW-2X}L
zrM%Pf+cVEg4^OK!__Xa$d%f}^;8H$0P2_*kzKk+?ee+v5?yWFsnFkjlhi5>+7Mibe
zl<x_zjHC}PWl^wIE8M_x@TdNXGvgKR)E*hev)iNJGj$FwHFOZhz3G$6g$p!579|@F
zCgX-}b{l<U?xS=;IsFQX5)u{XQ2~40;}A!mY);zI+s~=lRr0)(rQ#~|6a5n~dC9X6
z)7zwaY7;On*UZfFZtMm2dZA}h00&8dCB$tfnr_duF4TVHN>WEmnR%r?&NS^(Z=F9o
zX8X#`))m027lDPTGJD`Xd`qa8%(gXB4~4;Qx8I=HHX{FyC@-&CCOKWY45+2EPo!cn
z<0_!AV+uhk3esD+<EObl<!LI$D{zQ%Uweb25?RdV>De&U2{9h$641-1RD1>4CH(*!
zCKus|)+b}{1ylN5^rCW81%Ca>>WmSo>%A99xGYBxsd>`(p6r?#SHdh8bvpF76EAo5
z{(or`WfJZZf$lo%_H1T+`<15-Nf)=s=R(_GQ^;m?_%+G`wLcptjjLO>Fl5mSrM57N
z3~BY+bL#KvHq$h`8QuEkMrP6KUcWxU1f=yQEhf&@inDvOU_YfdO2l`}w9QOz$BQoM
zEW&Yx2a$8;g8%c`a%j91PNG2QB|5cTpCK8F_P^1=`KLK5iR;?+E>ThSKfI=T%WFb7
zT_|DnY0cg6eu%7%Q%68~*o0KRF+tuhW%249f?~hSxCN)kd3@HJDw+{Y0waTw?u03+
z;n^6>jeyo(ydJAh1!4<C4-TYi3HWYZ8}=PdmQ2cw)TSwa{6VpT4~e|KQ(K~U9dGn;
z*08v(gh{R$&=)I0rAM%cYGkk8htpC;m?ccu1Bi5Pb6YzQOhy<-P|V4UBzANVKJ#Bb
z{SaNQNxE>|1dUQqCK#(GTUXb>5^gd_-&q>g_VNW4!sCqqsSNJYD5mJIr=|o(WGs&Q
z>oN$fcW8Efc?Veb+#{@hX9v8|+YdGWaNvzSzO7x|%6v8?@b7rGM_$h*H+bBgB)qW@
zQMTv52HJCj#cb|locnZ6kxo9y4Q>P|!KQHVg85oE89&t8g#+ztsLjn%4ZTsuMjDHX
z_F4*7xz#G;X{A;^eY#ydN&Op2MG5UX2jS7rm{&o4ykP60I==qKpGJR!#nXQIpcid#
z4aZ%sVnqB7LKtF!ccf%Yc-@7}fL4)^5J&GTN<R4;lf2S4X8ilvct`F3@{AJ*&@)vn
zVoNN;W%-|xRF8qlL3+WcVPo7z_B6(3H+jA?Cm=XKtO-hTe1aw~4<j7ta>K#V`!Q1|
z^b9c+IKNAqQ~G6rtEp~=(8HdNs@jt-O;kcnOLCoBV?WR{W8imZjB~Yk=p$cctz4UP
zLTNRkZ6mDkE+ynzZ>1Q#D^rd6G=rQUH%BfV9X`U5xZ5i{+$gH;1c~8<XGSpniI(FB
zwyfUqji3Ym7i|+80BT~w!kBuG)g`5yt14MOS<J{X8OPegrIB0BQk*uV<#=iPV}Mp1
zCRPTpPlO`v9y)Gdv<cRi4s}a!^nGahzC_G^AbR*CcV$gsN(|E=F4_pyrVKhju@~e@
zpoyhz^r1Os;WN<3)0h)s>oynfy$d9d&#}RRb8tCI;oGgb^%>RJO<mv=i@z+9Kh1e`
z@@h?+1-j7lo`^wgP$$G00M;CRGxcA&F+tFd;W*bo1c4|s{79>m&u%82KdJcm`i@x&
zjRG29lW1rl;#1cr7wZ6txElZYVFaFNtcz67nzKBa_6mF*y)B$74Y;b&WQ6V6ts9uO
zft6wABvepERbQlI`OPI~3o_f2tEj0=Fio{;&eg-bZDhIfQ~7uMR$uW1B7xw}5x(Ts
z&jn>CukU|FtR(-r*FoZ8zq@xJ|6Y|n!<0~AAR>7Fjc<d|Y!h2_kdKdrFhKpc0j>g%
zEm8+ObGk8wWg-33x^BU=;^RrB5Wj9y{4k?6dCf^_j$eN;TvdKvDgwO~<B*ndpBG0V
z-}LXB@sbu-vKGFM55%%ABSAX+iq2LYjDme<u}OBvolDdUZ$rMgO_lLUb&%u<nizO^
z*gz*1eR7qm*jP0jdnahu2k1=kw;*f-&-HBPQ<_`vbKyOAabvW@bCx>7b;}9Q3uxKz
zxkmQC35I!#7ime6DKp^XF^Z^(z8u6-hY={b!17nPcAM92_3d?BQ`|kR?suAVR4B?2
z&6Wy+X+L(zadi0o5>75o0z4C(O~zOAS=m|8Mw5yj6vju&(QFu(Y`dKc$CESwz{3p9
zz(Zh-7mv}>wM#Am=Cc1G>?<q4t=nxJ1!4qrdKb0MW_P{Vx)L}K$C$zc`r>2dVhxcS
z4XuQv2Vj2DQx*tRF2HZjshq11a<3z_8w5g2epvZ$UADS)wWCN6VO%f`b3+mt)v_I9
zPbU+buWS}5$6one(35H2ZnvO(94flc(y6i(AKR*-Ua!2}av*T>9@%h7JT@gi{p}T>
zz7K8x?a^D4&i?+tvf;Eo@XCX~GArJGE&vb;8oQ?SM;M&vWJ~NS&W0+WH9&PzyH_mS
zr2p}aa}4T%R`Dj9e?$JG3Vpl4M-_(`9$qy+#@d|zEi3&?Q*m@&cYC;WHvfZjp0{P@
zb_*619J$Q-OqD3WRwW^vEkt096R#x)PtBoJsDmUW2s4J^<%XGjzW5mT>~2#|%;e@5
z%b#C7^%66*`lhW?$Z-{3l>ls39-#>)NB!=eX&6gh{ERCO1iIxyeX`s%Ho#L*3_g=4
zeVcFH5R~mpL)1o-VdhQwz`=QMe)`#A>R8|z%uYkia$d*|S$Y=pmQ5mAf}7xP<CJ5V
z^8~Geh0gQU8jHzVmn6AS_(*6b8Ug)lY+mW1lquLS4{_){+i<%{by$EUoBcD2*4nAH
zT;w0&HaO}7QfadGKMcR_NeA4_+W%UW^kD+cx487dyr>g8L$WQ?peNW`EQnnSh8}oO
zux)MHqxvEP=`p=sp<lSd!fGHR1FIIK>Kq7x&ELB`45KV^!No@N@$Z^l>KT?6s{>lU
zO>hKv&5jA^^yGO;<*Prb5_S&<4Ee>Ig=sEWVn8@uVt=bJ=Y&gQFmplwYdn-N<u#Ad
zq`EE4h#FhEq}Xtli;Q2-3m1Tt4iIG^2*_>}Tpb+<hasHnES#HV1yj&~4#!pk6^Qf#
zd(A%n?X$=_J`Af=daeCz%)sj8&m}`f1?`6lKIXWt3;Dja!|QO+?~EA5gYI7;+UG!(
zA0hzn0TP9-T1hn)^5_w4P3@Gl!R7B;QQbj#@?_*zOIt;Q;rsiz_YI<VzyA#^^_Poy
z*x$j#!jX&pjD~MomDJBsgX`pZSM7H`*xHrVUj9|nKDq0cx=1~|f|mxKJR6qOHFLxQ
z4)CyEhZ=_Fl49`;x#}J^S!SDsfyC%;NW#>?`Bxd(z(AQ%n8$@J(Nra!=4)9=O?)#?
z_TG`y;72#pgk)29I|nb3A-GwJ5{OdM8oT&pKqUkeGnM|d#3;M><xmyVaAc4&OZjTU
z0$8n*x5}zC#E!Uv#7puCL!rOr`@SBK@(pW2=wU#@L0~GVvV{ATfXvslX=kxooYKgX
zIekoeVQoe`z72?m_ad#oUT9Alucnp~Flz^@SCeg&t^#bqQDgk<i50k}Rfs%{TA<xC
zfIlhCq5MWyTvXgR=}O9PR@yRCJ$EROQrv}%-u5|Mi@Xa{p5kugb{W9_)TST$Woalb
zX^>79{9&+T;vcjx+V0vlz(&jQB%^uZ%u<k?<@BAMHUm8mt&Yunq=tC2Xv$}BvZNQb
zFSH*|O_MdEp9yRsR|K{It2X6&Y@2D8QS~(6q-m*6OB&4vH$K;_3lL~jP}p@H=Zn){
z8u1JJP`M&GtmU_Qr$HfVpr&v|@Lu#iO!|6aVX(Z8N3FOcgpR8t#HmgXKHw5A69NXX
zy+L%e%35t~Hu-X%h0W4S=*1j`16ShF<!`S*fVU^vojO#nuQ7=~WENlUYqKdj{sqka
zL(OzM9S-?tAYITu7I}H2E#ixcq!~&L5>%npyN9Rb&`|AvW%Quyo9uUT%qq``tkGuu
z#-!@QS~ty8_Qo(=Jky76I)H1y^T_tcJtScoUqZT*YT<jy%bw-}!@Vx)pRQqXwJwR^
z=9qH5RivfJpS@|WZ<I717IF<a_TUJ^dnNjnjT5U2eayCwDIC!dL+s<|-%BUE+lamo
z)}yoP^}}P2!8HTEQN_<szR=7sw|!C=t9nPoo3eu;h3+C|BkOGRU;UoH{hMF`m%zxl
zY=`Fm7<Sg!Hf`g$@T3Zf^Qx0a_cjyl_0m{=&7|++>bFm>J#NwS%3}U_ZYGW(5^!C!
z-C=RmFtS!1H<S7S{-Ou5v32psqFfv(ni`2(7o8$!?p$?i4wIC`k^@e2`GEPtwl&}s
z)t{N;=b8n4E%}8@Dcl_!Z3CaQ*RHk8ge9Df4ygVfR9|^=ZeTO7yC2bh<wE^aSZ%=W
z#WlC5iZR(_;*up?AXwIm!8LFDgtnKLkW#H;LLJRWdj<m$9&Pg;h6P3?u<7h4Bd<Da
zEnc~->M4&ri_$hX<5_6&@b!4q%Q5>p0fVsHqEeQuWy@1j;&?MMi;^dVLHoWj-WII$
z<(=4?^NVkAQ_wNJ1JTd8q+l#rbHDy+!ZtiZi)<8cuXuI`lv&ZUMyK}e00%V7<#}fL
zsesA$Qn$Ex-IjSOE7@Jh=Sn+Uv;5-!>ppMS5bu&4qiw>a{HY5lXX^dOCU-muR+n6*
z&Q77CtfL;69s`+EM4l>6fLwP&Hhm_xch47i#L-dl1rM^VB0a|2?>sQlCi2^I9=$Lk
zMdnp@X({O+OsJZS$`3vgq%>$#b00(qw3y0(9-{<chCduJTsdretEmN>3m8m#1v*>?
z^|&t9=g;4iX13IDnfMbp2$RG$+E}|vX4y{LK(X~S!_=xq8ToE+XeDY<m`O%t7GN+N
zCFU+1+Chw=v{@FYl7T2rmjc4&8Z?zpn60OI{KSamX<AOPWg5AdwgVTor|>gCSd^7Z
zBqU{UnEv+h9j9Ss9o6IxE->7mr(a&*){ZYFq#O2Xd-;|W&BxP!FkbF+fjsR{O*&M=
zadeJFOmoX?kuto!mlpy$Ne=-mE=XeBi?h*_fD;y??WFuF@ZaK6Nw@RD-FGU@-lp&H
zz09aM<aDr&TCl-&e0ixms)1=Dy5*93JUS7;#5%h#h@;IfhB!Ig#nMv&%1)VunXADy
zL{_1*kn&ZWjs#FU-|vr!8nu}@+O)|nBwng!5Z}(USkzKF%0(!=L0M$2td<_W>!_Z3
z8M^-@F1XX;h0+9J5IlC=m40ATeuZ9MelW+7*uoSI=o@<KaMmG(TPa$>@`xlbNRUt~
z!+Qusr7>3o`_|Gm$0lYi7h0>Qa#tYNsU!`JZ8NuPPp?`^|1KOYqHM`(=ITA&YNU|7
zVP^q>a{UW33ehCoSc#GLb`r*9Z{xw1LhYfR7zOG`jbRD0LJce?m=uheZ6iS{MNM^r
z%T+sCZffRBH07HNv;h0!Mn$9p4lYb>BKEOSH^1c@#Pbj7LVA7wb#@eU#r#m5ti=_o
z|K)1NY%(64KXtCCzG?`%k-6}0xGenTk5%c%V>})_xPp59vJD%lJ+5>*8N|tI0o8=1
z({%s!W`AXduw@pQfQ`WvR3Xv1&wJ<-RUPfd4QC+(r+bJ{E=#T76b#at8H5&SLjoq-
zjKZWCs+7v&Yv=TWMu_+%A{7>^7botaO}u=Gta>u4iuDTsvib4=<rfhAE55mSRht%V
zSBL@NDxSg@6Y4=5Qax*IU>ocsnGY#UiBBlma%x2uEfVs-Sdd2t+mParBbaKJtF@%p
zcU;*r7rN1liz`KOQ_;!TWh`3k9v>g>hGU=}sw8cYk=GXms!q7zfP_r2ObC%|+LVGB
zNVnlb9}*`(<ZvZDWL&a8&gJTN*@iF2;O(U!%51h)$V6_-Z$}7Q>y*V>wt0*<Hq+?7
z=bUe{F&&qKC-W(fjgQ?6glo!tSqBmTO9r96jUdQ-4*e1fx!1n|`IP#?xp~mzL&*;%
z<o0OM*cYp>Ipg+e!U8%cb0KQl^-38et<?)gg$wW0Sc@Ntbck+jrzs?90#-X18gB$D
z@f=PXzv#_$Y>Dp?=uiFkZ;T%HU;5#n&C7X%ior5U#t$vWB?k(?D0}9sck7eS3SS;k
zatiAGfY@cC_;rAdfoF8r8wg%7YJ%AeA}_#Cj@1e1OIG8PlbaZD(di*KZOP7l$UKkU
z=9Z#Q&(1F*t($RZK&`@sFs4G|%_P$j!B)4ZuQJ`!fZKp)e+k-$oYH%r%OhWc&Gy9H
zdPuM@RG>N7z<rwTtAyb=W5l;n$<m#c^=2pBsQ4hlfvkMvU7>JVeDkQZwVx%AH|?PG
z{XL0)KuNk*?PXNa!)Z#=4P2nW1^4iT8&0p-;5AQsWZ{st#R=L2FCsJHBG;JpVx=nF
zs`a7JCexa&*u)%(#ogxL!kyE$?vyUjCp$D9X9BXGC=VdU?wuy|DtGAsm{twfh9Pm)
zl=Id156P@lsOp_o=jzSQexL+LVzMbL3uolAuw>B(m_hNWRzTA?H$u$5u<1hPh;sR{
zE52UFKMx>DZaPi{upYd#X-s&y(lybFhH6wyqLP4DbS*;JTp+kNq~fCAx^3f&5k4Im
zP#OYFLu>28mcPecY)*iV4Q@H!TWV!-@`6z?opcE+lJVPV$n~z#ZO(WF>g==oT`mh|
zx4tzbFjc%GL{57A@sLX!2=$$<r4tj^1mVa5x!>dizJ2FX7jbj`eoZTmRM%hLJx=Uy
ziu|OBY2YDTSo;0vOZMgL$5%7oAHh(4_7&W=aF3)>gR9kNX%`YC#~oI8Q?H*1>M$s3
zVxm|-u{Th~m-r?ZzD~m-pLF!Q8^}0=eJ<!1Ak7!IMufLp@LH084*R&-t))tV12rqN
zngqGuVuQI>{cQbTSgLTj`{}Qd=N`qhyRukVQZP9TtvE<zD_araF%AIu$7^wJaO2CO
z6<S80$M6cigV=W3Gc#=_Mm|Zee-MNVTN~XC_qITE9;5{D$^pk4E0Cdrb8l9%5xb2B
zbUi={@yDV>0oNcFYoB*C@A2Lqd42wrtIG{;iTQV8kf5HK_3Fk5_Vt@2_4HXDsjZ*H
zCPROZ%(FId*vC9Cdr%J4>qw)=JZJ>sT{{MWxmMajoVE>DtNJmS%_ZvUq463ja|dr0
zNxc+raTGM?Y!|NNM$<!_+pj!&PlSP9YaUoqFtyp_!xb`}ld#9<_#w7x`9_0AOMHB5
zytL%!ct@_gui2kja2H09_Pw9mmCu>h%o@HrjTc0fkV5g}Wy5;V+};MZvrgE724x2e
zcgS#uOww~;LU`K)Kh9K2Zy0PnJ4pafQ+ue3hcCJS|E}OiCmoHvg8t%%>g5xy&zKPc
zsKh0XLTt|u31*#|&xrFp2cnDub!<2;WnOAFKZ}uOJ&d7s$v3S*RV_G!vsB<WoP1&N
z>*DQ)-+D^PkLEti3&An{PwJ@Jow!g9M_E%4KI$<OwmRcN<m%72*K<I)e1ZN<#(84l
z?*M8BDEon_;Y@9;9Kgp09*~W{rznG)fA0EbKdlF+Z3aCRRHJZ7GUbW}Qpkw?U}>8c
z2}3BxtL%_h*#ew%hqJ%Vg%N2ju5spHndCFuD5@i!`Lzj#{kWd5`;74l8uYq;(sHVo
z7tE4_B%2lM9(d{=2<jfZp0!l?cww3705B?utcr@IfKUT6z4SHPvJl+i$mPe-8!BBj
zRO^iG!Eap%Yz@@xvB`hGr7W@h=A83oF&wV{;T=>V*DQUfS)ik!JrYQ*dZX&lHJeqx
z={w+;$O|P8HFr0Di{p&oqGx`xOj+>A&y_vXrW?KDm&@#-02|O3)&&_pXQAUzQX?A7
zS@$7w6o@cuU5JgkbV;17<pYk6X=O4DOkbq=IFH%q`D&{ap~OdeNJC=t432e$&eJ)0
z>DN-3!t12uW@z5NF{Lb?JeYI4SQs|srU1VHlNh&b)5pB@q|O)9!rV2~Hf68#B^KRq
z0Xr`|WJnPD68OMe_N%5L{aW_g85KPTDTz5}WDt7!Ll5G$6uvzR>XXDfvMj6PMU8!)
zSzq_nb#ft8NlFMR$MbqqO~UW;5Wq<(ZJ+uSG=+q-f&qae+9<jX82|HnXu>8XJFW24
zYGJm{M#Ev`97Vr5bq_ORL;`rvTX)RrSS12wcnp}zcJ+YKchuI+v36@-CwQwT-I&xl
zv|tC+I=L&gjsDxtm%1qtN$8#@_c$~adUp7E$6*VP!641_N%Zd?p5Vt7JKw!BlcccF
z+cVYPmRF*Q5B^jmcBG-X2WgE3(h1v?Bd<;$HD~Q0-YUqzo~W`(GjFNB@LjuyIy=rB
zUg+v7nAH-F7$HbJ3Z$p_`-c#BMY2-rMB34xikRr~Nhw^QI4qtT>x^h<1o%61<a&L^
zg{W#^dzwyo0wY50qKSQ<mU@00<t>vUj}-<O?A(kN3j}c<(=nLRG~K*(sboK!Ysl(f
zZpgjMc*=Y1LU!Un$b(I=-YP!z#cUs}Z=KPJRunirJ$+X`qmoiku$D5}xv<xNe^u7`
z0A^?VCbbWc<%1wcSs(=+XarlNjo3FbtlsKVSD3ClR;f!}d;?Y_voZVBye7+9cD*g@
zl^;AVA8;s28}ihWiRkghG<^;MrEKw#g(%C;u!N-qEWOHeeNvo49bPT22Gfof>-%Y$
z!T^8C?QW&QZJXBG_;b5gJJ5`+NTodA(PVa0du7d#`~>5YQJC_NI?T05j1LkOZ8Kh}
zZf-QXtNPwDt+;xwTB`@pY>xHvOvMBX;{v1@n?#CtDbAw67*5-7SS{~yBA@<G8+^KO
zDeZI7`5E-ae%tx%Ag^OM@a$vtGa)X%u<2I!cG@8YH!pq|y??Tf4HUSzJ!4N(LrmE8
zHh`vfS>!_HXLz~|6cFhi`<6;m)1LOfB3_PZSP2kcDkswgaXA~zet5q12-@-Xx2yuk
zW1_+0^$CI1fAe*G1M>gm>9dcHkM5T`hF0zu{L`nV)bzb}>F-BZuHUp?%-gJ-IQ#$V
z^NUpWznNOOILW?F&0a(L-yu8p$8P>^1=5USMVsTtNQ=BOK>SyZtLuepMc-23*W?-S
zQ7*<@<*i%7$<Z#)_5j@zw8QM{QUusituJJ9lvUPWf{4+3aXc&?QO|lJ4}Upfc9eO2
zWffdhgp-c2pL4>AHh*dGMt9uB1t7Q)OcrJJgZpWNJ<1;Y9s}QbJq0d2Hw-`Jmh3b;
z36?uTjgIRmCA6gL9WiU$Cle!;B-Dyn$O743$vwkFP)fzUJI)D!jXV<8@6F&e?ox-C
zEZC*v{W|xsZUdPx8EmxDth<WjY4R!CJ-|(qRGO0^Ff_rJxYRzRt!7y0K9)qs*ZMo4
z+oKM3divp#=QW%%m<3Lin7p}eeG#AF7>me5!yQ_kkq!*D0wNRO)pI00yO_X&ek!*>
zBu{!SS<cl`Wo}3f-#KTuGBtTOZ+0?u@xnmw;f1RpRGc&*ZKpwoRO~D;ai~MBGNM(8
zrmpporR7x3@Xy(XQP&i+P)A9C2`{>+YKBKJrMJ~npu5J!4V?p4#PuyV{Zt+ZY6;PR
zxKuyscDjaQ@huT12rImrjS>l|c{(=T=*@!n@skDC5YGUd9t}t)$nCRB!2pPNaPG;2
zO;(ZTahqIsvf+VWwQweoaW$JXT2{w!IjVseMGz4Z`TX{Zk)r*uLeb3t2HJwMs_-XK
z&drMBgr1FsM^g_@2R~pBjeG3MCxcJUN(()^^-d-}_d2ll{yqG=iu3|6aO8<~7i{gZ
zJraBpRwe%B2v?yavNi)5k_S&Q+|~At#Xnpg=bbCY^95p|0rl^@a21|&w75wZx}la`
zQpH=zaXOV?Rq+n9NeJJzO6N(dIkL_cgis0(3K%d*dVYGA5-hIrPFusYiBw$?ntKHQ
z^-LG6U(*}wslj<TgS=TvDl@Q-un_AKklnvD38V-_2W=pyaK^FWug6$-LO)E}%&Js2
zjLeM~^xVgKAnF$~FYPHu!B8Wp*K@^}2E1>2UskG{ChR9GVFgCLw1e<oX1S<u+BVw2
zztGsW4EBlqoe%DsVB5T0rT~{XnZ?i(SQ$rMqlz{F!hyPk>!L&OsL^N_SWFkS?T^>p
z(sK|QYsJL?iuPH*h^;HQgBm#3hhD93kOB2*vXDflXQHIssPL%lWB^R#KS^dsj-$kO
zi*fteonlf?4J)w{X_Goz?l%DQwJ8Fe!)8+%u_yfGM;|Znhsd4gM1L)Dfa?8@vB8C)
zEq~iMAU5in9`m%W>8z_mxBoSE)zys|(fAG~?*c^?W|5utt|4lqdPbrcI6t^Ide=9*
z5aJ0(WcwKf-Z_5nWb2u!o9`jbWj!rTN6$gIyF{JW!<<QAhnd$*upgC43QJlx4rRzy
z5ByUH%BMNlvwMvYXwaBQMol<R`7<irsZAIBvo1I6sd10Hfk@H6gZxnBX3Zs|3uPdx
zqV^y2_<vVW|D$6~zb(zH>+HXt_xCG*|M-CQ<J-x8yUABS4enrv%WXsR^Yab9_bx*8
z(2o0pY(<9F-;)Zr-Tqg2*9!i-oW98HOM7p8&T8*V#+McfVm*KtZu=CXkMYsp<ZB`c
zJtOufFE`!5s3fSqy0C!3$yOkl6^Gq%y$!g5;tT8u3Iu@<mREEe*i3~Rezq?%bwN;T
zOsfMBx)B9k3A374N*oeuf+_MAj_ZA-X_7m(uC{>6B|;5%Ct&1$8Ks)Jp4=ddU@{r!
z0&87W-1GOx?X{9qe@qEkLsjz5?U5?VLjv8(lB_^z4b=&kga*Y+bxlAveTa}=afB?S
zwQTx}hyN(7f-;~G0gy^cNPt#<xV&)@0N=}y8@dxnpuZJO>WS{ljg5PuP^#Vc6_pCu
z9+G<P;he}fg{+FuBB^G%x6j@K3z_K9*HTLX^w$89UbaL`zuY&%HKM9e#euF}NQ2B|
zbUT?2$*GKClfw))0lnVgfKnw3Zg3Zh>0`S8y6}n@>o(nFq#Q9}gBrE5zdFO%_}FdL
z&ojF7+E#zj`d>M8rzR*^FVDgt(?7Ux@vb-ZXQ4n=TK(kq=IBn60`|nPNTjtJ>uYpG
zqF@BivXJh7d4i%|-u-m4+^$5TgV)*1BKaG;!vT~sn!w)cX;+~TJ1F~WYiV?Ag{>cJ
zJ#Te+E+9uWv?im#&H)%)SARaU@awfDQc+?2DZl_6c>ojxVk*(IMTU->o0q_90nDAF
zErk!|^w)QbE3sB&9`2A;zP5A#)&~8K&hz=VO|P$VC&q4R%Ki~YqnW(_)Gg-tW^k}G
zStz=*lKI$NPYo^&=EKmi>{y~x5o<x7fS(^_;y=o~`V`lo8;4-R`W7zV$K534&@|&7
zRvl0i<Z`{ihO#8)J$asy7S%B2tU@FJXQ0&og)JMPltP$Mp&DLoI$PT~<8C?6U?RF^
zimz~C4^v5`_bhs!5cDxjr4pK0Zc@4IW*yqf(cD@vptcWn^IG1@+rGaJ548Gh*Kbq?
zBl#A>3vgfCftT$c;RatizT|AAflJ47PAHycYjgnNjba5*f{B6bBhQyr!GXa8BcRRx
zgUC`*5AYF6lNk<(%wd7s12Tv%t7<>2w7*<$iiB0snL6t6KVw37+}MO-b&YpcSOB;i
z1+!!p!SxYQv7%V~2%yp<Nhvy%5+-C@r44p|w?%98bbr<@!o*%~XKnz)*F(>{wk}1d
z;oR0}oN1zP0%9X=1$&OgHgS*sh(}Utq+a}Y!C!_so^`-6DeTgLw(GZW=7v?*E(<}3
zC;j$27?JQe0HNDzh(N6j$LgELd_JJXY4ck#h0vqESEd5*YVsaPEF>5Buj0q-Pb~>T
z>sv-uKBSQMorTKeK9JA^g8LaDZhhGSfc(CVk9Y8X5B!?aKoa0NeV3AOThfBEl>ebW
zW|l}aIlDXC>do>rCFa-p>~4$CUYVNqdE5D_?GM@kEqlz=^)-Rjx}|3*vECBRmPifm
zz~B+*bawp~?yLVdXwhJQ$Me7H_SunmvEFgZ`B&k=V6~}*MM%eu=3f^7gNnPlqnq=N
z-wGW)ocHUTPEJk%DBsJF&KDQ7!B(RuxBCI}txRO~l1D@ZT`cp5LdoM>Ttr$&9*NUb
z-;2$q<Y*e+$<hy&m|;Je<>gzi-E50%-8E>LiC2`Btn~&H*qmxcDHv1>_baE|2g1Ye
z!b0_Mp=(2bcqho+^or$wZNu)~0{+C1LI{h9F(M^>Kc~UlsO1d05eDHDgiC^XM{a{l
z(qB{K=hnz_wJUc9Q`e)68Hx$u7_}sKfFwg(4&=X{@BIW@O4Au@VqeE3QI61tSNHI`
z&X@1}0&$_i#g(CFLkV$G87Hg4hMRrWyPpNUeUmjS&MWtriojjs2|xXD$E-s8_{Wr)
zq=tI!{`;F8Cb&H|j$1A}xB8!{*b0$U{Cu?fq6x12SYK;^o${rk513(nr<1+fyN|rH
z`dr2ep<1)GbA6Ip`Om5|(`ZVa^ls%juBnO!C9?Chxfnr+R`JUX>|iu_lZL%o7u@Z)
zq_WF<@gofB2rYM^{9sqR<BblyJGqYESq65!iL>&#EoI$1`626UC|2d|MbWeCDvRJQ
zHwXks`_+yzRFlxoD)aWO|A_(HX1;8s0-;a+{Ih+#ba*V6HvPXUJ+>@RutoYS=JAJ8
zz9#?ro*KohIqQ?rxSi`sHb!PZu24Z>uLhiM*h4^?c?%=9Q1n2q+;<@VexM!f@^}4`
z`|r{RGJ%4Bq8ZbggH>%I|Lq`X+WE@GA6GzNm;bwmaz#D**@kCr-_+Vg<-)<Yktg0c
z!8bDx63t?wbJdc#Fs0nn%@M?L+ml3lLJXd;6U$#`9+Z!`j-PCn7Zr2;qI8V+dj%+K
zvs&-^NWl~-CZ^wS)hgAp!Yp=%*FR55iN|h|^7*bVD$7<Q3hcOSQy_YQY@GY$d8_Z#
z(FO@1L<nu!c}G`WHO2~Ht2L%@#pT!L_$SIO=d+wEh<#Y_QcPog`opGY3eSlY(#v@k
zh#H6rheB)2x_XpTgD_!ent%S+DrhfAXf^<(%PoZpS&HJ-;>|iGT=Xm~im>i+dRD*L
z+F8rNhu(>sKp!dH{vWAObe=a31u?r!q0zwatsJi3&j;fC9S+a7*D`8gkp<T86YFMF
z9SmYRB;1c_{VI+yeOeZ{=61aRlE^Rg?6ESUz*Yh<Q`<(pxm7)vseHCx>J;7EM!~9L
zQ*k)I;R#jYJmK;V8X%Qj{gLm78S*{Ww<J&%fix)~O4?S@pdgeIg2Xs!#O+Ad<g6MJ
zpZqqDuVn)Hg_P`jC@J3^<$as>r@SEI%KLFlg>Wi91Da+tf5qXsY(#jrmsjTBZ3T?T
zWXC(UDInDWh{g5uvC@+T_r68ZWYh%OhC`VdFCH)Ooy0edoe2NnhHYBXg#?}#jFd&8
z>rjPM7YpdnwK~-ec_bctP)-l)1cJ(qqqosJzI8LcQFCc=Tj75z1YCpm*=*Nie{YBW
zPZFd1Fr|RGeE-y@<HF{4aqMn!tIvOp#P8F!&#34%<KN8|zZepIk54gvzjJR-Iscxz
z-(J$m$s#sSJCcX!-@JKq^oLd{FhS<kU!E>$q`I5Xvn;#H$QlK=?5{=VOmqM}RaS|?
z9!aA|I^tA|XP8g_Kza>r1ML%t_6laklNN#NpX~j7))p)UAM8B?kdNqK6>eg%okQJG
z^<2`oav75jhQegZ(O1MqgJ(K+Y0<Vxeuy<>m2!B_@x=aH02r@964I;ptEZf|j^ZNj
zx9*$9SVn<|*#@ffJui9S#ma4E8xtX5<bm`z#3b8{4rx_m=WJ5eX}(}YoCk*Bb<>wk
zKi=qxJp*8xP`y~}zr+KO1FvF)I0Ie}c<f}Q*=Lc8x|A19%RiK*bgI;xpZov4Pr*Qy
zD13EC@-^|t#$WP`P7tC&6GPi`64KAtX8Xp~vzDx@epU$@Fjy7`6tmF5<)oyNNQ8@*
z5L53?8SL#KjyQAU8?9Xvbg4j~uA&ZHk93Sb6CV7zvyz2%DvfPV2;{9)0)2(g`9OwM
zEXd$d)DwfL1CIyz48lgI^2R&bCN{0d@=IBOS&nTZV<dV6!>t7;9o>xnRg%~gt*m6n
z`Sw~Qz=zf(kj=BfN>h-hlOV)u7^F^2ak=IQTU3WzwGeaa?jD13ADsWs(2tQzRLoKZ
z1EVq5Id7LDBqo90!f%k#9#;_lgA(ck7Mc*5?KA@?e8VWHuhHcv&z0?ui`<p2SM@_G
zesOBilO1K<SYnZq$bPx3Vr_HE?_^=V9<X@bRL7vA)9_lmwQ=b_yAAtl#SQOI)a#i1
zA?o}m15(iy3NYpj4*K+lQmnM3&JP{n%@mt-VD<XRr-athjLJ`nOP7{^<;8w}81or9
z=VI-_O+=0bq5|{xk8j;VqHVaad8ApnE9Jz9oFTYj5vuQ|(&VS`OXY`V(2@zZOySYl
z;k}jLSb}lWZG4l()Yg}jh*e`V*o~9oBo~5!!m>+ap}b~{9dJA@_}s-2@?$ei^XIdB
zTC)XyVK?*E3Vr6}1+Je>JHfdHDm-M!@|J5V2Ii&o!=uddV>LOVdFwZ}wP@J9*t|e}
ze&Rv!MV+Xe{`cd)|Fpa4{_95jXtRqW{20j<dX^!Vf?qgkwcm(-H!q3$Sxu9>Y>xvq
zA~I|BZQXPrHr`=LDqrst+<*<z#hOONWmk*hb}$c)j(<Lxcf%>irM$R-0=r&1RY+8H
zRGAG}Km5I<8kj6Vx822N(PmXOQJ?yR+5QE)Kw8A-FB{|_ex7zPHMq;h#f3^wlOz2J
zP!is}7rFx38Yi^t2f6gh<}b})6ZJTyLIz%BW=!fCf0)H|D??S;^wtWqmkd+;Yrmn$
z{<Sp)Mp){=bLs-!@JW5zU2o;YqkPqWPL0<qqJjYIALoJF6?VnJv4O97Hme>6RPDWd
z(257@b7nb&R=Ol~OeD>S87s?XU4|K|fI5+VcIo{%z4l{N^oU35@gM*_l6W{-CTUIZ
z6A7i1ViR11#gL2oAc18Tur((q`@Vh`iK66GXLz~~J$!tIFE04AuZJzJc4S1ou225{
z68$rF^}A4X1SR_S+J6J-zlLqF+Wq35EBo&)>Tg|IXU9GAuaN(S{cV)}pQ}G@q3+7b
zpg23?MlF>qgqfDt+fwz#ic2}r1u!{6GyNp4V0xPKouC<XS4wiV4*sGAOt1#?zhh9`
z@Rw3UK-^}A_9y-$5D8R^KPY0?d{B~9aBN9%YPg9lk_+G<HJIFB04r5;gqr-&1W@Dm
z`f11bx`<plXy+P3>YF*RR5!EO=^35d%$#k?1}d$5(+#gFkZeHD6eOIA&i9<TfGKZ8
z#CGl%?vGd+uFWUNX<$ZNEsB~kYd;nytdU9BA}tQjHeGyy#+DQ52OS*?M=(yMl)MF9
zliSl8L|^<nrT(N*QSbc}PSLcl4*)-F@j6YoVX3nQvMizxBs`TaQAw{T6k@jlvp*j_
z3a(V3$&Ts<#B9y7+xT=iq{4#-Q!q{Dsw%G;P&8Xj2hU6ldb@MY*m`_m>klgMghBep
zKqz0FRy!;Mm`0HyfT4H@;2R1FWy<yQ^BEYL=xCFFyD@=mVU<xo-|cuEH4=k5#1+-F
zwc~T!`w=K36eYqwH_yB83ZPrK3KPoWtcm(={6n=9qZPnIyJnU{oQ$mqZvpTDz$I31
zgr19nHS-vt44zD=QLtaP?b&V=6rHBK_x=_JT+jFAFTgBHZ`Zd8(l_j1jLL;$<(o4v
z1Bv2ZBgopgfrg0wIVJA6^^UpP91-z9b^sN_nA}_|P4a0T{R+3>nhw%Nb0Lo&q~U%n
zd`%7nm`T1?g=Zq|H-9^@zxyw|{bheY^_tr??9*edA7q6JMz?%anb9WiHHbX*5t=!i
z`t}PxmJLH*I>(Y2HQ%;B2joY<X$K;PcIv!Q#vj#)9#SeiREaeu=&PTpZk#`xJrrDD
zWM`ePf9ciW*|W=IjxTo;`of|owNYZBKIO@Jn(_|vuYQ8@`=+*KL;LaXe0&4tS;62N
zC;qzmDCX|T)VL*F2z<iQefj%jgh7_k7GGT`L!c2L+X((Hc8e*rU1JJ1O6*jr0Z_*1
zp|U6j9ln)meh83u19kfeghmky0{tW}hRoYYMhV6+cA={$3Qan$R?j(&!-Bqv+-9*q
z*N_X=MwRUsJEc~t1Z<|i>4GdC2mwxCPQUP#85%~-M+^%QtEer6?uV2s%-lr605FE0
z(3zgIJM4XU`)STv#}*l!NC8Y<hc2_WLq}v5^nQplk9A<X`O%|nB`Bmr<w&;jR^PEU
z*0|D4!PYc|m@PM<G<u77A3`ppNmfr)KR@{yY>TeMN$lHr-|n6gOC)%diN-`fyph00
zavNR1;v)Q5LQO#Lsn~}9)gCvy4(j^gnPZYOPt4QFM|zsOf&rF)csW0S;v`LI_Q?*5
z<sfy0p@s{!@!G4^x4J8a$6RIbns(bG4uKL(*9%DofYD+uGvm4QC^u18t!cxNQ@x_)
zEc%Pq`EF&WN%7$@MrTZG2i(AHFTZ~Kwl~K34B%-r7nm>|Bme$7)1`teBSnt8A06w=
z7~Bu}r;hnu)c73O82mx{jdmRDUV9LSGYpg5cp~|%-hiDx_~!I7yt^|v<iwhPj303G
z1yyA`L9jh}B1j>5b$%0xllxsqwyCRda<L3&*Ht45@=GNk2@r>%xj;K-T*;%AU|kc!
z#$*<ju6o;PLf{)WqcfS%IVY`H`S1CFFk#!``6n8oWuY_^ifYdlQhuvhA+a>InS$u3
zq&xdx1-hilpiViVTsrU0K0-)Fkmh|x6-VQailZ7ibKnj{HRu2>OFLzExiNlHb{WAp
zQwc#mx+|+u5IppQ)lNIQ^F;UiWOu;`YYP<^de-{;CQWfp_wA{jFNZ-DI(hXP14Lc@
z4^d|s7UdUqdka)bx{>aoI~3`bj-hJ^0VSmr1f;tgqz0)01{fNoQ@T3^r9o2S?Em|o
zbDeK~ur6kv{oMOrYyDQQ_X4egi|_WZ2$=Eyt3}!Es||ILV{d;{CdBySuN%>ZJX&Y4
z;&VveUt|QgZK@k3Ui0A@#h%)uR&VYtGd8LO@cjWJ7x-WWy`avmDxt2<21;_@zi7me
z;DNLGI|ig96UxA>W2_N|%mlO@D2;vwIC#M^6468@8a=$EW&ZtD-PAe=I+fKr4Da&H
zqluq+;_+Z=>DH$D62l9;!r;Z?Ez1R4q?r|<Q-}Ee=z`kWe&=|)KmIJad&x5A#mdgq
zjNt0ITmndXpG=U~7#v?cB_Kotb|#Kq8YPz=^I<jdC7b2}uM4ZUy|rl}Jg~wOzL5&o
zi^bkC0npTEP5vyqSIs4Wf4lmH%>y^eC}ED}nadf?)#>0~>xHeWEF3)_&ZbxW^>-WJ
zKriW1=iTDUzt19X^wBe@B6jVu&#zyqYT2X_QgTU+=QX+&z#H7Q)*&HE=1ume>q$6u
zLO^k|%8I^uqN<^;9g0up@$prHPUgw=T|j4pSun`Gq_aV*VXVhaDAv?8OiO4*tD26^
ztbeI6$rDu*-iojX5iI%N@ff@`rjZo6kH3dYbqdyt*dDFF&&aiLZdIDgLGQ8yUOW~I
z=tFERSq&YCYOeV%zheTpgu}^7E8pnkZlu;wVh_fS?Y^CnoiX#Htdxj4vX4p?*56%e
zvjkW&V4JAc2{5g;<S>TeTS4*?`72UGcwbWfV-1G8IR4zY`YD)QXig!DZHVzF!K<Ua
zd0K9JtMFw73p)~70C`9ljGa&<m;ur&U_F&sc=YQary>=N>{MUW4i5L%<T-_{<4&vp
z+3@O0u=}eT!VR?j-T#vM^U={TCDh^5j|<?BwwJ3$orphc@L(d~9WyfkilGpWwKyeO
ztz329ceax}I#!GlQ4?HQ+&93%10v%}Go@nv0r>w;?lPf34|_@<U$ZHnL*~LC73vyy
z^DIU$9UZ_GV^`WFa$yvy=^g#NOh1*e(8LF=^69_l7fq)dnSMBG2qEYx1)Qjbn?-iR
zPk*vpJ0ROdIVRgPE23*_eKvU8Wde}9^$?G%af0^kPh0(wD0=4{4*6alH-6c*`O&R|
z;uy<t#UqELQD=S-`v&(!Y-Q#FrI(SihI7s;U?#E1K)>&mINkd<|MbpA@4?+d<d;Ai
zGkg1pwkx4~ru*N1*Ey|+Iae(vT|zid9=p1h(^9nwZva4z8`+q<YWAE%)f{q{0psDP
zci}E+3IJ20gEkZLGCJ28NoYN8Fi5Awuu-*54i3xBdUQ#>(Biz6N_8TJKwc7#NVqOP
z<Tp1ltiMm~w-*z*sB)eKkT|<e&~)=^ALK;b`sO=UPa_+aJw!|xHevlq=@A!My1I!-
z#UZ+7G1KGKPmI43$dizBHB}tg#&J~&sU%>b4k!@f)ngrkTTjRf16qK1^qe5950f`6
z+2F$*z!+H+ZE%+ee3gvfASGA(fU3k!YASl<`I-0lYJM!7g~QXisT%;(ex8{Jvakzh
z%iavPxZ=QRfq9e5v5u8ZI8*kCVt7f3ZFrYDhh=*`zav=%fEr~c)=s~TH|mh-Q2<jj
zu;9q1{wk}3`0f?cQqS)4p7BL*A8)P9eb?ZGL93Rk!cODIc-KBL=EnkN4zE}J+uKl|
z{7_ipI8xPQ(a!br?CSI=J*N`5k9op-df(%CzON2~hY9{36aImSg1qR#zd&K}1!#+g
z-3D7_q9~~(IFD3n<?uyr^m4o+l^FNRn>Wpf!`opy$6LLwfR1<z|JkKhFY?z@Jc?x?
z0O(3<BIKA7u>5_eLkGnL0WZ6vK@hb)As?v*ONA!E5r>+&oS?r(2@?1I>^FgG5}PYN
zXBhXLdN}`SXI?C{dyZP;60NfEzf^i7Q=n9yx@CG}UOhY(Ok@sILsf^ppYH>kTW4lY
zF7Bp$C`a-QNRq#?v!RN^od<b-PvO5X*4Xi6GS<tx1w24I6L9N-n+KP)%S|L}W6P)J
zDw@O*ub0GV^;e;=qa-_Ve;)UT3$yEg5TQ%!!_fZfVKWI`cV==jQzmhoaW#|@OGTSw
zyin7kWoJvS`UcB`Fmovp#~mqdODY*?f5)hKm0&JCL_ULVLM6u9wy>1IXep6vPF4?>
zGU*3JJUSrDRU$Sg33+>+SE5jjp-RxY%q~Wi9^F1?#PlGYE$*9!3G$g%a+U2TLlifz
zyN+-<_;!CDDT&humWb5egfC*u9S>@;l8OY$7fn8ZPtZtNwAuwXp?)kubnsPKLV|8>
z53h=AII_T;%!tZcylL60XggjHpWAGg9F+@PBf`dVDcFEL+;s=Bdig7*a+}$`uU!aH
zK3_BDjE$vgvcLycy-WMZIrmDp{ftd%-}s0yn^k1^*GX=Jo~{OQwVqZh=BMuJl3^}s
zkO^0n?4|d|{_Ms*8UqqO8>zvJOff<@#P!J??GU=3Y(~;+?ZP0k`%gHSROW6gix2Ds
z-_QKAeVIS`Td(Lev=-I*)WZiWoe6}EfTV)dp0zc^5|1G9n1^fxoO<<1`0aj0*c@R}
zFKz5w=782@ACMO2Co#TK6MwaO5H@m5w|2D;U^fLD@2sHR8+M^mj69hfA+8eUHQ}=d
zjW>lc`^WaCuN#L(mk;J|rX-Fh|2M(?SnokV>KhZEeS!bc^uMIO`RL(s#qPINaYLa>
zbH77<OMka<e<|^6XD8IR?Y`A-nVJdn(w4KR^;>X`355~vxGUWM+Y3>ILqP<~_F{~)
zo4@tJ6i#J=>`*@);^$KgQ!#zqEs>{McD-W3(J&S)3RI*wHh?s0>EZ|0V8IAQ9F$ZH
zdwBpLhjjX?B|;&fkJ+~agi7n~-Hr2#?QGdRlO;8D^#v+VDH4=9zn_%~Gb<G6LuKNB
zUpi58V=|^$m0E(v`Mk0M*shz=*q+fdIM<INXQuaC!0Y6dIzuk>ASj#ppdJud3=%1a
zttLx~AAMcgkFeaYzUXbf0|=%>@r`6lAl-*h`*?@K2W+kVh))j2;>Xz5l9q5r9P79q
zPowduYZ97el&Jrk2slH5#w^5{YIkAa9M6=2)+SiH@|Z9R2M<i;_SX8#FYo|U0q9V)
zodYT$zv`Mp+P6}S1*!}$hHtyQ8$>pxiOnpj!DNz|1Po*`%y=5<)l2SId|Zw$KIQ$q
z!@%t~)RAZ-Wccai{7W*u>=oMp(0*C}@r9X2@Zz;@7dj13rxSziV>EN|6ImVnA~X;m
z*ft<-1o81~*TW&3Qp;+I^Dq}Oyw8{3Ibg^y8dONtv6Ji-2t(or#rwH?sLT*)YxMG@
zqQ~Q(B&b#J0S!(#b=+0;w`Sr|QMTd67=w%!;W_%tkq>oO+oob`TUmk?ha+TY@gE$%
zo9%qlsr07Cqp1rKv?~U`ePwkzU1oO4<ip9y+wIR}Z|=DNU-aAbUv0%Yvj`#=Z;Mgw
zPjL|9i_PzPfy%k>Yky&kvQeo+l>9ofvbikE%VICUljsgf|EBAWjVV7f>bBz(Pw3XD
z@GjX*Rx{?!NHP!Aj9F83c4T!xS!}dcuYm0=c?5S;q}yI2ah;GjppXY6^6X07qH!VW
zu^6tLN|+#k-;Z(5yasdmYH61U3qzC$Co6men<;}qntml5ctDQbJbu1in!B7!`PA@0
zd`W?C+q)rp)<tBq)hM@JQ@n0o3C61#pq+ywnZH;a5Pikd!XvGkm7qPm=*A!Bz<?6W
z5n)2rM$$#*q4ufu$g`tQ%_%a%$i3g(Lq&4g!Gl5V^^lTW=~q~$aYs>+WCz&)kAXWY
zJdu&#VT@RiG7G70D=G@+KR1U2=UDC+S~cVz$Viu$k$RaBd;qJGKTTDP_05Ru;^BW^
z=7mMUgss(E?f$WebHo%d4R=6Bm|qsI0|s*qw&;T;#83VqV0!5v)U>!CkIeTPYUj85
z!`uaZs~bMeuL9%0ZwJ`E_jT&(!kO~oWFcq&Si#?G(mcMNMUYea>V<Sduxjk)9ySH=
zS2lmHS@q=f8iqlShpZ`#_4MULl5@i)kpT;}Y1_lcnH#VG902kK8I;~CAjCyJ-DaPU
zRxD1;<iH0DQ@{oT3>}o^YC1ZqkmrTwb$0hLT^TeH!#vYGEug1C;D|JYOfnv10Is<1
zI-dUDWX1o}_k&C?(eBzME}#BqExH-!y2Zc0q`tR2SZwILz800ZKb05(yCc8d54YtX
zt{c@ptrIwW_;9*%S9iaE@%QwOTf-@D33vPxR3`<RL}H7tMbfiy$Ivz2{RUu3XKkUB
zTox~*A3dMY)j7y4_v9x|2rVqd`jc_<OTC8|3U$8!#ArDtEUiO?nECKa*VrgZ_j$!k
zxot1uk;{TW=y)b9v3kK3Ow=iHir|^E3k!Av+1PNU`VUo+VqB^6*vdtZqCsj*B(+N7
zdGVK@$*S>0Lv)H#Uhi6DE6hics?+wdwIys%oO?hN$wLkJhyvi{W91bw#ib6lpq&2g
zE8@8KIq(5=ag*maq5~Mf>NT)Y72~Kdx%_d)de*J}_0gXI9tI`G@1Kdbfl3fnX!q&N
ziSyA1X?j)+q$vGlb=TZ9%FQ~?4Hu~?YEDFyoJ+v)=mhBS$+^y-S1XrJxdohy-Hjs5
zPj@xceCdL+^L~IT3r*(G$lI^-8?i9R+X0sT{?hO`E|K(#1Jsr2Z67wJ^0voe6lsMf
zxQa-)RF6z#a3G~f&49*h>ML$T&WQC&tEa*XK!?i&7{~3Kb}J)Ye@Q9@MTRh(DCWZ`
z_GX4U??+NzczG3dFZ8kxXx^QJ?=QE5{>^k2pKr6EYdyUE{m2?njjOrM{pf}U7`SW~
zM8Q(hcwuwO{*{TJIDP5nn&=Ds^yq%jYv2jk$3OTkH}~Cr?_2PMrpW=O99;fx)DAN>
zbVgCy@p%O-UZ&4dUl==9g)94>%whkhR_T*>!f#v8{QOe?-97dCTv4&bb-N<b`<qTm
ztD6BBF#xuPK*Dmfmr(Y3xJ-?JqiT&|ElOlqNktW}IouvZApo6JN{bLI-*}&EQc9QC
z8e@b%LLsV_QC_}Hz};uUD3T0!;>Wz&%V$01F2w|gQ__u#M1TJvhL+Zt_QSG>D^B?t
znS{DI)SPWZevP9tdWDI7DTy<9^*HZZG4c|~`bWV7a|g}qAzQcNrS5hpTCYV^F1Ldy
zOZaBR1qGPc;EzFVcS2B?wkUi+ZZMC<96xk1uY@gZa7fufdel(Qn{x*JWoK}ArFkwJ
zMg+^&9m{I_RX7cY?Oab`b&V9iuNTw>70N}xk7bOIotU=OANRMPM+f?P>En=KMwu^|
znbCUNooc!~|Fqh#Lm6ZwwYn5i3W%`fi#B$?xasN0%3X38(*Fcbg;SHL!7$B!8(;qC
zt;J&BX~Y+5;X%2j!s#8~ww&zC+g&qM1)DfgxyWa(=S2@eWs<FaSzVcOO=@nboACzz
zk6WirSK$Z|!{>z8+a}BFfII|%cmG}5pT1zyoZKY`d#LVmL6GvBdun-t+^n_#yew)E
zn?ls|Dg~MvzrAifqAZ+x%?}9#j&55o;urD{b)e(NDr=07>`ad4YhNelx*j?_*Xh<f
z-!7|jc<t6QxFp(A2w~4Q`M`sCxTfbLxXu2%l6pw`+#3gb;fET=d#qRIMQgC|Y1=fX
z4}{1u-mAm=0p|a89(@?$`1erXZ;k5)O)ej=?J}aR@Aq$3uC^QAwquI6_Up#csvL2z
zfD&KV$%&;VkE`32`_He>iq$@^__fueQ|pyTLE9_MT`j-+5u_Xruf-z+N+24eP<!Zq
zCZmkQ%qfZh*43?i(5dKT-_rk$Hmf-+TYaB8`jhYJ5+AWfaG~AaU&`0M)iI8?5IVdY
z3G$~hXT=vuRCu<gR%bxjUrQ%9o>?^`pSINVuYO%SmS?^Tmde~e?F`Vf8C66UVtMvA
z!bXY?qdh^mf^K+ca$`)Ef#E*Iqsak~8Zd&fo&D{?JNOm;Bi#Wb<5xvm?mv=XEeB44
zWXDPkOIl>!rWNOPZbaEwT{JOp7BVYg_DCuv5`%0?3c{^kFfpK_jAjHAY@=35ML{P}
zW2h5H77}t5_>Vvo6%EM1`6n;--JDdya@k75|BE)dlMO1ajLz*hl+~$NOAhP3ur6DV
zT*@9)do^K!2RqxD+yb<s&xgCgsXCd?7j8i%EPH3hJ-!1Un%z4X2rc_&qKStBv5d8V
z&80d$;q0)vzMt#U9^1c!P`~48Ru~%f;_gwA>8uqVfNaWr+P9^ZHG+ub+N*lK>GD$S
z6uj&jj%{h}h2T?d`C{X?#X7QuckE-YnSA#0>(QckdEZ^MywUIz-Q(?Jx7u|pvg(Bo
zC-g%9ZX_g{MryMM1MT{B7p%eR=HNIUUJjtk?0|m(%bY-AzfHztV*|TJ7?E;^a)mh{
zkSocqC1nW$i=~zhxTuL4hbYFLSQ>|L?JMhA>$qZI8`H8At)DdI$m@WSq!AHq|16?>
zKNQD<yugie<8#B8d*KQYM*(-)U%BF|{6sIKqyuW+6vnrxT!}t?JHs!SzHaq2h(=Y!
zxFF?wvES7JnfKD?)`n@b)${^bkXGP^p7Zc+by7jk12JPBCjlc7?OO-sp5VhQ4zFEL
zXwh>4)_0qcCEs_*QyZ0_$wQq+kyySR8NJHD`-2|ZpJ%6>VqMO9oF1K9wG?-=$B^>7
zLyTD&K*z5~Ep1jo<wR3bh`rrc#r83X6*)YUa=5r8-n`QqFZ|cS<k#x!+!jvNUYbzX
z{{aQzIdg{mOr{roW2^;k@hSIB9*q`7yQ8O&_~sv@{N$(}<tF2F_;8<^uwgf8iBg;S
zI5J8!tDv8Ms2-GXtO<YnW<^x|je$iQeChI`KPaAx%Iy;_JBXzO73cf{Vy3bfkhN%j
zf{95sb?nq?q>X#0#*s=#3;LDypXC^CiuO47P9xC1<VsV=*tV|9nNGQF2pNErMb3Ai
z9kRYkQUA5>Bz%nR7vGDX6zFezjy2Vf_Ve9i;*FUmHO2{EsjJK)_8dTkS;!MH=>f_U
zNQw@Nvylp7-OXh%-!za2U1f!6^$7g+B8LNPTRe`b&hhkEzzgHuleB*X=gj`%9s}We
zh>!ZgB9fWh&oy7)d-gm<eEcoS^z;;$j~(0=l=a34?Zlfu75=n2h!IW?7WkRJ?iP1a
zWV#5j{vh0V2vqg>sN^7fmS&&Ej%llFjWyg~L>tasl`xlPT<d+hA>mZ!X1BthIbL2F
zYAvqk@~9U#Z$3V_|2}hJARA&AUBZ$QG~V6F()f1QD&l+a#<lY>WMKB4*rL&?LUsI1
z&XB$w*@$CAd+u&}ty$x*g>sAE>J0#{h<6cUr)Z*dqq@Mu5YOH(V$ucdkBwNA)c>9z
zW<x+h1)yIFCnO}60Q|q8=e~DJvZF7B(O-)~XuID<620$fmX?E7qgx&qcygn1pN13i
zWjH>S^*?4?8!mp5Q6$?A`0Qy>=y1+!QB|GRV!rxtJ{%i7JCpHKm3RnPxfW556uK|h
zxOW&3)KAAhYuSKx&CwBim?+CptHEhMyyQ5t?7tM?SA9EEGP;=aAM199_e6_0m#_&$
z51+*cph~OfT!4469fB{n@f)=4Zh}!uZ~^tCj2&a-eXZATrGJ`Q&nrWD&zT>>5y%C1
zAe-HkirHii!!}O<tALwhc3&TBLGWVakviesBKyKk7oR;S=vY<$$EUrT7R7{Oor5Ea
ztEPx^Z<eZwE`~q5$+q+AfGbJGf&1LP7l5E18}Y{I$TohCy7mW*);xBMBu&6(;pklj
zu8t445n}r3wq^X-VB-X|g!?<NPAvBrsrNeoycEotNxKDWW75lJq4>aH_?vhlkOfSD
z{#X5QboScRTO3?+Akvo2N|o4AIiIY79IyA+D2wD{qq@$b$l09joOt#VNJ3=YAMGSm
zc{~Zx;NPYXU`jP_`3-EwV6)Jq51t)8QKR1tfZPCt!dt}3<2z4}2lo~CCkRqH<ca1@
zMe%<Yd|#?<kC$DULv$xfx#s=C0z(4^Bh1l0WCZqK?c_2EyMn%qRX2Bx>R_o#&TQuT
zfqU#(ZbWX0cu93zXN}n`($hL0P{<k<^*Z@CfM#7PyggUWM8T2a9Bcd$ox>P_$Tm_-
z_WY|uZ=qd7YTzUhX1Bs}cy<P$8g|0&Zdv?7iYM_vuLUOeS2JLlk2#~R&&ZfwGVY2!
zP+~=vl38J4?Jr~fQ5-Pqc`*5Cv1=gH1~}M;$k|(M%H|0gbFzq+m-F8wd&fi!yD8sV
z29$l<C4W?KfQe#L0mVAWblcAojLxBYv8u_ma%s^v@gzatG~u*}Vo4om?u6~PCevPx
z>B07FHO3;DA8H^*gWn*9-0#HjPf6e|vzh30hPlhiAci0`EzYk4!?Sj0=44R7V51pS
zjwYs+HObw@&|rG5I^nG%Zop`@;`D7oJBS2jxaVd}DGk=G4#q`WJ1Lsfir9^_kEuWe
zSR<^smq!}}huhAES%GeP#qMa*FurzpzbuWoc$JQjOedlo0tS(RocmQGkU<2e9>CWx
zS9yw`K<!07*;e$5F|%vSinfB$C`4;+P{<I+3u-||E&-~6C45d3V^(%f7j+{>B(m|o
zl2uR-@AwcKkC>IiPDvz?y>vsDhixu>?fl?>FX!(>SvN08Ah99;{)ICL3FbSaa22I>
zbRs{~(y)a8sJ_ovGuoE+IaKR7ml(u5qCUIsH0i>W?xIuqulJ{&ACR5qeLw5-t;EMK
zwRW0}Z1qFdOP}(+B?*%WY+!<O?$tA-O!V*);z2NKr~<Ru8{<Fjf4%N`v^myCl4{?J
z5ORJ_$ffj<iYw3Pt)y(X_rfobJ&!`&R(yV?XmSK%R`+My7l3R0A3X$Nqlg7u&l{w|
z@Zm~UgPQ8X`Wd;_%4-C+I6z0Nr;6GPer4by737#U98ZGSf1ErP&nuC!ac+zZDks`D
zqVNV+dQf2lbR<+GPG>No7x<PolrxDfp-bL6)w?!Y;e|(AlyaDFBUm%LDWqR0&j1AH
zP+%e~-We6ye-dsIwb}5daHk~T(W1l!tl6YVK}QGBf46=|5l3I0S@^xgm;{+b#5v`%
zc=~Cn>FZ4^Jgr2c3DL^isTNMrqp5tx=-lIKE=S}vi0XhJ9s_#q=3JDZ(C*@9ZE_^#
zvKS_~{IvJe5kELHNJ#nEm~t>(4I7%{l|$EB_pWh|`aUft;vNE>nM{{vx*GM{ompAC
zV=>EzRVjh9jTVkL9;5h-p55(taZOucyz7Egx%%-w1bBx2T`+r!e!4q0k{T(+(9!@V
z1}oWq=Jz-t)-@uLK6>hKBTTQNY|>082T;H+mb2^su>>cCqP?z^GPn-8NKsgKy>-Kk
z4RS@CvqZ$l9xKrZ6o-=RmW|2c>aTQp4L0#w+A7V9;ZNBs@7pT=UBtLPy!5*~Y-Q>W
z%pG@3wU7(tJ;*KxO9rYL+^AAVTr09(_hFO`83S|}8U1mu2|XA=V1p*<9fwys%CEyF
zd*aCz9vzp1Yr}{_On<q-bdCQKIe{$Dp;!XFKlc5S9{2^6KV@96$CGdnjE86ZDeF2D
zgyinLAtgV+e6K%&m#3|QCrM{%-#a<VPh_lNtKUdS2d6a1R-q?k`{?uX77(<PA^Bw=
zXOkbra?-LZQ)+GeGI%^PDw3-FNL%;K3g5d%e}>nLvWn>>GR|*}sF$&Ia-}3o+1OO|
zJux&daWc-gNR2Aee3>!FmggGf7Kl>pGKuJP#-^0}d$o4y9;Wq@kz*)o9NAJjiN~Z~
zvMTC6a8Ym&V?rUm(2)}7($Cfh5BKw0H^;Z>Xe9p?kD#!}Mm6CW&oFQ)0e}p6JAxh`
z48E5m(gS%63a-TF(Wc5_aLFF=|5j7O_dT`X`vVyH-^_WB07tUD5|&HH)SJ*fgY9lS
z^GNo`70Gc{<0Fol&7L2?Cy!i4UAA2G$S6j``}tP4CA^dFXSQ>=BMKg>>#U}UQrFJ|
zT8j%e&fK5IzZAqT<k(xUt#F8d7sZ0yir+C%;`D@=LwQ?h`86XK=DKK6L&19YJ`N*>
zp<@I;*Br1@UoVG3DvlsYfzU4Ap?U!~4gM_ZmMNRG#80d5Qhb=%U;lH8EM`^7k1S#F
zI#9}xxLp^dZo85vu(|qQu;_u>BlChSG+%>CEWY0cA|Pst_UsdX4B52z4*&{%8S{8N
zk3q?`G26lE^N5TN;oB6tN{-U-Gk7;xQ-B0Hf|cO9q=eH;**CsAyjmH1Ey`S4YEB1W
zCjW|SA#IgeF$`+yPl}^{6_RbN`=A%60@xt;Jy_-q9778BNdZ$YK*>DcapDxqNLe0j
zSXZjmbjr8|<i3E1D7)A+kE{_9>B;4LFU&5S8ICKagt_khB*PkMZbU6OYnui_i0z%u
zKMXo8Mr&Buujbi}Ll8E+20i0=V#aICXnH;`saxi@i_g#{4PuGoo8z_^J%)85(#`~Z
zOpsy{I46Z`Y7&5_5w8Bs7jf1-S+jrTU5hgy(CvH^^?nCRbEG{o36LFaC)W?#P7gx!
zO7!gw)OBRM+QCi-kk3g&Y8<mjJi0#b$*<!&G@whkxc&g{LPEyKo77&*u)d6I)(sl#
z()|-~E`S3vcZM`=!%HTSI5BZ3FtT}5Ge_W5i><K4nTYq!Sv3x0F9t+cTU=Y4B{sVE
zi9P;61}9)$wUP;@m$N!|2ZxLDS~4Dg4_j+~_Vb#PGcnk;Z&RxlmxLp1xKFfL*ufq&
z*z(9yt^By*X~KMAxl5}O@P*`fT4#Z7GSB4t$2ou)XJ0{mlrU0u%C#Z_LXcb3ZN6AP
z(wT84^1g>Ihuj?|al}A1j}Kdr-AVKHHDe!V_nze(D9wDP_PC`!d~|ewGqPe$7Uuf#
zkByA`#kqm$`jeO<+88*VhVFQcyFdLd+XhV$1Vem*CX#RXCK35v-+loEH?2$UT^qx_
z{&{xWrp5l())6i}K}cn`z#IDKZ486dtS20liDg2hh&A47fWmrV9Y@u|tu*0a0~@PS
zagV~EXa4e<p|RAP4*>Z9V%%r|GCq6EkUXpf_2&ueC#zPwk4!-)Ktb@Y-XOwHS`z58
zuzZ-S>rcy~-Q1GZ-V=kK!PIKe9;{WRUh}$LO%{Cix@VeLK$H(7<KigDs8iv!_Wu$I
zLKt?oLyvjtSwaQ9^7mKgG*^ik>*1+`=-3208f4`bzs$pJb}e?swb#XMh^@eGf#<m<
zSjL{lwY)TM8F`Ol%<#as0@=p5DLkhv^eQfdLCPGfyaUjwqg|6IY=XKfH<KpB@ctQb
zCtOD!eV}0L6KnqEF&n(htZtY4Tnn`4ASG}}%Nc&cbftR}nytvKHdjWeCUt@+q+|9G
zogWSJju7YKjoJOPF<#<f!l_SKO8TCHh?ITEz3KTKUmx;s^WhV}-hcL|%NZzS^KXQp
zEx-7~0Gc<Wp-c*jgxymL^w`pP2K$=i`=R-1v$IQ~S#hsb%cSO$pc>X5GMw^xEJ_TF
z<hZD{1wHqxva-KvGZ`C8zMg7~&(sb0L*GZ$uTuEG_>LIJ@H)x75I;P<xo^7%GRw{s
zBpm-e^{|tBkTVOe0fo+;Plff#py?8Azmn2(>)gwhSD+`rhLVmxTtkA`=iK&XTq2`#
z^f1HFW&iff@kVAD%j_Raj*l~Z0ACjVh`MEbBgJ%i+ht_q_ZhmQ;dYL%uQK5k((^-?
zSN|1CHdFD_<sbYBnyu-v$f)Gid4u$DYGz7RdL!wL!<)ZiE6Lk;6f@LBUq0=UICGQ8
z1>Q*=Ej-X=+g}2s(*DkDdPmgbv2G36ELjNn_0$%*yE3y95U_bP(zU9)SXNaGyK1hg
zD*tk=SCH{SRNHx;NdFn*UR7CmjeYbw^=lD%8^8^U4fE^EkfpO}B_DOY!){pLe=b>L
zHva14&2q_VpKm<9ve`Rcy<~gRbf+Iz&L$0~N^UxN8)-&+!k+Ne{#PH1%O_RI&u1{;
zcjD_)#_jj`8fxU(g>hRJ`9|kod$xj}N~5kZz1<TAjbF@Y?~D-SvX+ohs?wRPP`mOI
z<afU?F<Pge@!pxLeGtcZMXH*xwcc$(!OL9zL?^h61-Qkd5y<(<osmZe=bs^}Yc1O;
z!V@U3(To1@fI;J!Gq}6xOKMIF6#8qs<Z6MVtFCfOoFY7B#BdTK0)p8&LN@@2Z$tT5
zBic~=)EcUUw?7LE`JM<j!zL=vFm7YRCwy@A&@(ensZwH$_SNg$$Bkr+s7C0<sG>D_
zcwNz?@~lh5cm4qwLGr8WKY#lvN2Jet*ZpE%JVqf_$uNMBqiauWy8IH%V3#Ytmyg<x
zuX!r=qISHQR@|#P>H6)&$a5NmaJlFFV{Gggyu#Qzr~4l%CVP!EFv&ts*7Kjc8#0G$
z{Za%rE9+Ww7w+c`qj*^joyC2A?kY+7K(b8vwATI%8W7D6Fsy8g`zsEwT7J>eo=T6D
zJtKG{O&XS0W0rqb8#rNjyLi9d+&Y!B<#%;L4{~@4<C*D<ye@$L4Y;=xES=*|oaU<d
zPikm}x(?+N7>%@Ohr8A`J^{ZL<^ffN40S;~qh6$2jzMX@eF--=L<(%$hG(~}*yIAk
zv4Zoy71J$Q_Y2gIp4Csf#>UQ)p;Nw<wssDtF1&P>G$B@aqjA<Ilp(m{THEP8LF-AW
zZ-fDguV_oYDh{8I+SspZ5r4DC2Vy=;yTByIE?sybJ_`WmFG=7*nvhS76P(@KH&0cv
zU}}^Hm<*_E3kwY)A^;A}E=+(+wTVC164!z?FpbMv6UnZ9`=eeG-jS9pJE%*>A;VLz
z;L0znU@e+tt{N{R1w4Y29JxYjCy&U&P|V9hC3N%;h0cqbQ4-yJ+Svx>6K+o?+*(*4
z3^Foy$^!ijkhCI%=}Q^|8%xY~c|j<=5?3z143}zvn`lv6G5*PqCf3DO0YlgvUon(C
zfo2U5sTvDUhJP|C6aD%HSW|`}TJNdDs?I!V!}Dr*i+{F0lv;8=(+XU-9>kNaEOOp(
z3by}l^r!x&i6R7xc>Uu0!1^t4^$&A407ChKyp}<9w|s>==tAHUCq3N^9~@@c98<S?
zD$zE#8I>=>8@qXAoHGCEz$B+rQ6XJ=FxjqwI3?v3L={FD+Wc-t9bDMI+Vzj|7i`%^
z21ZJp9bh{cZ}lvHU1|0jlA8dVGAvaK4^rJ<(zO1bF6F~X*}Yp^<g!pFOTV7!{(xF|
zyWf3pwzpySzYj)!+7C|3O}p~#-2R+9h=@k#0@twJJUP_lXltK`KtfIr#^fs}*WWZE
z2(_PA$B4={lXiahTdibB$rx|ga7Mc{;Ltw5m!`koU3gn_q|h4UF#vm5qMxSH^fw@l
z0KOP0aKu_-YTr3t((x8|gCcp=2T+emJbJ)z!OuRI@g(W($hyT0V1DMY(a~$f=i2=)
zqW|I?>!i)75;{@r%+_csLsqdD?dkXdJu%Z!tc?HbTW&;SDvZ^rsA6ADB&|2oegTA+
ze~b~nEjh@`XWvupa-SyD{~+9aZxLZLQj;2NG3ltCagwzm@-zY0Q1AB?7T6;O5}0<a
z!9+G_U@f4v3FF1nN66+m<ki6-mALZ8+7Fu@W(R%_&W|lhwwx!bANEw|$U^q8K7V9>
zNLV?BoGX($@CbhWZz9xOK<RzM|A2Pw1wh8_qdVPCWZw95)=TrAG4YZbpMcLF;jq)a
zG;|b#*QjgpReBD&nUa!s_Ud~MJ&_5n_c8T8yhYZwG|yjIq=s@vmnaN9a>h8A#EMtB
z`DgmsJtw``Y1cNGiT$RJuQU<=X5RiO5Kn^Mt2$K#B+gtE)dzQwP-7{E#{)}v6Z0ts
zTH+ch+kZx%h!^rI>@(Z_bUnJTWsj=Pe2q}4ek}@BM7j6P^iN9!htHyT>ew-s?`c%R
zz#_w&W-lnO7r}<9aWEf8hAUObVdJEvmKCC6dd|qjbJX-`F=O)nRx<nFedbKvb@dJS
zSI2UC!Tgs8iUaZnFi3njN6MyVspLP|*>~|1UNz>^F!u9mTZL;DsMAQsMZ~R(jR-bO
zI?qytG&l0CRazJ~HjWOJPpgP#``g=1a>UX4@)D4iwNyP=l~;_;8%PD_L|te}TG=Gl
zpXBJ=evtCkWE07JhBakIYm{59Zo4IlXiqYUnu3ou2{=9;RG0Cne>5+#EH>W!ef7us
z>U52Owk%VBmp0vqa=7N3q8B}Q5&Bk>aIdWcRawF;(=}^@=-8b;<k!ODqL3;ElE!^I
z8kWa{0trPk<eA3EuyJ4-Ec)c`GjvW<Ad_kxUsk|^;U+5?|9wc9Mv;yVVM3G}jj;HG
z0W|?}?+i!72Ybq6{f8s;aX323HLo74YAEpeGdC@|g6H)6lmHn7KiK!vd{_v+wALt5
zT3&!ZrejfR&7v68NjQp)1*O=_7vy6={4|80%TxGhOnZz}-%;P~6|Hibc9=%Rp=!Nr
zjoD6woKhlD<ZWVp+782OlSVezhFzp4N#mhc;?cLm2^Y5xew$ir<h0AIsL5zCnSSs9
z$b_3wa${@GqvR$Cj3;N#t&I#nUX6Fsnx~up5gkGJfX`~^)P@2aB(>SSB&guU-H{--
zV#w9e7NXCU8gjYJoZQf8wBdSjt)nH?!SmH>&5ZE)v04SUe%D_cp1K(8=EDwrF=^^I
zshT*}Wz#AkM`FA^Una;AAY5cu`pL^@sAYrfKBw){zizAHzUNFHjb`Oqc99BArBl1h
z?q|^G>6%&l?msO;MT>p`Zf!E1xQ50(=lA^ph~n(($F{SB>H(px!uMGxzp&0E9Qr()
zLEq5ykL|D|@?f^pRjYPA>L|N;(d=TUs@-F!ooXQr1s5K4u{s7)m1sn!?|xOzQBQlM
zKXtVc-{n9jUIc?p15=GRk^x_l^P>?6;*{Iz?K|VCzby2EEeMV2q#+5J>xFP^no&aS
zYA&4a#)$ICJ{KNqgdusSY**ARtEB`?pHVB*(8bf*SpPhR2~TX}fpKygGwUgfP3?OZ
zxMTG)q5uAby4Ad>PWC~|2AW1PVC@MS5F>JdNf+RtWK4)Jo!`y~JR1+Fq6TuDUdpZ!
zl$d_QnFa%5s8^5Dl#XKr2WL|+U+cAPhb>-psO!2)vl%%k#T$I2bU?`=T!uv&%tkp+
z(#ScIWtUql#hvquj`=KXioB7n%6;8cIBJwIY%AU0^4(6!P99)NWo?l85Ui!wvS&{R
zMN3y$^SmHvZG80$(@+{Lu~iMZe|QU2HK2?Gh1oAprepJ=3c7Yl-q|~J4riaE;T^a&
zV$Zz}8^FuGu(P7oyKkArtOa)5o3paIB@eJMYr|ui@_}D8a9wkEU4+A83*z@%V@ec&
z2UjKuX0k+PNyg@70@^4=v%9B=V#<B{=1~0&(aJM?msy6r&D#!u3Qy_WuzjM|R{8kN
zLcAiTjON>;pY(5%DT%bSt}YXdInUmG8ND|)Jg)a!2YfT?Pc|f9d^Wy#CGw^+>$KB{
zH!sEQG!{Y?>#SYiFn@+zC42$-El;}e@LtVCXeu5<fNUB`<tC{S)LCQ#d7K}x)E@v%
zj3dbl%_V#Qlh*^Ts~W1uz3O1{hZz3?A28Qz@(A(uyzjiOvL2(_L>P+dPlmZfGVSJv
zEd1$2NJA)9rEd7;_0R%G)OBLN>GnGgdl+Tw7gvo}j#~Xr+?xOBzRqt$4CCjZGEO+&
z=*5~$gs=m=ln+4;lqK{ziq^$hRaQkYdK<4PKNjfIgmmat@85|x{HoHei?a%|V<4@X
zc=Kb;$34O0y`Yy%?c0gz`V~_rS3jlLUy~e38+Jo22_h3L(rk=>P~n!}rM$K6{)8ie
z0gF;qZYF5ME8lM7RE0s7ZcynuA{3gqN=k05XGO!6m&m{cQZ=RS8H{R0cJ{W&A#Mdy
zPW&E8&D?$lajkDn<KjgUB#LxM$RqUg9`iFXV#~Y6!rW5c61G%tS)i9=4QO_Z!^lz5
zOoN0!#Fb{Lg220@>J{!-0s;}fXDYD;O1A7$KUXhYEtmwnU>{f5ELkP)Asn-Z6S5j9
zxtn>Y5*g|pS<(try2t&7q~n;z{1;q9<mxKq2f#TBOgZE?)@T4%(CCjSXaxrq6xDZs
zE{xaag#2r+^B)~?=&xM=`h8khjwfxp`50v3=NLn#f}blWC=9J9Mg<4uJX;Eb_Zb~{
z85y6ywRBzH6U_`C`pNz(W&_$45yP}8>ug}f*&xb%N;*O)wNo@rZu-;>ou{YS4{1ip
zP{t2=3w$`<!o<I++zFI;ln_u+(AAyVE3%zJc^#4JT<C06oFN>{_GHPm{cEG&dUYt@
zG&!+a`HVU&NF#QYoJ~M$jr~P$^O2c?3E^z|8$$7q_`T3}enPT>V7f4waPjQ{q<pWy
zm`>17l<{qr!j3*M4-k3{37$W+F%V3rJ1z{N{}}wb1i}|1iy;>;HK+lqulXIZO9M=H
z!&Nf3z9UyvuM|Nuy63G~55>QUKW&P@sukuYI)0K6S@D{DB1iVJ>heNQmvY!cTz8k)
zi&YfSlCBv0%QLv+`y(o4otJ-tq784~(o-6XB;n7d<5~>HuYQ1e^&cBGKLE@Jjd;u7
zY<eli;CpOIkr#o_V4>}5=@(87n`EmVonw^Wp?G1f(vj%um~pH4Bc)ExXwUm5?c)7O
zN8m%QMe5N?9p7~-ipUN^J-y4xrPt};7^1{wU>(VFJf1%QY6G4X8619ctHybnBv1d=
znkN~-Z4&CRJviw8%UmfFQx(?B$Bx&u6MLveM+jZ7TNai~HtU{9>mgHua(^axPNUaI
zEU9_Kyy7mmQ^+4woBY!S4d29G|HE~|^;^ydeg~$l1L8_{vvl#AdN2djfSeX3`K=6L
z;6nZvn&G(bJ*V+qB{t*bVO17!nAp34zCWDssuXy~`Q^1>Sh6x{b|nfXEDx0l9{grt
z|Elq=PlJJMMBAA^DA~rvy?e`L#<E_a-Wi}J@c}#}Erc#+o=4kFmWjiK(im1Qy4RTA
zalo+tYb`O${7p8An>{+H73OD;We9_?t55Z^;0MBCN0*B&bNJrlc8OzWSh^I%NCuj&
zQx+I;VNq2|Q<AQC!+TBQ4h@Gt7@AOC#NWo4>Q&_dHsZ-#n7z8vSEz6LX524E0p$(w
z#vA#7(zn-l!2W#RL%SdDDvXUEH_sE3iay6Gs;FgEps5n_tIA2?sa<q#pV7`-=7*VB
zm^q=33KoDSeH3W$&a8Eyz{t+kOngKOB8A+oN!$^<a=t2=VaGm8_?N^w{%cY@y&;lP
z@!=KugL$XLEcoye0HC6mXrCn1{X-o7#Eur@Dk)fTMnE0r9J45reY#an>rvX-f>7Xh
z$bGs(75492BQ{z%_VH~nQHG{eWRE|FnTfBl(D94GYj*0EF4L=>tq<21ci$g;!kn)2
z#`8PvdC2uT{hR%x!CdY6h^S=6*ie2fEe%iMlQF_H_Q99-*t@%w%^yC#+Tlt?r-N>e
zDOfdgiipDD|DoR+3qgyOfAIie;WOl2OU^`2uXxF$eh%Z@_a!&Ke$u<?<eC>}?g}@z
z3c+5&J{{J+bXTsx5KVacILFm7B)I%J=`q7<dst}nKm`7bdc&-{6IPTWV{>^Ln@>Kd
zWqoDtgNxkU0l4^BT;XSW+(neegyFYdD#<B3=|f^8{6WzSDp31Std(0u1&u_N&ablC
zqfTtZI)ij7E+(GbW*&sk_80g?1*|~C;hS|yUi~;)$QYk?kN>EIK_jl&%B$>?KL?5J
zbj6}o_!_!-3ZtcI;P{HX`3jzAXKrl8aZ0GJaGjNNF1?F;Q+7Yt?%2C0P=siGlr!HN
z5B!*<x=<Muwsmk!sZ3JT3nBS23;Jf-8Soz-CN0b9FsK-t7h0Q<)A2R#{`n%4PGkk^
zH;X&^bCBwl2wpU!0?v)Pu-Tl@fUEvmV4#A1pga<*i$2CmqW1j=WSiry5NY&?uhib}
zf-tE<#F#OyCFO977R3oQR3-invE?+($a9+85PmDhS3^uYyCr2jbyX^U!<l=?3~4}D
zD+jpTX0rxbH$EE!oSbIYA#n3-A;>|Lo)$NKK4-!!d9suaQp(TQ1B6_M{k*3bhn4R(
zn*^P(lg{8fxK9;_=TN8a@Ap5TS`#pTwB3xDTJ7~(p-rk1X_5Ss&m`&sdz6R(r+z+l
z#;`J6Ys>}p{&1s6W4urj>7<7FQjKcDW#P=@li{yRRiPw?cKVZUK}Ptp1CTxz<*?xL
zaTk6iMGK??89Hdkrs;oEcxZ1xzAeUCI<*W<MuTd&h)2!l$XY?e;M%NIUd|XY2fC0c
zhM!)yW*zw&U-SFj-VuAnMGVvAIelNih(^#V<D2W^7Bz$#y2_Z>FV}clbOR`2Q7Q+n
z=zhjO0+acycq<Xd1)2fZ4q?#1uKSLu<dIl@=)YbNRe2~$N9w;VDrmF72}h_WGfr?5
z&=1=6iQ|Wiz{{QCj94OG8k&6W$-7J?bXml9;S5>4Qg!H<!4WcMZiHbqETorJt2Yfy
z8LAnl#w5~lvKK<c*6CZp{nkN=^I2a>OTf=vd5|Yn_s4!7p-)ThbL|Nh+|`>rmSad#
zZKLeq)boT$P!ZeQDgkT(f8fDv(k2)Tv0uB^7`i59ufq`ogoz4;q(Xj<n-K2V-cx;}
zrm#1z8gCY+z*ORo=*Rj+8(b1g%((K|F3hyedB}Qjy{W>@%g$z~C7;YA`d8+mu*Kmq
zKPm*zIUS!Nue!JxoF<?!9vpxZEU=f>^}!$6I=x48T)FXzCNQZeW(XBt7Je-tgeSIW
z{P>n90gOz*94i@Ly-7wIKE#hdqa24u+48=4Gbofa@k=fj4yL2t_G0{Cj?X>RceOaF
z?&kkrFFySqi?vJ5W#v^h+&1bp#{&hVCb20}W(zKE+4|*UAO8wbsiK*>Zb#pwj;v54
zLp77%F{vsaEm2ArzW<$R>Ou;!_}EJqS75hWlp2d9<owx!Q5R3bs4BnRejMO)E3%zT
z*-@~nIkL6V>hc3jBGkUtK?z;~2Dxe`D>*K{{h;~k`MgiE79W2cjz#!{c`{~VL~;LK
z%p;Te3@84mSV=QWhZeg16bDWb@XV*CcXX}=%v@NrBxc$*iTk0#O8SbJm+e?WkMOl|
zIx=;WB9TS_QTVJR07ApH>dCfqd^qoLG|a~V0M0%eu~Et0TiGFW%6pLkO7dDRho0?G
zw`3s+4;Iwz=A`uSN!0mtINgj895eN<;@~2OE>3<8xP(7P0XE*b*f4*K+z7B*jm~40
zE*Ebf*5MQtq2=OQM{q73LoA9c5Z!je;`s1%VS+eUn{+`EZ%z<O<5FT!jHHNWqT8s}
zVT@vl|C-AR=}1lSDC+WRzd}iy5cVMhij~FpFk7=rG5hMY11w7-ZxWZPMhMv5m!0AL
z_C5r!UF$}A5$;}hMj7C`1$c&B@1s&fSm`C?G#EHPii7<y-`{TGzH!3W-sQc!|M+m2
zX&c%2BD#LJZbMVkhgcN6R)PeB^D3A5ug+#-uCm#cKb_xwVNSNcxgB}^<i|vz7V%d9
zdPa&LuTw62k0EtiYXC&q$h6!QK|d<F5`OohwHous{hu7)w3aH|>V-x&izhm0I&yAF
zxR#Z|?6E}8weYesp_%xM7TSkoY}kf{?|1dnN821xXe|24;9^5@{98;qID&Kr7tM5#
zXN<@{mi#hIVe5Muc;kT7r!!4vY6J`YPvstSsl?Fl(pqR0e_m%9BJF0$PZd_H^Hgqb
zAzYGdtS<h)2Og^86Qp^iR?P#}ThmuNX9MwJI@;q(0Uiz3lXwOaN@01Dd=u3;XmLi3
zaUiJK-TYHS`B<4~erDzz{=jH&c9RjTl&bj<SnmXs-wcrOya=Jw<W=H(mpnVTfB>sy
z1g<guMf7$NnsKUiA2KqR&6Zaf-V>h>8l5PO3mtbs6hv;qqV_#_n+~)4XuZU~(Xw?{
zRLoB5!(i}21KPotuMA0DX)G$kTyr%>SqV@nKfH-EnGWR<0^W2*nw~T}lW%PJNCwe{
z4I0`KqfLAF2$P>O@k%Bu1X<&~4w5`?NtP6p6Y58gW35OzzYdcA;0tQ7C^yg5dw+1J
z(&vJ@rn~r*zvYAs9U~0Oic?z<ILgoTr>0H9SNp)PaFThe7+TCi`S$S*_r})k^AW_<
z&a~qJ)lZN#<TuV56Nl?f94F)jL@SH8==Q3~Ojr^#D1^%x!Q>@cbS+!foHm?#<PmOO
z7eT}?o<3~5ER?Tth<ahPcQn78`1G-)<Rt>rIC)nzr%9qq-Ad3vu$RxA?vtHOM+2&J
zP28fI_Q4g+>B4=;i#CptTOmYU*@Gcuq}t~{TH>BZUnFje?)kH~ot>QCl<Ri7FV(Hn
zFQd-NO}YXV6*7tPs?nM+ZM*Vw>&Ju^aP;)v)qijq`X^jsj}Hfloo3wmS}}P(1Y+9r
zUHR|<i+B`C<LudYDBFb^ab9R2=5Q_5JVHa`-$upn&2dWAxaatiB_z&Dnnbnmliy;W
zUwamJKIP;DpP6JSxll7w3>2@Uu3vA9;kX-G;#m`hYBA$md9@!1o9$WwBX5v%xB9aQ
zsW%bWXj(}Pt-9Nv7R<?Ei?7q2|2YMh^nnNCIGRsYX?5*~d@=+2R9_b#+BO@z(lEax
zCx?$Mvdi8_%VZ&gyM*O9f~C-o;?rK(%R%gfCV9+X8a)&C?wsf1O#a`-ZM=qoW9RI#
z#^0nuVMXgS&-TCXh)Ei8O0g;Q^t{%5bGTlUhCfP=Kc$M<79`hC`gZ%fw$X`7O{%qC
zz{&@Aw9G|$qJ(AmG@fRVj1xbezq7nP5Dp)8dSY$U0#0!>!$7nq9gE;%c9lRZhI(~(
z2?t}zWb>xOQ$wMcG@0kGMhYOx5aaGi2GrAsz+0S4&-*2zj6snPReJ0?F<!rNaaH9}
z=5lWZ@a(=_6ZQhj?jN7)#F0JaxlsFrY&QF_cWO&|QoZ6D=Xc&sH7;(+9W$m~Wq0w8
zN$ABdnY0Ia7oc#^Jpq#{raV_|V%+nhff&?0f?Qy>OqrmH<Uc*w-5Gm`Lg;Ztj#Yi|
zz};TPh~F!!KdAl?E)P$O+Gqx4a67F>?Ntdp^RzfKTzJq12Auodilt3Yrg#eUr06-p
zC;7tyGV@@_+$?<HCl7rK4Y0BO`wkt&$%myGZ3Dzu#u3>R`Q#^GK9}3M*6vnLyTr13
zdU*^Su>2T@4Ga{}D613V=hHB#JfR;MwMt%8N~V)cPvUoo0>#q*p7-+6KJ=n&urA@{
znEAH*ZZ<f$kWB%(TJny!OwdtWYt1UXQtWw&ZSPkwTIX&63mAv|D_N_#^ngP4NMcqH
zG)#bM5~;F<s|1zfp&3K*XtZT*4+k{4MTwy6X=ms&y5j3)tCB0*<}1v@-Px#6UUZ#!
zIcRdGDbkrEsuMaisWv8%#>Wv+8~i|qT5l)WpPCO5URXvLPd=mV>=tFBT1hE1k0YR*
z4^i&H%%xa)$w_*Zr(i#ap5lYp!R5ryD|?dwo6hJ+#D$}&Di;|Fsl&qBrBrgJ$)z@K
zgdegFyTFMwEbwd!prCq{=)U4jav_GjCB6)qysLS5!5>c?5}t)Au~Ua$pth~;U}(ex
z4f1Mz-N~%7VH4CHk>1akENxG(1=vjQ>IZj}Su73UJMcTac0K&M5aNoZYF%r0nj1mo
zd$#qt^{-QCaoE9vhp|I75dZ2b4*0NLvgf`>d>?0#j%VcbptMr#0ev@c+63-$q_^d{
z94%C1v~=Oq-T=PO1ZfM~pO@!ou2p4H_$sRAlng|ZrgAe3fNZX7s$l=I7mwNx*&DH7
zDT?T`IN!*8|8tf4KIQQ9|420-6dzvhjM2B=|7*+1%A#5q{&ld_lquntopU>v!x(M(
zKe*GC1Nr9i-51_?AE@45P42%2*T<|3Rj{dYPPo6<kD&%sqV?XnVzC)M?HdM54@zEa
z&1JtTY(}t+ezp|EAy`(%0;HT?b%uZ%VSymZVY+Cxyw8b5MRtq{+~#;N{gRF~@jR%=
zAFxMU56OXUxm5K03}m%2cock<1XdY6uLY#Pn8lFrz-V7qLK9uM!*=w%(oqOtV$XEu
zRReK6y_8q{vXR8%#&2ZRta%oaF#mx6U$%k0y5`9@mw45HC&oIqbL9CzHh}%NTpL4@
za$3B;H|rZ_q3L(&@s-kj#=P^769N`gArrWY@t)o-!!ADyBx2@XKI89P+7OYiCLO?6
zjRUtxVfIS5v?zF|_h8QyEZQ0`r%hsWI<Ak`9XseE8~6(_Ro}ma#4?8253VoI<^)r!
zN#hjpAfE^*cMI^RqZHi6w9}R}LURhrgdEdRZ#4qS!}QB!Gr?1z8vI;9Yg=W|y2Xwc
z2cJo@$3V;B26UZlqg7C;T&@S=sk<~V%iX}~@>zJFPT-Dcds%b?fy?!<i&=p;UR<87
zo%gGqlokUghsaQTW`#q{vyH?Tru)u?+sEU>Po$IH`cdc04i%bUt0pzL3&YfZzX!lo
z%Nfx{(epWKG?s7`TpJcm#U-8{FHxV3&~2d%(#C-%LBpF9<tFjAvDK+Sk(TuKl)B!<
z*Z9AwDX;@IVsqp8BHOM`Q5q9{3~Bc#w;7X;R9afHs1ln8m70F1*Akb*oDc4MP5lDp
zJHM1cs<h*(*bAuWB&FhD5aUG~Bv9rw>F}7yCRoJw2Z~EZy|!d&7fWGb`Rw({Usjdl
zzZObN?-5}UKl(gX!w+5^>CyghMG!9u3H;(uov%ojt`R32qK`QFYHVD%^~bGbaW}p2
zIH{bLOPmrZf&cUsG)@CXCQZb(mY3Ma-ufqrD0eD};d1>J4@i8h0|CL7G|G(8h;=e@
zw16V7){V-!kE1$aPnloG1a}DY;eb-M?<YmrjR?gFyvn;gNE-U{h0ZUsiLP)*XUD=G
z@b!?n$y9jMPE;r&bk6+A*?y=WV$tyas70^Q2-y7rvl0C@_(3voaQ`2o-a0DE_Wj~j
zUQk*XIs_bt9(q7P1nC--hM_y9Te`bp2nhiJ0SQ4V5$R^=66q9>R6^?9{=R3O^PlT=
ztyf2$dG71p*WRCvC`$>ZM$qRn4bScH2){oxRLeHs;m>$2SoKFt+`UP97&4v<u3ubc
zj{76tF4jZ2MKmZQP(O@vir)SB4439FcrXfVccsW|cD%L5W1-P%j>F@+mGd72D<=ru
zg9=F;j}@8PMpB=*EHM;y>wzK)j<l5JhjuYT0Gvu4xtD;GFes!sj%2FXspob^d|cMt
zEXI8@N}iQxeR#Z|iAr19p-tsoaeYah=*WNp={EC_cgz1#p=G3*`+b}dYdtH3TIn`V
zSyc4NZ65p;S5qgk77551ub_E_Fa&aDN=OS^#g^ocmqkpL6%xHuT6ua-qm}yLLMK0=
zOFu0bUgQ&1X|F2POZGuVSp$4E=6l^J)dmnzO_P-$1xXHM<#oL6KKEP)kC0Um-6Tp}
zt<N9-2wZ!ujNit1`H`{nL@eNDD=^RV%Wy??>%hgy*2QG0c=yQF(1`QF;#a4$$BWbd
zW~AJLk?L;&Q5=JQSN#mTDwh(O&;EMnJjH{Qm@1`EY?qJPPH_>H4ir-)h)If_P2{lZ
z*zM_rqoJL}Rq_*KS@+n_E(zc&CD`0v&%oE3^(511_$?d|7@?1YHB&4=fnAc;LJNW6
zeb51A>(%kH#XLU28sMqp2A@%(QqI8&40J~M^|q^8I6fSXstdV+vb-QL&?pn~CFRY_
zU=H;e((7Eks8i|q#1wTh(4L;ODH7__0mpdXyozPju~$&|F`Mta>-<8Se5z4l)thpn
zB1^1pv2D`Aaqs2Dz-bhmiYIv?S)p?F70Lj0>WpkBgAahgCxcnug0r>$>55Sx#<V!t
z7fzW0qtPJalMqy9Vzek*|9p-2#IrOmG|VSWLqKa$<8G!`h1e)-hCO*3nB{~MX~Nl?
zp)UTP+ovoG9;jPT-^Xisu!Zb+C#4-^yg}{7QQ($l3Do@*R;G_G==>g*$|nx!-Hu^j
z1d)%6Z`UvLu`GYrV!6!CwD8eEz+`@9NU4Zzo}Zj=su|5+4y{}K)Hv$`q|`rMe40k{
zxrfd*%XMe>ucEyiM~?{aonULFH#R*nXJCe>#QidDTRIQ6zsg-|Qcl|#>Lf>L2~Utm
zy#*K9xKL`XWUyYq@I-ST^^K@fL?=EQGY7FBW7qTaj0fk*)*1g!r3_!en?>vTE;dxU
zjR+|I<$oZsD1${m(@bf4?{@0`NC5{wC(#=>hezLpms|pj5o`ay%_aUTdy#6*@(U|c
z;NVb@S^T{SvtnS@!ir$KGBlLa=~hPGMX1Q3EkZP@AHcf$JV;B2q)E9lK#`D6Ls)=E
z69DIZszHQ5A?}JoC=rae2>*)ndKL<$N(EUszKT9xSxVV$C`Z(>f%~9VI1nNko5wez
zDKfsgbd~Ps)Cd~z63FnYqhqB6*=5|bs$#GTSE3SH7`OHkq~I!MO{M_zX`21`5vetM
zdHE>l-Kt6`j_zobA%6^UR+mf&)w~)4oiTLy$_a7{{J0v)$%X>qG%CqBW*iv&6fvst
z+<NKGaX?nTo8Ky6LTYWG>k5{l>UADF^&I0*A|73mkw(&3;UpB)e6M*=1GT?!=J2kG
zAp{W-<ltsm{5>kPn2?_F#M@q53^<j)x$S0F-eAe1f#>1hHnT40seeAKV*Dj6c|^bN
z(+)#Fdkmyk(^k7*HPJ26P`Z+Mt|KSK0-EB~s6hRfbnSUtz9shZU$g3-;)G7Ier!j3
z^3gxY9W;K542kvBdn0g!t$tX4H7^JDeM|W{cJp5ZbEEUe-3j3;n*o8r);5G!neAf0
zKIDuHzd=|z(en~(py@*;&J4HCJFdmQxvWu-q^VhkGwYa>#$KM<P!?9;|H8+u?1$tP
zgT0gEUR-~YZq>Rt0IFAL(eLh$mg>}bzb<NbEG810sHHjQtN!&OTI1+NrmTo6_=|}F
zYLP|2-keq~eZXGRpZ{&0(d0qDJi#=6b7IisiZgQZf}ri!FNZwuRlE?Ub5Ds?2g9Wu
zSw9c2IBWdi;{(GbVT?aKuX6r7d%?|bfugire6yCtPoRJ!mB|u-L@#Y#WK)2faO^`W
zt>n3fID}M%D&y~8tzXLy*>!fQcB$M7%TKFXZYIx;CYF*GF>Ux(d3w(pJBulw-|P;~
zMo>RC-f1ozeu&F^q_iMYxNO|w9>lE-Wl+#Tb(v`x*k^&ugSN3nU+eCx)Knj6qhC2;
z3iU4#4Qhw2evj#`qa&(qS?`uN2mjpa4T6}4hdl*ta0C4^m!{o3xs8V}ptK2cIqUys
zx%;leCY2k`A33Wo5JS2SnV(+e0>g;kN^n~dTZ+suc~RU4l1D%D%5oF*SxpC^NbhTB
zxj@R)O&pM@I&yD4vr81<ij3^UsosAPwl9|+or+b5|Jk_ejab9nTPwR#y{vYqnKj$t
z(}y!{6*XjU9O$Ge;`$VmG^6pSf8zO4zI&C3MJVge{+Hm0e~}S~9%BaZgT+<K+GvEs
z-LclsPt0kV3*Ue_CMC?iJPp@-2ZSaMkt_Ha%r3{DJkr@HN)iJD*OGuYZ_J%&nvSx8
z*vi7Ee^L1DLGNCiU%=IzVEw&EUN5cUIA)s)IQ;frC;Ik^f01T{fv$rOeE@Y@AsAX&
z5cGz723}P;eZF&wD}m8<Rk?RUJ+FEm&T|00>!2^2qC_Zb6^K5z5-J!}W=SYJVa;25
z`+PN;A~E(^;(b;p@!PYrs)Kc`!@%>;Z%X-6to4QR%J>rrSR@UeJuT!cRKl-?9UC$$
zmd8Pv6|>1KL=^Q%EVC3)cMJ>*w7cWNZ87bzx8F$axHUB=BDBa;J(Sz0Wxhu?%DY{?
z3)*P~t0Gk9YN<#JtU!S|1#WK3pAFA%Uhe#K2TyzF9@y|7{nq<Wkpm1BCv0oplwdVb
z5mlRp%8^8ke`r>Z!J1$*r+uf}+(9<Tpwj!iJZS>3Zcv%geFz<L(PobPkW4O>M;Il&
zssXDB7M^7b?mPP+_sH9S)S31Ujw(YL7+1TruHl<WW@-|)z8iQV)OVahcd2pOa~DwC
z&vG2JeFmj`%LWP`bBM-^U&~6VMbV>VDfK=l=fAChRF#dDWlbXEo56%$FZ^YF)B}gg
z1p@QiSc$4!30W{z^|;$DzCpuL4js#9CA<o&2Z(wEpO)<Ns!(is9g!sun=hh1+-=`g
z<fYRsfi-_PX5yK4W>A-nzY~;28h}m=o_L>G#ePN)Ya1hZ5n3VYq6^BgSaPkxA5O)j
zgK7kVzrS{H8aP6%7R#I0)~<iy@dDCqg0+;qI3Z-3R*126d?VbK6__YV4DZg5f<LeT
z4n)pJj7>i_9F3_t);=%|yRk?r4Nqc^2$H}Rq|+;~Yp|tLv%>Hg8)t#*r1lvu^*P?w
zJh#MAm7s;<>Vg09-3RvxKz*G6m$C)(CA*u1oH{xBS?=zxM~wRBg7q8k`T_YW?*EBJ
zPVw@`j3`;Pq^)oHE1w%0BDZ{;y5+B|Kj+Pn@Vs%N--;3X=()43-dez^i%zU_e*DUC
zMXI-Z(qcb1iqw2!+S|w4Z-?ITOUGT$6BCZ-1E_?uvbd12&k@e03(qKRsPuJ&=j(B3
zzLIQHKSBtCQ&ro0r9G*+^2mQCLyaP<G8M-|V?#2q9V7~z$ietWkNoduBlgm7St_=c
zoGxI3&wIotqcBXzla-wLYWa_)XYV7hu7S$2mPDABugkV~wvH%ut|m$}^uUl83r|_a
zgi$pHPkgM3Fzvz?)UX!wuXiCo#7GvzRVYPaD0Jh;CzkzSJ5Y(S8r;nKBkE%G@8b9R
zo^n-}^?uGjy$!$`zS9CW3nVJ)(?-NL0=z@lY`Zvp$r7wdK_DA8oCKq3ilO*!Gq+tn
z^lMEdzja<-YIU_%xmxsM0Mv&_>5wZe7t50!V;vVVap42_N*xzOTr^K|a2f>ud~&cL
zBaV~KM9ITqx?L?Uyc_K%lIV2Zh*3&Q6no+FRT!|fF$Oy3{)=D8z525SyiYQyj;4g+
zri93<D5_UU>p8Zl50hK~FnuFj_%jL<QZ)c9td*TK5Wu^Q6IpVxZMp}5x~F4=y{e7=
zFkN$5_8$>5GqhKmbiL;;95skq<op!7GHxM~iNPXjJg&|m1FN7Ip;;`dip$<{+O{R1
zXS6wS7U3=y`h<8h2-p677XBJ<XI1mu(kmMNFjW5^AfAUnLp6_nwoYM|7`c+)C2xvE
zt|eXe60Ck`;|qJbAfSanvGkjgN1US~$~b15N1Kce34iY-p)^?l3Qv{v5_76+F{0j)
zVn1Fv2bDFW6o^(2fiVxeuF4dqG19$If~yqU5;S#)4JR{$J-4nS^-xHTB*|k-sqf1f
zs~gpQ5Ht(HThP$%WF)ezWAwe3j{>;tT!-802*sUUqNkd>;mEmMnLR=(G2tVcdw3C0
zUIair?y>rr+CSG;fX{d|=gBTiH@50MR&f0rfF#-KQ)dqvoccOjBlu2zV@&uhTiYc0
zbT;uDWo|Tg(VaD>HOx`mlFwSE@KdmF0rh6}*5Dy4)3Pb$9riIR?h&xI`|&;hL?Y0Z
zN(&8!?^Np<y7%SC9Ll}~nNB_&O<&rN5(KJAs?54wK=`s;FSUso6}o5JLw8>)vXVVy
zreUnsd!~tUB;VoK1{Ot-%CxDh|9DZDJhC&sq&t$Q9`z3A>=KIQQS9hawwp}F+&~=m
zs$`8utV#WG*{qMCMv=b2`|CX1@#-)436g(Z4q!4zo>B9DO&+VjOA0VjfcxA7F4@&0
z(tsTE_xQlo{S(#Yi)lDN+~C%hRXFeFc_uYup%(!pbD-P9Q0>5gYKzt&$LXB_8>%r^
z_x7wIoQ}<oltI9f3V*q{?^d=8d<My|EM5x3X;jX~ruE}w`)B*~!#)m9$eeUwRjH2@
zr&Edal_zj~h$)*F{rACu!|nF+V3F!$L5YdLiz0FfE;eO69J~ooY&wRRdw9Rayr`ci
zz%o+e^!F35+g7bzuAg^q)?ePh{=CiwjW@TC^0tn?+gFeMbUz<syjH{Hh;;rR2fp&$
zXbI@q+GCLuNx&N{igD0g4Lf9XyR>TZPR<|dj&J&2GfJMUcnPjm5tSfAA{QXlE7)l8
za~lNzr^g62Qs+dEP{)KAFM7Rt40!%mMtjFKxz7wzTHZ<|N^)FuTo=6k>h+9XV<UY5
zZw8H1E2CoxRBSZ2SwDE{SPydEvK?+zlFm!#=f9qzF<Q($Y>$1IXH@OR8xLNNA>|4N
z#uWgb)8vIHb+mP_fF*$#S$WR{)fes);Vl~$If#cw=Ldr_I)KVW4;K}Tw8NydEHqIT
zH~WPD0W8JL6w0_(WWwG1b!876Q!y6NHJa$NF&0*Y3ae83Fo`T|0Yd8A?_(62D6p2Q
zk0f9??Pg(=p^1Sv(?D4Ux<<(hHHE)xDzb38zEY8$vMAK#t?q!8TG@k%{Se7lt3nVQ
z4O~v>g2OJ?_*lagoB4<L_Fje^6~40|#an48=v$}A4puiS=6?P?R@a>k=vwoFx(<5a
zJ?m_MQvf#M7YUsYBtC~!m9?}8SOr3u6wT7oGq9LPT@mQrY4))6<D*ARJ&Ecd<mZ)O
zeKlKW|4xLu92v+^r=ac`iA8X<?A;!C?VoY+OW(i#uN!He5jnsgoEaN$nl4#euamK$
zEQ$MOU-=;D<0G$ete^ff6c(~WfU)wxiQ%MZr0N*tH-z+sZnwi7LSXV4rj_hQJ^VP=
z%Wv+tan0e!d^n^MuZh<AdCTo?_`Qmx;>>Qd;qPpqJcxv!9xQH{z)~RZ@Wr?vCvTu|
znx{rJO|i4O%btG@<Kr(}J_R7YX*+|xCK*M^Z!wgy0(=+`LX?8uYK(Fz2XzcGsf1ny
z$fn{&yhU*A<HrN8Au{yuEuuBNYL^?+y_?xsfV-X2A`a}mjnK?4Qy+(7u@oT>p9LYv
zjP-Ra4W!#AwZ11T-$wiEN7gQb)QdCaD(mVlbuAJ_PJfAoeXtYbtiZY2^ap9UU1*uC
zPFfNk<@p;2yV=tIlVPra${W@hfIUst<vZ565@yEWgMJ8%jbskm8#I1fxq$h<w9Qn5
zO!m8#KFsr<hjvkhellOL63jysZMv%ejeh`;mE#&lligKo@f`d+-x#-m$M5LBnHgYn
z_?Mg8=!<eJ<QN1hQ*$xI8mGb|w#wT1)}sAjQG6487nmWyMgevPJ)@wTk}raTm~a@M
z!V~<ybMm;WAh!oSX>xg)aI?}38N15u$Fp~g)_#A+qUe0dHWgox6DZu?=kaPB^~l1;
zyFH=uJVfGD$GrT5j0ZV`+9$+O<9lgH7oYBxiU+K}xA?#dL`^{OUTkJ9T=6*PSFrc3
zD!Dr?C1Z0-*Y)FkaC4l2*dkM4x$1m&CBSy~hR5lFNERztqP0V%w8-7@YSpPrXzMXR
zjeVqXtnWlm1{2xhh#opZm^CUmtGLATco>LjY{Wi@Obe=F^ZOmmT~dCm^X*{Z{}YYA
zetjl>bCjroW<J|rSi9KD-&*|Y3+LiG${O*DaBR5z70@-`+&UGji{5x6^+m=up!|SL
zO}OXhH!{BG0^q%ZVGDv(RlP2~XbqJe3QfgH<%qicF9QuyIPw!oK#fq~7yGDQWNP~Z
zFlT5uKt5-VF&tcA-tE!=(GMtRWuIv!v&oooo<7bmo!QN?73r((vOWNXfX+SIftVEg
z<6r_VqaRtvMF+a!5Fi%yyMBBGUrqaE8dMaU;WQ(VDbIeHN+N*)4IXmLxJx^iFcdfB
zg+*3;gvnh76&&&}R$%*`d9B!GlB|klPm~HoAW+LywzVQp^W;HksbifyAQ>l(xg)^D
zn##+}i7oPMkl<LIeE#fG{3-CT^^aNLsidow%iQTzjh?{LNJTm5O8JTZ2Ekx1eL!D!
zE3qu##LwmcDrW?I5E+WCtivmq^)%A$7=rQ2*|8=P&Wv~uCX=xjsUq2Exw~<ccwPxx
z*tR`(!=UXL`0;>tmWl%4yH}GcS2FIuyO#0tP}L``1pl_I7wZe+z%{9rVk?@x?f80?
z+5CMB?&p!NSa(F?OAY-(1S^j+h@Je5r;UAYQe5J;W+Dep|0W5b_6&7|q3sV-4@5g}
ze>EKN{s9=}V|AHVUl06ZGdxQ>HtwyXUQ}BpkwZL`3%)5_%#s;K1PuJs=>l&y7Iw&u
zfOAGrdgJz@L+Fl&bdLZzKa3?n>i6(}&xL1%q6E(85{;l0FP5mjFx0VasE=HW`gFfB
zT<XVJmN@_IUn2SXweI#MpG)={HNGtOM+J?$L1-9ly<G2D**w<RZ<}oaeD$7IHY?G*
z_Khm#YL>!gq{@%P#}bszXa6F?9WrsQ_GpD1-7fI64PxTv6>OJFwTw>93Szi#8^JMX
zQ!8UY4lAP;c-o8{B(<u5k+D!#53uHiL(PRo-0qfh%V3-`qp{S~$af+-lqP`sfsMgd
zQz%Ja)YY+Y?K*d*EY2B{#s2=++)X=MgB)tyB2$C?<PA5GKJx(nsodTBsZqRKYy((z
zO8Bo&k+JrJ3`Fws$uh@Xz$ud9RRREP=@GnX?|j<fNSb>cntZX4ln-61-F1vQFXl>$
zkAKX8wJQTNvm(cFI+x0w?FTG()wE+K(Yz~vuJJmXs4s6v|4=OY7Z~0xe7u<)RjkR8
z>6(fA0W)QpxNGl$R9Tr`Ni+7p2)cib6Ty;V&@A|n#w92T)~D=@2>(EC*l)|0^BK(X
z#oViR9!eLj(R_M%E%82g<T&J7v2S14f}!MxXbxZOczLHTD-QvX4N<F;Ny3Tzc@p#M
zrz<%+px@@;(x!aKYJw71Kw|^ssvZ~_tv=59ay7ttAD9_jcw=>I1<`au2O=};WDPH7
zE1~P;cvaOiP4(l;o57>|kJ_7<GjiMGp0eKcQ0^Yr;{GGXLcj-MIA;f&Qlod&<zbd=
zlgM1y2#%3L%AC!--_c4Gg2(nderPzZzOzSszV{Zb#Gl1Fr$M*d_+75<w*O-XR=57;
zi0uy9%JWUgzxkQpQ`;(Ny?9HzTNywqhIxa{XZB&<cnK?Cmfb6H)22L#ieORw8lm^=
zbSy5?KVLTGXx%QCRc9B(VW=8<!PhI5(?ckm5>%?ULwnRM1Fm~Wb$q#ZRp^v;yd44u
z!V{e39CFXWos71={gV6B%iMw6{hqtuF8-2o5-F=5s<<d!-0VJcWaOddfq>HcG7j*z
zBE4I`Xwg3*l7pW*&@a}ZNdXRCP1GtlFs{QDcASY^APIFLh2*rkY+Q}tpfrJiWXfQk
zS{5VR0yTQx8_+@3AO|l;V!#QMO$8!TlkjAg6>5C2#paJ^iwTj{t9klcBqvOvHmrgj
zY?vg2EA7Cqiz>jqNH=2WS11lxWX-GkfNfiE6&F~n2qbo8diL{W%CTR|#&HwWbygN-
z;c7W$f*~XgPpoSrueZ1=vW#=6DD6f}oGT`gGo#MF9NYE<i6xVAnBt2WfyPl6VH6C>
z!J-E)V$AoolEIie0H95u72m_P{&tiU=coLQS9)2j1UV75toTN9C86lf+MNE;s>$jI
zI2OCDP&6qlrX7pA%jWfwk8VpDTL&a=SGo1CaX`h2^-m7Qz{1{k#O6wWjt?|!-&F5#
zNH3m=lZMOedsX|H(JxowSns6^q1)v;{j2b^{Ea7$^kX$1^^rh$otM8Slzq?CXw{B?
zzB9^c@itUhnWen^*jmCl0Y2DNtG=~U+xu1G&h5c(kt^bVHslQB4a^EXY~}gh6yfVU
zSv53FHdswWqI2L!O%(RM0o8S-d}gE5`bTS#H1|{QXi+3P<`v*-Y6dUk53KT9zQ@Wq
zrWUSbhm>PU<^Uf8FEPK&Gp6y^nzCj_-MJoy|IKJ-MbJc=2Ujar9B5Rx_1ZDFyKp?Q
z)k9mLf?>*&?lc-2c(pm9(cUa4g<YByNa^k|nVE@6@?)vzKZvxDWU#ksCcKG9sX6G%
zIs=L?D*BD8jPclC+xa!K(Mk<gX0`wW*!_|ppwrr3_Qhc7%xq5+VLsDypUSj0lnwll
zKsk%WZcoW{nd_yc2fTo>#8dJ;eVojjqE)q9z^IYN3`}v0zK+qHoD4BAzKVCW6VJQS
z{)pv*V#?ubJYcUesZp@Wq}9<jsDjV<*r~X4ByO!f=dKO1w2}l<&h|7ixp%%s)e=Pf
z20#qsi}}ZqDWquvP&kfWkydWW7cQ$&1T(~HtYF>0J7-)Ln_st<WxT}}&=MoHcO9vb
z%=z^R9w0X!e5>igur22Y>_3XEBcaLq_IE5cEhWexaAPHC$@@AAU38*xu=p0f!szcA
z`(pU*E)ZKIpvtcvT;V@M13i7tCg!|t$!{&QmJOWeTWEt2MDLWQBLfmur9TM|*EzFZ
zMI|wVJdb3Gn`IMb1+ffjyjgLS6SVJ65?Cl^X4#dz&?5(aHrU`?Zq(V<MZ$0m2X*JC
z<^1b#2^s24iM`!QZ&I=8vtNnQ^wO3^99^w0xkHaZNP6`X6PFSQ+i~7`f5$Rl=LI84
z4F(js=r*gzUrT`{xXZ^~?WT`zE8e1cZr=`0dh?8cgXcXzSVNtxa6Tl|+7$4*d?(xf
zWu*S4<XAw?5rMH^SL{=B@tgJIGHPKZH=zAf!y&H)L)l0#GE=Y*MCUu|5a2>1EpTk^
zuC>w>e-DTG-;c?Oe3gd7Vdm||wt^O~cjvS>;?T!GuxB>WMJ0WgjtE^&w!HTIqSLl*
z*H>e_@VTTJq*%Clpg3U=N}6mdsPX7UH=gh?_%zElv|6+X8XcXe6RA5%T=ClBFKm28
zzpO=FvbwP`3uVB?U5k643gN^azQf;t!C&x`lU>a+nu$GA7LQ1IYc5_i=0g%?$KFQ+
zs>hjXIO~gaI9M^+(Mjh_m~i$kauNf)===XM%A&WMMWu~1O$tA>C90DvLfMrmL96bI
zB$EA>^av71<Gq-D8+7c8rP-IVzqMIhC8=EQF^4GzR}SXbKAm(R<$02*4rDJhhs)M%
z>6WdZr7tp*l5mr6<kjTG=mtaJrPn_1sX3uam3K4?W6SI&=Tmdyo<}}Lnm-?KWV4L%
zYMxq=GB(djs>8&+3x;>^YdG$WT#ObAH6(6=XTbPsfFV>igr&2oei%03>>eHI6mko_
zTj7ygIz!iTCUsvc!k!a4;p0T_VtMhPMBoO*85pbO|H~hd*mm3U#=wXAIDDj2GP-ON
zDl34+NKATthkG?nHZwV$xfZt8E7c5DhCcD2XHivhH{j#fx7C~E%599%pt}e;qAV7%
z<*wCn`084x$003Y#bEX6VLTgIa%w=1fL>RaZL~f6hbeI1d1t|^OnuIWN-pZ>cd;ZK
zBQH>fABVif0+8EBpWU^^9;#^3{7n{yvqNk-K#(H<E*FhbQ)LphD~FG(4yeNl-=Tu-
zcf8RtAnLpwqtI8D?<|~%Ypxc=j8bHmL<JQnOghAiQCU$*^dFfgr@{%%pfgX0m08pV
z!kHt_@|@PPY1KILj@T$92E~wb4`a>DAr?Qx$xY4ymBn>o^(tj>j+t(R{J1bzEvQ5R
z*k!@P_N$2f+uvh`$cfD6;*X29FWYB8PzUqb!4<|LDVO;!3?-Slj^LO~FJ|8^H=qcl
z1<*Mv#(3duxy<sRrmDeo%V4&_*pTFWetdO8+0~8|0QwfA#X(+SOKWUv#|PCZ3q2gU
z@^4;Tfo@7JS(A4)Hb*d)wtNz>^4fMQ`BhjB*_sz(I(lg|yNMh$tNWGoPUP$_0bFbI
z+169P%|*UCHnPhAwOR>MzBizIFGdFEGiCO%qe`YIdKcQ@Uu>1KfddR>HZclyw!-(k
zh^>*GaV3>eU6~x#fB6KSR0+ozh2JDw;7MREf_FJHvj^xRp#kpJ<fHkr?gpWO*WWHj
z=l=pn^FJrhliSUucQd<3TenyFx6Kl_y|?F#69;D(P5GBi|J$_xGU!^WH~{@)6+|Zz
zv=^r~+b1>O+FrUggPd#m^y^Gq{Trb3cIU;y6#ozGb`D-B;8o-7Um0AY0Da&}2@-YH
zzv4kj%t6z*sr~X=FHI5Cin)$2D3fYEckWZ_Twk!BpX7>1@HF}hugw@eHKDWuYBIwE
z`A3Nn;-IFdQ|AjZG^_=FaoouVEMkjIb5g0yj`TyB;p{2!-_VzpQ@UX^`Cpf;eMwa7
z5MZJK>(X+zbj}aU#kqKAmu|;c+;EYWEx*_na@lP$2yV1(TFo<DnepgyRRwUid3q$2
zkV<BovH@yDrfKH$^rKm0SgC`T21VCRcI8KdI-94o3006yk@@Q&a8Nb4b8~$G>YjCz
zV|2@srf_^@fI&1`){BP|``S2znKnT(JJzvNuTx45#*i}v-h2b5I5=K}V+HmiqTs==
zL%>y-UkKDkwo=b4Lpg<&@F`uSGAZEDk7b_rptif`84g&SuxjtXpCBi9WOyJCTuJZn
z#5yWzr-^Oz%+kgF4l^2FU)Y)lDRv4wCQDzEx110Vquoo8&IW<>HJr}p%l?!uJ%P8<
zGv~hfuiHY`Xo_Ly0aeAv$()9pE*Rjh2lq8dG<?Koritat<FV*bi)awrh}XX_5b*b_
z^+llHF~dlnZOs!(!?o<PV=sv&oS+KD50gZcE|O?#C#BmhJCiq}fv3R!gM6s0)A-}W
zMx<l=SXbz%^8eOc`(tq>MH&L1l?dL{O`XlrMoEEe0;Znc9mkPjNF+P*T~=F$7@ESO
z$=hR6g=S_n+O6GLxi$m@RwDwGC<bGSINj}8I0Zu^%-^y@YT=|ZG<Ccs2eHYnDEv63
zv~_G-$q@mt22eD@|7xEGDQOB4--YVvYHdt(+U0CEMzr-1Qemy`QR7=aGI+134Jl!F
z(U#2N^ifDdZ`XFc28}DW6Rv<|POa=I2X3Ze>X80NrYU44G>Y2K)mAfO`dHBv2t(mF
zh8om%%~g`a)3~W1Xe;oW?Cy4kY+PP(h5Iui{OC2Gp_-WMPl6khGP9l@U?R~i(}y=<
zDXwMFDn63y2G_4p#l98TI_JFfc#B1)?O09M!Hl35+MXqX1vE~Lm0x-Hs)|tG_dCYm
z$=97DzSnK+Ag|&+eSbN#ONr(-=mXE<CpO^;$m8EFLAGhU1R9w7hoH#s>;>gNphK>U
z!r*vg^Iq+^^Qa(=R!9*Fh;}*`y$sofH3Imv`PfrN<u!U#ePwRSP&;yLt8K9D75Co?
zE3)m=u=fzkFz5;B7)<PW^}ruI5s}<HM;Q{X(Z;@if{d?@AA=WB;O&_5RS^Jm{*=uz
zeS;6qQ?rrB!}#J})gz02r3a<(8PGZND9UX*o8`KR7iFY(dOkjltDFjA-`xL6tlS(S
z6%sjmEPKb2DK>t)Y04r^V<0XjxnX;WenR@(3)%f=2D^uU7vr7~v(m>sW|17YsOsjq
z?takvKa=#kJ7T`4@Nb6%-`XyhZf~kDPk8))DqodJ{3-Q5KU`_JBz4;T8*0$SM$v<v
z{KkIl;@RI;Cj(hE*AwRf&n0<ni#)?>V5>YbSEJ7vG9%Q2h*ZVI74(5LxeUPZRMn}#
z4VjUyfYYPk`thy={a;Yk0jM@0iW~>>)x82uvu?8t*ri1q))+($LY6#CU%LJUWnbC{
zA-E?iDKqnowq1^c2#9#grr?7dM{mBqHG_3a^V7SqU^RAEG=1)kf$&BGD=Ba|TMN-n
zay^vRa#jfku@o>)skzhK1H*yOmQ`wU{-F)IGvZpWey!i2tI+}B0DL4<yT4vS!FFmy
znV99S<h^>9jvo(7`05nmgph{hfgzaclyv4a$YY63Z`#NGkS+A9s%jyyQR~U7m(HR5
zA(HK~3(N)S2K!AphOWSsR7?J7S3b37;1xM;{#q7^hSP)pj%zLY$C@VHn#NCbonn5B
z0rwmb8hPU(9Wu`=A_HQ~`@pM(nqr_mQC8=V`pDYXt?%3Az_vm-n|}}B!MM5cZm81M
zaXR?2D46gV&}AGpF?URs#B<2DXl8Wiux{+^fVn5vv~6LnsZU-|W^x87c}|eAPl86T
z5ilhjj<qk<a2JocwS$}e|Ez+yZ#FFZT^BY(sWQ%wPop{&Wil;hBjmKnKTN{OX~1YT
zl3P|c=Lr@Sd1c3(JO?Ke2Xo7*4VSvvQh~{bbWakJf~!i4%0PTTASV(&?sW5^(whSZ
zB5`(90bk9TDjD9$1L^NFLXE(rKe+~XZhn084!<#Y8}XVbHlo;UN9uzy`7&#vx3X)p
zo3F=|ESkI!ZapXP(Ou2t*Gp*zdcaa3S;%6-^eZRx>}AusWIwuF!maDM#>Ao|t5&~G
z+nCOvszs8o?{UeZCYKlM9t|6jSuxJ4mun50r&3q!tuh-U{H!Dzg39EJ6?kQ6_0=&o
zT?)OH%^#5FhDbarLT+14<g~z5DaN?yBiU_@%*vE3>DAuPi9VZY8<h!YFvabc4Wv8*
z&xv}05eyHh)iWt8lUIgs_Z&^Izwo;AorXYCGW7+xKFg#~S%kB}4W-L!0JV_#MAsc@
zoxHp<h0L|*bWAsa)>?STQf_|CmQ2a`A(90uJA8b+VxW=>lkZVjI7K?IT5H1!!~T>g
z&W{0bg)ggkL!z~eVgyXl{aI@4Gb&|@s_$%^fICraS6lalI1`&@P%t+ht<pJ89m(D)
zt5nJ_N?SgZ+eu^WCu~LzgU=v^GG6VV0|sgzUWU9Kb9V<yMnV-o9+@ZWBQlA~K5ULH
zQ3en;$p1RA)w7butS>)d@ozu)a!JIrQQ_wZ&d^`he2Gr0`IW`~oHzFc?zB^4w_Z7S
zWub=wl43_GH`Xw6>Fa_#2Z6YkQ<yymcrXDdzfk%S7k)%6uEdZ}@7z6U32AQS#gB9A
z4UP;zWc`3}C`6P>B#u!;11sR<!-pdz-}^!pL*N-;Wvf@!W9@V%a-<<N-=X5aW}-AP
z*{9^9dNC7t@j&hVC_=??p}D)sZL4*Eq@ljvdo`Lhs&IS$I;z3rTC#9iF5sAP^2~ot
z=_CXs6g1~P5d%>?g@SqhzkG;Gj8FwxV=+63yB2Zs16x3e)z36bU_v2RbF7uoZh9)=
zS1fJR20Wm>Cz%z43{1dS{$-6sBh@FMPBzKnhj6B=TOiBxO0^!>1rvtNFdw-Y0g)xo
z*?gH-+;iZ6BtWOakOQA(_}j8&0A4A4<<u@`iL%^>R`XR|?HnG;%t!%XGX+Lew&1Cf
zZZ|rjs^4Njpx`-(<J|o^s{lp?*Wq>K{H8Z$suhl!5pQ-c$a;z)g|k?<W8y6Pa*)!0
zPp~b1l0Ubz<Im$xeXNmYSv*L_yEE1N?~mr@@dc(lg_D#otZw!9q5B_92^wq52MEje
ztYjlHyjbF{vB4E<<gv>!k1~*_Usk;a`6gBTJ9}gBO=pPnmJO!u<@-0ZgG9+ne0cnT
z9|?}9B-23p@qBr<l&%wNP{ua4Jl3}bHp!*mqI|$Y<Im@NB^+9Wy*Y*NXc_6?T338C
znlu>mn_<q&oO8T?TMxUhl4>4Woro=G+5*nnpqU*IuV)@s4mjd`YVfPs?{pE*)Cj^%
z6H$I_9Js`MQ@YgLd%Z@3euTXa*pmKK8LF@>y8C3rJq1x%*m=(;mW`Hrr!7K0Ed!&v
zX$9k<_0qTgZ&y1EvBGPvBj$}!Ad+Alc%2yuE|Vby%9H_6QCa=#peHN0CUpfx<s5-W
zUwB9?MKYSDG6QZO^xgWl{9i2f^jpi2m_qbPh;15F5t3K+wj4@~nbuOmA2)D*7A{cs
z&Pt2KSHrT>)}{5=xU0vGR;lfF?+=ljqn?$VqOzzPWzMLN`M3%#*6+9RWo#wW@H}mY
z$<1DCp<PteaQpe`Rm_M>SG)InX;hT0i%pX<Agpy!3&}jKj#DbBmSswYBrNeY95$j=
zioFxgDnND=YxYQul+I6$<~9FA1s#8(!E9j>Oc*V|hEI-?)4<6X)x>sX;R&a)EVEku
zi61NT{Ge1##z=wOQqm|YqRrefzwd%Fav;4~vf2ud$WkxKn%Tz)vf-01_6Qf-R-5TB
zKL46*H^q>vfnl-Egvg91%86hCl|~N5k<}Tc_#X{ItxgjL8>lrYcEmf;HgeU?LSQ~2
zr?aIPejj5pi(G=rDzKiGROC3#e-L_ATWGP|D--R!H4RT^BR%IJ(Ph!6T9|2J({&R4
z^sM~@(OMb^6*Fi~yo~@OyucBh0tE>#%dGA5*5XIo<8>~2le&o|X35h4s+hh2E+p85
zU?PsS0=JLuQeZLAB^tQguw5{2)^zR$skWJi%Eoi;<C=?st=+&7|9guRXjL%9O~2mN
zEr|v$#5=SSXh9q&`{?z}?J|tFAJ?q?&JaOk>mt*jo1A1!S=kU)a4Mf>R!hvb03ton
z_<)aC-ROJP@;1Mzy0hgRS)8o0`6UPyEr_;HQGo5V3+PUOV;X<BU04U+`2W8e^U58^
zSFa8`2zahqu@1wjj7&T{{#2ev-F}$1>oog(G(5}|kal$7NDtY<VQd^N`MLe$snOd$
z2-p`zN<XrDmLr#5)3E*@)%<U}Rr^Rq4Cfbf5WUjtTz6(E%ntZHCkU1X4a8g6iX5vk
z<n4E_I;KhFbwHC0CjJ_)dS692uP5g~bSNb4N0P{goW=vOi9$>JAG?=+=Sq=kNw~YW
zM<EL?ZBu!JP77k2CXi%`)nfm8bGLft9_(^>odBV?t#20$$*UE<TOpWd;51qzxKEF{
zxSzjZMq2VtR&s&7FDD)y58jeu<gA|%T-5UH>T0enQMjp6aZgTiati`?uJm0AVX_^9
znzZN`_@HoeggylqpxXYn3mqKf1_3eUj5$VmztvN`TA0>j;*8;XX*Ei*z57t86XdYk
z=!YewWH#HN7_$9g3d!y9%u1~Xq2%LW>`vsMQt)ow8LEnZAie}bg-34>PHad<E9m@g
zQwRHMhv?E(zYCFra7$-x>B;JT+U>rMQ7M~po?i9QURDHd>sj}oZ<8mW<gUKNFIRcr
zhW7AG;t!Ch2!KmL-1D>ipdhbcxNO?TCwPi=;Pt@mARjjjYD%hqsQ~NsvzJ?As&_i;
zT9{On#Q}_eVGZYtd43B?p^v_XP4hiYk2fO{)Vu_sTK=uoPxvZ53{PHzIuZ}4ikh<S
zZY2fh3?RlgOcxnDv0d_uV}F}(?AtpS&+!wjtGnwsIZmiB+hC~6&bfb`&wSvgfZLSd
zHGB`i4MjhTErgF!M3+>M_29(Hxc7H_G9isgUKS<^WN99I%d|GfF-RdWA*kuDQf&GY
zMDoF26oI^GWo-r%aDo-q=u^B24#{^G)bIA<q+<%(1O#7KNs?4_iM;c9u`vD98?8W@
z-7Q8NF^NpC;AA-XQWM4q!v8q}IlbE)Y_&2ecsX$e>@qHx)4nZ?VnS7)-^$G$EmGRK
z3QnlYnok}3@_=d25q!Puk7x;pPI3y~1RHZT_I2!sutFww3po%@K#!7tWL3rGvd|`{
z&X$@OcOe?1h)ab?^k;^#SMSTV5vtr%*7?YJugIRQk5;C5`~35~D7pkX$ZWfVZ{m?z
zBY~GC%ML;zJ`RB#&FF}Mx%zHbVE*q`m=g%=VjIM7_6P+}bFO*8;IBr}g@j>vpSO!|
zn7ti*2Iukb#UnzkWSgg=`k_IW|K-;gxdimD0VyAWq|s)7iRRXT0{Y-<k2khjB>o=0
zGI%+*`fy*CnVB81GG2D3bpbp08(}7zn3}MJ3iI6#wkLp3D4nT<kZtD^%sa$p($`@&
zTF7$dCs5PMLS;G&M=K1b?CuI)32*(?%SB}#W_|I$SwGn=6xOiVRD(j^AE#K4AiRZf
zg=gx|r*w$Ji9cJVq>p;KYiezV-}gtVta@{ls05^}(a@UaKr6t_p%4z!t9nuGSTpS&
zm6Btc<m^AS(VdI;aAvuQC{aI>paWqZNN#3_52U9$-i^O@@oGuVmp1KU7+YPAH(Xm1
zbK-~Sy|)r40ur6Zptmm^p>okDFOJp5Dt?4_^Vj%x_v-Vv)0xGX#B=EeaEJN!ci^t|
zJTc>}O-}t7&mV&j7yvr1+3@EF@;;YZ^{4t?+7Gov9Xl_d0KC=bi0<yH&f!oRfVFv9
zHR<CK2$nhQmUsXl`Y!}Ql)e<OMy;=tWO+`-v+s%^0;0Ou7Zk_~S~FOJ#e0`F28e=+
zX?(=Xy-Oc^_JzBqwCeAR1Mgk*L<o-~(cCK(d?sFFufNusU$+YgRirQ))07oz+Jo6w
zE>#YI`wfiBje)goiTva^)A~Y#d5WFW!r9#b;ufbVrA{fen`#00O(}Mvx#8?M{14@+
zaz1Kq4*|AGEp)j8ylPv*7Sw{u75tXRAzZ!hM@gh+O4%V?@})Wg{jn*{#jCv{`7Itp
zP)yc6zB7)QSG{D0LK*;Mtd<TwW3}tW_Xui<hbrO87L*nSdZ$2_3AODiGVS|JkqDeu
z9M?WWtd99L5|lVMWSk5UCCBNJ(!kPW!U38kCYwep0UZXtd#|c2nM!Iajj9+|`GEjS
zL7NE35O?geEg#noCW?QFDXu_rhN%QHnA5Gk<Nk@?kXQn6Nd;>HS{QB-u{xtIFmCI3
zmRJ;T3xBAGw_dtN-mc6nF-58CRe|OevtmbAkegCauNAXq*y^j5%F6%-oL59x^n3-I
z-|gbT6<5B>qp3+kS_h_AA6+5;4%ME(Lm~lLw&6z+m&59!7Q6PS?gqWU2SCYt;JX`9
zu7C_lD=GSBn0RozR9&Y`jOOk!SYz6}F-ZsuD;gA|z0X#=Js{{1pWYm!LGztbQB9Fr
zX3_#rMiFxzz(ijaL2bI(lmr*^Bs+-v4^W9K*?)_XHSNs$Xt92Dq@%{)OL$y=DgGcr
z(o$+1RYsGU9apAA?LgD3Ba#yVq}Dq#s=^cQAhOap=>h)BP-mQxN+%ySuLzF}l^c0f
zSA0M8QM~M<Ki?N|l{L!rMY9Lko7$!Xub^u`1ylfmxJQzMX{IyiWBywul8umE5d!`~
zMCT+zFVCBIGkFvtu6s_t0ZQ2A<|ke@1-PI;MiG!^p)HUhVQ@}1W;_+NvkH!m4zGKB
zGy5G->j5YYjgx;Aj=}A@Yeh^($HG9`YPFR91=IosPNuPnH(`oOBpG}0kM$CIBC9>O
zM<8JBL#SnwuVM%o*l;LVOrx!xd5O;lPI2X_nAVD-XSd*51&$Q}H|OKpO}36Lv0{UE
zu0B}X;h34(ZMhBUMkqD<@UBgTvODPaPExGz+zh(DVfz<nYKNK%;W`tB8x6mEyFR^w
zL>lV}?RvZ^gg5!_0jG__OqUROz<QZhz|ZbyDIeZ$Wqo)eNa`e<c+nP4d0S1zu)<1{
zp(CbOg0bFz0Ma+Q4!E6ttd0*sG(frPMduh;O><s-nR#$AnJeB5|1@K-l6nILlVP%I
zqYFa5xJ$@gTaSh6DKYq3{Cat@cA&fEOnLOh>&5J=Y{F}7)Hoj^TmX$ZLRffFJ6D;z
z6|B{&3!LB2pEIK@`S3*LIl#jA2iE#8bFtsKeRI3{%~Y8FIOAsA7>wG_<H7^?w*nHr
z9X+|}(LW*$d_S$jkZ`{DRs7)FXwM}1=9I$FBsl|-Kaj|7`nFcsmZdX9K{8AD*Kg28
zO~pXSK9~Z6pDx1jjczbis)r-%6J9SH{afDt+V)$I<~AGwl*gNH9WO9j*1}d;*#UVf
z??f3Kg+SU!$!IAO{qtKaR58=lr>O_5x5qSvb&dTHX7<bfKEpJ^*5_U<uX!Dp5haML
ztQcGcT(flxX%@(j$sk6p56(pLdY~y7Rqumv%7-Hskdhy>bSKW#ZxxvpxedySn9M2h
z@>1b<i;;s<oL@}_`VE5Qbm%BiM&qu87!No@9ya)@=II`tCztAgJ#KASDoRE#%{zib
z^(E)Ck-7TNDzIt=>TsAag;*!}>o!;&u8mZ>IrbN?(SG7|anGh8{O@Bj{mA1@R(2pn
zf_@eQ4a;Ij{s$VAE-$KCj>ViFs$4Ww)7L2Qizc*(=Mhkl{5FA%m(g6cr4AbGy>2Xt
zfewNJAdO;(rC{_viwG*+IcXi|{#5k!kCe{9qgH_+WX?vZ$#-U*t!F^t9)fy;Gl;F&
z(avuqoj9YJp}s(yxEZH<?*+g&BQiz>v%K={@mLpxpC`-Kh8D$;orPBoZXFOE6{SH3
z%jQI5@$E!2d*ecLinA3vN$B=d%nx^2BLPxCKU9QB|1r|d=tu6I)ieD%#Hzvi$Apk4
zrD7iue2d>E|4?6)MP=7|^hkCnT|YRRJq;ZGY|t>Ox$@qrN%TOt2&!B&btZy>v85>B
z)BG(~IC>uUGp6==vc<8~9gkrS+TE*^c>D5qO$u;5fZIZZo%=(gGY4gJ034`s-uO96
zU5Qzi`*$#!C2+W4{!?R%UeOT2sr91=_k(#~u$dto=s312!s1hXhE7Mp*+x}lqfEvw
z5YS;_H13Q5Bu;H9d7;K1e@3;kf_b8d)_jl#QLCz{rj#%-`Y34UC_wpbHo$0RATB9}
zBg+@V_hR5~VzD&{3LPC`J((Y~6^IVm=t_HcF*1C<>0E4@I)&#mUIP$AEnWnUA>~75
zn2YXbk1S|LW-neU@0xesTZInPZg%0sVm`r%juGy@aqo1Wau#A_&L+Fd2esm>ZZsd?
z7Yor!c5x5?5EOX(moKmQ_~t~}rgjKe9#KlaMTyn)c#p^cF!$;7Op*nu|KFZqT=?v(
zOeEHdT)?O?h(X3U07-SuK%X?XJy)%*`kN<UV*R#bS;%fz8n)21%$wmN(ir4J1e3+6
z-+d=OQ9DL4o8O}qP%AexL8FB<kfpr;jcxFvFP&S_=(uN$P%CAWC_L6MgN<zSYYV)|
zZM~?^%xsxP6l7Z>m`_iAdqHDW@SWe*I59A@EfBa2hEPP+($A}Ik7#ha(*rM~54AaG
z2=>Z0zlTO|b<mV`2K{~>{Cu>Vnb_)4^#KxORRE?w&!FOD2NMQ5o8Dj>?1>0R42=YO
z5)=<eWn<&1X}gyE$tTV|Lz|gU#^N+ME_Bqwxec4cOsSowG6M8x^owzTTx1;J@((cx
zcwoAYv#TZA6_bgfH2+DgZsf9qh&y1{M5@g#41!j|uj~!P)>?TLV0S5E$&S$BKjbm~
zQGCGacUDXcR`F&G)_X+$H^cUkGG*KQ7cQ7|%Zvxc`>tY%AhtOK2JPH8MB;%?ecdq|
z?w<XC%ImoQ@?t&Ld=;@JRodfbZTFiBi_fHd;HUDJwHUf<3kI$1hw(5T-7XPD7MTHC
zZnkq^JQ@h8-bXh2GsSV{6{Ju~g$m<^pZ2L^cVa$kVuT#v8Y}xpA-our!|ZHxoWpyu
zk2@(PhsTT9hRIF!J1JBteGzGvNR<*(TR?Z!M6oEi3S|rJs+=II_AMCHDItKGrkLjH
z(A4v?#f`{ND@%v=eQ!%8N;lqjAOv75+cky#dAtlkt0M5XH~EXPj&xhCv4*6wC_4jk
zXu<A31c7iSUuA7V;NG0zk<{~7-gT;NmDe0g6M|iWQ(9RC`M{h6(6xLn=%d9GDaz4u
zg=^LmT^pL%6j1L16H(Czdd`~JO!s(qlVf94H-SF`Fz<@{9)uwV)hwc446pO}&4}EL
z+@7+WC2oG0=3<Ed5A*pI##4;@G;H~`FFBL)XpMXFVL_Q{{EmcQQ*3eB2ZRcU<N&oA
zY#&pB9+6!iQN(`rYsvq43e`5hf&MADub{b44q;3IWKT!ROAFZeB`!byvpWqg&;mT9
z_vqO-u7Ai<`#ZN^jD6x|(@gDS6j~JBIuKM<li9)CoWLyBGIR$=C0VzM1_6&3<^<8B
z6JASdSlpFni%f-7iB*LvzCc8o;s-(526M3b^7L?*-L!bbbcizIHRu9Sal$3l?yVkF
zbmgE;jPRW_=BchntUUOHS{<W;X&hdD43=@B*h%Hj`U>Gn_W}kG5Wa8+ec`;+mab9Q
z_xd*B;k7gXXtcs{$u`^Nh$sg?8yYR|549u-ceFy#=A+K;-xIcNE?gpt%YfDX2qQp5
zsfYoG=q#PQMR=5=U(ViRMEKi*5aVs_zdsbyR^(TG-(t^b+BVVVl_YP{0AG$$3MI7Q
zJrI=)C-_Fw%S{Nw?ltbx0?Ji9i}Id|bcv~y(Q~yZ#{hK3RDr%(v8ij*ZZJ%M0H@8n
z$*L}jtX}@^=kmr0!7P9-rXQBVAi8e@5rk~9DfQ-WZ6UVW>J3`lp3eaD@`P=JbN(@?
zr`Mn{Ae5zEAygrMXR4Q^Q?ti5JjWPmAscO94qx3vC`js95XwRo2yTA+3gP?^RXEqF
z-kWeC%9EH`|GYp0zKbx7>$E;Gk9PibJAp&M?leGks>0+gMdm~AXa&PK`Lwdma-34B
zn2l#?3zeA5p=K)fXp?)35-6s-ZyX|P;WTiln6mX}Q!*G$IqYa%FIPWO09-LpDA!wW
z;xueP%wNwrXbiJ$w;|Q}Ycn>i0zORTp$RDAeIhdBa={pQmG!^WsyzMk#cZl}AeaN0
zN_L*_VjJ8y<}qbYT}X>toCDeLqdzV8<{4UEoY~-Bo<xE4NZt1u?|{1YQ#KRcmtgn<
zZbUKKi>vVrFHnErx7Tk8TsdQGCbamMzJsLwG@$?VhCmr?Yr&pAziFNo>4ur)68|E&
z{8*>iLY94;mp|BFI8P4h)ZhT9!kOV6K0?Imi*7I64;hHBV^*}Ivre#*cs;`3b~I6E
zBZtQ`|KdG)V2$kfJsIo?o+B=^O_vyKghMc^jW8`2Xiio4lqdB;fscuDbw}?~K76?a
z7W|13h4xL0wRq$7FLh9kxTJrw>Qxpdg)`;O(;w?g1t}RXKGZoviqtW6j#@5$p4;ry
zL1)U)$Ugeb;KEHpq-Z4dkDon9gQK8%@vbVNU;r?U+z|1Hgz3P=@iiOh^x)DSd)|yT
zbTA`zF#?q6f1lYVW(YozsUr&2APe_c7<VqmZAYaY*L(x16mJi&fg9NRW1TPm$ks&q
zrED_j17LNkE$kYXC0r@48f;=dOB2b`xmQGYUn3$mlXrN`G8zy;cAb@SD1iUw;A{O5
zLg`#S-zNyFxNaHEp$I}dtq1*<fC_20F26_Y-|jN#8gTfRKfmRPO!D(&Sis>=t<%y+
zfvXR9uX^GigX7B!pi^zu37*)f9pB!J6GN7>AXFK4gdxC(?JZ2Wq<QbqPQ|=v<2Y1H
zeDlwgC?Kz2`*d_yUEY7K%~<^}tNQ0F9?t=OnS3@C;kEAb*tHznG)OlNx<K`b;3||g
zK^a!oHL6AGrroXZ8XaQ5ajrOkVY8@-%9Le~Wh3KO%8H5B0YMNnH}*|l*=UUiIjTq1
zlKW4X@=bg6OR(D$>gsc1hz>#Oz&em^;iYOlD>A%?l*xU6PLVjuWKx%@KCn;a<iZgN
z_H3Z>FNxfyJgmY#Guyc^BMtG2$dKYZTQ*07oH?z~_s`_(T(-fviW0_Nx4Uf*RtaU3
zfzK?jLZc4Rq3N#dU}$rsy~7d!JspS#r4M7XG`Lb77+w?F5*P(*slLF!C8$Fts8hJS
z7Wq;X#ng|Ts!6(LBCo?1lfpY#9>_EFFdGV1$+~4IaBKUj7*$zjVCd2~TUo#Xk)H@_
zm+~8C8wT0ILM*FlAvnO~A!s6k&xigkm{j8RDMSh_cgfp8rU0dOoDdJH#hkaCdw|I!
z&C)J`S2iZOVR=R9VQ!IPA54ieEs0HPq7d|p0BWZ+J6~c~*FDAUrIT8aGXU)@EAaKj
zqmb+}6Wc;zKOe7<;(?a4E674~G*t0wkjhK<81z?KH!6VdX-n;s2z8_$gByLSm^Ep%
zMPh<Ii8YR(^drvh%SxRYnhTnM9D-UIHIp5VBBpL_X3QsgVr#q<%>{{=kT99yl&oG6
ziHj<MNtN_(H-pt00u<n7hC(M(0x2qbf?Orpg)sfhP}gj|_s26M-b8irW0{QG!9=m^
zda<*m#>Et@(KG*>>7~E=pe#Lqd3`;s@cecC%=Gm1|F-s?+-7-HSC@2=emh3_{{W!}
z&9fJx#LIN|1&q8$<<nt88T8`rKkaXy^}GH9_BbaWKGSJti|GPT9$B?<PY+;%_;%gH
z2Q;0gAAjD@i`r`AZn>}rr8&&BsagX5+K3}Ul_CpFcmbF3=HC;*OVTxv<goRGD+cmC
zK{Ak32kWnTP)KlwXAf8~AqHk`^Z-Rsj263;PA)pETtPQGrV>xy02PDL3hrzBx@4SW
zDZ(Ys`v35B)=^P^;no)flvIXJi2;d`Mp98==<cDUyF)=rx*LavAA=0iDGkyoErQZr
zN~hF&eD7WNKV6H(S~J6(?|JsK_h<LJscak;O;(!PV?N0nNDB)q8vv-)u5_0K0LTEp
z7v!#_+0RusQj)rpOX%%Ac`?sm*>xqkNk^--kF)QG7@WC4+<+tR?A3pIr1FxwmOA3W
z(L;iO3?5?p8nGUX6B$v+fv?wX=D~YnPAkPYL@b)o#BMSWofc;e?l+7dS>DXkgN-rJ
zDH`TVmiwrh$)cSH1i?LYiPDbDrx5H;#5Kh%eFC@sIz74%oDC5ZL4i-Ifbb6N$xi~@
zsGyO5I)b>J<GqFT#%e?Tw+v@ZTUi(?yw#&p?!9@~buJ^hyZw|3KE<d7jjSsK^k$hZ
z4OfC}iSt|K+u49qYB8zYSoAOe=kHW2)>|Eavx;48bnP5rlSX~ge(hbs)$DM7-%7oc
z-#P>6aBG$WVnPQ0K4k$IMz&M4C?Fe@G;6k13Q8vax4DhEmx%<xTM`K$%~9RH@r%4d
zpDbG+;jlq~ceEq(t^eciLlXAy6v}O>)DG<G$y4#zf@x`^*%Z*PI|07DVt<s1i-|z_
zHv*JP4Tj9tS``=0aEy|EeYBhclfxggB(_b{fsY)*tSTe=-TLtZnX9y@l#}J%dEM;w
zF=-{@`FY75X2Gc%6IEEZP^)Hv=%%$=xy;7FHSo&~Mo1&pw#Z8T0KFnyekzibH?D@q
zRNeEM1>qqvgo1Jc#ihWq<huhQZo2SLS|3x}-tVM-lT@+}wt|pVL_hdD&5ueiEP`af
z$~80A`>QE;xx7UYT*RAH<>ljN1a?h08Ux}Z>3z6eww;Gcj~~-YSv(R8>sLjxB)0RZ
zY7#x=Wq#wq!qP4%NC1~i1T6wqrg2dG>*#U?tf~Y<Vf2`>@m3QcJebhe@Dpu~!Xc{Z
z(P=H~8QyjW2Hin2W3GfFT)>5#Q2LEI@r}PYA$(#?1_0`R(u4yJS5)ca_j}c^kSum#
zGWkqG46i3A0Jazeh(PcHDhEzEu%7*{L<mpR(|WKW&%m_vgBsaCJWub&7YoXW{F9Eq
zdM~n?8XnkefdJA`qyeBqWe~H9fCotYUyS*%jsXGmx*aew`SkHa70wE^uk%=$**Q#+
zlYF7ch8iY2(WDt5yX05ad$TGcCBZnPh%vh-3WQiwJHAv=`e(YECGlZ4)t#EVo<YLa
zKEl=kwC1~f&qgh3Y<c6|_GqKW0>cU*>tn7S<)vOghycFh{p;`lTXbchaqsrK>J4e`
z?MdM!f566QWp(w@m6w;qRgJ{|gureYD5_wJdvv3h>#1i|f}wXS_pNPzy&)-xzexki
za++-#D2r;SHe8kLdz}Q0k_@r>S}V4({Y6wa)gE}I09Z&TrBnz;F5k-?;2C}^4!YG2
zzsUp-wSwvdu$YaxaqVYP0rOAGxnXpHUCYJbNCa37Cm_qJ#NxnKdB61sSZ!-4Kr$UO
zt%3g5B4%1C*B>1Oh|M(%l`aJ7sDuU|rUd>~2I3E!|C$!mOP+7qrn!le=%s<F*oEGF
z2M@`?Pox2E4GhD)+M%aNlNd|v!t^hIE1j?gJe2N=s5ILQYJ>Z^W$Le{r$A8@l5(_{
ziE!<A(y+pOr(pL(Ajn7}@SY8}11Ch<1!lP+KUqdyhk{i5M<Y$<5x9S$I`$m*N(#U{
zOAv{q_t+G7U$UfTl!1K!g(wlotpGoQCchTfnW2F8#>Q<7sMslb`rq`^nviVQR~cj{
zJFKb|(;~BPiwu6*t}W~vA-F_I!CEovoB4ae`?!HTTT=vW=lb!<i0+%V(S`1(G1&8W
zzZ7Cm(XrOW1b)lDG=D+)nC`TIDg9Hd#Qs;$6t}!^KpNLIqecu5L_314i?5e7ha67t
z-)?U{S)oKl@1_B-`Px+l_dk|6vjhNAbqRVxmtJHl8~@5RV|xre0I|Cle<03}{PyTa
zz2BF0w2~~}3JOF99BhYwLlG#09<XpWibD-Bej=`iF~*RXbw!7Z{&!>>>Lb8E43<#L
z63c(}Z+4@$CHnvI;_vpJ?%`FwLnNhQzdpUvv(+mE&7zSFuMKq5E^wR8m-&PQ(-B#2
z0Mn}w8CPg-v$Gv>=1O3S`!zhi5XD}(maZ*S@hVZ_h&If>Q+y;rftZ)lQUu8wMJO>S
zABCx@&Fg_{r1RN;qqV8$XPxqBjZI@yf1eJSjA>_Z+vwXxYvO3BgyM)58y<>sYeh}r
zblE9-CR0011C12y=!VBJ6A%k354ekrr*?;6h<k&TEv&fqF#=PW4BI*bSEF-v*RvWl
zH(<3%C$fyUPH4Ys*c#&CQbmGXfA3hxxH<j$R#Zbd8U^`870yq5#eNg;^Xk~-&&Y8p
z#dVpt@8-MFNufs>ECz0C>?#h{JXn}(CxM@6a+wIGw33`Y<Ee#kWfV_-49h|*LEMsF
zj?qnsutSvE1C^jv<_2{Rh$O*Xn)CY2<?L|5_?y0(+TiR2IXxn^jUtVYwO<xZE6$DV
z=SThCOz$UI{ECwq`aTey1$xW^$X9#VtD%!`1!;})+3;D7gHqQl;DP0MHzrABq;lV%
zg@S4Ur(rO(%uq!Q+oxI1G<9B<55G6y(e$Z_Kmd|6Si*q&?m@f$eeMCTowE-M;%4Bx
zdvW<ZuH01s7YxnCr6$D_R93ROWcqbF@biyg!ZJianp#sIq36W2S2H+DJcH{v^^%cj
z2rGMdGB97b`Sq-a`gMH2Ik>Obaz26i?may-_eQ<*Lz1;+sL75skbFTFF6=lY@Tr=4
z5|e2O-{@XUl_y4`L{5)@Qs@*i_=&ZLQS9g(RkcY8OaLZh90hw_hNAvWBpy|5QDo)f
z-8hn+(`PpPc>cArRRAep#mo48Uz_yVr#X~`G-er}h9HGC3G;dFP{&$FE%_C&9vntW
zWl&tVKdA6E8o8=4c2>lgad>_?_2x!<;W+2qw@mf)jYaEJu<(i5OV66V*1$gI-De8{
zo?zDHjn*(0%Q~7^1{N^op{d3ey?g^t&cDYI)^B?OKD(Lw?W-V~r|yPEmejo=@&oS8
z%zGy<bf{SaR`Rszjh&O$u))hue&z8#j+Jr0d2T(CFfW+;68G!nAn}`D=?DEX!EDbC
zCS%GSdEx*m(mOExM<ywJ;H%&Ahtx+sTs8r;E}lgr{qe2@Wbh=f{dKN-l@}e`yP}lE
zF=YdD+gB}>Ld!0{`D3M7f795K1|zz~A^E|`{7FyGD2PQdZNK=_0nS@Zz$&twETRHt
z@y)k=0Cyx-r{A(`m<e=L0Xc-Tfgz%JvKo$F8HbWG9!Aat(yK~+(tyM@wX{!>!Zay0
zm6XZVZCw?JgfkQ}D6B_%0-rAIPg!D^k0=VK!=>eWhaPBxazIcc{*=3}$bdTc>x-q{
z?X0@|8a@694c;g_eQI#7KlqV>O1rvmvSggadtw&HN_^uD=+=ga(!@29H*@rm0_@GE
z=l)sI9vDU4@$OCK9Xfriql$n1JE4z#JHra9w1g5B8pIALy?D*JiQ)C=$|+aiKX1<e
zklnWQT(l|5JVIsYjh9*QlUrSGH$Gg1^X%NVk9h3@uw#gO=M6nI7GATVRwu`H)U_sG
z_r1DNyxxm_i9l5Ts`avIl}{n}y=a5KJbfQ;o8oFYZb^@?lL}S5cFt`L54q<hQM<5Y
z!@-wT@AGmz)1<vtiAr<&VtK!zcSg0LW$UQlD*8|0C^OBlkoViW!f7EljEmzpGp1{O
zz-I6SpP`z<l1ZnG&vkpstImo|4=pqQUClC)<%Ob~lD4F##rxw39xj9vrJ=EhI$6i+
zNpkJV=*O6#6gD!rMoJW>>@;bTR{Nq@c^K1BH$pyR9f~mCu7Tp2Y_VY3;)F`aMcAiU
zM%I<ps0meil2PF*regc|klRwh5j^p33V3+LbLz3NpgWhBP$OraYPFDxgU>%hUB~I#
zJL_>_f)&$iXc1}&Dvz@I3=;yH6uiDXZ-`G_e=N5Tv;l_|fH$LHeG4>EZ$np1R`?Gz
z68`(9IDrDF5mGo)1op8iGPL<5R=uJK_p#^XgkCWXWwGR9^aw;GiZ-t&PNRwubezI{
zoeHo#rIen$pG=sS58)ar6(qDn$E8|85pl&4FdKLh@NRT9Mo0xGeF+<AEbmg7?2cC0
zBiQ`r=bHV-Aq{+f!TFrxUpa*<=yXj=f#j2{2F;aA>Q?BI{>bw7!OI8QbrsN$as}O&
z<y{{Os0j^LJ?LHQH;HxD9!Mo`&)~!rsUZ*2GhZm+0!Q_ikJLV03GbY~G|ZmQJOQkc
z0D4X2<3bTZu=sET`Lk+RjRfA-ur%^yMin&qUaRiE?5k+;{kAYBC7G7EXJD%aDBG^&
z1xhVgeG>;ODIc2#W+);BHh+;fp${6Hr3m1tZ5z3QhkOkIWhz-6@5C#mAG3(JzF77k
z*9m(Pi!N|;e*7OB9}Qu0D0s0dvEC_1^=-$0r`KEVc24UWJ{Xjd|9N}$$E)~`^#_gl
zuSWnX(Q(Qe1p_PjcRo2=5#GG}Z4Ixo$PC{UneK9cJlf!L%7XYt-WMDLBUxXnO-Y~Q
zbOXPtD;EMS_=@Vw+RkR~XP`+(OzIA{u{E-t1=^_gWT!(SS8ZgBQ+$_=b5~uv&%d6F
z=Qeo)kGOTCRbCO_lcqMgjMVFJUDb<W-FPZ5+()H904VE8LrcGaN@Lv|a1D~eS3c&8
zfPS9DVOCJIS~Pp#DeK$A6@tMe0dDytPz6;@6LGDknXKoSF5oC-(`}JgA@6Q8!vXBE
zn2Dpd_I&Riyi}d1tNc-Ufv@VYL@g?$vhh0xlU~oU`ezl8m3aodw_OHD+BZGO`A$Oc
zFAe*acQmqB<zFRr>CZ292DupenfkoK1;fho+2_TukTu&J_j+!=kj1L0jpb9;8}IJT
zJtE5YxXf?1XrCB59~9Cf%vbuZcS`zx6JXYf674s)yfWS(*kCVt7JcQI9b4C>Fn^6_
z{02z@NKhSyYnayBR<5(-jn0;Vc@)QB8+-m2Q@dualIj6F!q`oma1@n!4=2auew%p-
z9N7867}?fW)67iNi(xKNz@~ttVWl9a1HhviF{x`|>nhWu97QAgd56cA*b=8j&DTHs
zeq5fWf`Ub90mtlM*M9v;=rG<w<J8LWIZi($L|zB?J3be;C;+C%<jF(T2ZHSh2O&C{
ztndEPzt$gf^(10l5gtaRH2ZbjCPQ762#ZK*0vQJT7ohjlG@&3?UZJ1@^avEJDE%(`
z?)*!6u<NUUoQU5}vIQWXS3(+9iYYWyEp41~Qu463)$~;s-m6AvbVxpmZCr~=lWg;(
zBpkG8j!b<N#V=<}icT^-5Y%=h5fU8liLTBPe}G7&^ub*f?w$C~&o8e|egXnSNbubk
z$QGUCVTs^ZGGdf>`)k~3FIy9JTx43zhs-ni%#@^Df!EK&-mO;`^{&)_`sw7u>F*b}
z=h=N~mTg|nArzY)hKF~5Ytn3?d=A!ATig|*XoOCs5JZoXiV8t~K{`unat#~ub`4M(
zCKYKeFK~TbRy6^m)P;xU4PeG<rKAW(vJHd3Rc3B0H6-fZ1Q`)4afZf(r?Q7~o~>;y
zZ5}>XAi7Mq%F~S0|J@RAwA&6EFiSq>AXezCA&l{Do91kg@`6>H&i<rN*hw(_`RsCH
zOLp1hw65*P%G#f|Iq(UC!-aM%Sh7@@Zl`LlK!8l9&n*8B4<SjN&&yWZ5XWM&5bD3j
zlnTqWH%kil5gb2hLbFDgj3v?!dwiI|rQngXK!bDR2y7%GiJ<R0<pXxv!n_1Jc_zfV
z*!~++a_j5HzULQfv6?degJ#1^$M10FzW!Lc{fKj4V&im`y}X<^x?5O?)9q~-1|Z4M
zJ>r_IBD_7fKEp0(AFI-eY_;un>ss<42l6-t1+DQY-cifa4r)kp80Fk6-a2nh4f?p0
z%Abjz1e0p(9HOv{G=EazTtC{MGYKGd6(&pz#r4n9W#i8t24@m91L8Tm_Jui~dtcdm
z<@(8+XQy=@_X7UmXH|iIbsZgW$X)3OPN_EmrMrM*^kDS^NX9oXtaR-WV*M)E?`dcn
z)ahdyl<wA7N6!86rj3!>LH0`R5+;={xxd3XX-!MEjo=s%_s<K>dr0wHr*9(?-x2)}
z4`$JbKJM&Dxlw;*+!<`|`75qJgkUl&*YAC}@PM~X>p-DXR?-^FzVP%);1mugVRClh
zXic*@kqjf9?q>F8hVSiO0{%SpB-=Wj!81cC%XeG(0{~yOeuNs@pWDK<f1kA4S48=A
zZ5<Jkls5<yf^~e+u-K?-Xyk;{(qi;Eq?tk0?!k=#&1Kvo^4*)?N8O?1VqFP<rUr0k
z!MlWW9Os7?)JN;BxD`)S)5F+{ncOYM*W0FWj(L`&#V`}*-wFV&&5wIyf#F}B--4o%
zew~bS&clOZZs~DBsntv_kF5q#?k7rKmo^*%3{CpB)5Sbzu+5_8`zmPS_}u-2J$$-7
z!WZD%1CW;!q(~Zad*$8pgBRMV&a@G(uWtzbjL8tX{948$Sw}x-{o)E41vS@ir^x9H
zx>S!jw2an|G-3;y;FIIUl#GYh>KVs7%PH1xWeGNjNhIL9A+}Q5(6`|jQrvk4WYMYQ
zMx+R<rJgyF(~yIv4-%J4kbq-~HzsROOY0r?h{Wj;&KmrJ3qD0dU%i(TR^_GBt-r&9
zc{nfJ<I9dJa>XRI+n)<8_`UWh&utl7{+LhIER|al?Fkxq%JWci2gjqo-3210|MmsN
z7U5y_xX`=(WjzfrGFUgL;CA&yIH2hD>{l{&ei9zo8!h(Jwcqsttno~Am1o{+^5e#a
z5B>Ac6p~=f!=@#UAQIlIG2cBw230_-_YOc;T$mcH`f46>L)y>TMh+q4WxGE@k&R9O
zy)TJisATh3&`<eJnAga!9a>kGr$sH@H>L*4Q<4HHz{8#LuY9OCaZ()gnTrycGC6Cp
zd_vMDwsK$l6j>7N={LM;$+?w~mlaShyfqEddIlbWy7GUw@F~kwV;1$sfG>ds@+TX(
z2xAuklJRKplD$(P1)VU|pqd;6&V@&SK&r7^Dn%vg@H%f&&_17CF=QJ|s>-bvI!ni4
z3jAbc2QnF=<s;i#OOCz%;6|+%hSjeYWoQXjIA(SnVH}J@d?m^>vbrvwRNMpJrtb!q
zVeM1e8F7Q+5VifLJQJGYnFdb2WHqHnTJoLGyVLlG-CKmac(rtB$dj~I;8d@5mcm7_
z3qFL+;)HQ%lSW`FCq|T*`&Zf5jJvuJ=55Dt-gwTBjcH*p2nm|D;X-&f+PwKd^$fYa
z=%y%F6GtM2i+@KR+wzw#+%U8oc;f}+c(fvOZ?1_DTv;orsaWfhA1wuN{nrbaCot`r
zf(keZt|{S={i_$i57heN@yI-SAiSI#lr@N1l-Rpg?3)+uAo4e=%2*$)riB%Ns-nOC
zK_vro!r9UGlM2DX^6h2GBGs_Ng&{gnFQF)y{Lbii<@|BO^tpU=kT)9?!xlH*QQ+sE
z4<HmFU5=LYgYT|#{SS8C76MM+8{oCeKJiOwez&iIWZ4|amYQtiqyGa~M|b*fR(6{f
zc3#7xG`brJ{m^+cRB^!}(fC~Q>(z$N2wD}%vtA51m#6@Q$<Ept0ygFU!i4W=0V%-y
zRBr1py<UZT+iLZTgCZ&p|D;2EZaPf`77&EEox%YgnE^awv7E!6KNui~yBtf-{X~U*
zC<x!k6$12IinG3yf)LCL+Yz1FjRm$P0K@*FO>by#Vq<pRyihqqa|xR!%xMx^^P){4
zkI2>}V}3dS0ms(_D!PO->pi<VQ<igyvL*urmMH0Z<)M;gxfaT1@?4B)(d|ogp6%Qp
z9{|bzs=w(wP26$d+FKG%^3uCTHXvtd3lEf@&eiMW_b*Q+Q`N^=e}WJ6V5f3MbG_g;
z1%@!kp7p~rfI?dL;+W9|;w|kb*bk*W1G4rBinFKSNu`Crk+;DVnvJ8Zu@?DyG(p^#
zoZ#G6-OjwLY%HDykSCxTRc@>fepWzfe`B0Fs)b#RhAEon9Gdg?vMcVL-6xWH83n7+
zBOy{?tfLS~cAQ-bJQY<GJm!??2O&h!c(1dCh-uZ0m=?Y6??8Gmi1@G1Kgixryn1}8
zs*IejPg<7+Pyo(x+S;Ff<^!kpcr4D>+d<sBzl|@D_T+uX5vO6HMJ2woH_<I`1vb5y
zLjDLU#>5FA&<6Z7d!zElNzP4`ue!*`HKuj#=61Zl1%;>#cTbUm=`@2t;<CB(!m{XK
zNk2O8i8?>FjW6V8{I3LA;^JlSlNO~cic$dwa6#icA^{(jFTC$m;+-Ee#fRMHsm-Z;
zs`Id%$8{8v;b=KP^VY1_+CjTKkGn}xPcX4z+EA2{)MY8<NG!CZsij-C%ME8QbWg+;
z$0socxkGV$AV~IPL>9|g%L}YDF-`DlAv_K|v{6pQlubIs_qE84NThT}%y=Id8^U{w
zx@~22<Ql?UFs|f*P;a)@^EpbgkGRPaQNGg4$5``lCCeKY)zI?zuZebIC1D=-oE<_*
z(iGj)roXuo<YV>T$K2dg9JgWWm~w;}rL~4@J!R1y&u@W{L>6DcjUfe88O39Dp057d
zhwt8pCQ$?WEdwzdonrnU>i$<cq=1iZn~<l~OASxbE6=3N_*Sc6eOO}g@e4AzXy9{F
z#5k=tZn$}(lyg5`jj$!P1p^r%<l1$awp#MxnZ4&HZcN?8F6MZSqn?pW#!|-x5i94K
z9@EP6+#NQ%hB9+YcUupQyAU$|Y@9s@UP^R<yZ89=F9kXTa!c8~1)6VVBj9NI+kRFw
znT;ieAFn+g+jz@^Tnb^Pi{|c-(gI*ET?$(Njhb^1jRGSHKt#iSDDxI2Eq%#nSJyE4
zjQ33TKfkwq7)aDX&XAF@L{60$!9{=Yke6$RyN{HEGK&H~2yJPm0x8&W`jZfe8<jE{
z-Ara5e-cVMd8zRR_nBsalWj~8sjTL>0IqIwgZIv)2jTNfGprXx5NhP$pPW9Kp-w?i
z9fxzpoQsby_%B8GMKk`%?V0_X)Z^Qd=~<w)2L~KHx*fZ_JHNYK{J&#JJR8mwh?>X&
z<)AtrZGGBy)lyi*{NioXmV4NnSd%SS0Bp;5wu4W)a6glq`vX=5y$aREzk?4f<z-hB
zN+16>ydb{(w|WSWV-2S*TP<Dr2rvIkhvJMEHL1V*i=uSYmJ41GM>*L6U^Op3w(NGt
z7tuE}RL8eNbV}OmM@fq6op-mPbRKXk5RD&zgeNCVB3>Ck{A{nEBJfQ8<k=5tdtXv0
zfIzInQ>V~y_Q2x)6D#6+RqU^0YS@47ZUr&2&QK|UtywEr&OL3bi&|?<>~n7YwYuKL
z4siiR5MZ!Yo;A@3ML$u-jKUn@ZUCnoxYCWoRy3y9JA=BG1-j#pYv{K66+uslkt|qY
zR7)Jm^DW;_X{Aqv4VZ&=>+i}=*&`R{+;sf-{4bA|_<(M-d5*S;>RyM3GdwFE3Fi4P
zYeYJtx%^>AeI9#=>j+e#ANa@1A!}5vw%?%N5*{-zNhUS4EP>NJfmyVGg51z<L)R%S
z%>SF>m8S8`lbZSCtx@w%fqW;L+LGV4Qpw(%I^oWZzbx~dveOlvvfl%<*f$Ot6`2)a
z3!f?wedoh}Tk-V*kb{5#HS<JPyi%voaku4MBQ-wY@<Xz`G_qhO3|C_yyn_3XrXuIR
z{#WdW%U7u&+5!@`P^B3<VAF_2r2Tt}&I41cz>2Q1xD4_?%OY^(tvP}GoaNrBU#Ac+
z=38wSSnSuU>}iox63~vl;9Y)Gw|%!n(|qqbUe`F~-xx9T#GQx8RRCLxg6gL~jG>hU
zyVYAMCza_dS!5~c`_}5xZv;m)8E-n$h7><k@7yZ@wRlmIADo{2w5eL6?rfwkq>TgO
zq!ocU#@_<iI4LMcSoJp|%O9vKTh-_A<`PXoqmy)PS-lx)x4^%&GJ5k7dY;pQP30<O
zn3bsy&E+UJV=$L-lqa$np5acWKuL;Nhm`LtlPfh8Gz}<pF^%AB$)k+;E43N9MRMR-
zO6E`noy_NSvi}vG^<k(dB~xHNcpj^1^*Rg|xvEi*t^Hb8i&PstKTuF>a+VZqusAf(
zJe<tp8Bj|YAm;ilK+)7I1c^~l(68(5(EHCz$w(%dR7vXfYo>vowE91kGfm%KPjzJf
z#qD?TYaB7pRGzqwEj@lLp#mZe4D2Hi<w2D6lm*vY`<7!v5{1ats?m|LLJ5+2&1toU
zHjR9zo!<oaQ3BSG0=9>(n?OlITJnxeliDE%&F0sRBx5ze^cDu@6wnyyI1YL|kACkx
zlv4LHX&vcTz4#tWZ=7;?(fGXHhkk#?4%EHoERs#;@eFwNH4*{wq~E#qZD;B^x&aO3
z_UODpEZdQH3twTV+s=?er}Iz(1thN^uCA^~hKzXAcsH*g3fjkK;Y2&yP`fpe+t=bI
zAdAEUIeA6F;CF9~p(=PzryJi^3Ah#HqLT_T(zpa+v0d{0pt*E(U%65+9dvOd05Qu}
zg!)m1(A_Lw;B|f5!g~x#Hy-j=Ydph<QmT^;-d)1v>9|cL$z@SDLJX**9?D?S)zc~4
zi!!78hSM|8n+E5^nh{_R={HNi5D+Mv)3Xw^raT>~KQP1C2MgfQ{Wx$O8M6rZ+qAw8
zpyxkgUUr$(F0T;qaw+o<;~qH1lt)r}<31`AnxVpxtCrT=hIneTD)Eh*;%UCcp||Do
z@&j}LhdRgiHEGG-F+~H0Z;D+#zo@9I@fP5LKFTf*(m7rMBsDo`OV+PbNs6SqlT=<(
z_t3WTSvoAz^%1WZVwH2};-2yv?Rl=7O4H95(~_1$;ePeqe2vAy({aR8+I2p&4=~f~
zo>2xDGdU4Ssm3T<;Ktw`WRbCC5JWM>4S(#Rp4Qp%a_yab7uMpz^E!#;w)?)~Ojxht
z0}^WFi2l%u3++-LN`FAHQV?!o`CQ2^EG#dIZay|y#8p3Q8)c>bw)e$1_rPD?`|H@{
z+|9H~A(s|4mfDXIfz=?iCZuZRaPFsqHeEk>HcqN%{}a|A%ED(r7yXp`?N5hUQWou}
zsHo3)5SVfYuU1p}$d`3c1C|ya1YNy5&8#@ye<NyIV+pp4KkW=CGX}%=4AxW@Nr(;o
zax;o1cmz=`OD`xyug_vNp-}Kp)w9+AS8uti4E74|eCR<rGQ%*)#ni$srzlh#yhUXb
zO-d}Qf7ES{lAxgLHq^=2<Akq%p_S^L4#(f+K9#E;#O7vd9C^2xQ~K@B_}=a9#|^Vp
z?^Q1qb<iqSX-i3+%+{rbCFbnBrd#C<_S@OpM^@@8304;AGo;pVA~|S!gpZa*Xf+I|
z7{;*nZAHS&FJY06pFciVyRUF9M;iD-HH~Af8al|gQN?4%psl}F`>GiF6kr&KyY)^6
zQJ^+TQpW<oU(a4G<+bR&@*z|P^>VaQdFC|WMgf%*PREyZGpstghDsqx4V7blw+I(H
z0#ws8-U*rRy4E~?qMEF#RPYR))}?MSuaVavrHwNO=6P;TKF|R;pQoBd21Ow9+@PfL
z9!#DHCeDyr=m>sO+=$Gq1-&&vYm6|N16-S!d7_2w=Z-#UV@SC?Xe920GJjKGq{8L@
z1ld6b!8pOLL{izhm3?Gh_fhX0-NU6D{jQ~sFRT=)hDoJhpTKjX(d+-MKfc7NFa##w
zUoAZ}(O9|U(q_>09ExMyIlM}Bj)OtPVx%agV@Zwfm9hu=$C?*4?}7O%QbP>{su*?U
zV2cNyQspq=5X49<Nqoy=c^xV4g7we7XJ-k`tyJ;3_Q#ja%nPpNktto#;UqigL3bbX
z5l@G<^GJ7JSL=TJK6jNu#wb{}tENzi($y_&ExF?SASos~8q~UVQCnfhJAd)Ibm{GZ
zheL<q#J2rx#ho_frdu4|iTo#64p;Z5o7s{EZ^!x+-C^2R%@VRmOQgLhvOKRCPpx|D
zZNq%#cy7-Odhjw%3N-z-Xszb&Hy-r4EqryvUb5g@UVU;W{0<<A!vikDZR=;n@2-yl
zVYaHS?(A(#_p~Ees!_5soxB|G&i?Jf^fvu96KbZKmVVZ*dWrfsfuG7IuYbzKF$6f#
zJPlTjM^`QAc^_#63g}fL+?3-A!OEN1Qe?mJJCJeQue!zDH`(#lL|Rr2Gy6urcsp6Q
zJ1=e#cVj>Cwh6t`v9p*`@oL@#)}Duu*8>!d*{DqIdxet#^O}yu`*f0Y_OURQa^g=1
z5wN=7Rx+$DFKZC5o9g_brNVK~J|%K)!Ipu`>sg;kjbqh9ZP6@sH7r3$P_yd?>Dw@R
zu%K+5&m@6jQ#3taJZ$IN7oUEe$nw|t+X9b?nyH4&>%p56L3({>h_y)J)Yhv+7N|VW
zU*K>~_hH^6^3~U-j>sFdA*&cv?GD`odO4xyYC$mj=o}4SG#}Tq2Jr@U>i|l(<L~B;
zL2CHV^ZrRqR(+uUEW5z|9+)R}AN5?kLMm#>%OtU^2w0OsK@H!$2BJ9Ap^lE+4)dEI
zU$l)r*n0e@O`vTS05$f3bTc6smDr;&jgSbZG!Mh<ID5A{g@1WqG=~SQj|Afr9k3H1
zfqZzcYi#Tu#=U#;(yz5Z3_+pG449u*K^u!93l%B6WAXMPTJnvh9T*EZ6}h4X)zxl~
zjvt&@F}fnnwGMT_IXowvPF%o_z*u~{2=N^Hj}0_EnG35vtR|3BpRq}wqrm}YGF9Il
zA~<dn(@C+d4(c=HK^JHlmJtv5zVnMCKG*qpMhA!;-Y%`|PCw!<#1`XDV*!q2ck?W7
zN5|0vMvaauAKZ%aWt9TE7TrhI4v&0A<3>WU6LUog`3E(+W3zW7BzKicGS*^BQULm$
z7Z386e>{=;veV|Y*NT1EZ9^Rl&|rv~=Cu*+B+S~h#C9gvH%Ek_9<Q2CYb9Dr*Oj7y
zT@MhkTfL*C<+E_*RZ_@+$q%fe?EXQ>wA9X0ct#S*`l~Uy$^N_WPDvsysKiqYaxp35
zv*r3k`I_mdTN`c$L4OD-xL2)Y%+NdCw6#WN`P!MBKiaLSycbCKO@fKyW^5>*R0>RU
zK>|ny8L2T6+5#;Qir&_aTVYcvA=n@V9bxvF4Pi1G>szeeud~1kkn+QN6WloBqyWj>
zdkhW<l|o)_x>ynFzO*K#E;Ti>K=jYwGl0A{(!P2bjH|BoP%65}CC^~>S8+5GJJ2_X
z`sg2!xT<opu`8%!O)C>l>*%>T{?KOyOB)aJ<G)o)N9stWP6FkRUM61#?w5Gh{UY*7
zPDqVM6Ic;@8?U5E>L;X{LAgb<AmOVKbtUc<n|Yjkul0G3n12{_TDbEZxe~_Syl!9r
zC<-2O({Wbd!WpF1{pE`TVS-*!riUGC^Zoc6Q1Q2RJrJ&&76t@t6B@jFRo?_GYfo#+
z%Bd=mj4)b)`=I_~B5BBN-prb(&~n$|kyHy?nTs<M?xqJv29r9HQ6A3`^g~aHD(e=q
z75`#XU{^@%qY{)_9&IttzFQT_zH7=lf77~v#5^cSHC~Z=^7nD9OfYD0bXsT`2*szk
z@Cn%8ud1)_8@-tu^(*5(^fYolWmwoA7Ri>ORgz39Wf;OfaD#jT6u&roo_Oh)SQxHJ
zF{}f8_`eG=%V{@(|IUZ|U1JNIs1#_7Xl@lT&Kj>2jO%4QW&@JV7(MVa)Uq|!Y4uH5
z`0pK`H_V@lOxIoR$?C8}sf<20xbGbPoPB|%Wz?x<NuUhK-gUpYhCH*I{o-}wrB|dg
zQVrj@QjKW-H4cJBR{m)!I9lEvf@#r>EQqs?gn$7DkJVH%3)gRzW`GhPD|CAJ(laD8
z*5|e_X_etJ8<N9IMw?>du5H2QpP`73ai|5b4Db<x`5Mq_fPU?=XcwH9I+h97wpKN3
zedeTdwfZ5Ih&6>pylHC9a8@M0c|Z1kwPC{Vin8y*Ss}il%WPcQL;uN7yZ;!DcMfBO
zhZ<)*w;MzO>92_~Dt7&@|EB?xCHd8yg;Vyo?JL|c7zoUUa190pyGmz}b^f%rS!u0<
zO&FRyAMvEC6J;t`>mA7A-j5f_kbcE3`oUk4J+KHsX&mY}_Ld7+Qp?SGVl{|*I|YqN
zS!{LGQnWH?zPMpVWg2KWDy0necpqDg@Xy|V`Odn%v3z)s&7#E7d>e1SwdIvFwfm}9
z?}%{wIM#O|CB@r0^OEa1iZfS_bRa4?ei<<vHfgmbUKgpLkMLFF9%-o6TXkAA-(tC*
zaDTw`azxGBt7KrWz0lEqj?OElOGxwfb)U}VR`TY!XZm{6fLPrxf}4==1y+zTF{k+O
z^?I&NZX&|nO1sd|V+2l74*E2d4qNShQk#MW6+y&NmZDMWDe~T@eS-U7!7)CJj`(oR
ztYe%KbCX8%l;0`m{Rx_Qm1EJ9o)ybBvOobQq@*7Q$_c?(&>cSzbwserbZd0|ZhHLZ
z)(kj#5Gj#zgT)CgLTNQZ1P_xjwLgZi>S*$4!+EYrBhiizhqLo*e({W|an`GB%<Xvt
zuK*Lp`^M78`Y#oAi?_psM@7qJ5GgS&A&4Yx91>olN69hX(3!AI&;>u%T`TiZT^m~&
zXGjgW>%YW3>-_?kpPs4`{3txUeM#Jx{{c)1b9<K{Q9G=~tNpQ*l${fZ`It|qw7MrO
zQChy`mbl>Sm^h3>C%f4j*$4K7sM>bkwHoEc&`@eY#dYJw)zDyeK9)Exl1&DAh=F7D
zE~g-PC^YZ811U3M?Gr*iVM8uyY<C+}nGeAlaDZ<_n0w|4PD%vP-Wkt-Fh%{3ihm>V
z>A(eBsWYaqKD@<Om_-GSu~@6EL;c)Qb{&k+tV(!b|4g^590c3!G>hHbk5#jX{HVZm
zdZCjEc!<Ngs`AG%s1!XH&{{dN*WUGAkQ%LEOqVHfH$u4J+BQ7u?-X7eCHS2*@{RxP
zJx!9dtUqm|m*o*biq)8!x^gv1_MBX@$&~Og;HCxvb$p}DNpDaR(v1N1U!`AY_Y6vy
znLSRuF{Y9BY*DZn`0LK-a7{z0^~Jk4ZI}NiOh5A5k`h0gF~80aIEnu}1$>Vy(S{8!
zf&c2#eD}}&xy8i9-s~#+8ciF#i1>QCgZuUDh}TxVa*4W@mTvm7eqp&i&w&|exQ+It
zQ$~ViU1_~Ti*SJxhwt`6$QSfLFpY8@r-x)#R9Z$@{y&QCTyw;5-B_04j%dq!aj8TW
z;};-3;SomgIjZx<-V+K95&+H1qirct5|#ywqGoaIn&gl-zo@`B36w3#Sq_*@uKCWc
zTafc);QP;Z(L5v2UXh$$_Q(M>m&oGB6fFL*w@0=*m1;tI#p$r`(oaXC-PiCiefS6g
z_+O5XjB|9k;}j>;@j-#xpFfZ8@14ECIgZH7MR7p(8$s!lEnr~_=Y+KQ^T{Um$VnkB
z!`kkZa0wC6&pN0hm-<A2*qo1$n;-MKHYAZpdLNo}&bP%A09(v<1CA88=~my0$Sk?1
z)ZAPK^6kcRY+gK?_qct@ALdQlN7&lqL197vQY~l3?EnmAYFBI2@c=84q6pO3Cd>Uc
z)=vMEcaf!PP0VNJ%NGx~Acr)B^#UW@W%-cCyhW>HB1Hrs4x_Ms#j(L!*_gQ%E#y0Y
ze5{dFO|<xA7eWf9s-Pe7TDKV#SBjiSsmj%0GR!fc%LLUYVx#q!^`lQF&dq{^Uu(ke
z`3!#t0g3Fu2T9#NT%OxniSpp*O4-2j;=8gk>0fVoO8oN^u&*S=eNK%7sBm*T!{cP>
zH&v?e!%L}DR;eNK>#upaF18)(hsShpab7<9stnp=dx5!9LUn%K1lFco$12KPUX(xq
zG0jNc5RvB+c4x_oy$0;MwH~HPgf_xHz1G1fZCSy~tE^+aKvYoi1wX^hYlm+_ItdRc
zuf_$SG45IjGjf)ew|wTmtv~ZI@l`1-riBD4{Kt-&3)W!REXG-h6hNCVu(iJckt)fW
z_fJ4d72iB@lXIwZJq!P`2R72r=i)&%q!m)9tEp!fv6h&T#}~^-)Q2(2z&HTXaGi$&
zP!)b)2mep{V`z<M9-3L6Po9<G+v}uNS*l4pb3-HxoosiXZ8gs%n_CZNDc<na^bY6&
zCeS9P<I?Pj@7V;5FcOL7?LUmo`S*;FYzm2gN!CR`<L@qqdKO4tnEo!Bo4U`9w)$*n
z=+-#q3d#z>N;SPFC?IVObKBFqV|^v4YlLTMiiB}<YX%3wD2lXX7md|L!l}oW*_h<8
zQv7M5YmFwf)Wwon_~<wFXwCW)4k05<+_*ks4Z)EL-Hv-D%F&#Pf`Iwh*veZ$J3@Jh
zYnri4jc05K3QGwMHvcNbZoXq>;{)cosuu?cdwp@&yq?s(SdsLQtP%K;wtQqrcYdZa
ztGbS*>JWH7cYz&uk=T8ZqRsh998vZK*<;tW@{<mu_t8XFT>f#OJwAA(`+7TnGgB5`
zr9lr4z|f^JKBh?+5EfLshjJ)<`agpj;YH5|Z@#`CWL#XL4xWZUEcN~aY&^&`j_ZoM
z!l!Lqn>V@@^#%7F1gQT_VumeVhn#rr^YcdK#;ppZaOdsI+ii)P!t0s%8_K6opNd}}
zbX|9=USp|>ALhPfNFE4za+&yKIW^!il`2-o!O`)sD&8+9-gk}JdFK?Y(yp_oRT80I
zUils1QQXLdmGe65a8^hq9h=JjS+ejmSEbMrMK0jup;cyj=~ps3#gf{S@vz|K>^iGR
z6v5^%BJL;3fM{h{BtRK07z|<(gD^RH{KP717M236{pRm^+gm`=6(CI*KMfnU@6+Uh
zHP9+yp2?yFvnj@E$>@QQDTY@CXzgW(e-FW$^<Xv2eu&7HOLEX&`@5zyIUhd1mAcGh
zYXI}*91;XhqA?ypv}q@goJdYcmJ;y9TQ%m@=+)T#5~ZLp5{*+q;N2g#m*poaTt*0~
zYmy!s){TIP9ki9q=W)ky3=!eSu~*^-n!G3zfF()BNWcBdxXQCJ5j12qyWv9ZcHGUZ
z0Iv63m0a+^H=+Ib{@LIY6_2hJo%RGg7Vorb{*#~d7hvx$JDFyiGFI&YRPz}dgHIw%
zRV>_kPHCT`V$gCQBEH+rQUhOMwN{c9i)>9qa?uuPG|V!cZ5Twxl3?RsJh`}j=-Ec>
zCln3*j>XzCK*A|ptE^lEA{*vO0#XZxYU%FRC@88y8x@S%8eCTnL&c<2HcSh`lQ^yG
zmgM*1vCb>3cEtVLhG7ccwgKNdpxtVmIy8`gzsqF)%ZgWeQ6a^}z=qM&7-z)zOaL4?
zQP6vfk4gm_yjK_%S?+fCok%RPGceC9zPoyKP||3peL7HnY(h&$PV+oq{e}Wu>*Jm3
z8mgi9iVU*rW$zv=&R{Pc`!G)^d4zX2@S1Hs?M4LZdrDX!SWA|+;RvR$#1CvgW8TZ%
zT<OhvcoK>3$L!HSepen>w-@X5THa#P?&j@&I6Xn?q1~(Yk$<4KU3F&9hoB`y1v^<H
zy^l5qa|Mrbo~P#+WPuZN0Wn%Vix-ct8fCKb6s&C!NEZPr1kzTYdb5c<qu50>RwO$d
zsx3RF2Cu6s0Li43Re7xfy?aSo@`4uBbt*$YLcO8nd@0B>)6aPSzd1|;xlZLIpUQ%V
zH9H#L^JSh-OAD~i`NC?_1RhMQjccd3U}+;mL>P(lowXD7oM&qtgHdeTKM1(Egn<>U
z)NWM>pWpap*1_O%47P5Upx{|JM)+g|1M$zM4A3L{DzTtAjcnB$kz|9Q2JQ>siRqYm
z%Kb_^l+l#>6K3uF7-+R3;U;O3lIr5M43eW<9MnKraeRl-5vFI90ZOEmHo@m0ERENQ
z+noD`ns9-|p#0TBXOkYS-at75K=A=T+4uOIQT@WwcS9#WqsY{T@2V?7U?|Zdk`rIh
z_+^gSXS^p3P3`VtBr2O2ED-gKL^R8;W3oG4qImUB*9KQU>oQw6+q^J)j+NtFnTD{=
z(DJVAw(uvLfOiNCejrV)l#eM(vG_oW1m4zulk(jomTma5vz54p$Fj(DtXg2)NRad`
z4#tpYZ~=NiW(d^MLRhee2=7Y_3TI#e+e9>B@S6+H{oDT;+Q?DGrj*~TY8KCr4eFh$
zDk~3=IlB}AhZL8Y0msfqlM*+RExShhE9OtGC>L-4ERGCk3Ly*)ul>H}LUEtm9)yGc
zto_rV?CtG;@9j-EJil|l%aU;O+dYK?n+x!xUm@+g5d#&izTHFN{Pe~Z$7`YF@4_)d
zn)wed=CU!4)gq+te))_B!+_nL>OrO0L+SA!0_kv2kvLyy)h}N4(_-oAKZ3wijZXtw
z!AO@K&qhy_kYvHHs0*xfj7R&PV-4*{POnFU_Cdq!!uki_I%CS>y6TM$J{hTROZe4<
zmk$6ZEpTWCp5mx3QjNbAxc+3(YRdP}eD6eDiz=xUO95b#FOl}qhHuLGPt4mYUn~{9
zmU~?IT{$?mZorT8lwWvTW)@{-7b4$#>|+j=RNX}`dsEM1QEJ-_)c$G5%|h>Eg0R?Z
zMR5X-@&rdb9l>nsL-zIHR=WoB_`%$tE4J!xU;~&!b}aC$>5NJxLUm}!oKS3$i*k2t
zOMz3Y_R7(Kcw*f7esaYIau`WLMD-q1i-U(uq>1CW!yezzpgi}_918=d+jFtEr~52|
zQo=nJOD0TNlVKbn5={Ak7tNG4R#*NoU-_~5E?#OXV005a*C!&>EZFQ3Zx$o}J5<*r
zHu@!9mi<kW(8^xB6cUS|-hjVSkW7k%$U;$<@;&j_(_bHS<5?|2Lkc)c>tW?wQ=btT
zGwDy6jz+&r>sqxsR;f*1th6$&GQc?s?36wcZ?Zff37i<Ph_sIe043fN4Um5X#$T%p
zO@2WS(!@eTB!ORE^LyszTC2}79Db9_!caf=3kd<ZS(z<_l?gd0s3qgyX?9~jbbGVE
zd;X)fZ3|`{y46zp#gTsA&a@a$Jp%-aRyYu7tEiTJfT_jDhFH-;zXiuIkQLE6yUZvV
zIWPev7$n8oD!z$*1f!V1Rhv8v0!NM&kC&FQZ#B7X)CF9+fHA7EdJ7Zs^cd`L=TTv0
zH4Eh$ZY^mAGTvzX5-QZ^-=pSMuD#m=Yl`Q+pWyPdKWu;EG;yk{t=MpgfVOl=`2qbQ
zivr8Xtlolnq<_0?(_Dpf@RVx`Ozfy{8ii%;&@w~356#C<$MNOx7!SQo%0?RNq%t<=
zeqa4QdvPt@-S^S*^(e{vRja9WrRtzKbGPz*9wH6kSQI)20Zwbm?m6KMlr7&g1CPhg
zl-JC@UTI{~$ZGDLiSHQ{9bbeJ0oIw*=&A?nm`DakS^$tP|C+_VP6J+<oLL&p6cd`x
z&y0n9`!hD~ZT5!~%PD^XO>mv+JbO)p47Y_akanyHjAI_tzT=Tf(`Sn7hO=mC%C1k<
zpV&f1%Glu4nxva8!s$$m<E{dOlI>T@MrO{mjImO@uC{3acC5Z;sP`Fz<to22>Xh^%
zeRP2iC1)Z?3Ia@H$7V{(mcRR;jWgdrM@goAgf)p?N##D!v7=RG&iP4&Qs4s`J^g~s
zLHRKqu*w|(OEju<X-6aW(Hnz}<8zRuOZ!i+NpgDp@_IgSiFb#il?f;($s6o+vdr@C
zc$x`pfcqH6RV769@Hl9<M7hnephL6%0Z16fzyixkz>Xb^;CD0S_nj3V9B|*QfBohn
zyq4;}ooZ7*+j26Y$QIT4F5vv#@$LE7+<3p==FWc`7f#FH9bT-)A16y(Cf@_<YHu%&
z#6_?~|Kg=sz~=eQt?J#0>an@RnfY9U>+1R8gT>$POSDgJ*-Rj4prS7HZ?6BtQLT2O
zhD>r?oVRMUdO|JX9W$yxtd4hjE|QK%mrY>_j6t%PafQ6iT_2jt_|jxU<U7P6M#;7*
zF6EAsm0A;$l8K?gJA)e8=x5zMa*$%^iQwJJ_Iyp^lHL4TGd&`I)lU4Y7|S4^Q;rv{
zvXloRhJ#RfS=U&&dqS)2^cNF0Ho*O>3^PRVdmuKlDSr9xPQjIwH#CKO#NC~hQq}?g
z<GI*#14kNoB3yQooItvDLJ&;w=4*WoVUk@xLgmEe1~}`cwqfL}$fnoRg5D;r9SZYn
zJzc<Qo6wUHXm7)`|05c_m~+pei32m@J|tP_<Gme8+H&@w0SZ=;u;PQU_n@=qCmj}P
zMAdj1Nqi2)q^rksF>+fSz7unh**jg=C!?S+*ZMUiWHtlr4od*o+r7ueFd7=n{VW7o
zArY?z0u%QC4Rx{z#l!2(Is`u>m?-@UOqX>N-`bJcYNxpHfhIm_TGjh21kY#5AC-R*
zp8jI{3*@oy0}AD$e_7w=KGDOZm^jDE2_A&4zJQ~$hhK6jrmiA*=ax6Nv(N?OWeD-g
zUksdhz}YP4OmE)*;nPwEpK6x<@wV9gWqR3_c?vPZttJrv1MYQG4w*fTkI!BlYNFK#
zqaSF3yK_9OL*HmM@RVFgbraUaTCZA2uvmIKs(w)uUkV&4z!eI97yp47xD(&zt#;Qp
zwIzyukT_0n2G^ot^t1ndX}2JmqDu`CNV(%6VfKTh6k*OWZFMi$^l;4yt~_LVOP+1*
z>3P`8aAs33e2!X$$sbjOA8Xypeh`w)s*QXzEZ}LE1Y0vcyD9HUK`IGy<`v9_nYXUb
zLe&{O%8Ox1EK16s@IR&qXN1|aD{&T9K_e6E3fL*N-@O=>{)WmW{T+W_#Tu3N89^P!
zG+F7<RPJ(lzJcqLp8zGvey2<h-+(^89<81Du!#-iQ*?f^RcO2QI<~0}N{)rD&!Wb>
zMvEVQs7Y00MRvBU5)eiaZ4T=|^ws>?($?h()j*lz(`u)Yfxa1YwyDqBXeT?Y$cEm`
zrf<yl%0I-OzbBi1`YnXG8ma_~E3Z})I>-k%!$^(s@!?{wY+yHQ`ng1B;JA0@yLZxR
z7o6b66AQ4FO{o7Am<ptI6ygeT&G!lKD@L&n2+$o;r58(<V9s61_X-RS90#?)z#wRx
zmZ<Hr8~(!{KybSvh0oB%C^m>Xli8(CCNkmCCCb2o4U)`eIW7ox;h?1D9|ey=kq?MP
zOhN8NaTJAfBH4iZ5wcw-x>3g%H4Gj0zVT84sZ^J;-dnlrg{hC!pipb2;{>M6s=$9O
z{b`k8led`371$hF+xmFqe%b)_#qs$s@(Nm&pEMVMM@Vi@4kob8x4i@{X`rTh#wt3I
zVzScnc2nYb?0<tLq#q`?Q}!=&_LqNJ=UTqZt!Zi+92me=eR&D}dS?GMY_+AaN;wHS
zQ#afvaqHT4)%^A9_vVwEE!lJ>R^a%~z3_|whqIJyp6{)5e0dnDT3K1ye(t}2epG*V
zwkV^XB5~lkBFq}MT632(Y~|qK(6SwT(YcT|9dAAN2RCA;F;1!wjUTI|?q)f#t}<F^
z#i1lc%wlfCQLnlQ_2!D9%J^pO{ODnF)-gVjfn%mF(Ni$8Rqdug6%@eTnl0%Y;B3xz
zwqA}LJGO00rk0l+Z)d&$Cp&;$HVQp#2mk==k2t_!Jzhrs?#eyqC;hmaKu#OK#Zr%h
z_WnETCHLd*)x0m%i|1?JFMUcz2LynBochhnw8OR`qucJAA>T`r!9u6g?NKHmT$jU|
z0sEXWXru@3&}dq%#h(^ti~WVD_?36wqs(xoK`!Cue$uQrE{$zWQDwt;#(z6+Kkb}t
z*+gmB9-7JcM=?dNaB<@i4`_WT!Y|B$MbDQMajwB5%^%9-!Q??M4Um$P8Or$LL(6-A
zgxa*Q0iz77r=qz|tPFAor>?w`pY(S%APNc*@CZ^-ah&{%>_Z2-V;viv)Jh<)KH;4%
z^@v256D3LOkoL>P1g8jaj%o3Vjwrewf6nB14g0U*k)uFklVRU?EW<Mka@Wo8tD%5a
zOs7nR9pvs+wF=@#NRDJumKvV?)h!24C05hk=qt^Y(xr)@F6XbHESDG5Om7aNh621=
zWu2)Z1xykiWsLO4%U2Yam)yU{`2>;FI205+rF9{JmNlyfzy^zE6%7De(X}t;@hG;-
zpM~$JE^(u<fJYDEm=0VKpCEB)jGBx1w(o$crhS?B$fy_JoeS3FJ9$Y2GJGKQ7_hdw
zIQZAO@!qw5g{i~|{wVJO6RAA9wA>)Be5{&q2UMw{-EFIS8{?F_AVr-1S(BOg@?Ee&
zby63P#BV5$m6lgYYGe;rK~WoJtV&Rb>*L8<4|1R+azxf(DifQ(k`IDDV`uBfV`I+?
zfl;u=CqcMXM(=e-mk&r9Neq$wAmvr?mr|rCK#6_%Rm-F0N?1i$z-Nm)+4CnS{>kQC
zy~gTM(vU!Z!(=9iTfW@;kHbVsJI`QxTNH<3WVtC1RfH$SjegVdvsKuf`V$s{t0}z+
z(~>>P5WzIWP#hK8W5=wQVixt^;KRCqO9LV*T>?!5U6E|8ZtqBQHfhAbSad`UL{=9*
zp`l%?008)T7N8Wx_+Rw&U$_RiNx+?2>^VZmO&k|jlebQSG8E9n+D}PAjRxEHD6@(+
zxElj=KQONzEbl6Uk;9;nK8!;+YLX$}w&Yu<VMAv#5r|l7N|**f+|*bav^b~)<U1xT
zGvX-si>skt-g*BibQkd`DVdv|{PgN-eMXUz)G0uUZts$h$%`5ugemcQ#uj84ca$ii
zO=Yq^V#yDhXFA?~XSOG3iO?k~1h$JNb`9invAD6f&7unIys^L(MOXDpz}r{RAw<=h
zF%GSR7$pv4ojhkAlmxeMSK+gCSrL_~vX|LU&33e+1}6P>WAR9^W^Y8U^6x2<^k<gE
znLt#6DQgty*$0#E)&qP`l#L$TmQZ2{Q4{*_*84{n=Dd$eHGYpfdhGSPs_$Yf{N)KU
z$C`&SVN*m=QSr^;aX|r#(_F*hgO^r5KAEciN6Eg6=YCKA9eBOjV(%_?18ypk(c1x#
zS2y_*fAa&NEq@;wfaqA+q)2M2z3+(`5U@HHgg;3Ow*fP$yJ^S9<vcE+J56+uYLhdE
z$1(FzgLwP{U7FG_4nXd{omcGB%lM|>g-U^XXm3;v|In2N3r*j)n?N$4{#?w>V|;bb
ze~<!eV2vw33|@SnUY>uvvP>%V@iHJ_1KhN`cu7H82seCJCU{(Y`DZq6G3-#{Wg93d
z^SvGUAfI-$7W(|Fk54<W8&*riU3|T1p577`1+FhWTdv6>0`C9E(_02a*~MGmw+Je!
zbV?3A^pK)R3?0%8-Q5i$DBa!ZfPjFEgrtCUH&W6Kf;5uCyE*51&vy_Uu50gqto2*X
z4bK9<@+*<pKDsW5-}039b_rGUU<ms32?|*iH3Oz6EZ0=ZN5I5SFr7+CrPRT<Fo1n+
z29mY^VEs`^m2@I;<|Pyq*4Bp0(DuK6q?f`2C6-aQ78ZG~?XFLZFfIG@Y)d%r`-^33
zdaAafqEQcAlHD`5kk?K7V0wg+hy8f{U73K5*5NUdrOF*nWUt^6<mQmH^_OW6w+rK&
zB^!NX$?uO+Hguc>x&OhlO4~bCwv`y<<=ufJUz3C?C<UH06x*RY&>qLv(j-X7E}tv3
z1o9|(G1U#{^D6jGg~>3eR6^1bY>c4C1(i>ts&N^w36=||Ga=|;7on@wb&nxFK#GVC
zBE?_vtD7M$$GJVgaZFPv@wntkG#pbO;d;5cP#NyXZc<}WJ;P1zoqw%O#Fz+?+VW5!
zl#S0fHr%;^w7!AZOIlJ}z?4!7<orm!pUH3-NUyi$$EOK<NWOTOaCd(8cGs>sw0UbB
z4P3!w81%MZcDG>;<>PG_RMGE1`+?sTcscd0%V)O2ox~Oh;6Gf(GiE?YeWp`6%>jqK
zfE~+2k9pX&jkH|ERh224Iue?fA=t0xd_!XAx$%XlDh5JG*<6+UKGC!gWai@av>O+(
z^J0U9L3XgC4MxOjU~%LZ$pc>%0Lp-;O`VSgNY;>bb~+T=Y5K~R_#ng$6O346YP&)q
zF1*B9>ouwH<juXOlCf5FHn|0|^cKvru9>Obz5|;PZVnon?NM&A(hmgRnsy*k8B^(9
zm$xEXyUl%uJt3cP?NV9<`?2ss!Mgi_TsIH<f_;32mLS+v?%_)M`VXE>KH^um*u!j^
zn2BPIFC`Pmm@H~ziB6Yk7=>D4G-r%HT%^1JoF)ki)2Q)4FZ&u<-UI_wZ{Q0ZB2$rX
zV=hltlY>+{Pu>!Oxrb?`<sl0DIms^5KSXeO37}xg`kOZo_la(L86F^66UpFCHQPgk
zxl1%-S>4;2AQ8tqLRRIskO-<|>`&x%KdfIULpt4GF%`4*c#);^r#TG%`m>w-6y>8T
z`uwAKfL<OJ1l%h-u6wMZoa*v#HU~ZT#{ruYUI=FosHUjOG(W;o_Y1$nqmy*5c}i|v
zHA_v{@01M@I08KDfUM@BzTr($F<XiFLOCy*adfkvuGdVCD@lChCp<5D1^{+o*QupT
zqOw4JxHlXgT8Qk|Ohc%rOOrR!D+(|9H4N;_uI|&eD)8YyJL>s%TZSf1^ZYpwOLcFF
z-TCDD9rsn|-If11tGhTY78@e{anZGOp3bg~dIrY_3EnSuFaE6s+{9g0(p>Yz#>L&W
zPG9PJjXW;+Z|;rFEX4%(#%vB?IG%vtDj%A_;`0w6vP<=K+-JYM|74+he7u6r#R=4!
zOgI=Hp1t}W+C4d$`I=itM7uO<V!`lrDk%#aM6R#NR0_kTJB2aL4V$wWWLJlCU9-#Z
zfi3Ev3E@&U`-?<S=x?-cDLW}gbZSkn==WGs6@60*0{K~!7GX;bJj+?bb(o3K1m0T>
z0^oryY?MnR2IfB){7IRWi*AE=7p|c;k-kFP81XR`aF`js$Xp>~J7}3%d8ke5f@>mu
zxQavmL5fxWv*vd%L?|%Eu0VhbPAij14(y=#U`qLp);f*LZOurYy|+9Tm;jFq_J0#z
zj}<#W{OYJXjxF(lyk_U;Gh9KiqPKY=0vv*0p1oL`23{DSzDpDfdJu#`78^_ufJg!W
zi#Lk8)P{hGyh&;Xa$*}>*v}ou>lymW2M^mLXQ;O0U6}{y&&qJI4G5PeUNj4H?WqhC
zh##YYF9}<F?~@ML1t&V|=w5ararQ7(NCc7#ed~mUaoE+~fa<#q3juLelW~ECsV#*K
z-Kl_s-Ujv@1GSD*#p1V%!5liUyA;9~Tvr}4CfT^z3DdqH&09j6*yuF&1D_-qfX&i8
zp0ArScG}JZ%dJIg7g1agI}cU`U`q^2t0?{yJ9ZA`VtjU`oV35iiXD4O4?fya=<7Nj
zGSle7Q&*1Ma@gE{-MounHQK$B-v95FN3lv~>N*k57UqkGaV<0xFn&&sDfB5eMOioG
zIfeFSWD`pTdqW6)A5r4A>-+bG18NaSHs};C_ykBv^(j)7KRA~xzWmp$e_B+?Q%+{h
zYYL^<!f;TqDzh7NQNqz=;*b&vH4@)S`a&zu7h6ns9{PZQU^j(wHi~bPxVQp}WWiS6
zG7bF5w!EGBnVf~k17rFbCY_J!GF+mLnVthcg|#-uDS+ArM)a!(#tw)e>aWNi1KqP8
zVZ`gIPDI+6f-Bfh_^4LRRsq#N$dMaWz)VW!?(37UfQ1aD4(ZMTe%PjJ|0t>7q~M#E
zs6tl{WN1m>U<c8h21saHvkBBlAhm6jFqBB&+%65&jX9DL?{6h^@ymLy$Bj+=;gJGw
z8-szJE~H`9cMBmNPlC3UFP}e=xxN=GQTrIAp~F*5Dgo$Sz2hfFj426F3<cGs%B4m*
zP75iPP)Zz+R}eX74}***K@g)po-c)zcOD~-RD+3$q{?fAOC@smZ~Nd|sA}N9&tLQ7
zeab_eajiuOM(h(_kGvzouc~1opII&sYv3UEy)Wu{sDHcDk43Y}BH-Y_7WJtdD0Vt4
zGk%Orve#Xu0JMjr{Adi26mE0eH+>DX@bVMuVol>~?j-bb&x*-?SS$6X6D%@XNZnk-
z)2XczWSxM3@gV`OLb;lWYheASmF@b_Kr>q~E&589$vU07d-La@>9z><@UBbbz1Z!V
z*lkh3d6@rp*y7IvkM9OV@RifkQ+9uUU*G=Y<71kO^pq50n9_Bm%fZ?1w6L%+og9Jt
zzjH4<oQ_uR+Y^ipz$E}D;Dpa$9}L3PrTn))a6m|Q`<fMq;{mo|eZO_hm_Ridtq|yw
z2o*?$Bdf8vFM*yuU~`NguRGuEn~tFOKkx2x^{09%vy_j4Wq(R<Dg(I;P6_K5kBUL;
zo9!=wA2;oEwVWsIx5*ogZDx8L`QMhEsSf+5l`&7U*`qvglveboEd$7^CMz4}PTpvM
z3@z&8m1`2<i=U3&yt(WFRZ}{cz<_~zZ(_Y{-ms~qiM*Wn#1z75Rw$G#0`#ATxzaeB
zn81ZXbCDXQk8q>~Iqk$IOMtp0h=m^5{>GY&!b0`1e=IqHi`onHwdMpaB2k|yO^Ieb
z-G{Y>dn^!DFL0hDl4*VbKhbYBL=vptE15jKP;}3p0j+(B;~V<1EE&a)<do1p>6GxJ
z)wW<ne0~WpI8D4kKYGvve5}L~z>MTj|4n19rI13FS2Z=qLt5H@aKZo00~u_AUjhNE
zZN#8ZvOrq;VbpQTxxET_+X@zNL~sJm-z;l2gin=7HpyiOV+CC!dTKN6?+ev;cry5i
z1UFv7_D%2zwecUB(ft+BNC6sJ>(cG~bqnzx7z;l9OjjL!?5)f2OdNZ{3kaOmP@Aj4
zLxeQD4A^Vt@*Q`webKBYYBgH+7-t#-A$|3Y63OemiJiMWKdjL7oE0fj+ZVYR6k+o6
z<ATh0lIp}Ls7SC)x9P$s8@p5X=#BuRA9XELc4QWaqdTcIhe{I@Yp{X~H+r{`-o%0^
zCI3<b)qoQ%$K}|8CFqr*tJcN)_oPvQ4`B%UKKX9{_C?y|X`@$2Xh`nJq$ct~b*EM1
z%ftFwjWytVnqaY_U?MAxcLoAf;2eM_EzuG@(LPc$Fs`#VfG1&Ct^p5s`(khln3C{M
zWeJ?f3W)@LGe<_Q^ed!?k;igQlt-XBZFhmrMwIkX>Z0iC*;)w;o<~e*xeG5gH&Mz^
zY?sy?J`0f;V=V8m*Yj+wG2Y|gpc+f8<^o8ezCkG;o~dKDueU0wu)wIVi-;bqC?;4e
zNIrpe2yo9rmEhW)-Y2VpQH9uEXE|**igqj*1Sc$AKh1#~uq5~!TWK(pnn?*#HI>(c
z%Mw}S!W-e^?-h;ggMJsOwHd91nPUR~=~xvm0G1@unS8=vu70LTvLk7!4hq|oQQzu(
zm^WBNS$T%`zPBrdDVYdm$$+=q0L3Ag-h(C`z>9rI(3bYoA_M^`lCe=8&tnE(U72T^
z!S-CwilT6EEt{ST4Er*7xR2rAiX2$>VkA(|DQDaeQLf9ua&=$BZdWCl1Pe<O&;tTR
z6{^c?zOEhnF(H_{+z(XDm)Az954rMY83wSKROLH<B5cUOV7^I+@MpAI`>46&;;FST
z2n{@;{F?17gr^GmP-^=f#Fwci)4%*w{fXx^L_z-z1q?Wvd-y)|l~D33O^aj1GClaO
z5t8}nUV!cO_SLS#uHBE<O?_v13lBsCZrSA~|M}<9J{g{`OX@?Ps|AZe^S5uWmBbFd
zFP?Gcdaq*ouQL0iJwsa?8dCCKe(b;e6!zg};m5VtWr)!QMr~bPf(7~efMb<g%)3LH
zTglki*u|fAEt70le}RCJMsZt7a*X%!l%7DdO)qrV+q)T2=wxmWQ=*atB_XgGR`6E_
z9jy{6s9&a*VP_kCKKhm5oDL!!otpjobCQG^8B0s!KVdBSzDw9G^I%z7iD8;}LyO%G
zkx7;JKdDKI#7I9uxj8Jp_4%N(uzmD<Ectseb%@3>80CFo5nifn@c2DZK@{3D!!g(R
ztQUlP0o;h!s|v-;G(h1ZEE0CnGPdI_LSdmkqJVbB5HiYYlAXlxzE7KU*YDyMo=UMA
zC3<QKG@D={e!7K*KK%^__#V9~-ZlY&rTsu;g_)R*q%pHW`}WJzW7ta{j2igW-SrP}
z{ffi`X$tpLB$SJJQaP&^Yj;HG9g9^NU4%kqIl;EtKpVuUscLkYIZ^Xss%@G`i?`jf
z<O{|jT)Skw0*F@A*_J%tC2SOtO{&~{A{b<*Qeps>6$?yIm^n4H^wt?7I&b3CB**oi
zWY=9&12r1>EI*U8@k_!r85>c3T0Pz4=D=CJ|95j6_+9KgKrZV*n7T`WEP5zN8a`eJ
zq_UFZ_7Bg9zlZiqg93nf^_ww4XEMPcgNcByO(vk^fOn??H^pinlf^S}OEx6f4)b(C
z&}iqlj4SV>EPd|I*bvG2SHT7!(X}JD=g0Z$Q+?4wTso!XLPh;MFiYyfo1ev9!j3G5
z+m=bO1ppr)R~o&lddP(K9v_o->2C4FoKauj-RHcE6ZG7VrzxF&A30h4tDApt^ZQKe
zE|tq3`=zs5p8|_`YNuEv+ZqOIov)j65k3l8^@dN@?Sad%LZzzB90h{z)K59YoAp0=
zI<HYpo_q!1eTQ`0i9_?}aNyzz@Gz_Go^SR!DMcZLeaVII@#+kCa8=DS=t<)}YD}*z
z;2$7z>6!G;_%am@u{=vjt-d+L01JqLHozHTc$V^hbU1$(Ta?gyWk|4)DM;q$^B|PT
zly7PD@mm*VLt*VgHTN(uHK;`z_wvdWHrvgzRY4&Pjr0@WLN4j#Lli_<u^g#~X34Pv
zKRA3Qg8NwBct4CIv<SphKs+l7lwoVE1d@(*##E;2#VVV0wmA^}LZeE*MuugFOQ@t7
zZyr&_($J()V6h9F1S_IoN|+bq^1cYOwt%TaTEY5d2QXxk!|YRl;Hv7|dPoN*oNfl`
z1fdUc>0W4&iC-tK7BHuJ@~9jdRKlvS^X2HzIadKYK@Z_r`B{GSWN4Rs$MR-LkBP<U
z1aXE-dVCsMUP>Ax!mw)jH_Se!B+u1pd4-wY@XtGK60jd_J6PO1I`zy$=#179Ux|kz
z-xroae=3@EM!3bx4}0jN%6|$|g}r8l<%vrdTY_y<5EugWK{tQ7##nht|5VKB%Ipq7
z4S)y=zPc-_7&L)K;Ec%0ha#Na$I(FZ&lAT>qW`?-&Grz1dBZ1@-=g)4>r*??x1wN+
zZN`Qr%m@xIpIp^y_o+-04(*<C4d5dYyF7_|_rGkqfhQNF{wt&xC3#n@x0hmfTl&qu
z#P72NosP9{`;At<{;T=Xeg1Cgym@Eqs>|pG{_1Kh?wAk6PKh3M68i0#(X8FCA;AXm
z5g38pbp5y>Md<0R$poCrJeXNp+Pb|y*eam;$MQd$$3Js*mkI$pTt|{(r;<0laW}nL
zzC(4s&%^?XM@aW^S&tJ>?uE@=%uavf86$cG^r)248u^1@?7}5zZKV1^XQ|fJ-;4V5
z%L+6zE3}^$gJS7xj-oKxW0qchL)TgBEMUAg@*NQ-Lacy_?{-q8Slwl8em@E|4|b`a
zBL~$MmKmO+KC<%uI2@$rB{+XXJ;AajIZYZqA@N#>-K(@9>T_;ksPY6a@VBgScZ?DH
z=Ke=wgeV;=At#!|7>wiF1O>bb2nztmMWWQJgPM&;puh~*i}`&cv=-8gS$E9`#y<!T
z0)5GqsgjXfh;a#1e)ARI2NSUHiQ-Q>F$g=sc1*!#tNM_)MB^qJ!Nz1f2<Zic7Siwh
zV+d}361(o%u6=I`Onmo={Eh7kmu}A`ASAb<E_~BP7Ge_$CQH4XZKEocu#m86iow<B
zX4Uuf5n1C9#q>uKFX}X^>z!)4fs&RJ8lKL|B`^HBwbo3`f`n-=2F()5@>);?Z|gf}
zmIF5MZ)V-g?H3h=67^8sx3AHWc9CR1z>$|kDgr)A?=$P+G8w)mp@QAW=N_u9qxVp*
zQmE9Zhr(K4)V*C;sY3%GWrnRBK^j=dl5cYuO~jaLI<aWu;-)$E(Qx^rAE$*68~Gnx
zv;>s51=;}#0JYfCCNL-wkptY&TTV{^*UOM;XS_Vx0{5}?3%2Q0icjzlCGzvHh=1@Y
z+-?s<<3J}p{Fn8(c<^`uD8bdIL|WfNb7}wHuhZfQV^GGBR7waw4J5z{E_*4bkPl@J
zSKmK<x1Z|^@flX@Fm14mUvKB<QUJ)wj;Wv3G5JNZva%2#@p!C=V1&ni4hHyHN!%S#
zd{s?ER}Cy3Uj@4O^*TVnXnkvxy35c{Vwk4=p@Wh|gN>$($?7`=HG~C{oy94;82O%3
zuoUxsaB!MAOfhD%NY07m=w}XH7-1fK`jg&7MldUaQ_=jnb}_V&FC&De@bq7BL|Qxv
zW2|b%86*NSB<H4zGuN0dl3p4O=>yBO##rMgkq~^I2MyC%1se4bvEks|Miy4=&l#rq
z8x#~uGQ}1)ELeM|55W{bCMg`iBvJ!~DyBxm)f3csaY7OX`C+=L;1Oz!XJ`n$$o4&n
z=Q>c_1|ypyZanqXPaiGIU`ht&xzkEG1{Ni3x92hvGEL(!Fcu5tIR)vebVW?6f*wWU
zgSSHK#su8B=;t=o4ksVVu=RDn05lsYMe5)EHX&y_=a@ds2V~uFv(HipWgUJ^D#_jm
zU%n&6uf|UHFg_y0rsTPY&*ozATBS^)<TQVUdrB%tCdHD^9j|x@A6i?5y@<%$wb$1F
z+M~ILE_(m5`^`zRg9p6W@Zf0Ml8{T@=lAgYFVI5<9=V8#Y{b3q9~tH8;g~y&Nl%B*
z>k;8NWKQ+T+~xzcV^jeM8=;d*oKT^gW4ACVgetO2?UTH}i+|e#2nsrhu*tB8-2*ju
zo(kFh`1}7j35HGH32^~8Yob@%+9KMyZ}){;w%)dcy*=M=F5J5A%e(30dv?kYC)yr&
zcM*4j@@mF@|I}~!VfG8J*t~zUc`vX373cY9-p5rv;%YiX0}3#|yYqlMn%`)*;O2e0
z%53yvqh^g9Xc=Xz1FgIOyWA5qiGWMa#Un*EKA)Yy2Vrc(mWV#UALWz1sv#Sk650mz
zyWVHc@>-EO#n8T{Ew!{BfvhHP?v0@7)ZT$aFpUB(DaYJFVY2Hz97%AdoOp#nr*Oc^
z#o3~;dsoH{PddNmX0<IHAd)7Yq7GNbqVt2}L@Ne!7PG;bnG64^1Tda&h(6%=_6LGB
z*(N(SFf4<Zn=k#zXY51Ls$E1FHp9IN4J~SG&1odu9xhE2P0I@Bp#ZDX9>x1B7PJ%j
z&}FTW4i!EUgB%v(f|JI=%3s_K_1mr%CDIclX4m;F4L%}kKsa6fI{HOBj9C$D3XE?W
z@s#wekd|PGxeDE~Cg3gmMmYd3<^%w+GFqgpULY+2{3vF2tRivg022HyVTYdAnw{Y@
zKc%E{=p_fHSR+9iFM$;!fwRq+MH$w<04xLy7DR~Gt_4NXUUIG0bdVb+=0qSzJe~)B
zTHf4=){e8>!x2r?7h3q8+CL`Lt`eL8n0QlqBykQk>2Pu&YKU%LkN%;lrM>~+NktOH
zr;nDeB2MU)7>8U9UuGU%^<@JW&h5-rL(V;e2ExS}MSw`Ko_my*N@XaP&2vadgh9Pv
z?Pf8%;Z1PVs+&~UILvHxP+-T4Fg)ImdT#iN2oo)@Wp=ADh&2*@I*m#|gmdm;T$Di=
zxWNb4?nh;~<ilJ-H#E63^ds20O*3$93ve9|Pn)3Mfa7c*?;X<z+0eeIs>o`hv4p#$
zM8--4RDSH=_lc8?3m%M$H-xjZ)jH97Nq3(w*du{MXGbW|pyYLkgFL9C^I24hFbzm{
z=uc$1A2EwdL<AQ`?WztD*^s-Wu64D~mh_qOk!XFLu5j0_jYP7^zSoaICP8~tup#J8
zO}p0i?o=@Nsx!e#kFbMGu}KM_bJZ{8RWFn2mpRig^BoD+*ssiR$d_jSNnDfb+<`K&
zW2Np&<R-40qQ8hyi!B?jhsvRm=m(mGo0hIo=!8lrkv@vlYZk78xW@LF1kzxI)fBR>
z>gq_OM@mQ$&7<kRj6?SYG!E{=U}_Cba~#C7&kdy5(H59=tk|kLf!bLQ#hGJ6abKR-
z3!=$WX@U*nGYE$4M|J{-9^*2H6YvK>mAMrkR0?7V1BH6idcGWiKRrri?wyhQOAk$B
zH;oLv9w3X&jWBZnd6ljWEx(Mb*IpxzZ*G$pp`{;x9nb#U0yK;`wHqOsX>+YilB(eU
z-d)pAT9b6?WcGJtDgCM)#(qo4q}u(Eou7Kf8bpB?yzV7V{03olEi#PG?*-O3FxfZT
z5%Ey42p|fV;9|Yl%Xwhukwi;aU$$lwOy8zEfLx0lv=<v5hHYWI5v*Ywf>(HGB0w+>
zFOC}ci9QL{W3Jx*ULg4G9?l5pG^FRn^CB51x9q0lm6I;_(9EUN&b1$rXrn*?@oz`v
z1ZcSex<>MRHpyQdet7x}y>0r6z-apZ=-GjG#xuV1aBFpY%b-h~SH8PVHDW5?Iz|kZ
zguiNK^*?8{dw7_%tu*BDfKT=7bLBCOjFe1qI!xW`ya2zT%#x$IuX)G4zJy9C<i40`
zBvk%eXSp*T41}KozyCdy4g$W=^v!SfF%fykT?=%QU)x>2z`&SHAAveDcv^l86rA;f
zI4qH92m)3zeswb0V}eBJ)^}>dHDjv7z5Q?q;YTG%d_pCzc<IxmK!qqyz_2jW3^B?j
z_1oWnzvQ<87UbW~HbUjn*Quhrgx-{JDi*OxlWA)KwhXAf2H7kS08;_jl!NRt>K{MZ
z*rH*I_esyKNKBiQn2dYiLORKI%t?^!`8w7!`F7qg%!|G92OwvAvcFvPsT&^igYzqD
z#wTq+!!h0li0b1mIN%R}LUH3@po(5gh=VvcDD!aJ=sKU(5rE*0nuDZDuD!UO5i_uY
zS@J6f0(_)c9GK|=dClo&LMxsSDAxpR@b1}|+1lT&6Jln$`D3kk4AZ$W&G4A<RdvTJ
z_-%`1V-p}5GJ=`<+alJbgVJ-Bm6imZ<XqH>fHbyUFVXr6@7~P+^@pn>*_IV)S(pID
zUL3<KZ-)<gZ++90m5Lp$fkgV=5CuSYG)*n;(j_$4nA)lO5IM_rxMF*9-qnJ`vD$n(
z&75`BY%QRdvT?2#yGi?ZXPLYfRPvq<frE=;v?QbHz%>oXfx&h-V87&F7|IC*I(Qk1
z(Q!$=vhN{|z-VOa&Tt9Qj;Ib4BBl`k==vTTUCoJt7Z+{{*CqM_Qv?M#3rD&%X5%MX
zR&H!HOVv<%$!Q)N;rw*@@oesah{SAK_Rc3i-|!G+_+(cC#Eu)?)ilv1RF|_)O1tN4
zw{K&<M&=}eL5j2c2a*TKJ~jTam1yuy_~c(3nW|?A5Oe=k3w2EqgN^J2Vh}J*+hJDK
z9AptbbOQbkdd&TuE`a3!(>?`}Ua@dae#u&f8#I~9(Onnb?nF{_y2WF{|LE^TG7BsF
zO^j4(RDw+^sTKJeut7n&gmU2MC8IMaevZso>7i#K<6j~d2i(%71B-vlU`9f__k(De
z6x>!Nq!P*nIl!Vi>G#z%H-F2<FSNM4-8X8-A%F0JX@0CK#PWQ&WCW@RIE_pdU{0~s
z0-}1#o-_c+>G#A*7{_zTlzPEaVG0?ry$F;dk}LwF7uJz|MY2cq#q}0D(4Xj(E@ftG
zA*|=pAHu&6xN>q#0**8Hq15?-JBeB{!PcArbM^bBhFPl;%fo#IY`<msw1De&9@WJ0
z6F})Mj4JH~iY$<;cmep`Tj#~#KOG5i0hbXN+vk4$j8_GvS`}2;{NxisYOxpdu3ZX7
zNTtxBMtrZI{7N$-l*4FlA)Y~cVWa9nOB|X135pHq7B-lMYs0k5zXLn!=6?+Cz*f%+
zSTP7_8l57VkpOSMb(Ta94gv_`o^C64(Z>FNyea=_uXm(jVPQAb0e?rl_jq`|`|mew
zU6ci!eXRnk#jiL0U;X}n+jC33;3jGAEW;3h%TD!11{TPL7nWon{OW9&8M`eBOPl>&
z59UeO<@2pu896)%44ywf-35NdhD)fKzj9w9$fUn`ui(+<eQfps=ikpE>Rk>>%}`Rj
zqGXFc-w9Ka4G;n3<5xB`5@3>6e1Nyz$P{BML&YS#Mf!Y-f&?fffEWxqsR)tVo@)QL
z3`Fz#+VkJ;1`F)w|6Ja5mod8!OGh)7LaPJ^)IP%ZjqI`;c%$;cTEnyb$J>=>awItQ
zA(`;>@Tei!7y1aIh>*;CEbhLML&ZGzNMuVwp7V)$H_%V|@+G8J>P)j@eHam*_P8gm
zN({Z$G;b1Gv!;iogT44hW4Oh@aUc+_h<aGPN7pw%tx@n6Poz()01x{^fP#^v#AT7s
zUC80aRo^!%cd@AqO<T`aq<_8bu-f#|@^iB6H-ntV(-Z(_qp11X!Yh<oMjhFO&O>78
zG-s95V-wG-q893gT(OzMdJZsSal&L_fp4{UA`~@C=9}{=U3sInhv2X=fCge37);6r
z;{`z@gO4`0r%mfnPRBiF(I0{<e|K;s3H93=IU2tX@F(3}Y8!!42UA%WK|snlof>Qd
z(o}%}dxZi-aNHBWlF}G)CD#Dz9!w8S>?9}-9nQZV9=9b6$q8q6-`us-htW$arTJJ_
zhZJCcyoC0X`QSek+Zxh0Z>;SG*!1Uc)%yDF<CX#J{FCOte|P_vJk{=*H-Vtn->)1^
z$0GggoSd22E9f9h;h^Wdt8dVn%m}HHfc)6Q@e7173}Zb?p$uqn1hH|e{g>)|u#fA;
zRGJY)1OvEHDUCDPvQO|<AVgJ(KJf$_j1%0JB$hgjQn`}ZY0&9l%c-9rK+Sv`y8}3%
z`{2nWDcD-e!8l6!pkb>&674H>tGOqk+A>i-f2^l27a7>ZW@#a}`FAIK0wFsvJ_f5i
znk0f-PiY`E^-D8A!IQnYhxbr<3%?9EDr{=^C&$AdzAw3yf!`k|C|Q)NG<zekk)ylX
zg^4$)$A7RHNv1B~pqhM5tODaOX(N8lt3ar12Tuf&qs$WC^pGacJ)Rh~->ISH1L#0Y
zEh}8wnDJa@t)=FXlKM(_$wwZb7szcI1;(%r(u0`U=a+#Sq~=Y2+(<D}{0}HrHC++@
zB$QpgY&=oEII#Bq<S3{32i!CawnhC9%lM^DJf)e1azmS{#OM^&O{mm}O#H@(No?Df
z37KaRrK7$IL@S~6tC%WVJk5yClNV;gT#YxQkGW3zDBs0~Qvajf_?PxwsHoeRdJRsl
znrp^_#aJ=*Y=0lrdj!12d><U&cFg|y&HH#FJhg!D%nWe#)j9|g!$oDpSDh&;Cp|!J
zlFuLK$S=L3wTx)g0=i5_i&@;beB)5700Jz&I8N`@U8SU_hDMtJP%CX1O+O&oG<KgY
z%!>;g*JTR0BB<`AtKsYaa~|%UU;CXF&}<0$uL%y13spha=6?K}yPO3bn;3WdpC3i{
z<<SqH+xIVu-Q1Pjw^AvXrqgVte8f}JW<uNFa(RGy=f~|aW&yW2r$THI53Pt*TAD44
zBPU2MoJqlNyTyVCu?oC4EQueZX;dWI?4s0xd#2vwccmDPP6_2+w8XH71E2xy0<^>J
zQ9ykJjfLMtmw*N9(amBwfPVsv2H1Lo2)}^K!B??JQ(FQXt*W2gK1&-IK(Pf_1ztIJ
zM}R+E?D!ODN<*OoWRXy5{*<G&el;b?W7^=>yUmby{%e!!`UV?da15r{Hp$0VV_<he
z%ig|y*z21lBLMf>j)8?mfaA!$yF96(;<`9+bgAT&m(kgwVfbu=qLFF-Lowe9POUqj
z#KD`{B*F|kC%j|;N~L<7yS`S9S>A0w+ZXUrc`(%k=J3+e?3Ign&aYm<F~Pe&Q#81v
zm@W@Ph-=AfAOP53T~HHLPoNI5yoV`0N$VO?iY_|N|7c<Zd^>CD^y=Q*VCXJY<YX9b
z`rV|IYD9{}rh|wnUC^;bRxO1so)4F=+oV0wtZJT@RUv`F*QaeIA%@B30ocW*V6`;A
zJiIs^dIdt4F|@Zn)PM%e7BEABm2$o3W|N)y<~j}U$PC|5>?t9td<wi_O$WzldEP5e
z6tqV*BD@?ug??C{^tVNe9gSjDL5>>c9%6OQ?1O94@K+WXLA5k-MLJ;AkJjiMG!AwI
zqs$Kb+laX2GNik**#1kirMTOAs&7uP8Q4{$!N4v*m1#tyJ%1ms@Uy%{w*Rwhz!_WB
zv(Jef@;5>~Drz@6pZ+{?Y4L8-ec3&xsR3UPAK-;f;;c3;n^Vv!fLD)J&RaFI2+J=C
z+A$^sj%Bbr3@t~iq%5V6EKqBTivT&OIArK(s)Sa;54k|jkyh3eLQOhiTT!_K9ZrBZ
zC@3Nv2`ylG#z;R=4-LU{+AAJ3OGx09d&9?X8yc7|Ka<(7J7L;vp_x_lL@!B|IN}5I
z>gZIafO4UYerbf?q7bq3#rNV@iO~AYG}2vngPhC~x&dy~ZBdeq2GW671CCtVww`1X
z7<R3_Hg%hIb;m)hx|6tyr~u}CCjBVF6C+;RoSZx(DoQ|A&AIr12b+*cPC8YAN>WjE
z|1U;2Jc*Hn-61m=68W*oywU`#_*}G|$79i>r4N)3pcD4;b-NKF-C+P3FtG!sDcNIL
zU1Xa16S@jS+uwuU1CTbbC@+wHm$GR)0v`HausikJocfOo-xqF>3JAdZ%bg)mIkrA5
zfNum?2ZX*e?~%7jz_8cLCcr^L!veVVERHCk8v;-1Jh^wTY?G%D6T5toNvzqEllUUG
zFM`lalyj7k0SZ6~x;;m>6uz5frb-*G)3VyvaDdI)LoTEv+r5%9+c(=c|9XKNc-fxo
zqQr@|z}D2foY7Pw%wHm7Yfpt9MBt=K#+%!F8npcRMgQYnwTu=VmjX+1HG2`3zxIUh
z=iSL$>i=gk)WTaX*Q*<TKQxUpSnE#_obJBiYQ8r!$7sj;fW(`w#<s~W3Xnsi{fkQs
z|41_L2j8+ciA$v!VG+jLj`!>XoHR|fQt8C)P;ShUEQ%#x2Yf`Lx1&=uX<_ug;Fdh+
z`?wHH5=|%0nE;zyK1{xrwOb1?ssR};;eej5<Lp8G%NkfWL5$3a6^$Mv$Fr_HrjxsW
z^Dj4Dx-EM`)mFh={3%M1-5^GGi|iVcc_JKiWRhY4n2>`%Q72;B^rS-U?`4<OUvEb*
ziGeBayA2G0j|N3yniunh4d>#BZjw{u%*uFRRDL+&TLQ@ZtDKXoJ}NPJKo)^wniX=I
zCc@yGphUJ)!j?+}wqjtk;qC+YC@`@jLeML^?&FH)&OPVDB0zCW>!1*Oaqk?@Me%yj
z_(8toXL5>%VodUGkl<Dwch2DD?|;H$%gjE1(Z&eVusH(XDQy^tV*AO_HcyBi5OJo(
z-yZH3+x_4r7eI+wtryncgTtbb3VS%g4^bv|Ab}<c0k+s*es-Z?!&tz95h_0D+hJP6
zvFO&rqvJ@HeINqVH-Vf}ILG<RU^oqiL39d-Kp1k+^7G|{3ea_f64K8n0IVd18r%8K
zr3dqgFvXVwj{UN7^PEfUP9oA`ceK_HK${1HBJcbFs1HcY63pp|{^5Qbg~^i-n%!Qi
zzd)CZW$Jh#K=GyK-W=v)Gq%~V0356g-h?Mc&vkZZK<I}sSUcSxW~KBW7oB8r7XN3=
zy(pS7mepfjHJWUrsr~Y6<G)b0f-(Zz!Zd<4<F=isi9i@>mqsFqxrJOVcFZJ1_NYBN
zvjoY;eW(@)>G-)G$-!zmkoubm)Fo&_z+j{qL7GXAy%&WXuL^mkqG>+fv}{ZSVp}L`
z`)5>2l@WpEW$w>o3Pa%wQ(*ma3$-MA;@tQpoXSj(blMu88XR31xen}us_UDkKiTQ`
zI}gw!WbZAyWQ_nT4k+V59rCyUU|yxFe<$kRb9z5H+P#sFekd-M-i0XKn&{-0*50X?
zsMAlzSbMFA6)1l<=L!3QMY0DknYoC2m|5GM4}c+jmg0jeuu^9}39p5!8u?=AN+y<^
zR=k!gvAoHQvj_4d7k$;O?-&iPRRPQBYNR_(=im4eejgfA&hLGff&rvp08|5B;qE8`
z5XE4!<4%QtR@}#H!Ag70SlH^q?~W&pe=gmA==H8fdEprm4kGbh=*FgW#S-3iRMa^o
z<@haaU9r@0S-!*EYq@f^w!Z$Sah=%O21DFa!#})HSiexx10bJ-UPf1&_(4mZh9mU|
z0Co0R_N+?Ob!+vQI1wG5{(1E^Pnp~s&GMbk*oHNM$qt(h=kc)iZ<12A<MvxcNEJlb
zB@^2%woHAU(aV>*s0i?>K-$xq!VAg2J@|n82Kc=CJpk+98s!Gwlh1Q(TzkrgNlm}2
zc01*UcZ=P?5OJmFHcTuR5`S0vdi1Vv3Q%40zE4;Ygi|(^uznGM?w@Y{4|b9^+|6Dc
z4lh-<4k`=<b~O<&8UcB6LJ=Tt2B2k0<pN2cGTkRTSyLPz`X<>aaOO1%w~k76jvPVT
z_@#PWnk{(Z(9OTG4E$EuaxFe^;f}o+J%h(G;R942UBBz?2I;wQMr2k){_ANTZ|CN|
zKIkVF1!X242W>MiGGkk;M*7O}iZ?_VL$yQl9S|=0mim|LX+Qj0$ArwAElw5(4D6Lw
z+`Jyo3mtLuM}$ph`3=-C0yk*hgSai#egJF#s<A&A%BaW#QUq*v!8iqo=^$-dbCyWU
zcv1^~G-90k$;jQb4YWPqUGr{FvDc|9{rL)7M0<D`uYE}@J}0M#x`ra(n76S;kAET9
z*45W>)^S#%s0BHr^lboUW=;)W*lwE=M^~S*F8*(kSxk!D{OUj(j||I%18Sh>)q_F}
z*OoPBlQ9$O4g05}(v!HB1d)|JU{FYHSi%>@Yxpw5<-|Oe9uQE#Rh0!G;QSg?k``qQ
zRbB^w{l=lif^ncwPY&s{om=m7dtTd!mx%sQC3@^>H=Uf8h7ewc3(GjQ|GDK;ZEX;H
zL-{<5rO(^B1~8<}+QfenEmlqfDSU+(rq4(ho`Oe`QNRWn_(E@<rL#jepkM%8mUmn$
z0oQvk@!L^AnC1Xlm#@!NN)RI7U*;k64w-OTh1kGWc(Yks1f=0UU93|CE355kx&Ngn
zv=T1$A-Wo}!Kj4i`=eOSRmvpWd}M#(DCYM^9_n5E8pl|7$Dvx>@@|`|-J-)!OcPvQ
zo)Z2}X_-+_0|R_rB5O?o+3XGC8SSnqjpO6Hk;Q;VbE&!K@0Oi5zPSkrFJw1fdrB3E
zOFjAz!pPl<LjWK6R-W)jHAbILlkdc&pUqBU=<xuB!g07-3VJG3A)aNN2s4pZE$Fr2
zXiV(+-=U-R3OAj%8{9OD9#3U;NGQ=V87TbA95v5(URmXR|LZ^PySR0E7jQ?#MUY@x
z%S6CKBu9cN5-Q)P(d(Abg(~BX(%jYIYkEPH5L!wf5m;f)txbX-{X#O2g+%zfAyJY=
zIF5;CcO0B9YeV>-&1)wT%uvhB0=xWYvEyJ|Z1P*?eRQ%_y-Q>XTehw6cB$eLc1gO#
zQ$cM3A^rnn8w&qq-eZQpp|1>+KbE!Q{vl(NU6w&4tkTgb(BCSnLi$7deakw5>Q~$R
zhGiqC8Q<w38q{i_J%aAPG~;3l=i`;SiZ3j~S%{A;T<t>Bdd@_?>k(CIwoG!<(?a=>
zmh|5xER2Pi!2sBDK4)gzrqug$TUH;q`L_(tUJ{hjkL88vFJy3IO11a-N?H^37`C&7
zgIF{-izHgPO81m1&T29I?(~q-2;>vt56DQN)`bL>M03Cdo2Rv?-WtmZU4Qbp9ugs`
zkiK~{I5FV2+xGqzO0VP!c!9PyU<tU|!z!Lgq7KVBD@tDlt{ZLYQ=u`=Bjf~!uVb&B
zp5#8YGPo_YW{oRt4~HI0PW@&gO|=7`6v*W*NEoU4hhLg+T*_Ns)s`U~c9z<4(CpUQ
zCbV9#^($NNv-8)r)hqm=DtMRL#jih{nhIMcBNh#gu3Og9&x&vM)Y<LNt*DgVC8EOs
zX_5*g-%!*JTiyMByqNUvE@$^z?Y|<M{;&UjeLdRX3dm3{mQ4NU@_Sdi<v8WwjB#<|
zY{9VePg``5>U$*X9<EhW!^`$J=310!Q2X6{V-P_pJ%USqjXCmVx{IX)SBUI!w$Ycv
zJ>MYcz(uncp10Xs*$JMDz8=z<y=(o2=W6_#zK$TkZd7B2{7SS^ziTuYlU4y~&IYgd
z0x(wC^PENKW)Dr)+VA0DZNbOhChgU`{2;O!sC3-083nGXi?`$fYSz-Ld-%<~hPEA^
zrVKpQT)9pG-KG6Ne_<=@{5rlvdT#SrKBt7{WODAguVH6jQAghQrp{|+SlmLUW4iOM
z@nB6eCZU)K%5)U94KNW`j-7p45zN!2QgXvEMV<1fcBj2A`u{|SbUvpKrr<&_A7MBe
z2drarSCoKBNu#4w00j^95b8*lR3w~=8fVLz$$ZAd%xVvV4s&4rF-d+Rv(Sn?OYKb+
z?dJ*^9!jaN*C&SOmAL3cXESyVA=r+?nK9xH9jzisiw3m;ghaE#BP48rT>G!|DfXsG
zf%)N#Ysk}FTOK@u#aWulg>#|IU=5BCg(%HbN`V1>y4$Jyy+hXuWI3jsuA2U*p*kcp
zR(xS9xUI-^+(XilMrlV;qBJ$Kob}WS|FKzK8&oQRcymCa_x5sEW*h6&l85Gp(8y2_
z<_K!~g7oX3aTJ+`Qp}c$K*raMrW(f6kB(#~!~5w7h#J9SKvi3oo@A(DXY+9M(htvY
zwhpzf_4Vu^RIhge8K%-rIEwPb@yUB!$<2|MHHwqt!u&>?3|6{eAjiqkEC?uAZJhrt
zEmE6WV_{?1>m{P=To|wO9ZL0Fs0s=zqNADZVH3P3#pctV94~qC>&R++RLbihp_4%w
zkUdwOKgklrBDqiL0g8ITMk?EqC2_l!JX5jyQ?*u|O)B~uZ!btBr4qhRn<Go&YYjwa
zGhD9Uw9CA~1W61ZZ_}|>apPUpnb-Rr+dX^9TZL|ztSXSQdXV(StuxmLvUM04J0XtD
z7;qaCFUu5>H?bAYeiW=KooMCGL?@n1JSYZso>gDo)M@@A4q{FVxn2@LFHk2kV<m9K
zXJ*ARS0YZx1NanfZ0uL2!KmK*d6EvwwbY<pK%@4t={WrW!4`@91-N-Z3-<!GZQqTF
zP5jU`qJ>l)4BMol10EYf@zNS4CR86}6xhh>T3<#>gk=*6cfS*)A6VIIk|ZmxSMAo9
zV@|Vhr<HeWbPT5H0>Tp4g-=?C=yliBv@7ZD^L$49+(a2+FKG(NkYN$tG{>T1Ez+O}
zny)Cypb*b?y|xD7a14qVPE{{(S@qb|Rnom2sT94J`hXAAcAemJRcAv8FTdW}Fj67e
z@<P=$yX~z*K>rxb7zvFIi-fMpV1Ad3Qy9=R6q*FqziPqLhKY>TD^}YC`O_;c@LUB_
zS-SLRct%Wy4)v10-2{*08tuwg{|ofpasnqu+{>#RPB+FFx%6?iou(Gy)+j#ZItO4w
zoYZLW*ajLy70$E-^N=4$B|ZOl65ZDP_&a@3V|2~$f6N?^*wxjwM%h|$cJEfcH8fg%
zX|VJM?q}uInWfiZw&K{O7J`EG3BNE}o%ah#Y3xc_z|XPieXIgf0$p|#{Q$v!B+uT}
zv8A)-AEr_B0ubu574`d0tW329|7jV3$W}%sfGQoFOro{=-Ekoq^wATm-?Dj(#ws;J
za*$y(R-$#PR^LIXhI`Zzd|BGZ%r$(p>$LCYJB&29_FLbhiUdx-mc~A)V#Zers5Xau
zk#yf1Rr8Czy4r?!+vA+Oz>lnJ5ft_^j{Q3WATNY*4IZhha87VacNiu+pBT4KF5sJS
zS(aE<>DG}y#(esnPXoT%5c9}tQ=W>bxUD8sscT$_>ajbx)L0{}^t4?%f`bx9z~ARL
zb956}2&%r3A*2&}8kNo_CO_&i)-E>d2^6o3Gt?%Z+cnR&*G1C1IwO|p9Puo{$<`1;
zKqERqVlYCl=<l`7gGt~Hm*RtmxL7%@yMbS2G|EPZViz|nv4r<8A-IjaXU-EoWyfem
zoSdMMn%QE#1*zJVJ^QUz1f7H0P--KTOX;6CUhbjQ;gF$9_l)wEo}hgk1#O+rzzo9l
zO<MMwya<8w4?#0n)aPF!ySP!6G$((D!-FD5W)y%o(Q2#zGu!+@Bf$CJaDzD5Hyks@
z<%z?cQ4Rh=T|7jb99lyB+H~QjB&8(9Tm4N}6>%l~;A_EyXuKZf*s*Wi{q;wsb~OIV
zv1>*{=@7tdNKI_kggv|wGFi83%6b;>DYv2h$~lp?&)c_o_%Z}b__9*{q*+I-7h}0?
z#;Tf|QZ({R_j9I1fi_h=9<yoZ3@0Tx>c2^JKL{k@{%hvzD7Azecj=S=P`fq514EJ+
zZj%1Vy_l{R<!XzP>e7gN>Jfx^oxYAUAHE2MM8oIxCrJ{FT<8bMi%Kgr>a5dnc+n|O
z#hs+B2)MQJHU6tHwu)^EmZN3D$wUh?!!Oa|;U9BI0?=-zTGT2W?J=J=IK~W<2|6W6
z+_?X`NVfGRK`_-x*T%W~$glAYlTAP8Gk9`ijcG=RCXxA=tSD-+kn=L@GAe(Q2Xu&b
zaW41Zm_+f+NQr9I@sc?afZ#3`MlfLwmao3gd~iaXprlvLuYvf)ZXqn%peC8PJd7M@
z>ttZ_{}EE!*QxL%@l$w0^tK`cgtaH0kqIN=fz+n!<Wxi=*k;6Pl^9o3NBfCP$jTZK
z)z|B6?CD3mc#dQ}?Fgn-mOIII!?sW!N)b*E4zBEtNM+nsX1Eb5Fc@dm!wN#Qt92O{
zFk&GWkY3QUN{^7FKLjAwvs#Ki1|<u4#Y^=|7S+thAG{vDD#anz+gc;tgeQMAtN2y1
zqxkS@BA)#PL!!R%S8xkfk5ycD7W(+#UEw$8kGjJetHWu&bE`axw&L-SF$-7+4`?Lx
zWT!X{ynCK<ZV#}H?u)z^pjZI{K%iu!S^PZ;jkgE+FLBvfW;Ko{{Xwsc8fK9OBnja~
zrH8BKcu|^BRfW!GOBM~b+zPz3TZE*f7fbKfPv2G@{BV0`gw?85)+Khyd>5Q|5s`Oc
z81VDXXv*s?{?`Lc&yj`w<e1bTb#v6K7B!1zlL9tgC#R=iwSUvwAdI4khBMnx9gdy2
zDgH~}y;f&SMPqHM5iGJTjs6#(gEXdpx7RMJKpZtEW41U?adIIa0wvTJ5tFO;SXSdD
zqGlI?a4ibcE4#h&bi@->oZT4-AI~l=JYKhrS@L$vBUR=XdSSFOo_O)%Hiv78PMMFB
zW159hVuZ(tXQ%$bkF!l`U>qx+vISF|w+3k_N>@wNppowqMcmgZ^eA@uh8D5S{^4|&
zXO(Jo!V6WvpY$g|hRAAAQ!!V!Y|Poc_L{m{mX-u!?6jU_*^ZcJBk#27{@ZG1$8EM3
zf$Hdui%KZ!t*{rOFwA!Sy4_OK&$LC_58i#jzs&Qyjj-oQR)V%)1fS1yn^<ipSUi)$
z%t3^sSVRNz`>J=(%=}j>ylFp><?Q3UQ?(AmJQo3`#kdl77)LYhpQya^id`No)6_C1
zmGbFIcMJnfR^sT^q1)(=)kiI3-x?}3Hcs`)UF+vEKtNJx=`x%r2mz@Njs1XR{Tw;k
zCo5cE>1;1x&+|*RI(zrpY;pEE->Ipq>r@t=f@Tnk^0;D)hh2Ys1SNPeRQGpgXXK*O
z%uRpE%g481_`p#3qj}_Z3602tDazK>;_cvwSal$h_2%~Mp`xbc9`ntq`aqN4*Esr!
z$G?~Eem#U>LNRg|tt-^3z2ADZX9#Ca)BFV|y7*w8&VTZ)@(asi|CZ&LOxk;fhZqzH
znCJ4n{l88<b$w7jn`L54Uu(aP9rMSUX^<FK#;TGjQXHPTtt=VKgjV`W8$WMIjKb-{
zdIJvG{N!BuR#hT%)4DVo)eBrd;|_l@RxMggsoj2Cs`mek<Iz}oa@HOUgpMClG9xU)
zm+s{~>d~<028E_JpVgG7V=!00)yHjXIOLQT;&+ih#VW2cW2qOHAHnUa=MUqSXVH_$
z5<DWrQ34KbK5h3-VmPh2r%d}d0_OaWtmG70$j7^lt5@r<pIIs1kSP$$(Y`gyE0B`~
zqovRyR+xQ|XP0rICBBugFVD+<Ub0J`S`@D~4;En>dV^GA#QtUuLs}*2S~}Jc+QLRC
z{^PayD&+BPl9D<3K2<|!1j`f=qG)GZbk7IvmKwwxy3f*FCo86Glq#i1KB?=-4>$Od
zxt%F7I~t67-o`UnoxPCI({`as{Vu~wp!7m@7+eC*6w4u)CfO$mZXB=Ni~9bddU6;@
z3&T@(6(u}F`SOJbV~=5L<MJ`p*cLdN`FywyRs8FQR-veYFQk|%MIbQHNg}Zb>_%-<
zBSDaKr<sD^?XL%BT$@~?KZY|dc|)1_v;{!P>~Qis=mF>?OpxsWSzSgGzd`RhsYbjw
zPkq$J$D#FGK$7c|Sk#X4!(9E66%yWzuM4~Pq3u2E#ku<VThgEpIw~1ZXgdBl66Mu6
zDDeTG@{)(xV5RR&1RUv1;)9?paI(ik*ks}ue1Q928pidcmGvYQGHcm3)e0}q`$gOh
zsObBu{Qc*JI#75}|Lt3+_K&McF}|X>1O58~$<40!PXVS%`Ok(%L6Asrbn3fy5$tDS
z)d}V_glQz;5$#><EjYlU+h!;(0^<Bx>XVOxYvAAgS4rmO_qzoC%GWV=0XahIMCHPL
zoc>ie+dgHK$R+UYpOyDpU;r!V@K@#)D5%=TiC>*{@?I0(>HPdshJ{Tmx!yX(kt;U4
zDm6eCPH@{cy-&zQ4&E@}CGeE9D%f#eGj1nE{W<x)8DG=b)f&a45fYc{1kP<ORy_*w
zWM_wf_LLm_HCoGZM;Cu&AZfAv-)6I+Hw#n3vz#WoSbFacE*^2no|=qZ{tQ?DFeaQC
z9QdzL_LI#pEE*$F8vo)$Ca{k>4dzIaDgn-;$QOQlnC7OqQ@8SnS&8G(pl=Rnu)Ve`
z`3`wB5)MHLWLrCYLg}GAlBP<0sdKM+V%cRgVd`cz<0{$;-Fx~jnLbV<74B7l7|-zK
z@e<*b%RT8-OGydkzW(x4(Ozkze=&S$F`BlIg`CX}H%r|O54^wYUH0cMVeshYPHVQV
zaUv?oWy(N^Sv9jraw_KpTgDfDXg*|Zvn}}AAnjj55A506N<(CyAA*^sTeb@7;yJuh
zu(dU2o-tzZa@nY*Hl(OD+h*=Uu3vw4^(l%bGC4D}ix9)h^*6I<K$V;q1u$Z5f2WH{
z9G<ZAFkGVG{mTGSRe#)eb7n#V=SR3SfbsJDtvU>D>9swUB^`W_|F8JrmNuK;&8lcD
z>iDRKL^u=ufc)-Y>`I*Tzkq|Bhj=}&#Iv7g@out<BtrPG&2EI?DdwfI!xH%KnB<&c
zcsoPrMK+bk4x5{I<Ndrmp*a`ySjzuRVQ9AEnt}y%gQdEFCLrwXb%B}Rm7_OR!c;~i
z)ud<N@}5M<MrlO8Z1g*m^dv|nSBK8)sRi1uQHyz6Mn$#KqJFgni4?{voXvg^L<i`+
za%u_7Jn3EArJb$|!BoJ7=F<@=l^yxCk{+8#F&8ZOAZ>rXybzINbM5d7&E;-KX9ePM
zA>!>6ndi!o1i!>W3CbKg6FbvS4!fpUKiPHyQ_?qK3L<Wd4-JwM7?DHRYstg+*P0on
z>2Gno>AyYpe_z_Bc83&Vxn#R?V+O115jSoXmGd3ayR<&{28Vr*<%nboB$GHNqzax~
zM1P%=omo&oGM*$;Z~-GMN-E*%z7$U!<D}H!$Vo)fC#;3=yeI%cCf+MjNuasLtd5?0
zycBAvKmZqb+dd{lJM5+`Kn39F3j9%3C9WQzr<s%dL<MH;E7tn6<dstuE<wq0*-oTw
z*!bF>0G!ZSoM%WU&kJe<7c-|pp>W|@l99+Wt%#i5fjWW5qRoa|SGU1Wa~;RB;Ojoa
zI}M;b#QSmf_f-d?lbLg9!b9>#HvP>j51$Cyq535k#Agcqxvb?!JO+ZRVIw0vr>66?
ztJs1fr$rC){QMFI*!g=`j9Mx??4DlrEPlvD9$x(orfHtp%vr#!<&0JnPdqvgxNHd6
zZMpJVIA3aL;lfy6y3z}Xdi<37@cedh={hea;12rb;4l3?j_AIJzO;X*%y6xZ&=Q_S
zz}vUI`=eV)X`4O{cD*_DXGq(fY-?eC!xCO>^%mFNaB#&pM9MuY%+6px8KBW&j&^ed
z{fL!Qr=^N}RdC3wjj6HV>*r}=0X}k1EFN<%xX312fo@E{tsn}(d$S$WbUy2X>hh)F
zI_Xkfw_lfe!1~G8#TiCr-dz21Iyqql{;bKyuC1@nk3rq{T{4dk8*jm8@b0#N8QlGH
z>7Vr3Mcz3+t5&OPXwq*};RU}v1Tv*d5}F^s@kA@F$oVUSGkPgK4%9Gtlm#Y<z;h&%
z6z_{0&mM+-)XOg7*R(|0(_07|HBWC?Q&nk<*-T5-ReUNF-Fr)g^A7yM)3i&;N=Jin
zUOxD*Mk9oF;4?x*1vRDPFvC*i?Y$Ql?A&B&vllDca_0QniemNAkxOd>lGQpAY=ZJX
zmCe6&ZCG+ENvBXjX_b&-(lhHZ)?Rad6(!1SPDHJyV-c&HwTvPz6JF~8Wro^^_;mDD
z;L*)j`2TqN?s%%-H|_=sQT8m^qm!LcqU=55Bzte!GeY(r2gfFa5E*4t_TE|9TiGKc
z<hgu*&-46SuUB-==W}1z{hmW9Sk`zgooC{50-tLaeJYRI0r+CTNT*liu~sms*IDSP
z6C=iT3}=)3H2NAoCVNioW4OLUq(@Gd=k5wE&Et_3UE8*5iA*KzCJYIc<DCgUXDxBY
zw{Va=nw;zAk77x}slA>-Q$4U^M@Z1ddfS7i!LzU#=C95>$9=|?0#XaCWJHaeF1aIb
z$Slc~nylvbJ`Py*1px*v-FKu`5M@U+R7y<1R^;~Dm7&arKyjv&V~5TFh4V0xiHKA$
zBOuF2h0i&LPnAc)QzI3XOM#0F3n(DZYe<PVt?s}&B*H3>ZWcXb$CqKWL`x&@(2x7@
zb_2>x=y19db0)f!r&%L=C-1;a#ej3z|0GlT<W>HJd{gj~Ae(nltUG)LqJ_pKuXsB5
zsWs4G^9RiH9hp%9o$I(72~3E{U8D_5&4Jd_7-njQSw%|0L5EkLk`l|pQy8ctMJ08e
zQYB{W{3~`1Y$fAvPU}Ow=R%wcIgx)>GtH`?vufVf$ML|xtd6zN?=wB>^70CeL&Rzn
ze{Fi|`-u7FtLt7EH552HoAvxT=lX-{%XbOMF6A74sOC@a@>NB2%$PE=&vRnQ#;nP<
zzPdjxo7J8P=K{s*7Uj2~IVQjBRTwJYI>3^o$;Cz%K71~tocKq+E&aIXW2k)Xr}8KG
z9jaATYAaKoNt`4UdnBr$)^3bW#gY#b?NT$54a7oK70Wq?Z82Pq#z#BQx?xgmX!mWg
zY{tnPeQSt!d2<>zYyTp)I7C2g8Y?<?e=(|GwGwa3)y#84JB-ZMwW?Iaq)=or=?`UZ
zS!?VAWM5ChuiC2@2P1bWD?;-N5(9l{n1t#_=fun=r3tlXGwi%Fd(Mly&m-l`DEAUX
zN*Z6ziXruv7{`-)VC5*A$beDpH;-7*b?y374~{AOHw`Te()r0a=Ndh?(?0Fr;IFmj
zwPr`WJnk2ES$R6~e-EJkv!4E)g)ojGbKWYaUqf8DJA;3FXU}^72V-yhD(_^)!4AdJ
z%EHj^2+LY2PvY(m%``Gw7ublt^Gcs2<t3b~O}GyUEq3!i@tprl3mBSvb*M>!P`U2v
zf?<46Jq6)T+o+H2YkYTyoUT}_M-VYxSs}$>(WSbM_r^vVh4bcLT5&f~H}bDd2ZMC*
zaB7B!OMsus_kSH`9&5Ym!cM4~hePb}v$I&0xmqn=HyF-gVSl#4AOGDyV9b)#L~!PP
zcC1myvi!T)Zo3$EDeWQxCJNUl>?$bxtmTryvI#1$C-Zc&JojA8FzkKv^E*&>acS1r
zAEeAExer<-qlPz*u>^5d_k@~<kSh<EJenzHtN5Rx^uHQF4N%yl6jPbi!^XJg{C>;k
z`6*A07D{K+E2-uhaS#3SGnC8J*dHQa6=;5N_xkkc2T2G+x7#n9DAz-#4-(;8Bcohj
zrPu|+S_@juN%uzkf*TRA=B)%Ne&yoDlv58bkK?ZlwEpvy4J|99nk2~zF~cqmJ#U7E
zy`o26rvXY(fM#-XzfK$Y(j=+5RB0>%&~;EHwE?zV+|4V^LdCXOM*i?teWF9my1PxL
z_$*1JgVcq~#^~tx#kQH73|E`e0<pcw_eT#CZ>L)Bm2BJFm$uk8eC-!)eQ!!3<eH_}
zs#g7$cd=iO)P0C!=J(A!!_*zvEZufp$mC)tXfV&F$+oZS7w#NrrUlfrrimrshB4~a
zOfoCJ2-JmBV~ZV6Zl0sg7EdF_-Jz&2AfRj4Y^7fN5;5xgZbW$VnTPg!oh<PWNO|8x
z3zj>yNVx~vDcsjN4A&(5WfYTfa+BG_;*7)JHsXAGrKw`R7T42MuXR?6_b2Mt&5Jc0
zqxnSd0iwW|T>28Wr~xjo)S;9>@&eIXM9o^|?L0U8u8`d3fE?}$rqtg6x~5YRFtL_%
z=Gd&mU0RwGH=w)f(}T4FPismQ+_y<IIX|q^P_(IeBw{4h3$ejv`F1rrrr_x2=9SHr
zRg-_p^zL}ZKKgx_{8N;?_T+t8Om1cpn_uyCKX-o4BsL;)4GVb@*F_X@(3#Roo!s-<
zqa-1>q+f)($wtK2h}C{F`p8zp#6(ehJxU|4WY7sVX`!{6smF<-O`mKa(hKN?mF{bH
zb%VtCE2@r+nv8xyItjC09vN_CGALljrpD0l?1}DaFBo?l)LRf4({<)p<}X#NwW#25
zUg%Tus6U!8*9d=lHT0<I)to9SwI<u%AXBu^nsb+^BU5*1kc0w>UaO`(s^wua=&5!I
zahZ6}A4X2>9atVsQ_tBySxe8}qu$}uC%)ql<FIwNdcn^6W1M`1($mtGeP(d4N+z40
zg^gAGQWC?@&Ek1ern9h`p_QW+Uh{c7a^hZJ%c@eWyOl>1ZXQ(tWD|}|b3uLixr$G|
zOkFv@LAF8c3SHE3-+UYkRIMV>)fFOi-Hy~cVh<$Ozr8id$R`80G({Cx`F+z;oAS{T
zcSp01sMOc{>cW4`K3+MC*E@;#UiZG}F%#aeZ)owTX_yR4G&w27VhRq(T}nCfIG75~
z8?*FWw8^+}J>hFjulK*fVL)f)jvm9-w_CDZ`jV&m%L_5lg^+j<J1DjSbn1OJ#8ml&
zNcfFu<X3dBUvc}EvbiZc`%5mhBuES`Uh`ajN!G3mHM-HsIlKG)MZ0aI(XNM*Kt90A
zm}`m4V%tuV_hs{6I4U1ynze6g!EI}eXU7}AnW@yOW`VrbB(MrEa_3h{cPuz*&V8dB
zewIKcW6)C-ry@R#+#J0rH@Q|U?N7;S^6@i>EFOZNRAMo3ITx)&HzO1=)H_G*c>`Pz
z+ol8^@&(zzhK4rj<3O|i!Wp}r3yo&zLQ(qRlmMsPtjVJm*+1kzEPplq;MC)yAn9AM
zxRWhSs3jIvS%F0@k71Z=X&I+LA{JkQ7ttE}9NpZU-tJc%hLH2zaf&q?qmpm(*NC&@
zyj5k-$<&=`Q3wnPZ?FbDo!({dvJ0oS`#>OVw$iPw{VCd1(A!NtRhSVwT$Ct&^_od`
zKKE@8c6w+ZSx1aX{9E?h3b#lsQ7#qW(KD={xBK=aXI(PR2~&jd*=!@i)O$%~2x%-q
zhoFvqjvC02{6U@8T-xv1YRE|x&OJHgb`gSAgnCDo3v25qzPnp@6`*0JKGVAg%X4|&
zM<beZht6x0O3R#EO+KKz@Y)qG6WgAUz-Fq+r`jS0q`Sq`-*HLOE|&b56EUzes=nO(
zo^(PO_;&`XgKyMcscYbG;=}O%WSHL`BBYXh*YP4Ni>Dld^r(YcG??~!%c(_3b<D*`
z9ybMqBB_TMz?t(L@%V@3E=5LwOpMusd;Fn@XG0ambL79dD07Spyb+?GOmRaxc%?f+
zxBrCtTIQP$Iyiw1_6Y8?$@ZqYPD!^E7<OSDd}H&<C%bj&zHC6XVL5+un5t|b-OhtD
zhC|_!kt4fiNV%Dlq%$3UftkY5`Qqiep?p4RM;xL*=@?t^gez6oZs*pS^oI{ZYkh75
zj$kJbuMbA$lxjaBe*<KzVZDW;``5_Ku7ljHJ2hh@w_eD_aybpgJNjYp6ZcIN6b;(l
z8vV_;8LF5r2U<21BJbBtRk+^{IMcHQ%8tCK9wtccs1|)N!9oEGD*NYAPPHVfQ=<ll
z>}G*ra{)?5oy9-BSNZXC-BfxUrdNgxDY^#+kO8p|Z-~gc4W!xg;*D?N{^&GR5O2X=
zQV*udtx^$<{_v%nX_b^M)ke_CEQ><_ch@d4ZSP!L=6&f@9TxjCB;7;pvCkqKJbI<a
zNoHvGfql1MiU|RsjSRu=18IY<^bk0hvr?j6={6gJmG=|>;_v2LVeid5(9P$N6PSZr
z<ydaKBbKLGSO0X$%C+{aBLI_HZ2spAxds|GgS@c<_zPh^oQc4uKj+vqPbYJA_ykii
zejp--9T~S=T0CBI6MJ2xU57nOTeEm41n%n7F2_TEuK(K#JHou86+;eKSDby836CAU
z8gichUl`5*FrkZx>m9D%1X8*!|5#WAUcQLAMT?V|u>I#|Y<PD!GJWRb{^4OTY;rPa
zo@5E3+QB>Ycv@z~pR!qTn1AAHp!X{q>@Kx4mS5XMdlw>luSS(kh`80Xj?2NJY_@V&
zcz=({J$?V+&2=r+0p5U&x0Xp%r$h~Q=Xd1P-CeR3M@Gr}=o(Lkh#Q>Nqj`h6A0UO_
zxI{Aa9!qgN{h-6=HjuL0;t+2uyyj?H+~UYgx$r{(KmQdavnJm@$zVJMDd}1HzDb$T
zYTsa7Iu!)X6*`BVguWvSS<4XU&DUT%PvT@E4-JQS8}KlQ>tmj=EdFdv451;ry7@Sg
z*3m{+95pRMXhDl+w=rD2PuuOF_*G6@w}>l@&?QLNM0I)SMSNL)8UMk=Zx4o~n<q2W
zRa2W){wlKdABNc}e8qCu%#;1uWZYurCjXexh9Uiwv^78cA{W}Ic48u^ZKjYKz9Xml
zk(_%`**uy3vzp)tI}aXFjp{r=eHwAM0ybmXQ&Q;h+vsB5iVSeK4DhSt961TiQ#EDG
zir9w~-joM7tN48SZrN*EoE1xNzul3u9k@j3Z_m`!NwVu9B5OXe(YvMys(sms7vtuS
z5n~7_JSaT!fG1%0#OlTm1g4Ja$-smlHFDm@wX0vaUbIu~Wb?r(9L;^aOR;3D6ZdeQ
z4sVJjz*_2Q<>-?s4H~x5-=vSUR`oYxjqP~<{8AKcT2kWk>G=MuqPJhDJd_QPP9iyP
zw*Rs97eThd`ex+Hd|$5d>KYa!%HGHY{<V$U6Ox?=pveMll+Uf3YR|bVNFBmu6ZgS9
zbS=`VOgz;DA~wF}_Bu>ADNC#>pjkSwn_@(8iA*Ij$r)KyVivgejCduS6q6aRE1)|F
zf_7A5&J4`i@-+Fv9)*D_)^Fw(5`9VJ{?HB)3fg>Um$>5IXW4A})xhYZEP)cKodD|s
z&0=c9(WdW*?Z6ML<jLbkxdrdpP|@k(5~0JM65*|6Y)p(|re?&Xb8gs3(y6NQ{Tw<>
zi6%TEmmBv|-g03|mxUMPmp@7bLB+>clu-}WE#jQyk<bVm*(Q9uhFy>lof}P$(^pJ2
z<S5p}0t}YGS7p+pHEqKs+zO*!wN6HyXcx%BH%H^yY${1R!sYMU#Wff;J1v%YQnBu@
z9a+-NEZSOIxq-d)+*FMEJiR%@S~iwq;un}*KI3!@QXhE*{gQ^0_w5nqtjRCtbQj`S
zx?3963DzJ{cYm|2QF1ubxWIRlrO}ua)a>QgX0y5oSIMYrY+CBGDc~3GU<}F6&}ZZ3
zkIMbh^%&F}GG(c=o<<4GW0t8Y9FT4{PEj?r-Xk7Zn>A37SZ7zzX-BH^=n;P2)ijy!
zIjwA=h<QnD5;-YVhK_l2!Ygq1cDF2QtGtIFpq5*;dTnYvW57;4@%w7=*8a*0Y$;Zr
zi(WLycP<l4`V3i5QyBd~v;>eyrbt+e_`~j3tokb#(;UM4=^Lj2J6B%glp$>RJM!7#
z;@{hU3wmc^aor||M@RW;Y;m$V+wF37X6v=}!Mbtx$@n|qFmfAk&ex(#aHeeONl*ZZ
z1@P@;p&O$yWR3>;O)~@Se8wveNE6Tna2!^;N}i2sd5HRJ^Tb5E4IBx4Qx=Ue?&FBB
zauuNObo`C(6S){A%`M(BA|vYvd%2(StxnvGawjQ)1}E#pGf8%E;>|!x>12le`hZEX
z<sMB#Nt<T2ct758tzJ#L=ZO9WBVt5cefK~#DJs=w_wQ&lyz)!NT07U;b$dqj1B<&p
z@l!W3EO$i)V`$}TrCKILwzDl}MTf=4d8APcbFPRF(*~;+Zz@GIU;qPACW0S46qjqH
z?0AqB{Dn5iBaQWm*zG&XbyDw#nO_#9CO;_UC*D~NEOPY9luyLT+#LvuW=xd8o_reR
zXK|(aio%%eV^vUZJFr!B(2bDL%F^7x&RfMm`~cI;R=6NJu9#RH&g!e)A@H=LZpo40
zb}LKoQ}f?BFLsg0`w%ub2<SJze5iB-ArZ%lqM7V$g!EqOOPN*U?%b`!s&@n8?lB#G
zUwH}i;$nGHTbJAw>ltc3tUtWLFTJKf--Xun>1dbtJ$|))+~R<#@ZfzFW7@Zq>h&vd
zz`#u(>T04MVjHWlze98ieSIl~a+sN-=Efu9o?Fe(S;ifU`l!Wr`^LSC^ROeCa%!!B
z?g!4ABz}#EEE6kjkwii!Y^g<8Y@e5N)v*uR+0_LG*>;;(SJ$cuR!O@q6g^IAH7>7P
zX1X=~9ddTJ;(5D6?K#z6$G3*Iz8N;nw;3+^<84GJm%I$E_tO^Ai8s+nnIRjynPdcv
zf-1?P_!{hzEsOUh6C{$sz2KEk*WLbCU4zgxSn7g3(f1m<?nx@wA<_Kbc!`75E6BX|
zfPn#`pVN{ud8QV0vjUxfz-em|VvdYMc~!qO_uzA(!M_#5XDaPGG5FlFn6bskAqncK
zXZjVs0*!oXE3#P6>1vj|Qlm|&rTXLLbAt7{OYkr$IvmX76_O(rTYo?dJcr`M@BVQr
z9F~iG7i|9U>KdL*!+fCzZ4!nnB23|gHLZ=xsQ1P!U%Hq~bWB}|ciIkvx3Uv^3OeK^
zQHfaCB&zxupnx^wG;(G65|30hnv;k_*!TZgAeQjm-!~z(<f{PA%B*Lywkdcwv-~wF
z8<SrBfAW@hWF$s5F~sFVB`2j}j;T5y0WHh;=F-7Wga0uZ2fT^dGA$iYsbS|t!cQz=
zfLgvbV~yjro|A*KSr$o^<HT(E>D$OrFyU_M0B_PeD%A3QIYKbz$4iDprP=OV&5kp$
zmaV=Pn{WG<csfl1+&L^3SmlWk6}>q*l&<*9mnI9ctS-AUWE+=S9NqMKk9N{|NI&d3
z8Fd2vTLX)?w-1kx0|XWOkT(tI-yG3D4|6lp>S!P>fKOsxvE3xcwDjqLyG?eVY;W5S
z68K~O!}ha6ewUNw)v@Eh_4j7qa6VHvNf#NJyExdz-R^9-==<}Q^-5($#qkoeNax`p
zp8>hUYTsSObfHeYi*@fb&RZuVdgmkm7YlMAsQk;LTjvb063+9F?Jl#+Q=*s(@q(16
zU*xcqK99ILn#uPLt<}&H?v(ED1K|Y{ICWds37R*z>OcCVm?B24VfYIqOM-w8-b(*r
z*z``;gCGZ$hsv2PR}%m1*a+86rOW$h`N>*nPe3b<i7LG!x9%k6`%d4;3}aUB-4ncI
zMtC!pwmnbqA4;_p<Y@GndduEfJ)`kj;{CR}BJBN#p4mbG3-J5H@6AIbqm&CrUXE7M
zai&5QuS}{923-RtT7D2CbWEXD=INbMlf0jNV~2TDQ&a{Nl%a7v2jUnys1+C#&RVIF
z2?eAzZ1%5hdp7A45d9iODPQE6oTV-?EdYdcrGjf#yRaOQe$($b6hkE>EnFAM9~~t9
zg<)Y*KOenJ5fpnzE<UO772V>i7|^3YYVmFkTh!;5RTcd6MMYUx*v38ABM9x&9louH
z2z)H)lvQ;SALeE5#Nc?)Zy<nMZO_??N5syKHb=LZeskx1xX^12qE5-H3itlfVHK!_
z?>_=d3y(2Il3qL{dR%iUWaF;4Xc>>PZ;l@0m)gL8TTs#Fxb7_MRuNHr^L)s2LLAB9
z5_HcMU#G!sKUno)N~KF+i+w0V_t|(WCTOVasQ0qTR$;09EWBZ0iE}CzwH(+S3qrPr
zS>|vF#of<dn^%+RU<meq8qxSPp*>CGXpz~(zvYT;NMK7+k9)&zB>EVzqss`oQ7TZh
zzJI{&7=QHg;)Vz`<z1Xte&*T{e`%d;wj>>QE2%8=Il)7wZqc3DVscP?jZ)%omhFm&
z?E?)n9`#F6EZ4v{yD%|Xo|r7a1TDrsgs&#Ugh;Kd$W^XTLm|j+qzQ>czEZn)9>c(Z
zoyEY+g0h=>_CA8Npy>{6)?~}}9!r3+*?p$lDs)p-^dgzoI9T)5E)>Jc<UMpTA-cv3
z!oK6o0qoU0cq$BBGx1>?1|g~ErFM*~Nz%9$q<bMjPZRF<ebg|to_22b9Z8=1%azN1
z%OsGgT<t6PhRh2*g<fnc(I2*5Q+IyT_Z-Pkys5yNf+ON!kp4B0o;#U-@Ohob5;r!c
z_h8hE`Q$!_t;4*7T(0MFhL+sjNZf9<4jyHa>MLFAaq^i-Q@`Dteo>mGrf~o`*8|1C
zhUVJZg#G0j<c`4nx_C>r4jt!rcZ<ukCD`1x=f$_bl~lmk);o5{9NpaM64I_-kZS%I
zc=7f>eyr;XDKNEG{`ohD*u8o1LNQcaLEV1w%iNvNj7sH48pRU`b-c&J%DgNq)D>7f
zHN(+41)WhknsfAKFi>c}{Xus;>3<h=P9e2;g<t+vq1JsF__6JMU!ecG7u5!8;Nza|
zlq(42EjIMKtkBhK)jrx$Pv5Qlu7%Me-uiN9G5pP-lX<+nnnsqMG#)P|+LBJJiXYD+
z-nk&Mi>A)8MM<cl<D?Y_Xkk#;R8|B5{Bg1~!K7C^Y_uSC`Tn0}c%Sv(!CJ}l{+q{|
z(EVSdusL6fcC|x5EVI(y$Y-zkZ;%g%<GMX_{y%Z!BMEP_jt57lAIqv(ryi$r;uHQf
z;p(y$e7r3(VVq;I&!IQls`l+^)@f%uDHm@5JoNy~eX!_~JGwz0dS_8rH{akl&Qt!T
z+eS7t)xds37U#F!wtHwMN1MzZ%VD4og}v<4H4^ut^t`0+<XC@s9ax_Ao)f89877b-
z^F5E9qg0zL*I)1aw~o9{rr}x#cX*vo!ZzNCt@l6fZ*QG5Xaq(2MZriRl5HKTe*pc<
zBlQG~V$1noDS^A@CAEKwlr!3B2Vozy*MaITkZpHl@@d6ic1AH0+HDGu{iiPR%4O-?
z%M)bQ;*&KHs+x5MttIoz!c_D465e<oTt;vTZw;zBU6V()j7MBY3FT>@r561Q?_&^0
zz1QJ_ez}e(kvy`{O2xI4{pEeLnLut6yx80h@M@b?ZbQ~q0h=mJ=PoQT%+2tASXryP
z5ch)(86s(Vo3YbGrC#w#wEP71vjM71w-S3eeEqa3sY3aY;2<HVr36ZX8cnyujY|?E
zzxOePRc>J&PfDP@p2b9fC~c_O{(IlEs(ItN-FUW)s;Xz@)^mYPo^;D}MW&TL1+u6)
z=9$NdWkV&$QzLIz3b6_w1%b|lSjPCe_kngwa?$?h*`Gvr^+Z;fY4nzK#~2kHZKxH;
zH{3nv+X!;mfH+P~;wV>x78eoFpn;452!x%c4ISqzT8t4FYv=37GJVW6DN-nZcS-e&
zUekBmtCyVp#VN-uc%|;-(os$RKw2xb;i}dN(8>^Gj2L2|P3ep(46$Et%k~-;u{x9#
zq9cpD+OeWESoWN0?pnUq5p?;-r14LBL=?4ZOab0J(GkLr6!%#O1<zQX26C_7d=#GZ
zU}0dT751&ovghTWVoj*mNwdXUY89IO|M7~e8iTgFBMg*kLrGY4W0T4(lazr7)e3*w
zIVUbb_b>OHG3%v|CmdTE3rwx36|TLpVW^o86rfJH6%Xo?jwlQ}9ra%|ndCfCZwd6!
zkQK`CLV-BRrSrXY!K-BQsHm{GOaM69D86)4J`tmlQay<Y+lRA-CTdMemtiW8TN>2@
zI9`3svq5=ZkfD<IiDP7BKZtu6k(>j*ky8!y9nCJ3q8Z`N=vMF+!@`+oNbJ4|C{`oT
z&62~CXa@YYtDkN0)*{6~z<7LOmQX$W2X+hg;rqD;Al<!dSpk=$L%~#kI$TY4bFy7$
z71Zv=o3xzdR{rgy>dB-mV$CP2@fx3!4s>X>hKNI|L!<GDo4vP=;Ni_OwE$!=UAHik
z6=(8#0>`va7Lre30s+?xAJEE{ujBPSjH`0zY-JfRQYuzEX|8=LsVhG{8jK33sTb%!
z{GM&SMxN|X)nY6tR6h8u%hBufw{U-k`A&}{QCaF(1=f^M;)_b&T$LC1yN!p`mMAA{
zx+J^jy`<KJ^WJebr<!9shUInu3$9*qL*Cav!`(7q*3&9MBYLm)VmWB`|2aSSQWP3?
zBW&%qrUbi|_w27$RabYmhLP<GZF{^8IdXbiNwynNbd~=$6No9bC-3g}{ljQi6CAX2
zn^%i<#3h;HoTvGA0G)L_dIXYzG_?Z#Em0>w)7T%$Ew0S_+s<q|W-U`234ksHs71k@
zl1Bk@awJU*i@0O>D#ju0FRc(S<%g)f4w|UXH99U1Muk(JCJ1%I+@>lbG@E2xKp1&|
zYEY1jv{0y2bJ3h|heLPLkTmy;tmx_fifM+e5cQ7t>F)kUWWebB^*l|dT-pcDVNaw~
z?;V(a`higFk0(EEx~anVLsXB)h>!tND`X4`S9cfE@Za*cs4e@Reo*=P=&y}<E8$`k
zLjcf0f6IeKs=T1SKtJDdH=ArFsW1`q{x}Ho`v(1ssc2(x@sViN(SUn-JQ3q}ac-m<
zne$o`RHd(g3z4pyJCKbF??skC>i}^dF_+ut*Z9?A_Mq5xk#Qn^X~|T>=J<JaL5czx
zdttkp&Ns~l-b;=-oa5HSA5%6P<;7>*xsuuDu~>&W+kkNXB4^F}nhh;Y$%#y|Q2F?k
z+VAvf!h<Jb;1-gdHqB;zOmJ<2Uh(9pM)h%Qum(9%j-VrB8t`Iv_SKP8Ow=T$R!{kj
zNpspg8zXYXh6jUHGvB=d-TXjUmB*6dugmd{uM?AmiET#Og&27~b>F~)Gn)rb*(5;E
z%_d;B_RX_GW4lV8WgZS&E<7S@JRn>W3pJ(7q9x@0;u?AHXmst?^7wojQ}CL9yAFF!
zzfgVCphxP(sni(o_s6acPC-x7Sast_xq>LlrR$7x=~T<q=6lrq`9|3mn)kGg+uHJ@
znSA{7yLhR^IM>;7IcYn=g++VSCXO4|P)&p39&76d`ff)CnX27->F}@=#<chvh5F#U
z)9eS`;$%B225S=7H?n=QEqqi<_^i<68k058a(q}8?m6ikwt3Ogs*Kb8r3-pQBJoix
zu@;=v2dT|4@-+%Vd==G~ZCx?Lf&~~tYKow-4s5F)Vz)&IkI;W|6(+gXs%V2^eM0j=
z56k%|fKz4yv_Iwi!lQ(8!Ml0+VPZvOIZe+~gJW#H!H&uuMglDwx0dSc%36sEkp`*5
zNEO~gGl{FFgN|aw+e&UmRz768(W`Jg$wU({(Ge^kr26HvY$*C(3(Fq1b1LIqg}V+Z
zWD&DK*6qnMX#Y1uLvDdztFDx1$&PU6hh>D4w)WggV;s!-rE2M@q|dL^5b=<IIh+a)
z#FHS^#}fF?&7<yy$Pc`m_)qV-U{-l;w38<{c~0E#c>$fWWejcA#DktPsQr0IM_%i+
zLJxX13DZ20W;~*p2WckMir0Ud#G*0@TCl^AvA0ugKxSua8#TW1=D<`q0u_+iwdCH$
zL!|&STn^M0$*nh(Dr0<R8_L>GPH=Qn?Fw}0f_ZH2DpbD}0R6ofX(8#h$cS-+HT12k
zR|NUX^f~kzQnvxLWS3>)c&guZs>tVNTJ*dS!Mmnc<K7|S^@}zE&31O?SO|j8lYpb5
zdgjxAPsWaNu6!C>uRGi8zN|x!BJts<KOoHbcaSIe`0?NWcBKpUn*pUXV81A9J)a2&
zklE|3RD%5>NO?xmXfO8u=%j0Wb}Z8|EEidf1<9isTz#5Z;lHR(11=V4_gTrbqcn5W
z2SKS=pGc}Z{uFsDcV%m2XT#ji=kJIpp6n|j0M?KV?2_v5EgebORnnVYFc{0(BKQ28
zMAFt^m!)_Xd}ErzL@s-Nt<E;P@n6TP9RCyRKRTfP?0?OBTm+KhjebzckVbKg0o&LC
z(obV=hOD1hx-GK3SzBxJz8I>pDkdl6Xj@Ahv!9H&6c)brUY)JtC=2Qie1DLU&Oy6e
z@@R+CDF9WRl}~7e7i(2Z)ZkLftltHI`pGz54HFTgU^kM8L($#JszZf<-_#r@mtCIf
zc`CIL9xpP>f~p7Ek}qp)8Cslyct^$f5d|BdX{n{%QOEM%Mt`9oo=w6ClwxWq+RK)-
zbUH(&-1(kXv4ef7D7x7+a~KY;n(X0}K<!B?NC>=1H~+>4eal*`ML*3l;Z=c_uev#l
zVccB%7iwCB*K(teRuB=d#{8aT{Bp8H+;SYw(L~|xzX6;&S*BJjsDZ7bM{*s(H(B^Y
z!xl7!g5h9>tA|8fn$j3v%+-7!{aS;jZQY4H%LhN?{ml3%KLmK}N^$h4x;hDapGVge
zIdI}@*p7oqXZiNFQw(FwiyhYn&)X7cJGD#K9%6+0-l6FyetD2p(1iV-#k-<;o&EC|
z#*aFK{cJKaayJ6A)me{onlpVeAH6y+=~NT7TX-Cbuq}Mj04QlH+OM>~MF|)@UO!Hm
zU{tD5{~$3%w_A%JQQ89;v@+v>T0DxDSMh;!ic9DBQ6#ZOQxxDpgS>a>yR(eQ$#kBJ
z#dq@TEPNQCn6&<avAU_W(2YM8eX3I#9e0zxkR~rcGTew*A<-CyaVUdIxkL?!b6JV>
z^wnY7_{Bs~F<ts;w2JM@9#a2`%v9#)EfX^13`;!D%GThM*kzR~;kk3F7z0)hYP2P(
z&HY=JD2zhqQ}Xn*Tng7(cGoEF1%#O7l`{ThKBQbTuIk_N=twFD4{y5f`xdgjqc<0*
zWxA@rr~$wckX!`%<*Qs3K;<G{Dlo_xy3XzXfyKkC<wsP~@wI)&SBVuM`*$!XZK|hF
zn2z!ORDO$&pntTO9JJ85+LIaF*oJV6w;0Ew@V#cgs2@&dda2}x25Orz6W6(Ua-2<+
zOu4b4Wz+GP!!3|>KDd43>~PY<`u1CEolKn}jlw4ax#7hu;6<>w6SGWsKkJWashzh-
zgZAP%a|(61&^s1fLt-RT=nF9Vh@$7#kKUxFJ9h8@W%Mp?U#>snP9}yQ1(4VNCrUAi
z97#gey0A(_4G=?pRRCG;*-ikalf=Sp8|tAFJXq-xF|k3`2{kJW!S=TxSlwjnZtq$z
z$jstGUhE1LCO)>^4h@+Ay^`0_&#O3g`*vN!>&~N;e77Id@z1<8xqSAbbP^YX`g`}6
zYK1T^JXEgyYvMS&Hewm?eSG_ZcG)4RQ2q0eP31{G#D0&Cj{Z{*dgs3RJmcaptP;xG
zzn%Sa7IyHur>dHW#RcTxQ$Rdup3LT2Tf%v`)6|Kj*D%7!eCb?jee(1W>z?;k2qPt<
zLd$P)(I-5+Mv7nV)+q<)Z#(6}=XJ3mmgjb2kP(sFOC$%saN6|-m(SfHxnjMeCrde)
zy6yw3NfD0$D*}zZ%LL1_q^2*RREjQqLJt-p;Kv)uY^cllZz+?6TpasR)j3u6!$A%Z
z<9U*R^P{`<0}&nK5_Yc1w2E#>zYv;z5JsxINWow?Y&Fc@w1Wp?zGD?mjnEvGuTZ)L
z@&39sW2;*+a=k;x1Jz96lQ1PU!F;4=W;SR4*4vk=?HZoJM}e9Frn?HBE<xAk>eR|#
z6+MYlz~qo!BNO~Ke4i<lhub09B)0r@W_wkQ2yG}2WXXUlEZi?~5Rq%mCfbww-9cv#
z@rCYsl}N;z2QBe!R63PCF9E)tgbLi|5&VcDc_%{nmIa=uXQ_x^+ayzQ1=2eZTirzx
z-s$QxS`<HVQZW_Ei~{skRSo>KijHVDEE6rGg?3|i#pzKQZhKU5!Y#$G)_gx1(S@Wa
zjM{9U`pld198_$%S)8ndchLo;`zC&{HV45{C=DO<&M)XKAYfd0=U|^N**tn_aPI7Y
zoaKy~x|X8s5NR60)S-+(fGLkk!nnE?<xF50*ltEue^Fk^sePKoV73%+3<_+s!kt|?
z+#dCS@~xp4B&w?B2@xVEDc^3d-+&@bDLwCJu9U<~x3J}inDBMtRqjxHeWYbH7wXoc
z!B9f4!G6Et+(O!wbTI(i0>6SCj8D0_(J@5gz7JSr2m^BmN+fQasZAOb{VGxBdq%j}
z$0?)27$VTUo9)j@Qe7k_LK^9+Uu_!V#*HL64mVu-M8@S<!zNF!f+~GN>|yhjcIuY}
z6Wwp{>re(G*6%CgScmzHR?3O~hTJ0P!C4EaYk0Ke)y~LlmScK3UaM_XzVaH&(6P4t
zyw-UQDYs}`E8aW`HgIdar$fAEMJ&eq)PU3ta~kAZa;g7w`JmfR@+w(l*F*8lf4K%}
z-Xo49ml%u@ny}^KF0j+_g5U{%m?~)(mJ*bZ+Ws6*7L5!2w^6-ty02|cb`*GTNwD%V
zrP$2aqRX$LeO}-La>#u)gsJ9QS_q4wlvhuqz<VA%;7!jqV&ygznidDY(#e5dYx0Jw
z&0R%6n`ne9eHa=e{IFp=E@iV5AW>z|$3D+&1A0Ly?6`;jp2J&P(ZrnN%;O%!OT3p>
ze7TwLAK-wJ8Zrjx$v(!@C#2jU@_pr0;wmMqfwis>mYx5sWW1&;K*E67=~l$;^w$tb
zdt6_5W8q#tYhAgBsQ{kfBaYBo=s7I)h1#HyD}LpUjOml+cIWy%DU<Y(lER^M`}5@w
z`H~sCa@{1I^Jp`^d36|$EWs)mnYs6K<%(Fm`x}L-*l%`=(Y?iT6U6Qt0JpcJ3)Egm
zc%Kw0c&+~aPqD4{#%uBaD<QYG*fWk>m^=1UYhAbZHp=@Qj<$3(i%9iOe~u-lq@4ag
zEA82Z=PqBG#dq(MY_p`oV7Fq}GC`DJm}yut$d*}12rjO*f}ZoO@S_ZK!{6K&8zT>$
zBvz#Q`t3Oha-&UUm^<}<w<E?B?61o?f;s;=YENCn!FRUa0j^a1zhFBB+AJ?a^U@g%
zTOJ7P(_iIt?-JSEm4^@pIF&jNzl*6dY|&KLIelx-uV_}aa^7;)(Ve6wWTb$iC@hcb
zQK)<avITQU;U9L1W(Z;FmD#73@I46)$?TdEOuZCvr$|(8oEe6+qb)!2B6K|jmf9M{
zF%r}jni!0P!rVM;TUv-79^6j1u(!{Db4w3NGQM?6ttf79_OXf0c0H6c^VU-)FVXT3
z`^OGYX}uVT4cmWy2DEX}i$G~y;ES^@9R5S8@aDLPz~1T;ZW(qb-+Bx2=D}70+TKC6
z7@7P+9H3Oy{Tte%xCzMk@scn_J6@CZ{R{|wp6Zt?t*u=EBnX`%51jqrPY@!DHscS4
zPE0oyI-J$A&VeYm@ieYviK#8cUkdb^k&Vmo++_&I)ec5kM{9-fS@b%{#Vu1U5;&a$
zw$y|;+n$H#u&KSq6N89h0vi^?<)4h#HAtImDUQ9*XYk(X4iI}CPf;77h|E~QcJop5
zvDVwGDsRU#3_jv~+nJ*L=4g_!9NIbQLGQI*^cMWlj5;kl5BU+46>X**?`;YJRNy=m
zpYzyZ$N659vqP4yLx1((8Ijz5nN&T*1xJ(bx7o`>Y6?iLdp)o~NtqKe1@Q4nJZEBN
z?iFV`q{mBS>y}`$@u8EEkk{<{w8rGz25f<(it$gz2%;jwLgy3~1qjx>c13SP#X%W~
zEru+?%ZYXjKAIj4$Q>PQ6{|C5!LV=p9>qeHyjDEcL&v^3S@TWmB6Zo&cp@qIL*$>_
z<~Pm%Tq?`-?_t?9@jCfYJh@JzLvm7TfrdM<m2(P+3D@<TtWw@$*@S1}h6fv2_8>8?
z<!G1S-flM_P`I)FFle}lkV4QyjD}wxTo8aK8LhM3kdU;S9*x`;xW6zDb#*EzYM%3<
z1^siDjENyYIaWYhw7sDi-I*kbJuM^Msg|$lOI!1T2lA3o@T^&ZBx!+n@wvL42|pzt
z0K&GfZ-*TXEPD0R%V6r2EU$eUd5xwbb$!AEY^j65pH~H;&#q{Vxfy=#5*>Lw%Az+-
z0o-(sX+Ee}@!k8H)0J#V2M`pg)`9{xvc05u7#gToR^!x)o()dfVsW|R_HBi?v8b!B
zM&=H%r+`(H5-pp$40cNpmXA{LDSaMu2aOE?O#*$F>(1vN0{nUgrKWqZ>J9PcX^d8v
zJ#88=A!Q>4uw;xs=*Mf7o^M~DEHHCCykk}}R8m#<V)a2Kes!q`!7nDSt~cG62K85^
zJK1p~xzRU2Xegqa|BaIo2XTrqnK{nwHbh{RAxfVd2#&b8jXS$98)Qts>pxZE<9rm)
zJoX#&C(EMu@d}e};@<6ldl{K50(Zz0`@Ig!FAGw}oZbKa$~(!?`^$*v{_Ayi{6EA=
ze~Ik2b^!1et}g8#B^+9^je`=lj}rfKfmX{Cpw!P(zjyY|?}09;x=bcrPL%4Sbko;3
zF%XHp709u)ocdKI5ZDxzma$q{ya7IFj#O(0&fTja8D0({v$Zb4*|-SVR^@1%y<<tS
z77LweikOwAn3EALol!~<926RB^lfWD{XPK69qO-ae}c*gZAE&WgDzNA7!(!%SiTP!
ztAq<a9=^YlbE*U1Cul0nH1Xj>2Sch@%seiC<oHJKxcN^+oWdcKu+*YYy@5}bthR3P
zrDZpupml4&jIwPlv<VSDPK6bI2IAdvO&*F9E)OFb4*62&csV&xPa>24a>31XvmuBE
zm%3jc^dJ<tB(c?HjfdY?%_&P>RKk};FE8$POphMEl#P6ZA?FqI3C=h=#i&>xbsBsS
z;A_x&xjG>eWftN<Z1AR-A(R+fk<@oa=1$nF_e~QoA52BSS=WeQkdas#o#!>C4nBVJ
z?arh~#jBO2)Ml#NH?&0@R2b)p8>+PDPlyy=ra%q#yn&L>%d5Z9Ao7a7!@Tp3iqNbn
ze3(6KD*_?oly`OVdIpQkNszZ&#>>t*xD2|qGKBL$r5_A4NYc6ZRir~1**329Dt}U2
z8v#pZu}{<~f%kftYpZ}lg1cu<oAw0Q$kdQ4cGTQ|=nvb4?QiYYj=P4Lq<uSR!m>}^
zb?h%5N(%J0{GV1STl|B(c&qdm5nSxBn0%VNgX!pOnT~O??7sV?ono7zjUi}-W-e(S
z>ESpAvcGk_{n%mnXa5F_%*b!@hB0vIk!IAwBWhkI-;dkc_^>m$ESyCVJ8a>hVwfR&
zSBHu8P^v&@w23kfvstw9HP=VXJF<+br}WP(a)RyRu(SMc)Kg4er%{b%rTAS+RAKzF
zod`>ns5Cr9$NnQ51)I$sM^1b$-XutINM4zGoIEE%_gf{#B#S{J2J=~i_M8|7%S?Ni
zxa-?|8zUBO_$ESVc=Wj3P3!v%d$YC!HzLGmjxC#o`Q8hm?B)M;Y;)uJyS(D~cMqye
z3_rrZw>MOOublIF7o)t6X{MnaCq8f)IR);X-yZYmfqyc)W9R)U>}H_-s@Lc!^23%I
zj^0e?M1`MY@JnNG=h^IbhB>+c5%<}3{N5Y2hxXGzM@Q+UuKxOpWfQm4*B3m-e`6!^
z8hV|n$dHmdRpNuYV!l3zh<cK$xgAMjdAl_}B~J<5q1adYq{i|K=i=<^-Qumt4##K{
z3JKp!mB6usAK*z2a2kW1dQR82i{Z|8Si$!wk%kstX$<wj(dfO@5E?FQqJKmCRG$0K
zX{6L|qT<#)h=zoWyvg^D9Y{D}Wm`^6OEE1O{PL@YcAta`S*S?~NfdH&NNX|LP|S_f
zg%<%rL2w{Il8@T0yE-HkKU`~7({*g4O(|@*Bai=XvS02I@LI<yL<settkmBFF#C*|
zM)=QvvZV{Eo5|jl4AJMimB02$<njzdpZ(63`Tlhp+gWPSEL~d?HJ9OS71W3@&a^BW
zwJX7+&_N~!K(ObrvO5<4ZDvpg*lZ|$)TAt6b5#Bw;eD8ognOA>?WcWkai09077ikc
z8mSRe4&Q~(Mu0~h!f{r|sjxFk^VvWnmH;pGuM=(vzJO2Y*_iPH@2XT^<-~>u;CBdl
z$h(cw)1Dg%^1DV};ArI9#7}cLLGia8f3Ew1;nYA1mgiQQSfEYINh^0QV9V>=tl@MZ
z^UEJ$lXJd-4BUKon-GD-dmI}*DWwXXf0oUn&kel^qjLo#GfT9~-3*a;CS7j`gW2Fz
zhYX1>aZbv%mrh<qj6KW}F#GntRHR<B^#_R!aWV-q2BJZ_kzwzpNfHAVlIv}=Ws*%i
z&utQ{PU-UO8?5IN*^&{lOV>&P<J<3f1I7ZtRAJQTX2Wj<DO9rLxT+Dk(g$W-a2md2
zCb!T?eqlb9uf-<ABv15mKC;dc_+`VFmZXqRvM>>W<=j38BVe&IE;Jv?ft?g+g9EO3
zZD2p_r+@A`GK9V8zxi%$2)G}I5z+X0*BQY%(5Rw|_d?`w{q1p31FR3#<f(P$mv6jf
z^73VCR(ksN(p6=UYJX}cd&kUkg(^j$hI~;^x47)5U6r%I{Dx?f?bLc~F2!B`mZ*IZ
zf-87-(DGokx~fs2>A5gOw|IZyU@9Z>-)*lF-0un7E&(xJYM{#?Q8jn*6EIQsS-fgl
z5r}b69}V`P+mEk(q|OmN(yUbLLI?M<r9gt1Us(<pwxMMDdjdCuo~`ez26`W$I0*3)
zk*@n0l&~Z^$Tg3qyAL|`^9}Vd?7O}U6z96@8n+aXC(EN{AlC}4+B^};CYiktQ?4ze
z$WVAl5^R|wR3uwBLz2ENyP7!fQg)10VzAPu%&x@baH<bIS^g)XG84+;Qz;yCN9COJ
zx~`ImA3&LaP@oORJW^ISVS8yH+U*dD^L;0q0fj0av~RsSstjZ2<*UL;bBQKZ#Xiev
z(nxcexoR+ar&&l(sCAZ6b0`KRNutPaV87TEcFqN;a;zDbhk-(o_pY#KZtL_`cwsHQ
zslkiHDa6ZUJ`*u9v_2)%Cw_^o;|`*Z&!uGL{CndxA=ftmDYc`U8jj*fu=t5}R6~c^
z^}`sHxyfvf2AMrU{a-Y!9&6760?~^f%Q>GvLTRG9Xwz=u!6}A*Z)pX#uR4L%rYshD
zN0}_s0#7TeXSRg9^kl1-C(wqL7MRP1*nd?S$kuXdQAfPR(uGEHkj_FWcwb)vE??2R
zb!PJd`8zrO7Ll(zK&fu$6fY^$KKKY-!1P2cYK!-~61VB|o<ZJ)&z?INtE7cf<1{>Y
zgkBp>)Vir5*eEKfo3vnG!mh;6OEsV9DC-#%=Ole3VY(5L#UHFn%684<5Uvp%P(8Xb
z6PH}Hdmx#X_tL5z<EC7KMf}WNto7NHYdeRVlmAs%R=e*rT_I!Fp8KyI?qq-U8;!`n
z{}PU-?!FP)=!L{y8LS#M?yDX|87$rKFrJ+-lb-yXzNc3`onylasuXT@z9flGd+PzX
zzAm)i(&k#wE=UHpqKS(a!k!_Cfq`O`46a>fAO%RDHiw+Jy$FwbUsDc634VNL(4j2{
zDOM&)U|$N?A!dVPcH78;Zs1I<wuOJ|XiZ5Bg;?BOrbM$#x<+Q&WoazXul}st(|>k{
z89o-rJRbUfM_8Ev-xjD<HZ3*0^e`vbxa=&(wUc)DWW{Y}-d<X&4u1Z)R#_5jLH<20
zX<-)6pTon%)2Cz#1RqH_!eflV<H=6(<w4WH9Y$8y$D-?{A_o@nc$R3e2TGwrQ_`tg
zI?7rgy$epCqD&SqY6XKFgKUwXXEK2~VX9uq%U~eGmjZq=7tAi5737Y1^%x?GCC)KA
zZ(NF7CIa@jrrtEH9oE-BC8Vd8s@f_;m{GSdSV&$rb6~7;_A|=nik5$J<Ruk#)<AvK
z!Vwdp?e<q(o%n;Lc(rY-njn-;g`=j&_ZxFfsxC_jIr=irqV!Rc27`W86v{%A5xQI)
z@}4X|Z)faucv(RDmS4Zv(B8D6hnM);3b=fJ-1Qr~hhKshvY8&0+j$1;#x9lZM(Iq+
z@#so)!qy+4`ZEosLH(Dv(uEk4FWy!V;Mv~kJ-yEL+f~eT+wjb)qshX{G*f%+!cqW}
z_Cm?({63kF$kCCWA*m_v-3SN{aF{mDv;@#6{XlyHL7hr*U1@FV7IH~p3}HY17A(^d
zZ;z^)i@HR~nC0jU9qNDC*))iE4^m)apBm|r-|^&HnWC;TG-m4tA~s!{<cJK3+9fLG
zw-(FM867lk?3vo|0JOzT<&98M;%D+7*6|@60`FWJAH05yWsio`L%IjkO4z8l`~MEF
z?(_h?zR3;#dR4NZ2mHjH48!-?k8=rfnaW>~N!}JA9mbOBa<?*TX=H&<Nx70YMsXp;
zBiyLqi=1!DS|Ha`{>P)#eernkeU9B0pPJCr33rRZ@M?+T0oXflh5C0zN7rUGfd{?b
zv$C*{waGb<aI<4H*y-$-xcRPKYm$CX+vzw%!*(a8{56VqwRe*ifMQ4^0vHv%QeYSP
zhiFnJ0ai$rbwESwoFY4EfS_HYt#M%Sa6azF+uIzW`IMrfL}t})8FG)0p0Jm{S}F+Z
zP)-pdikx8S6BN|TvIwDbvt=Oklh*-(2I?BK7QQ115%dmh@mB;UCO0VwHFON}f))C4
zsvgV!%ApTr`X#dsr%9geoCpFCQkugC$rcbN1}PUj>yIqaeP8Rj$P(D3m(J$X?kN?I
zTzaaP2(|o@)`XZ@^{QqJDBDug%TGEc+`G_iN!3HU&zEAJKkSDvIM4|Fy;TPhNCN%v
z00<Cm4|Y~GPd8~VwX;G%Vt8F};S(cbWRn_69p`*+8o@)UY(tBVGWiK*>9W*|12ONw
z$3EY7slGLG{9E9-gopN;a5=8*bUpaI^V90{?nJ*ATFLwG@a(!=UEX@_y{l1g8Rm4Q
z{7r3dOSCBV4p}7N^>M@<f3>xaiQWHUdH!nSNwNfFIs+4R(TR1thG&affkKg<+hI!D
z^E*PpB-4=iE`LB7Bllg7&o99|DD3K-(OS@{Ic@53nZ@DelHG`#rwMQH>dZvmr$X&X
z8Zq|LpI<JMyS`3yxkl!a7l@;q+c_TYDSfN%oWP7ho^D9`X-TO<q>sIcl{hFOOEu@k
zYnx62xFgy27pB=cW=RJ5n3H`~#vJLXRlx&dS>g6xT!tx>MJH!Im}+ryQtYn{1_}f=
zfnA2kXX)VCuV(k30nH(m*7f4eDD3Pn`Xh~y5@~=k_{~-pH_V;_HVR_~u9p&q#nvUG
zU*PK2UaW&eO;EmZi+MHzM~XM8YWbt|R|VWV^nipFp&H?rDWEg3VgvBDaNAx0Kl($e
zO!R;&J=i5d1Od!qGxSDQE~<t1)CKe2_<F~c*`jTbmzeT>XrpDXq3r_CJ`wsibQRp7
z!0yO@^0=rhlC3k`H<43z5Q!-l8-sJ-W_pwhnM6DoqFAQwgh|Im*Wk!dlm3QQ2CN1O
z=->EhrTw!fdpZCJ2_?nFm=Q*lc?l8zCHuUsJZeJdue!UToy0LKM7Wl&(Q<KmWP4rt
zkx5^l`m0aL9fSw<<Z|)Vv(W26@H}pT(R(lb!8)PH{sB%+!4By6?l<Nz5;Py=ef5gQ
z{r+vgRST}%2h)ix-5urBp^5Br>iLsdDPwj{mY55WergWQI)y0li67>V7xvu*u&c(L
zDLlYC!g)vsrNTBWgp$82+VZOj6G$IG1ls~4j|WeX`lS7&v=*~!`Uet}RC%j+Ul!uO
z$s9(KgcLARZ!1WVt@w#0^rp+%lE#oNw$?nnmm7*0c5s3$O}LdbeJ5lB%0%Y=EzP4R
z#liB!{*e_Q3VG!gglE{PzJwlEWRlr`<^eg0zJ+1x<J49_QUgLe2O0F2+EdL1K0Vr8
zyjc#a=4Lugn+C%LbDX>UhOY)lX1ohzCe+-GtUCN=!hv4cQqg!GkQn%}tj7@P6lIC_
zxrh>at`9cm*k|%YiWSsq5=`<huxwhok80{1ynSex!9=K}ow|BGZxY0DcBH#OuC>||
zcZ~`z4I^r}!%;@TA94Z;I4fV1Z1>WW?a#NY0&JS1lQ6?;I49w65tv;l^lk9Hq1A$L
zQRvZ}`EQ;*|L|w=2B4HyF5k_Lvq*g3Nf=~Aqk+b_hal2<RH(1IB26869efhzKObKl
z=L7}JGn-hg?b8U=`;z4#)B>h9mOY+6j`fK^_d;BVMf(0=`D@6Ku`lef)%%!)Kq4O7
zKzOSgctj%nxIWwBvr-stcnBPxa0t63$G%oMc)AhNxBO;Ua7JqEsNEXR^wZc~CIfj_
z!m^n!F^nhV?Dv>hVKRe4nk}$P<oM@x+1xJ4;dvm9%zguH(QnHOyW}Ohnd;&C%i=<p
z*e6dhZQ-&9_c+CS-}YbW=Zgp^iagG{ILH7~$@%}J+mJ?^9nv3GSqrW`p$*)tyNeHy
z^Is@WrUVlA=6N2H7Wh(g$QTmAe;i&hh}+s7d2&~`TI9+zRsAB3g^*>o8;d@ntl~}f
zXd0_5h%w~aqXP(CcoWqp+dThsEBGH`q1kLwsC-7BvgI;u$((wCB8b6VvfO%Su-p+A
zpYkUM9XJf#93tP^W<7R=MVx9HPcR#AZYYTTn<&(R)SIr{rhVpQ&x13eMU9qiJ&~o~
zTZv_v0Ic0y2uF06@+R3w8&WmZ=RW|Mp(-(xMPK8ORWGwPJ5pZ;oG$>9hLK#CA>FEv
zuYqjJDo26k!pd4}JRty10DgYTztFW6EC7K+lq7N_qEiJCuRDw}iAqeZ)<WUzqUZB~
z_}}(Xk~S~Vk(Yw}>U7fFkhkmE3bDctuWjYc*!Zl62LoQR_kNss?5ELzAZB<A-ImCZ
z_++sWwQ-uDjR?$OjTnDb_TYf@&WFmBxok$HeO<V?CSAE*>Nc+A?p)w=1UjUstn<Z4
z(+?6X8|kr#4uZ(pfKb*UjL?q{wbTK0Mzb$UU_ZH@0hZ&UchWIjk77<d%9izivE{P!
z>>5^J89Mz5{`|*ySGckpi+R$-GEy<_!_;=wMuZSTXVEM}cO@;2OBZo%rd()syWS0O
z0Zt@ng8|>Sb~(N0thM;uJCH__Ivn=Xc5gCF{qDiz)}tcS^=}7fkELQobpA*6fAesT
zsyBE>o+s~@n91DMAcHUQRxEXyA?es`GTK6rsR$)GGRXQ1gsU+o)!(mCWYO!CBeMiJ
z?2w`4H2rioG>Tyd=~u8>k{%a2$Nn)5`4brN`sP79#O}hraC#Y~7fqaT-dfY^Iq|{3
z$~@FJE7;JW*t$%+oF$1(H?C|trIhD@+>OgC4rNS%VmR;u^b-6fop@2t7Tr~dXP!7r
z+zWkL#AX<UX)hTUs<=qsqdGjOLYx>tQNw1y!hLF#xj9T6(g36<-Yct{ozjI2ZHoZn
z0GGAp?EkWNl;l|!KQ6|sb3QPQ$eLpEul#rYz_gW^+%?EB=QU=sBr#s&Tg(>llH25W
z`+b>DKVB1S7-j=6cz$@yDgUCuC9ipG`(T{r1O$_pzIyP^gFcA2+s(|oDJ4aW8zWJZ
z-3eXzD6BP^k%^<CZIEH<aGqEprm9Zch_oGc7Gc<Td2y?~I8AKlAl0q^(`e!Ytq+6{
zCftthmT!*xJoDS`v?D-hwZ}6FMxKNJyC!85Cfz0|_LBVmOX2_Ube3UJcHtIQ1f&G%
z6cC0Uq`M@B?vRp(A(fPF=^na~{z#=$x*L&_Zjc5ksk1%Txz0a-h|VzY?ES=A_tFwi
z4x)G3?Ooh&gigEEd(ueG|KI^ah?8#OZ4b}7u^oT{(W~+B@p|iqmnRD@UCiVIfNWX?
z1XXA7ET@_S?LnvaG!6790AOBEhdxt^|2dQ!25$pq_I7(_El3ZcS}pVkC@OCv2BYuM
zZ)tIq)PVJGH2#gL0bMwPnCpy&_m&H_CagcXtfwS(8}!ZER9Io_K?LETc!y5giMl>4
zky<`}XMFNuVf*)VS*EDZ-&mjfqotPjZQeJ2pQv{$9<Y=G*%l8ba!~FjN_ZPC`-Kl~
z%NwBCVw1ulg!<`SOEHbP272WVu5CT-o<1oVG8)^%^uT`;IZU`cB=LlW7Es7Qo7HXk
z3m_mihPA-1!C4yzw0?jz%8%hbmvfq+0(JDL1IL5oSD~$}C5{?)T$97*Ar;ZK0f4pD
z%|89MUoD2={$T!@ZRy}-(K9P@m<OC|ctISG<16ieO$#ilj@C``Wx%0g`<x=mM?R!a
zeU!hiO0hLn+2oX1)q+rNh;$4uk9)&v^YqK{8nIx9!m!p1xpM9b?o#_!b8zm!KLXHK
zVUm~hObqF~FJyLn-TIbj1Efmk$!w$Xh>3<XGg)$ns{IL7Bx(cMTsMp*;>?AsbWHLO
zC!ArOx((c|I#Md5NiQ^Y*DDTjK_r&?^cP)DV9%}wGCQtbovvu`hLhi9hQ(rZ=?5E5
z{xGL7VD(?>QiBfTMFNcgCLNICLPs31(w1`jhp(Q?0!^&pMz97ekef}huNgz1lmp#)
zok^7{FDW5B9zm?(hCHH+T#h;L`|>UOkolKAG;!Y=yB}F$Z@sh-199wxY0`4}Ci}7u
z!hH#Juhs2Su->q0>;9MO{qXnoIbA~58WtcN6#}PG%GX1O(!-MrFN2Sijjb+UWxC&0
zeF7p@o_G87vzPQ>D|UMJT~s$2)OUb;;imTJSZ#8;p#q(Nrrk9$B&UXSy?Q8a`d21B
z_U}{z{$zSgW6pnP4Oz2>BwUered2)T$QTioD;wlERL*YuU$xZygifh0xHC&**k6Hz
zA%2@bM2{ehRDQBk?}Ls`Y1bNrAD$NhiN;bQ^s=e40gxUrvoQyepWe&ANS%%cM0gW5
zKXAM*`6U=r6fcL@3|&iQ28gkXqTnFeW`2su`vaLk%yjd#CtQ3|V~&H+DC(k!d?qa6
zNm5O7D}F(1)G2C{&ETjz60spLjpZhc9c5Gk>KXisz$Idf#23N9WjwQntf0}GJEE#)
zv8U60&8Y3;w@Ex1j$1>Wd$bECqE5!E(txxEJPW3EuPIsqbah#?(PubWo@57WF^)VG
ze~~@+qxDa_&pW)o{S=isE=@tz;qb|;<8!vm630J0^)OFla|NiU!So5V6Phprf1zgq
zkH8-cHq)k7>xLxK=nXON!H~~~Ap==$h>sEtUp{%u`$9Q1<N1^+#_Jz}WHAAd!S^}@
zczO0BBw$-qDH}OS0009lJPNX|>xrM`2KMT^)u(C3oQvAxl;t>EFO3K0Fn2xB$L)Wn
z!MbX9`<cee`A+k|tT7ZCPBN09l)=(Mi+KLN1_QNV?DJUSRnEs#M2)leULQbCD3t$%
zX)X!fnl+o{Xj5X}%-s-a4X-Xrq{ZigYZq9w9X|&F{SjEW0ux>bpgR@M{EqU@eJYlD
z-S3ni91%=R2LRbUKc>4J7|%l6oyurcjgmt?*xdDtxdeus9~H3wE04rZ@x?mF?-$2U
zS=!H7{0{bWZbmGx4=lxosqQ+dyuY%YW>*5>8fDBg-;L;3ec~gS#p#QpsnvWK*Ym~6
zRWFi+$J1a;FJ!ioZY4pB=5AUj-_7w)tN=teQ-^?8t^n2iFcdNbrllf$c(|6jV+ppO
z5Opo@Er{svmW7^|&#;CVVtI_Q;QYNSB%Wz%)C7{@xmVYV`^xi2kt56egEzrD?>sE+
z6?@g2s=12zUJPKSESZjOrbVNIDbY!O(Bpo4b42CP-9#GE&dzXa6W98_n-@ql;w4+x
zoX-jp-fb}_1*rfVcM@p<R2<A-Cakj^xiR)UaZ8Z_851;}C?baOVO)rWCiRAS@=;*4
zkUVmGPgYOjgX#!S=VmR(2?_$m3v_S3z5@?S=MO^9DImn_9CA`jVjtU8L>fvSk{KH_
zDaF%v?MJS1Xp@8WRcTo{0Alu9Fip#iR7(&&W@kes^8vO<Jk&DRhUl5@<f}RI>p!h%
z2DsMrz(&0^mY6RWRLE>si;`{8X0r}Mzw+XO`;IJC4!;{W154JuVr`ud{4VtCCg%U;
zMSAA#8JA+nO!>FSym`ikcUK0ufYYO*vsZn9%}=`eAI7Mu6t^}<iC(%xJxbn!dff{k
zLhV)GQ>MMdFf?reGK|`PgkIx=sW<wy6rVwOrp@R&-^wpoJcPo)r0#Wdm><xE1Xp)H
z@UPnYoMJnsYczmx2JEs=p%E4TN{~qx_Y0F8>OqoGn#F6Cwyc2;u6?a+oeI#mXMt-^
z6T|o>XZwGBB-{4w-7^!C2m%+_x4)&ard{SI`BChHafK8Hh(Q1~c?l6nM+zx@Nk^KX
zYV~)WTT+{VxxP<g-n`AmbiXu?ZQu(vK16~-kF(dSVhFDqlIjI2?KLQzsIO7Hi(3%^
z!98vK^-aefjdk&8U;|uaZq08D8P)Ce>ZL~~OLX)CoRM@xN}2vNfxX>a<es8rHIzY;
zqy?P(V8OYz+uv8tJsfGuZ+QSDDLvpT1n>Nj0b}|Yw@TE$(?h(J1t|J%%<Jt?1DDBk
zy4;ilWm>YBPW&=LEG-v7OIRq!xN{f3$XN7Y(^JDtwFVv_xC!y?0WwF(xQA;+A0{Y%
zhEZwR;C>Xw))9>5Ji+&l9s3#2JU_mT$7r1c!W5#&E1Gd<FsW|9?gJx=8ZNXsD*^9o
z7tDUmSdWk?YEe=5P{*O}Y&6mTevy6Z|6&u7NGLWoN9G9}rEHm2oRN5y2Qh<Oo&I-S
z(b#U-^j7H-%3~5BuHd739Tug2DGl+j3&3F8490>p2?*4`p;2Ttz`ZJq?`<mNOfCg6
z2tXFYej5Cz;N5_$;R*!^i2$o&9^aiI(bx*K4;!w2eHHiU&RUTwqV;dHcka%$fdtXr
z^?N7Wh{O;eSvNsf)4!dO70^Er32MI9zYZP!V|8HP4ANPbegM7l6*v!u*7%K82XN@c
z0gST%Zmpw-2%%Mg$*W6@M}y%MJ<sqeuyuf^Z3iq}G#A#KwO!MK*|jHiD(CNZsD%EH
z1dxJsqV0E~ee?C9>@gze@g`^I{@729n%lqqVZVLM@80Nf7Rt){>ajiN@6Y14*_7M1
zui(^r_-nfW#t3(R5)}TOsV3>C+F~vdS%}y|rl`Br-s(;<uMZic*~6iLBAJZ!u$m=F
zp)oO;qxaR!76S;S{%(UTynX=CN{7}mf@RR0da!ls$@5w4rdGYJF~k7&127E(M{O~H
z2Uyr|PPqju>_=CKn>49h1H|M`)_3%Z_`b1@ouSLHG456K1#vZVpoBeLlupR%i~<ID
za_}j(UoSE@GoLTmA5Ue#z$SuHzr@kZZ)6Sf1!%w>s^tbXo4hswJ%R&6nM+z+@nvMj
z@9=lwprad&_y2|({!$6}o<M<l-QDskkP#Q10zC@;RAveQ<a;<Yfg}v1rrf7`d;$X!
zFYex04e^qcp0FYtTfoJ1<VJ6D=7wpT^Xl<hqkbfy-TEew>Ue7$k;?=C6lh^(&`Bp?
zV{F{(EitAT%d6crMZ=O#G)J}$A@AvL?JBJRBWK-;1^yyg6cR67KAd7KOF+jo2GPIo
zna>|CP{q>|;G~f}YA1E_s2kKpLNDg2JiuRxfFcS6GJH~QMPk>Wkl|cYYoHqbM^n)$
zrw|JR-XE}c1CXxL<RST`ToYRgWuzmQq9DgUPmKb1JMLGlI#GPfzbL<nVFdFQ8RXgQ
zgHshDu#^UJ;2%3{$^dm3;ALGyDDm(fM~OiO5NG`x5XS@Z8B7VJV4TPY*p`2aamC|T
zrTxx)YcQU;fV&D<a-X<+H+f?9-cRZMLj^ou^|pXk--P<i6t0FGA5{_|%PE7OifD=$
z*m?CA-4G(3VxeKxDjhs8&Z?8J31|%#o8jw>gDBcHaBEG)=w@2VC<~2F4i_S3)JPso
z4g5V3K>h@;*h^GiJQ*z^aaqcHG?o)ftH_3!Bh(_g6I4F{oiF9`gBzA-gQ4bA;xI%b
zKE<2*m9(J5zbPXk$bu0Ie#$YnhwCCR><l-FAg9BEzh3ZpYHwH`c4XfB1KFm?_zEUR
z|8@!nSsa__exYs-Q<QIIge5!miOVm|Tz3o-?GO+Dw0osXbd{mY7{w&1(z<-ujgE6m
z*-Uf(pz$KxumU{psr#Y$pa61d8F@eblD=6sYH)lcg)xK7hi?;{t&%D|lJ)ZIf}p{6
zhTYBGSF)J0x`j|EB8}A#d#|6O5HW9abUYaL_cGQ-)-V=|y>P_hXr{d7NT%1Q`2;GW
z^m3k{eRn*1J0i~}fJzVE4+YOd3{$+eof6ZC4Jt_caa2qKLuR06NH*^i4<n}kV~PGk
zrhmrVTsLoRcTvy<jDiJ2H{t*jW+_Htlwh%&9T!emLxR|kP#m3czYquNtEO){^rOhw
zaH2x*Yr9CW#Q#zI=^-B{+gMYT&;?fX(e{7{Ifg~Z43~*HoZ^<RUYb9o!#m4}F}Dn<
zuy6?QyttO#@=p7!M`|Qw=F?E6tow|KXc6aiEg}W)X@*f+c*D4e%1{Xv9&0R+5RU`b
zfKY21FO}Yw5i}vT^e@(EO|em+-QhkAh^og$T85%OGey(%0toTLjjet`i7`<H&nuvl
z14P4!k`h2nrcgsFVm?pOe+5??NxQ*NlCdOWsHH&QaiqdPYtTPs%d<RBK9z+GmXIeY
z*+x5n=Z)i4lV{7j@pl>70O7EaXyp4y#)kt$Fd=}dx=pT+jG!K#ef5;)VfTi2rEy`$
zsI_fWuX{UXrG1QifwuAhjo&2`4t#_~zFk0M-xbIxZBMC;z5(=LgD?5ywc`I1A{+h$
zxYKl{MSa{apZsv0@3=5CY+cw|UL$~oP!l<77!quRFvU4nYpws&VV>E%(haQ2IT>3H
zdu~8%09cP{no1`0z>8O8=D>{(^xzf>7R!@+9=KprUrA2%db7*@bW<-Yy`n}3IAp&$
zZX|*L5^hD1G7<>MFDL_}6~O#M<e*VkLZ+Cl%YRb*jzMh-BJV9oGcIfrmvAFn4MFn2
zwV>JCS~Lv~j_8%1GAS|T#Gt|Mr{QlC!{ny`Sh%M(jJT`6pfbr8U`{BnL68C{RJJcB
zvq08dgf~!>`Pc#{EdP>|j`f_2LC%5<*E(dwd|_>m2e<l-9Nu^9j-?oXA=`)%U^q#{
zN~)$aZYEa>^}FbM1I$^u#0i@M?SI;ML~s9M{{!=Y7GVss+3=NOg4z1N_@0eOI^Jy-
zfHwU0zOK=cpH5EQH6(jN<2&XP<55ju34Mnk2!A|zM($FK3mCm^J3e<yl~kKZCjjKv
z-$=}chkcd?0%*@Lr_`afelVFPxHY#tJ&yv4To(=qdH~Swj-OhBQ~u!$4$Nxtfe^}$
z*!Oj`D#9)`isnt}fa1^8kF$rVV3DnFTfO-0X-tiqTPTP?14&;V*q}`wd)vdQ>^)h+
z`6=zDnyD{Au4@uvO4efzma65c%+3}XyF1)zTj?O+b&<_~V3?;Y%q_A+sEee28BcdD
z_FeLp9X8zVqWAtDw?Qc$VrL+*jh-AubNNQTafOi95QaeON<4x>!ql;)Y2>Ubtci8&
zFh=9$TG(^=L*C8<8D`dJ4x?5&>C%|olOFY8J-HnR3gGS7Eo6~DrG$fSfYPW$jZ+hG
z$8iGaaJj^4buNh1yg}|)uOv03u0~@$oNYTM@DoaaiNZ3UT@V<6fe1#f21`bpKNUHu
zHHbmGblpQ+l~_}kMphk!JOM5<;8zpHIXo@&^Qjq6vH!$Kili#`2WB(HugFf279Aql
zt1gB3ItW71xQE#_!^bqOA|L$@?%lps?CCDQcvz;GX>t1Hk@H;zJ4>0NE>16=Fz#C?
zk^uR$oDDvW`hdJXi&=s<ain*np(4G;sex7fM7+Fc-{RP*1Dse~jzo)FRFs8^lcqAd
z?@r5?)!h{SBZttYxR+`cE?XYgjd@&o=3Mz6-8TEqE>nL=<74Menr@CREXm9sJilc-
zR2#M~VDi3k4ZX_A+0o=0KZR%a+q+E=rj@lk`$dZ=@rKtANV|0vS5lm8Z)hFqLurTc
z?DZ<vLO3M(;FhaEC`Gf5G8NTE>@aSJAa)CF-Ls0ue-R4r+&rJvZ{0b-s)k^*JeSP5
zH|gm5<4n=V^o(tDyAzZ~!MP>;&5RvD%#&pOd3!k+&zhbyRb_M&2d8i4;Pzp9e_X)7
z^)&Ljcc0bp*DI&Iy^|C9q%to_MXl|^tk)<S8yjp#K;7+k7(i!du*R3Y*i%J_5z=>i
zoRPEsL|xQl&p)13xr`y*!!zNDfriJfY%m~p6&^nAC&ym!$sw-yvU3q$_NpITQ04_>
zM(O0$U(3L5ZCtToYKuX|-`;U}85CIPSYUJhRm_)A@~IpEeWfsjtsXM$-yLBJy<d41
z&!tJJR|4q&Q6)@Lj)CMwBPm$FJ`SX%(HQU*aA6>e=P06;api6dD}8!fY<cc4Wl7UF
zS4snfzCAVjJ~IpWt&BwUp;(GqT@#u%F=O3jTFb80;>oWCpNsb#B?@e%k||h_Yb02!
zNW~OZ4k&h^K}b1qb$L%DT>pwX#+#CICg)ZE7^qTrW}F9ncudKpxxL6y5cKJ7F1P8}
z_D$rA2nN_6H^uj=^YoXb38N_Fb727EJ_xa-pMshzXakGcg2>a9WQ*{c8W`pK3U(s`
z|E(nxFt@~$I^}-GQ^18C2eI0GQ7Z6EG67+Z-v2VQx-eqYZW8Z+M=lC`<`+JP6D9Hh
zCq;z-q-56)^s@3=J^%4wc0J`gq4`)Zy=dJcA#B85+_4h1;$@OC#dWi1C13oT#QXP<
zs2N+HA-is+E;Pv*h?OX1n0fj1%zU@b8%Y0hXes2T^XY!zhbBTn@4-Lsa@{rhtN`n2
zR<uUf?9(TiEF`UucNBE`JGVz^{gLYLY?|!vKCf(EWW9g#aBVLfN>v!D9;CNS9R-qb
zVF-X`fbWI?ZH{9;Pmu^l(IB-(sJC*IBdZcjn=T@`Q+*y@355_G)s{_vLU}S5tNwH+
zfs@|XDOF(~24vEX%n;%b`b7i|G)cg(pvO*5SXL;(|Fg;4!~#tx!MYM@8^7JEnNLm4
zLIgCoj<Sj7RuSx01$j^TN7>({vXUhhi|LP(L$yA*dQyUQ6M^E|-Eku>2t|EA=s0VN
z2r+BR6IvpD9_Up(5GX+%{_K}kM66uf5Xr_au&GYuNLiJ#C%%e8@=Wj_&5oohguV+x
zAC1*O7-gCRN_Hn6Nt(2Hs?m(#o1E6xe&rPg`-s}aT!uN{{vwj(rGPI~BSfr}B5L`+
zNy_UO{(L`qIh|idlZ@BXC=`aD2cZk+0$(gZa>mrkkcpku&HC-Mu;CwFsi3T(3UP<$
z_XEj83C-nDYhm?F_O~hG#qLWTlF_lTq@IAKNk%;~Dj(krwq+|W4e&vt-%NEfzU+I+
z=~Sx7EJwf^y;Tich?L#H(!~`VzhwNvgU;_)&?ThIdqYLsz?9DX4?p6g=Dm%TMCCQ>
z*yd6*3<;Yc2%c^pwfy?ZfgOWiZ+J3?=nG8%tr-P`W_z3<e<3q>a=<p~o`5al`tE{P
zQp~jO2>7OC_5}nt;-+d)%;naPg7J#zxzMHFIV1;^Fv-du{4pRZBAtkbfH)8jYuk?R
z3b1S=;_>UHT!uVinWr&@r*4$GH_3}vUkz9OAWdBPD4t)4W3?O1kbe(*#s?f3sZH^N
zT(&+=Ar6kr@jxZikjAR#y?Q5X`wE~D>>@k%@_MgUg!<vy{yaHLyUqnRlUJt8o~<F|
z^an4zmp?TR>REPmP81tkGggl<upL)TI(qb%2p$WHbc1M24~?~^3(@*F=V@iEM8mwt
z{7c4uBxp;D^Xp<FZG8inryGg4DGGM3-a`V>3L3b!1PrZ*Qm@#E@{=yXGUowVGG%`6
zHZyZDpM`USy7Pr8aJh2sl}0fSXIgCwOuFnU24T}H2dCa_KZNbyuIH}xno7N|)gp^g
z?&P#4a3Q0MYzpM@Q6kjEdp%E{g<jX%X~Ppr-UFK1P}sYVzk{NuqM%CiudZtuo?7Tj
z5y5mrIHD`{DmTG_fI0;o<{z|DLHb2YrIEC!M3cR+i(kluQ>n>s==PsVf>a)4YaJd7
za<Kj><Y+iFxYZ!xLRshy`uOy@FW-x~7Yt2?NM;rR+!mvfES~I-#7jo=U75`fNg7Jx
ztyG>IS#n{8gWQqyGNstjfsUz9Rzo<z_+oJnt-_4NLKaVx;su{9>0exDx`Y~M7#x#6
zpStKc9NXNMfc;FeRN(WOri!Hv28sqq*g7pF|A4!|r}0IN-QBf9MmeFWmaXTv-+d<x
zyk@XT!Il=ShMWR(VcBj(Q2roA&S;J1D1x38q%+i5oh?cev&oWN<L1A$`)|c6D{b)n
z#3sorCFp5M{1nstqw|H|NtVHrhkPTSh>Qw!Uq*V)qBqqBvEu~%%Px#0WZ<C@(S+rh
z?hmK~BK&LOhR}n{XSsXBJ=a$j{!U{rIl{ry8}k!6^SuV4Q*$81vKIAehEHrRIj+hi
zmoKC*QiRWRGtx>>jOcG;AYjts&aX4T&neQXMHV*D#FQBq1f#$HPVb%aUL=1`Yxs*2
z3~dLb$LXTDC=5D(f|jPp)E26=argr(=BJ%}(b(urm>gr2;WIWk9=)cs;MJQX0j`Yx
zAJg5o{OCdR)KOr^Lsa*b)4==F!wK-7b|bSz6@RmIMAIZh=Tih=0xbI1!#tt$dEr#W
zQdR5=?rMOKALJ+>_(W?zLduq4f(%_Ww@CKjMeq5}fp-A7cL27bl?#%v+DU6cG~7i(
zOwWXeH>{<mL#3!DGRR$rCIy)XXO=uyx<>*ZRheV$8*HnVXTXOct6R+zY&qNFLZ7!=
z?-HDUL1Cc)0u4_`q+_!x6()(4kd$!8qtXHgzlt=C^^S`2$V8p5IJNdIckE#D{rc1T
zlI>;itleP25sAA=#E1XN>-q{v%Xul_kwiNwLw0}YTWfh%Bl#8<8&uav>=10rAFI@d
zt8}WE7Em+3qc>bK%!PC<zf~0?Vi9<Aj78d<$M4o1bd{CS$xFD74Cd1h9hPpZjy8R3
zHM76&miVZ^(B<KL!`7;FA{bM;h(=W~S$UGEsl$zxAbcH*O`cyw|E-4kFWQ`OLtDpj
zZ%`wlwq6}T1vD)1dhU*bKZpW6cm9h|zf%#*HoB|(vRS^uCyMbwIrsB9TPc1wDf25U
zJAW_s%PT9-62xv2C<L5>+OI%ZNzNr}Tj74RVb8G8GhQzV;bzTI^ok(*s;Toy!te%)
zGDi?81P8Xjg7TR&?-h}*=oYb)Inry251a1RUOmsp#_<_QRld4;Jc>9kS}6?_k)}kq
zcFN=|Z2n$OJr)z|Z@R$~^2xDL-U(lLmF@w6vcJsQOdWlzqQorNwy=_j>;DS7R=_0s
zOa}qje<0||tqyHJ_%M!zr+P3oE}x6aB|BQU`F{W&kD$Dgbs6+h<|<8Bc5cM2^*(cd
zh5#1KqJj&)8lJETQJf=hT(njqx#lom&;8+o+|JDRX@0(_uPUT_(lbVW%(zBqEg=O5
zJc3YdXhglT7X<}9?$2{nOu99!<)ap$ZovGI4ms@rp^EXH)A#5=S3Yivm~Lj&Gm258
z2?n8=j4yFAXV{JNcYqQ$mZHL;RTmV(adeGK>0o>nJ6b7YN#kEVnioJWb#NgmSoyM%
zHL<5p!xalxBPncDFbpp@GC>Nma2oQtu;c(sSa1WtbW_>%q1w*e7^s(jOgJx5mRDVE
zM<9ydJ_R$%4gOmaQWKS6mVM&DqkNv%LZgsZQY}PCv!kTtfb=Q1>NP%DCiu#N$|To3
z{wWtMVIx}jaVz{&y2tB}ZuyDr7ZZ5x4?j_5$OR}{miuDWABOSwxhY=B1!{~UF<73L
z*Y#fbrg_zg{Ry8+Y<rAe@Jq*vrTogFU4PwigA!+c^uRUo*y-l_^!k+8ZLjIIV|?|(
zRpT^S&HQ-}C_17IhOEoq1#5oE<^#Ae2)<LGjdK5tCa)KM<#F^J)pK`oDfa2sBBp_A
zsmY(Z<*A^0A@i7A`bY$YW1yxptK)<E^BEsRUj1$Df#8P!Qd>WKC0_*VB!NH)?^e4L
ztkZSMf7-znh)@)sOexeqM068`jTu=|oNQ%$5cSLN&AJTYAH$wXbt>wrvdPx#?1+^T
zULH<s`bc7C%}Fa}@_iq#4?WT%BUHN#TyL~;>s;i0Obb4ahv_H@X@NJNJd(wsUk;I|
zwrS+P$p6SKfTZQM9~nRq=JT)p-4FVu<RmiTL2?nq^on;<{Nc|<i<mhQi1kDylhRlL
zI1Y=H0!t(hD|v`$tkxLFTAFE%?@OVRWSIC0UKHQ>PKr%XkP7olIAQu5h-v2j-yWRC
zHTxi-_+6>MI=5LqOCCT<^t>CF)P{ibB!t=7!?I&^CWu>Ma1Y2%oSo|5F00;LD+rZ0
z!0@K4q_f*yuCT0=&$bpsX7p$)4<kRHNDNA8K@8`}&N`gX`+o#P70Sm@%ik>4iNx@p
z+qoG0O?at<5D?H<Z7amfavT`_jEnPCvvJ-j4H*}vuyI)Y8$h2yV8dM)ST#AAEq{WC
zT#N;&z2wraJ;_gH`1lEDs{Y+Rc75d+W<0t0vk9BvVIea+1He%bS9H@%6Z)W5zTBWG
zYR)F=QS7Y4y<DSOpj(T!ktlRnMN-GWM9h18N9DJbZ9nMq`1?YM=?Ta}#yKl%|6AsL
z@|kLr>h7~&hXUL8p5MQjsUA*3Z*FoPJt+O|PQ?!Oy%lCb-Q@PmEL;D03QFc8{aQeh
zgs@M%s|4=bj6l5xe`pdu77`lXv9HB$Q61LGVEA}(jX2h$SLUQ_Us_u#&RE=2WSl3f
zKsbDUm@M?kHrzjRYu`r;DZ&G(5-6X?SFD`Nvp4WT&&sY9!-FZMQ4<J}+f+)56>V4Q
zeOI!D)y17J7B0{dS8QedevaZ~6~TGvIM5N<=vOk*d-${h(LX_0Y{W+N$4OGFsF*GI
zrlh9+7-m%+gmn<@4VwHp^1lU4I;By$azD=`M$)GU(@`Q@OR9D1m=<U>u5Y46**;!O
z@}xaaX6`!RIN{&{=uHlSG@IoPRdWy;^e@NPB;KhNu_I!c4yfD2yk0*kJ|7DjTI8Bj
zMWhS{otl5&SeFU(<NG63WRY2vX-C9AGw>8|-yj8qVRa7VM9c!!XQ3`T>5T-n*cojc
zO!OFG6O&m;nLVTh%QODHS?vDYXXu4pEDCSePXeL{qG6uXmpnYVEM&JFvzPynC&1L#
z%QPt#g>rkm#i9P4TDgPmaVa128jVX3ze=qvqDP&_E!Do93C}R~hck@7rGjO#K{L;k
z{BEWu1DeQ+Q8XuPH21rP>t*m2=owAv`j6Ycb)SZGV`lh>-W?Eu>tH@Grld48eY8e>
z*wNan!-=*3oLr<DYE~m8E}cj0y@AADEZ6$jCIA#sGHtfOWBi-hV$bfWye)wM{B&(+
zqG*sZrUWrdc$O^PshJjpp5hkre(s=Z>gqJlJpE~fnyU-AY61(x*I~)9pLrD~q!KdV
z+OlqWElv`t_Kl6Zt~pQD5et0tcw~yVcL`Xt;GO36=BA|>h7?m|A^$S4A=tGc%$NB>
z&uJ@<VsT#R$))EFai*6VlXnSg4~IY77}U*uTDJYCTY^=IOE2Gld}VH@%9b?%L)?X(
zP8VKf?sR2}4ogl$UZ+lO&~(x5`l0nNhHkZL`s$=BXsdLzn}smB47-^*K{<Z~d06X9
zmLYG1_jeu#AY#Ms^Hoe2_u6#!A`@!R8+P<9k>at(Xo`L_@`BBcNT$u7MaFT+)9c@}
zC>Nh=%b8r{P`c+(izwy_*8v$H2kyYNo+j|z0Qb2DELngjVW;wjCWIb4`&P25_!wK*
z938l^)iS>@Vt~+DM7~xGUvREz1+B}jOsRk%{;9BT#k^EFBZe0^KnVZ41$Nes7bMp-
z(Y-%4-<9&_v#<R=imM{V>tQOkjJ)gcIWN$N8g?Du+F(1Z5<AglOpEFc!kT=ML^qH@
z242|5wy+W=@3XTvm~@et8PB4uX$HH?x(e<7<2<L!cp!Lq>~IQ4u+hH<{!X@)gS;}%
z6|ZrM`pO>}L9LBk9X|=c`D)ZUPh-CerdBB=e{r?lLWO<HhD9P_^`4ZYt%GSzNQQ|g
zKeOIkw0361zI$3Um+#-i)Ntl*ruR*m9J7et*~GcSNSIG@?US-p8!JYKS%1v`3J6-y
zvyLG-<Ibn+uy@+pO`aFbvBGC)-wlsdZi!ZYTu8I6>v;uM%C~;w=}?71#vdja?86dv
z7{hjzXejyBw{w#)a++rHbSLI_-J!{>;C|hv68;yk1#(QP)f@J_I3g_~N3&&kiqHS(
zg^nB9o;FY#A_dV~=GvXz?0-mRA)@kg1P~v_Ur0-FNX3?CAkKXK@w%G@;uuoN<ihZ@
zxzwTj;|2p$mRb*ZkD$t1g|%+!WZ8O2`W|-?Z7|Y(EsN?-O^6njbug8Vx{rqx@?bd9
zu0^#>IpN4)bZxTbGI4@JOmVQt93B2X6|U)w)S6ku#P$J4cVC%bZ`{%Iw<+awnp|21
z6N)~~S6Kn%r}He&!2t%3ppVV|&w17xVL9Wu$WP)7CT6?gg2|PN=gg8$%_E(R{09Xu
zzN5)#b9@TGVN7M7&J>@0AwYt9triu;?g|oKA_}>*>(SN;#G>Z+yrQ}_---+q(P_AF
z4hv-U#s`~O{So2VR+<7OE$Pi9ci%X?@Qu`RI*p>bB-?u7p=gR5%+(>^*vW}rWpJjk
zUTAZD#-Ck|X+e-j=4~BUb&0YCb9T<hAD?6#O3sQxh(X%+Qv*IJ2^Z2Q%G~F>Q{y`y
zjXKOzN_hc*Ql+)F+insJ<dEf(uNQDw!9r8zt-^OSSp0QLyBE4{U^QLM`_UR5qF290
zkAjjyx>G*Z4pv9q2TT|#8!|lk3yNEfS{L8dJ0uW=wK@Lw@%hR;*t#9^u{n5HO1<o#
z+QdQl=aw?#ePM~8=uZ9s@-37tyMjkW^$k+w5~*ob*^?_x|HI&9V#D_*kSj7SBfyG=
zD9QaZ+L8n_(ejy6(8r-)=9r;nE!YjF$EcyOu3BWii$`nxI}wSNfF+rvYC*~tHcxd^
z+rPfbdlsBwRFF>1**M6!%X}=nwCZ(~ouB*|!Of)n!`h^37{8$<%6n}D3;ZCs_%n9?
z*ac84<3x5kjeR*Hc`KbRG5Z4COvq}$X1uv*7a%>EXO51)2uhG-u$4%7EUny})icSg
z7CNzt4HkyT;{j{u9I5+A5;SJ^RUCwqe-mA>C~iD8-6_@1$<eIC(^5ETLuvpI!1qr<
z4l}rLwcTXAPvIH&xhgfr>f@12A<%Ac_DjqQLLiMM!1Q6@dMo-sU~%T;>(u$feg9O@
zaytTI-FJ9Nf*yVoKe(+))ui_rfhgd?`@6$Vq}%+!?R;enfhZ96r@{|V8L8j5IK$eG
zv*)da4HXj$-iwhqdRPuz+~X$G{K&6-6(ykN2GH}-mHlb+G`3kBU^q5?B{1XQej=Mx
znt#4z30Kuhv-P5dtfpO1GLQ>+`u13YVf2Je8s`2p<xwJche8kouan71+wR>WnNBb4
z(bA9OH;YR)H9j?gd2=1<RX*dq_^}025{K+kJn2_oIYD+q5hZb?%LtAwmT8LT8>$p|
z1SS}O_q=UTOm5GJeZG+Qq!t<pFiF*|t#=VfCudv3!ha?tqxv`gztAxx0n_sc4e8kz
zy7np{qa?U`ls^nmI$AH>qnq*uFMQS>nyUHo5PSd}*T3R1g{Ovk`z(iDc|tQ7Un#V%
zKC%S;*L5-5@p)T|Wr-jvxLe`Nt5xKzF`9O#xgRjNIu;XOT+eLqBs252-!)}`FLXj7
zX;ziWG9GMru!tmjEQK}BxNHYpGKy{#C-#rKs1ZocFnopCMz36dayp8yJ0vJTrGrfq
zV%X)tOz)bXT1AChyb2Oa$gvfYD3Wtkt6k!HE_9d|>@I;Wv`o?-PdSn(>E7fmU9ods
znq|#?HQx!KA!JqdJL?-j<klD82Q=i^eHp@2sk{L2(IW%q(-V0VFc~=VGAAV|<qrO!
z;_x{Y3zsf2-qF@?WG<HD)7C_qmGV+37|#e{|F`*K8-w2K*=ytF124$b%1%=-QW?)v
zv<hM@s1`hbziI{-opk*??h7NQGc3VODYW@K30KvU;ABxcw(OQRUTV{;a>f}zYeqkF
z#IF3FAj5Q9n7^TkZjH$;k;R)!cV|q2zF;Fd61OZp#=@H%1$RzSsj9O}E9080t>V~A
zoGotrfK5Y2sby<=r?S{-=evetzCm(kn1aP7!(DV}^7U7ha_923lirLEV3zoJ-lPv&
zuIebWQDHE=r@)fTgXVs*ZEc*gJirVsCT<gb@<)Dp{S^Zl2;FfSH%;?g{$h21iqmgD
z{Z|=e(mo_;%c<jvU2_^a`$3wo!T*vA>FD!~|3MlW6#h1Pgir|tSd#SExiMgQo2-A+
zO3P%5w}Hs*ZZzZP?qHfC`f`B^@(uMCP&1h4u@KX(rzXI^lyYk6Q#2}%Zwi2+{|>UU
z4QI6De#`6oz5PCGiS+PUUhXt%MJpv5M@9uO`!*>%=DMU!X5j@_vv6#--71OE<XN!e
zP^7+ysH;gRXh6+5*OBSf<OkmgaA~AEiQ+FZ7G_Tg{2ItD!7wDDRZ8Hx3=ze^8_koW
zE7lm>n@i{S4w38gmU9s`I+FFRNU|W!3SS*B)D58@Ud*J?Pafipr6^k`Axp3(cA)6d
zFwB3VJxlR9fcR`H<5j&ErFHs`Y&Z|Gb0FsR2?Fq9&VL@pJdsriSn*J>Xfx+BBJvc`
z&|utg0WdUITPL)y*!YMibz(1n9h?taP#TudC|Pg|NV3!Mc`cjNW^wZj9bgNGfP$+C
zOISQYC#R?(5&FU)2?U?fhpGSIrc*@0x%6=xjpO33JizvFZ0U#hZ=D^*aSu1z=Jtqr
z535xwGvNbQy(!X<t46nyp2L}->WcG#b4-%-T?q#3Aud~yJShT~V)&V`aI~e5d)*N6
z*9PqQc(D7zqhkV3U>u(oSVh&$>fhxzROTt-Tsg*{?Xlec7ZHU7<|)@KI3rxCJ$DZZ
zLUH2x#^3nRRL!l36yMnjl*V*=$~X|k@||a9dYIi#G{By^Ki*%RZ-_l?+^ym~tg>Kq
z;P{?=_C8C<*~B!u?)v{h4;Z=R6)q90u4a=vtvI2x$$~e#{$XlNGt(S&L1kWDU4F$|
zXKgKTYy-3Bv{XunCi{{R_$(cW`8dLIzjIBQa`uo<9)quJAd)cH4zWBOntB+%*18)v
z^(Y8Q%GXOLcU~<C$xQyZIke=xxhHNo4Gcp5PY4wZD6X=za(S>35b>0iBQJ1ssIqJM
z?$tEN#wU8E3|U1;F%@s1qD%78%bS_&!tugJ>&0*d!?{NX^vR0TJ(fN5lqY`_=c@)-
zgD+ftiNo#NRqH%%i^6zGHJHww4Qs#m;5w)lh8Ld2as5q*yEh3lD_v}du6BHv8vwWR
z6PyZ70TAmME1|dHA6UZv$CR>t+9aRFGysgoljLL6UOE<wn5>2*zWSUNxZ7o2`lR*f
zj}N_gkKJUR82UUZ8nk;LmV-O3WhcUW_EeAoj@)U8`jH|EcmQakG+bcx^6C)u?IGU&
zeQcioxA9UtYhY$xJkz{r6OEwb5H$S2&sZ(xM&QuGB>>%~)Ol0&-Z@M|t(=fGI=*5J
zgyLSk@VP`4wb1XLA@r<tP$t46wEaalpH5W3)4!_IookN%^l@3$$^&#Qz>w@{>#$33
z5ak1sS9D6|C_V^+i`w_9i+5)`LfEmT+p@B*v6joc&HRgOUWU6<{lN6nYYPQ4Kd6g-
z%e_BDVYm5f_XDPIqvmRlA_7PFn}1Nx_sBxA%Yf^O_bY=at6~pz3<4xF=1l)ZNQ~kZ
zg_W93<OG2b6N<oueUGduE)?alO;UG`hJ`@uWb-fbT?1>V=*tA>V3BH}U#!aw2m$rE
zcv2h-KCufQQf3&9*BOH3FgOQvBl2XeswFx#tGHob;Ooc+fV_}2AKyC}(%sZV<2=<9
z#yu+ayGxbdsKHC3g@uKa%Ofk@3`3%$1Bh!b|5Eb(dZU)<<XYKrwZb>P6uBk&485Z#
z_PSfa3v2_Q(uzqTI$m)7fEK=E*QwZ2g@hxh^2<yQ8da+S*j6qb!AB6H5&g0ACpKmq
z_vp!n6p3;vuQS~(`L>~#Ntdfp)aOrHCdK=X-|5KEK@Cg=UkPM<+nk^pWn1i+Y@O`5
z|2Ap_Z0&N1P4)wgwAp#aMhJCEy<vhiOXoVrjR3W{r)xqVJ{1K)(GpJ$%p=g(Xq=|p
zkU<Zj*;J9uoAS7l2oEd)c7^|b306ZU(~-+(oI<6)cGEO4Sm6MId~No|&{15*bF;G$
zY0Guzb7RHi@ME0yM4PdK-&Vl5Kd2tQk+|}`QD%Np2RbkOO>fOgWL(;-i`c$ir}7u*
z+fDb#p0VZqwi5O%52!>J#+8N0^(`sDL_cl$%e-R&m3e0F53+yO+2`7&jpD-7XF@<Y
zFJC`q`IS=j6Zf)*=Qd$`4JNtBzvyDcPHM|iSu}|1EJ)4$)W7LIWr7EBp?^AV|7UK`
zYZEJ}f#PeGDp8EXJ-O`jhl2kbIN#0ni}v9>p5UCbJ@mG(ex|yk7Q3Gk8?CdQy&iwO
z5j*#4zw&CmT{?c4YU^!q`92xx4t8q3-}@DWK<d|s+pDs*esp5KTj54^lS(VFWhAn$
zpcU=$P5kHF?)-UT_WJq|2<^mZlwW~Oj-9K4YiJ<)PpFOyy(kU^)A@m?QDa<XJy_DG
z0n#rRDuS9`NQIXa6_?7-BxU^l0<or_7Vg^W1x2wXtnC|A?z(EO_6WdcpHlgHW`V;C
zEZgMV@V?ETSMh{9rgG}DlPZRv_$J0dm{2q<EA7rY5EyQ3iNMzs0af}G8Ss|%KTg|2
zjs|^+RqulMZVs)<%gUW^Ji#hP^E==mx=T!(PD4(#%km;t7$Jf&e9Yf5vBkK8CVz8D
zi;n4M5Bn$Us2q;e_gPn**isoI;OBgW(#nqVC7JbuXS^1pE=B^VI))_?UyNDoF1-Iv
zRaFsNZgK~Q#=N@$>UL=2AyZ&4z}cdSMNMj@nKzTpV4_CSkxM3logad#dihgRn^4hC
zC|J_$z%eN*cEY%<qLHt@a>bCi{}VoJDrV2e!v<%HdzZ8+CDh`pMYe3M*({yuX?zv5
zdRCrE<w=iYmdptj!!ZkVT?fdaDp_t%7JP8~vMop%EYI%xrpTd5to22^neF!{oa2>O
zyNVKv?BX;Vj}MayPkw&LU3*^{9(P;TLYr*MtFp3}JM?Xy@b&4dEAp#HER<!pkIC^B
z=T}q)g+k~}Ub%llj%V#>{uMp#SHzV0T-ZQ6VyfIdsr3U&E6WLi$k)rRrn3FQx(}<X
zAMRv_jNsYScY@aVij@cq(?+xVJ!kPZmZ(h*La~aon3R&)wQ>t>`B9w-BMT!~JGS!U
z^lxdwVLeaoj@6n<9T8rGMLMCyv0m$vTSk~JiL#(+wauD5OccqZVD7Tx!hx-2@oU2$
z>_pxiNBD4BM#TK>AY6ZpsBkc`9q?R`oHYj$meb`LuFPzfIIgxI(QVG`>t@7mph8fg
zh<v6ThXJ>$F<-SJ^U-cZ#LhSix9ZU<2Z=UVS(B71+EUOcfb>nKbTSi~A2ELodlWlC
zfcx4@h4YJd>2QiDB-iK5+%p`TXGbq(w85$ub$D7^p0OH=ujyaK`1h}25OI%HX%Bqd
zY@Olf8JS^f?trlpEy$3gw`aDy@-bY*Yi<bD9^#H5a*XHr6HMu8PtEBXS^s8rbxQa;
zs}1bstqHKwVfePC`bL#=QAnLCA~O!ypC+g;n=7(q-JbIE>6_OaS1(<qv68)Ce&t~8
zC6!pvBi99yy{>Z0>G-L6RND{y>XwxUj%6Zi75%or2;}VHHlWzKx_%i4kt^AH0k+p5
zTK2U`yjr!3jRs>qF&*dt!{IS2evYy27pr(Vt)6eYC#_XzmVJskrDyYbFx<~4nStxe
zq-I=xbbB^uarO2n%hGY^?pf}*_c}aubU7%17}209xDmD$(W^MQ`GFxLbfrb?50I~i
zLc705(g7|&v^D%v#rvwn-W{jALGCHqNidstz`q&@V6+jZ?jwLQa+C6SW^|T))w-kd
z$Fu#9LVM3YOG#BjL%+z?Vt4zk+ha^9rRT1#?{TQ;pWNA=3DS^&<J+(M-<Nk`t<8L!
zZaFJmk?Id!C8)Gw>w9&gU_0l&X%)>l#!nJH7ERu_sT5sZq4o(Yq>UZY+Q_rY+HYH+
z+jrEFFNVI`jJvZP>8^&9CRJnnZTUi{VgcOL!{0IK>b>5?E?J#}QCLC^>?~x+863eb
z`N8D<QizvNbF!H@jd-m$S-G`>5^bv?+?=Gs&c)q8AJ%AVi1x0Ve*<Brb@}}16&2I@
zqI3=z5ZMT11v32*q10pL(yv9yA6G5ebeR#v51AS$_#QBRIFlM!=nSv-FNqxZ0`L5?
z6XLzzE5hqVIjjX#LRPV`hLe%^lm=?xl&|drIYC_d-$3J>OQkltN1&KgJnIQGAYfj!
z@gg+1TJEK#uut)@r%X?_KoNkh7wNJ~Et}bby+*&P1-ORA!sR%t()#84?B>@I(-=4v
zKjR>scFoi26IV!ivYEfSA^E`1TNlqtVQ3QnZ5}s^4^d%D+3a%9FEDa-5cQ0{&HcY=
ziqFrJS)f{n+IE3#Y+~}*24m)5nDEk>&Gs?sE7fbcBkDum43~dm=#lG_AEghg=|)Tr
zBi-f_tE`Q%8`XmW6(~^wLZIEvkyuu@baNLl*DfAl5BmxGKyf7OSS&fFdgJH2PF82#
zwvR~f9h9aN8zoVeAY$e<YM!Q7;JOTbs+-J-)^^IP&TRqa4+5nE3GJ5?c#iQ^nsx1K
z<343GZh-plcR#KI(iNqUWa3tPnSi^#I_}k`%Q}DWi{Rup_Ey~IK#QC}EjLeoT}%Jk
zF*Ck_TbcJ9%+2v$N%3tOGsgRX^2ST@sey_*y3$%4%yJ)NG;k0M?t`ntp~D9_!T`+B
zG*323h7#ptQXend>#n56PKYk4Jz-9&PYg5dRA<`3Div7ctW**57mzfm*>QBNQF&Pi
zvF?JeFvzLL%)E-{-s+OhnSp#UC0_)5j}M^d%w$?kU02;k1+DPW$$Z)KgBK!qswAq4
zS|)8YCv1n|<I8XP@2L!#YJNHFDw;XMBk4^te_WY@p0mV8!x?$t;dk4_Gq!hjxNa#^
z4>}tI6R~r@J>7F<;U11&XdNxeeEg@M`a%<^;WaD~<?Og$wgQObr%-slZ~-<P3M{0J
z7`<F+QX-iP2(-R&C*jw400j$*h9W)qMi4i^JVS};0a~O4o^A3HEP<suHFX&mG$}S-
z4*Ap`z3idEb}2D!@4H!{d=8$%NMd>l+f87|!up)L6)~DMlL~TK0ah#L`||pM-BTqk
zos#*x?Wv5Q@US=KbEs~k!~+oYe$`YKT`)n^7fC5}iW!3#=h`{9B8q?S{wW^#vJB~&
zft`AJj^&EQfcuvG?x^=w+rCv2Xyw7E>*ra>6H^>-z!<q1rS#j+r7O4oxqP1C0|qTB
zOGhV7`_I7xY2;;$5IRIxAk4Up)nw;E5tF~pANJyJ9{R3e!7Sht5s3TipdrV`cXVaT
zRupID4azE?v(I%m)pn@wZs_%v(Srr0@3o2V!;W9}%HN8W^<Q<i++@+S-jCpPdhPp1
z#xK!!w(^xP(6^?K+K<BSt|NtnYi71U!&*J<?BQ844%Q6I&aT@h-7{&(e;9BDG^ASR
z(gTZ_!Ap=H-+F$O!SF1*b;_iSB`;*zshocOyW$JBi16ZO1Sm?t>a{`NEW3GXOJpbA
z`zlSBQ>t<;Y3X&#GZk|}BIuErde;okwgRw%N&}q|YBqloo0c5`x~RvNKmI6%)&jNU
zdXHkv{w@AdI@?JXJeF<gx6`9juuROmCb!(EJb4PNXQcvJd4*2ama<+@RVJY1I-s=O
z<`>a;DPf}|tzrI=GRoaIKb8tYVZAw^4tzWR#gp*03oAz_L<Q1Oe=Dj4xqM*8ExoW6
zTY0b{!(j=>97$m{mn-5z_pokJv4u42gp@j&(vL4;!v207aL(`vF${{DAlrI@BBpE=
z?ai&VHDZ??s%-;2Rmv+H#_1GzJ?NxB{8gjp+-R#-;3*Z}u>~e(RWq(>K^!mze5AF^
zfI{@E54Hr~g#MuCC+1Zk__Tg|d`=ieg5B2XH9&`TY*pO<DyMv|nt+WZLAQ%uANH2K
zK}FWL1Iu4$Ng{dPj|A;`3&kv)UXnTAb5%MCJSmlM9ge`3#cpzQ{^CtFGBMwCijpn<
zX@b8+v4(ToZNW4-?ep8~(mCMQ_w9X^>{Qfr>8Qvs;YSKAH4BqVV30H2gkkyMnp}83
znW#a@a?8nDTYQfI!;W;6bMlCd_N0BGrYGLB?f1}UY!XQpPR-Xj%)^apT^*62OynN^
zR{UnDwIrTq?Ky55a!iTFpta_QEbau7#`P!UmZUMc)$gkWrlG2ABm>nQnFp6Xyv<;7
z|K?*{Sh*%lLIP+uuwc1~_+E3~^IfVC@LQHvp@0IKCm+FCGGs6;0omNbMi%i+DLSoH
zoQ`4fG9JR8N~ga8G<M*&2NKqfNT79y9t%43rLMJyMP3{QWjvSg<JHGoY64wKW;&DV
zH*N&#57P5sa@IE=Vy$9*`h24x3Q#S(7@k+vkd2|`)HdQDBm0XDI(@vFU66bPU&$Qv
zVfgA1Av9?|57zl}ZdVydK9H2m82LlO##k2qg4E11ZEn^soRF?UACsEcq5<zmA0WY)
zv2A<Ys#~~n_ig5L?SBet@3vmNb2{9zqJj0RS+Za|7kV(2DZvzR?8eO!G0+-NnbMN>
z9RTL~k0gNm9z|75QKDy_)&tZs%GzlxdQdG^)a)?yask0on!b^XJKL5V;7?hH1M|l~
z2D8^Q4%2#`w0`^WXTL8#BiNt8n2y`+iDa~IaEx@!R#95dH9zNBoLp*r3h^D`+86n|
z;cxQtyeMA_3)aaY<vBc<hLl`%awi=dKy^B~)1YdR_B>Y8LD3`#6nD#x)pE%ei`FX0
ztrX#4_}z2J8<KaDUBhK}^)K&0LI2r0bmNa=Hm1JMjL;jgvjN=t=b~zvXS?BKIn7S~
zmGaO;X9`<>cMwb=+@O>Q8A(=Jc7|W-JFNg4zdn&aj@TVLj_5UY`*mbH8XEi7`Z~?y
zX3ib{!;RSE&exj<9G_juySdnVl6&7rl|R1re`rK~ia!?tN5Cs5ll<|kC9dI&*lEDn
z6*Rw({v6a>>To@hVrS!##V#j6@;5V`6%3Wz9KQEvjivGjw7ZXE#jX(0OyyGLEAo9E
zZlRU0522zvxByBA3NU@I5jk;ptQqG=Yg%$e*$322yCg>3rS1T7wL$AfwTQw&@7YOa
zAGU2VpFM08$&CkEx<z(6WsP`bW`QK|kG(m@_L!Q)@Y@KmfEwiPOCtqAyU+K1ekJ&~
zRsw)Rczo2k0u0Pt4jsdwl74tq^PlE_dmyVtneBC%5Jw`q?FLQ3V%~-WB<00(4GavB
zBDI_ZlYcOrqh!}-Z?esOn}~+E9*6~&U~_xMm#8p>tpHk!7@o8DiwB@+4kL!hR$E=L
zX*n+nlGMy(fu8~T;I7HiQ3zh+GdVokWOn5$m!zcT?!7c&0ZHm7uFuzV>*fh}Ljwz|
z-`9cr5M)rRe!LJy)!<g1pV2M-h2(?@1Vvxj1q!VDIwI#%BT^e~-&pAH+&QJ0y?EZQ
zYLp*7%$d8huy?!pBA9*9aPkmy{ZE|N95fa5iA&)FPe(cb+j?OWGqG~l&$et*V#jp?
zqZz^xZ5;t+>1C+I6k#8e*hNc9*i_U_p0ctwP*`VXGo5r@`b#gHe;4Y*32qcs!uxdu
zxVFMmE{&E=KqJPMH{e$fsI_Atv2B#0^5hO4dn6Tgktfw?NeJso&C@n#>c*_G4q%Hy
z6;R6E+?{KdC$a32i-214hbZ$Dt8w$-y?kpaMnB34s^j;%1tW=0g7uYgaHb{49z9D%
zF9JjX4IjUpMjkgv+7uUHhf%uL7=z850pDuhLs&(dA4z^HNNq$mNynon)nzVV1VIRu
zR@JIiI6`R(TNM3Z8}w1plI90<7vtK{GMax<Gc~1f%d!|U?KqL**cfr6Wp3A?Zm9g&
z7e4f+#_65l(WeXdRbl0rQtxh-F#qKzfT0?T7wr#jvmz^Y?RrFjld@S|@^dD$%}5D;
zdqz!XRJHiYPhfo-m^%pMs{BfW88jY++@qY2zo)Z;x@7tOxPv{l#Qe6`adz*G;C3nZ
z=-h9Q1MJok7i++HJOP?Cd@hgIJL2Bh;E1i5N&T$$mV_?R33X>1llzRugfT3byEhV#
zTbE`Sa^%v7grHL>|MXr@(r?I*Ten1L058<v)<BED{k>_u)pt0XRqQixw<7!<sm;-v
znWspfzh5F(#$5JDhecY>2;EIa-U4wG*wqw2oaz4kC^4p~!>9|M*P10_v>bemZqrYt
z3m=ITK{W`nA5=ysK~gPH%L}*9Z`_%pXd&Eh2T>W|XQ8vV+nr(TV>-}`e|?U;2nd=j
z<crZM{Uuzhz6H09s$ltd9<ZZbE4Yn2)$nY?ZAa`T&5Mj3k4-^b@Xf&ag6!MrNl+_{
zyuaBeKJ`og=jKp(|0H>MYxcW`Tg%eyeX|LP=>l*2{X+Zb?aJdr_g64_dSt)%&589n
zF#CFw_P+<wef#58T22mwlEzq$*!cLv!MM}q!Tjt+w!JVqWR)-b*ZSXwIJX>YPW@`Y
zra!nah=U<d+AO^umYP%_CkIhQJ~n}%C|Px$#ct%ifBCrOwc~LBk7Sf1Z3L(Q<0?BY
z^mywxl^qna9k<6ut6z>!|LCO%Eur7u`^Q({7Lm%7UarBXDKRVNL3ub--*;2UY5PA+
zy>~d(|NqC|5SfKzQ{tG(mYuTqmh3&VM^;9%_ulkIju07<z4snvlsyh2Ga=;n@csO*
z-}U>q>*^@y^?E)Z<9@&0v$01$9aDHdxKei#=UUmV;{VMFGkzH@v7^Td>(s}Pd?Y>J
z8N1vObh}5B!V5d)0kmwb)3Gz#k!hc4a{}e*$6`2iDJ^7Rf4_)3(bbnOqQ-_Xpc8Ui
zYl`H$_fD<ItE!>SSV7Tug*g+ObzFiHxlj~2myEQ@T5rR-sXQbg+bdE%X;HP3V9Da8
zu@1wajQsszDw<b+?=e|Y4-b>DtV8#*V%$$Sz}>rv(_?jz^+C9o!(-s@y&t}*z!pMS
z0n=+mDgGkOqFf9QQb%o4Ol|w!s2x-*+`SED$qPe+-;puGv=Pc`D_!R1kL$u9NBe5`
zFP7%?H{*k;JJRp5t&82JAMo-8H&a!9JG^Fc60W{9ox~8XYbbD|rQXNM*RdCdn-o}3
z+2Od&#od{0>HQ>HF+3(6W3}tX@?A*mNQ_;BSIg}dNNEO0GV@B|Q(hF~2nJruR#qhu
zAon%9cz2`mmuGNYWyj^eKlv~MZ6J+U*jsP@zoO`TFYzTpF&Zy@r*nO>E$t|0Lp-`M
zNkE~O8(Si|mx?u`_^u?OFNX=@HVP#kV$aO$f*JB4AvlovCh{O<gPOEZGN4V8T7X;9
z;Y6_wWRMm3BnE1#=DB(=V<j!iQQ~c1F@)Wz$6aqGV6YjWQuxy!rb9ae2=0b09vn%+
zhXLCxen{yaIuL3p?-;QXSvYat9@R`Yi>CI{Fteh)_77hMB4r@jUWYB6+7Ugo<0aBG
zpU-!2r)ns6l6uF0D4f_!W9_+5dvz07a8iGia#0ziBKri?tLprV0_~Y+9aOCGZ;rDO
zn;Akm{nduclm|OFQy0u|N&&M>CQ)X#Y>LnRHL+OTx$Co~NqKnSTDu$a(L>*1`_2`2
z0hEKgBBHJW&peNCmV@-L5?S1ytQ851N<SOyLrE|&5u`*1hUb@nvDslec6tYU^?BEr
zfcg*@7&(ep<1E5KF#C`5KdxI_v4LQl2RW<fws+(@e?u>AxrAUg)Mg!H6v5$XykA`h
zNxN28*68>LDm#A1=dt|uNPUIbDJRZVs8Q*5HD{Z7`H1u(@XEh0_YVhT6&;xEK6y6w
zm1UapkRdX)=IZov@v@Sg*R1oeYmc`jP|%*ff3hqjN5r78wvzfMjUxlkH}}%d<YlYR
zFmL+9nXlB&VO~FD(opa`^pw#!nWDhc!2sG%XD<)QM}<7U&0w<c`BOPmLD=*s*nh!S
zAvo{<?)Dp4G&J-^z_?C?_foV`_u~INAO6e#v;5z=^e>!PvhqV)o$T!XbGYZVN_Y)4
zjn4nzNQ!%R-uG^`$N#*?=y+>qdz({CY$Ubie63^Z#);MC_PhPl5&s;fcdrll#12nZ
zVqI6d@pqcf95CNrzeF#bQ{D5~=3+Of?`&^>DUtk^miBgF$K^eOP&Rh<=AFosOcnbB
zsPGnicB$7>S%HPW&L3Q$ul*HLV#uS*zPjL@hjmpBNl&fq1k=SXI2_LEz6&hcn<U21
zsh!Ie#->dqKP8l*A}Dok?Tsy3bE{-q`97a^8WDV-A(UENPXpdFdNK5@=D-)tnrRA5
z;AO2(`V6Mp!zp4eXPPgJhM|I~a(+7h%ujFdGI{8lnQ|*iSr(wn^@&A3Fv~%iXNl-s
zMV&(Cv-(4Fx1I+M<cXQK*GmMm{~(NvZE$#QNz{&aKY$f!R`x^K%6(tCCZ=4$ONbML
z%+i?pBLlhOMd>2WnC9ZKZfC}Wk-|66nXei*XsK`KCuQgMDEyEB)-hi}??S^1bN{=0
z)6t~)w9B3yOOg}0#+w)uF-8f`W}fSmp#{d+sBJktFrPFz%TAHR%k$id?23@3xV`*I
zL=}mRr$i{nl+muJHv2hKl+2NmD%n!J1bjdQP8kpG<wXcw8$__7<r#lYk^gFJWPCOV
zl0^tcR~$2~pnIId`xb10|3-UM%>h146v1vw@2J|hX>{x^cK)03sIj{%`noGaX4N4b
z(GLBH1IR4~*URpgPdcBzB|Cmmq$8=;Zpznlp+Exdl{zxRE4#<>%L5Vui}a|N<LLF4
z6deEkG95~($d;o$Zk_IjofH0~TmEY|_EZ`|FGeqSu>lQt#pmS~oOgOriu>QbKv>;G
zjuWgi8-#PWx|gds6cBjHmd^^UZ86l^pO55=5++@{W^hwRqe6*<LYBL&>>)XhR?mC8
zmz{F*{uxfW@<b#uRyD@zY5g<_cnEAp6Y;wYvI&Z9n#>tOYAG(92o$TBqAH@Cx2+Sm
z@Yp8fO>S>Hw%DF^w8*S#>(V2GrRD;<WA>!!``A+-3;KB@F~@lSmF(Q!9(=wak!&$M
zgl0jzf%9+Yv%z;Wu2awo@<=_{y^@?*e9-?vT=wH_ZclLZle3B1nuhhrtgQA#VTY$y
z)P(P`Q}W`>DU!3=iYTuCoI<H(%B93xn9gpIv*KEqQ2NYV%w`8go!^#9u~s%?V~`Wt
z{%$b6F#!W2@4VrQf#gqmw+R}<^h&wBBnozM4&aQ;vL$uyxGJZrtesxf$d2H8y=v--
zZXz^HQly_pLLReZRcm-ihvmr1-+!Qa3mcEhDsu5gwZgEw>(>L7Quk|wOoDsFn%FUq
zU3d=}k60?6X<NWZKv!Azc7o<cR>MkwijgR-BGLTo&(aIgCl_R|bc%GDi`#9#Nhcux
zwZzVu)|l_-?}`k!#!}Q0_D~W4Cbvlx)a%Q)xV#j2WyaY$9&3_5d_3~eJs!#Y1M`Qd
zw)Kgt5NL=6g^eJ`wPWji^%r{zs+`A3r)cf;hof$uFm||x^M7?76>b}U{w+t7US8Ba
z8s<nT$0IQyoErpLY*{7M@&@;uT<?PCN$=Kotma43t@lq_u{?W(=i9{o^<18i7+rQ5
zol8AADpJ`Wi}m~M-ZJ^_{k7Le5qV(~g(`>qa4fAFU6lAAWr$r~<-hs;g`96bBmX9{
z{nKdnlrkq<RPVpm6-b5rfV1ir9mw$UlEdzf0FR#W24F^owtxGARW<qo_Hs6<S-CD#
z;^BXI-<WR5)V}#|;7()QQH%;hVLmfV-?OrYD)gOeYeoMvEtwAKwzL`)7_5EyWJG#u
zU6Q9~=q)zStwge#vn(T4P#b}Z{=3;b6~pfxxWqe#%$4)$i_$D_<$^4F*`7=G^sAbu
z6hpZ45(vPev*Z}9cyYwBK8xKZIPyBQh(r!RAhBh`7AzUS3Z~~9+f24U>Jm&`b=on!
zJRKX%I6P((EhY15Jp@*H70H%FM4UL{S8DGa+{cSQlqUuGQ`a|z4a&Bor+3eg=F~+(
zBIWv~C~FV`8L<DtdoYm3LP1?hMlj=flfQf1!ZL#Np%&E-RhwAJgzh%x*3A0XxHZgC
zN=j{6Iqe-t26=T}DY_xrq5n*`1`$dC#v31wYo=1KaX&z@Qa4P*IrkvCqvh_xUB%Vk
zpXqP6oz97H-;9!g!cC`ioj|GsMw)NcXlGk1qjLLr4@_6jqVv9bC&pd=`HETPMPpvk
zcb?7v2m%W{GMMRjoaq0Jh&A>L$i66hu`!Vpe-@kMWzqy2NvglnKH?*|6e`uGi5pao
z547~wnmmS_z|8&jp%z+U^5GuU|L-Jnyxf%2mAD+8yH|Ka!0<D#Wkh>XgDN){UosLr
zfPjXeaw_~)DMnP!BbR`RH*KfRV<kGvkGsen5eN$($4kf4O5rrWm5!kasSABn;#=CT
zZs*7XQ347d!eTbHflS^*At^nk4C#R`n-Dg+50$M--wrTqiokn{C9(AS&N6wtO43p=
zS3XNVr&ZW;xhTP2UYv8>QYcGw<FG^tNd#(RwufbdyORP#jn2j3QF8X1>HMvj?(lRG
zZf-34cV%Q_R9KQWA*ByrWQ;aE(td<Qo(ptEnmPQ61(r|`QZb!{JS%;GN29!#g0e{t
z^3eZFO29zsm^NE;=^+gMNw*0kWt_zKW*HO-<_$b<TF*yXeA%6##2QSNL}VkwH!5Jw
zoujh_N^+1FINR2o@Zq+7+WE|Ph{+EP=}xqOdNq_VijG9iHdfb&veF-^e@3u75Wo+Q
z&6THi6pA{3aqSd8+4WEgA(w#`4*kqT2)bwb60PEz;ox6aV6DMa#r79zI<K&<o2Q>N
z=ku`&W^<CZ?s?C*j+6Y`C=mVY(0M$^-veSyFpk4xIkTf&YgN%M=CjLUC$V-~%`SG~
zlUlUy{%<K?M(5)+G-#C*$7<%c^D+39i`IDCu4g^m6ZIH!akpZoTa;(CVa$8%_@!I(
z5K}IL%ZOVInNC)N1){Y6l;O)Wo+*}eAtoytm?MLtP>OJ-m1Ec0B<sM3W1Bkg@cKEj
zZA_DhI|4PDcsY;TkY&Ak>JwYmT;{t7vPn1UP+T&0r|szq?bjFRq0CvI-6qG|7#B<w
z*ZoV6v@d@eNoi^MjX&wcymtNCH95&=nio_47kvJI3PnW=;o(r&zVrHtkG+xPn{+xH
zyJqwK&R^_>mel70E8F^S=8<5^4TGnyuxuN?Oz7@=+nC?)V%q(Ye6911i&HBYso>>V
zAddqeTnk3WJZ2)pIMo_HYQ6P#=~!M8<GG`p9@D;6Sa#BXqlTOYQ7g($o5wU9mgmP-
zU#ykHmH7|1-I%rsf`8N`ObBwVT!V;``*~VmPab;Rm6d@nNvJgxzHd9n1)4|r4Ffcr
z+Mtuu{@6Exccrw)7L+`b4W>v=f}q4HO^ZoLIsHzYfcL!Q@jx~8O_k1S8a%1Jn)?i2
zX%aoM9aKL91?!Zi>=%AGF;_yZM@1*tlfQQas2JK*@H+)h3m3XtqI$<d=)22SpmeTK
ztx01^!*F)lm_N8d_?unf#_V5i5+e;{Nb|{AS}~sNiF06CFAPdeZDVhO*bagz8snL)
zZZqcIkCgqAsv+_ixMfM@FKU_}%r)`dlD<<o_WI$`>=L7&ufOC%Gzc5K4YPi?cq<+p
zrBHQK<4KA4np4`b*N-xU81|m+>Mgd2R18<tci{dIXWbhSJ0Jq)4z~sP!n{N#ooYV|
zo0pjbAzm`${YMXjakhM(7%x$b*0th3dKugE)|<GY+XG@-M=h{!E1c`_yIg*(CmI`_
zGoU<@zb*xat&H1W<B@MmSu|<h)>Zb2>+~wv8PLnBe>aPL*+B15!)SM`&!~a<A1c6%
zIm!N}MygyQ9*3Os#V2xwgdmm^K`C(|UXno)He0V~W|Hxwig*9@Dz1^-NKP`RXyax2
zF>a278n%<$*iw{2@q|mV?@GQ$^gT^8cVk*|L$XEPHm3j#x4Q&F2Mrq6s>55-&8<_+
z5kc!cZu(;;dTg22mF<(Q^br~`=PWi$D4>YAA;Ce4;-OHKl2A*dyp*h9=$>IAtMP_*
zhF+muv*Jb3nxJ|$3*^BY#9rhoBs`5nA{3RZfAIz{BgWaT3*U_8eG6+?`s&%-L3$+~
zV2GS`1f=Kn**&9v?2{PBwBuG=J7ZO5eXwU6oUHrrC262@f=i)jTdev|uk6BIw;7rU
zw`d<y1sD{9^yROi?96zDTd;5`*B4#0l#;+*R&`bV?q`Z^<Aou-k@5E{##~v_gw|8#
z2T2vnwqqf_0YWxl2@A#}NTVF)q*b0MlF+9nZK6eQ9zD6-urJYcfsj}5p+n4KF&k1l
zI`EmY_Gyl5)#{_@5yQC^F_BlH!4X3q?=sVW=JUZh2R_PC_nY;JabNDeBck<w5>o`Y
zj#0{6X&JpnM!Sl@U#6gdexUKtEz#9g3bK8S{+%nqPQ}WS#&DLRQtWR1IFGB8@({WR
zjR%!m#nvO2kU%m<JGWb!K8_oS7UjAgd+ljWMSTZDJ+_w<(ya?g^q$qaYi9S)$gl8{
z`w^e~x6SV7>-zd$#-*l~ZO!^$UV!Z8-4nKJ6TE8^5W)RYD9Xr@L9RWLqq+C^^~mw-
zAu)cun;o7P&JRBsLK2+V?~}Kwe_pWZt>Q=HU2i`hFC~EVjyLx;T-~l-|45j;BiA;M
z2mq01T?7OA4`hn<iBX9>1QrT1Q|67FYo7Ey!?btFNo6`>)hSZ~WHX985A|;H>Y&uY
zYEqBals2*x>$O|QSd&|+?mnF@eS0RP-BGMc%8H14g&dxo@obl4%gD(k%;gQ75k~?~
z&jOIS-}|qZ*6PN6R~J-k_=MMcJa)UIy7C#OD2+Jo{u*h2qIiqFvI!-ZkbVFSkU88g
z)nO>a<rx?b1IjZ$-&Nr{tXfyYzYOI~cP{PN!%F0Dz5xqqC?*b8h4c(!ZIxH%yNK3f
zlWbO%nmzNoF1KPG?sl_ONkx8>&XIZklG=V+{|Ak$ufO6Dmq3eV+o-wjN`M}g8Iof8
zg;PCb2TJmc%*@;fL6wUkDwM_jJ93rBG)dSD?2=5_@Te=1&fp@tU*H)Z7Oq=_{itu0
z4D(JvLGeX|@n7vvoXRts4S?k!?qoS;Lr85KvOdn(@}Ylf3JI+)Lihek*Xwayxz61R
zFT3%$lnL^Z>MB`0gAxN8&|R2l%zaI}dsUM;*XRhl?-08g<i_()-^U~`j$(|Qs?<zJ
z-cpJ;(%#2X9oVD=CyCHku8Y&ZiJ!!N-Fu7^oZ^TwcKW>Z-@uI*4Ku_s|8l4}v56lG
znIb;+8c<8*|Fj+iLD81rm9?@ZWg?7NE$M39Y2nfPe9EF>ov@H1At|>HLTgP+aEj@2
zKM=Vv5?WEHvGV$WhL7*y`kN=LX_P~OaVaNi-IJ{ynYu%VA@akjduHKHZ#XeF0NbIN
z&Jr#$a9ThFa?ldKrQp)*@Kh~ZccyoaMsFy?NqrAg-x?&~U-NXxSD|;@Luplv1U>J!
zHRz5cg88o?dH4;J8sle}FH5nB(v7j?^L#S+HNlyVM4aUZ2Ns86E0SBWX}M&jKOCdL
z?dUFK-(4c7j3h}U9ujdBn=xONnayBy1ld1ACoh9%BT51$O#vWD+iJIbH<Q=ptH9tA
z^7PXn$s>?|k2ps7kLdR;`>tDFA#_v>=hyQ`cAadc5seDTvk!5eX~hf95AtA}Uj4gx
zj8>c1xQ%MZmVxjC1G`ZAiR$4XNbVb4Yiv8*qC*`vj64?P(T0V};<&JHtX^q_)4$<W
zIUY~Hsu6sNyF~M!bxk0P*~_XEjNkcup|FCtNm;(NJes3OY&P#)98+5I`scwDFq46M
z5~3a9G?A|XmClL>yWQdN1pd5VWnUZfB2TK`>wWLY-3E`G94a0SUX~dlFr#&kONwl&
zf9qBt>W}Ds&Au+V|KnJE$AEc%v&q*@%cMHRtaQc2n{fp6QHSVnl#Zfm>s8u!B*zmN
z6U{dt%>Vp8AE(f8b6KEcJh^C=Ze+deo+7>iz@^Xf1BN@-&*O;>E+x%o`3r8IRAR6A
z?)u!n+O}O&d2))pO0_*XzWldvNDszN$Bm;W=Sz1EJH*aA&ig-|<J=MbeOK)AP;A8U
zAKKC7=%-f35Tn>{39-)ebCJuxMwc(IJ-Kq>qWGJ_!38(Adwg2=?o9B>-?7-M6ciWq
zV7lXppX29<d8(7wtEZ3Pzht^aC*oV}a9CG@<u(`ghRo~(kxm8|4iTG@`JMTu7pq`f
z=zKafcKSQ{CT1DuO9Wf8+s|oz$2zyUYk>M8+_$e{NX#X3IcqK0l}WNs29ie&);(N{
z%$EH{I<<^kxpLv#0$hoOSraEih5U7mJnfUYJTd#rL8l2DVNh9lKP%+H@p_ft^LlLg
zXImJM8jPB)V)9CzZBeZ-&RRwXFcbd%mrG?rfkO*k-W?sbTv?(oy2O7l?>R6ZNL_H|
z@?rOmN5S%JMu*VvqW#ZZjwolm;6|3&q7cR^%?dA@su5z3dd0jT@X>+iz$Gu<%46dY
z>eKT1S4HdiJBA3n&IVF>gi<uqWc`Urdwj-_Vq0gdI;^1c!dlFHmaE48aPf}$3h9z9
zt(B?;&?$sI+n-}$dsShl#z8L5{H`sz>3VjBu@uQ8lkyKjM$dTNx<PBA5)`6s$956R
z|LV3tRoJx9bB{C*9gzJ2dt)>wh`MMd%Yl@6t=XCRM4QalBrD#WCGly{02$kZfU<Rb
zzvE>BFNrl{^1Zi?x@eFL=!x#CVq6Tw&Y_JFPK^!#vvg&M1Q5^FR;GXzL34H0(-Pl#
zEtO#16m!kk;EhMl$0}9l6JbX~>3FHa^AYchPqWuwAAw>XkMWsEv;<x_>b0s8XwZ>;
z<fj$nk790M`t|lk-zMUU_@<fOm<xA;ykK${%kc7xxo27y1R+mgT&q7%0qypvEB(1@
z>ua8;2~X856m_7_wKI@Nki@d3Mk=-kF%#P0qgXXQ^Zkr@Q^10uF>DzZ$ojS@Fz*8l
zrI1m5+~JKZQ5Gdyp;+1vlN6(ScHQdSIU&xA((A1N2~8grmYG8}<5TG@RNzb2haVr?
z4G)p_rL%r(l<OFq+3^u+nYZ%aUa{^^B4OsW1;64Ye*2IOGk3?+|32WLB2eLPHK+?|
z-{kgp+Vq1q@rw|<GaI9bj4gqq;B%`wQ&-5O238lXyebKpN2m?x6@y;Zqpguetqq{@
z1eg7(>dh>Rz=^uK;}3#DhYYw{@i34-k0q<w??`GOlOjC%DZaP%jq(<l?@9Yf!ORG3
zkm>q%16Qe-$28-JNuhRGZVk?vox0cC%LFzl`!|n=TtDYnQRB?bRbOyHjn7MA8S&UA
zL*3aehL4(8%7jQ#Jo1-xR^|Xi^rm`*%0BJjDq3^b)Um8u2b~Jp<G+JRZ2x{<M1n%a
zxk1#j#(0eK6gc`Imx2ZnY!&=A(IV6u>tieKAk{&S|B1uR6rH8b75)=%9)6bkaQPH1
z!+XG}U=ewFvc`uK(*IV570Okl+{Z6H&{!5);+z*WSgDbugJ`hu)AksJOs|V{NGB-o
zL#a@+-|vsTt8_VY=(QY?NPb_0Ex$y}v6C=C(tO-3*eDy&-Lk$@_y4;8J6c6fR};_M
zInHlrm+4<${JlAGKH>jY<qmE;f4$#6ng9M>4By?R;{)85J>1YwM}NMwWYf#U?B(9^
zy&Z7>a{vB1{r%J1K0ZE40~R|imwqr|_X|1OY1nmdx<UoSF4&sc;P@d9&!ykJx$kW6
zp_qiCE?M_nlm!kHgiz1gL^@1wrr%gkRXC~G$Po1^95`yqwN?<ue1`0%$(9QulO(T?
zU<I>2CTP2X;b!T-=*)RL4soSD!fQM-&rTcy>KT2M`s1^E$AhUK$nc(=f2AaQjK%CC
zZN+As<uQ~hJAu$rvnF+){Bch$gY&?YO=79x`O8`V_k-S3U792g)l32*?X5t?xb@XV
z2%PIYi03Z>R}BJ~!RJO%4#8vB6D0q`jp08?%NR`b5VCtEic`~F;Z4wU*E5}yD044T
zSOVmMDU4vtO+QPC?3ER%o?n}ZzQ@gdhdtHPsjgi>{)2Mxcd{>-@WfTr179SnV<T?t
zeP4p?i%l<}ivawg+%0+TjY`t{-pVV=j)%n0n6?U3hh3X^N`b`hevpEI&ICL(@wPi8
z`BSOb7tmiM6hOb&Qa1Xxq+=dmt2<IsQ*&(c#mfc|-s7OU4Gz1eg0F#{{cXI858^J8
z#UzUx+YE_>$)31QtEJJuGvNBSoDD)=RtnhYuIyu%M}6+kw**1%$NYm*Y+sM#zh0s$
zuW0pQ*rK{FxP-%I0rm@OY2`m$Z_NsYVJAeH>_z(o#IPUESK&iGljrJ-<bsHch}A@~
zlK%s>ZHd(mj>h)du&!l=nB6{n_L!tYFGF_s!NBd6pdF;C)ndhjt04IqNZ+oKNLWJh
zjH&t`Z2i8wr$6BAf&!0X6~9|uJ-=lsgXt<}nbYT(i!@rfLgH%_S|nZSW8<TUa87xO
zzu5F*?cl3KD@h?RZV@BNRb1cjAP+*=x`T5!SpUr4mWl^Pn+Wiz@|kRZ&!&wSIdUC{
z1CA*05vSI#__&{d5kXo2v`PAz_?RYmHNGA)x-xL_0l*9pU<1Mu-OmDZ`!_NqiEx`t
zae_?b+n%ShWSp3EGe?!<2WJ$#MaDgsrIYI$Bl#0@eWc7|=GU3|Q@+CeR{3O)9NRd{
zz4pm~q2@?DKwW9+F)3oS7hBuZnzXmZdY*wz$*SdmqPH??@^^r4Mx_Fu6dDZtVxBgs
zc}32)k2xZ8iZ3WxpJiA=!;NX!8DHF0dNVaL)3?|wyQ=C|%1cw*iZK%D+OCM9w$b%3
zIWwJ)5%+Y8xHT#X9-NX*o`HV$EuDfuy?BHqU!gOo`{b1ig|TaE`Tkp%23fe8vm3h|
z0T6OLo*^wk1rS~I;e74zFf?3-|Bm#PQc?C%kHbRGxVfg*QZBv4zY!IEvu5+Vg+#6v
zOZv?|&AlKuY&qIy^tlL-QPBi74TkJTrNeMsl&9g<6dCVz%#_RP9ZM=P*^`9@=P2N}
z0W&Oks5vZtB6zhwQvHeRQ9xxW4CnL*=H7&{v8`(=f4xP&r@()eLUkd-3MYH{)M2e-
zQKzUv^8b!2`@ZYvpc&neI`2q*bvhkv&uG#P>n8>c`b6$|U(1jcrTs$~Vx4hpUx}|0
zdAwlItRyvo11tS^4V69WL%BKp*0B;u?nZ%Nw#=VHn%v0dvspZ&`WY18Z=K!x6SekN
zBW5+`E`CK(YznaPB9dhB&?O#zM4HZ_bN+>Cvi7s+-F?wF0wfD9n&uwl>V2Cc-j)54
znG((k;9&$Mi0dK4m>{m1d`d`BzecK2tBclqeIUarw!v(Db`~O<_Q<j5%qq<zPxA8^
z(zLc{QpvtxfzjIFP-=J>*KY7vje^f=g&5up)hT&|<Q$hZLJ>^=^eSl&g%Sg1@<wk=
z;*y$)Js1iyeqFLYK9~tQw#p(wdPkc<QpxZApnWkDv#j26@@Drxc{++vYs8<u&7u6F
z6#b$oGHV`5va(cK){{Mqp{I@7cl*TFyNtB3Vy*Y{>O44d5jEQRQ=r4xd*~qRL1a<{
z;bOG%+%aEWs7r*6cxYBjQtyW=1L9{{s*e|+_L<iE4keUObaF-z7yL+~u-(mu2LOhF
zJ{3eawlQs?@8(yRlDXftOj=TRgIcMyLcJQhJuJPltHL|SMcFEUg<HF(_A}qp1a5~f
z2JhMOMPp97FT|_`C>@2z7SBdT%zm=2x8VQ&AAYT5p`p{Vhe4iOlY&>NP4lUS7D7il
zK{CdR$B|p_GxPWMBI~WZ$m*GFC+Oj#Z;pN2RJw7_ER_OU0QoV^K0Qhu#(d@M`bd(<
zM4nq2UeSM$tKWCh=4^D;vpT_YGAZzWH{>MVX%-mJZK}-ccx(nawoYB5h|cy^HO^bN
zeNonNNV=i+*h1=O5?!N7pUQ=_((CfNekgy1D67ST`MA_qAsdyF`PRlSSxpk0>^<5!
z!<hA04fxGVw&({`M>f;yBMM5`v_DFe%|RN2|6-y%MFw)Q4^j)dxf(hGD(-8HY{7-^
z%aMWZjKr2jFol{2W#pTEz=6llaQde!C#lc~^lctAr7oB8IQg`z?cD*B0<eGpU9}oX
zcn+01r)JF=WYr#$?Pi$N{FX2D1=P_D$?#-B#I-Il=fV!MvM)dnVN<$OXH;_ccVWJh
zQhfTu449&t<!_C%k!R&RDp#$!gWCqfB-I*Nx1cMwu&7_61!vfx{}D7O$e5Vk_7?GL
z80wA1PQ;K(NM}h~rZq%Jr>ygP0|qoJ$DKJ%$f$V^WgcOAY+z+oyRM~{(BK<8V(`-W
zvqROqvlN#QfdEh9I+>VH<QD*RT^Nf4e-8>}6c547hH`DS7+OR7_yybgi_&$0@4Y-m
zm9TpBce9DD{sT=%Ut!tKVN4>vLru4)-P`*<$Jk}RqETlJ|CwRPiCe9L(6F|Kr+ri_
zk2yDK!s|!4P))9jjZd$)Py4^#l{fSx0EHj_6Off#uG+Wn?{)U@0<rM%{|L0WPv=jh
z8q3ElnzpVsztCuo*DPLVe@!yE?5{t&v{_Bi^D__Mev_Ms2=bOek+k|%F~@Rjv7q!i
zNe`{^EqAej*6~}Eix<bk`|;}@lJRTOim1K8u3{y6$GG=<t?}F;#aEYrMwJ)vrTjJz
z`RnS0jU5rjwscx6P3sx6B5A696pa_=(Xk>e9m)&zF+AXecJixJ%)nkml`(&~nsl#7
zN+60TxCISN99LcKJv?QtjEbiR$sXMuy3sDM_H#b&Lao(h22E2)3HQv@u}ow6%=+_}
ziX$Xu@oN#~=PJulb%Amrqsb834D8xFACtI%_&2?+1oZ;Ah2$ZI)0)o0)J;U=GczUY
z6zn^;acY*KKmVT8!mRM0EYUN%O1TT>0%IF!fXpe}80jj9aHYb0`X&Xr5_`Oo75Ku4
zA2-~n6)~zLsKEVXXf*fh;S6PU*6`6=uB~ir20fX~_E)i*Mw{p}-L18}Pof0QZ^P7R
z$K~0Sz)Gm*?gsf^)X3Ug^0R)8^?W)x&`CsZZu;ihIfL{J>%K0!Zs^|Dukekhuk@e5
z${yNC<q;ta6Y+fGgOq$ehw7f*u=q)04{z|_WtMV<&{go^8BJZd=Q+3@o2K9Qox+Ad
zjhQ+zy?+Y>SAy&Fy@@S5cVBK6`rE|uOH`7eX3Xn-kr5RYL{=E6hLvO=eVQlL#N2K6
z(0zS$@AD6jx<=vUnQC^j8Vt!9Tr+bFneq&N<{xyroHvm=D6}=4kY8OeefdpG*8z?r
z3mRQW6N)bD?R=Fl8TElAIPignd(fGvgh0}<%yqxC8x%1QgyOz6Drd-hYINt6mS8xG
zdl20gXlAWa_wfAfNn;T2Wn%TzpIUV|Q}fF3K7H?G`^c9tnbG2-5<Qa@E@pLL3>8ka
zS=%_2AOfT`8gePLiD8v^G8)zh_ew>W%#d?Lfda>d?&L!yQpmWT?yg7#n}$0t4oW!*
zHEXN%LYSQAHl3o@qA^I?DpiNU8&f++2|?okjAYgoa!9(%!~!T=wj{RZ<D9xFSvi>d
zUO{6agLAq!0qN%M1G`MJ4{LZhDSu#1)=cHnAPVC-a946H*s2`8XJoR(flbpgi(UPC
z2=9Rq=;>;|W$g7<Py(4*f&;aVW4YNGYZEQ17Ih?|DFq_kD$N=$yx=Spe2{6$dVB!|
zz;Lb#%KLuix7J?LeyU#y7||CtuCWHdYTe6bU|>omFHYs?uYbyVClgQNP}6gjc$Ko^
z4miit-`{*;efFYk8k8_l0fd{@njXGub_*S7xH&wy>SbEmb3Dg4Lgx5Il_4C~H0m7W
z6PWVt>t(+V+tC4dOQZKfu>rsIuYg<oRVdgGMu|PM{SSuCAkV4ga@oF@dmfTIP88qp
zd~%IFKLq4@`IP}};vW3~-Ps>FWIW!bd)=DYhP1SkugRh@U#W+ONQ%(6LWl9`|D9C{
zgRcl#`0(|PZYCTFnmQW}nmUZ~8c>*5NWWQVDpyKict4?G-#5-#IFtG;!D-ht9c2cV
zl}E4`a8c%*3z<j?l~(RhP>1-;y1AVHXq{g*w*q-KznA;YCwQ-;zhr&j0yy(FK4_Ea
ztEpAyH_P>FipZ2;x&D1P?ngGZAGO9n`fQVGp0Taloa*$s%ynVkT)o#5#pKnVCjE&X
z-(N3DXCBqya`Vz8TDG-D6XgNce&WH<;?H>yZ(N}S_F3w?C(XfqLafN5vGzh#W{6AG
zJl(quh>QE}X|QGl+s<jxhv}~*2sv?Wi#()6g1}%8{&JGMGWkpCXHmKR(;59v{S!j!
z15Kzs*N(KK7zOe|@D21E$)p~Cn`^E&0|Bx2arft*yu@@KO+_V>T)P%Jn$>Ruu?Z5B
zGt6NW^$%85<Nmz#a>YR>$KR#>m6aQTk<7xL$3w8GoLf<Ya<XMt>EQC*uHX25(bZsi
zTFo>i2xKKQf5ow{Clw2gw_9zyIw!R<7$U#Cvpy)$!$P3wFRiA$uZ1qoW>2YIlW*0F
z)q=dF^SLh{^ZE~#deI}^zD#5R9Cm-4JHq7RaQIgRTBCz(=yT9FkDKBI`bK%vvru$$
z&|sdwkj3YflAc2R)S#Jy!VUx?p#S+pzw?<1_*g@UkecbvgjYywt)KycJV${stPdq^
zZiMx<z3zLTMST48)dGE##sNy+P@CMFaOzIC-H}1}8|K><*f=QpCSy)L-uS+0Dh#`0
z2}To1%o{m|yycoCa)jOsyujg5mKmiG1cgr*JPglkzk$q%?*V=O(gwo*CLo_^PCMmb
zUocJOIbnV;y6O<(n;-B_7e{UNY%8k?gs}cD3t^o&HdRytNLoDY*uB%~kT9&doyaML
zs;I_-yG(hW|6@uGt8UaytR~4}z&0;>t+-=f8UPJ|Nen8BAg?;$?eWRqy2@M!YtO%N
z1J`Nv!{eINr7;v^_pD*sUU=#&pc8<{i<55+_Fg(EaQX<4-}<O&AJV*-Vo9M%)5dFu
zP=wO7frFp}#dx@_o64!=H6b6pqJ$MNEMVTSTRHivkE&hv%gp{$qWqo8M?~R_YMo41
z&_K5xL!!j=v(y12R~POB`X~NP!3E*7^Pw?G!DnAYVR~@*28*<|P7K8dwTSlOZ#zPG
zqS4Bs&|*U;48O2)^&zkh)>R6duF8O7&P<F^SzUQh=&FJpb5opmbqE1N5fMd;k8@~K
zu9#9l1t)K?4=IkwtEP^=bn20hN1IVMpEzAo*PsqVQ$anp1^_@4YaF$#eY9Ei|GJp<
zTeObnbkff4fQSc%dgG~e^tXD~6DyT@heP2J$185QcMWr`U-LCu|5MA%1~d+6Aq%H`
zKrDNUgS7UV+%SO7l=*Po^tqPt=uGP43a?Gll2U<m3~caRMwPYo{L3%sdtAEEQ}>K|
z@XhiB7!re&KV|_6g@`RhDMx?nim|9v!Uy|e&=%UTKAngT91ge4Nv**&Pw4&8SR-ai
zB#BVW&Q}VB@d~{d47FT14=C!}$GqOBto&l41i$+9PTl!%7fnv>jOAja^MUENZdD5;
z#!b6iQvi|bY}LYC!|{o3eirc@n?&+yzJd=VQ&nj8Ilp38nMc&sRxuE3Xo35ucFrZ7
zh|smLMEp#n@|M7lNOe#}Sk4=lZdK0bgkgmjsa1A`38m+#Uk3*Kl+z}}7E4b7II!^W
z<=Uj5`<IbPlI!C~!L@$dG6+QygY0EqdGwUhg=gKyk1Q0|=kDu<ae#+G4@{aYZEmz*
zFJcw{#P?hB{kK-)TOWf?n$`<-@As&J$P7$gAQ=Qbt6Cq$`Cknx@Pwly`GqB&`#Jpn
zkV5hzuL1@}L)OZm{+md>b~#T@yp1SdU9%<Xn1rOV<7oOtFwMK7`s%uf97QJuRw@02
zKklyHaLQk)obbRlt!S~XTwIS1G-uV=+t0_tmc1?(w&^v-`;h9h&6KXTxu#S(2Gg6T
zH%Jtrp&F8(Gs!hq`|NIx{U`K^%G~C`An<F4({eN8=*HdpCkU&`<ZDgj6C%V*QuvuE
z0}Ezc=#dzx;++k*?g!e(p&$~L_7hLS*^Vq=(WHJm6Bv$ajh+np!z`&GRalz-r7sgF
zgK-73Vk75l%!I6InfwZ81>f_CJpmjH0D2i4ufPwcm8ZbE5hw;rpPkrF^E_KuJ4ThU
z$#jkk^7?itssU&QY#vB@E6AX?vH=~wMrzUYUj!#dh3<eF=eNzztzD6(dB`?=^#rz#
zNiEizm;yPNqU)Mnrykx6Z1FxejRy@Es5J6eqHu>;jL-UnY07qEy;iN0kF(OfqUG#2
zWIlcgUh%B6g>D$c$uQCtm>Hy37@uOp$VaY0ahqk%rfm;8#WB0MTvMv73=+yw-V6D*
z4NO8XJ!(>C!{R9Ar68Wx138|u-8+TBA0>ULIe*hfgml~WwMP>HzBn4J*q`XyZZ`tq
zfR87b>DkqM51gT3!!C%B>f$<GT{GDE%-DGN8&68ZtZMh5w|Ml@{#gHB%M4gi;#&SK
zg8y%Rd3-nkSj<)5d-u0>W5F>BX5BMT(ynE7&vUNECv(H{h-2ZWdQI0ZOXmFj<44I}
z&Q`q=GX3Y>8kyF`gFYjCGuvLXHqknzYo5CXuCs#VhwaMmx)Qq1c5*KNCu26ecF@hS
z44OO+hdTm%iDGu&EOolUcm^^H0-)G}r!cX~3jj5l2B~w+GT+8-`IKxO<nSoL8$O^V
zGy9q*6`Y-W9YIhX|L}cRP<e|8kC9N0QAkJl=DUe1FK9R_#eVl?vSvHC?uxWTvPKR$
z0+XNawt(P@Q`K661&hfIj}bDRD}Y9N;Z*&{N$S52ypPD)y8Q;hj0)(63f^GYJSu2q
zR+hRJilWhNFKSm&cL9`gYsaVzmJ-dctGXPYH1z=#KLJ^x`B>)z8sjC0<0zZ%=a45~
z3g3LULX8PVu)z|`f&6e^;mpq6V0aJ#qEp$n$vcPkK8Dc#7pw=H+sLOybOxMWp$cnF
zqKbPb_C5Qd$pEw(a69)ydiRnRM{A*Ff4o$Jo4`gqrt!n~jviUe{3osQ#@Szo`K#sR
z69@`DZGvaD1)C@T?KHj`1oH5lT&s7F1@ys`*k-{fSvF0(;X%?DMS`DZJ~VNHIRnny
zWzDF{$cfQ|$rqig->9>lX`4C1g3%~gk9|${{rC4G5@@tehsKiMxN2u8_pvy3_Mzv4
zivh0*_&iRRE)@KVC^;k}8UJ5v`R=2X$UuSW7J^GGDUBtXLaNt`&9_`of9&%w-|lYT
zaH`tMV>x(``H=^^kl?$GGZQvKeIb&f_Q5Z?fcuQ0386?Oic>BLv36YIWj6XkR?R~?
zCeMwY+GjV+5%$@2d85Z@Jba&@DfY29;K7GAjinX&=zXl9qS2>QE|^1sZ;JZx&dK@Z
z<CZvCBn14r$USmRd9-;qYOia*>JF?zcL5HF0-+eZ79o|;o?W)Visnt1SI##qht`05
z)LG(O7Lkyh8}`0r&HXBQBIir3J10t*keM|8X`XR0Yg}$0H)KD6O#4Ti;2c}|i)tIv
zkDK@YbG2z0xo;o{sO7u?oNtYgTt$n`0~NztgKuLxWz%WJ?K`bz{>VRx6-*I(75C$B
z)Swb9v2SubP&Cc(iiEFsrViMF*xIka-VIWh{*Rijng>GfJe^~!W5BZb-1A2`=B=&d
ziKwyPIvCWpC(r@__%PDqLHem~@Z!Km#<=FCua^g8pFFq@)TT~^Su(0!|JddE`DpBm
z;uo+?2X9`e@xhNrW7!G`QK_%~f}%rRdHu!Ql0DN;07lS;5q)<ED~i|)G;?3)JNrn+
z2ou(!YJ)MFCjX*#t53h2O<a3jx|T22$-tljD{}OzT@K)7wh3N0J6!}&O<)2;O9FFF
zE`Te05olgOqRBV5h3O!?*UlOH@Va)Jw1bhark))LWTR~4t4i5o^^GI?V+b4d;`WLV
zi7`;JYA;S`6HM<kygs$08c;nQz5Mfcu*%=?nji61riGrtvktKq<U!m7?@CcMJ4`m|
z!qj2;>U=C75A-6~_R|}cPw1}10Z{8BdNd;lwlP=QTI$a8-pp9&e9pfoZgo1hZ`RR?
zYDc-KJU)c=AJDv4cnF3X<?a?xN8OzKVFqa=GKFp@fR}Sq0(LoCffTw$Yn(yEBUm7$
z6cd0Fb-@=+3K2U2Lhh_B7v^sxGew_xG)3PB4uC+G5j!D@CG`syaC6+G(6Zkwi%eZV
zrdw6-jMAK)Pcc0{X}>ZFphTp45NvIi!u&@6S=*j3t%<?`x0W@0bHL*HR?JfhW^Qo6
zUs1vqy8xN6Sau2&yDqyiWCkTW1~&r`kwRDY=(XaX31XYEX%X&Ca<*l14?njxnnC12
zX}La@9%M98yMNIAS$MAq?FJ>nLl?Ja2z#HK?}l;|5-dSzVpa?zIY9$7;DTU6eJT9i
ziD2ndBCQEEjs(cX!Bqu1Q)Cn*0Q6|;lnPQi;wXL7-ojZy7OY217s`=G?&~NQiu^Km
zP(JdlDWg+Zl)j;B;O^F6<M#sKWw-j*o7OFXNa#?mTdlKI58EZNvtML@s9M3Nl89&d
zLAl`jDg|<y3dP-{fmF7t)aJmel|JOxx}_7z@D>c%=Nk>moY+LegT7}7T>yg_FOaCk
z`0_u&$WC<?3Ss4~<haktru0n1oSPe{#Z&m3O5%`(VVIU+K~c=MLo60y6MYlYFp5z^
z&?@c=LY6`LO`MTH2)MeGk4i%F;^ZhKI3SayXrNXUaUAyU{fA0LKC5zllDf)#&iA4*
za;EX4iRN_ZM-e=B5|&`J)8=-5n+DVDS-S=RSgmbt4zA)8bUy%ZF6+RmQaTtk(d58e
za=ts3uIPNvrdxTWBa)jn#;GaAMd%qaLAu^SGNT~L9$!JDf-e}93BQcnkzr{Mr>Wod
zRp@#I5l=pb%je;l1Y9K@ZJbY^r*)&>u)Znlhy0wNkD4?YQf~$NHtGm{RnqL=+sv*?
zk8~WG%`)e0b+6(l6A#@0igCVsD7@9Pf_op<-FI|YT^!03=haW94;uir1*P91+6;gJ
zha~S}<!H@0Y*K)#qW_&@{NAZ1E+)Rrx$i+r_dy9x+eI!1cgMpU4DP=EZsa3Vy;Km!
z(nrCn!+0A9s;}8uDvfeYQiaFWfzDy?oh91nJ_A}p86;Zf6?F3|X-rbA-@L5?86!n^
zzuGO4BxWIpJiSXJ$-1GOx{?!HSHo0mufpn!&WG<Jjp}A~w);g7v7k9p9wx1=5L<?s
z2Y5dYju+}-+TZs1%o-X_hia*a@PZORIo|`t&HY*rAnuzYg6!Nx?g^=*!-Z*}i0LKL
z&`E5MZ9c@nB7Vo@=pm@<LH?J^e6MeyVdDLJZb#vn^Z(B=h2^)1Wq5q-w+g%6#Kx}6
z&wyMubtGC&yFs4*o<e1Q2N1h(VXmFKgdYx&u-pBSBI;-bvK;uaf8~T0%g0#?q_C?Z
zTZL8lJsp~QWH`2+0J_<0o1A-}3(s_T80`grn%qQKbt^X%e6LI2ugGc=j1z$qlZc$T
zjd&zWF&z&9U7##RthNg#ps!>2X`qsUT>^M;Gagvk%sNC1&I}&@7_naoCKQy|tAQ5V
zsQFR%TL@d((Pd8H=L7$1J>C9v^kaWdo7n7}(6V0MNFfC+@K~om)PdsO;>-HM!n*~_
z)t;;jLW6<8(#VO1`-kH@0Y^0o2^Luy&LGa~7VqkwwD=R+_LT<TOl;X0$txO3k~x#t
z*BX6$83@3f?#Y2grknjqi|nm~8FY{1qstyX?PVYoJ^FSlg_*PK_HoH3WX<|62#}9a
zgpyBWt)PE<5Hhj=jca<N7V;C&{V7i?F-$d4{X`jbOVZdE;l0<ggScwSDgnLTnC>5Z
zk8^-KWI~qVdUi-<sfPP+2{(BUA}4!C7)hCZ2rN*qfiU=Qf4fqvBS99L2ehGl5;mo^
ztFhQn2}!eCgG5Se%f(lBRzVsCDS?VZfJYY}p1!m%7R09rkNk1-y^VfSQJ-Do@QuBl
zQ;~JF!g&Xfqxs%+p1jhz0u4&{6!2gJyda@8o<}neS+E-cs_b6GBN83bxtiSjD$%s{
zvL~|Xs9z`B_&0Uw-@TzCWHfzWBHO<A__x^-Z_a^$1%|3<8Y&RM7G0Mh_s9|UW$-cz
zkjSR|dNW|SP`HIjjd$z*FuoZAiBmHy|1HmuH;#w!*?REuGfE_j1GA7<3L{|X3418C
z|44t9d7_4sL8s;c^iWfOav~&44f}YI?S%8^qM~gA18G0`d4z6PX_-nEGW4{(I_p{A
z7<nxh+_uHHo8A|(Fmr?W0~)GgI_U2OMKxA|!5mnWtE6;)>QCtQZ&75BkN;};_L{)J
zZThw|J!ZYEEa$XG-#z1<6kE>EN`!^rikr2GwTbnuxxC!Iut_ra@Sex|x8t`mKmBEP
zN>@$msmTIWld5Utds#WrB*jO(aykf6iFZ2h$E;(v>4)RDE4J_a5v(z%=|hjcw~G%=
zVAep)3>nJu)l)q+`y~ELD?QoAQnI^w);ImmWiQ6)M#1O?<FmM3!INwManmn9?-ZLq
zYaTf~_kIFF36Z$BNc(kcYlgReh=Tf{`eWN$+*^>IP1MEJ;#2*GxSK86{+<qN1+U5G
zy=1CRyI0*9YJs`Ev$h1mWiElu=a$jNS?_1<qn&R!6VH>MQ2&<VtR;(}{;r`RsHh|u
z^WpTbw+~TyR-rjA#)2CCG_|*#hh$1_ZCefGNT6GV=e<2m6kn2a>T$#C)FYUm>fwzH
z|F4G^i``7jWFuHVi(}OJ#uNYQwyrKZSE<M|tHakz-ifcJExv3yFMJ+)nJMa~g&uT^
zDaz_=#XUX$#^@^bhG*_)ZC068I!pTNo&M`2dJKxlf0-lt?^iI+m$j!T1fFa?ik%5;
z3$KsPEecX=pGg-n=vn`{0~4Xebus5a?>5sf>wCT9p(2=|K|x#XoR0h(=eQX6IK1m`
ze3hxntnBIBGz3QWU+6gEYBrUR&pmys6c9`xn8*><Vo+AkY5CTEhdz(DAZ?7PPKhv<
zExGO+wTrI@XxpU#R<spLF!W}?w9B$(x6)4l?KX$9rZ`_b6}tQpski+{{~%>}hJ{Po
z&#Ot|<tn`#U88H6dBTKNoW<}6n%NQ$kT4ba>G5QTn`eb}jdLsAZ2@Rn<2uuLI^&HA
z!M87?AB=p|tSwjV#Fms^QuY}f@MxH|7xpv>Dv;B3C3P*pvrt%{)z6uzrZ&=Z+;7k(
zkbP6A-7mtKo<ty;TcjsOO(?l+tXFJ)GF>EeSb%51wEEv}JlzM{2Aa@rS!x&8U_3b>
zdr404B=Q%n-6DD<*Y$?Di-pP7)c!jKE=>XJ09z8z?%1y2v(8sN+1G}YY!^7aaPxwG
zL_hE#km*h6bm=n9wF9EgtSU7lhojrfX6?sGNw)xdYjgbTC~eq98t3y_ep%=KBN0(E
zT;v_2QO%a_!E0;C{M>2I`Kdsk%E#$tQJ1e#1ha9%Lxk{|FER$hy<^e?<>tT6e=EPo
z^>&w6rq*Oo@kvKf5u1e`Z?mJ_tGY3A)e*{BhZxU)uOi7N;X+hhbXwIIu+2(mS0ADL
z_(kSx6{E0d=$owxatz`^XL;A(**A?5i**xP_Y8{m!g=_VXC=N4-8a8*cDNS%^TXzr
z5P(R<GkJpqH5j5ayb~|WdWm!-$e#s<^PE_Z>vQt8BP|v~pe1b(4V}wj$~q#pEU(>F
z;&Kl*7k8J{*pDR(r+#{rcFk0ED18~Ftbx5ONj@aKT=!Y#UUbgL{3!h8_t?b!;vwmc
z;ko?+YVD)ZdmZFVQTr=tF4FlKJTt!pw~zGORe5zXid8ug=K>~+VN<%FdF6H9Ru*XV
zfdRDtc@KjUQSQ>=jlarx36g<fEk!e!M_Jthis5El(O;1i<`2@RcZ7nR6}j@O^al=3
zB6=g0G_ZT_%xns7ziPtPeqmX-KISVU`N9JccTK}9oA?`bl?ARrya4s-XPN8Z$g4)D
zm2Fk|37EF5Ln2nKfYI`AO;V*4PuD^fvqiNbN|74<<@7|pKY!Uj*}savuAwa;rnfq)
zxMM(|#}PI2Qy@x&i|YyXUt*2CK*0w$EtHN7qW_}`n@)>j=Sk+7u@?4Jd#%W*TlV7P
zWK~aQN_Xp8<W$D*tbiWv0p<kBC-mmL5o{H6&g7oU)NFcS8SBBN2TwZ_p#Tn6M?hlY
z3l5G53U9)7t3w30(3eaJO*y*!hx*?)UA<Wc9D0_d%{qIOwZb*h`ZCWO27oN9nC%TQ
zkf#5>l{Dz54eWOuj~pB|*w4@IIDg?oH>Zmn53P<N!lP3<2Da>L>Ip9UsveSO47Rnd
z{I`yq*Y6fXJRw-Dlf6*a$l;*R==HWbt8ap-c$J6MFX<lrzvg%E&hq~zl5j4}N#W%M
zrB3SEdFT^_KB>frElZjfpq?QA8>H56<YoG1Qeg4E|0aPNC9Le%Ef3hVI*rQc<m_Hi
z2daMLgTw=0rp~YC?Cc2aytI*K_ot7n)O)Q(>h)Kr4Cet&Oe#+SMl7CSyibm=JYB0o
zsiepw22G_fi0(9{omslbRtBM%&GCc3y!v~7LhpLs@PE@(mtsp?<Q(&VyHc8!l%h<l
z6YHqp4dJkvO<}t)qC)g{#N;?NgVWoC?&xyX`IiM0^_~CG!_xo0lQ5>0BW$oVsCm5y
zh;Tx7)l4PwT2NGh@JS{6gKDq!HV|tJWm)pM4(J{DLwbyaLQr=l!^13#tkNG@#4wr2
zq-%&W8CPd8SQRfz@b593W86HUQX-pkZdHcSY&x&oZcej(4Nck35yctdL@*v0%yd#H
zRHE+O<TwAczKp>|Miv1##J$5O7qaAnlNk~<4eW*pjggG3^x+EK>aprCyKxOMPCsML
zZ)<49XxGTJ4H=S8xZj}kRjGZjl9b4oIiVpK<!GIE&TI+K@GAk+4Q%{{BHa|-`PV8>
zmReqWB_>yf#mdT^UY<9+jCz}fWqB*FZ(Pf7u^FBydUKWnJF3RIxeiSRhsV+%5+r}@
zJMX+fTiWdP@l3o%#|z5W?B|YOw%y;wRGkfI;uoinBCc0{>PROxMlyC6&4j-c&n9(4
zQRzucJzRz#b78P|CZMbizcNQ^C)hER{3>7YU=p-xpAk;v|1gcY0>cqf<zr^$uw{ge
zDdqqr4#~iO)-N@ygw*Kb+;8${{r8!O+URcRtNP&|8sc(_)DLxT8ffk1g$ZCXMX55R
z^=E=YB|B%DKvrFzyZ|UyD9&_R)Qb%HZ<Da&oj4ur3|b$j#ClOIuGl|{&!tJBlVUA=
z-6Sbt!Xg63u92Q;S!Ko>IHk{8f1zdLau9N=(dYpYk9zsG@y!_DKi>DKXbAOOELlVE
zGw2JGgZR)*r#rgy$fYAFs!X+-52$~#EZNSIGvfl(*hA=&Od4<HUimKcZ?_oS!&}P&
z=)Y}g7tNyu{@do!J=*g6GsKO58*|6qfl5z$=MiaNBmHLsNfSZd&?ml3NI4CzZz`Z2
za{PX~M&=2X{Z-T;tAbrM4X+lcK=$&3G>+I5NX||^e*YPD`b*&q#-mkC4{LL4SqzKA
zu5FnI_KSS~eU19JfdfB=*{K3Qr8#eW%pTc&16nKV#)^qGP0>g;IsG{r<DWPj^EO=C
z+;w$gt4`ETuUH4ZFQpZ#JfF$=Sn-4YUH8+*FWec5Q`l-2;Tjq<*gxhaG>{1bCWY&2
zpEbYJS2T@j1*ZqTVUvSL($$u=hbq%R(ZF=YN%3xBlAQsqo_lf*@j>62ZPtgiW_$Tg
ztkYS4vkSSKy=N_95E^lG^teHkOe{*rG1|+nK+$!>`B+s^>7G$(<;=Z9$AxK=YFm1_
zy-`z;RJ%;=+>1$jl&-kWqd&iS_|?{{GGwq^b$IcFtL5KM(u1JbhkYzluR@b5>8ky!
zS@%RgIeb3d{&urnV5xaiBmqN5!mL5qvLIpYT>rWO$v*8xMeW3H-SyCpfY^6sE~6e4
zSlBhW+-?&6I&b02nH5Dm`?LZ{njxNV{&Q3$;|BN<(@P!p!8IK*>cikId;+W|0Ducm
z#)0Ur?3j*hi3b^zn(6Hrc7~Pi&tE>ypc|vcOynR>9!eP)>JAW4mNZ;IK7peoW-a2P
ztYOTpGGcbu2Pz2!kC3oIXbu&3vi~k6znnUy#Qypr${H`0QB6i^q2L$0ut6+TgJtIG
zHA8(5@{B)l-&Fa3c>3;eEZ_J4ccn!377;zR?46OlMM%h=*)yB6_ud&HWELWu?EM%a
z*?W_{^*w!#-|_pqI^ucm>$=bLHTm93+jPAgT`(33%Ik{DyD{}gr<aPyf9^}-NJ}56
zcvBW_Vt|7i7xq4bjN<DsqvqA?uMUUplaIm(-;mQ)8$e(HvxXJp1Po5T^jYk7nzsCh
z>0qmLC{Hx1HE)!zbG;Nk5xZ^=`zYv<-4f<9*;UZ?@5GV!n^S=MD|k0|r92clo+Hon
z*nID<+o?vNJF^8*XQjKCY-;Nn<4x-F^96C-*JGIi5t*OU(y#5fHGR^^fLTe-AdAhr
zZIH<QYtiSIOTCXTJ#MK!T{GIu7&?(+1RL)8uMbQMuZ+=TQb@)Sxc6<>;q4NOP~-}_
zRi~3ghI+?_+h3MR34aBhbfABU{tvA<tMQN3VTyC;>|yrHDqJh{&k9}D`p%7xe{a{0
zLj|2y0vD_j;#4`mH|xT$lG7Mow~5BaenxxCD2}K4G})Ld;aLQ3df6vwM63i!?sKvD
z=Qw$5j0B2)^(8N>FkdkdL4#QlznEOWubf2`SCu{|UC(>GlUNWyn3NSiTCk!JN}9f2
z51|qDmE)_XMlv|G$+az>?5&*xzg#VH<pj>2)e678RC-}jW6Rk2YJC2cql63jC#3;C
z-tf1WUL3VO-R_C>fmypoo4$``C|o_Q8zgTRt1*KYaCH^ETC@!;RzDV%2S`27FA2=@
z59%8y#1pd8MkmZ`U`>=xlG_#VYvD5}TD1GYowte-&kSiaA!M|0PsiW4K8OinNkOoD
z?-XWI|2QKtZbgPZsYCpdkMn6uQk0Cs_%W*Ix|Ux-Jv8RhfSuZtLEI}_W6I)LxF;2B
ztVu}3Sa8zVvtqTU5aBD@jW?_DA2OquVv~Fkv0O4O40$~Pl{}g{D@86?q<pylhVUO4
zb47%+I2N%}hvLCl%e47OcV^Abo23&e<PzH&)45v8<!V*a7Pe`9d_m!}KE$8V#9$Z&
zf=z??$K7~7|D=(7WA8G*mQ>zfh)>CmW+P-yhhsTRad+!Ps64MF9j){F-*rCHQ&0G0
z_VsMVy}x?@HhucP{yYC#QCTU8Q~!)|>u6l>pRQorNy*Hv<Qv;a*L53ISvlAFcOL`F
zB|Fh%RIAxOt}dxzu?C-($a|<&+ye&apu&wn7K9Lk>M3SQuaa4ePhD(xK-9NZM-|0F
ztYQ*SM!Y}4bU_|@n?jc!WO%_P53bZ^!)|^PwV5TT_`95mfZyQnuMiHU<ONh*)H_j~
z8jOPVYIqUte8yLHZQ6BS8N!+dIovB-P8pRe!EdB;GDdEybrw7`&RzyuUWnW9N|9sT
zeddz)c*nC>GY-Dhz7@<yr=spnZK3e)hikvoBFl5_5A@q?km-oSuvUj$!yHSFSVVbi
z3tv_$_x*+Bah_Uoj5iYRWqX`{E@TBzhb7n!<XJ^|@;t+;EEc88DI0MP!833@GgTb6
zLFRH|CHFuibJPc>y!SsIidQ|us>-rA;9+gaR6IHza9)z$a*{0c!pQ~H9X4EqEMh=!
zn(W&M;nyPx?%otq`}eZ3PJ%xzqaw#@7qkfIET<8&pl5wVr#}~n|1?o9HIx-k72ojK
z1wFNuZ{BXLV(1|RvnAJ7-^pMy-FaK-{x-ieAzc_kgg-{$ezkHsT{xn}W+5=g)Y&-0
z2+PI5W=(QCvk3|<!Iyv<KWX|{oK`c8ggWb!XTn6y-Ex7_uL>_SMeN16QqHiWuCcFO
zdEf6cvl0}Q$FQ!ktHgp3zbY&Cqs&8&R9}Sbsfy?2@^;|u53kCd<PxQD@(<Jh@`<p?
zs=j>B)t>fvwdZ(J3oe@a`RScx4W2rf_=YG@&C4MSX9>Qp39);;;jcx=D@L66vQgA=
z`*U5DK6P_c6&TXv6&6P;B{#yX<>HT3HgQ67&rsZubOUjyGshlBfL_%jC#(EqhpV$q
zRzXSwlF=B#>Apu2gcwRG2so?5UH<J&_i!4Ui?z7Hk~I0D7lOeg*Yrs~-lny*ISMYX
zPkV9lW$`QPK%1^2ZJi+uX|zPJ5N9l_nr%*rXy`A6Iqnb42bY%|{0c$W)X^7ksbeXL
zdBHdTfD%ivI7Ff&!Z7V|N*WC*%l4t0Nshd?gDSr&TMCmlT!E!e@X+%1X+<uZF-F|t
zAQ#>=jQYn>&&+#-fdfxd2|mW6gq7mE3qAwbVoe2Xhn-tt{y10Z({jXU_UtyUV;b;Q
z46L(8Rg9^&&1+WEm2rUA>!!<Oec6PGo5PrTdB+}63IE}1%Sd)wc6}4=#jx>@m+KcB
zRxUca4U<yp5Kg(A)lMB|f-6o)2ePGeW9NX`+WOA<L6K|SbJY_?gAH`OnrXu(x-Z(#
z9tHVjAU!7*E<H2SOIs%jAg}m_?fheeEr~aqpef!K|1`c_t>WQVo8vY!D#aRGo=%(b
zMLG)YHV#6_4!}QRHekN5VG#dqh~gmb#Z#YIk(0kd7XSSV`Zw=0C-M*Z&+#VjCDrE|
z`Hx@ZfK#OP&!2~1>-W+B=RWi~(!0{2Jfv#;dgRTzFmCh1)qXqr$GeMo>+}@EK<J27
zRwt4t(xIDcujH{$8$6)S(0UqbRYZ`;NHlVHPx7LM2{rtZL^0uP2-Vaa!)rLCVCkrY
zB|qN^60luYj0r)jXjSLAukc7Q<`nx!YJKC75xKu$<C~U2HQkA=mu1cw{F*eDu)g=~
zPbQuGHh=aywFSS(#D-Uz9<SO!jHW1#lk4G~s1&q`g`)_HMK;52ySA77;_W<|1txAA
z#&^fVPPdsZwvO6e>kZe}3VoS>p3mQ>Bj69H4Ec0%=#=<a2ADBp=LCxZ9G@!wm5!gv
zI7|wnjQm10)Xt5gYaVL>#gyoR+5>8DF+3p)s?4f^u4y;e8&^RrlTg!eP7*eMAzlLh
z5Qq?3x_mP)WI5Lw^Saf%Tp~8FtspoRLIG{9PbPN5zMqdAk)ME>eHkM*Ue-JH_2x#Q
z9ZRAWrnq`!{`=`}0-gXP<BZ&HatzN@@yv~-uzB-PVnwXL?SbTDzUMJCEAlC_1V0-}
z&>}Z9^k$69tcJ)>;Z^K&5->@Q<wc=~=c(VT>}FBPkwxPV(o&a6qYj(q>&MO5)W({%
zbg17@1Z$I5Q&gR=f7QHIda>L{LwAN<bp83(<rpE|I=hk`*H5&(x1m{~$k~`NZl&SH
zz`-y>1@$arO9D=)EJqM~X%w3*N^k`fQpa(AKk1_Qk6dIy%Bb0r?9mPvt*TsG;g*w7
zFKkqNiU7fsPuknJqCo0i3iV6LL>l9eZi$y9-+HQ)To5+r0+GlrOcga;?kBPg5eCir
zzJY2cA<GVd>xO~g34~F)Q1agRi{LmC5rwE91e|Vu;aS;C=w^B41lE2q-QlzwYR&AK
zV+mU6n3NgDe#Rqvn)HEKA^17^U3sI*Sd#*<xI9E_w2OWm^YXjD&1qQz2a|eOdPeK`
zcpaZ;TTYnCjvhV45LkUjdAk<eIzw@0%=s`k{Qf581tc^rUQdu`HVn_bGLo1c(=65k
zK`XO3?pbce{kaSV<5PHuqfJsuN?U74?FV%J?7)sWFtEgb5Q%gGaP)1st4T!!wd|CL
z-X~*Hu0V=ZW}EZbiTlUX(Uq;ZmBk-Qn5>-0dlDll+o!D2#JasI+zR79o;Q;1k7@@^
zZHM!V#F^zK8`M~p%{b7akco=HjIG+e0oyeI4j)nP469`u7aR&v{eTn+xVGxE_9r!t
zZ^260@w_XIS%)yvRWUiebO)1OQ^03FOY?d?X&l1fM6UmmlRXV$)v^6lh#tj4D4GaR
z$l-MK*RMZPSSMhu>me>Bw=?kxHt7Ov%;%tn5Is?J7V|BJU})k%$~w6c&YlNvYFe?0
z+p*d<_LAd2p|>yu3H7pq2k`Gr0RwGPul@Qk4HR#lUHJUl9a(oeh0G~RpTVHh#<7cr
zV7*%V<|cO&GKK(H|ITeB#?;*G8oTImzGPK)1UC{p6tblKK&BNd<>@jIuRieHlZqp|
z@PbMzJr9SFqr^dydb_&z-D!*1QIi&w>#aK#hw0J%k(^Y)U+O)7^!EEvL51(cI)p5)
zGP!77&UtZ1o51Chn76N;bOICli?4|_@7O+By@kg2J_3%{wZ6<Dmkp)(DrMX()+fmW
zWN$=pTBc#hI=FM9xOZ+Ew7UB}W+eWg;%<z;$X}AXxL5YUg<UP8+FT4KPLK^-qSAbH
zMP;I0EbtLRzcb_o?IUF0qQ|yfIo?%LgmWc7{=jGqyHg_Mwdx6ZOoy;WUwwWSA4xSN
z0xtwDd}F7>zs;&3quQ0is7wDnS$@=I^WV(+z0%)?&gHWwGB0r{M@UXz*S1Wo3!0fZ
z#9JGcj~%TmqS#B{8j+5e;7xrN$@NlF9!0lKwC~-x!IpF%A?Ghz^BO+cw^3~pFlhLp
zjJRdhFl+wdZAW_LKRm;2x~#+$_O5Li-+dSJ-skV+xQcPeXrUxS4#X9R91oJ~HH^?^
zs(|x0smlFzU{7U(a4K_4*~L#6VGb@*fVr1hg==LR^u5=>SkqZ|Ysx$#svoOm%HDRD
zfi*meE+8j_<yB(acje+J8ejUnP5)E;#uB?PmiXJ2GDtMoJa7><7969+d!=an^-mqE
zC*S>sR@)8tt#>+7x(EdJ5p#aNH34Ew3+qB?H<^K{Wu8!t)E7Y&YEWZgK!(nH41-wb
zVFzxPaut^a{~c%++zyv)B|x-TsaMLrhtJ<vLMn#Dzojc$f?YQrkB*^TRy9Trk6(3h
zON~f2W-a_SdSZ`0xxKRw<RQ{S@2IvJ$$d?k(CR$}{4Q-qCc+2u<G(lf$PXL3C+P4<
zUNQ189WjF=d)Ouu&MfrYt}2b&`2|mx_Wm5C3pVs+%iEPAu)@^VnAix1gtrQ}xS%BT
z%ibgx^J5`sLU%OLUW+FT80&(ZtG5y2i9JLju8ooSwAG@0eh(xWs|K_^?z|Jgs%Xa`
zUYTtSQKH=()!-Fa0(-PZk1<!kbmK*a)S&XR5d+epJ=Hkn95&hSfj?2w3+!hVGO}s*
z(SJNMy55F0?lL&Hupe75InS8P)u3;8)|^A5?{BkM%m=dlXSUl1-4)IVG@D1^$|lN$
zq^kJk8RI;4&}m)31n`8tqfaxj4Hudp34-rg+-i%}2dJ8c1m8j#$ciT(d-fR-?ln*D
zvtk@jZvAb7RJF_*VjSM9)tkNi@>>Ab{D#s_KG4dqbbVorVoB;U<C21WSi1Vg>BV*R
z4?VqCD#QP2CAJ@Jj%#XZW$S)Dxan;PBW88DIJVC4+?V~ceT?cAyldGo8+HmQ-ax~8
zb+yV%$|e3h)5JZWAPsWr5Jt`le%`%njG6FQ>1@JG11g$IVvwNugaC&)`AH#;)IOxo
zKEX4gBtE56Y~EDQvv>HsGSz$<b3k~|E7mmf8ds(Qz!Ay%#&Ve4y_i=Tu(Px9fmfaL
zM){1cDsobB{N(r6QWc&uS*_~TjxAkUedv8A^r}b`?<EI*8*&@%Bgb8eH}rpVHx(5W
z;on$dqy=xH-pM<lJwYy`(HTc)Jb6WDqP=V^^kRZ&sz*y{6>@_Ayh<uAjs2l^$tIa7
znnVL#c56~goC4Uagi#vhv&Gqb1{I6JYlP~wN~)&lw5t3Eub*)22Mm}sa4&rFL6Jfl
zuA?jAjGTyB>EuQat%XCjj7iP;PWQy%gNgM}yWc6<vIOn*-&po*viYcN*V8f?>`VoM
zvbvo*=k1#vcE}zIl-c-bvmW*LJ^EXPD;evLyG;MMQV8q!CR(~e?EO2?(O4|lsP1=}
zz7ec9MN)0$GPQWLo4ZZX<m{H|$mw=35~;u--adRf6xQ;XOS=zDJYTJyznEgO>>f*A
z>vb+t>}Pgwa8N$eS%ze|5kyy7Mvs>U^54b(P>siOYc*otY=_xK!{P?}$P2e78tIW1
zS)(-+cG=WW!8)`Vj%0=Dr(VN2N>dkx?rS6Vx8Hzvq$03KVaOBNuVyqilwm&mM&u^T
zi^=0ot<7^OLi&o))BxIEF6oxfbN77Hb;_O9Wb+nLVDy9^_s^3|jUTmKw2H4-ayUQo
zcAKfqo9=jUol}UWu_f@*n>zhbdQLz?z4Ii&{1A-twBh)d%EkF+4@ezI6A*+doDY^U
zyb)cr3B?22&+e3NAIZjiF?Cjxw2O+NF>c1jBaq}8Z4Z|s$!~F84ntd37&5L&H7T=F
zMQ*_8t=l+hkQF~4XSR7b9X!^Rvl&?J?T;_v8_tC`)Hq5q{P1pCdbfSH>E2hpX``he
z2u*<&pOKNf03bTf0nK2#CVlfhJ(mbj!7(CZ#AZ!|7Zp$!J6zK6%)>92M|-|WF9*{^
zEBtU!t3jvH`z5qb>111KkjGzBA|`-rEDDcEj_Be}5}#}es}}IGk&BjKRauL?qa_~I
z(Ac@0E(|>-NI|Y3vW}K1I$1zU#=2G&ggU2>lgpoeALGRiTnp_RBsLf?;?_=%u8&^b
zB?p4+b2IT*ZnOmeKGmSN6@^cwcmI!RQlWJ;C*ie^-lTr1{{9HE^xL#EgR#Wp)2|Kn
zy%luw%<&bT$0c4smShTOuS)M^!WwtL#5%RSKk%M&#0}7m7p3uM3lVo}bAp@BrWk{A
z{?nA>e|8`pe7C-~wx{Wg_g^6NGzCjjdc|j6O=K>wdhe{BUapEDUtG4m7+iM}UK97!
z+q}|kn{6nMB9j8%fmB^GrMa1R(x}nlB6_-1Tnb-oIrHBV!?8OH%sQ0b`hl#oR>12#
z9adY|y-uls;Oe>(cI2$h{)W!{@AHsOV!IZ9--7*Nsh|t92WcO>_bMxSrpu6I+E=pG
z6#=Dk#)=7Pr4AMR3liL&*co3B;Ex~B#6ceupl&?1Sg^Ps)6Ypfm51k+#GO92&2S>m
zc>VULbY3>+lmU9<%0!b&Q%*3|P+*v2c<z%X#8no5YbxR>WX_SpbJ3y>>KRjC2_<C~
zrzv9k-g)behD4CUvLetuGjYQI1_TThN-0Cp+GF0fa<u)?p9^J8eJn?ECg$m#ood2p
zX2*vP%}!WzYBOzWCOotTdGZNA$eLgTz!JhbH=OeM&Um_CLZnMlp;h5#m(^x5N~`cv
zz$i0|ILk}C*&+7x;hVDShwfB=+SoPAB2I7wBYpF*@7a>6I7#MhqIA;pvp*u_4^=(0
zwO*-C66pDlTmi=(7VCDRq8J!e<@M@%r>j~4Waqla!Y>ZDY|NQhAXx-eLfSyrgQ(rK
zAGItHXZ3DvYGm<+8Fz-T`l`squkOWfTiDt+q-s-H?33|Qe2%{@XrYQ+kI%x7b=pr(
zWA57!o3}+jmiwENOpnX!T1c%@7O{v_i$%vY@E{)~aSf-F{ZAVIpWGPK$agdj@}KB@
z&?+{;1ZH7;x5~{j|ApgPRu6|Nzh6x(-xHS$mg%x+{$ExR*NDu&Qwot<N_^9-)+Z#W
zXnBk(NU}a6W`es|@TfdW8XJ_>pfTOQXMkeO``GMD-;$GnQp?Z5_H-_7S-d!I6u;N6
zj0rR$j)NjjK}pSefo~b6OyJqq3MB@V_ZA^wL&fB-Oo%Ky0jYXIa7(-j&O8cQC<;id
zCHhTkVoj=&4YuvN;4}i1DafW_yw!P_{@ltUEgFiKbJ3tyhh`aIhvTl*>xtWWrGG1*
zJ@utG5U}M)Kz>u}W>fwCd$$H?iuOJH0R@wQI5_M~n!tn^e&Nu56a*5e=s7hZ4~5qb
ze!)O?Pw|tp>-GKpWsySvVY@Ues|#h<9Ayzf*#Efz7U`X;j?U73_&0wR-&c&Zurtfz
zrN<zfyo{1cB3og4y`J)O{1V!>TDql*Y-(_>o+gLVd0>BGa&VFo7T4?2+|cx!cp3Oo
zYVlzF;dd!vmxtoN&^~=KU%yOwz*#(IrKY(++*CVYIyv7-nX=e^ghhJu>SV018tOzd
zQll|x+I2LjdY{|9bs~6)SO)S@MIB8i28=V!%98h%rd!#uS%?^3Gm7&rNC@>>4FRlw
zu`Ti@?p&Gcz0acUB_<IO(P@C*>D}vYku$o>^rizB*Oe4A*P~I5vpzH7R@UnUR`37q
z94>w;0_33#(RngysXcoeQXh$Tze5rT9Ln}#lk1hv7v;PmtEUww$iKtWSbEj&@2$UF
z>Iu8uR#Z9tlP<2SWx3b!4*TV48R{wtD*I<N%cnt{s+us|zqSmo{yy^2j@W1=l&9#3
z`C|!g=#A{`dbsrg^LY8?SgsK6R#8|TpZcr&i?FsY!AhK)P7!7T8uvvyG}%-;N1v{c
z)PC3>+*B9KmN{}Bo8~jMjXT~A%U4Q#U0}K%-fqBQ_vD_pOzLNPNG2DpPKD6XD<F}=
z>e?YNsla_lNUOoNc%^x9-zJj}cm2Rk^+_;(%WH{O1RQ_LVrXRz7;WdIyZ6-eYxQ4N
zLCv*k4$Z@+<w7s=$SoAyXF5qSl<Wz-Dm154qrf6asg(=BjcwLQ{|Z-HputcoCynq{
z<zw=o^pJi^T7kyyfB#G^xkT_>@~%R}-5AZ-{!#<fGS)?RgAX|8>3h$m)q|7s#7<%y
zE!sqOsfdB2mynD#)a`AFWE$eq&&S7)z*V0o=xrB*;qvbpsSUUE=eKguYhoF>*Am=N
z8bsE<7TPu)>GA@(*cac9J;mFKt&=>zrLc2{yJGh1dE7l%SAIvW)fk$PNp$aFDuK_$
zdATl@`t#dR{CaBvP*%Asy(dGM7fLv{^koc$Z9j*-6-y8pO32AlCfwl1RnseTPGej7
zBqGXc%9Zz^rNL*=#m%dzSAO$R>YYr^5uSVteyn@Ko7=de$s8JNvItx*z3x=d1pt)|
z>Yjxie>3e0XNxlofRN)q!zu#O5>mcfqiN>4lQ}r7p#`Rx&@&^nzs^Uci^Gm(NT8V)
zufkx)Ez{m(&zMAISqQiKfRb6wq{C@XiFoQ0O7XZMgv?8{x7lWD#%R6D?8YH%vTrek
z?%5zneg^i0SR{_9YJxUV!{vMnQ)nl%w?erpB$i;DIO4728oQIYlVo-*_g`QCUJoI6
zo3`#$h;_HBL@{cJ<xK7|OJ=zrKdC!DpAbPI&T5!zSoX4_+oUb@xKE8jJ<%LP$cFsB
z*@4=Oer_rueoa<qFIjqb?f6Pa0*4Y#n3OUhzNkEL#9Z@(iJfZZcn-)_(tS}JtS)K#
zm9AX7s7h3bbugppe4qB?qnlS|U&7NohrHkv(z|IN`*XT{iH0AZfePGol151~EybGA
zRkY;|L+kGc@Y<(6XvY}Gz6U~x^1vRi5q<-U7@9l>+Pm6|xkp&opzU-k3Zv`*3@*K~
zbCHQ=n9rnQbHA^fQN-aN#=oszOSoz%MO0VI642dr=Ko#U&Qj;dE^!9x>zX-B>G~tk
zFgKAt8;3%`Fc4ts$0882zJ;Nz-Fg?5dW!+Xx^uq@R4=EHSLC(VC3M1+f$mMj_D!n}
zfWM1%8&;Yg)M*@_ZKMN&xwIg5i&gG4M4AY?9X*7SUQBDO0>|@jBf%OH&H?<0fU2(D
zH#3X9yx5wSQJO4f$xlN+&5yn23pJ5l7+KACGs<aA+U434V~<E>Ooq#TX_L%4&yjRR
z{`N(M7IJS`>w$>e&Cmy12`IR$%X#&ef_@!PTb*^XC@lu~>Z^2_7e9);;b)3|JnJLC
zq)28RcDz0Fd&Azlv>8qYn0{fUBe{V1<eg}4O+%kKRU~QJ7c>5~I=K2Fyb}i2r?%y4
zRkNI!9>}AuXM#ROlCzj8>5u2+;vO$~N{J`rx{z`z==f_gl}iYjk$;s*f3Bc962I8N
zD*F=Tm{*N1f}<pSw?HuFIC$b}T+kITahrjf?=W4{6fOvtau>(E!LtU{qC>@5Ro$+4
z(vKHcXn)sqH0N_3!&9nGr1x&*?t#}w{zlD5Sds+&T~S>R(aV_)-a?-HX0rGG=rd3|
zQ`%LqFO2I=8`K!$(y>2~e4rC|2J_fiKkvH~Mq@U$l-@}p^u!LJd}|q})}PW{M<OM`
zv&yiaM4@L~7Q38<odtx)=Ubjl%hJ?%g(BUz1+~a1715fsQ*PhK-02FFc%zDh#@XOl
z|8g3IcNKTziWwZlPTX1Qgp#s|LCXSx;16;!ptG6e95jh0!;h0w#VptMv~73xHeMwG
zpYYcZS2GdPC5(UU#`1E(&jLl^+-`>DJ$8%w%^JJk*Q4%0VA7k3Y~dQ&2d&8;BdEx>
z+3nLf5=Pj#B0f-NGos%WHzlV%Ous*l4LB$((fML%uA%vc@pOqw6}Krm>8nk{xV=B7
zK^i=@81?tX6C#(N7$iqW8<J5c?V0=8@g1)5Q9t?n%PH}}fh1$%Przg#UCM${`l+Qa
zg0?}IplsBE^n3`B{v3Wyn0@FTif9|Hb-Id%<G00^3&86{PHA<^DH2{2N88Yj@SM9y
z6`4{zW=)~P_#)2{z|UfC>2Nxyg7$JrA<J)&R!}~&YPaQ>ow_`fizhDP{^Cisa?c>i
zIiQ5l&pNOC`0t85bxii+{X6X?9nlpdZqRW*TVECt4$`V|6b_}BiM9K8!tmPwkn6V@
zH()g5(CNXEWYA7$4-$v5v-}OAy*31Y8re2CW4}Fwc;g8AKhzytUxjOAW+OnOuHV3&
zXR%rT-CM?td=E}OHZ##{+tArFapJrfV^UdPd2JyUmiuvw#HHM5Ey~IY<+Rf=pGNpT
zG}8<aDPHF{9sd^cvVqe-M|pH^gYqknb^*ovXjA*m%nrK_f$KL=UKDU1q2=(be5j#z
z(SG867H|WJ^@!2*o}~yjovmPHHvM@ls|pn5@}HUaajKfJve%PcQKtvk{rhJ{il9q^
z$MRVORc;zXHm+|n=!qN#nz8XSgw&V&w<#LzyTs-1vRGq(!%WfOMM?M^%bnbla+yhy
zvO27k3Ah5NkLxz~mn=fy;B%z&x^75BY?rp`A{X(^EI=svZk#HIsgIBYpuEF|=*fOu
zcsk|P!#-X-0pS>@B)+z4m-czt<R%$Y2F{)HQkll4@6&#23wxt@CyEau7l(|gLgS2L
zzEOt-zK)S}wRmbtzM2qm6&z7ze)+19r;4hN0>qQIn|L*cQ1LUC=IQ6UM>+B>CKm5i
zCF1K5iN=$6{%}<Z&t;0IORn7IYIo1~<I7<n;9zV;6X6b3MWDspdBxUb$=UggN%ker
zGf4Wt`mf9oHRjG)#34RiFff<*whfwX>&sfT>+T4<<#n9T+u;@I@9m9nH@cQ|tLgIX
z!W9NncYWDt>loslhw!z~{r@oU0<p9B?`XY(J`U|HsO*M{(04x6+^L~{z$Pn6;`9?w
zpv4x$z^5qz(Y?3&)+ct=%wpfZQF}l+M0GxPc3mL6KDn4-|FAqS5Wh`ARt}!=UQ+Y&
zJ`Ed6_|8yWRa)*ix<Pc`<c(xPyoD{#eDl3Ob1o$LLKzC9eEqkvg^iv6G-H#Wk|-oG
zLkS&Y_aXUB;V#xWTs*2(&N4L7s&a`+5A(OMC4PI6)18L!F)2VRbK&D|dXC$bnq2%s
z39AxMJ<@nRynKSmLQVUJ)yYLn$^Fek?Y{r14c`pl>bM2fB%y7F{5SKpm8w~h5N#%|
z=pWU)_>zeMV{D-{Zfu5ILHGojJ*Jb(E7v%RxJyL53WMAsF?4@j;H{~8H!h*DdcMh-
z&2TP=K5nWLl{gFldv8yoqqO&RC3^@G&AbfG0u$$|I;_b@K?8hC#zMXAcMbBU*1;X5
zfD<BRCtwT9dXz?WNsNp`7es>*7_HyFRXa*lQxVCUXa1r%9&I7MJlC;~ALlJuDA!pH
z6f-ZNpq+)w4gKOHcb~EfKXt_JpTCF8T3Wf6oebLpvtkFPDofvztC35y+RWTn!N%t0
zY2l{peIDpg?sH0XA6Qay>|10a%n(pnNouV6`2zk1jCTD=wmhe+$geSZ%H<d!eRCmK
zW>YUu3@mxOq>ut6cx?QBk<USj$9ec=7WexkR=t&$fCpXWeapQ`P1~7Xb01I}KL(xN
zDcW73)}7uy_^LNc8;Ym&tK!r^Y!c1Yuu@@Qu-lxPt&(&9Os}#0XJ(D^s)UeqYKw;L
zqy4i@ObRV}js)J*KUyE9lX?iACcqbs_Y!`@`G|fNkoY;&wvfR=RiAHp^CABg6UZ<N
ztCY;h_uDKqe{X0%F+CuK?Pg=<eQOr7>p_?Hy_?yDZi9z)>McGUJFD$68(|8*s;12j
zHiPEF0~#{P4+tVA9b9driq`|5dZb-G3NLp%DGgHVHyYns_)nWzyVIpudNbqt^ZbDr
zdu7<7@qPu~i}Yd&N3y7%2Sz-Tx3luP@Z7|6baASyTv8c28k0|D@t8HXj1W^=juNpg
zh7+leS8mVz*&cGF9_SCQPT4M%pMgHzkOXlva<;T+Y`H3>_OV^JM`Rsn7eZ03F;Te;
zG938RId+t0m>+(eKQ@hYd4)p*0y6u2hF9aNZgG$C6t8(5r76p}##6;1-Cg8-N)dvB
zx)X*a(lYm_))`_E20n8ydsCgHZX0}Nz|+>jo{X-ZuK#h_<O@~RDGb-PB6g|SxsRmz
zJ%4Y!c-Xvao%^<jFL<3_jM`p;j6uuL@H!pw_+<Ygs4>`4U@a`+o4-GQmhkJ^PZCZY
zV$y4p>;Y&^ZxU%@;}=6ly^+hjtlQ-i4_}ZpxMd^e;gk3K5R}FEx|Z5pqEJjq6>8{!
z7Hg~>gBZ51R7R)^b5~NJTSS-?J^l7+@o~$TvFLV;u|<_=_p7&9Uhg~%O&$fXmmzX`
zeKUafP`S;(K-fF?%PA5L6gbTKTW>TNQT;|o^|wAdrQa6`X5%to1OKcjikF5$U!kU7
zoHLm=@o@6XSK>)`wSNynEs&KeeEhin?3^@VX<2q&K3z8BlH<mKrN6yj*Y-5y_m9K)
z3#xW8sZ0fHu$qJucfjSe$YdhyWpK{-5d<byRR9IxxcL%(cMgDHE?%rP{#$hU4j^1(
z|2s2ze}%JfF8W2db1NENDU^{p2doUPE_i+@#owp2p@5yMjSl$>*kuqf8LI^p_=Z@c
zW7!T3e_8Gex&D){ANp}6X-uHLW@HgR-x8i?5WX%+K_#TE%@G^zfSsgD^(e`pBTD9s
z8sW^;I!OGJnP7b|uI5Wx9@m|ClDsD1;=#hfV3v0PuA|zV_Zn5TuoIjb^m)&AqCsRc
zZPRh4cNf`n%|w1azE8SeA<1Ux<l_~iY;5r2ZavR=Yw?K(uFpRyKaHJt490EKrNr%9
zC!XLh$u2<8-_V5SUNWi6eGT5zEL2lhrvk)GatV9|?_dyIsje4lLu(^5hnk0$bRIN@
zC}v_eOUve&rO16at5pU2i`dA6HHQG!8FP^OJEcW;CiR}G+(?Y9baFDrtIMBJhdwFY
z!RF^oUlqCf&0%RBwK6Ck^YTXU4G}{in>NulYv89gqW9c)N^dCbeTOt@jOW+Qk&=}o
zpzE~Eyop9rhbgVYK%sj$h4oPYbC{{^8rouz^?hQS83J4uTbq;n3;vyny?x%|i{oU_
zSVr8k6iZWlARSlZl8A$hgj{;xT=qlQmb1FwK<y>x(}&yI5mcM_iMybe0v+rzJ&YS0
z{Nm=Me!-z7s#QlPgh97#5B#4k{-+)jjhph}EqWlW(<91XeR~5ISVAC9A0kBsDU3s4
zC^1bRfkfaUZ|P@gZ$wL@V%gZW$jp>U1OI}C?w?e)H-toCr1gCBBJNfptG15}$y##V
zirCv8P(!DbH-S+h#}Dk?l3Q=+NiGNPR9sGkRRH$c6vn(~P>3x)a@bIay=l3Z<BEs%
zh~lHbejd-yUM~;Qc9%jfMN(H^mSIP+DK@bf6E01$ScyP0pBB1EyghzgKK{LfgyP5&
z2G6rsVM)Ay0K<MzaThXAr5yYESE3P$UFO-EqPJZ?+U<9|OZxE+wJqe653@$a66Zrw
zb%Uj#H<M)FME*u2zB#KaTyX_J{zR@kq-|lLQrr=t4L4d+Yf+}@sS>5m4Zlow%69gK
zRjfjl-xp%cY?64Z?sb}J48Y5(7UL?pxB#4d0#9N%`Sx4U)@9SA^YS>-k<py$3tT1;
zP}BQmtx#bWlaAa2H|2@E;UGIYbv9qts)Tv0f1zynC5mc-W`U`zd6_>kdY;oPChg%D
zbFn7I{q+ORK7zi*_m@vR^O?~ijAF$|Xdp`fUjp)nl*Y<2u3UJ1Np%3V8zx1qZyYr#
zFpT3>AEUiBQCmp9`CM;yZ9hy;r}G!VfYPcm0eicrQ*K)Yon?L1#QZzQHkd%Gvi8nj
z1-TE*olQR7&`kHs==77PZgOebwI3wo#Ci$N#O<;dmL&R{96;C#hGP7?6e*5-O#K16
zgL%}m`ErfJ5IB;)*|`;??}A6-FHHA=a1by`(%`uWzlu)FisK{%X3KjXsZ*VgY^Kz@
zQ(aTj4JE%2t58#OX8mt#c=-Q$o<QNV`SB6hWkVQyWy(QnH{VzwXpc_$OdXmfqY>;d
zDhuC0{0m|+K0a<yt&PZ^U5h~jihKJ~QK&?ram^-S^3Q~fNkL$8qh4bRU-!6$>i2#<
zts{Rx@t`l{A!~7^EUgxgP7)hoz^2C?HOYk!0eCQxV+FB%wLLujz5dYWdfpAjMF;_}
z+s4hT8`du~yxD#Wv?5t&_?{;kU&omxxm~~e2#9H)TH*@Z+#<}-&7)Vh_E4&IdpDLi
zeTl6$3!ir>l6$98Q*ItR1l~6h>KU(_Cc$k(J3+3dur{V6G7)GlSe<VsH=oGjndLh&
zd*e^f;a)4+xf|u}2flAaHB_v7UB-~i*ss!jJ|BE-tf>FRkcJ>1E#gB|NfAm&u-%!>
zkqabWXgSItG1<A<8D#LIP3R{&PTKnH)pSWHpJX{fWnL;xil;HL9MRp)2IMUE&D^}E
zM9mtBHNo@@tnEkdiN~LLE+`}%tprbq1^VXhX;5g-!RXR2)-9%r0MQz7Sa9$SzToMS
zl!=7ztn<%`#4YD%=+3vP?T8os^TW0sIh_7QnJ~V;J=W14e*K}PC8~>hx*LV#E||F0
z5rFfqG#jl+8q;N7<$Th6@0;r|cyPLr&I8s|iD}bZi;+jDaP(l$1E+|Xl$DEABZ);k
za{`;3n{`pE4YKu@iIQ5m0<og~Jvm=5k0O5KrA}imG;#Hi;dnzSBgXv2z?QDDhwjy)
zV$q-3hR#RJ4a<L%+>34rNo~hY%QU3TH2Au@e{P!8Uu7O!i0|Tv#IGr>r5;HW+$|%X
zm(;hw3q{1R6R`9Vs1U64(PWvEg2M~elZ1Y(x3La|8AfXX5QhD!8?PkB{BL=!LVSyB
zi;nQKxW#pR^>-c=N;r5@1*a9>!G(-z={*Ni>x>WLDujaZe%er=eklwmlUR7xt{f8Q
z&j!}2Zii9BrOvJ6HcKbJ<c5r1x`AKaW?Z6=(G0VC!;N6g8b3jGN>^S&i}dW!ZR>WA
z<Q(A#|J`q^C^xH`z8of~T{y0pCILXrZT^^wF0b5hcuR8YSTN%?3+G=k;fpch2PO{`
z)%nP)t(+ghVPokdJUw@p7F5E??nYN3P-P-fD<@68^FaV*K2n!?((3!8|G60xwIgtU
zVBPDI**ekL$wVG>IRrFaZ-sccBYo((-Xlf+3BC}yIxIt5`jSn*aD!#ltY%z%!rt>L
zfw$||Ehjglx*%p*tExw)=wkX81)SWal;I?2R82lP?cxstj4(&vb$%d#uFX3znX5$-
zk54WsMsO+qZ9x%{LopFcr2TY2pp(G&Omz-8yZNIMiu0wNcRA1UQn^X^T66@oxkz1a
zK^VN1GkJzq!HxT%6#o-Fp6&J_*b$e3G9Pu1Oh|2C-WIOY3Ci?#5BNH}J%_?xAsIKu
z275tSR}zQ}7X94^J|*Mg-o5b349`>UeYnK&@GSgX@km&Dq-<pnoH8J}2;2S7`-il!
z#yb$|2@>@Z5wGkM6#wWK+ZF6PxnS`SiYm@>aM$|1Sf}Jlp1&YXU0lP7_J&2fwoBkw
z!s48UBwj~CLA{$*sH+I!i^F%k$3cd^$)iKG{!wWKUdTRj>8}mk`<l<lhe%H2t#LV?
zp)Z_#3m4@IxeC!fkg;_%W$g5vpe`zMntl({!16T3ziDC{PG~uIuIiXqPPSkfAchjQ
z8|x>x8IBa1#<9D?DhJ%&R3B9)HU+|4cQ2<1=U^N6*$o<qVgKb3$$8rPMTAjtq~RCk
z=(t1tD#J_@HO@(J<eunwhivSy{q189Lv)S)hyG>v6Gh0+>C2fp6zJ|HB5?HrMAPh?
zU7Br>S&5vmUbZU}mTnzad+^apanaFecVZg4mp;!VQb?F?1UE%+46#aYYFgnWfz=~z
zRr{9UO%IrKDatYU#c32pV`+Lwy$o%_NL&s579`nK;_+UjE@{1LMGF{sooKw1*?$AH
z)P(E*(Vch{Jqqn_qymv_rRoO)5_r~~5D$zo9ZY{qJb`w?pW_6X6l}<z{-#tj{;6Ma
z*0<RMSA&KCX4xs;Cpe~)=}^CD(!fotB%uWon>7zH)YC!35BW;JYDP(7@<QyCP6U0~
zTE^4|`a={(Q^ZzY(^Zc<!cfjpD`v}lm{yrK=qexlZNJ^&9-Yi-&-RUGrjR=3b|51d
z_>h0*L2P>d1>Zfcgb5NYL8hI@@<kz#8CPEa3wlIjSOL>#+0cTbUlsQ@t%G6!-6K#M
zb9`Of$3*rhRUB8TVVY>mh>jrSz7omeIT|pO*fYG6-D79dMVL|mHZA4!;02Z%S~804
zH*-bAAU(TjF^qI)H#qj&1nRjD*Q;-xAPHeou6R>*s^`73=iwgZL;vu(q{{Ev*?irv
zO`A<WCapJ)&ZTYcUaq{lv1SD7N=OmA1iUCtoLCotQ>|M+GXG00QENaSa!L1h-WR03
zj?=1rU(!@I`b5r<Lc8E^!bi*<a&+MZHDNzQZgi;nCg_BO-Kx9=b!WHH?`==4PqF_S
z#FSL)+3tw`K=!g1$V*O83T5Db8+s=>XePRJAXNUwG>3bkM>mr$vG;KLK5wvQ!5@sP
z;Cu}xfQ5$~Zv&t3Hg=(%0d$aY++{1LuP}e9blKx{N|Ze(Em6BW?gGomcd4njq2Wyj
zj&84LQ<S8wrgFNUqtoL-7?CGv^*K?P*p-q~mfms)2?ssN<WfX4is|?&gZD)5vSI2s
zWi|sId{8M2-Y!MIKakyE>!7eVTvwwr$jZy}yg7l~hhkuO@X$%pZ^`B2zRMb#((}=m
zi>>C_cAf2NWlE}8s@OQ@YesFrz}oQpWi1hWU7?6Qc~+~x$V|P?VZv98zJkDv2_JL`
zx;%W+LZ6p5`WC}$46CyR@#`J<akNj8|0Su+<D2E3PG$;=Pw@CZz82T=QTv!@`-(Dm
zeE%3T;&~TCxf8ka)pTh@gR(+ot&+@BOXc$Ki(Pn~QD2mPFT}R_k3P-Q_U)i*2}@$)
zaWskZYeZUpoJLEKS(JGC0FUc};?9d=VL}>{qV2MYcPP50hGoh`nxAW_E2c@IZ=!1u
z8#1lx3vg+DiHr+F?<XnYcVrBXO3I2wBR;b)aR}swi*=%{*WhaE%*W*7nVC9(rEsys
zzxlgXB|pjipuZ6Ly(Lv0r@qmJDf^#J49Qv9+<OhqYF)AJF+QP$+*oc9>5-M-vKBiJ
zs)qng1Y7{=T`Ewzb<F(tIliV|bTY&|(i4FI(H{H<=)51r2}oPmmIvl>G!q$;AA-sx
z`Kbw;+Op2-neHs1YF8E#60#DxyFQ}V=26JIAP_u%p78vM<h#g;ui)gTNMm*Qv8_*v
z?t(2X6CCJNxB0hMZ}RgdN?OJMSVIbagJAHYjUiqkA&FNYEv={gO_jaaPa8cjl(&rD
z<9#Sz&(^UQmAdcmF6>uO`2A57cx)`V<p-eWNb@p?4D-e&rJh3fLJhj5Eiax3dzG>{
zWj_iU=lugKT|Lc`w2Ag-C32I#Yhs_{OQD9ZA_SP9n0vV%Vp>$X9BkT5*S6<G3HRu;
zo&Nv#;qFmQm5BbDbZ0P&tS?NLr8_5OV9JXPplq+No>{rB5tNi?!-l=qAYMVD$jy84
zNQY<Ah^U7LgqA2V&vzjEFMorFQiQV|SdXTIF$&9m-2ND3MUgreWa7Z5gH=e~aZXl3
zWfRS&!J%3BJN(i=V(dO$@Rwx;(DzUZinLZ<2{-v%MdPc2DX8_Av1u47xk#S5rHl+f
z(T>?Raop;Oj8&&|USU#nvM;Gnb{?Eq3f9EwmZ0go7|m6TQot}_!5`;X;RYJ8U38Y6
z!LaD}aQ?V7-`cqN|DG65Fd83!!!Axjw{<+#{irB>v2%&Tkg#FsAwErkzU@QCsHaI$
zl3ly`pdkQN8Q_~Qlb?oIu}da4<(o$4bXs5tVg*smAXcbsEf5ScpX2;Qy^}&{Z9`dG
z(+Z%4#8PoyPOsokTro#bDZb&`eS4*BN}3z<u;;afpP7={Z;L2Cl?vJD(}Mh@vxsq*
z0u%1?yn*x%9Ffq8?OGJs@Z170gwBp^7QANk+4<>qJtgk=CYi|-Svqo(BP^#lBE4Se
zX(VayP<%w@?yKeWJ@TpX_mfH<VTX;~lbcxI(&V4Y#xQu*Op_Z}a5%POq%u00l_S`(
zsNdDy?oGU{>h{4{+!x*mFc|7vaIBi5)E(Oq*bNd-WPU~yP5U-_yykH$AbcRjJ<+!$
z4jFv32bDeA>3eOmMcIr!WkvwIxQhBBLYSK#%Z3=;$H%>=)3fnK#_C9jqp**Yu1Kx?
zUXsf>sHu{fO!!oP)J%{p`a>se`Z3${#OLY=$UdSLPkWEwmiBQ#rQA97Nv7P`9Ox&t
z20j_DPyOCKSkOZ4ResH^Ic1JPH!H;O1Dl*c-LOjEfzNJ&5=G`^PyWDI4V$)p#iHQ`
zM}vLacE^9Js`L~0USt-tX|qAbDUE#Fd^nw3NhnH5*^8WA1B({dmwo17N-XlYg8*N9
zZ`wx&w*7YW8smc7+0jF;c3EAppd{_Sb2H5CQyK?1ns|q|KUv4?K;KgOtt0YK(g?uK
z3h9{>I6?kQekf?AYRudr+xKd6S${5uliQe|MiLJ}qjA|7b|9&yY4~+_P8!Vx`;lB#
zN4GnXY7FFR(HiNoRhos-qTb1bu6Dp)h8*D)xdR?aN!sP6)1iyq^{Y!!krT?U#G@3X
z=RF*$sLbp2%nr7h>+*=j<lZFKgTs**KG8v^%C80pI{eNY0*kY#<o}+(I&lr<h(pc=
z;=;ftSoQkmO2^U>$ccxmEu+a<W)|%%NpBW@q`lt`At|2sWQH&Uwo7|0+E>6I<CvYe
z?Blh^;zj2V4HnGQOnBY|5GUJZII+@V73p*J2MhI162#xPhS40K3!{1k9sh=XF;Y5_
zF*&tmIh{hUZaOb<gkQZkhe3Z`EZ(b^^d&H4l#vjRjqe--R54?)1m_qjDhw8PRgLBX
z`Un(!>6gzla2V+d^d~Y>d*)HtE%8z90CD?skBN{>Gm*rV>f_TOdRb+Y5H=-t?sRx~
z`#&Hf{N~HDK69f(*zt1c&)jLSG`9^JbELLlr=_R1@lSaK*o%QAu#SDfZY>6BZaDg>
zXV#HZIu(6uQdF<F^Hw%O652d<6*&I}uxZEGxB81o8RS|tS#FcHqXFq?6K$7mgHOi*
zd-cR-U38v#s2EEQM*=6<_k*qJFUOl4AK}#IjwzcEARyjlSgp9)S)D3!Q8vfMZc{(l
z<V*Oa$L}j8MqB!{;>phWN}rcLFDbvPNeXAgHcwhE&LAG_m|HW86>>9QY5SDO#sX&y
zA5-;*!Nf-mv#fo@B<@sjT*%4YsaZ?)L6elnSn;~*?V%1>?%-Ab@)cxJ9l)0dWY^|L
zA>GN1G7zL0*T@jGDs`IJ&Xq^^^wH3Yi!4QK3(GK61YkTc2|Iun4}{Q6I#!o+>I?37
zrE;jmzh@5V7GXdw2+GKlDGw)>Kq>1Y#aMt9v|Y`((XK&sbj;1OrcG{i0l2%_I{U8f
z0&?Qj17MrzWTCx-pZsn-a|At+X(}{RZ`w(G%lx}!fhHe!26_wyo%!ntL5OH-I@?nR
z)Byy8kF0cLKI~L=PVpwS9vM}6uml#`{qy{=^LsUH{2|yJ#f9IM_oS=~zNOv^so)+<
zHG;qExl<DD@M<+Bj?lq0@MtWUO(~#5Tjjry;?f}(KiS|)HuM$D2vt>0bkc~9=0znV
ziGi1pb1<6rbE1H`$z%CF!RhHb<Ha5OJmGq^jwlJ&Q+LNM@6P4pdDn8$mt|ZZPM&EZ
zKlM<{|6Njxw7j?@P&ybz`&KMg=2yAc5BKcE)(1}jvI!{Y1~ehUol0EFa<4}U9(ZZh
zwnS6f{6z_dI-N{94S7&>*fx#oH%9-AHqoti?3!2?sV1w&{`DUby01o39z)HWNQD(^
zjr`Kh|1~^qA+NVkJpw6PHu!E8jWP$)#^o^p3q*%Yh6}TcOfX4wF4zc)9AVmvgEnT;
z2))Zc0@~I=HL*Cs!euYBQ+w?k7<)}S=liX3NornpXEAX!?VV%EroBv=^LTFhm6c5v
zn>C@DO=FJo>eyMJ6e<3|#CxrUE6Jr6E?{WIxonqSfd{!+!{!Hx`|aF!Hnh;kA{8U?
zus)aR#`+OgWCA9n^XvB_js`2*JU&Kx#Ztbq<|b0T-zF$>m|6%x!*D;_zmZM>k6&D&
z5P-4_UU+@Yaw~2@^N2RJv6iZ_1Mv6^;^bl6D=HmxU|F`A&DBT8JxQ1r{`!ncJ^>>!
z!=TKHU8mY*>37VSO+M52P~GHcUiHKxp3$k3KtN|5ZsQt0`%7)5^YeibFo!nDIPGYa
zX$+_Fnv1_B4kDb511EO)01&-uY~9tdjh3^6U|jpsE~{>YS$zW|nL8EhEFrZsw7Fq`
z$;VFx2jVP1dH3i*Z(4ax#RLrx@tL6Zg!=lU+D|-G$C*q07|NSNI$H>D12YTo4Fq)i
z9JC~7Jh_jz`-ZG3j_xudVTe#^tO5~|jIKM4FI5zjB=@_v$JC=Q!ta(rn$rF*23+n6
z?5*=%W<NIfJ6;ist3pjUSgW}V((FuZo8b}p6hFkr8@Jk+LM|jN@oM{++)14(T(ve5
z2~4$RVX>MOX!SE|ZA1BFtLsP1ia4b-Xk;X7p9|-(CvZb(47n6Lzs{2g)p|DC`>KI9
zGVD3jar+ZRm_)rde)SYq?Wq(5wIG|GepU4Ra-EQZ&YMoqgC>lVUZB~>>OVwh{pu1!
z7}A+2zl0G1&fwB5?AOeb0DfOryvSH;ud~*znWV}u`Vv>Ma_L_-_3-?6V9Qy}Z7;f|
zv&kS&>ymaqXvR@m<<~^o|9cZN{zIUcHz-bd9V@CN(ap6{1{nqsCLn#b8fqIR#JWcT
z1WF&VHAQOMqEqKB=rIDn+a2D=p?B)(7vRJTdKAie`Q8^Awx|<rv(@>t*rOX&Tnefj
zrC(MV{3<lJtNvhl(arUzQ)z-7&1#=juQrT0SFlQLVKLv*^IFUADkH40i%-=55hF0p
z9fJY!>!ywp)Ia68!^mjf{Mj<JQ0})V=6=*u)p6(GE~x8DO676D9Tvr4qQ<OaRC|aT
z)PLE`G{>cQw&`>A^+4o0>fe)ztssrRSKAv`kt(P*&G4*(Cke3Hv7gaZ_<zgdxWVJ=
zK`__Z(y|+fCgNa_`n9z3EWOtQx{9tUG<9bmf<7KnCi=j1;l1k~*Ixx?sQQg*%dnI{
zD%Yu~zx9UI_m0Q8kmFfUKyIQfonYd<+>ZHp89#rsFP_8dhmH+-a9#bsPV6*8@;1+F
z_dAM-e7xGv@44G_IZ?H~oG>H^_N0xW%>UbAniL=x_#!RE9l2@qB9%rlMzi>J_*F1i
zIEK|E7-gOt&9UzXq!<?%s($}&OZs)#Mcn2Q###;al2d7NmenRGze-e|4-F?jO@>7@
zJtM4uk)8vbLmi@Ruu3F<UZS(y@JQ!7N{?=6)b|h&Wxg$m86zyB+IO>aKBje!$U?p|
zQy|EP28w-BI)QeG8n$5UwW2z(*4PpkF@%+Tw50U$?&rs{pvQ7EE|3^<KPjx+4J5}3
zpk=W9awBTLzF?hM`FYp=ac1eHQ$uv7r$x%DEqN@racJUCn{kr{p%yg)9swhV+P?n0
zUZEwCobJC{1&ja%Ays-w2}-cPUSkVuS&l>vTb}8v5zrm##Gvw}$~fokueb<4>F{o|
z*|mPN<TPVUE&=vIxCFpsN^5IK+GGE@TWLIp^B4d$SezijjJeyY65`|{0nr>SWc#UE
z3v+{G1-tRrT@DBFDYm@LK>FCh_sA7R^eVXtp5N=}8}XZ+w|5H|#87p(U@`444O^A=
zm-WSOlPT}}jm@Am@J%JzuFzB!Np{#l_0c+rFMv)Xj3l*($>rH?lY$})zx%2Q>yhGt
zqW*&2&Rot32b^v_ntpt;GI-xG5^A!$!5aA0#tFof(ozYsG;Hc<5R0-|19be*OVS{;
z>9Q9{T&w-u&}hN*!c`gk)BSDQ{twbmZIzKBIGdfXdj5k%19zq1ykLb6h;T_SLl%A1
z0nq#GcQ@m5v62hHw{aJM>{O6w!iT0FAZ$V~R2%p+qE#J^&=tXZIB%|t7{f;K+S`QG
zmyIU8J=8U4dEqA7(V)s0$_)jpqd@_cfz;)n>$rZ;ChMx-*w!9!J22bMEkCT3L5(?L
z{D%hwYmtkHsY(PIFn*xwjk^`1Kp+Ya8&e8`VM>r>KcMNV+rvHcUzX*6afqK>ilw>@
zJq4=y!d+j_#ft!db*VFIbh?T*on4kPx}h_{Te_AC(Y&`r(;oM797x8oMV|GfV@;Pu
zq5N{ZH>ravqHrc3g{)962bnRCWzdtts+-gau;`O@wAqWL<$-&gN}<hZ8FgT6!s-FR
zu#_dFI0@FjMue+sJ^9&Jyo_<Zn4jckUn2CAhC$G!P|0%19&T)i*6DMKY>jDt>fHHZ
zQ1=vf{P_GbVs)nKQ>Sdt&*mKVzx1a5QbtX?`!E*J<7x%+5v^6gQLv$XA@py1SKoc{
zJ+L-V7x#v}Rc@48wjEVS792{Xj8#&OB;O{8M3LnPsuvEQL#w1>gYT_@<mH8`0Bg8p
zhU#oPwh@$b__wM3gmkB?j8uLwIn_OhUB7@^|H+{+bRo#9a-^D9HO7qHvJF}f6>NV$
z#`cMmNXgtIyJ1KzAfUW`3x(a~(9w(cGyhwbgFLyXbZw(vyi0UGdz&}$Hmj^k2V*k;
zo5_x;o!xAm#H`ljWd1gk7-&X6S+R_$NSbPWY|lmH1CW9o$S9H8wmXgi*ztSEs4A4X
zdFwfcT8-mJ1DwpfXzbVTEEjSz|L2jQNM!Mb_5rMO3J|mM;aaZ7=DtW3h??xD+ExJx
z@gQNqXsV)+?eJP%@sGCDvec0RW5o?kOda*#=KOD_cZVLH!X?%S97gw^ruKKW4gR6(
z#<EeP3V=&EKub|Z88<`-30<kJ@5j1Y_nzQdKau6rm6>@d8XIi%e>{C<RFzHG?yDe(
z(v5V7G}0yAAZ!qjW>ZRcgVNod(xG&BcWk=55s>aW<9EJu*7Apo1-f}+=AJ8g)auwA
z{AWIWkgd%+TPkAk<lX=pGLBbRVi0M=iWfXYcgW2eNj%j3_zdUNq4PdJvKa7F-sA$<
z9GDS-@g4{$U((B@kgQIvSVpqpB&&SRgx?88i7sLQ-(s3y1)huE{?MDH8iW>GkW2ug
zUu_K;X!E1;otpDW?l2PT+(BkuA8cyw!0qAhv}X#IkR5g(sEeo|wXd{-RB)<FbWDiD
zYY3VG7*B2$iDurHEr0=1?$F5!n!bT*u2=mpk)3ftq`<fiT65Iz((jD9d}6fdBI_WX
z?@NMzNO(~Z*htl{-9V!euy29owwCxy&UZIOWLdt5*KDxl&NYo%&e9_kd_;P{C9ab`
z_=jF!tUQTX|My}vQyBJ7#(%aItK>DNcNwU1d`;oC@8-MO<kX(t^>1W0B}wTDRcUo9
z=usWj&oUgheF^mXIkHiloP3-5mF@l@T%_%+o5IxeE%@!=Qs5%@(+1|}1>jhKG^(#1
zW&X%@Y3^XtkF6lTQlS3RR(un&IK|h_u}EGWSmolW4EkJd>?I@ZA6};>W+0BLpCmd5
z%Q@Ibrk90kvrgvh{G&;;=rMr;K%-~@j0$@7D_doP72+Ke(vxxFwPLdvcdftzKaCH{
zK>bPww37yMzOs8Iv1$!_KwkE6f3{x(x=%b|Ssw1x%%7g(zCdSY`(F|Jy{9u~9R*G3
zWm+)SYnKOQ`)0l$Vctz4h6dm`CD<z>j!sQ6r2)|>F;1rU*xdydzVzzYt}K=)Rj*--
zjxBM)&|>LB2kii_pwfPMycN%q=1>2c$Bni^zk>cnab=%;Y4NYJF}ZQ<e8KScJ;N2`
zv<eV_fZFe^!&j`a>>_9LY;T97>ggSyOt_^U+`~l(Z3)}tceZ0Gqoc9lx`NdcuZ)yc
zy)sI=p?VscZ+3QB0|HYp^N|ip+~0mLod;nNMH~Xv8XFs;r*Jy6!a;X*oC^Pd*T>(j
zgq8da4axLx8poU)Pr<|slCHRel9`j_x8(t%&KyB%ANSpu@LeS`VIZYbBeRZ_(av-w
z0_``=UaZfq=sh6l3cwjz4D#D3G1$!3pce{%!Wr%^LTOcQOf`AFS%lj3)9~I@B?91y
zxSeRWBC>xl4<Y*|VR5Ej(f#B{1*zLoKsAh~T<)MMBMLDF{DMK84eP~z!wV3j3GBV#
zCfjf7_p9`Delj-Fq)5TXUEKUm76H)KMhY&xA*@|<wj7MS&Sn}pOm!!TLrwg=y~be>
zbwH*F^n3aYAPT#i!J$pKcMRJ>LNcCuR37gp>m1^+6t(IK#3Wd(>X=b}z`W5VNBu=C
zbGcLs9n@rrd;OjLO9YbbQYb=8Ad=8hGoTv2_rzlft3Zjbq^`dP3sVzQU|z`B2}m=+
zY@fjfY7}hR>SgrRNagLgeIJRKpsLT&68+xvIresg;QvOI5@IQFEYx^+J%Y0r=8*!Y
z)Q;8Z_dzC@R%A>4DxeTk9tRnSC{ldzB!P#)A3<5Ya96==TiEqa#tMuy1mUF4>!9w&
z`g!p=!TV0>nI4aMNz3Olun43GJ!H1o1wPy44D7o_^d<Z{yy$JNS=@z~7O-;q!&+)!
zCbR2^u-UJu@Ip+Am_BqnPSEa1-L%ajEHzziYg-=VF`RuKpY8crH7-=(p8(+AyR0G1
z__P6sm39;?3o)F4r~#eN__RDfKtEUVU^FKuU{FbPm``_{B8z-PK!iN1;0&77L%Szr
zE?}@71>^=%nHMuAFaf({9248S$k;e@cQ^#EkET?g>Aq4Z4;H6oGNFs_SIk2hOuhPK
zV3Ouk?8s`#fq0>2A;Ti8h@ccqrPgr*`!7*MeJw4KV?lu(AXShBvJM9JDs6c1I95OJ
zCg$%O8Sb!Mh-Opupg^U2E*!y-diUQrl@!rHD<S(C1^7A|w3#m7>?@;sB{l~zHGaRG
zdCiR;2vY%8z^{GDrc$}s3enOKjHL9Km~mAQ;lQd$!-g0Idu^mMFe8I$K*mdScN6`y
zyf8Mc4{)>1)hj!np%)puf?BM%(f6kRxUkRWxZtU7d+5dl*izOpq_TT(>(A5o_{yHy
znFS!x4gVoG+miT-HLY0dp))eWwAH9e`*CI`3r)^5E+zlp(CJ&~yg@irJ8Cih5OKIc
z$=k<I8w}z!XU~Dbm{DhIhSwl0t3S*q2x!GYO3~_dR9@Wnw}i0#@_Ac)cZ_pt+l+8p
z6tKqR@?W_+{_6YeKHMJ;&Uo+@?TI6}99s&}3Htl&W!9JT`;Fl9fYACO_MtrFCQlIM
z#&!V(GC4(1QmtP+mmg9>4{8`Q@W%G;#X+awyuiP8W!owdc+c&6KmtKDHGTsdfZM%b
z)Uud6Z)wOe6Y^~utspEKMhZAGKA>OE8T%xjUm*)o33Cv5#peY)mCUX_58?tJNeS`k
zgM9#nRfsX*MO?j3mavbqV2RdsV4lX6BT$^O2hfKIjF#G<3J}y75Le&b(J6XO1TTC_
zIZPc|vIwk|xxWBo296*g#54JIcnt>}9~!@nWMbn0T>0S#xI4(rQNDkZT3Y4eWEta`
zDg%<TpQIvT(BgNqSN9*?1PbTB@l)9jo0f6nyJF<YTxdEU+Or}9o~GCs=qsiw-yI=|
zXT{(UgrqCC%dh<0m5?{)#8%t(zMmHsMbR%(?X)GzrW2jLbWRrek1u`n{~P2tN@(<&
z=Rw>{#mc))DI`*1HE53D&Mki+esQN>Jx^av?|+~eDB@K*MGyph=38&CyZ?cdfQlR(
z=S}$f<|1#wSwJ0Bx-TAsWC>Nv@C>}lFC(YjvdO>h;ZUVW&Rs(fM#y(fD+Ro(B;E*=
ze=_3BT!&mw54>YAk6pF=^F{_)uX%he*7VOmPjeGF`x7`c1E-!0ri*zUOwY|I1HQg;
z5l{I8Wn}CnkyX;C$`2I@z#wL_wWx+q3+8AWN9&JF<^V)9z&s=ZK1355#8SL&?Ukd7
z86pn#!IO6FAK4`wpzKUuqe~=BrLx-k*E!)m#=YmdbVO_%Sp*UVD`#?E<%mOv$LeG&
z<<T<63Uk<Me)@L}cSL1TLF+fzsZ<k7v#9_44<(^$ESP=5sT{9}+an{!+&~;3gC?<@
zI)U5EMD8of9XVW(Ul5i*Msf?%B@N=%B-rZSi2>jb=&vG6M%_8?fG0OYo9TS1=$%qV
zAvh=BE*#N#AFt{H5*L>x=yqNyZEdx=QH<%A^5C9@lOWd@F$=M^c%U9W$`^A%qxFLq
zsL<ZR^E!M9%4-zqZp){oioV{_wIm4l^EDbQrTR(0a5%0Vd~Zf&HXJ%~za3Q5W=#UG
z>+oKE)*6MovJaj(72EF`sSkmz(zyKuKw+|w1G;N$4iIhS%^BbMefhkrXy3nmL~>;N
z(H)ey2Li*U{CKv>!CbzqvCnz)3X$>$Rb+)M{e5{;PA)!XLGEr8P|nsZb88^lGebXr
zw}qgR&uarPQT*az$|33mSzOraSD*s|=8*hK8n&@_jerGPWI{CJ(Hvnvfyygv9I8g-
zeZy72Fq;d+eFIX@y+G2S`3SiPBj#HY-NN!oEJ%H<Bb$VWW6qE+j64$qqAC{M079`S
z<v^Yu-OMU+Ac!AT{}4@Oh6a~V{LeNif*(6%MqxrN(=h&T#Yui|0&M0n+&IkEmp2dL
zG^T%e?Ggq-9z{{;w9PC|EQA(_h=nx8c9k-JLI3y=KNQ{nd3Zv>Ag}sT_Z6kb*$tl>
zC^uRub3X%YJ+k|YW<0fpj)*#yo35W_AvRMws70DyW;Q@3fKcbq164x>%TvZbea<u*
z|J$4N%b~W1q1CJB=P9MfssI1l9h6yA&J)=4yav@*RXxacDTWImtq7J;JWd!G_VE2Z
zAQz8Ocr}MqDmkB0N!$9{TrC5+>+Z#62IrCH4M8~xFba)4lG~;jroI^wX@`bVdu;?$
zKh2+;Bez7+sxve<j~g_kt*-*C?VGu_HsjfxtUr*QI{+XfLv9p%%@z_1o;=+1ZDe5&
zgM^_b&lsNdGu_NUTv?!*bWRZi_!bzTw3i8^rg4~$t~83~%!=h>-Xv^`-twE9-dXiu
zw!B!>3GAutksr_9^#djo3pTNV32BHK>MNvNL0o6(ZvHu4hzF7NgmVvUulRw<(d&44
z>@Bpz;DYui?8^+mgIx6_{c`?Wc=FR?keguBs)`yH#}-@7;d}$)d*9A;se+BZw<F5@
z-gooUJjkZtG>_qaWoo}C(U%*OR-{}Z5imb~F@*-zT|OMcze&sR3#tp)2>n`iTt00(
zZxQ(_x8-^z{t%SkELmijDt*5FQ3a9rjHTZfzA1SR#5EnCa1sW&yq~gopL*m7fzT54
zzBoapW2h_~1v}|_xzw1CQIGc*2;Mx*_3%FZfeZmxv<K`WZWZ?bCeJ`MT{=SpH$`bd
zQ&ydu(QzfPAdcb1b~hKXaj&NyUlAebiVM=7>1pMB2ei^$BxqmuWZC$mOG@wWCStnQ
z4YJw>Ye#cJsZHb(cou1}WEFmZI#8|@$^~d<?^xekEDIfjA_~!G)3>;QiK#=Ano9{Q
zGG|O+$1m>wGGwq7+N`D)J$LM$E9G|ZTZH=8CL-ISPGkn<o*0PGkW#10Xj|?@p-@!X
z)hIOq<+dPUX1#y&Z&qljEU=$e&Z3onh@W&!J-3(W<PlLb0!jTgW6pqdb!w<yrVN5I
z597K39qC<`nEf80<LQi4b+ygyd6570yUNlVa{?Ac8D@Gy$+j#P<9C_RmC45~flh5$
z&WehUh;bkL_0VvBBdZ5fhQTLMPs1E&BZrJbuXywyZKP+W#*C0?(M_uuhWreMMW=U!
zy}l8E-k=fgD=-|i;@O(;Nl8NH4&1?!7K%Sc;$~LdN%tD*;k9tceq%OTp^?HFf<v00
zwxox3eo?=(0MV^O07-s+)QDNPZHRGUR4-l3py~5W-|@cSeU{mO&xNxfBPz408Fs9@
zT#z=6v;hwyp_%ZtS4qzbU<w>(qBCEhD|3gHv-Y<axAbYO>jfE_JK^EXEK5h92F<XU
zL=ZuNVlAym4n#Bk2k8a^eI^fN6YF~V_E^|EHO{Z?t{d|o?_h;rHR~8n9c-#JeG?kV
z{0WwNc%|LA`5k~ydOGF?U)d1_GarBSt-3h*UtZx0hk?s%+8Sq2VF+CCK=KC6ki*k0
zFJ3{`f0-X`<{u1KYxH<E`JPA{c{Y(}+kilq+AJ{6PyhD%T|?+X&QF{eeieir44krz
zajQ;kBo~5CjTF4X{?^6kaxo5=sDCyk*mAS?LpQL+VnJ2>wSx`mV`*2kl0BLJX?{Kt
z79yh#FHKx^FW@fGfJtdEb*`~<4DTQv$jAW`o1JHC@VT217#C+T7J8#e4T?J-)hbs^
z(rIF+9afQC9jm%PdT~QGt0mAQ*3wF50^^QTYu0)!Nc>6P1dMfj@d)(2F@WRSn7D?J
zD4rIc9~_Ho$;ck&5qJB|-dI~X7Hum`P_36NzW@fx#F+(+09o0=lARWek9K?yde2h?
zBzi?p(PPgYQosK126xEEhX*7+0Sp{^jde!fGbrh#ZDZKAhCyx<s#IsYqY+hbY#fMG
zItZlvjimJmJVI8moFwHSObveQ)JVjgP=VGjA}>L5*2U5%3qzkE!f*_J;=q<tF<7v)
zbTTu(ys?6!;%<HIBsAwp|BnxQ<(I0{Y(L=K0KHH8#eOn_pG}dG7-#LLs`xIOWdJq`
zVpfR!0P88_s25&}b^1^uMzsdw-SKwB!v+8y^3)E^;mp_nla^6Ce-w;!3#tXA66G#*
zvX;HR)pS~|Zga2BZG%Whcx8~>M9s{MJNM_cl!M?uPq&9>iQ{kRMVFs{$jIa`V?L|?
z5?>5Zp%6kr{(_PvoAV9GQT?4IkIc{T3ep8Y6K_XSP1*T<;Exu?rpC4~FseWaE{TwG
zK7^XF&%oy;a{D9M<|SZAHY@ilu8OE*V{2v`9)LhcTjW_n&kY`fvSW|}8pD_+gH9bq
zkv|y!TjX<MLnSx5-3u}5*OZb%dkk?^z1?Jmz<_!)o#}p+vGR+w5n=3Hz=zCb*89Gk
zRQex`@-#@2*;1)AGp0#rnlu=?bKE*FU_QbY{zJ<Bu*#Na1D2L4p-i<iL=dexh0hkt
z4FXaM)MXz=rxC<hy4h6|h6q4YG7o3^wh<?g-TmV>zKpKH^wKsH^%jiw5DF`Sk#Sqt
z^DjI|)nu$>tnlKYkX2Eft^zO~hz{*f?%a6>>64rOYzHM6b%f9*)(sG!B4~Xwy~s2Q
z+dAEh<A891FtQyiQ+T((Sv?zd>04nQwzMv+C*7}Cu#@;9kR+Di{Zzc(4+tb9KfqwC
z3{+5;L^aM)5|Xiw)&jxYnshDPurdJOU$j8I4Dmyo`DA=5^Frfp+|#a>hti)g5=su>
z@r!+U*_x?}#&o$5J1zOOKJaqHca32$P?EPMpqnrIP*+~2ef)WqMkbW;saIyu$#mGn
zG^`|8|3GU51E_=#uUni2@icvaMkhnN$i`Mg|0qs|<1Rg$DIlZXDjX~z&m-@(lSZm$
z&u``<%g*Apm>6|v3Vsnqc$OZa^0FD9Ka}A1ge83;ipX(q%;#{8e?DW;-cl8W8TK?5
zWZrg_MygC+D7!I{=JWAwrjZ%ulOb!Y%A6C)C;csv89O`Ve=Vt-#3~pa1}0cQ*EsaP
z09iDwxUo#Z*aX4+OwknO8nrHNHG|2Z?Syu1rsF?ud7!guB?L)c_Qp+8=8P`Q30dVD
zH4UukyYVH=O~iPb(Mc>xjOt7Pt_+{S7nV(CNxo?L&L7Tp=l4r%h+yh10wL(%SVT2Q
zvI9ZJ_62{Igow^2z7Xm_TwIF!;p|@lWt8(fF{ef-kXqo#nO3A^vAqW935Zr8(PJ8a
z7wqkA*f0!ohgRL}BWJnz#br8s{4oKz`2$acAy<w-XZ!4PqAmXBE2PbjX(a&sPkEyQ
zD&0Y%2oR;Fu^JQ^BK_6{+++c)#+UpnrajLanOJ<y1Yp$w79Z%IRI^<jaGGn<5#Xnc
zdUSkLGuCZhB!7dop$flX&JaiA`$8fH$W1{Y|18qUCiaA1a}h^8H>e^9805!)_D|^`
zzfj*FN=dnXovI4tIr?<$V!{-r3?MsXk}y#99HH%kLuqO9{#<PQzlc8Q)8Ey{riZ(-
zXTBq+d%@@DoZgpDfJ~VG7M{$skd{19u}pCOXPH@$1!U(YfMpkCr`_Bnzj;fY3g7_<
z=>NbMQwF*TU^{sDHS0ug>Il4)1a}lvH?gMQ9D4sb$x|4VYZQPA8u+31c!Wfv#_Zvq
zX3#R$lVcqxK5UyngJ|Sb?P&tVB3C`^BLNWuA`oc{#AhZ<+`Gp@@f$IeGK){fM1$42
z{c*f%&1^vphG`jzjcqH?EiuORAnp?aA1}xl1;QY_-mph>nV52v_<0}n5AS2T*QMo0
zCO%Nhs!$;T>+I#yVz>@C!&!#!^S&Tc9{F=TaU1;8C^CX_OrnzwN2w5pog=odNnKRR
znTV)N$&@zP_n478NJAJ~K{JMx{B4k$rBDurB%As*0+?*)v+2c~b%*=WUfpOx#*Kn)
zd^%%k>6*ey4Zpr2&>;_j0xieDZW&4W>h;`y2up}_0)-q9kDYAECewZ%0yGF{uS7+-
z=86WvRK8%*+FO@+DN(aba&BH`0U6l`DdugbhYZgLZ=Gp9RN{{!x+LYkuUHdD0P9@M
z&g<uu2U38>p6Wf{z%i;FTHm$avd;YP;S3EbQE~GCq^+(H<5IKe@wS$)Tm5-$v-UC&
z1DByx21K}Lo%c|Q0K$7QFpIE*F^>*}dwQisDH{Xz2M7|4y$80@f63WQueImX13r3!
zg(H0cC6_P+$0+8F&Pqcvqh8x0)O(GL+}4RtO0b}6`5csi{GhTe&M+?6Enjiy-+KcP
z>0xZfV$JXacQL>y8X<uZ4#fK>t8=;kBZf@B3AB9K*)R+~K+Fk-bCGu@QW+qS-*(`O
zSVPDxGIez+5*YvZ4qT?KosOv0=IS9WeouLYogcR^Q+|JsGc0^Vj~PIbVi2A42DmQ9
zvE{)L35Rd<J3x9=stOf8c>o5GE@D3n1aTt(@ASZz1f}#MC_Y1YYo9JGi}y7UPSo4;
zNUC)o6TA5^Y$GD+#>H@`odHG5m~quMdM}1ol24q<fC&$GI&~PiJQgZq=Y_$+8^G(C
zEE-^U=ORV<6p4&;5x9fM9Q##}yQ>IbS8^G4Ijs4K=ZDG2)yexa@Pa)(z@7!#p6}aW
z!ViJZQJD_|nOm6;vEH43|1v)xW<GK}jSJsNJxsUVsSE$VP_Q9>a3WX-nL)r|e5+pe
z_aE%RLCf$!ns_yL%q!qo*(txWQ8R)nAy<&VZ5H`bQa1>zdr6SU-Zl$3m<_#SD(XIQ
zHGmX?Uz>KeSuV(dD~F(nA#BfeL+rBOFc?IGa-2zYsZ@UeiO~cKJ8x&-O`ZFw9r&|l
zx|;o{b8_D}dFKLR6+sL&!$?+NTU#{@`ks1JVfXeiH4x;h52=CG<(h;jyqyL~CzPSX
zOYgS=a0P*MHlI9VL`HW5(95$a#b!m9GVq5#srfipQyOzFz^&O%XmIr#s>j3!iG){I
z%ps5o?2v>}Z8m!tjW3<z+;OOHlUXp_wX~8;^@@&p*xXT*v^1ld-zks<j$w<q&HoJp
z=vrWx^2ZtMLuE(NcXVIg`G@ttta5k-fvn`I%TIz~j4iXxt^Ih}%nlq^`@OGGq8a3X
zE@Lft`VHVdTpb$ZXFjbujvVm%?l%YDp^RAJ)7Hn*C4*N>7slUvOl|9vMGsh>x+^zv
zWl?kNlR4RBsRY@Fom6+m1WVxU8twSFXWYoF9AkLEZvA&D^+^fgY511j>uQtM#b$j`
zDO_0oQu$kokZjp@m#!zRQPTlYncFYTNB-8`M>-*=x9S!;<+HTIi&Qvm1#uW>VZbe8
z8$&^^QSXHhsRa){Osh!@=xbx9T`clC!5uFQ83sDD87(m%*6kgy%%&r@5P0Pjc2rD~
z6qZVr8W+c3?(!ADZ5mc217PU+eV3$w!N;msD+@TimCZ%Al?ldP>}pMwi>3v;U=xTe
zQ+3fNhpGRqZm8w|88dwgB3^ie)z>CWBGtJU0Nc{6F&6(OfmvEE091>;1RXxr#L>d;
z23&oq{GMLtm!Jq*a>^nd5l=3x`L$X?6zy(#)__@vpI?||e}4?<_=x6IZDuv8VzM&y
z<)KzIL{{jzgc|W~Bf^yIixcwLP8t?TCQ1D34f=uB!5o{Y%scjjAeZS5GL8;L#Ks5z
zuza$i4LABqe7C|7(HJVw`al?TUVCs;h$(uWo}XKxfrwYd_X@J@APDqX#vsJ8fmdb&
zxPy+el+xOP#R(+oJYm?qU}0uHd*62U-eUtRlMHx}?heZA&$eWqdfIOA9{!X)F|?iY
zx1EcSZrpzXQzV~AH`s2<EFD~obB4M?U*R%v={2vkUs5P4eD=}85?u6e5U4}zwWtle
zSW0S$u?d%n@vtChYhK*^c|7h22i}SzBNV{99#!-y_H{(Ab@&U!{~iFm?Z=r<4U<F_
zDmZ(r-@$Q1d;Z|G-6I;TxL#X72pSdt$y`lVbMy}$_Ifk{^5E_LH53qa;3U17SHtB+
zbY2mw28_)1&)Vztx3082=^udZ9bYr1p7lMtm~9M6ne2JK<opcHLS0f%<z)u_V*A?$
zPckBd6b5juDKnM#`N;1*6Uk)9a5TGriL=SXWYe=>Hs7}9dr}G;796bk0~K)eHTBR$
z#lf-u+IYIzMFImC$mu{LD*GnoMW;R7B&MRGI8~XJJjkw9UAP@?O+XN>h#HkRv*>Z@
z6bhVCujwB6W!3*RI=L9;bppj4i7o?4s(CBH@0BU+)8<ZmZuIBwBGwk<78|=3u^1`|
zHKS;Pr2In%9WELW8Q~h6fpyaRIWg1AB;GdzR3YY=eHFJI-cEn<c6<0av*kIgl2I6T
zKeno9dR(P68V_-H{JZ~mfPeu0^^2c$!PwN%QFQY$B}Vsdw+rfDVyIbkR+=0vqw_Jh
z<z!kbnS8Q8qLYcE^CFv!#Woh$7ycbwpBk{;Z8;8n-cQ+2MwsJ(m(kaNC2B%+9E-1t
zAotDB*PfZHUFUL+w$NB(X3N#K7hBe?&*(4`=F@i7tJc<?K#}A8?EPg2CSeA<$EcE8
z38?B+_$s38jw0&KXRSjUB@_SX!;#3zI?tI@F$@N>jxUqhEA|7l@PD^WN?1-P4Z1aE
z*1Gz7c~xaj{(MP2m?jY%+Wg~`dK4iht6uRCRJwh3Vi94--kVoaIycSnA%iEX?^3MP
z0%_NR9%pk;JfNzM@X*O0$(Y@=@)IWiCa|y{UJ=DK4pN@;qQ<Lu><^;W5sKwkB4i2(
zf)?Pd_-WkL2F-#6`>5R4D8bc57=9T0F7o0rxUm<+nfuKHhP)l;awtOA3pCTELk^>B
z!xjmpINALea9a428^!X_y>%C%qI${Iv%NVI!S^3$seI7m=+^S4Q9q-zB}Q^&VDx7s
z(NIeKEM^zM&yvP0&bILokg#7n$z-;K!=%^gnNpLbCzgoOutc3R)9l9_-r5icoUX*g
zK>H;dXR@mH*n389bpCBN93)mX|NU&xg2&63v2kb@x?Jj?MPE-jnw>B0he5|ayc7G-
z^nCb~G$@SyN}oM>YcT%(@bEAzCufhT&^ZB#=Y?`YVj{2W8P(&HR)dkSyjzpf#$`G)
z9ydSZ?47uBw+79qooD+PQqTeWX;Nb%hr5SBk9n$2g1ILqTli#-|Ez;Hv{m2G>`i7{
zxqqoNfPX3~J_Pouq%E3lx*IxoycHu!huQy3Gh#Pq*ORw#{l!11jgww}x%2(`<`0b$
zDdVSu2JBcq#6cNdUCYWD{3vdy$Ke!wG2?l|xt4T0t)h~RWx>!lXn$aY8vHL=TyE+<
zwEF6K>dc_=3hh$MYdR_@n6sn9&8O~Ar%PSdBTP5CIg>9FzUPf@t%d~KK1OMAuIodL
zlN&^R;<Cz@ZTB=tY!%2wKgI`>QBV|x%CGUy#jna)Yp-y9NFGrBy_g;smR|ZF4K0*G
zY!;Nu8HNaKoCcS!jGcm9Bxnw=Q6u{9fIkJtDxxq}0LHA7TbhoQ&6QJZ)ey>;uBNv_
z3K*2jrXZN5_S0HN@{*2DCs~*onvT^h;#4DWtDVK-NJzePP{5@`Qy|gF8T#hgt_b@<
zN0Aw%toR39ZSZ@J09p8KLEMC^5i@L{2$>hM+je`w`vne!t9v5ejM!_%sqoQ@85fw_
z0tRlpf=aLTb52y)_}>N#hxB{5m8yVlmWQpcit@(Eqt9lC&mB#_%-pzjh9Xl|w5Jr7
zcZJRfoulPs(ksFgXRnUkh$Ip;%u$hif^x};3KQ!Zee@DJ@f*Rx1L)U7bF)rk(G+b8
z({`35Q$g!S?V?!e--~@s<{s6sQk=0`ZIYCMXjbe5r~6Kiq>%HQA4E+$iSM;^-oXXV
z8~QYertMnI?DN0ih2*4h@o31Ri<eg~=0I<A9m6@QTxskP^qQ-?6c=|s@^)mpxujH7
zcQ_~h@z|6MFJ*GV?xMp)q9)&p@j=s&4<YUc|6bAW4)xN*v+ei%4os1^Dn6UB1lTf?
zvqnyQxG9W)yR#euw}(s=C}EMcGq1*WC2OSQV_jf|1Y!_;9+L2^p|Pt0YETbB3W+bg
zBHAb({x+1odlOcGoWb*HOP{yHZD?n~$uzEXD_i#aZ%l}h5pVZp``echVbq#mk@vs6
zxHYnH5XKvKe7g-FI7lCsG<dWuR7XgmTLrv<8qEH=9aBO6{K*{kwH&#}^$!brk6j(@
z4<9iX@o;f9bahQkOadsSB2Et4o(=@=_HsO)&NFFKM_P0G9!$Y3dm8zCDr{nE8VD&+
zJpIQ)fbTgL#CxvMyXORLq+S6^UT3i$S5=vwQmgmZF;qVUWxe~8-suea(sh%z5L`2|
zrEogNmub%%S3%igLm{L1OQk4DEHpgYGLz1)PCY0rZ44RYuw_&y*hWj7Z9Fqu2{9qY
zB=|!o!4NDD=Y|1Q=F#mERH?p6cby!Q)%DQCVj==L5TcrZXX8N86Iz$L1d>7Y*j;UC
zfXUe7-ULN6fEE>d?hv^_`RF_v3kshzI+?r==p0+us;)5%Z|U<M_)H$oh$ATrXGX3s
zxZ81~!NDKA=t5BY(2_=ISJ+d+v(2!zxc{RXq$d`G%>UK+aK|1(k=2ixK=>UytAFB}
z$k~gfWq!)mgCact_3Y`$x5(0?&{Jzkzj~JGyQCDd(p`Cs_|FG?OT}n={3K}d#<!Db
z3WjFZkLRyDgMKaQw7NHS%x}}MFmZof&FOMBRN0ac;z`}6?CTp-C0NASTqvpa-+o0|
zSPi^2=(9#3#w3zyiw$acG4?-#l36=VbJY3$g$u^WBnAz?zTQ|3-F`P{fA}}FS-kQ&
zQ#2f2j?~oy<WR#_g`8v7bPi=~2`X&A4L2fg(P;gG-RvQDKMbaWrx0QEfJE7yr<<?$
z3$slMF;vR(f3Rck%q+#k+a7hV*~B0`pJe}XzAd%7{)7`e@9|P6cHT_W!BUxMM<Ki<
zLV+Ux46A_V#3!igy#m)l_K7>7-#Kk=>sB?Xanx_O_GP?wDhQaZ9FqDS#mn2R(1gQy
zG~{iXg!I>3hzOK8;&D*IRt4&M40#6-BjYu#kip_<T*I|CXB40+zjM2dP-pJ>&NI+M
z#k4i46OE#FL2PuTG&;nbD6{|M&@h-qXC9<1@~iXmCa=SU*t;NsvL25M%&X0<p`-!G
z#0EicDO~=Vq|4HbM$3k5FMO}L`wrgA3?7ZO^W8evuXzKKRf7)X5xZ}{_-^|jI=yPL
zt<{eS?kjX+YOIaiy(eNwB}yGuk$(qv)!z23q@d^#TArqv&Qmz6YRgX6&bMM4#e-0L
zUc=(f+i+aHwu;b$xA{M*M94A3AV%O7Y(Fjc3~tCyl1|q;CX>Om)ybd1f{jEHm;LO+
zP7zc;w3!L;`_Ax|aTq+B9ImnIGbu!cx%I8f+lR7m`>qL;`i^hiZsA;yp2+yVKZCO+
z7PiE?^<zm}k9zAFj?mL-B$fn5#_nm`1IzPU?|bX#qq<9%%%_Wj+lI`0-ly%hheUy^
zIq1U%%M-=(y6^*Hsa~66gZ5K~UR`Bn=cJPGNN4BETaTyKx9xRW605(jZq!$sd)hB~
zRTV21-i;C85oJnOO%RnY*A8^6J&e0tY`KI~fBnh8nj9(y4+e`bRZ7~B+mll^t8hDL
z3mFx%Xrt42MhWmc+`f&3o#YKQrKQPE*7RIAaD2mW6v`B=3(l7Ych1JX4whrj`#BYt
zW$7C@0gDM0*%!uabV{bK1gQ!EMafJ!-?5-JRU&$6-hg&mb(WoNRhMjz5_75jH$D*d
zkjEj49tJj2s3BFu1pXQ#2FI)O44uZ&u}!y{p>~`mA`^KDh|vby_+4$we-a5uK2bNX
zV{O`&g<O`WFeElyM<V~5;MYlw*p8l1V<7OY&Z^Kpwzn!#Sop*%luTYqZ&A&d0vFg@
zIOGQKs1Ca`=EO`=Kkb!2c1{f)&eq)ExsidjE)tlVu?N1Vj|eMpu_*mGstO{Q^5P<d
z{#<+wseh`^|H*(^0u(P-1GI*5Ds0-q+1B?CmZ$IPCv7dRh`APVt<Vj>`G8p;Q(*|p
zR?I+r;S+tX1FjssiKLD(`sQKYa_fIa*Ti0r+m1rEZfV*Az1&t)$^E_2)TeF;8rCRM
z_ug-IXPcG^o_$SGcW%psJ^T<&?XDl2?hVG8ip;O98WWs>2xPmKANMq=Q`IVt?$q6E
zF9dZhy=B!-rNG$LT^ej@iZIl|OX5NLbQP7|JH^D~rD6^TvEB*32vQrzCOMVaoA&WH
zv<2)-zponV$9EpwV&i+!)nf36@Byyvh0dB0o+Dkp>}0k`f;O^~G(@Pq2(Y_fV(<b2
zsd>Bbz%roUYvdq74gIF7!uJAWIV%uPvfI(W!Q&+d&3X8Nqs8oQ)~$f>V?Q$ciN73w
zbMulogo<dj?@CnV6hDQEO<kT<Pc~aN#wAVEhc)D;<UIBF2ZU}l92txv#baSrll&z8
z8lkm@JOl%(7{ylmCK2+i^EuHB$yI340&!3I8x9$3y3gK5nw?|VqGV9BZdDxgVEIzK
zzehi59dVE<&S!~usu4&+u-60h?>ocOCvb>_xAt7#sl|W2BL0}oRjy3sZz!4*HPzFi
z>EkwQ+-i2VAQ>3n``wQwR^bm3BMpy0#!gw=^O$GLs281NyY<YUKMeoMs1VUewtgUz
zxGd^D&p{(k`za$&l%Ds6Uy&h;<rfs(rh1=SpRGP$KR*h#(Y$_{;d!O$^|b4qVeNgt
zSiN{8T>kPz&%-kq&uaFzeVsj(I5jA$NE_rz5`R8`>s4WMx@RuMV~yrW9;GeJMM;56
z+-WzUFsHz>wXEywBTX`D;@>h?|0ckmXuO6n)yNz|Wa~x)gM?B$HE}Ikh7P0+oCaU4
z1W%xe3Z4I<Y48gu?l;<`_+~)?J()jGO-mV#;r6d`$|JwR2()VVVCx8dk1{I1n>3p0
zz7w9a`L*=ep`m)7JnNV0gkga(h52Flhk#O;tfw@-#4zr#d!sV3@Y;@IX8g{7i*E~a
zk$bBU#4&aH#A16|#n2_Jn-;GnAyPD@Z1GqHdHhxe9QmvXxB@M#KJk*7>Yu-CVX~1V
z4^fia+Sm1cfws4!(B(UjBn`7BW97)PylyBTpC{aJtz5`H_9*jV8PWJ+%5M7ig?>{)
zANZUge@5$FYtGEa?ygJA%-pO9mq5&7#%T6#tH`5QOLIrI(1O6DK-CF8DF%AO;$cOY
z!7u77DtnVKL)pGDKSDQw^E>-Fl17jKo<EMq=1aUNCAq;-R(d+S($qh{h>}a!Ii^O$
zvbAB7IzL`WYgyPIGwfe-zxWLK-gdR}O32O52P-!D{w{?Sw<R?s-+8j8dC`GZ?q=k^
ze^tpkmuboNNiL>3Fqgi-F<08Kz%-WYL%tmXAya6X-Y=Dd22~XV42+@h7xZGA7pe)d
zkgraSsAVN}uIrzDFK410<S8=4Kz|c9{X2ps{Q}EP3Cv?&RSM(6G(B@$&w@aVSXV<K
zSV?1#n;1TNw9!O>JRgZzf>ATsM`a@;@m=|7>QQEH5uXCwkcVWc{QWN^nfz@%BRUw)
z3$`9adSS89kj@PapM4J>vatSc&U||soj&MVkT{lb7=HbWgoE>iZ>iNqBG`bJMBsQ8
zi%3(Yz4UF8ppi;yUv9RxsTx5@8C!YaoxCYZ^M9zHRdOAG_qFF}A*%I=av-MvQOyV(
z7Kh%}FXOK@k{2U`qTDdiUKPH6OgL_;oc9jDzel~+i4GcI&kOsJEo~MOcI3aMfWv~`
zbdjDl1P77XNST|@7)~MUh_3kHeN{1k?`HWJ9r3-3b&K?Q(|dSV=$co1b{mfL=0ES}
z<7ZDWr6#Ahw7qqy0$+_+*IoC%A$<yLi5#dLMijmg(REq(KfZN*J|TS)8{B61zGZnf
zZhbtm_Fg?+=Ru7)zXP+`QEN%&>4if;5(C?Xn7o3?o{U8e!qVJ6dqzanQnfT8!*=z+
zzyM=#(%^hN4VWC=QR83#fVDOb^eKf`&A&U*DziK<AW6$Ly`OT&ZKOHBFbuX^Xea7$
zQ-W>r@FveU+4>oD<FOpE>ac2<Yk1sf_?KxfZwrP}KSfskV#JOGe|Wt5zu5Q73ysQ(
z{e)+>sOHs>1!=7Qay2Wv-Qoe4J5h<B--$<ekK4ZBY=vVc;SqWt{z4h3rMm@{s@7lI
z-JN~IT0{Y81iMg%$r}8{k-ibt3$<ve<=*?_t7SGv4Xbt+n$e0=?90}n=_6(~H4{cl
z;OdU13jVQa@7Od1PI=%->{C+IcNa2EB&1N+fLuvdS2EgqwTxJ2EN(T`r8YOtV7doc
zf6|5|y-!I?>Oa}~@$zjR)$;hwvp#(=5bl;<f~`-U1^a54##b464II;*`Y4Ki0VOqR
zTx_mG{A=)Rg|~J!)Zh{_y;Y0m`6Pcf0NLabp_=;+a=DOHY{Rjz2&ncQ+xLhw40-Md
zbhWHG9N^e(J385+8#~VnZlR*0*u9IysmY>g_4yMOjM{oenD;OGV??hl!@M~OaEItm
zCCkf%KRkLJ8{FS@K?cWo*z=Ewbz#=kQW9Ee?$kst`YZRw!i~FM+y7!x5oh@7duu4F
zU+ea=o@mwx47}2PoKG9E=N8VEt$?JI95P_QqJ62WX>QjvQxFq;zNejmNW+n)XMKKY
zu4Z*~MNb38w+&~-<W7t$%8|xct^eHf;e(Jz?(pdVGRun9+4e8avtayVL6wb~a8%Dl
ziFoLR4k1TAdMxz@$aJIUF$jW6fxrrbd-Byr&D}%USTrK$^K;m6)dK!m!))y*1XS^N
z|HP`BQ0tqeViaC&#h=puw%gG#12gt6jL+K>ZlNHqi-z5n>zx=_r)3B_{caj&xuRc8
zj+tEwrobTz`$0G96=5d$JmlO?cX5+!wTrlTcc`9ylwS*n@f6sD@o8z;<w11g-L|8_
ze@G&PM`21vhExvz(*fc6gK;rzjKUw)DJiIsb@I}tfDXt(zT=aB@Y#|ZB1GNALMXDh
zUwdiK&W>%ssm~w>%2Do_aNv8-{GWH*ohQ!s=RML#%BQ~PQ|~Qb(g(6rmo&Y*JUv5a
z=S1(vvgb>FuVJ?qMRa&ohWI8S-6GX`r=tbIwjGcnwW;oXsm|;D^x)|ksRvpW)<_?G
zy-y;%C27%jenf1O4PoZ_bv|ECa8<d+e(TFl`v49qfhP8c7Z#0I-wW{G=9uBfdtWcB
zk)(~-8RT>U_=r?cw_^%hSJNt1&P>3o{3HF>@=vlS0MU`(+H_v$1OZlB+~25B;=jyp
z&WlOt)-SJtu-NC5KYR^-y@(MBuEsSAY(@4S6kR<EBJay0-z4U2Zd3#J(Y?n=R+p)+
zCN0kF4FpXgCO9-g4|b(t`C5C(gUuF3S5FqEQG9j#*N&Jaj7LRLMILFcW-~HLReTGR
zFFwv|b-9(liA^i#REoL;>Q3=gvXV$w;Haoz9Br)ppIVdQ)<g|F<6W|^1Yd2R>8Ryc
z1yBsr3)z12QIXw!a7*>R@c-rUx2B}-@PyhizAK(eigrSvNn@+XmwF?mQ*qX$2^t#_
zy2}zqVKs~3r`)Z|9I+XwU3LXeI*zgzd3CuHiZm*3ieFEzvzW8u?7!)by-y{Bi71Pd
zp-nedE12Mf=&@-6rse1<oQQ}5nf-SOyKtD3l%J=F?f)$zvrdiO*#2Oq6v#)1$M`A+
zE#ckjlq8%5;g+dAc5t0^6mP^{OgedDzxqZTnMC1>J??^)cydpVrH&74i7%=z6V~|O
zH#M<1NN{z(S;l>@vU-u!(W`qEb}c$U<rps(k1c|)e6*mwf{B&V6_4$O5Bd{B;-l#{
zI!zMoWOsK-GC}{<dAK=eLaL_<TYMH?`2zaVN3mF5_Kcl5@swGv1&_-=%<nISdUAhc
z(Sy1CHKhd$MG%G8<`NB7G7l*)Tdz~!j=z>;tAlG|a9T%>z3XV1v$$0gl>yc|BKZh8
z2qp8NkQB(@2w<kA;c#FKTy3o$;kKT+IjMN_T!+SZw(%{rOlOtafNri7COOn`CaPPZ
zGtnf58cq$AYO{&+W?I^df5(aHjW*s4T?*A?gyw(8g4ghklGiqFjcUJw<4aKb{8A^L
zVBrOr?(tKGRaZX7;~*;>;*7?3^SD}U;Xf_JGsx8lMUZCo8=gTFUvDmf6!W@W4n0|P
z2-Q2WOi|H-q%oPESD2P-xAOVOr54V6UxfU{!y+?&moHxKg{Q%eg|)X#EZe~`^tFRJ
z{N~rkj51&0*N?4|?V+Ke4|UH+-a~qizr1n2mw9Zz_c(--c2zPO-9D{st^O`4VWOd>
zb$z<sb=?S_)#Q0z_P$;{9kKR`tulqJT;8_mMuNHbyY~~QXC-~ylX`Iz>V3-MwTf}l
zq6nLuoU|T#_3-E0TeW~80DyosRb%1fEcOS@=HSsCtnb|&vOZxXUE!h;VR<(fp$Eu1
znJ)6gY~fTwHo3##d5_`bB*_>8uGrA+SKuN}IC?w{i%q)o8z^I+p2^1cwo^h2fwXk+
zlcqT!%!2h#pEm^GdTcK+Px8jb#u8Xf-|=MFteb4SQzBsmlu0S6-Pd1Y3!iSUSxRj!
zC`Obrk{6mVbmlR2{BI36GR7va#*FiXw%lq$cLT8}_-<Xo{Fu{J;AUw_urvp(3l>cy
z!!S#-xWBZ<gMv>Aia~l`L-tyA@m5<?Np<6Z2qpTWs%e6=4cXLsn)9h26jHS87PX9C
z=Swtgu2#PBy$2YknvhM<>MFKP)kn)Az>}I2!yZ(`u1od_?Xp#ln(3+iJ_2_6E%GR$
z$Aw05@RQ6R!#3F2ODA_!^yslyXJjE1!$~0ov0qW(a~<}O>Mpba@N!YFMyT)lB0g7f
zX6p6#xYD2IZ$^ID$gB<+{+(aRXewK%0ekZBmI6559@WCBh|iu8G`^UxAG8p8crrrz
zOo*#q%a4u47sHt5QAt%XgTH}DZX&;46VCkPxq_pplz;1<$usI(b$kEx_*A|p<!W(v
z-x7Fq5GZE71=q^Pb$+cfdo^dF$paMO1NaPtmZ7bbe|;11*rPcMa~*2=xM0h_t5isT
z(qq=*#mJcrZ(-H{pz~5CD7A9B=?&-8(ONT;7wqj&@hLs3(msZ|zKnJ>r09Ymo@+1f
zpLVLGmtxtJ8^v%#mwuzk^kR3i0xA(%t@kaZEPZ-WT5u`-lslJ>9Fd=vm^v$bWccZN
z4o<5>&7l7!mx+AdLe-oF5>Y~_^j6FrNz~%0sGTheRdDH1bQnFTZFgZFwGirPKdrA^
z7E{;XLvRuCBZ%c?i<e07(}ZTl>E{?n2meBs$qC{LPLJraYl++ps5c10CTCyYT-dWA
zRT;o<-qnw`;~O3NrNGD{u5N^iy$OU#@yjpXql6}G+ZhRrxTE3OdURCQ=opmB_i&;n
z#9Gubeb&OYca2pUBr`62IK;BQzP!7oTz)a-n4XtM4UUS}LMHeKa&mI=3JNVbeGWQ!
zlOJxwS8q_fHfX&Jk2f>#rV7mJ79VD;A7^wj!3OC%n#Sv!-K_i5FH1u&(Kz{Fw#<I5
z{m1dW;OgJldhj%Ms`$d%YI|f`lw{R4=n9Ii#HkjNlZoOTfUD<Bd;af6AV%cq_RT{N
zv2OG9+Llp@hl=<MPcWdawls@jO9<E(4r9HQC_Qqp6~LgqxG>x=)u;|7B%aZNdxHd8
zk&{91{gE3AB(|Y^e!Hyg_`kB!S{ab^G>W&kKr&SGg1Z<L!}mfEL)~?n)z!5e7$em`
z6Q-pNpRI2~-FsH8VENYfjBzo}I`?Cj*Y)0Ti+BU?Fov-)!;BrY^-k^wzxr0JQsy6y
zG(!Kp?tINblkn}4H9ILxnFPmZ<N7_H?;<gwLNq;<p}X*LZ@+ExC)u-=E^2JejXD;K
z^LhK!=H`Wkd)-Ec^CwDUl2jM~lZG^Ue<n>`^Z3@{0q`GMw;RPig~2Qtmfchf1COD<
zN(`M3c>h4b=FAv1;Vk|TsIj3Cq$*|j%?08}|BgF*<lEWORN3++^nHOOF@wJi(nH~G
zBEYsSpm+G+HWmfT0-7J3dkx#q1^c~4L$t!EvAKO7iyEf&QhPWM)4t~t07jMm%I?pL
zCIZm%$%>os#94YO{~XIb1ns3mc0JbHt8kr(d`bBPmu;aJWB&=??|lGC(T9;sPoxiK
zc;BsOn)bgu@i2r?{PFw|ql^<=^xIXh{GR`na?Js2I+o(Y<<6&ETJjed9RBcN7bvM6
zsk1t)TzN+!H&*kKrV62thTVY}tq%pMGb*)-;;-f2ozGFq6t7@9@a}B&DkioveLM&2
zrJWn!kh67gg%!FuQ4!VaDB$#*u=QW3D`m=;hQ|mi*kn}2-20cpYf$sGViC(~3GBvQ
zw7o7aG%!47OOTPnrk1P~{sI6HNfeVc;<A16BJFqFUvLrMAbn010~gV)R`^}Gd0L80
z)igFFD~ZPTXOez{!8mcqYgl?|k2ZOS(`@fKr4OBv8RFrU#RDCcs3=W#+Eg@la>h1B
zWRb_l6c3`TOsiK;uUpqf);y6^Z%lLL41lcOswsb>asg~r7l&r6hh<`VMEC(BMRac3
zR&=sJs$eQy4?HOkU$1k*x3ty0lN;@f9@!Y}ji<)%-np-SX44%W9)4nozY<E2&8xRr
ze|dg>etcozdRWotaxpG-yRG+R{1MxMgM;HCyzP{G#qec?WolktUX#$3^Yh=;yxD_O
zxwZ>AB0h&8jl=uMf)yv+lWohVTW%|Z2GF={|1-v@M(978jlPITj#MR^Z+s)OWz(;7
zf&kE>Td%V_l=w@)v0R+5G$y<+imK2_f@qF*R35ae49R+a?6d2kAHlA^=5N$Is;FM1
zC00*np@A1@u$-*X0f77M6jqh$Jt|z^?xPd~IB(vDbEW@~g6^F>QR18h&t;4ol(7Ar
z!Q+mIOv7J&|J2F4vE-IInezdY1-*Q5xm3`OR4Pr&eBAk^qM{L<se@bX-qo_f@p9P}
zMM7ikjW6MJsK1{gZ^>#6CDgq`$IK7RpiY+U(<P|Wt>MRAl;@h@=Wef5(gC_ZFk5!K
z@oIyb`9<W&?k#{6`)DS@jFG|H0um(F5ZoIYh8!yncv86uhTgf)+M7=?$wdxxO_#RJ
zF6jNHx|T%LMXy}Ix5-&3Fm(q&N&L8d^`c`G6@~<NayW+#FiuAke3Ab%b&m=-Hb%=R
zZV7`XZPKxtSKNb_VsCPk@4nc#*Q5(6qU-mnMuZo*I*`MRW^zs40R<)fjsYAR$U^D?
zx(PmLr^f)5W%t}44ej|qHg*(U-i_5_)YOfsa+m9=-U`=oFCRKH`8p;;jcF>qUG{uP
z<5m<TCjY&E_2zpa59=#tyT@Lr6C(J)Ej9s$b6F%zHb<6%Wi&t&bc)i;^iF}j(PPss
zz9a&)dQ9Cj1I$0r#LmGob7w4X+PPlyz}B&;Z|sy6YB*?9P-JUaAZNe!^B04#Q#1JS
z0rgT<-nJD<Wuu6=SO?-v)-X`tc-1aOh=H!EX&^wnsn6@rAkuPdYfz=&x%S4(d|~ZT
z>OH^`JZ`_llLdXk21gHiQ{)U@ja+m0W}9CkTqK(fd3?N(E+T~tFwo-<B_R`K4ry6-
z(sh2qv6V*f;3oSTG!>~;^MXx{JYQNP1J;fNhFD=gdC^TX>K?br&M_P?ka1*i-3(6a
zSj?IVnA?@7>6aKQA1sJ}l#ivSF>Jmr*DzQ#Lm9^oey=4M(`ch$iX1=CJkX9RPCNF=
z)j<e0eMFL2$Nvyt03b%tR)2gS@g)5gMD^e9NePFPw&|950VZ<y@z`59SlsX!nhEOj
zR%SN6?j|#LAendAKn&E@`7d1Yf2{cb*h}pur*PT4PT1oTjE$Ff;9v9i()odR8OBNK
zrBi12^Q?E{u(Q?7Yl@1<=}G}rsth634RD<bW4h(uGyRVKRkTudn=Z$nmL?BI>eIul
zY@6d0HfC$bo7;k2dMZaFnfkn)P#RDEwJlW>w9raz(sJzs${|cZ@M&73%5-D5I$0(z
zXN&|!d>&Kvk2pH*7JVTGsn<D}Bnr9+1n-(D1D<J>rSee_?t>llBO3uIJvjb5Q_9_=
zD5DLz;<L-kjlhEccYECT0JkVXZ@hLx9$zi8ZRvE65D_G@aP;P;x^y7BT;{;}h+u`6
zz}i_iFo(u5UXGBswP{Iz?=~(50la3Qr%lM+E0#rj4mWFql9-PArn?A<BxNW%s0dIe
z*jQ{SVu{~o$%7~agnyde8;#%x_QiB_#%ECCd#cwp&FQwwcTizW;`3y%8f*=>q*iSr
zT2u!B9^jQTYpJq)ugybL={DsMYesOS=xR@EHdTj9???4l;)|DJz3hA<ujQIN{|E+>
zA=|=r6sHXK1{Yx_ay2b29L2J#D$3`&8kM$NkpEzmbEoG_Abh>cxnWAkBQ%oH1@?;L
zyikh|fq;SmSP5R;Law92mdnC5;&+*ECoEonF-S6)9x@sVg~QyqaTJd8&F2$8;Y+Hz
zbMX_W`j!683-A!;^=dC!?$C8pmXDtQ8#1S`h|*7E$e0jI<q$tI6UCxeq;kx^mm;3S
zg8D0Rc-Qvv`!|?z!;ERsf^my6n}dL1O7H$YGTMlV%h$ZFnj_SjA7ad<=U4@_bX1N)
z&R(JB{5l(K$qj}v8QU?eAF#zM=!XD3J7E$Q?2tC#K_n5gr-2G%;{1xD{YEA{Slj2C
zF+uvzo3he6nSBOL4oUaDv%zmzuYyZT{VC*<QhGh;J0OT}hT|~-p0;tD$7dc+&L%-a
ztUkX#=ZCS!jtmzxfC-vHk1|HA`g)K=9SEp24QNc!Ba6snuqAtpJ=fL{fBwQe%8D!w
z3XTdJcKN(J;7d)CHh>v$Qj*FUU<KYoo296EjM3gxQLxieQA8*h6*?V_cgt-9QGvaj
zV=X_|^8rl2MXxMcSOxE-K<Kil&f}2f!N2X~-c|ZPue$>8lrJT{tM~5Sk?HBgikU(q
zcx^X5g3$hjg$4c<wg-{^DK+H>W0Z>BpV9*CdRr_4LPOK(B>cpockLn=#4fF#zj`%#
z*P?GJj`>wK)LUIc>uKzvbujO%!<R8s(d-{uZ46Sci5M!!ajgfPL)rkic@#md;TX#s
z9a=c(u`e4xI<`IRfGU#^T=Y^2NO?T_Oq(q2f6eR;%O}KT*;%zH&=)}l@PX1rMe)Fz
z{^_s7J!%~0Z_Yq@lTDza8ZG@`cBKPmN4PvHO^)7f!ba$I4qZqgFaS^AU09Gp0RGTR
zFobVyi_>O{R-P1QTOVV%_6A`ZjPjl6xI;3sU(<K3xNhX0uZ-CuNMsn7B2J^BP^!&$
zjSL*goF`&G!hk&B<w{CO#dIyJ;Tu2#9tAEHhHRz$IM}?yKpQnuHUc_Sl=iHO56R;~
z{XkTsQX>3Lfo==y9q6z|p=hj~lM-N0kW}vpzj5-E21~-2Mobn~kS(&0>+~K!7|YT9
z#`kW6g1&2ub~(Z6YFbt<I^xR_dTedsPu{{S!|N>WK1im1>jOydaN_WIYj{}c;QyiN
zDx<3Gnzo9H3ewVD(s1Yo>6TJJ0coUDx<R^;E|C(DmM(*o&LbcpjYu3CLHUMfz292?
zvN&+>d(X^OL!9wQCE+aDfn7tPGafGO=#B5<O$FW0Oan;piw^=_Iu0dd!k#|yqmSzF
zg#pf>hAjr?u~Rc0PYzj#O7zr|o!78^?zJhtl~;XF9mS|Y1}lA9FERW$?U8>(-D|r&
z_=ob8B0EkeM7IX7cX&5Nl0?N9hJ#WMJV!9gTC&(w>8d$9Ny`k>g;83j+CVRH&`!r7
zel&4om~#?T8T$_HwJHpDdlBRKJ}YLgFHgS_rV5A@ER%-5GkQ&<C4@GMF}*zvwJ~se
zyxssf31+eG9yK*CM}u|`tJH%sRcnf<KA#|Yi>pPj5JCuAideG6n=}UfruFeFVqwZ}
zG;jOe%=I;oT9u1a&W|>k#nv^X!a<r_)G!uC#}f4E&u;F0^CG(yVW!Ve*;SPlP=Xs*
zKkr_xU%j7?lfZ3ihIva1fA;|-rhP9MD(Pbbs%$kB2GP-m*Pl%rEP2v-JWcWfR&W0(
zJt{dOryO2uG<>id7?n%_!8e_-kvhENU%PDm%!xqY*`o~@0#5UbfMvftC(=3Vo@BWz
zJ?s7TZ>5A|xC1BO-q!#mzc9VYy;6E7^Rhv;;nj|_5o2JM&*p#@tsiA9`sVYylep#`
zmkyVweaVnCk6x)*DmA7MU$!h20Dd!slNEz~eVF5(NiJJ<Dem4dIyk@0T9jn&f`?Qd
z+l8j3_YN1kgwJt9<a*D;u}(-cO5FJNueOq1L#gZZAa1yAbtc8FOLf`JE}*Cv$Oq^0
z$2hXkJCJ{%XD*&MY<QfdbpE3viu`9&P@maO`QC<zP}_4mZwD3Ym)y_oqGs!^H(2{#
zNDA6F2=jM%dxl|<Ww%wv+1z_f!M2xVS)>bX+`X!hypauzQ_G=fL<;|2P{6?!#qEmI
zEEWCVzYFiHR$-Az@87*I5xx-<&0tD`h9qLT;#!YC!FKqF&gckPlY(#+Q8|Y4BoNb~
ze?fUaL<-;|-@8$OCWQ8_n4_MZo`<2rj5gvGs@9GD`f$BG-BKVCmyW{Tcc0jjtN^jC
zjG&5dAHz5@?J`UBvy^5Qy*IC<Yis+ArSC>u3^*#>b)ibG9gzxYNo&y?K5OyM>g|cu
zIo+ZX?5vKU8jKzPAFuA~>)`pQcPKnKgRid5{@;ma`Fhn9Id=|%csM*%Hh<T#Z6<%(
zJ)IT!;S?z?vnf*Ff8$*gu<Be%vv}EqIO9tMuAqOJDQe$2DYL~lG*;Pi5m$g$+|w(_
zR)4vrU)BC;%SZNV(~}6Tq(;9_mWI=B`S*H%-8&13E$uUzD_kETZd8zDps-}2T9;%F
z+{TdZ8aX=saYjKdVN~%FJ_ht5uSaE*-k*@?2&*sksKw^>28o;&_KW<eH~|io#<JAI
zPht43pjgss51=G<WG$d=;I2Z8PLq*B#?x0v7L?qR@6qfuizu6<q{c-8eHYwh5)`{S
zM(IV2c)e+xWr?LDh{$1#Tm&KmS!Uo|6uK3482=EP5cG_43#;E{NM#AHU*r5?kMo*_
zHHHRgp=GOJ2lmMkf^?W{azxczIq7gN;)CA$PaHx|=6~|#o=*LY0&a-HMRpv;@)pt^
zG#lGFstfxX1nEPP(b$(mxq{suQ!tRm`Cl%v9NbFkAyVpm<&+4Rj-37%brRt~!MX?d
zzuoLWSph@lnPwt<q52s@9e_uMx+t)8#>kl-dj2v>0N@d300i&rLhD@CSql5fb5U`f
z#T8e#%CE;AzFUNbIH(uxyNk~&`j2l^>>loYb;N#Vu-zysn6x^fa%}hN{F<9EA@nkE
zkdlO)mRgB+KB3U+XZd3SS)J{)5Hl&m7B&hti?N9c`oN37zOw$><Af}kmnq;CTh{oj
z$h^j;C?9dZ@lL^Ykeu_a={YcA)zKSIr@XN^D|ty8#Q|0xAjI<)7|Gjmsh#;s;)LRr
zy{2of_im<o`sMu+Z2d3Pw073sD9J~}xN*$da^BwVOY=Jfv1c-{xK`H+v&mqqcCxzU
zg#3cKAU;|{g+u`tQy*IqFM`GoC)_^kCNMmr!ja+L`G8q$Kg<2VeC*QxLZf}Q?ldg7
zgXQ>giB!ZcOgWQ%Xkg{^;*6bTGfnXG4%W_=JL`4-lOF7YYAr4I^f!KQ4((Yo=2ek3
zmFMGRN%2L_3nGySOj&Zv@?EYM3cq~~y*He#w9MuxNgg{j(nh{neU`h>UZsk3PG@JV
z`f*M13R`I`ts1X8-6_A_uyx~bfwYl8Q&{3?gclycw`WO(GXd?}h)jXA-UugB-#CD+
z)R6mqj3xz`$yp^yd|eeL<-6}#@L+_J2BxE%`$`%@ke;`-KTA|9=qIU?l0HCkYTTKG
zt_(wn!fhBr&bE>D8_z&HJk!aWP+Y;ny({soXy(=FjwZz{_9#JmLEhpS4Qlj}7j)?D
zX^T%)f7|Pf3B&Uc6=jWn1K8D{`nQ@{X+xT7ziw)Ds`V7jyTb)+ZcI?EqefP6KY0)Y
zd-oiVKH;(&G?(JY(yAS<CNq}jZTN``zgXNty6M?rdHCdgLXV4?L)sX1T)Viwu`M2v
zDsW4%ai;ccc2}aazl<Gd5{K^7KSu`KMM?x;G|PvF#SlOq$+kHhY!V?AGfAn0pMk}m
z_lZ;ry*{T|dzw+3`Oh$12vhqQ%0b4ClFz(C$p`U6a9^}NTUAbNNatio8_4jvn4l*s
z&kJC+Ty|woec|my(F43p(&}ax^1g7czg^OSjorVUTtT<TYZHZoSdWOEJ!^xh<O(oj
z;>5U|7K(8L@|MAp*<U^r0CFr7?yqM2e#1_(E(C8WWw~;2C6}{fRZSP+pjWqhKmYM^
zDCSy^yQ>*n7hru?(2Nj75jt}5rl+xuyZE?1F=nHy@uPV0R=>jb)m=J|mmI9`x9H9u
zfp{^lu1NNqdinq<bn8<irnP(pP?6gLe56ddZSb@}pd+b8UsLpHUeK_2jp>a+BE542
zv7$j2Dp=#GLuw9w{0IT#0V9v+&wVZWK&lNr%2+Cs^DlgsFS#)a33?vBi2K_@H;cb~
zQ!f$j`uY7`>Xa|HtRH0^J7|0JdT~I7uVAsQzirFU?%YRA#4lo|n!KH&TDRM_TD?=4
zcJp${l4_}>YwToO@`7t?PMfzav%>s^!(;aWC$U*mZldBh6T6=NCqeP6f36|1!QngP
zx;;&~J~gA=!VmMIE?Y6|56GOP%$4W>l^s-v?l`F+_0Fm;g7m@Tc%78zwy*5|q{!QB
zz~1w-KAz=?KOA5LpuPVMBHq(QUEY3rHomkaVEx@*az?$claI0Hc4*NR7d_>!u)%^J
zbysb0MNl|pExC@r6w{u1e>@*&skysklkKj-&z1&KKtIjOOf_}|J9%H5Xi6uTaTk5J
zvBwlMqsB+}GVn^82wT2m8L9J$f6I61*cNYzoh&E};(>e8y+F~+sjLD&0_v!Y_GJuw
zbC+J3?zizHx;A>Jk2<Amo5Xmi!_W8BbX20P4Z<E`6EMXU@q&6|*O1U6tDs{4@H1^p
z;8%<OwPMP2zy94_!x_$N{@1Rw{0$XP&IZm@=e^&{^xW%Z<XyYRc)3K@(7A8jZB{tH
zezo<`(#6BCch_(pNUpuqT*J4f-ntdv@zu>qR1>a*(iu`TBy>f+ANWJR8wRZ20>V>@
zvJFS#a)vSE9jK&Er+X}_TjL?kytt^=H59AkPxgcE@!N$Tht9Eej_*}E7%Z$R>y%zd
z_+2oqi4jJKr1Y?|-iPQm5HE2Tb$esYEo8P(g<Rpi^S_t=;D+@ik(U&V?q~=`;+8J{
zT2+bpPy!sJRM&yDnVv}-ViAL=%i|;_ehqpIOTs8o->P0{zCtS=4<C;b;IbEg<4%q?
zI+6`X^{C1J#+B3PtnTAAb+|L*=JWYktHw4@3Lzby0M6UZXTy1x=g<wCl2DLDOpScS
zp#8ZLY4g4&ge=6BEtM-ILO-j-mYk=kYX0Nt{v-u3x<J#T6%z0WCHZj6ao|ed2Fya&
zCS0S6L`#mCG(t*g-H7=pWv&@0XuCV|dx4rJm2}<@UlrJR0u29bm}L8L^ZS|s8GbcQ
z9nj!@K9HSSt;>dpgzBk}j$d)K5}K+eyP8C9u6mcIsh<}6|HGwx!E`a7dm&xwt~GiH
zazo!X9(aY<!u5o6aCd%sQ?!Ov3_~PCC?k)K%J00&(AZ%ZsN974p$w9jh9gG-D(pC#
zA&e#3`=otaFY(U1@WvwAemvz#4kCVGOmg^4@3v&vo@)B6=)Rupj-QPYpJm4oz_F1j
zF6Pl+*hK9=!$4vvkeX!Fs4Feqx-L-Wn;{)%B4t{gwba+ejjqIKnVQ*YM)L5EkWjKo
z%ktjry1=yUEDU^#c*GylTO&$Jc*8hAPr>v?q3a3O$})lGZC?TE5>_}#H;p^1QS{1u
znVQd*OM=a{_1-0pH9g&Sc5(moI45Tjg(myUH+4%y>=#;CmuBKuQ)coPyx1^d+Un;c
zOvrAm3=o4F-{H^GT@jF?$M{ICt1~DvFw8sno$T{ha1PLf^^A!MX3s@2{%!mC;@ZD!
zS544@aKq=Q{`-{KoSF)q5XSZ5?AT64b+s)_iEYwrqx1_DGU5!jtY(8KiN7P5Xh=(V
z9Y$<p0d}>hdn{Z*qYUXL1F3j`u|QITKpC3&@h8D=FK?bzcDQ8jj97v%q*B!7nkp{B
z0&wzrVM0z0fnH)nWC?N9>cqU_^6zco7ftp3kHvVqOjhTSCuawl`g1$ZFWI^RH4~*Y
zFWX<nQAfW&v4x%p#xgK@fGl*S1HI$D*sqtN2?0U3^BhlR>?LJFwqN^fS;TM}&P?+>
z4-l;O-FulXzqSu>TG9yLGimoY_u@E0=YD32Cxo&yZarTxI*HvEPw7s!w+wo{hJTJd
z4SP0>JEBSZODt7xq3ak8?)Ri7!ICtVV-4PP)W^@AWcnT=2GwCn6HgfBn9>KQ0vr}>
z9qr3u#S=NDvXe8nGy|Q&rIoaK)BL)Katv;lVSXY%s26W)gwyt&L{CT2Iz|4lo>f`z
zqBQQ=2WlL|4!ftk=U4zBfwIQ>UXor?pFa7M1Ud8wM^1t^8_(AIdE@HDl`UlHN)m1O
zUoc6X*j)EC;pka7Ux{dwNvD$ipWyHJ)=4o}^!J^6CyvY|JPB>?Q(i8MHDko;O254I
zzEa=04-MY*&UaDKRDfz_31s!ZzdSfY*58p9<Vm6_V#ZPYB5KhB0-7pqvTL}R<unj2
zyCMTNhLtA0ZD{FUV(ABIWNnxrJ~zeiD^H3W5+{sYU1jL~ioz+wb=e$ZJ(k?2jxX5;
zLF_q2hQ?v{s*}<T$#{pBp2Rs+Y5W*Soz*{XaoIW21R(|Zq+u*p^X~6^4;p^)qF>Ko
z8Ep`wO#5c^^4|KCuqapd>dz)oh@8aOqTrt)1koRr1Ogm6;5i@>6i=EKL}&eC&;}?s
zr5F6v+X^-CLU6b9lufDKhVMVgON&m3wE=nzPlj5NcC|2Z?Ce*afyKTPAQ!k=X8*><
zF;0$J=KsG0q*ne`S9mF-V9U$i?%^4ACnDG1IaUWLd+lg*lN^6q@I~8Gh6m%l^iUh_
z!ky;1kbuW?c5@v$3-FG2c#@_P(WtFr_PWF5mD|8oH4oeehsrh|CeY!K`lFZwWAw@~
zVsqJ)ZyNWO&xymUr;ymWxyg17oENa1F&hAQQO~!hDfq?&vRC2|%hWB9sP7+|1P-PQ
zmW(1hG>nNy$(ye*$yc`YM%9UC^?QGq_hXuC0r(Xua7{=VB@TBg4Uri}HMUOvEYoop
zYM5FZ!ax-Dw(crKT$NFErO38NMrC%mGGAy!Qet<VZlK3<(k|K>?<;JwJg2aE^OMz&
z|MMD|Gg!$+P{71_Ep%bfB{p|^%q;HdCKgm^P=$xQ)>P`iz+xXDeAoD)z!l0~o#nx=
zXHghrc3yOj$F*Tf0}t@bmPt!#O`F6yE9*VX3uR?GRXB$9nftVd3T&D%=i-}8me`Ak
z1LPKdg4CYrhF&7Emv|0m6_;<%g7koj#S%P9$Wj4CKujP=EMPE^(tL=kb(*@+7*oRf
z+cgwek^M!L*x{T7Gg`8qeNhg(xrKjJpI&&o#daJX2JYGjpBW=tO8zr+wfk{3OB;ND
z`P^)YbX@3weM?k>pn4LwYYF@zX+D3iNx1qN>f38>&wy6kGLQZ}8fPW{*<+&mQSEf}
zVwcWWw@5VZ*w=-L`nNo)vor3NNgc$JQhRI;KtfZU%_zCBpkV-PP4s5u9jdH@^<|BZ
z6uyn<GBKucW_)O{t8j1!by|AXXEhxovmk0SZijaYc*_9erIMW*c8(mtm<ob=6Naww
zFy#+?w_6*>Xv@n85)lJwUsTixa<BRS$z#US(QIv=&zq?eL?p-|3Amk%_YZ@@RrFVl
z^i<4u-qwG!`kc^4HJ#J=P~e=ykec>Uty2sHT)6e*IYcQ8^YG}IL=OB!ouh5-5(P+H
zYfE~Fx_A~<+mNYGX)N*&`ii}=XPJP^X!8%1Y?8ldZ~jULdH05P2PiN+!7-!~H#BZf
zGb^+Ed5b0{u-|OTsCaoX3>PL}PQ!|Ql-4q|=EozeWJsIVni&}#bHDvO=GYj^@3!}^
z0HX0I27<foukyQH=jjV=1_)I(L%Yk$RDmE^NzikO^gc+eq6+a@wN6G(&4LQ!VkI~U
zkS(Xi{XeIONZJ``hY3_o7gZ{WDJ=8OIZMvY6JU8GRpWFYiM=T-Udr|lhr{k3T~CAk
zl+$0Nw_4o5>M1_93e@q8I}v6og?3F7UK~!UD=+6azW^F<kIb8L=Q6#wyeSA)1ZUp>
zq7ZT>hKL>lm_P^TD0LcyKe7|`U_{U|6=4>+IAOg>gw406i@0*QT^a%?wbAuIl|<DR
zDiG=X%r^`Kjwv>vl9FdGJQ(;)3)?+9<>y+sqQMg3@iXO)@o=0vKYBlNWFN7JLD<xY
zc}4v9V3+5HlC`^4vo1itec%m+^dH;L^mZ?qxDT_g^<c<F_1~^NqtBp?$n2MlrUcrd
ztDyZv<hO{pa*J5I*Q?YtOn{8akK@53F?qy4?GbH_G%54;C;7y^H~jB{MBLH!x7~Kw
zzpBmup9my8_|Q2Zs4}#OD@kCX#&8M!OR}(|56E8KRcW9Pnx0jK&Vh$Vh%hq9>g1wa
z_(iimLka7ZQIwaLSC1e_z^&Qm%sM%TB7W+@=k_Fe$~2WEK2vB+blc--nZLIWYfsS^
zDO`!LNj*FQ!@r2~r{_V3@?^`UOq@EekDRF(TKAQ%N&dM&kUL(lAxPjc#gwC4T*fES
zEb5MGAcL+w&>0{SX8Xxd-cYP|D%6HgtR+@TQ#&j9D98q(gMoS10^?od8~O+^Gk(m&
zmBrmst$yZGS+Ve!1(UA1z9)H`W7aM+c$NMvM6Pyse@rO~T#(olWj~On+y5Q;xreB4
zPs4^=YaaC+!|W&FX%R^_DL$J`<D|O!Ih}2xX>!H%AuF81n1|RaXG5AvCzKpcKHR~}
z&}#om6p#&jnQu;`?C8oiNW@`((EimNygG2?e{VCcPdyWCA2!91od-e6n0?&r(e+&k
zO~yNPaHTB%p^0NBcz+bKdU4?um70>-y6Z?Xi@kt1IoaR~fsoLE?Nj(4qlDlQFa?3i
zkU;iF1@=p<&Is8X0Cbi7Xsbr^j-V6^lTmq;z~EIMN2_eIVAs-*(U9S@B^R@{cO+%X
zLbN_C%G(td8v}>*5iyN&)ja9aEnJt^H{lnF79A#?wQra;Qch}S_5@9rtc!}Bpv(S;
zwJ9|weSU(>5J8K(sa5v=9epv!&XNyVV<iPI;xfD`<ms|nXYMFgX_D2<RimUdZc<Sy
zJ41jS;>e5WgP3De<K)_yg$|qf`f^sr0OC2<s9E~#kvLcg^rfVe&EaMGgmSu2zovZu
zfxw-Fg77}u8~qqyC$8Y+7I|26d-V;1G*9G^+~2|dWBiDQnbk)R+v=&R_eZvpwD>wr
z?c<=)gq8tVs<OfBu~bIZEJOmcy5c7cTCe?&O5}IBiT^VF`u`K|@Azwm(k7q$a60IW
zuqHVC)9?IV2fnl;wzyQC=Kw@Je8@9Egb(nGK;`y6@KEql!Edc33FDcxo&mTsoW1bn
z{*9XEZVMzWt`>;fm*Px#O>07|-XGfKM%ZTMkvZD6eEoEHPrBOW@1ko;bO~&!1_8K>
zUw+J%Y@@M)elipNeRzVV%R#@ZrfXi#azNh(BEQhs9+w7Eup5EPqr1kb6+C)3jr>`1
zjjS5NLOP`o)nRkr;xZU_3%1-PtPd$ttWD6{EYCmi-v;n~SF8q_@w`m{WiFA^5A3(G
zz+Yo=&#5`(vCGEdx#Qc!l|O26;02B((#be&4|qaJs=V7~Y&i2<U2W6W-4@2h$#RZi
zHo~iR2MBeL-;a`bWwp&2mGcz{IdbtOV^dIlJ-*@kgwg6fykSKAvuw!Q7ZK>)O;5-i
zL4jDe4Pe1D(OO+F{*rr+TbUP|9!!=!0hJE2RP5hjuonIO7;uiNJ+=g)KG`JYQI51|
z5m2yNYI9_wUJu{=IKF|Y`Waj5%blhC<4Yv`=znNt*R5?i_N(YGr+3l%?~GclMJc*9
z^UL6}_vcUH-<qzD;4Q@c{ZWj-R<Z}1DJHT`grYe+R&o`ju@}@!7tb<*?nH%UNAct!
zMzn89Yo!gPuizu#e~=b1;wCFj^J$B+^1coktE@-Y{fKcxXa?Jr+jb#00MxXF+R@G|
zLn@zth%kuweHzRb*J%*3b+Z+ffE;pe9Rjr$N8ZR1H@ljm`ml!zvE8+sjiER6XCD}J
z`AE^>YTczx(9>5VF(Lo?hxZT1BZ7no8v;DiNVaizAFq1(I@b7`u~eEW-D%%4qAl;^
zA(AUrQB*JH-a@}#GqvQ%(1)ye%3LHfdb+Pfb}7CKp6I*(khz}mbP`|A{r7hMhh0i|
z+&9Rb?8f1|lirtUHrP@uot@XS?UnoE4&tcW+uw3yA6l%OWfx!s+tWHRq}qK7zbW_M
zhfxnd$x4Yi!Rw8Hsi=Qhx4(yZFxMSL(tCFO#aUzN1L@i#H_)`yx)gpPn(LKm^0c;u
zK1`hJ`Y900t*E@>y?!@>84sb#i1~Tq!ELoqzc++0sA-eJ6DDrmTYG93Ko&wy^;k>g
zF7>h56t!h{Z`zjuWP`8v6FHsa7LQVi_Z15#S_`;ct3Rs=iWo899u!gqGBvtK&F>`X
zZV6>AysVk4AALa5L&QG!rUIbme?LrluBy9Ti?WmslwH4zPQg-I_(&haOc1X1M*A(P
zr56!e%dM?r^d)6L;l4HX^f`H&?X@7AOqbI!x$=-IYw5d8Xh;P}aMm|GB9$`@rZUlB
zT1Xsu@<AhKyDJ6@z!ic-unlrO_Yh0jg=-&!i~_Kn!8~Q=l7n8KKn$`kD)6_TlA&3P
zg?%7dHaaWk>=c<gwnolx?KM4;=~myIrhL(=x!RQZio>A|>p^S@_hX130mL4giWQ%x
zdbXBC@ZEdRw?PdfR8oASQ84Y+QzUlSn?brvvdn>H*tWYj<K=Bejr%^if<j%bD~*#O
z;vEh$j;X*E1_xeQlQ!l=sUnw3^m2KPPB5c@V*RK+sz{PJQbGygr^=NOvwV7e(<8c1
z<B#&f@l`GLER(L=TKsbzY%frQBKmkyX^~VipxT_8O>t~`fNzG67|Yjkw_)n8pNu80
zT*s@%jn|`(B+if;KDNS^J<Q7c_Odj3E6m$KPa?WX)OfBuI%~i|bTZ7i)2(L-h@R$@
zehZu4<Uw#xTAclEQe|{@(wA8I(ft?g?d3$e)4yE_zi8)Q^6r<|^9KLvjE;}O9tOsf
zjMe`2rZw9!n_R%)UI2pl$A-~44J7M9hD6@-L07*B%wC#Go8-9zTnV`8>yWFHiG(;-
z9RXSK$|9T)t|T7t)m7LYfp0Ua@aA2ELeq146Yeec%<bcANTNQ4v&K-XXj-BC@OFHz
z1cV;Ba)JO~saPg;q{1dC9v<RJJikf4+$VkU*bjF;9hmVC!L3(Mb#;Nsti7sTq|!$2
z8aDGh@G(WA{0`+wZ=Li!@ts|rd75j&9ZDs4D9<U=k8!5_c*v=i->3~CN%ohf?J|7v
zAAm=&qn?JQ`xU5!4^}?k5GWH1!U2KNc%4{#t#&@op&)n8O0Lf)qr`b9+b~;7JpChb
z30);;&$6hkTcBX6y=Ba#0Sj7z;ZZA)^%;Je%0tH!Mik<6;%3Up*7{~b&qmK@#Up1s
zR(56t5Yt(TC~&ALzI*Ub?#Q-!7D*N3mVbv=50<SA87{1{>Xboe4fsU5)D1Z5d)4UO
zLPE^WF~RE5DYGeI+mU2F@@mmZyxOgj)F#$q=mrm#htTqq;;BNsbyFb#<`E7}zexi_
zQ7`#>`o1nx-cXJTDsCMGu51{vXkB!XD=pu3u7SI~-MKE3K(=5R@o|(-)y$RA#e|@G
z{K3aRZ}P^dnwF=Y6P5=;P=tEGSx}z|gY4AAhR7GE*?L=Oz+9(RB7$Wl(-d}oFM)0D
zCY!)*oV&JCP7_@}Hox1}y+$xmCMlATTBy*Jl%>Jqo)BJi%6ckwg~;qK0m~39gWQ<4
z_|@)srbrLB>R!uXyRkai&{enLM;hiAT<;LvZ9gi(jG4r<d&yW=d8HB13N3Q1%KCyK
z#s8eJl7H5n#*D>Ex>bia47mMj2nk4^yr-CbLe8PFEevM4>dBgtf>(Y^1N7z6=3uAD
zgtQGNdJ|Clf{tHJW!-b%D5BD^4CtDARPfd3n9J#hp7c${{#^2H-`Vv;?b}*9B*Vno
za)Ea}dE~+yFJH08rM+`BV;jsr@Pz&fZlUg)+hP58mk`ORIvFcV2*^CsV1gE!(U}nf
zTmIA!>r_4w(o{rEXgkHtope_$jA}G~q5A0h<Z8_|6jC<MQ#4M&-IpC`y|b0?I%>5c
zvIpHj3)QOLsDD4X<Kpl}!xq{UDH`~0?=feGt?zqNGy>|E*!;yF`T6dRAjy>y!?_*Q
z^1~ituj>nq_5CA9LZCVJ_i;~#^6;<LQr5c1ORF{AoelT(!Hid!#L*%_*&xkU46GFi
zn|{{^C<K(vGnr^9O(Ke}+50AuGMY#@<V&nNrJ|lykZ#;4%0fVv|DD^D;A;w>DZrBU
z!xk-eKX232(C3xigMSJ08(4nW;g){8+uAP4|3YPNUYg<Gz*nu_8kt(o$eJ~y5Iuxx
z7MN(0qwr<bsRXq}uy9A`??05Px|@lXnzQ}rW=x=J-8xPvE-qeEV=rBOTfo`~0|k*>
zlnJez-D1^Tbze7M(fo?vthL_7&MHNQbeOYJkF#`i8rz+nW;xSEJUQ^+Cjl7!Bt;kU
z8w|)6(P>I)$A7ml)#&1I6l0Es$_ZECg;q_GCWbSE!`G|+anF7{jy{ug5(JYF3AB9A
z>CJXUt9D^hR#(_iF<%a%GJn5OD#gVz=GKB`h!Z=e8Qs9<47Vnzi;HdI^xfX!<b3O@
z{t@uPcZYBtDPrcDd<4|n8MtL8%PG-z|JkG6+jH>yr;#IenBR7ZzFSc-4I<6sBl#<M
z?w7813szx);GG5LXmW`~H#<ojIdZm#yDZT=1u-A^!(@zQIB-oFsfvLa>^?Hc@C|#D
z=j0{Fwe!oDNHvtxVCH92A*uA^fFq2%!N5FbupE?wlU-sluSwKGh4ADoP&wbScC=pB
zIE`OWJmKe*eQwUy!FQh=h$A^$Dx`oG`?Pxbfj&0d@VZ!7KY2w#)!%s5=a2}%zbmHQ
z+)wUVyrrYGsWSI4Y8Smk<b3`0N}H3l!;~NTUYulT(Oc$w7Ak9?&fu>=<w|%5#H4gt
z^qaKU@TA8jMQnL@y|CQ)83xBImpanXx7x?cFaImtQV3UH=v<VL$3nVz9L52;{z`iR
zA8=>Bs(0;t2Xco>Q1kPBsiE@vc>0Lxce)<0E^vCVLF))vSMYIryZFE_4IW12C%9Bk
z70JNXU!_af)C7aaroPV`alGZH<&W!F(}1KbN{j``;Khux%@u^v(uNd~2YW0}l<d5J
z!2nMb1DMqTuex3F7)LG*k$jfeo6T#=0v#8)T=K{OCWYtVC?~sgd<T)lC#&&f8&jxb
zv}?qydyIDO>xecfTA(0o*8^~}YedFv-zJlS#VQ*XxwR~fg15cH#VT`hE#8?nltfFB
zlyp}jQC`E$+JH<EFMs0=joeQtDq33+Y%e4TvL_-`vY-_=1@}6;;t54=e<p=3C$;f=
zOnr5;$Jg>7D7r5Hx{@#E5oxM*^>%hOy5}*uJxCUK>v|z+WNj^$@Ks--bKZHpEkAHM
zYUl%<$*!mbaTf~%Ni+^G;I<mjnI>}nf?)Wf_>ly2#~t@G?B0h2-%8F~gtl#7oc|e3
zWK#wkEa;!X%u{Li@dSsvxzy2BRCV}RhUcq#`e5e!$Y}U0>#E&6x3}1%tb5kHAT3Da
zv=6-);*dP8n)*QP^vgO5OdA9|cjx_mq8}0|`<(s4s;uCZr@3w-znd=25X808^qkM<
zg-Vj!4F6zWzlo5diWbRZGPtdIvSE2`-0)KK^^S`=29+m2!L7XXtZvx!z%$2??dK^C
zvJA*>5A1&%5!G$yn^dw;(G%%6+oL+DLXI&$xxdx?X4JlTFiX^srqFZ7xzM6ez7E-0
zQ7qC3T+@x)K8GO+ig&(-7e32vEvX)_P9Nw_pl@uPT;0?#I<z#L`*cJHUxqt-^6YtT
z`{3na-!fCA3gF?D$6S236oXdfa-~g~_xp&!r5=HQlboekmz=%io}g<G#+IG6uT$dq
zf5J47S4@JHQ-bfa{c!h4i0MhUTO2;V6`WX4BbKj_#{;4yxR}yt;d9oXUA$(8ASmx)
zY1*Q_!2Ac4>G%;3F}mnMH|5Y;Gc<A^N>jA?65G?nXv@xuX*FaK-)~3<84C@~V_DJ`
zxIssRCk!JE#56Tc5`wrDM6N(X5$D@~f7&CZB^s+*kUo@tYW$H1Pdq*9U{U3`LM#PT
zz^NP#V5gP`Iv{5%otfUlr(Kfxv$YFTTHx!OlR1On0SD!fQS^^GaNwk|@Fc~7=6EM2
z!H#p$t(+5&KPVhnM9puKDTX~9a7(&j-7vnL0lFexHBBWF2a;K@e1v-&=a27aX3=Nw
z?k0vnugOehh-2anH$Zh*N!h*E1#)e>qZ_jXcsESPH(CmoSv)}jX&VT*fiZDncu8aa
z#Dn3J-Pw!ljvj2PP94{NAlv6B6X_BE$_wt3`X*R8y!@8qaysFf@@yDvU|7o})!uB1
zc%`&FFI2eFV`|5UN9tg)$d_xU)d?CXQ0MH<+vAR|Ov`@u;`?})GPD0+(&kF`TRnqQ
zol*Px=B|i#k?9$nK-_UN=*kHczf73HJ3u8Y-VF&)b=w>-+M4?lstLln$1wkfcN|WL
zf(?Dmo@(}r=o2_UmGQ!k{R@S>?X+Rn0#K%1$sDRZSb*VZLfe(=ZJPpmYk<=r-|d3{
zJ0vJb9PJ&>b8jmBr?ZA(7Esxzq|$$4dr%RWe#HA7{@%$D113*&Z(t4oVs6-Qr6)E9
zp@nnq0CyWx<la}DSqydaE6EbXC+IrpiMmwvdhn~wj2uUkfJoZI@5-4G3u_dTLU2KS
z>g=r9c?Mq?^hIDJTRxS^h^g;#6(6!6Ui<M3e6XjJyKDzwRCLs)XR7Ih&;UqR;YN?V
z8_ArFivFx=_o>pyCW0#<IGO%HwmLX<<cXX24U6CCG|i=N72cUL_K)TCvk?oKJ52ro
zDfVwyI~L|(C@Lca5jopVEp}RJB*1;s=1z1eew?AhG-h<D(QP8Ls+yyO`L0aA%&m`<
z_UX&Kv=uPNC^*_Q#I2$;DP%2d?!S&AEYYax;R#dPh>Ru*HItXfr&R_*3;;kkxjpTG
z*BWfkaY9XUIU=EF2DwcRdt`mYhtIWP{Y*7XYjBy*?h^}APF6^_t>ehWBkIHgTx)m7
zYJE&fRczl-P=vfwo~aXqCcu-sAG()XJDyYl&wa^2fE44Hm(B&neS1I~^4)%VBa&<3
zeJSkB{&@Q&^{=-)pvryoXu54zz2{tsl;*;7hG#R-uX8<DP*>U{z(au4YaQTgB0TTq
zi8K~S5F({WM5jBgemth`Ih=>5%!jz|`R^%(MD*#WgnQV-tIS21_S!8Vtsjze=~Z(1
z^b~2qNVNXBiM)SOJa*mm7wPMQq4a6;dw8%y33!Lk-ay?Tk{NEOshGXQ)St?bUb`28
z%l?^Y*T3dadi>}-f5PwjDc0ea^4pIG);*o8WpKWvR%pYQrfsk`x!m=YXDyI-NOYeM
zKxRmTrPb*IcpYPYhcR;Di?t@wVU850LRX7}yY0^AVLKzsmXw*-J2%`55!8ic`Sjrd
z_(>!cI=58L(8Bs*wgFKwRY+^|%)9t-T)aFcJkw3VanI_0N7Ldcw$GoPBcUyYaa+gT
z3u4jk3Y3t=Beef*B2%D)nK4MPEiU4|=uB74(PiQ~l!<R3(Q0fTU&pAlyY2`c_RmU@
zsA#nX-1OM;JT5D6z5fpHaxeabx`EQypCRRT^#u_*JX$RRH5Cg;>1grI48(VAOk4sM
z-1SKw!ui@(hM-;h==bYY^W=M7$`H(WPX8(<37-V(jg}}^@-%B~!z_O(TTBft9SpdH
zk#{jwA0Y|;)Q1&)+ejRMNqETu{hF56ov%2m8laB~V0%=)3%n5Q_N9`Q(U7rWqMZSo
zkZ`V~&g(XbqR3c#Bwi&3ETS0d@_flCFIv=k>jQ->uEDHj)4XM!I6dvVEdJy7|KI||
z0IP8DY6Qo!S`Xt5Z9T}IfN$>B*8)@Tm2syOYc$Q~AU4^FuoHCu_WWi1Y42{OtLp6K
z&I)=M6^r?x`1GMnuqmZ%p+P_#`nSr3r2DEWIe%h@K{f~ZPE2^KxYb5hIv*OZ^5z7>
zPkT}Kz(D@l2hn#{2Yu=64fJthyj_uxVU$TE!AVI%7Ez4cj`nSXPiH|#lnbe!8D7IQ
zso@ZIGYq$cK21vxzbHS=@yb6z_rLS+|K8Z6*tkMop!CmAoPOz|5cm0@QodWGxyzF+
z0NCR7rNMK#x}HzRS5hFC+a0K1!nb(j+SaRPYED;56oZ&d^4^#go1!*Tr_Mwq{A??6
zJnW<!y$>RYkJ=aMp0p5hmVfkc<$rkU=7}$H=WIWjq6S55O#ge|WP+*t)lvIbd*oWL
zZpN{C=jMk7HY|dqGz5Zwen<CW(*J}z(@x{`K@!4mZIndZ)B1*%&h$<=t$e~Fgh*}2
z1{2-P%KGMP!Xix|2gy4JV0hpu;?Cv~5=&yR>^^Svb$6R}^X3QOoJOvsJTGkf7C?Ar
z^!3jOLyN)-e}3SK(uTzl)yo?mf7qD`fDE}9g1N&NLMA}UvFgY$n1CXPq(=cXH`yJ~
zT7tn7v(1wngFkL&Qp080D*Vfb1@bh$0T~JdQp88K{(Y<a2H=$@y6xSOnZ}aUyxRJR
znIvDxl|=A5JXVo4=p=xn5^^1Vl5qB1p>t=+MpmQq6Yg2N*ep>;ZBB5spi{@L!d=-E
zkPAR(1YK;z2O77BVe^I@)n`Rg+?aGjm3-CupT9M8IrctuHIoXR|66n^CG|Suf+|+T
z?gb?4VHXdieH%r6{N%tF#)d?P?hlLwY@CZzJ4U;Xe(UYI6HfqS7^7^4KQvonfj|~<
zC2m*I3u<~(YoG0B4>$uPP@e?-k4N=lRV_S>OFl!21@v?ta&Ui%);8UQl$kG@Xb&+V
z2+S2o%H<g~Y@VWbHutcH!Z(`xpBp$~4|kt5`@ug+s6Koi?dtyngeUE2uk3Om*V4ba
z@oQMz6A-5*{|U(1J{XT1`0YDyF^iYglGh)XOT3IUeD>8%dhFIH|K&ymVn#X+##`Ir
z^~>{Zb4m$yqYC<HItViuZ0`@Q)q}G8^ChN_Y4Nbk47JruXz2mj14R#0q6&RP{N^%#
zy4ZZyGWwMl2vTJGh#{XDWWjy5S<xk^n5~wJ+DcB;m=eY{o}#^6iIz{p5_~W&%0xdN
z4!!dWR8)Aj&FMa1w8`48{|fo(i9CFStjv(%x*!{m{UzKQY2ogA|E|G<qDX49#oM;-
zzi}_s%=4{ylH(3cdwLYQ>cm{t+IHv1zvG7DLcY@Jelm+a4!A1?xl`8mbR!{P0%8H&
zn);shzk3sl$mZFc7B6MNl{dC(09hW7qiCdJHpqR7oEgu|Fb$rlGepE3#{>HWn`F|X
z=x88^MNdnn!hD&uh7DpWKE$Nyh4n<YX*ta9hueG+bwV#CmBp17cxxFI<gjrsCAZ=m
z>?P`PM)0OXn1wOH&s(?+TheUEV$(D7QhHB|YFDHS%HmZh<@0PB=sBPMlGeVJ7%KdX
z?iyzZSK}q}FA)3tbVL7Uk?qGv!3lZHY)@}WS>&68{SX}YFnt!8dV&++&O@i;SM`Qd
z(M}eOa{p=7HI=%<S8tBKoq46OK?=P1Leodpkj@?|VnAJDiR>qZT7H<ZSnhq70u4l0
zgZ%_k@%?}MS}qVo4h1!sXI#(XUPRlI%>L&T7W<5^3;cADyN6WE6$@l_SBI*L;0oA5
zwbPk{c*oK~ps{s&byz}N+E)7%d}Yw&`Sz@r1k3s@Pu{l{OD!gf0#4RvQ6arNb`Yd;
z*T4=G*1mkOPMLkXbK?})KA`q2+RPKenLq019;I|iF+XxybcVuZG5c=RokUnXgC+0V
zmH%#c#uDlpetx%4(CiV7F@c8CX5*t?|D#Qpi>W4dTFC_ZD<u4)9Gi_+d2hRrwE129
z!~<K-cXoGGqDi#go4MXLtRfO~x4e5A8w3E@FQYSK8JfJ9<}|EWszomv0rzv*$HJ%#
zD4P63%_~+>7B}TGb8`NK*G<sxQ&hidyOVEfS}9j{6dE4bKwM-sR@xfJ3P#-_JrU~v
z@zOXO=i5>IOVutuzuz*izpfHg3y4}u_pI-`r9yn|65_QKt=!ZEQ(?5m+tV$7VT%4`
zH&yim>^FE|70uixj7U=wD&*1Os{RY!yISq7-3Q(lD$~p7uZ?@Ro4IPBBkCTzv602A
zjn@pj`RJrovU=7Ay^Csoc+T0m+XC(mBGcIx-!T+%G2^)Yv~}%x^(_@)$p01H>9|C<
zw5WAZ#Qc7BlzN`e^qzte1nzX-PhjK@?85}W3Y;Yt`)l*O_ujr4gQz1Qy!fB<YYE|=
zHT@uVwrP+TuBcX5*t-fUGEaY;MY#?mC~)1CZLLF}P$2OghWvxGKS#of(yPnGjK{*U
zDP)cyD9(%Q<ukOwJZbSe8M$hww6|JBo!))|!x)&)l`|>8vdI1Kw#V}H>TNX~h<*SL
zmZB#P=xvi)b5tIkxxGNBw}{m`at3mkQ23UN5j`>+hcH*|bq0#U=-j8DrSmki(XUer
zeIM@8n8#Tei@%txI<ci@QeYzPis&0K5P5dR)t4Hb2ljrq!(uLW{kMI<dNI4M<LIk#
zL)3|S`10Vf_$U3(E0cAE4esazsj5?%gsd^*Q@uXiT3iq}01ApXT7L`{STX5-(HUkl
z{ke&u6zyXGq-O>#3L*okF%uZn36jw*_f?{ET5`6l`MKRx)|c60yEok@D4Rt)wATyk
z7_AJJQD@|Xd5?ukcnr9pHUaqfOYCQXn7$$iy&%WM3R~>9y(G@fg8MNKiK58KhJQcM
zg64Uq4zd}cuhvrf1ulicEX1+1+X7QUCbRlg?1hjk>Z|(ixlP&!=C23>K?q&ScJI(k
z#dz_yD|<q})%lIr2fg~SOPLB6hB!Oy8XZDoyRjG41Q#9kvaeQXE8O6jGy4BS=|M&$
za!Le5Cr}~^@}%hR9X904uVE}SEH_>?=y8>A1Zd1SY~4DWhjb&iD*|R05CegZq>s&(
zqYLjt|MVO!?93!cdX)eI<K6|QjR~o%?2&Ogykl9a63gb%V`C^W!^~-!UI{G0baA-D
z-e~$%NIZctns%tN0nWyP^!cqtk&!$lJjFzal1Swjp@sqEoA;$3F!ORQ;UO|ZBTNtZ
ze))z`2Vql|JDMIYO%SA)*LS%n^L6t$0gjEf2V5vrj*16dkg2d1Lke+r3CpGzJIO5I
z0&Q#_hnR3p3L#r6113-$`07}`2UQ?CM<gaN`U|ZZ%kb<kQOWn4>#qq6uc-cuI^ozG
z#~vObZ}wT6ynpp2vIh6351tP^P{C9+uSi>wWbX6m_Q1II5rHG(TLDbv6}Sb`DlkjM
zu7W48sxAOB-#@aaHWoql6vQjExbrjyt45SK@VVX03)lAk`81n#)<B6lI4&1V6;`9w
zXwW)q%Q8k9H8?;lf^M#}=2b>m-4%Hm&U_C-ToO>C5G3<5vUk+LCVX|Z`@Kk`uG;>5
zLerYz{zX6(kao|d69lZnHR{+D4{ePtnzWdh@owS@8itp$cADNZzEe+yN8{qmQk&{|
zAqO4MQ+Xr9`Hc$i<qiRK4zh9@GS#U1RS-N}*&Y!dj{Nd2i~HM|M;bRGdbEdDKx|<$
zMonw7IDdJea|<iZeCPb&1c-ybsEJHmZUMLRG(=!L=m%k~-D^z~2UUd>23SOguNOQC
zJ)a2VGo5OO9@wx$*p^v6@wySB4?Lu}>S|=VhM%<1G2ERFI#<3fK_ba7rY@_Ft&_I@
zN{wQo46$5m&6r7>IqIrwY5Q5p_s8eo7eeqafs00l#^>i4A(RHs-JjghxnVuIIpV(Y
z$LHV=2c#NW%EdX=&g7eVM(b#HzDhnIvumy&ek>;1)pW!k$}--71{{7PERTkbPg>(b
zv5$ZJT_X?u62Q$T9P_u|VSWj^qI{t*ekxjFcm>z4RhlZwJkmD151Pu(t62E=iEFEq
zjG!DUCQ^n4rkwuUj1ZVG6^&|#-1V~%5EYU(YqywWJy+NuCpJH|2i2ObJ3#H<tg^T3
zy9692zFR+s<ayw^pf{0A=yGWsub%#;Yqmt2w-VEG6<DH4wKt{|y!uD|lQ>oO&!^?X
z{jBVSO3W4*j8@XCP*XA|#VsvrRM<PAoUgcZ*rVZ+@W=^0u&hs8Je9FHba{3aO+5|k
z4&U{HH^~{!=SlY+2S0Iv=h6^zS3N_lsYs@>wnSWN#B&?BS4JU`>ovUz3$rvsUfD~!
zo&z=cG~Kb#KPkG+0;N0<JhTfJP*!@o_u-to&r|sl`oZ>KHe~{p1Ty{Y8IO!0mUGO%
z*QJqRcdnAG2YWoN>LQlIm})}S!)`j5{>xNm@_&hvflj%7p2CYx%Ay=0lkz(A^xIRN
z&a|}`FDx}4q=SDqmI}V>r9U(v`x52NTo-Au_W8Mwbrn%2tPWZ^7iVr4ot-l|8-<>I
zNOT8CQW<jEtXU2SYreD_BCqPptNa;LkOo7DVs&&NQ3$k0d)t|O;vNOj_;<*AkrlhJ
z1^^;5%<XrM(>Z?_cgtRhe%hCvJVXa==gk@{T|94^9Dfa;*|@y?HjZ)SsH>k_?hC%P
zhH;m}-tFam@@~1g908k>DNAmft73R6F<eotPj!v98jSuuLP9w-cGmA}zJSN3lomCL
zZ~E~z*5yI!l|>wydn@O=PR-51{9yj34d%xM^&K^(mYj&AiHcATB5i;eJk3{RX58mP
zp17gkF};7O&B~26wxXEo1;|8TEP3AI3-IT6`{!VGJ7XOlKmqAA{Nnpn^qIujB5%!=
zY8FKA*n=xT`E{0+v0p{NJ|aM*$7rcoztM8Sieo?pg>Q+pwYesJC`p)j!tE3@9ptHl
zvTC}%o!1XJ8eYI@TADUp*Ahw!TX#-FXnddh{DJ<}QO@!-#AVgm{Y1xICJ7$bn%-^d
zSU<fEU8|YmQm2f?@G6^qI9i3BI|!f!{Q@djz~4iSsb}>Jlz@WH?-lCS)whJwYQ6;X
z2tjk5L0daTUZc5{|1HtTS<iv!TK2RH?=B4IGssyD#tJ8Amam_1?~uq6GJb-t1aS0j
z?#j%5;Ps$|2isf=zhnYKj=1hpTi7?uA4U$X=iLMlV<9dR&(E7;9sIXq!PmOunh*M~
zSxU}wehsq-FlyM=vp@ZXmf@RfNu>NSl#<!t=Z|L?s>z@!W(EL&#mdy-jk|rD>@ezD
zL@^e@<}+~^Y2dSOOpAxIpdfHg{Ah&@dIB)}KoK_Q&s|KeO9|H*e=;=CT6b!5NC8f#
z(}fqhtAGT8gKGcbu*yEf(!Z=3II>P{ta!(yuC%o>NUT%e0;{d%x+hMDa}B6V%0Axa
z0GN2>0ZF@yEX>fLvWDF1U$Lr6I^b?|4Hbt44<GnKZR^6wOAF38=UoB@UcRyRzD~wI
zx|BOGZ})pI7`*g@fL8z%KlKSl=PvZvH=cI+UoSn~vLVOyqRx_NL+Cn@eD<i;ILUkx
ztCMG0|K&yDRrcCzO=oN=vGf>IZyF;0mPSKKmE^uO5a4Ji-X%<?l5-8Eq_6?pKUxmS
zfs1D)&45XVl{`*P@i>9c)Zz1oR=o<&X5YGYX+~%ZRs0q*HHhUtG=%b^)8dHPy}bFT
z7eoV<6?{!|#6~TUF=ot7wBG9x>HN)T@l@9yd_2%)C$fPPe@QItrlS|v(qk<0TxTVs
zbN}W@D%lD2<C6=|ox+D$gzc6;o1p=*d2x~-QsHd7^eQC!CPezzix2|3PjYZ1T|}ZK
zgh<a5iOd-%AQ89&+(4~B;oYy{P+VPAA$?{eqh{gAn+zOD3xV*`Col*KWv`C8rGBh<
z$Bb9#8Vx4?Fq$p))STr5E$rc{$!Y}>mgJ9us~-fyZ27|u*>cf~YM^?;Fd1P(+eVYe
z>f>GVx#w|Q?PyJ-#<ojUmRUvGVqr3zX+#nc?n?(sIzuq8oJhDw09JkjqR=GVe4;z{
zzOyMBq+uHx+Q9b=`!@1eV9~&cjt_qZ+Ql8eKM7WFxS<6(3z1CyaVCx?ue%vmF$_(z
zx=<gO!XS=yqM0AyyZnnj^oJ+!hB|ZR;;Ez4hI)i_{pP+S9A$2!tn-Jx#!WW<fTKpV
zUEQc;W*LL1;fk)-$rY0{UEJTd#+b#AZcFzGdfp=Dhhjxq9qMDn6$^?!rH%YF@gztx
z8jo5k+1w;Fk_2DYShNU;;nTH2QnCIX6q=|6{R1QjZqf&cKe7>VhJ)h0ouPi)(E7ei
z1H?^^ru!zjoEcgl-db^GD+fD}qj7mx6srk2r!rRx;wmGmuY|Yo%_D-($}mLv*2d=b
zK}!wtmEX5g!;I-MX$+qKMq87SQt>sz>SMV1uts3!dfN*z{ue&ab36Wf_^=^Gt`~Bl
zIPB~Ch;J|)cUviW5r`q{#1av&cBy~eqJfUFH}NA8VHG@U@Bf~)imI^@IwYO|lpRxQ
z^gH5vCMC!WK=C#48P6UEPj=LhR355G!#o%*l@x=!oD<7t#mp3|ewt)_u_2=Xkl*T1
z)Y>;wD?efH?1=xa!79`g*4eohebpAR%ol&Yss8&LWfOwiiwBizSu9NHN@NY}0KYJy
zGd<*AOAh2LBXS~Ks@|+X3J=Hvfc`b1#b5to5e6s(vosyK(?z|=yZ33|z}*<0*&<dy
zONb`$>FCyje~<x{=o4HGJ7HzC%q%e03~?XG;vThg`Tlmbb!#RxzL}&KtfHR}X@ITM
zG`SOJr?xpcbo)_a5(Qv|VW-!%Iy_rn&^4*Ga{`Cvd0g~pbx6N}-FuHmk~lKpFoUX{
zB)s}Yt>?I@e!Nf!=Pt2(g(bKf(<VS&b457!ShS}#+MJ*aoi75w_u00a$agCh<wyG(
zuForeH}-MA^I3M3w6Jt)NFL-6tUv-4Ia@J=EkSfgnb?ZJi#Z6?tJYOR!FG@6w(D(B
zP%^jSDZ8GlMb~g-qJ<IJ^TEqw;*E=V@&zT`+VJ_3=K&n5`AeeT|5X9sjhH)E?Y*^2
zFTfM<J;5LTkWE+UFkTowzF}x!hkK*eGpR)12RcQp|Bx=d_Th(*AIB?iK*fh#w9m-O
z5A{N#DP}Ep)w+rrYFi(}=5ro>#gH<5W}nDo3NJSZRcdC)6Zy5FEI8kdnXRK-eG|Kk
z{yn7iL_K2NK_G#ly2Bk%2qn``uHvz@%CI8B|7h^}g~NhjH)&g0$F5@pNOn}_SAu(3
zK?C|OaTJ!fE{}6o-nH>%ZaOLYx8xiQuEe-OA~1rkt=Z>BL<bi2vTYx4&QKN>3{_VX
z)U|&FP$5H_h(z6d)LfX4f!Imou(_Jf?j<>;_uO|d2EZ^eSTAd{2#Gd0iXg23BR*}v
zuAyM0JU-z+P;uO$17YN}6ATIQS6T@eb6vV_N*wGF%o<#Oklh!R8L)k&chJ{@G!X^v
z-yiJ~Oz);O*PuJ)^ksS%cemO9*d1+aS^hM)^F2g^UVoft?c~D8|4s08wP-0tRGEkY
zAOMLs@#1;-Afy(o8R@<0^zsVnfXbG~%9iaKpIz@^)V$b4FV>K62?YPgY_05vY+6Jz
zUpgz(gcsKEf_2rjpV+Km1#>6wUCw}t2d(9QJ5U<N?f<>h{y3Uu5t|w`I@`imR<!(?
zWf%^5&RTZM4f<C_l<&iQZ7GNU5qyXLF@eR!2d9sEUk_qMt=>esd$oT2u>v_N+x3I#
z>1@U~k8=9^6g4hjtZAy_&tMT&;b{`IhYbc1S0<Y1yqd{+X8I6QmfVpmS9GR;n^S8%
zgr8ugXzAu3hL>dZUH-ecce$nGFh=~U7OWu`*;En(tJ{=-FvC_a<$iX+DU3_GFLOCx
ztfOAvpb|rH?Auf1vSvcUz#s4zNiuSn3=rr+90YXvdc@g7@Y_Ix8hZxi38aiAnE#w9
zEqE2gWm1l4Ew0YD^oEOq)HgF_b(F%(Bmah71$8v^zKb3=3i0BY=`pJ<1s#20FwbO<
zx?DJ5g*d@)Feheb;XGPkk9w_%w}4$O_ac`M=xe#2{YJxW=*l)51Gv47={F9WPUk86
zKB1bP;J**j(n{Y=f3<ZAQ)%X!B-j1j$?d3j;q7#hZ9^mZLXOp~*ZL2px52)QmF^M!
zI#Cf}1wSyJ1U@ix!1l;(gM?i2kRS?TEwfmaf*qCuJ5HkmER&vSL;@e^&Y`+n#nJv;
zi)38=IJ7#5#aYe^qkD;l<wri}pasyDaqjG1{-=jPOQlx?qNSVrc3sY3nT_2L1mC-=
zRhbApCtT1gy}~D&OibWSKL3w!BM8AI5)dJ0VuURXz{`RF_+ywjZ0f3-uSf8narCU0
zKg&H<=h_2oSq%q~8v!W_v!<KLn!=EE4sM&z&WYf4uYblb;bvls{l*980jq|dg+FyJ
zANX&Yd4Tc%O4<)Y_ww&@#!@!ikS|aDEg0qNmiLlvM0^%|iK&DQuixaf%I7m{v4Q>a
zVc?R#pIkQ;Ut*gD`V$yZbQ+pJ2%x`@3>SaS#8@qwYx2#A(P8NpO@lRBbG>_N_43en
z4_`AW_z_Xeem#;-M7-vy6#_$-_l8M}_^g4(1trB}u=Gz>e3j`oGCIFFCqjbv&-KX?
z94uTeM>NQ31|ZSK`(=%aQ)jR>Cb^q17R=;15N2G}8lcquHjz7C{!YeRj%wX9wPKZS
z`^i5Af{2B|V<sw3S{p7MkH0e3aqa%d{g>FF@_?Kd$q=?aNtuOXh%tf)g)qwl2x1s-
z5=1=7<NI|e<QGB}zOydH9I+L4NMZc4ud{CS`w!z<zD|_%Qbr`LqU|d-TTYo|2KZP6
zBpkM`AhoaylB91367U(4#^%={dt59e>Lbr<qt4I$uO_^Ij!%?GF217>fY$&B0&zm2
z?|OZHuL#~X?QlJN^zu^A2YqI9W`M9WbK4<jf&A51dNs9zF7Yg)y<UC$$|TT0LjHN$
zU?x~I86W6?eZZtiykYCz0dE2VbrBC-0>~yHT7|>D{_M1*Pb}(USO3RauFvXifl^<|
z&dm!QD*crSvy7>cb;vyYGh)m__AsxEsRp*Z8|MqHs0-EUT$J<Q#>~3LdFSl1_GIBv
zBEzxo!&A@sg5eKrFDB<RQCRb6HzhUbnVc9ZdY3EQ1k^wUHPS`lNtP#=ecZc>;s|o0
zm)&u8(0Rywd<Xa1UF<KhUx^0}RZdjSclMv?CC`iR{vT6c9Tnx;#jBKbcgGOY-60*4
zQi60!r*um<gLDZ92nYxW(jwsu(o&K`hcrkd+=uhsweJ1roVAXu;|%Zn>}T&^?Sm9U
zQ)^!GA;D*kyl^cAWnaT@LqT=y=4W3+HgCAd*(2ybYxBQ!EUVc#Tg2QT8+GGL(>MBL
zlg+&iK9~`ysD$Z4qaMaub`*4AW#CJ}UKGJA)0l$r=j8p~oJF-X_%Ob7|AKZhH3bYo
zWm?VD+AG6G@k0q=RbtjpZ=8Qd%TdFE&KFpu=9TeRj%+epOX}MpOhl}eY4t2vHZ~MI
zi4NGNC@4QEGUSZxmZMu_hk^!v3t-XsGt55-qxjQQ8$;b@YdWY!RFXcm`#9PJdy6M<
z#yh-F1u;~SFU?}<6d@RJ=re=#r}lccY%GbZQdDoV1@rvx6&weCc_BFllj-2cu1+Sp
zwY?W8&^moPBO&b7FNg|^kG0NtJ~(>15>cO6h?Fp7iPNWu)zB84jV0gICXszGkUW&s
zlYLZ(*tWhQxS-=~obpXYIa*S<HzZxYd%H2;ru8F{LSPeS%UFFJ*Vu?%5SGZo5MQ@g
zOmVtNfz>?^GIH=0_(HFb2pbh?<TG=)D27O}yL>!A`ib=>$$lHz*XyBmph3q2zi0)I
z9`EnV2<dt?&FB4ypaF|cYu%q8!AAi@^r58WcW>D(PnYidqb3f?fKpWOU9a4hyi)H^
z7N7h+05EnVSoS_<6L(cIV1U>RM<!{=RUgGbU(43YFJMR}sy)CScuh#5u{t*r=1X10
zyFfxS;h$tYy0NZaI9f1o!(Qic(Zw${Ut#}*I(LXLl5fT*8otUwCUNc4tu$k|{6W79
z>@!|Q&5rN6iyxS6Me@PdrD}QBV?1AO!4`D;7aQaA!j~}!!3^HjU^Nb!W*7v6Iv?(C
zC9LJNzJ^B7#z@f5*8pKmsQlhQ7e(IiE>|>x5zK1{45LrkHbVZs6P!(ce0p&?5#Q5h
zvK`u+@*5U?ee6&5%1<ffdDeeTgJtdgX3!7&vsOwB)szbDubRxr<)OUZK;cg|k0%D0
z9p^uymVMnge8d<YUjwqIAmt~OJnL|*Rx>*2lr4|||K6LEY@GX!!?c-zX1{r&;L#@A
zHQqEas-NvTmb4J2X0Rcp%&u;OGc8VvXfcUVSUw0=#MR@+?tBEpqws$`kjERwP0pY3
z>8=!yTEK27`<txin%H`gsP$t%Dv;^CHR8gXGocdZX|OTU%4qIa{39?~bI+)}&!Z=5
zn6?T+h@1-n-|4DMV(YRImW9z1D-Z)VsSlrSCy|0Dan_u~Snh>yq_J5Px+xj@CWNDg
zKMv{s>g~<tFdlXJ>ne~Jd(xT5bzs8An1BQx&KD)xZ|7qnl=5^&#5?nkeYwh~;8=N`
zrq&kWb)7A}tcTev0l|?h3F~iiHrWb{f^O}IbRv<bIGtSbf^FaC*SdE~2lZPi6SIQW
zsHqF2uVCk=ZjLd<TUUt{fFFC}JO4!_cY5h0k})o5yN|eyai2ov!nuewa~{0p4VCVW
zUETGSoYgKd8O<8wsRYo!tTJyzf&nYZUnfv7K@qURL8_&vs{S=@KDdX?lD*FBTt9k(
z=RTme`BxN{A|ZH8kHIo1!Y~wImf)|@6(sQ%3qDBRSVN_)0-8YJ80tNii#CJXXl`Aq
zCnvI`PeI;kzu!VpP<2E%<t0X85VGwup&Dxn=CA5&Dqt9`d1J(ux6uBw2U9LHmo#PR
zwGO{tITmFmUBi+`>}q5_w}(ZK0x^n#u`r@_P9dFnDfH7%8v__7<6JWc%fRT3tR*^O
zAq^7~FQ6Ko1xMO$Xg;<6Q3p=#KKA3IlF94!rC{blvwdE3{_Lk6#941@q=Bw58GN|S
zGI3P-O(|v8%al=<&*M!%p!g=0Nms1Pw=s$nbMLPyizg-hQL9(pL;bsRHJfQqQ9nZ5
zFHY=L+ki{iyiFTu7(2F9GBz?5m7c#U488AC6b|Hm!lLdftYq`W`)Xx%nx4cJ+PzhT
zSbfXNI=d}O9<PGWoY)a7;I$Kd>^QOdTb;XrVX*4c8)_CWO8c!}s2C!#EwkjmsQn`?
z)t%NS>Vwsydx{;=<TH%ea{P2aC;3<>pB~8)$@h>$XV~#mvkNetfvnzt8@N9B9?s&h
z;G2YXp^UTSVXRlkCvAz9<x?8E8F5f=f1C36=@YiiH{6E1y`B9a%69fodCfeu(*5Ig
zuZt0-e3qA!sS<c-2Boh*(E6jlS0hA~FScM`7o?DgeU01M*|D0)QXGr4yC|3=Pa4%p
zF&cvlyfFG65!v9?OGGNi62IY~P~dC20h=<joE!>|YH(f+;{2jxZ=l>ygZUsH-Xl9<
zEf|KMYg2Dj4Q(5QKV#-4n?0XCNG;g{xqRJNVp12srWd-OOFh-lmm}08eDC(H_39fQ
zb#>RyjrCAApY@N9vxmyut;Y<7Sc<#3Z^@-i`Luzg;<fZZnDYgQPAhsl#fdQA(H0~r
z#}wD%`RX6gI()xV3N!j>^U^i>rArc=3e!vREzIr9<%Dup{l71orz5>2cS|7u|8kwU
zjq~B00?R-YG@SlJ&BV=l*5#jmngcV=@g`&twrAvEbZI%Hu47;3;$>GG#)uL0@0Vs<
zF~<j00$E~fM}GTF%-<pKHIZC(GC>GKQ}$Ea$<F@d<c-FuH{|}(!K@#go+#)HIgc+}
z2+6<u+`dTd71F2-?Y1#sj!hSIt#N)qSB~6S&6R-lDi~#%CWd_|16$mnTN7bD+c>a;
znUVMt94UjxRk;({KP$NYo+THI{LH{U$mbcOzA{<(!xV?qsQ4CH#zvkq*2*B4l*}o(
zN7S;}f;VGk+%~SH68uus^_PeofJ+=o2A#5rkB(qt;}qgDmd8^ihVSs_a)R<iCoQId
zH2H)(Z_)eJ)sTrLPlk2%)>zZONjbZjHbp}ESVB<Np^#-i0w^25xv=|fY(mrH6ce{p
zed5iZ;&ceZ5*xNpZ)k)iK+stXk<!AqZ$rh7nVB0pbbja&;NpYtU`Dey%>5-$6)8io
zAsuzC>B{^l;0HyS&w8mU5p~%GpBY0;eq$X#5e2THh55sWv+>xAf>d8E*ud{r6R`nm
zd+m&}WaF=uZ~u;TNZBc(HM<SXe%rOZub+K$`7LyVU7}sAiPiq(c6yX%Yh69LmS&8)
zJi04}MS%KaHu%1q&(UI78YLT{u6LT7LdoizpqM%HKw=3y=vVek9_tYM*_FvFY1nUc
zef#wtiwQIOXHF{;($Nf=FB=7W8zU1slKm!@ik#!CMgjNw&eU@_Sg*^UJ5*mu-NwF(
zR*C*MBD=z{J@!MUpk+R0GLe87r_6di6<4kE#Y@++0k$GZ0ePVc&o}mcjo`$&DhSni
zI&y1`Yaf?n=d%dt3c?*v%pmyPasvu%sn?i3G*x2zB<63y@l5w^vaMgBOzY#-Ro^!(
zKe;x0??+t0^=B$GiAsqWCu$&HX(<)lna6)UV?<T6;g$KhT<<vu_tTV>vQ!B_o|#sO
zB@qeumxvZS-Iy#=@Sey)$s<IqS{<x!7#dLvdea?xRDhyxxBYd4f^3$<EG>_#m`b4e
z#9z@toYrOmxeNuu7YX^jx<<rLP0uN=2u>9XBuL}4;}u}%My|jw(eBMf#-kRCh`cXi
zIE|mo1m`u8;`H2-^N|~}5wVzP+HBH$3~iFMR@1yp`uNrcsb_?!xa!%rS}8S0X+-Hk
zTaIs&_fPTYUao5kDdQ{M@*rN7^ZmWNJ$?1*3n#r$;N5---4@wFHu7?P9+xvu61xO{
zfowWNJb84r)?rg^aMd^%&a=Qbm{utXnjyZUm#)SOoppWRO&hnSrelo}D_f%@o;A`R
zZ2i?YUewdaA*CQxw(hLDwV5@CH+9Xw6)VHeDa0QenU?44WOJ_*L{7%rGL~B25Tqbp
zTRQSzaA!VK`MuZ1`40PMM7IdfzPo?B77RBV?s)qL8H*S@qat%&vUQ;ck0@Ew-hbNy
z6(1ipe4s;<xs(xojPh#1`NRA#EavnsqR*3*X+YKde5VBF<l(gTRoZ>_k9c2q+Q8(a
z_SfuzK0;rsnslJr10ShB*&X!=CR~i~YUI;fHNSSh4}X!n4eh>4q6K;Mo=WrccW)wM
z&*Ap1Vm)})5S?yq2(hX~jnjB@5U#on8#!i;PIv!Y7_`#Ug0a_Ib4v+qeceP?4T~s-
zjzicfGO`6_%b_|LtFiV5?eG$~w=7KC7FS;Yyh`Cz4NLWYdrIk|ny4<sF!gmomPU4S
zQS(4uw6Szl7s-2(hXd=T`vd6+0EOi~G~lo^VhoX*JBpYs(h|blgIHkZ(&lU1HkmPX
zLLNQYPw+I{54-qvh))rr8<>c;Ul!F{k3c~RIj$9x{<l)P?u`mdxg41imk7?ff)0R^
z0UDY4;28I8-dE$lw?!A6vHdMVqnLCZy3eNy1Pk}~<UF))&zy@YG9_+%3K~IZ$?2}?
z=5v;d10GNZt6dVBk~N6?sdIjMI0}#U*GP*{k;ZfKrfYH7GV7ueqPfC@#-DX>wa?_L
z+vovb1ff1@s}5b-IE1B<647R|W#<&TrL5P8M%U_=+fjQ|P>mxzCmhY~abXn~%uK|i
z$(BPofeNCDCQ#G6<b#)Wgh|RzC~JyG17LPl?>+1*FLO8hLdbRLT0kk47x^Hq!r^p5
z!*)2lytrym?bDXZ?Ht)Y3q^F8*l(i^WnnM9v5EMXAqf3we<Rts0N#30zI1!@{txy~
z7Mv5^Z_$ki!P67~e<2}gXSYPIo!`*l8VKa)53EJ=d2qU4Sfzbp_p+>p`PK;?MWD;4
zvDpjaYT9N0bP!aQP6}@))4HhPc%xR{^Y+QlAz}?ei}eFLfk1*P#?%=SFu3g(KL4)r
zY7R=jd}w^qL2-bPIlrfR=##&4xEVdZgoWP1Zto87r6nXJ7Cc}yw<~ma3>S#ddnekP
ztI^Ox#8m-eWpg+WMYcPQN%@lOe%1Ua<!)T&no8^|z1qh!6I>0Om-9`o;gja0cuXv(
zl%NH6K6myENG_yt9}Oow6qo#d_9Hd)ZfmQxVgBbNN!tBt!IR6$#Z#4%uv(Rck9gqG
zcvcE8cPdxs=MSe(khEpOK<Z=f6epDnm2R50O`0!tA$>AYp($|T5@6Y;m+Cs~j$ko$
z`L~YTq-d245Sq7qmi?~h4_k>_WNQtB)e5=yX7{EWX^xWkc~;6ht=HFCl{|GIVP_aV
zFW}k+3zEOq3BP2cxV@zj_y~nN&X+i$f`6w!cbRS%E-W%AqLnmaVOTi^y@FKMY{W=l
zis4iHi9fT><G%;}|MtX|Z30KEw%{++#9AV40#r~BF81D-jd0y$7Ievv;vSWXYIKd+
zNP9ZQdnyI?9|ucVv<j8v>xhZxY$#dx&6qFPZAT4};4B3Sk_1B`tCs{+i+Zm`qRsyB
zy=F71OX1avi%a}zHSmmBu}AdB38TR)VB1NcQhIpXw}J>n;w$ZhKyRJ&BX(5k>i}4e
zna2)#o_|CX<qecgvXn+&zqEHZF8aQ7HF}4gaY0wZpEkkwPEf4>oUw!LOmS$3f4_M2
z>>6J_EeBral^^k_oTXHOFZtf?Japx7&!8CgzaBf~>xW0TL-x|W6*B)E?*3}szkAr|
zBy;&zX1SE^AKOf=skNu)2?#LC8oc!d`=4OPYcL)C44%-d!P_5hw`d@RA?Ppo$+#kO
zC<IJXUai8viHuDLE5!b$G82SHj>>!;fQ_o39-WaMN}4d(v)v-PymLki=OK?39^c({
zV%>hF6IT<eI9>?Iax_@m$q;U;m4{(oPCX&Z;mCb9XMA_i!~*WMrsV#C9m%ZbCmlEL
zZ@xy&Gsaq-MEB%e35IYd6LkePecx%7)b%eoG^t#><_kS6boj97GFjL8If<@c-9|`r
zWc_L`VIG`O7Eg8}F%k{vYa{8Yd^O^J$)HY9r=qnMg<`5yz$6Cn003k7u6s4+81#b4
zf%SX~tVl~iIP-Z<oFBI1Y4N@i%;w3{Yn(Gv9GN>~^u9&cARu_elbFLrPAz0UfA6r3
z24)T)G;WW4`uLuut>vYE8;0vMEUI#W>t6NKFuV!t5x2j;|A<${ERC~2Vfm$2M5Z)(
z&HX%yp($ieT9`O4m<SEkl*R2sYriW#13J=x8M{g%of1T~$3~O2^!Y~+ofo!MPO_N1
zp*!%Jy4H?QpF%!8b}e!vy5GxbvGb~HbIf`@W@sJfZXEgMC0ou***@cWtJV$rI7TC(
zrq$VP?8>u+F9jyU;G1YO6Il9ww^>J#Y?@>YB0>)1cSqc<=C$)ZfGT!FS~y_klA9I#
zwp+}@TXA#Ho+5KyDuYCyY)wi^dO3`^)d<=B8nTNc1zYq*46{lfW89s6yPJ_Y`TgDE
zS+Vwe#9{jA7GmsFB!(7QWLaA%fl2|ixap7lO*eT87nl3glYfk{Rt!phU0SQ=abAnw
zLEC)$Wbk&^Ty!|i^!Rem(VnN@Y8fp*d!Ur#S7A$}w`?zloPy4>X-swEf_L=Zn!TOh
zsvQ<<tf*+3EJ8>e?eb2IiYL+kB8`2<7cJ;2$|d9!Sy|gBou{y6cJ)O~3LEA_FAJ{I
z3+kL@aG`)!n_g{g<qvY!e%Ij20H?iAjYV^tFv5JlB)*adk&q`P?&E09uRG?;rSf9N
zMn(-(<NN}ihET>+%vYH}mpQHxRRqC=x0AA_i~SRlx^Zpt6KMwuQpV?u8H;~$bnt(h
z52Q}ma?02&M<*_%hOadW9N~vC=~6h$*wwh`2<jb}aJ~@Kr<M`2>e8bcqrHL?_VL*W
z=B!PK2`b9rqp0gM)^V(RQ<#!EBnN@F3Yt3i)7iVAQ`lKEV_W?3ycC&iWFk8OD@Fvp
zRTUv*jMD#7C$U%!41AG#+4h1trMhAeS&K@clBEk3l`$Iv8TF%HbeZ_$-t4QA?_Ke#
z>g-{?Y~K?Y10DuY5PT;2ZF=pd+H`3f*L~of|DE~WdZ=Fjhlj+)zgHPwCz>9xwxr{=
zr3<%K%=d>k`4Ya#h`W8n(oHs7e9*Q$-Rb~q(QR6a7BuoOXiMT@IS1mV6LCW&2J@SV
zULT2K^vM74oGzjpj@SBSs=~PnMYgB0+O+H2oaJ`(v)tJm^A0bW`{g@Do2Nn!As*GR
zje^JNMg^@UBTrJaPOMzI<8p?Ub_>i8>nEw}-Fg7oR-p|^`_}cP7m%c<eZ!>9bjKsT
zL?(53VxJ{0d>i>bNFQCcPE1Tb!3}-W^&|A`;BDZlTc`E$o?}e5xAoU)F8~MG`(G+K
z&Gd3xZAUu$xZrBp`7c{Uv1z_^Ffysa(Y6gA$p1OP;qnJLC7*@QTsRelfuyIZ&kkYR
zW^r`DL<A!E4TJ6|<x{4boAM4AUJUOHktWUzttG8RW^U9^y?Omgx+K-OhdP^gXHI%Q
zO>01%X+QH7lY;DT0SnSd6jkDu`)4`H8(wehRlaNcA?m&Dj<3$RN9!)gTNyGw{oJel
z_I^VV9P=EWymg;WbhlyuxsVoLZcFOM>4w=oG3qKiKk&9Qy88!q1Jwsam7d^y?nXfD
zt%3>%Snvy8)LsRH=kwF;PuY>inuEIhx}qVNHk00I3iygBC=IzPBFRs8DZQ9!jEIj*
zBh$#h&$RjE=KgAwY!0A>KC<6DmF%1g?maw~*FROJ6Z~(&3vu)x#OOAt2BC)P8lAsF
zogE$Hs}1Y!#$abR!JC_#N0NA|&`Xi~O~la(Ug)Jz%9$kI=;`^LG&%Otq?qd&^=3R8
zlnvltC=8Zmw$@dI$gCEPTSqcdU2>ziVS1l#JKMLet|aU}ZoY|NTUj;8`Y~22?QX<I
zxzPSiPAH2de2DZ+N+Nw~GnOYGu`Z>|;&)j86jX(s4dEv&{NSO1A?$*#XnoZnw9e``
zWq`@=?c@3h0qp#+Z$TjmAlP;>Pyb6t>Zo*k@)`_)t^7GNEl$OKM9*EY6&cZqgadKG
z|EA*xjv~dn{L4V*>|X*AXq*8KVRZ14JbuVT)}FCKSD3`gqREyoZQ?QpkCQ(mbb$QE
z05U(Ul#9u4Iwr=SQ<kq|?ltedjRv~B<^YF3ir-U>*vdw~E_`V*q2e_Z%${FF8QD28
zds8p=_vPVlJ%08o;!`m5y-->Kn8DwIutPJa&kN*0>kjaUp-gO*_YVG?#5klr!|&V(
zU>Hs=33<Z=!4mYERvo-7@hg+g$rrRWCx6xQLNEF-W#4HF>Vr=4PAz(^QH(!+2;2Z~
zW`b@eox|huHG!VKlh3NgWt{e!jc{EfD!*7;(AGxv!|RX48mRh&vwy_Uk*ITq9w9M_
zl-L7Fi2}3_#Y4_(`Bx)ye0fhhuKwq@vAgX86>jeUZ+3IvnQ`UuEhg#V!g!_mXo<Vk
zB>8k}RDd$+^8D+!;fKBgmsS#fvvko#CwlH1df|Izc>l+I@!V;c*c<$<((VqpQqyy!
zPuMBC-q^<QbbXfh$_}v}+2yAb6YC!>$k`kUO4~1#w+7*w4Nr6UDKttfJ_a5230CEg
zyldi0X(xUDXz#Df-Fh1H!yG~@`OekG$E66wE-15lDL^Fs-pg5UzYrwB7D|HLEWJqZ
zrOxffv+fG4`TjkQlXdIYi8~Z>8%p~yrHYBAZ!MW3uYiS{$gqmn>?YNC{>#rOR0&%<
zq(Blirwmw-TKLMvt4W&C9!^~}B2X|h@Q1&FevI5dHllkJA(nUjcYpYYqRA}n=cRh^
zN3e=Ub+W{nzk8>>+526+XV<C88%*&?il)UjRcz2_!OKv3_lqt<-7C7Gk2HG$yj9RJ
z&hCU)!D{70^*fw-rZI;mE*Aa-=op96#sExlObu>KGr&EZH&n_{TotukMwxsvQoD`C
zr%#}p{?wK^_g^d<(jeYt^{EiOWhirRd$Kw|V39cU-l<0o*Y;vj1v|^Eb}K|b!F>lL
zZKI#OU%I{m30#oyorn>u{_aTtCjAUjt*~H+s2by#IKHfG((doWvG%#9W#UmFMOApb
z3MoE}6qN<Qf(~<9IF)Co<v4Y#3BunCY2|H&J!10M`p)P7y$r5sV;8~d6YGsd|2RW0
z4(}E_Q|_<p?#02{P@bR_EVpg{yJML<n6tC<(M;&S8JQ!uy0eMXo}Qk(Kcy)P-~hPb
z_hE#PpB@&O0Ew2dEubf2aNk{ZAFJYO*ojFOJkJtbowPqVvH@L0nb3*f==h%{yIoKP
z`x{g>4FTP7n|H{4JkTi6vF6Txw<@|KFzTw!|IYM#DnqF%C==k&s}QpWT{x!Cmi%xY
zU%yJ?4whKAuBhOf^qyTtgt$dv<w+eZnr{}wu!_DMZkTTN^JyAXP+AGF?z>6Wr)VJn
z4f%C9==Qr4@`7CjIYE`EAj5ozo%AVla}9Sf%owq;7ZkoWaRu<hY53Z0Kne(#dJo;$
z_VyX|z9LnH?s0{IcSV~n0AAAF)iChgF>@e4x%@*$T{zL_m%^1c=B-m$m4E~-=h_fe
z{fH;&EoRE%U_?uyAxEfa%v5AOAvQ8XhF%(3Q}}jYnKKgyiV~pN-Hid(eq@pn_G7HL
z1_~vDEwdfG8d|D^FFC2y$-Ge*3Ka9XU;hqqWNn;O@5HJ&p-iT0obdg<Cx%_{Nav9v
z^tbu2yEi|~#*Dd?y~Q$dBdSj|g0p;wLzbB7InusPE_42b4Y5%-4kM?C)jE*Gr8I8{
zc*_eKSyjCs-wSj;qp{JkV5~m5+WP9xB+|}+Wda()i{^_*w+@AQH>yE+4@0Szh+8|w
zGktxq?>X?HOmwn7S`*!zN8AbD?-l)P6uFi3p#HW>wLUm5h@){MOK_}2(fZgt#Dhu^
z|8n8<m1O-S+A~6wkd+Z509iMpCQY;-I~4`GCX^H(by@2#@Rp0I?JKC2x3S)}Wu|b(
zuX{)3oiT*o{lk&=>j3}p5p9<N!26Olvr-k~Je4+94T^fxcNP!;%fr^UDYtO94Gg1W
zS!jCMr_GG!+HJ@)36x<Q50h(wRxlEkd=@~70KoEmUVbUq21!IkRMheWFbrVGo!u5!
zU-y>wbB>58*5TR)*Yi<Cto=*?csSB;MkV7RgfPJDDVcma8;7I`oT8Gdv6N)za^o$A
z6;tth%NsLg6?8p^oeX^BkEv(~TJD>yMbJm;3<bc-w@~&tD$HW*SmKEc=1G&nm7#q=
zxtd8w@MDLu|9c#^zwQbcJ6Ec{KS|d7<$(lfRUOD<e_yJA8kY*LWu3OOd0+Wk`to<G
z^DSPA1x>Y4Z0zd0m2Wg0Pe3-q41hqL&KS&FwZZJLE<^#gB(*YRl9hY;hbWjCW)JNE
z6f2vHyiZk*oHddU(rshL1ooBeWn@Oxs^~SaD=OMHrv@N5bK!}^*5A@El7B@OR9&|a
z>OHQc)~fTlJjjqXGJW8k9W_hGeV0-%k+b26IPVdEZd(w&{}H%ctdAZ`|J%_zeOT(Y
z5O~<R5PJXb_B-9}<4oIQs&B_sk8bDc{!w~)dEFiy-f^ck%O7-JnBmfhM{fBComHD}
zAntEMM~iL;if)tg5vNpA4-G#D&jJ3vu@gX8j(Mbm5}lGB$3r}A_(%Ka9+{Fp%;|S;
zNor?eh;cF|8vVJQ&s&{}(Ld;a-pK6VnUx%^^YvRdPPoB)=QmnEc>;dzg|7}9rr!8p
ztgvh113*gIX}x%W(cg6C&qx@+KxVAc$}$!6pD~v<X!Ezss+>mCeH&lxWvu{d{?tXP
zy7IK&#^s2YLG1we&AUcQX5E#PZ(aq5b>2BY9XfX5D2<Bg0L?hpvWYFpR<0G%fuw2W
z=de_BU@wLp)_pN@li+-ZBkdhT=<_wV?g0C^2VDnW8xmejcHq*){SIX6J9~<pF_F+%
z5`&?{%Zp5Hv#|JVP!$>pCYu*t3it}n`<Mgy++L2&-CgkmJ_y@ZTRua<hwfM~hL+41
zO|OxCWAWUX80Jk~WISWLuCyKBEjPOiYTs|efiY_TpwkxxpyUG}ZoKMeHx7Y{;*G(m
zUB#iLKLZfE=FEVyq0LVW7{eFVeQ>;-A%|1~wjA~ZlJt^FD5B|Inqv5mv9*BUZmg{v
zKg*%<9ZpvJ)0e;xiTN7$pErlN>hC|Zp9Eb1M9N>`;R7h9ldGj+;b2ya<`2hS_I7Nx
zdc>+=p@B#J`HxrK);C!XxU~PxX(n;l!%{%G#@TiAyTQ<bkr7<qC4W})`|FttQu*nb
z8G`<-O&I6kYSC?aXwT**OSGl??WM<EZ|Ei7?a#W~pIWyIU2j@P?!o0y;brro=Wxg*
z9JGj*oxAubd~KYk3rZ<PG7|FE>pH`gCQF)9$c?!}YJ*nAHhWEKVA2Cmakhy&%L)0w
znA6jzE0(d$D?H07Sez;4-xPMR$a3-(BX}Uwnj&ew*SI2o<Uzd#>!TE8pJk$eE_h4i
zW?e9?&U9h3K$$fgz5|eM1q6nlXRC(S`syX$p|s@p;~dEn`lAXw#B!RG#=)%NYA~!^
zRP5FJ+zIrOY}TxKPy-I7U)VBM)Q{I+o=>3#vuccp3gxk?5l)MrA6P}pF_(h6Qo9#F
z30T6zJHm7^oGV+eFz}&f-+(1yr_67WiOObfDS?Uvv_11udBcSXYs&dKTqonYmETs3
z=fSe|l#W_}P6pI86%IP{3F&lwk?KB|Q*!NwrvM^InWd3BPt~4Mb@@dN?EZ3_P?p@w
zg7Iw?I>7s<C74P?zhe=Vh{`@X*RT<l&S17@bN=E|yNB)XQ}f&)T~xS0nGmI+Z3qY&
zq89&@uf=?7asiA@v4GmvRRqM6u^I|N3YGwrn@U93{*%DGADLv(30qRNN>$&^!xKQ=
zKLUuX$9)%}CjNvka_v4y27U!^IKtd5pzBApI<5G)CjF1W(CvHWigqM}>Q<i8HQVHm
zIFwO;T6zN3+o;H07dhoaoG)A;I_rJvg*SM50G*kPp#`FVD`*@9Q^5mDDRvlN!=@?v
zZCpMBr8$5bR^hdd69rhxOtOZ#X1F82MDP>Sj21U@3f+v>S!$huk8BNa>zTa0JUG76
z*PvCaxlEY%k_CU=2@KJx%3OM!+)IM1X_KZ9?o~}wXngP<j#MNmF_<<$9YTD{ifA1X
zsS|#f3-7O`Vsf}B|0iKE{2G)0AUnh~bPP6_BaMc}=Lcta`|wo{d~7!m04`gl;JyIZ
z4{hdTjg4#$3YpYRf_{uV)4b*k%x{k8O|Mcj$-sP&Kqh<*p7jeQaGkM=Nf^GDKKZ>P
zcIhDpLYk@)*)a7QDRupmnwp2bc1`ydQSs$f$=o*w(Vnn2Y{&P#u!di_Hn+41sDCZ+
zVvnu@$a6eJ4x3smRImn8{dBLIJhuDnIyR{`DJ-mVV2>v+KYMWP3123i$LHMddbYW}
zf*aOlwkuHt(VSqJ@|=%~63NqTObm-B9*#Twe*vBWx?&MxZ+%DwJ46n+TEPbuRXvAV
zU2DcF^fX<kW@KhyPO&K&LipqRE;2*E;3T5t-)GUWeC5zXjNA&2KI+Q-|3M1t$MC#+
zNXU*{Qa;N{6b5i>MDk_FN3th$oF1Y0a%KKLkdO>%>{r&((B&mwg)B~Xe3`N@|Cl~b
zTq3%%;*?}cWSIhbEKZ))-`>gRa1dOQ!@X^SpcreC2ge;cZgZ`fItP2F!R`DvwO`MV
z*gI+(Zry0_bsvtI3pv@q5>g83<5?7VPTr$uwxw1xdXZ^Nj2&g(<qI$cP5l1bc2+~b
zJ>&Gi^Q^1U19O{ug=i-U`9c%N^ofa_-Yyy_`sX@lPneuz$*pGxuT`0YN2fHSCL!Pq
zs0-C~dCWA3S;26b%ajrYl@+>}>lr<h?>ezc0z44g3--@4?m^HOO19OzV@=3{Mizm~
za!8G+Z(E_^wJIiaf4p2fpA$2$<K$F!Xewn5H~>DqATNm^W$|^rl)DadczzYHIq)D=
zK_bG)K1>fZpohHcKG+dLqphXW9bgA|#5gg9(L+m#y^snz7O~P_v{w(U?5+d^V>>>K
z9^G3B^q`-$N3;jnS}f)A`(C;NhEdxle5WI=IZRpFUqC1e*Jq27i}eF^>ep$!wYYo{
zh~Y`2N}aJeDgU`r$$7!GK1&fL8B#i6Eg~Uv(rG%$JaTU?bIz0!C~@RNsNq)O^e}Q#
zt+{+fIhp8QOknmw6Ar!G#zaE;cD>992IN`jz`!h$2&Fvt3|i3Y5cce1jiKtJK^&lK
zuk4T|@0V<!QzS)QMOg~|5Wk`9@Z5JfZQMpWz%ibz8I29vr>v<Dx_&|!#+FFngTQJ_
z1Z>>EF834%pkal<9I-}JN`WR$(fyjh?xWkH&;#;ampmd}ijkT({#LES-ey~dbPF-H
zDx$LRBB{2a)+TOy()f<8*Ca;7m#2qyM9dhhcAhF$lsZ0E<=s~OL5d_i$pCv02FLvG
zV%U|H<6Y38@6lABM_cebc>o|1fPkT^{HXfOxy#Qoue#~+sqNd8!raZ|D8eevhNb38
zJjsSC1rX4uz*hkf0}^b|xB}b?u<UTFIZTA1W>OQgDt)yYz;nP|_X!X%;2m~{502H}
ze#w2_15X*C0(M*N-FCp&uqV83fa_^)6$;pT=Bn3h5wc=nF~6P4?Ee<(nNiRM8^ihq
zA(1c6)t_*zf$WXQ#xFE!1M<?C36Cf7xF@!lC{^TFf6`QPZvbr5f_cze*1B~6=&zcD
z#!~`CVus?a%S)YzePtWsxF_c%I{Fl*nQ$upJku;UbEndut5nJ1?Pva!;MoFl((Qqq
z1Ugrl%V`|7ww}Yg6C+UM(MuwA1^j_(FI6!-@FMG?55bj5$)q{5N#W06`1Q#bE<Aiu
zeuokj#*M^2wZZb&?9^4c0NZQJR9@0MCuNwH5C%s9@H^LVTb4TAk*c5@$G3qOB(f@N
zlXoYr8h{VF|0DHe+xYQ+5`$lx?K6~M+&}&n2n+Gn|8d522UE^$=}LEf6uP}G2=JK)
zX;E3?&!F)~3?w>@?VB@Sw-Y6~r_h`)OxcYxCcIF~Pp_Ap*Q))Qdfb56sAXzvpiFo;
z90t0eZ`WhexcZF%-dBPtrJ)EK*|gdM9puF!Pr87Ou(+^!fMvpd*K;#NQ2|4XQ0BKA
zF)bnLqv!sXjWHX_09CyDT?s<a@ci}y5+W*jb$QuxF+H%G!k>F(YY$g}^<{wb7C?Oq
zt<<J#k$ltGKhoWQ!UW*%>uyC|1zdxMh1P8QQ0AD1`Q*{pJ1)%&lzgA%7wDThRaIW&
z1$VI7@mHjY*kn?Xp2@a--}8HnXI^(0--n``k<CTPN&M!PEvZl-JgOmlBYIN#F*DIz
zIIO?s(PMN~ZH2U}9;$irf6H-nLvytMdzLZ}&(d!_;O)StANNO<5=T9;Ze~j>Ocmu0
z*iy;rc1Es7Ls^N2lI7CwPjlF%B;HSZJyQXzhf62?%dyY~HLy?sn65qY5s3X9=dCIZ
zPX;eO$X_+n{qjYE=qxK)D`=)`Gkb`fS;uE&4g--S*@zpTJJJI$EU$kgS5)_{Wf)ve
z=@b5UjZe;ttNG<wm+y~f41peKp|Ni0-;+V=ypdD7qD4sEAQ+SZ^=i(SzU9YYxGSHT
z%Oq&Ddg+}UgR5^=KXmN4DhieyfPvgD;TjKJMkcXkiQMB&;4`mm*GcYo*h^l15rFwN
zBX0cF0N*n^R2yU@AUrT*Bm9A3xQh(1;;#Ml8Iq+pei5xuRi||i0Ca02O05F{2w1Uo
z3WQ~VcK~nCf0VLcL?X)sb%0<rZ8YezHRfw9p`WL|?!00En{2wAPE1E$?^r43KZvN>
z&og)-afL)+D{2J&`&c9{*jCyu&H{b8&Cl0R1q9AVydZC*_fY-g#Z={59P}P^g~u|9
z9{*nRAX;q7>wO<lNyV^cN<7Pb{%MfVl2{;Z0>FGVg-Pts->GLLa`R=(#H~fvb!i96
zAMWENp)Nltxo*{Y3m>l!UyOrcDTm{a-YHt>$&m(T+t7`cI-f3Y%jfYm_#}t@62M*n
zsBL7bgpUH|!uA=e`~nVlyIOeqpo^f`*--mg%ht5`PUfLi8Zwc?+2O5HilgS*>=bS$
znWqi=6O6qgmHDg&wxiB6(gwEogM@TaQo~V6Up6=(+QYj+CwqGb&bhGrU8)%eZ0cn7
z4bUDhcOf%_wqA2U`>DOq!IGgux-Rk#IPe@2ZzoGW<gRUdldcJEUYluBvpVCwsQZ=^
zaTou`f|~V{FBi94nTxe<+P0W97#7_b2fHwcU@0I9D4Lz<2cPLYYa-!;tdVM*JOITb
zaU+_%st;T;0;ov|uRx$a&gZwO`L&;4^2e>kR=-IwMwCOR#I*ScP{TLjK!Rn_%7DQ~
z|6YZVLLX9H4erUvX9%n~pC|*Rbl;?0MIg#DdxsChpfFV;RdANEDRpU_YgUcSj^JL4
z!T`xhWaVaI@hC1F;4&wS#DJoEAgL0jqa&Rh$f=rTx%YVg7)`Lk*0&O`6G)#KC&ADm
z1|3yrT1;hpWc@P)4n`?C4>m<>Ws>?RCVEKUjdA+&>bGn1_}d>M(V@Bi62rrt1+}&h
zMK|#ndpQ*8U?Iuz=GPV1oO}_&Z|kwQV#%}@TkcOhPRWJW*gNeEr?%<5nt*xl|E?C@
zcv^W$gwhgd?!ZG?Gu>EuW&*g=8;%|IX@k-4dXSMCAE1Xk_C3hf0L#<0nbVp#>L<x{
z_-&(I`(iXT0UeX)W>)1D#1g|k*s-P^<q7&0r_&fjYurX8b-_seH)l_0_Fn6mS9<D6
zR!SSS{fuBEdyifEG~3x);6>V>Le<k<kHJ#6_9E_ja1|{OJwzV?alv(W8IaixRZNYf
zk$JY{+7?4C*P}ttM6CKqmOkG4+UFU5|D|P*qr^WGQQ0r;e!8-xiiIYwr)5~e{_<AE
z`Id~O?zZ+oM=(vG7!FbaNC!!B4JZ*J?#RwOJ=)Bm6H_M<WZ@UrftI&>D@nil=iTmG
zY{Q?S6=m<T{C<Ejnvr**3EA`E9LYaK@zlqaWS7(~EA#DgJ<(BnRj-5eTmu9|5U}!<
zSEUNrpb#$b{=h@k<11K6XHHvjU6`+Im9*bVpw9c*Fz)4N**K~OD;GFHFGUea1pm-S
zQY4(vuo96=N8L;i_aOe|V}5Vn>8bQDhamJn)mG!k%STISuK35l&sfNW18*nrs6p?%
zKyPKge2^o7DB*MW59x#frE;0T$?RTv91$M%r=x`cvKg)Zxth$6WJ;3l(`C+8`T1P*
z=tySdI=ge6LQ0%XoOgK_v~q-C5W)oly`U|%w>5>mK}Po5N|NsHd!mwPL(hVeQVMez
zlP9K6f@asRN5bw8)2kIchE^j7KqowAlUZixD5A^z(1Cb%O0Ne5!{wiF43Q>@yu(g|
zU`QZm8IRhs*X?7-3}Ur32-uW4SblVv_p5jIY!$X>+y&5R4iDwakUo%JjsMIL0!Rr@
z>pmq||D;0si8v8O4mju#O;HK+_ij&u);y(LsIqXDLY)a`F8tS(@@l#U!QBv;tUlH(
z=@ZQzA|jjl;&~2)pvB%iM}ia+g%6vW4wh0M;eFbiuX9!Q7Kp;4rkeYcexy1YY%TRO
zYLZ*~6S(4o+%<4<eBmnk!~#y+aRmB&N-87`n21^VXa*~so@_VtRnAIAvo|h!?TAUY
zG4h)aeGctWAVkThXZcU#-CFW}?D+CW2HO3pHAkdJ1RMHC{*TZ|wy>uO5KM3fbnP->
zB7ulqoKHSTocGFP*P}xuuDc=R|J|iB!4D$K&$IOaOpx}rihsdEQ~OPa&clZP2DA9f
zcs+QogUXs3<K!)VmVL&s%~IBi)|a~zO~}$gR)0uR{2eQY6_l8&2jBUAeQmz5eR`{u
zl}2`~g($)tdEWHccZkf-v2}QOIu_t!P?Qs;Hs_-<EFmDU^@2kc@R`92;qKZrNXFyL
zh@m~~unxq;`su}F1gzRW*}(Bz?NLgZhHqsZ<OW8qoZ~Ug4lgbK5mVoMu|n3z2ikEU
z>Bd-#>`k8)OtN10D&_kyevu6?%TR))fEyK}3ZM;M;y55=rfs#W*B16)){diBZvhNU
z8VczkOHwhcbS5wMgVGR|V*2wY+WYeX5|v-^e@qh;4f2GyrUSR*;d*AJQYQ{cLCIQM
zqEfHuNiCM6KfJ|9QbhtyYz(g6_MEgxo_uB=)2gVfF7sY<|5-pEkT$m;f_D#99!Q+|
zOveC@>+GW|qoPZ|pxPKqBLmcgq75?2GZ+roCXny;#m=5~+*$^@pL+xu%Ea6~ddYPN
zMZeCKTgO~lDvayFKQc1OFveR&bH^&cp|SKIaP|^efHA)2wYbkS^0#cU@E4JEKqzr+
zT~Ur$FW~(jt!hj$`sifyW*caSPj{uHee5Vr(>~&vD4Dfn=~MM%82X|?Faj&7p=_Tg
zsrfP{VpmZHQ(6M+zkFrDtaNYh5CYrEeiYg2#Q~m&wsX8&4<a4nlUdm@0+fc}oN|gN
z2mM>cJ(L!gHOK%-%S@&5l3kN4WQVz>Q2htN>G8ybq-SS8m>E#O4AI2c8dx7&3;&|T
zU6%L*z%!#YX8>`+%g??|)BgrheT^+8+>Sft%pOLG?8|d(U2Eg+DXyiIWpw!4*_#tk
zM!Adt2mRccovrsw--|Mt%+s|-Ie6XEPPicD6Q1dan|B&Yctqug{SkVk$LthI5e%n`
zP8wzldwlx5en-zRWbt1*&Hk2E5Gw_~DDaj8dIeYS#eGj@geGw>9BZ5eaDo~$m>xau
zGo?HqtgiszqNgdJg*s=@Ju;@4p>VF1{jsFmbI_`$U<BVzJ?biA#M`VF*Q)smZXunu
zBcYN02bU**_ftO=UK6y1z{iq6HVVxk497}PzvKf)B*tHt6nf26g~uYXacd~L4~l10
zQ-+EG`VabRysAeX*)?5n_Z88t<~&Km5{utsx_55ej@6Z+OCG<XvuJyLJ7&G)+EK8%
zxuiMsAd43H;RnDc0#>`z?f6K#@?+bt#^0LhfP5GZDB7uM^atrSI|zfq{}HJprwIV}
z-8%eGfY+?Q$V<u3F0s8r?*;=TkMlX)fgg`H*8<2$J*r2;<FI#K66(wsmIlZF{poj6
zEB{%x&p;<3HGCI`t;O~AYCG;}*R5O<&=wcR?$|lyAB}w{6f-Rb_qe@a^60w1XY1SI
z_@L$O-GqlHiYq1DhD+4$-wV!Yq0w&YjIHL-I?y$tcnm4QHcxzt3POyha)0gupVU}F
zHpd?zrc;_(`E!mhfewJAWdL=zOp_mFjk+;n^6K-#KUT69QAO@-I^!P$C<J&D0F)B4
zv4pz7Ygg*DhRQdAT97W5RMXWfQa5SDX?vOEOKVZ@Jm!T2M4CxyS#v!t1V*Mk+P2ut
zE;BzFsq+TLX#<?kssCe_z~ZRhscSGkIHsm<av^{~SZ<&jQ_g@AWW)x%G>fb)f6CWX
zo3at50u4)rOvZrCx-qeRu_dv!D(WQOC((e-9<P_|S<DtQB4*3MY*Z}ZRjsa%e@Cg8
z(gDDH3seSVR!FS1plaWw!b#f@C@2;mMJ;>OCIS{O8jCU2N&eEE*8bm8%=J);wfXh=
zjm0@bOZnE1Y)Ey8eH7D%2XZ7$C0v9G3wVAt1vdu>ZzTTw?iig^+dICquyo3-+9Sd<
z<3E~)G2QMEeE=Z((Gwuc0br1raE5i0ZQ2Zpqg3959NGP|9Ppa`OqkEEf{)>V?aIqQ
zNCV`|9*8bN^8k;$uC>yqNWrXtjJsXfaX4xF>FA`uT<y`4yH>d&Ok|W4c#&adcq<gm
zooD*Ql$1b4qtaGq8L*~-=NT893~qu-*wjVWKw@#kmcx%K->;|Z1O=I|K#)TCmJ=gC
zTf+;=1J<ntw7#v41)iDgr>{k2`wV6jGcDTfPoqv9KFmtp#0RbJVe7neFBR)M*IhB>
z|E2KG&C~@HfXi>U3m#Qs4;{kS+DRi<W5Fi}g>RBLF~NHdI{`Y_TnIk%XMw+P2g0eu
zpJRAn?N#X^3b;`wyu`5ezaPs3env8&Gj0QTE+W{UD|qys(=*0@*JD0ui{u2d^}f4!
zwNwaz6N=G+f9ks1wsegiYQct??w_NkB%h&p?W&R(#>kS`0Sqqo+mS7@V54py1{}Ny
zKgk<n*`EWTY9C$VueKtAYS6#C!v(7R)Bnt6d~SFT(<)nb<##Wy?P4{DCfmvfeKzoE
z0Ae-8%<Z_VR39KL0Wdkzes3}B#(U)6^U+T}*+Wh!9zDR|P#A3cuo1q77LO)>8L+^~
z1qNw=+~~CCGuLd4Dl&6OPc??V6Tmza)dLIyE?|-Y=L1TODzQV7B$`h*&w`lpNNVdW
zIbY61Hb>shG4opiuP8cpq+I!wvQ2cJMDo@h=UJ11<zN`gQ}Ugy-s1-+C770*H@t%F
zBP+psE=cMl5H=$&d?uTc6EPw;psuJ+e!OZ|ltyO3j_JG4(eQF)j5?sBMX1{c*;Q~t
zeFbPwz#$fu%Clcr%71mp?<voMMR}?SY(_u;@jhGUN+x8TZ7P2RC={dt*=mf4g0R?s
z1lDIxbUiHbS~?(iIx5K)P2RcT^FT7guV0GaZU-o!B^y$C`En#5hX$5NqC?89DMCgY
z<*_XBQE#lA(#G9Cu}w9WdE~4vnnxI!WWX#6=v(Wur}WY60F`BFWc1fvbfFmhEgQM3
zjAGe_N)wI2#Pj|z@^^>=HX9>(mHL3*$MrY3`uqSQvqBMWV-gm}m6E6-?xMw`WxM}<
z(f=^0CSL74I2h7j<$S8JJMH*9Tah~<sy9ri#E0m3u~i#|<dbR;W23r`%vuG!CFlC;
zdRJxr@kz=3b~7t|-pO9orqUihX!BTCgIizywk@J`8&OL{Et?(!kTS{ArfkuC&^>_6
zg#Q(94hJQtfjV$|Vw6TYok>I4A|6RdQOze@UeX8<`_|~VGRK<K;`}yLW!f`KC<yP3
zI-`|h1+5KBgZRY(hwgCa`i{Ln1CYM+s)0TQyivb~mUgJUm<tMN8Wz5=<+4}UpZIwU
z0l@*9Ygp<ta#oN4CxNR!YtGj?PJSG9`U*IxK&_F@b_cWT{atX3sC2>FpYzyJnQw~3
ztl<2L3K@3mDY{N_b?~n0`i=^ij`{ypN(VF~*h~SeD{1<^UqgvNy0nvNLQLKC&hN_B
z<}XgQoT^AV@5a1d?B)O($&)1o(xW+J*8~y`_#)^n*c_Zhs|U8@6`1Szie^6li#@9y
zJbw?Os>kA9zoK=;XF}H&)<2jtWUT(_XQ^)XyfN~|wZ)!j?FmB!8-65)yh!))gd(1P
zw5P!aJaV9TM>1gpTNT=s1<ZnE%`f<9lpNmZrlCMD&p5)s=zl}hF7iKp>Z>B)5YzxR
zJv}e>EJx&Ib=!emOduEmTH=mcV44}N^$`=xgX!6#X&N)<!n=qEgxrw}UpndzI|2$)
z0#9o>hOa0zB%$$C5zwv!uXmL8;8+Bx@EravgTRH*_kB#jSzz}Dc*Zc8&W*<i-w*Js
zBFz9p7UTi6%K7PNGDiy_H(vdmv*v<;M<ch(M<4fNf{uf52Jrwuw$NVJXELBZ!Q%sx
z0D<jOR29Goz!_N*%ZNwKo<2bg1}G<&82NM~Fb4tp?rLs>riZsYcS0MLK%b=5sU8kM
z<YOwXyHKsVdI9gap2&{Mqe3QqfM~@#Ds2q%xSo=Yx&pfct>nwizb0wmn3~ZwOrMAe
zWWgpOv|UTf0m0}PL&VrXBe8Evw3*26WgFOSy@=LKXH|s>oG>|09lylTicm6%gIoaF
z@l+s52!H6)66ngXXjttq^Rs`Zp#$poZvnoddcQR}tI2O_iyMMuvT-luTizeq3Z$&+
zn(7;HN=&eu4t7#`VG7opbz@N+U~1|-($}F+J_3_!tPUi+wP9pu7KwH9gfmke*b~nR
z2=L6Ez^=0E<dh9%^e~p^px}!pAFb($-tpQHjEtmkS^TF8u#bDNfQn~oIe|SV4ojcd
z69D#CHuuN~AD&D2(BQxa95RhbZ&=m2r=shQ(7*gk<teC3o3f#m)C06UzC3~ddO&(a
z?PqN;F;e8e+WplmdD14QlyP;QAx<Z1wsBKU!(TNdAH(jwA0`^t`bFr&YdFwXvoT9H
zF7)bwU#zxW?du~T4{0sYl#@do&l>WHN#@*y@9s)SKgdY{0uXe&jD2NjOL<WccGis3
z?SojUr_#VyTrG%UL0^kZCb59?ABvkWXrg51%G0(^L*j`@2hl>_rB1C+-WrFEO12NI
ziGavb_3?Kgi)1UgpYVb+^EEcj-Pgmq#szXP(ZOW?bu$3ZG?i)0RHmZ!a&aYN#-|>j
zRs{WbJPNGbGZp>Q8_wuL6?`R=Me?z)$b^qs_`yaL;eqq+wVWp0gJJY=6#2}j_9TKv
zedpGDyh8C^P9<>XC5!DS=QC>1>4B+rD*tC!gDN@mCtv3|14>&m1-^22j{xDAV))<1
z)ULJT%NJCGSjRotylNtW{~j2)pA=Cg#mi_IrZ5-=5r13Htbe)$nHtbF&<t+%7-!X)
z{{DeO)WHiux3;3MRv97@eAag&-$AdGsfbOH`yJ2xoaZ&6*qSm`P|5>~ll>L%!Nv_5
zHC@!;(TmYzrUy$t$BxO%-SG=Kpz{$km=}J+%5O4H;o;5RKEv;VBL$2hnge=QN0m>3
zdJypNQNz+qBFoD*nQIgN+|$<}_Mpm$#E8&8plldmy8yP}lvz!;x0ekoCUOUh+sofG
z#R5ZM--`s`JAmsYzn!86{dG3iHqbm3!_~?3fD$-#c`jEz>;7_&#g~iNe_MNlL?B#h
z-oD0#=@D2qzBMfY8jwX`rk15g0|xi8DM7=F7cOv5r^q!$PQ70g-Je*zynTx|F1mc8
z0J8^Z?tvZZip_Wc+ds{vB%2Prt>!UjiM3kt4T6T=eP@cq0Naco+yQ&yz(b&}|3Nx&
zc5e=t(qeE8fA2_uYEu%6Bn7LOZj38IF@nDQM6SN6UON*QblgK)#i_%mkD{yi?pZ8q
z2U`yo5{fLn8=O2|VHES@8&hwi57|0KoK#`YMo(HkPgOMSo|41fKdpREsM`OT9tENr
zv8IUKnTo;-$QS}6HgZk_{!bU2&26~z!DOtl<2G^srZ|~(zn?SRU7z4KlQCgpKp}*R
ze6H`_$0L;`Mpwq~1eKF9#9=Vv_c837u?w4zA|D-1;&tKuuTanDJ)GlYt&Q`@agYem
zEU*q7opoI;8gf0r>lZ`UkD@aK#{$t2o^5jY{lG2t(y5{`3@D&Lml(Xvn83&+^1-5N
zLCoB$RqO#>7|4%ic@Ey)pGFy?(Ph#7uu)BcjESqw4e_M3LFezkNkTESxSIe-q)vtn
zK!BFdY22nSWrq}$nBCZ@Kiw^dUQ^K|@R)o&PJT-SR9*}1KY+b=ss6Y()=2^x+}+>{
zgR1<KBB^%@JL8iQ0>dF$vgG$hK(GQp>%GMX6%6DRy7f@Ol)d_(;YlNFe93pEjCWvN
z>`chItW1z%;#9xTz-CVp@~=2}p<cV>5M|-@%a^X*8T*f`Y4QdV)yVrzmg>dy22NIp
z(j^<GH+Fekk(W{VlIif7k#ae$ax2Nqf|ncT8sHS{6aH^)F5X0me!pP4eFPx@@c?Q{
zy;zJrYygv}-0VJKG?q`<m3Zr|$`q?gIA#+bQLN2tL#VJo|0Bgp5~zb260Ot~t!iJt
zA3(}wTtVS;GR><_V(`ePt>7Sh!isG}ut%RXNu-jB`q(J8zL)%|CfosY>QgKg3pn|;
zG{9P5iNAV@&qA*T0L`*bYKW8i(wPvM(B;cX#8A@z_53zG%s}qX$#E9KsLlw0>^11n
z4hwhzC#A4n$M*rhdxC8`P@vLaDT4O(EIFxCZHO2L7!&v_PA6X^K%zl#&21{sY}>RS
z9&IOmA(1rauf(ICV=4lO9BBUmT&FK4^lCjcFaT`+Yz$?5PR$x=IRs8Ts0lVW8M<Dm
zfP>Fw-v*$dvkq(NQ|fH^ZMTr=JjKy4#~W$CASjW*j^xY0_4?`N9V6)(N@Pb<GS$d_
z&**dke*h}UQcw8)0-GAJTJ@gKfAZxd1xR*y0AkB77~a}Mn@2-snjtD+zHX{q)ravV
zMtYDdAt`xy3}@DG;Am7PC=%#R2T%M%N}pRVd)NXyDJI*t)gPMQzG4z!*|`<?LM(_^
zuJl+)ZS3+;izf`0DCI1`O-geW3w>s$mBqENY~WL?!d$I0@)4-fkH*6JX*iZwfGy3`
zzAl4O!2v@fL%{|EMTfq$YH0n}40d`XsKT{EfXt1<HVIE7vtmPvFHVIZh2$@b5|q{|
zt1IA_mAoJ>e*aXZ;o1DT?|d)#L-d_YU!}Bx93<CN-oHp~nu++Vq9KuFvhPw|nVxG;
zD|myfVWK_y|HIT*MrGM_;fi!hr*tFTT@PK-4N7;XlyrAD(t^?*(%qr7fOMyHc!u{o
z>#Xxf7YjkS=bo9puiE$d8*e~MQ7&?qW%}ag9%wIqG3yuE+<k9k0vsK<B0pBxU5?(2
znS8|r*3N?b>ZL|L{uCggBsIX`QyvDw=kUbfO&8cX0o-NZG-#~FpCXRjlT)a7b1B5G
zk{iFJzkx#>XleQQay9{`Azk9!zBvU0E>o_?2NDk!-2G@l%ysiwpi_pEm2)w0u%m5z
zNu%jgr@rnCz4i91=uH|W7IyLqg>Fmf&Wa)5o%jCk&gfP6N;Visu5M1>6@J5nu&m@5
zV1e#3t<lTBYd8a7W@kWs>)>Wb2%W~OFocgu$|uwxks%1YwSb%pHwNx`6?9(yl+A7k
z4ls|#Bn<Y>XMyRpjSpx>$>>mQh+?k?OKKS#q+gC&KcKG1E^&b~O5R~B)$7MBp)vA?
z%4AR*839$qv0pfy__R8oDp$rxmZypW`T%IK*59!l70pve;g<tEOO@^$er=c3q**Z^
zMFhD`czULSR0^wE0;}w-0ob(X-H76V1upscMV100^2G*L8>E0s-iifc6%UN3d=I7=
zl~7R#6a`;`jVMO;`w`8!ovh#xt~uxx3yg@dLk$Fz>|2zAg$A~)9G=`td_vAWCQa4s
z^Dk;%?Wb<OssE{=uFXL=6m(4&Bcew~0PHf01kLRaGAV&&FUVi6JP5KeOE|p32&1h7
zLI6Od1d=CBo4ys<C@Th#|0T+b`bVq8p${LH$P0&k4o>)Tb&h8|d_W7+5D--+`=Yh@
z;OUq2P5|Qfe1rlja~K%i)Isq@s%T@SN!%9oL(teIzJR$U4B+7W;6yB&_`>Z072S)1
z-#ijdCEKxVZnx$T)b`Q}yRLxP=VenP!w^-GV90Q#-vY)hm29aN;_s$k-xzd&q?akw
z1W9F6;HMnu9%cl|9pK=|LPh2Y2dzk8zT;^Cd#O2cvSP~iT)4vrtEnd6M3@ya0RPby
zfYgTKOck(o2Z374(pguQiEux9#eBXe4pTu=JV9nwkSZd(Qk8UJK}<nOi;q2L`{S=*
z3DXLb+0~OxSPTdY@YBbtbC|lq%AA1+lWkBuk55Gnm9x+C<)|u=E6tdtR}!;glRJFH
zR=_#p>^Ccq&*yU1u|{zU45w3)Net)3h`?mUN$`b**3b-(P@%?9Os#(MpBpNii$X8~
zL)BOV(PhXK;W56LjvT!^1Xf0VB@o}KMhbVEFOM#kU36|gd{RX!Mg1**Xy))O+kZgv
zYV6Bt{P8O$$eM6pgS@J3txY?snW<&PP9VA>2ah6qp+TFkbI~N<w3@>)W5`;tbJgTS
zKUuVKY8I|EkR2ad2i2BcAbU()^r?mg<1fc$j0}U@_W92fvd>jN;bf)9-6ZM4U=uU2
zKOm)&vbX5%3veZEE=|No)0mXA15_`XrcT+f_%>>j)3<2O<F>zT=j)CuXgJG?xj+2S
z?!Be}vU>9m)V-!k77SG&kc4IKUa)D-q%qxmmAQd2bzT}cuFDx!*>SLS`jCxZ6i+!L
zfXfGY7?dnM>ePTJP05}BaydSOorq#?7F+jeGeiRn4@qVJ%5a5UPWOAXHy^&4z_XdD
zZT7ha$Z0O-biQ2=Wh-G!RG~H%Te+HzDr;f_4vLbGXX^^N(KNv0t-ngB!fIN;S|oIV
z9`ci&S_V_sy~}(On0OcbqA8+|STeRyGCGK-U3`x0CA!st63nu|AB&pPi&?-oj5QJ@
z_$#3U0wIzLSpN!MC4P@rndxj3Q<#iE@5c{-Sw{9)o{6U-3%3%$6IFphDwhE5mPQ7h
z0zgRQnX=>?F#gzwuu$OmHZ@Zdeu(qN9lU1wABQ*e!r>VRbfHqZQY$VrdwfQe!4uci
z!#z%z9GsjuqBaF^LOM%{iYW(d&UR`DS<$7j{5VEvq9*(|os>KQj!~>pVhN~L_aSlm
z@qbTyW%M0REfavv`t15o`1#_Fu#P&0ESjFM44<Dr$E?_{a1`_cQuKu&T4YDQ3j4Dd
z_GJp5^#TyGt#V0Es{?r7=jH%y^84TPq4Jd<+B^9{IOa&8H_1BvZHv{%_Am+u3)uip
zEYQ@)&GaYU6(}vE;WuLdN~zf-e+cL=cIR4$5{FC;H$QT<&4-eFBLC2)ZBhz?xY;wt
z4!+kj*@{SnlSg2s5>XODzmdfR^2Zeo?y?=um*F94Y9Bhj)2NPHB8W%hA#pOjSw;q8
zF16WDxFc|sEXH+Nk#!-pv5W}%-{OGRQ72eJX0HXx=q*+#8;D%y<Z{+!s87|yCD15I
zPXNVcARRdZWn&u@@v_BMN!(>&I33WfKtV7NvCbaxMPlP&0zP_FvXmm5x?zk)?JqOi
z$4_80B!*WaQ1wZuw2m{4mLZ|^q#p-_kv8|qGF%(RysA39xzTO^cl$TXr|&=5>5~8%
zmFz`fB#eVMTo6=<cz|E$d*OdUB|}v-u~?1nF+Mo5TU)jypi;U(=2Fnu`7O<`<>C7a
z=zff{$)b8HKgxl_A*C^<EF4%KcKQ@ZXv1%(&gp^R2RMF%gLy-*D*HiX$suuA0v-@l
zbn!6%;uB0Ej}}XT@p2yuunx`14a%1T)lmF2Ht<X9PkZ*a2>1=*G5-Y0sW*A)J&4B4
z4mWDvDOb!`d@=BEG(89Nk<zRoz+q65sks9h02^Xt5ruf_fcDCdpU5FVt*7y6_YVWW
zqd@x3;=!WTir!+Ma`8`l>cZ-oT0lF7QFV2f%g)QeYGTv=J@W6^oF=*R6jNAD26}r7
zryhwkYC!+J)JmAN$+Zjw8Z@z(OR1Ax$QT%OW~Q-9JFGulU6w8=Yg_x&r8cciB)qrO
z#%<DT!-*yOBK#Q<WQ+sx`aQ`}h~wx&ztiM1P#D1&;!{SwCQzh_rD%jF@#P^y!7QMv
zsFyV+9H$B2IAIU<XNaJy71h*;r@|OL-{&JY#D*mLkw|u3VfFkbsp>j8m2Dk=C|qi8
zib4>)1MJO##%V$x&gzlxKh2$ESEd)8xyfVm%~SJhPa+Wd1>vE3(O@cqHJU$!l=G#4
z5=ZEKzQN~SsS6}tSj_X6!LOR8viztX?=Bw(gO<-dPP5Gl56{*Ekif<?$|gk6vzPE)
zR~@$oP$LM5qqew+k#~z7H;piXQVW=)`v{KZ=|<ol1Ip^J&Y4UC59R+B!j!u3Sbh#!
zV-A*Ff<(GtCxyb(v!BjUK`AJCtqwp>Isuj*z)X>hGfkA0Z&o<Z#c`*o0uu6&Dj`V(
zKsRo@B?%(x#*;=dfTqpo?Dppgx^5sTwf^D>m=uYvn1aQj;|@pip0VM)HyF$;`R(EK
zgl_;+c!z-kM_Mxtu^@PK*%wR^$DmFqYY$~c{F@yQ><*X^zCuS&b_jKnRCPtYox1Fo
zM74||OZ_E+>X~}5;($z@#rFgjLUbu!qp=~y%|A1eYxQ>T5MJA-lgR?VfVi={m(@(<
zaevu5)(mz`d#=|#tW)4IT$v}9gbb#H7!`_%bz=+`jfBe5l44@LZd!0zeH&MZ$KhsE
z;VrgdISNwSYj&d>H^WU&2u23up2Ks<|60!XLN9;<P-z2VOAisyKgo(Xinzm;>``rX
z?)K=QJ!>}U=u#WC!<JnKl7EK}6()2PpYMEPKor0r7{Lk@zX_!57vi<8(NsvBQrtiq
zN$Sa{iIS+jK3t3oia~#Wa^xcvUvp&sL*gtDUbmo*CW}dLhHZ`nPE8;`0qQgIsMj3*
zZ~LRgu?ImeP8M>O@9L+oJz#(;pbd6DfAS$v@%K$ZX~^vbYzlTas!5>C`jC1mG?q>z
zK?s<@DZkq^yoU)?q6*trKbIZ%z)y3Z`W~GYn*)i~{DQH`w!<}F3cJ{fN&rHt;@_Xx
zoiHre9}NaoV@nJEm@zLiGWtsK@QvV+)!dthpQ(^OJ9C6w9_TnY93ic9baq_t@tehZ
zR9#@Yd>`cM8S?;3h~Vke*A)Rs1Xelp{$Lm=4QvqNAd3mCD6EkH-LpSJL3GDsGgX}w
z(ZHSd;Bos+w)ze3FcRf;uu8VqU%u7cS6J?!KQnP%^D{Gv_@CA8wo<<}A@jTwa}X6}
z#=g?R_x&H*lSlkQuEfIzoL0H(&=CNCKX5Z6Yy^W~BLG`OVgc}n7kDwvP5MCJa5hYD
z|1+$;d0HA97No|>Rh$C95)cT2-FMN}@fifY(1x2lvc6bgd6hi3%z=w$(IAg6VQI>Q
zp%j!oI&?&DbH~UKC{m&a{(dXx*J^z5d2V3>_KUoBlhnZcv!@j94;{+AsFZ$xVs&03
z<wlT~fn%Hl7Fs0w(}uZsQ(!WynRUBEO?26C^ke*ZCD6u`VCeu<=DqT`HD@Rq2)~=<
z<CO)_UL*mIPT=nR0}zRWv8sN_G^WYyNT-6z<Q5F!1u`mW`#K~nKktuFW^UUGHInY9
zHjV9M61#k4!EoqrK$m0<iQ|oAbIX8EK|y5wD_;4F37msrN(#<KE?tE@OELwb*dZuC
z(0Rh4pdviAj*6{FOBhV(UQ@CZ@VVlDVw<-Ar;*SV4@(77`eZKS>^@f&&?2qFu$><&
z{P~elkSoJ*VJMH^(N6ugW5UG2N`U-R3h&BxcQ2O1M)dy$YW@RULE2_br?InnNBDNg
zifwH3sVISg>>-<0BmB|3ERcMCnI$CbHvbm*Q#*N^hQ+79B+gbpBFJL$3Tdsc=%y8v
zIXlyg?&dBV@4-p133^sXZ%YClcZ&tJNjZ|9P}cDG5uoJ<Jv-o8WPsWO<e+V^$YKH^
zyM-n{$cMO`fMB&hSmLlm-#)gWlC(_&w=F>9dqB-_KY`AJP`2aB-O6plM5GfeaFf#P
zQ6sT@u}IP!%@252wFyf>SIfXaau7ta{_xI@rj~&Z(v2pLqGQ^&d-j7!F_RKc5)^^>
zUKpTERjcDemq2P2H%wbR(iW@}?29o*YV9%AH2bz&J4cu>O?_fO$7;apX=Kv{`ax5!
zjK#=qSEa&JiRgBJz@k|r0j&0uc}4Gs>}ByJ7zrq^LUe)XSv!3dGAL7(9%URVa-+Qm
zK*&7q6<g;}72z~$lQo#0V--*>Vtr@|*33S7zDMjBTpe;j1V7vXE*M?mAbxQlMNZ1k
zr}XMSTr^MPWv2jbk`>GRP0JNC(4G(LGqeN5^OJ!XCM-xk8JB*q<W$JEL6mmLn2KT^
z1vj1$#ux#5FwQkSYc+PRi7}kpL;2;O4``tjmtEa2QO;NQ{*WX-YycpP+Pc}vg^nR-
zL9e<si*r>aQL!3KWsL0uir;-LU--;2PXE9+?M+VlMrA)HNOk&mcW%brhWjge0;r3C
z&8HVQWPy;zRdOq&0?IF*#`!Tvz`SL#-qW}&sIusq0Y}ZG@g;y2>=38Na1dUDrdb!g
z;-AwozI}`#AQVOyF>J$*ysg0ho(@T22gOyxl-~gzisuk=0R0Al{MPJ20iz&Yl8rTu
z$Mfb7J2cCd)bEXviTDt?P{=R#M5>SQY)0u|0=-!B7kYM(NR<8IwBQ>%pT-v}P_+kB
zt>{jnPni*@$l!m1s!(dyGD~`;v0_&`u>U%=Z*=ECCS;E3^N+lu^)_Nm!&<+D;56Od
z>UT(=hZ<Rn=G}J1_Z^<8{+E-Zm-qr;+DCMbS1L}T0R|9O2DKmz<WCL1yzqsyI)H=%
zvTh58QsDhzY+}3!9iyEbJL_}K0=Zr89bLdR<LOj4X==E8s~-ESQ4*V4y^h~05xDdL
z2SlK;(J~Gt;$a=xo)7E>F4Rfw<aIyaPJ6rteU2Z#%BlB?$CsXN#3T>|)ZJ9c2?Prs
zV`;Zr#C0IjpsJ&KTyB}7YV_wld6G_w2?y2|S4d}vzD2&N@_4a4lcZX)(8Z^}Uc`_T
zKJD?uK13cp%1Vbfi~DV4{w)H?Iui^)qz?M!?<K4ser61n0SF_iA?@)2(3=%DAP1Dd
z;LojkYGT>68+7h6J{jU`fC+$9%ns&n^y!f@--JzPk;0RLccyyZm?$U~*$(}z@VDcU
zzy<0fby-nt*hjJg8}Ef~c@2&X*C3wo?qv-(3M)MxSwl<>M{mt{4#__Vy643xaB2&Z
zm-?s>B~&uwFLZaQ;|<ro>2-aAP<(qQ@+|v*FAceutq`#hOq;gkv$us^OJ)2(ZbJ;x
zRFeSa=!d`QN&e0*3m~>xLTN<f7=RJjEVb~oE{15Ty|f-lJbED5&Aj9*n5U|JxBwL1
zAm~N*wF`AnPDd040^qyPM{k$`#tClviE`kH1n3qBF$4OXQ(%gjZs6n1$hZ6Sc(rZE
zPnqY9hxE0)2LeAT=v*-jQeR!2ij6nI*d*2d+`8t1oc50Gx+>d^P`2`9FVf0-J3y8I
zUB}`9ug`V}VMA^Z83J(RL)3JTLL+Gq0o?6r|GU8QMLH*{A<@`Ma(quWK8s^iVFrdq
z$s_{E$kCD^C>|l$N!)I|KyJSWn9>wXUCogg3<S17F+wRKfdLz;B#KV^wYEz_%G`Us
z=1=Heq)dM@=tp@FKz@<^Sz-m!QKN_K?p_b?QyY0HX}bp0W0okD;^oI1eR~4#&h`dI
z#4<XZ06I98)rOD=hbOP&q_$tUxFN1^tg#=wpHI)W-blTU{ok4N-|hXvtNgJBZw4UI
zL3uTX<E9Dbz6oUkNjn%1?@TkP_KtXiipGJ?XZ;o%q@AUg_x+BNB<WaMm-2m)YH3jc
zY6V>Q;HwPrYm_HI@<<@>&di+u2j7pRfn<Qk0xroEYTS1;I*J3x6we2DsNDR_@Z|(R
zMNw=dmgx8;lRO#|skZBbaAydf#8Fkcx--AQC}rue!VcF&#SJh)TC}A#CTPfhM2B?8
zmzm>`bS~~x<u1bn*L(446=ka=P$%pRZA<-lKaW&qs(vD(OoL!d*!#*LpqAdnZF%Ej
z0%0(2N+Y2xP|8q^O&voqkvu>~jxG}uR&CpnOI{O$4|u|Bh)@F&Ts69`DGKPsHK-z6
zfm^0_zXk9@K4P>An60T{cyKW1^1s*u(w17Q?Bk^TFLm{BMBpe`id|O9sc0y?_lcW?
zD^Jj{X@)D;lV(I{MUzF*+ucLn!}VziaWDCf9j0Kdc^BQA7%u)#$YAv!bT>CtJO?JV
z|Gs<$LLgO#=~gVG9fQJ$0at;hO-J-1&=wHBlnesl7~nI?W(fj#kpQ}}lQIO}9to9b
zJOt3Y7YYBhj1dHJd$FXm9)gvp+K;_qxR$6wAg#3Kg{Qi(tPS3j)~~#JAyy)(UfiRR
zbZ1A_X$g7vIIS<na7RE0#%3_3E8uWh18q?Pz}s}!Hv7~=!L<+>8rktpP8X8p@Zsc7
zFhF9q_taCLu=sH0-~fKUT_S0+2L_1qm`*TmMNmKuMx_r~k=+YA6^?68lFJ$cR@POl
z)N}}25mhZIVgqLI&F~)5EMHKs8UYyPh$YW7r%Q<`nk3mnSHKfJrA`1zutEVED>Oa>
zbDAnpD04qpk(`WjP37KhjO$dcpfxR3HYwyd+&xqGjGA9<xh9jvs{QNUnj1dJttRbg
z^uzMVi`y#Rj$?iPW4BM53jf%hv<VacpK?U!zXzVvF$yw>KpxjSCgcFn3t55k#5j<b
z++t|K5Re4Sq=IY_{-T2yz|&dh0t<i@CRa&uEh^o@R}hRAhtC9~ZvZ*3%&9J+0%FB$
zz|iKi2-!+IAZQPHlb2m!$p8$e<M|*k6D%i))TOgfUm@FwN^?&iIW%?4h^8PvZZ!C?
zLimzyz{mph+dwkjkUMCAENRaPQdCcX;~?DF^WHpbZU}5FPy}#DCe2=cLXq47FKDiz
z{igK3c}4!v7CWX6#EhZT#F}DPv(;yF1`V1;R0b0?q1~M6^n2LsgsqI=7^@^r;KBy<
zCw)2+n4$m*Ie_8C!$2wGo<F{N7nwi}rKV*SqyJW99GCfk!k4Fr%WL1p)h6ktq0#j7
z3_?-p)JA1<8}$xwRFb87mKESh?L!wBK)UoeLaKQF(RLi%=nYez74{$5AVMVldy`tg
z{EZi^i(|)b_yz1L-Hm(&&fhFzV0;>c4jZw(d5bBkMU_bP=P>QS=OC942%Y%U$H@^K
zZ5akZgRMGI4$SLoKW4gqnytD~HCVRjA40zdm!Zqyi=y$Ak3HbK%-dlrbr*ghEeA?8
zQ$v%+FPGz^0>C`22Afm4?NY|Q6##9f3uF*9>du7VI16^_sS>J>vzTx76azYQ>Ougb
zumbsbQrY*3h<!%jgvR3011lqd8wD~R__)-*%+=iu^b-TjN(DRKYzhWLErC~gyjGjr
zGR~*tVgn5bHR)pGC0@9-=m7wQTQJZSgp-zJwdp9DmsqM!nOPfTg;ntY5he^2@bO{y
z0~(r9cCS{(d+QI3uyG<;#8h3O5Q|`hmi_LtqQuvTqGMeM)--j31oqjFcz;&{6RkD!
zNUH{xNb<-TopvF%>)Iqazca`_%KC|$1WQ|P-p>nJddk9~nK)s`|1mpuFU(Hgg*Z^e
z7~PTQ@e1|2uNAYMfp|L<yF80IFEFD4d$RHx7|(z{3G}&np#12N3+=D%w#Z!+O_2av
zawHZxHD+Ll_88l}y)PzSA^Vedk^@PiYSr{-mPKeIswt;(2FX;rdCOGMm_Uvk3KV)3
zW==jdqvrFHTGVLjt{uJ=FJgQEB8;lNGq_J3^qSuR{rGt0XnzHibh%aJAP@6t^>2?1
z79w!zqaah~MCl1Vk5lsOw+Uy`<sT7JSh$5e;L4l2Mhx0>-bJ|(MX!8h$)T|L%9zXn
zE+qwxB`z8z7iTxcZe=R=wu~QZtt*A6x4;plloIYixn(LV9)rc78iXZ^U$WSS!J<>U
z;)V4uc<|;rl^z8axc96ly4;|bl>E1)|E{pw9j}g@{jRu2d<`>w+>~kZY|dmpaq|4?
zKT*DkpZkBo)~FO<`zr#jEA2GXqHq_4i2z_I6j-V!y0(eKMgd;CsKq^G$sS5^QH<Bf
z$DLzV^5yah&x0%=E+@A;3Uh-F6!VRI9D3^lT6*}AeQmMomMDViI$QZcAZ`o=;7r!b
z8#Cs4yF9HguEgo2*~ecYcBU$1<lV@|as&jUF`h@}VgK$8;4K%0Cm9=Brfb7DK;GNL
zuVZomnCD2-0NUhvv&C%#=4AQ9&X2)z5LXHU`3wf%7^_Q4U{J9G@J(s=uWbN87s2i~
zEw;lLWGkm4i=qHu4uFw}@3C}LR)U0Nk^2&vR3$rOJVZd$&m5ed+2yK93D-`ixc6e$
z&ke)tlQ@4Dk|4n_v9U$_?h%`9e5fKF2^M0$-tQ{SyeyO#6Z>yQ;m>HTyChWNt=aB5
zznSzOY0?X<KbrD)4Th(br307$zm;X}gY46Tu(NaIcIvp&no&*mA7eXzI;5BM3=q+n
zKZ*%o&KFuas5a`WG`Rz~2@tqo?VAQ-G|0oT9$B|OKdlm;@@j8c`@&EiRTY7Rc*!7N
zESg5VbKp3OW^k(D2D!VJ&zow%-+cQaPak+tX5ZceSKxp4T77V={yTaKYgQ+6Hfsaf
zrSbcLp$w%iVS^SIV`-=$-ebesPINW?ZfcmGy?iK7u)S~Sd|Cs{Elg)2V8m5?<ONw~
zdD4|!d%RE-Y>{*rG!*^O17(Y^Ln$cafh=1JT`WlgM(hjMMKzETRrX2^(#T-~&#S8r
zrHpYq9XXPTSFH3}#LRZ6R<RCU(WR@Qt4-|5;9xvi>ASD&K+rZ6p?&B@W$?zc_zl_7
z#KsQ=&yI}C+bmqaD!-h0uOBxgIgfh$QU4<VHMg1Gz4~_>_b+|z-_x3fg+-XaNpq0M
zW3R~8^TX-qag%>Xlh4sFsZGRqEQT=ax!xy#Jp6V&{Pxa8m_3+Nr>GKau3QVjB$NpU
zcc6I$?<I(Bpozph5z#j<w5yqE<mE_{Rx?$XUDwmf1fiWFxmXn5<ZCT6TaLfBT4JC;
z(*sf*)i1%^YmflBVA!WFR=O_jbLmVh68ULw9i}$@%p$JB(Ul>JXy9$;5b&Esk@Y;p
zDza;_d)ogPEu7E%DJ}f)4$g@Ygb*kljbciiwOH`rf5HHW*9+_T0gPGh(Zy3yN|Gkw
zSC&eZFkCq}ZveA`Kssh9P=!;x4jgke+;rrMB8$XE$&k~O#GqEtC;*|_-`?Q!Vi>$L
z&;|8j#P`5Txjr_OcA$Y`ccpA+#qnjy+s1-Yl(X1>eH-!pe8iujv}#*&UAIhcwW6{1
ztXyR6zA0^T>#1}-%<if~g%gA3Lkf=2{|p}2p7KatlRi5fKKHYEJMaJfegA1a03Kwm
z-a-rCPsceQgQJS}bG6s`S?`<6H_zQ7>%^y?@U4}X-kmqzK3kNee`bp1$HvE>`X(op
zshI7rfWie>w9FcWHpcWZG%yyA_<;x-9xe4nu6ue54hN&J?3dIzYfnoWpp|q-=P-#B
z^r?s-qtZ6)Q!p&e2EqT~ps;B}d(W!He%6!zZ5bN)#qrCB@{L1_2S6yi@F~A~{vk}T
zO3&;!yX!QL_De?`kx1a-O_HwSnWl<p36#=QD>KCb`RmQvhsDMToG!fIMjtyneUoxG
zO&Hh#pAX!A()OGnZw$iIhfGPOBB4edEb%2%ErZ2k=z#rykpv7;yoP;O1O6a=wNE9G
zwBOa1R1q^{;!sAFp=wD})M9Jn9<jT!+^gj*Nv6r>i5JI81SNxK70&O|LD&=j8?TlR
z>uZ|7PH>xW6@!1<@`(7zUUp!d(?dljH<p;mieh#lVy+;RAQ1+h9c1qOuZ-4#$Rut#
zJp5}S9fp1**yVfkdyw>LniL8Q(m_T>e%x_*rs=%W>%5{QC6&Y|R*iTGKHZr2JD&I3
zIuGjh7ta^%T%*mK{5Q1eanD>rbN;9chT#d3@M@EjgSG^O2?0g!-XggyCWi)VI^Yg9
zGS>dharME(PVRu(#j%+Ci?HK)559;}`aAsW%y%Fo6o^2L7%+)`*XJzP`t-2U6v_e4
z2YA;(0DSdKqqdD6er;d=Fbb%SIN?Y>2zCxjO%|Kd0kJX2w)YwS@{z&k!8z!DN-RNw
zgkO8d8^jzhb5nsX>5u5*RzemNf2uS(HH=X-IWq8YrBf-GYV3H}8O17CEPj}n@_^X?
zgx6motX_0d#zsfCit|y|5lw=@RFQy#Ym{ETaPnv}`IDl_bWrKAYD2BPvXR<9N<<I$
z{FNF*@)a9Y(eJJ3&VB)LXc)HNglNXs=EQ3VO|_qIR3Q@aFc>=6%BE1@uUBH|VA4P=
z*2SpD<k`Q+2akW_22<}b9IiV)zq@pAeTkZHXY*TmcxKXq`s<tLvO96U-1)TZO^PL%
z@Wyu!B6zoz<gmwQyG(=ddSv^;=DBV6lV@(z%}w{mvyurX_LmUhuovH$pte{l44jh8
z8X~Y25D|w3=ErlX)6^1PJ^$<cArMzv&X4u{PpCI%FjLw*-v4zV3fiqTW~qC)RAHkC
z|CfmUldo)Y(U2ynhBQhxwyM!SONz{TxVdF(1%5<p8D%jEU_}^0=~uq=-2IdN^suRE
zK+p7r4yinTw{FhBakBZe<>tieMA%zIV7N+n1-fxu7J_tlv4Hv9UKwi_!<LvTSu@$+
zY)o1>`r0ycU<*uaNgaoG!{{+gYV-`XohzZv+{hWA`)A>bn5x>|R+!<*FYmT6T1_EB
z4Mlr;Y`OEEbgp2Za2tw;RugW@H3#vR822x!Pc|;lcIPODrBIfSC%+A-W_09&>=^>>
zGYjYqR~92HoS|8Kl!!}>{AUon4WgNDI2zf9hJP77G6&swG@SoiT~@S{PLyE}TC-Yr
zI^c9I5*>xrLhJ21{kb@cTS}fe4$qgDqctzaEF(`!_Uh33_)t;YqRu5N_qB+$7Q2~U
zp8qL=dH4y^T&we)8>*%3@k0^bLl!@;0&1F(>rTQLtPf#}#B3`XKR=q9v~=%JE?aEH
z1S+NoM!fgm1;wkB-ztXbg&~sneL;Sem6=)17*$(aYd<<VGIG0>^xn%Y`sH>S8Ihad
z+s!7K7r9*zddUd$W!_x9hWX~v$X>}Vnm;)_H$ay{&O>xaQ9m$?!g4X2Dr0oO-=@21
zYZtDH>G<cswpg#;OUvRPUw^eYsiYq*rTpWoY;^)t3zxTwK}M2>tBXqtCY+KsoIJ!U
zJTzM8eu@G_MEv+FTcs3RMwS6-*QTLb!j@ZwLUy;=V0-$ihxp$9hNIK2t&AmBWCKrH
zoFqwNfF^fj!|foN4()Yh+_h``gsF3(Zc0!7;euSD;GXePxbehWH3i89MLWGjk*aTI
zYQ8Yori?^s`S~`H#N&vff7~*Y!}QY$A59#HvJx5AnjF^15QKmHT2hNz>kylL4s6M7
zJhjp*%_XY%wm`clAXnjQ@N|tf*E+A^7Y3zN$YBs$w#h1*zk7k?J^_8i%85d(V9?9i
zFwqjyFySgdXkQvT_|r)MHQBDFjjNx&*Xvn@DwgU0J^eTHkT*ho1(H$PjkW~#9+Nr0
zPg@U}cP4VOKCezsgXlzvmv8L&6^0L55Z1fE?@OyAe=$NH?I1!#hlSV=3c^dX16m#G
zBVRE|Pi<{zGKXeThtB?u<M)JlFA?Z$ncX$gFa(ky{ZNv^-3{!FjL))DWRt>q7+&C(
ze^!yVVT`eR7ib8t78dGx9;C}@#^I|C6%kFb#irBtE#>pk@ka-nY)7&*Z<cj;BSOL$
zVtU;TEAAsZ+Qc!|4hjhfr>vkI#zp$AK25{1ETMpE^SJ)!HhnLw92IK6%52T?VF*%q
zNnLweMaEi3UDCB@X7zl)?3CquQ>%#LeHn8n0%eW$V*V|26G-rlM`L1RE4-Hwte6m2
zJKvpa$LcEQ*s%cbM%{AT-xa3qzZm#Grhl<}d<{D}Fxi{x-k)M~MV%dbLMCjq15N*I
z^^yL&^1*zgy%vmLasjExH;(CE9aVX;u^~C5DVFs7j8WPHC6rcwT=roEC;U2I8<V}Y
zC#DHT4`P1B$jm&zigIvx*v3><RaM&l*mXDR_qa)VIR$xhKHvGA5`jf~;vtYW8Zx?l
z;6gyF9)hoSU~!jfK0%^4CU0|n))a%P9x8-=zv5$Kj?+Lan`&=KmO1ui|Liz}($hm4
zxt|2y!QQT*>$qs))Kn%R%yo|D$ZdlYh33t9T}@?>*8KdY+rdI(0sB$^;m_NCqUCgU
zs$e5RTf1m1#o~yfT3efNBTK^GD4%R6akhR#yyG$y;$alhQ#%Vimu%y_AeDaMMovlk
zYd*wK=h?Df^;C`muE~Lo!IE<z#i6a`{75C59=W!Ryyx<pdvyide61FVBgT7+h3clD
zqT*`CEaT3ka~xO+EURag<?f}-Yo9Mor>?f}$XjpytwehQT|qK;rO9XRR~$oEF{2DJ
z#OCYsI9t=76n>T#26%H&%<nOlPV=!sW;tDj=A~8Voekci5U<cQk4;!T)a%<A&NU9}
zMt--$QncBlb2aGAx7o|N!&X5WpPPzaamw*Jc^W(TQa!BO;u8HWzQgDP_W#=)IZCvp
z+0IDvY0TTTqs+kAFPh41FrvuwyMFD7<kJ?$N#{R5L*FZn5~Un3iP7{1YzI55l7-Xt
z1ZpCy$P#>=Et8&jR&Z*A4=f`~5WEy>8^>?U$=)^bs1Z0{{yE|P&ZwT|Nq5y|8EJ;E
zMY7&Ef-cwn?eK6xm$ci+@@He|SEM$bAB7?Ju~fS`+9SEwROgmR23Dp_3%?V;fTdK>
z@vnPNXQZbV`2K*hPYrjpi92CPTcI39l~ggGGw0y-o2vxuHoc^l<w!Y|`CR{r)S}NB
z!hy(23>J5DG@6wwVtN!YWrx^aD2?2M3)=GmC0(3uQn=m-{HSyDl6Ma!QtUdCjU0pr
zNSUp#i=M{azCroH_i?o7b!9+~!fqB9Ds7S-MoxVFIx0V^9QtVIBz()?*4<N%AKMwR
zxY<be+Jzh*CBe5W*SA8U)$1b>LrziPuJp;u$n176$oOp?Gx-grnadhyt=4DUvZa`}
zmFF}~8KX=!w?`slVkqpkb)OVC(&CqQXPgAGn~C`$$Uny3s;#skQv{=LAUjOK`o98I
zgV{tjpF$3Q%KmKW<$_5ZMpS?EN$cUObRlnU4Gj%JaCy8Ed_3zN%i@W{GxQAj(;KFy
zr3LT5b~<=^bK`wCcVb`2m1?YJ08b*#{9$mo7pd9O7?v*Ps^Qp_9d|f7X4I&<kbWZE
zSfs(usgbKMoxT257^ILpbL$9e<cuG6b?N6&=n~qDPS&*qTq&fK|H$kq*If?6%$%p%
zPn1O%LCvusN|j-*C7)gXn#CeVS#GZOZq3aN`;Peg+U!K^4h)VMEv@ElncaI8T-VvD
z+#<f*o_w2y5%c$Ygb=O0)(Ke+v7KCw&>L%hPxCu8iZYZ`RbIE89HyFyvOU+xC#6Uq
zH0x#6@5{4dG+9K9iHo#nhh-W^b(6ohq&jJA!{1sqJFv8%%ILQEeO@%IzqQ)ZU8%2}
zYoE~jb;su|AWcr|jS+t$@U3#1;BN~G@$65>EI1{#K~7g!rg@R~^sLePhdE_d=~5@=
zm1Y5|+^Mz03^ASq6uC8QsyYfx9$E8dd!G6<esM|8rO;KQU7JIw?dSX`D@s^Zy<PiL
zPHwr&W(T~AL0EDR|JumF!Yy#pL3e$t>c|`zkOC7(qzw#+Y4d#8<x?4l-=LJ0mF;J1
zGVbDa-ooy?W;*z668(<l`26g10<vTMSk~fSzp?RX$hV5THckDUt2qu!t2vW9RNayv
zq&`r<mrIk<0G@ZYHV#GN$l21pS17jE;v9Z{*XhQ)4-AD&x#k`=xbpN2z4nb<17#@Q
zGN!dhma*(kG6;F2ro;FA{Fnr<?UR;Ld5<F)XQi=K37s$JQfUI2gGRn&b=rNGhM1_j
z#V<8-9uG6Zk|Z%*VW{d5LBb4}5VPStiD}TnCQvQnD9q^n9wzJ)7a3!SWo==nO;fe`
zqt?vWzb}9p0cKh@f#X*yoX#zGUXcyg^Gg=-4_%9dRNxfbAiF;PJf9`1NI73%Vck+R
zQ+8w#ITbZ0RCwuQ0L`Ue5;{8;Gvu~MC^tK#>u!&WCBT#CG|)K3MPx9`|8Q@WUd~|<
zHQRN}iO@<M`~wY15Dm$fxknft){qveNw({_>v&3FGKnbV*nKNdjyxE>fCbKsq^s~)
z*<~MG>wP0k;|q;Fo1C{zbyBm0@9Qeqr1U6yLzmo9P@ii`gRCf#&iECdtKC^b43iSS
zWDP@*)A`%um#(q<{b;?HF>UR`m1o}$9!^DA*!rL)Rj&n@(R2Tjlrd#|Z6Bm~#im~}
zuvhy^M=u{+fdJF^_bQt3$}g#*O){Hp-R*(oyTw&}6AO8iywNXwnNsv2Yj%l4bR}_&
z9|z9FSC<!+21aU2sb$&i{x}o%sp+gR6Ni<kEGHBQ<xX7ftF?SQtRu~=?C{JhzT97r
zTP?ERk9o@ZMoZWtMivSuH!owZUUhv;Sd@Qml`9OM)vf@X;^gl=4LLVB_jPtRALgX^
zK{44bN!c#Nh&5T+DM-SkY`~;!*LK$-=UFR?j}n$aTG1Hvc2u&!Y1CicVDulCA*0Zq
zj55VIeresYJG^W7Ji$CXJTvIB1@vD*6N}@1c|VA%?<>c(4NKy8u|y-Nl*48w7K;0+
z48c31)NUa4PN%LeCGN&uc`bp=X8Cr7`E_fA)q!a(H|ddeZ@yJM^^A{dnz4FD97_uP
zkS<c^ve?4$`sCqZI1LBBWuGc;Z)7{&@<zqDqG_!Tr$HfGX33NTraWDCNp}csO7#!E
zZlbJA%gCbe;@V&zB~Ien)Q}nMWui?`cNzrQUSC@2w(8z{5yBqst#oNViTLU|#BR^H
z@Lf(hi+#YnUZz*3<oT6d=HdQYy#<O0O>NKw|MqvRPuGrz0NvjV_271OJkPHPpXd-<
z!K;?jr03FP5P~iq-mUPuOBmsk;!+YWIOW8rHbv*jj-09s`34qh$v)pYzQKE6{-t$j
zH0blHf4+^vcce}CjA**+%i>6~)C%p6mV`L>Ce`KCo>MkDL{AwrrgkOhV%z6?CrSd?
z;d;sZ`ECz$v{Ipz;(SU;w}7xY&qzFxN=my$OQaF;+zHJd1UEaxc4l}2SU=PiL)zmS
zl)}iT<{Y@JtPQF}0^N-keg9%|9^wd;X^G7HSWYW>Bg+<Hr?Mygu*s5{+2P;D<L=jm
zY6p80S%*j27`>d%v!&FvA4%6cD(8d!!SmnkvO5U=!7K-D57v<$vH83W<+VLLtT3&l
zFs)G)R@H2L+0S5f@wb;$ua1H!@}fYk>y$#ykEx?zU?J^-QjGn1O?#3a(yJc4B6yMZ
z9>M+kk7^f=S~>l7n`Kn#-C-#I++6)Ez9L&mpESQs(y(q(9k;ihfOgxrp!oSB#`PWf
zLKn5cir>FO^R1$pFYA0vC<QW2zKGSZ@O0X6W*E6Zg0)A${#6lOFq_IAT2~=78HK(z
zpc#&qgh(SnX6cbzEn`X-r_uBIa>;X^>dj8;)C!JI=MFnA56v}htbWIGLPoFlT`XJR
zI-*nLDeFY+p#$yVLxgF2BYhRVqOmm!ibkx(Os0%R?RjL~W}90m34>E)BT5>q`qxt#
zwvwOkIaB(rtU>oROEOvA#1t@T$6h})c)D8Fy*)L&5Sbq$T$#i7IxJspK4JY0>CtwO
zL4{>kwK$Jr4QFbxGM#%vpH*xzpP59u|7iD=IO(}-s96>vfdL^|#>m;sgSSixqdpp{
z0f&saoievT<_3OA?kK6{A{9p@zB0juGOnXZ;q+U@q(-~QcIE=BC`DA2OiFrHO^XA}
zYmsHCIC?F9=qlS?6rvxpNzTo#{0~?AIW9L#=pfvd8T!#a$T%+wgW8Wq6OXl$qHBtR
zv=|Bxs%diwPZxna7+>OLRU@mjv>Bq7jn3kP?CNRQC6WlPr4sd7JEA*%Rp0ATK!!F=
z^^5cPN<pVn5rg4Z=PZzQXabgi8B)rXJC4dbg_Ct)0?L@(7?$3js4FECjRSNwuBjOP
zhVD&U7iVVd-1vn1jOro8pzxl?R#e;b0Nv!!694e`9y>zp2LtSTVpB#*#thJVOmW3r
zc`sW=m<@k@4{!a}HnV{lO{mGYgyU&+x_SItLR<gglnAi`rjTer4&kj3KeZoEp**z0
z*NJV*x2k)EoO?Zo$#ETE({83c;v5Cle#t74N%UxJh_7MPEs$;G>F3*(FMf1`KqsiD
zCP)frGY8qcCymIrVZ0kO%r)o;s<%_ab=}qd7wIE`NkCXgPe@5Gl?I34O`IMoC1_XE
zKp#18#$LlPR&F<L>kzrY>BcUnvgeF0Z|+%Df?8UsLVP`yI}$%(P3PRiFP-M)lt}}?
zjK&I9*IQGE;M9q1dbsszvguK>39Bt(R{C$dpS<H>(sJ8=EOSNj&h;W}G1IJj*d)WP
zL|33VJG+sC8oP|)?4{{<$A4H<11rYcA@gbh{`i~|l;jjPi{O?|QqYU8La4P>!%|YS
z2%eOeX76p9c-swbINYk${s%5Q^$@{5ez05|(s(snq*r~jI$(oil|!_VW^|$&gB8`N
z`lB(7bT<i}*SxvtocSIL%23GTf_~T`0zfBRoMk+L%6oY;$E9xx<=tNugS|qw#3D9A
z+JwfhcJk<_wrm`)swmU^`JCeQ8V%}9_XvyL7)3Vr{@Ke62wM5Fa%*vUbqs-iREhz&
zYf^cj0)gsKga$bSa(nWv0Xu~j_?`9wVElcn8`$LZ7$G)-sB$XUjx}vevO_kQ^18Nm
zk!%k?jegcsd0xa+)Jv#E(_MVrgPWBOwu-X28y_JI#rgCV<-rQ{zLCYCJgZHOspzi9
zfc?I(qTtIJ5GrLOD8FE$df+jeI<|FRL7&AGqr$j6kZ%Jcq!fyG2ig}My;(VS8!Sb$
za1ZWzK2FDs7OQzQ?uSq~UV(>&VsDN=$41Wa$m}r_Fg_V$&w2!0UAz7>{#gJF<25rS
z-+ULS9`XbGE<2G-=Vf^i5MMozt#bn);Jm&cpa+NUO(yh`0~6b0ZGVnbzOStIkP0s-
z)=YLs{36#cnFt;s>-<-Z3kgz_#*J54{=T-X6tO00HKl077}PtZoVxML&8?nVbpfds
z7c4TmtoiT)QdODCEGBJkjS~?eNtkSxU4gQX>U?Z5=f<!rGiu}rtaHgU-`a`TYmpp%
z5O&dW=Nd;9v|?vgEP9G^uL_?|-e)3e)z!acWMm%A_@}QN9VcHu(M<L$V_yJ$@sN3)
zw&v#|6TGK!RuqNVWKOk>uL-ZJ-*EquURcF8Q7B8(5XeMqQG7&$KUYt^_>%Rn&6Q(1
zxj)$m7DgY9u~^28G52#*s9Gp)LG_)b2Rt|i>ZI^$>ZpeS!o@uM(Lpb9M>`ULOBou;
z@V}E`R$QrOE$Q4jr6Z{96v(#RJ{JDrY*<M)NmZoCq0UuZVDNbOxyUDN--_3rEo<*6
zd4SM(?M7`dheIdC1AiJ2D4d2?Opd1(hU<8rj;Zt=MTO)-dEHG$l3s$Pc-G3=#wKEL
z6WRdV>6KwsT4W?rw$iRBp5$Bjms1aPkrsK%W72H`2yMictpjuS2qa-*HiAKuW~^@7
zP~+#Es0rALHP}XZU~#Z(YRKi(q$5YFvm(D_xp^1f>hrgj%)V8uYhw~)Fxg~v1H}(0
zfpiP#tuqVj`<Pr&<)D>RQ+sN0m}&Rq3Mo@)5*mlKTgpbs<|EVg`{R6*RuI-#V_%;j
zm#*?HEsQn%CAMV#h((9YpCMe}U#CQ8qOXyD5uf!%&yN_{>8tIs`NQ=weyEt48MRWW
zT8DH{v}&X)zoLl=<#LmQ1t7wI@3i4<$i#!S8i9}){%g)<iec*^Ny-^_l6$SBA1Vf>
zZ|<p*j`Hw;A6kXbm92J4+LvlLl4-m-n~*#%V)iFJQnSo5H>fWZr>Rhh=@rxl4`&d)
zvn3L7?ary%^wdo45WG<IOQfh?&gt_O#SQu>ZGJ_D!YNMC(e156)FAz+wFIO`S7%bz
zIHE)=nWZBS><GBzTLmRzjXtfx6c=NS%Hqk>#JI2MVU#&aCd=HAP?YY&D>PT9K0i-E
zlDBB^0v;z+IZB)EZSbu!8^}t}t=HWnOp3N&kH0YPZIF6XWGj<&9ZnIb?hPtOuKlP(
z&Kvw_>=s`U<K*5@pE#tg9?}l;1C0g^v8Ii0LSQ*l)}oG6YGyApbkYR<nwQNj=2|^8
z^cwpZ5qZ|1B2H>XvoDDo+=f=m1@ws&0CC-A4$BR`llcpCFhalQ?$-EsF<$B9^8-cP
zL|I`}%X*O<jd}<uH_myf4WiJW6m^r~L_Wp!O;bUAB<<tCErBEbS%3L?0K3qJaGM;D
z^u*rGPEmkX6V@q%#=*{h;HMiXun?M21exKLAiF7Vu4i&(Oe_h{)kA0eqdb>-esUJj
zBPaJylu!}C+GAzY^wlbFc&xiyA6dyJj<`;=cTZ>dZi@25pKLgba%UaRrY)|)^oTAL
zJKd+NGG{7x_mRUU!x=OyAl`#=Exb??lrNUiMF4?mIV*CuB}*MSdu(fywP2#rfZdnT
zJ(8uOyB45-<FzkiM)$W3G?OOYYS>}<`Sc;;y;Gco6jlHPF0al4wd-_t^>n}&WNbpi
z(je=qAmVYS8J!SSX%>vj0A7%HCu=V7fKKv7Sqwg#zcpB%R?F!kp0!x)4VvscfHTBV
zUQWc1wmSY2O9gzX=2sKAx&;JsLCgc#-j>m3Q=D|jpYSyfC^H)A&-k@FiX%)1C``;`
zO*uJX6V#D9XH74rUS%CpGA^$KzSbwizITBp$qrRAVN^8XjH!(~5{+4%^7p6YA(9yh
z9fLyJ^62_$6rwNehQJ|(t@hT@NTwAR6m!O1vf$EWn#TUt#8cBG6k@d(&2X-wu}+>i
zrhMdrdCiZnx@Teo2^O+*A{3Y@s?`DeMN^DDMUozTf;y`~r!k*C&&$kmD8T4I0u#4J
zJ@k7{TfeuzSY6&MS976Ai<`18^VuP<sC*tYubVo?TDtCdbb=(f#}f<RBRDBkGa@U_
z7O~{=G7puuzj1FIu!^vtM{ee`(5L@Uju1b#z{?qvK<$S~5H&ASX(A6v28BFQNbk>x
zBD;+;#>`3;Oh=XF)GbmsHdO{5{Ea?U7CZW=9~Jbvn>w|c%qdZlr-7Z7mK~koPnGs~
zqOpXf$pEg81J&qphI|P;Cy(r%Eq-umN|hyoTM6s$5vDBwCSkNQmJ!@*6d4)We7g~~
z*y{G~RcW3dzsHJ`bki{r5s?$`+xEMitXD;<^!9DX+~wuvXLml}oCz`;P2Fz#e7$Ia
z7}om;2BUp^yghH$XSTMs*8bte1qR_T_xJY)i<3|AA2P0j=_mHP_kjCI=X0U4Z^{B=
zH!lJ5i=#Mu`*iv_Z_OM`=G1rlr2&2tBxK~4zcccSH70`?Et=wQ8NhqfReXrT^q1Yv
z6bjs?xs5P{9}8=o^*B%ys0iMHLMNuemOoKP*My*;44ZlO3*)-6i*NM3B^(yM${s~R
z$he|Yfvmv%f%NIao!gIyCpYirccyi3oX2Fhg-IUoJy%;dM;V__6e@Ym5K@e)w(zlx
z<JIH<fE#xYDT5Hz;70YQ3wSV}0?7e@N+svrI-akpz`0dtK5jf+Ao)AC#q->l;oUpF
z%P|h9=KYc^?G|V1C|stE!Z_iLk#|*Ffym4sFL%c8d^gT{--tZ2cHaN0((d#VDlRUT
zpuhkN(0j2x`e`@E14OW)01S^jK?Y{10o%}h(aiIQp=O0{zuJ4DelUynv->-5nxsFE
zmy@*U2xoJ84mHO8NRNIZc>0b|`+uj?H4MU`lk~cLS=;+2uT$~cGL!mLt@P27v>6m@
z)?}y5s(;Xjn$d5hD$dmon=&f&e)N%y&rGEwJs$VH3sRENt1jhAAszOntE<}2S-!c6
zwu3J(kjCS?-Bpuzo@jOs#l0^Ic^__~Oku+2xB5-`shxDqL)g&b^xu)bQ$Nn{ToLdD
zl9Q9q&Id_deraeqjWRV=``lglblvnIvp9}(&3~~X<l^FD)NhW@%p?Sl466CKaap_8
z6cHJZc`E>(<dt*4(&Xo77aGOPWRXd&j*ArgUEhPUQiQPH_VWQEeV_CG_UHRU+uyAz
z;QdoEG(_uqx>=h9*jd(^#ra16LgRRq-_^IrxQ!5GJWRAO9l-j4=W1A06@)Gw2+GST
zLivOm>_*kS)>v@Wp*G)Idb~rO#-@vL9e9n6OZY3qFw&RiTa!mT`MmU-FD*hyjUl6W
z2Wsd|oxZh;`fWK@*OSY72OQ~c2aeBH#?+LNPwZY=f8`i)drxmpJ$yL{VTT;BVx-RR
zeBcEqESmVH_??cH+dpxD<v00NuEEL0WwhL)TUAmLnxw=ZaXBFnx10Almt#^e>3F%R
z&UstT?{4(N>uUSo{9wpSFqYDu`<Oa|q>75lbIa_$=|*3~@5ZvKs@;>eQ}e}IGfqy<
zk2<*Lp4s4OzT&oD83KIPMK@{J-~HJ(;?3SSBBrCMOfatl>_?M^O`o|Nc^iqpKl5o5
zFBErl<N`|aiDXyirK|%mT1|2YWFDbefnt~woQ_G2wyvD6jErBpHn-%r4d;$5E!28T
zkFAxU6`&IlVN1TP>7CsNtp!&pmTi@n_2TQ1ByNp~`viEtg#)j|vw!xDvcB^$H3|`T
zG&oH%0o7@{yi=5<G_oe1!)1fZv*yDQC-Th8(th?Di$>vNPcWkB$(u4G{O>w-aJ_qe
ztG@r}yajV7-SjdzEYQZs#~+U@L`)W{jlcn;Bye00X>}utk>cA<R`Yzm+ZAc5EGjCR
zAH>-B@0S6+8VaiMK1vo5WxdoauOez%eo_r~rV)F`(0JuFwmeV6_KnLS`UPQqKO+UM
zCE8@yk8l!C#b&Q1^vX>BWFzWlNUH{AFt591Cih<F8u;IBSDF)E4#m$=9(NbpzZ+l9
zr#mp?O)HgA2fWzrT}O*`mxna(bED04z>;j;oL1nx(>%M(`<D$qi#~+9V`*(YQbj7@
z&e-ODd@$!+=XG^pKbX#Lh7^UzQc_=^{E^jDW*2Y65`TrOi%erJ*H{e_EHNOigOs=2
zr^&6eJa9NQyu@weE`8~{ug{O|iOO$`do?|YQrrOF_%vDAd&CjyJpT8OI^I3&_H`w?
zF*+R`-IuRl!%K6$z{!pT4&`R>uC>{1Ufwky)sEgy$NBYzq7pUlWZGODD)Ih-F(1$D
z2fxLqzthqgf}hx9VqzE#T9W}W=R7TmcYLz~1g(FXI-i)-N|lNW3xi0X&%!(f^PWG7
z+^o0|a9Lx!?2P?MkR~C=jBvd9?E$@|hh+i2V<0SYGk8`lryUPpD=W=+Cvw`shbw*$
z$Ld;IH@n-I8}Y**9)v9wi`Y0rDy=-T;f7sPJx8R}B16+9)q13$&`=|P6ViQ+((Usm
zG;or*P$cI4`Zv>YR4epy!r;Ts<^#)CWfTo-I~7@X9=h?=lv|Eu6mE_AIH3~%K9&8C
zD@|NR?X#BSm)|ao)Nd%mB6oNyAJSH#*X&4MXEC`0FLDcIG8p?pf0<kz{K{KCsq=d}
z{4l$;G?GlOcDuvVIXup@reI}d<@dt~|MB^zla?@s-EN7qvh(@=k?v#DGOGXl;Wqd;
zrh8NQ`hIuY(L_A44xb;Esgh)^FtCFHOVHf(pYfEXrQ!3H!>_Ne4Sm@<pT-z8k1e5W
zRjpVRt8IrC@?yVS5wx^_f|o0tSqtJqbJeec;_)?u`6p;X72@a7nfpFG7v)-vEbRUV
z=@{By-Na$4mukN+9OG_X;%_H-*H~S46dG+8d%mg^-n*iFs)}WT7DCu^JY;QwU4bWf
z7h|_<TDQ*1rK*$izGLOrFW$LP+c9bO=j%Pk!@I&3uLDPM;du;!IDqdt82a6rfhz?p
zxDhx!eEvPG?M8D<?75uL_D$2gl(?$61VmG02^x2rA~}$}kiyO+nl`S5yMv@6?jN#6
z#bPtY2;_Yj=n)dlo3R34%gXYbo6}t;_`;+~eO<pqQA<?46k!qdF=n5aW|cOlb`Fjm
zNKoy6d_>VwA`VSrd=6tK`H~2}#ktY-(c*6@(Dr<PIir>m7wx-OMO_R2TAuxIk-D7Z
z2vytr_xIYKaqa_F58{7Vog653Ws1E7<Q@JwL#;hsAeiTHCnkPzMaLCKAeUHsflQ=b
z?)7NTDmcQNerFWYA){6HO^b({gi-n6KW+(ON+-SC5s8AegFFE`rJ83B7QPXYXS=2J
znBw89&b)75^=*QPis=4zthm8|_iG&Fu~aYRNq2tSwGcjWv$Hh;kMF0q;1+qQzMpzn
zQF{KcqN1XTGK<yxWC@B=jr>2YU3ol|>)#%nXpx*2Nroep>{^b-*efB&PL?n#lI$|F
zMJJ>ZQkJr0-}lPS453L*rVbj}H3ktfw#IDl_c`bN^ZmS^-yc8!_;|)V_x;@W_rAW@
zb=}{YJ4nOi>{-4^-=A=#*$%(MNRQY9sNxf{x$~(t^yi_Xr}NLrxx)A%x$W^T-mZzg
zLNO~#$-??+zue)Q`096rHEehGIVN|vnV`Hq67^s2e>xQhYy5(dQMU{KHQ{uON=%28
zV#L>Cu_T^}J*FD%unLRI%C7#MZ>YifN&O>BtCi$Mslt(4f>&%`UE2w2YYL7vm8~D~
z@A;(YtDJGN?s4P(M^qzMsQ^c5k=WxZ+gzfAN%9>khu;aGK}D*d99n{o!^R{HPDgfu
zLU)bz!K)b&90tH-&9KXeBd{E4{rD+_`2a;h-`Tx1{#T+@cu;$gN%{7xVw*bq^}!+$
z<`s?NnJ^8f?;)&YF8Pk)DLg-bAkv{L58Pe^-&0>|S-ClEnywn>-dD)8+Q7M<jy=_u
zx&M*d-yeC!3t4|7bJ{s^vY59m0Q({BcaMho%4T{`fa6TY&ECtj_sQ(Q7mzyTlHY4D
z&CX)Vsd38qm8{f^cdNf1*}b|=D!zeJ*#Xa$rMnNM3TfF+#MRYxG$l1Tc^RH-bvRfe
z?1%1R>&6&aamnk5SRJ*&!FS`x9yPp`tdhRctWoFH!y$I}93<OJ$sTs>r?%4uXq)?I
zYnR+=WU|<Vg{75{Z1U=4*jRIKgu>Br%@OP$Tm60tUF3?F3w_`H;G9rU23rj-T1FPH
z4>#XZhOOKd_WE!};E+3mRzDuD?$ff&Uv{wqJHhFQ*)@}L$y;sqx5s2o$k;2lT~tzn
zcKOR4Qt`%+PjZ`@Lb7B14DyBcoC)~mfmSWHN5#g*9v6(<+}!l;nV+9${(kyN5A6-j
zpO(dBhX*bU>HQ{*S?sF`2nbMdc_)lqP!ScM{VM)+N^Ctr`ZhY*0(dFVGbVdb-`;0V
zg(1Ek=TJ{boDZB8+DJotLmRw0MSb0}MHz3PhO-e;3aZ*Gy}&5dzt8wIx@KjrDg3Wm
zl(>7lvl*c?#<21-7xRR2eD8L2B2@xNWyjRb!PEYuo98QsbJR0D0x|f(?CmHXD9h=|
zP`mjIi<j#wGM%X*NkI6~^tX^-Xr`Bz${`c-+iBk^YJzqzHv%q}J2s0D$H}vw@&vx)
zk&rqgcoFJNS~lm(+&w*A##(~=2L}A;-S82=?kB7^gmVV4`!Yg8Hx^y#Q{BKEILir)
z<%Ed8?A$WDUb^<g2QI@-&sHh^R^9z5-jRSlV`eLj`6b)?=Z{I!)G<Q)dz%082oB4e
z9ow0c=rO20g_V$K48JW+sqmR<LO1<MxE6V1Z#8m@C=A^t`~=dc`_xCj+(E7glMZY{
z{d52Lle<ydQAl}(i8ec+#YuV6w}Iq^Wa-)5*dHOSNhvAx4oS{qPu0+X9U+2ZT49*9
zLVT#k>c_Kuj^(;b4`g0ALzfyxy`d7fy?KeC{!tHmZFQQ-`a;exbXTx{Jv%EB@HjD~
zI>I26qR!_?GDID_;~>-}U)AztDSV?@v$9?xHdd+WJ}jMKAL56bWKGg5vQ`g3LcG6W
zf-&T9$=cdl25Ez^(fX&--`RB_agga@F;;@P?2YHA+;;<U)dFneb9{aEO*?_?D5B&r
zxN!4J@*Qr_ku_&-B6g{bUk_arHAs3By~Fw#l<&~BxeW3J3f|hrCKb0Hi({SH5~-Vf
zp&<Uswgk<Ndpj+);mAyZCQ(un)H;+M^l3TH9eJr>ZD&^kQ{T>`Fi_(bG4587#M|Ga
zOboN?xmo@=<i`(#%v-*ZIq^?LbUw+;+f=#$nPfzbppfpN(dcp<F&t$S=ovJd6LU;X
z?h4Gif^D^q_W`sHHWPTk?|>F+4A-Vg_%&1QPkYa!@MS+a?(~>}pWksL3V|~uW|Rs@
ztvN3lX4XeU7Q`P+dXU{U0Yw&+iof~j<oUTd{#*mnqjVp_b6LwmGh5qt$dEeT8rAdi
zQq*R2oANhZkF>C~+>_$E_A{6XAje;9h0{RB7{_>T1M}UoKIAdt+p+Upp1C)HG5x2~
zP&N$V5<HnRwD(k`O}QH2@D(Y3jHxxX>&CeF``S!Uq;6TaRrko4Od&Pz2y+W&Et3oj
zEcjS*oi+JHB3T=%urbs}C!YbM13@o<z{(C01<c(lXJ-S=LGAor=GR2z=8~)l0f*3q
zT7i*kQ|}U6u@{4b88hdP4NxdE6ts`f%up`~eoxh&c(l5xdf@k8kB!z;G&G2IsL->j
zl-AbPNHn(!qloaV&=XhR3Uc7g(^xCHwdk~7qq%mD6yAw@Kmnie%N?Bakt0Vy9<KJ(
z0Es_^guj@W*fB-LEfZ@LoQ$a8iE!4>a5YlRlyBw2P_4%&yk-Pi@0Ll?Q%e)IXaJXK
z{eq#o4*)nD>iR#Iv$ct2O}pkm#VuR5;EE@9<x6~5wwPJVYPa||+P%wCQbb@)LG@*=
z2ZdK9Wc51m8c>500G~7I<E*0z8uYDrJbnbKrALCOakvI;QMNl|j&l3jhsqI*o?m!*
zE>=2pSJaQaC$f5oW@ct8AG;2|{y$;L4-(kUfKTz;1ejV}fRubtSfJYX)enD|uOjv$
zMcH2UqpH<gB%LjGNobX<gt(dY`{{eN)QI+j1CV!ztjPg-Ia3eDN6cnX#HI{wImpV-
zuV%A-S9msjnNZulP0?y#w!gy=@7o@oG`w(;`ZmsTm&T7f7mhxKRj9iJvfJh?oaqG4
zuyK}_V=GX8KF9p4Z(()9)8P6cua&dHiF8KT4Il$^@TXOZ>rT&hfoJ`ZF{X1ZV+Uuk
zCYBf8vUec3C2s2SFO+rqo#)m0|B&pT|8_f!+WQag^%|g#!P~d`WhA1^5ZL>Qk><(9
z6lyZRzl!RZ<Wc8B7=1jYsO_2rjchOJ=+sGUyc0cJ;8E3@Qdg%Mj!@3jod-@t4Gq~p
zjExnjbUw57(+-Q62e%D(WMdXxPdwLJ@(|^x(P+;KSt+~E2nv>ToWnt@Lc>pv7$FYR
zr0=ub+(Sx)pI7R<N7Hb`&<*d(#%VFImOz*zwN%z+7ikn^np}ony910uGmK3N@dYmR
zLIGk)6qlC1nOT7fZow4AOMz|l7W#YnKj7=+VK(0-N;htdg?3+i#aK%#Tm}@oTXlQE
z<7b!_640R9kz=2w&p&FTE~^B7w+&x^l||b?4u%eb5}zPivdu?01G*2&QSB1yO~-<$
zI}1&-;(KVhxw&_dH9pKBYcV|m1+IA$*@4hwS!HE-^WwLGTF&QsLRZ&*+=!Rm(Wg0}
z8j)jRTe0_qbJa^a;+@dzP!lyxCNAaGGh@T^qcDGPSG2i#PGQS>wZSXb*4hOmi;a{T
z&lx{c(X~q{IFxu_OwB(Z3zp5mC7;#;RiCkP5oM?Hv2dnWL6QZRN#vFZc9?;2VJR8G
zs<CeltO$XRchE@$FklA06=hY9g{>IZy7q|FI=A!E{2TiVjnk7XE+{%Qm_?t-VXgZR
z7=_#UBj>)jWm#XU_qMKTT~C8d*3blxLmx(!zJ6_`360;F|6Fkcl%ZUvNm=m9bjd)O
zO_|qlT?v=7#J%WLWuxfSVk=u<b$GYAjUzs}Y2xk<D=)8#(uxY}5~~Yd;{->ZxyW5w
zW*1~@VU}rYAp<*4k7TQQ4rU`l#@fN5e43n<rOky5UvRfQ8)+QrXg_4UqEp6}(5WHa
zs<nsv<dZIo-dGxIWw<K0u0DY!BDed2I6YJ*>M{2F_0N}=gES&I5RYrk<7|`Xg<r~U
z>v5l;%vPJB$ZJb6e*SX_jzdqC6W&@pJv}W^uetELuOG~e5QqAkGbXO0_l91%nby;A
z>^YkONDKu@n_)D@Kvj6{C0bdM1(ChrW38vB$5G4nZlt9|65=8EnLC1|#+U(NZJz2B
z;$H8ibt#xbieSZvsRx$7^in5&0B)ZH%lV~(k!*6AhFfpOmjz5;p|K@EJxRAGKgj}c
z&IbHOUk>9)szYuiJgU=zd=v%)(c+2z3<jEFX?3-kCjFgs%<HUKN&h~ac-@Ee(<gn#
zG{AJ!gN#+@#b;(|*#TgTxB}79(VC0EIPDR}zS`hD+Jv`F>Dnff@zE<-E;+ZjxR`WX
zi}mB|#d6zru;sFF*X%^PU7iGK#8c+CIM4N$WDRcsb(w&angb5lxkI|x%xOr6TC?-x
z6tE*OFlqn{#P1+De-n#)2@`rx2)(_n^TD3x1sij7-Lq%UR`vR^*C;FfmZpBpg?dXI
zjtm38p96=kqb<~;g|nVuzgz0(3GrR`@tOG=t2~_>tMoNSn>8twY`(f{8>;Ja^VAa~
zQB_Y-Q*HLC<m6;<3=L&6&tWs)QkTVj$N4G}&gY5VYX?1{BA7RfC5{uvZ-WYOQX;bG
z<cgckqYN%HjNm;lqG;a<fc_Rd-uLcBndg5Kb_Ss}y2)6TiRt*X&{*is(8da!3|;qJ
zY*{fu(g;Ed?*A%{ftXJ?cRU)SWPk=bT_p5MAr`3?bXIu>`2v6Vp-hM_ua2K>T6+4I
z20v-&W`NdKJ3?4+I2*<R2^u{kBclXjwv0)}JA`~gF{oC(H}D$U6g^~s_%eWnRQplS
zpHGmuZBTW!>*2nP@Xc=szhJmB19=5}v6;3h_%<qHRqV%w{aRqf*}n=aVCyfuV9Q8a
zJYO#Y>!?Sm`4O0?FM$$CRAmb1-pUu6U%cLbxqd7}9+oT>l+nCM|HC|^^E|N~7G`ET
z=H|beYOd|QROdMf2FO7Wh##tmU@#p607Lv0=!fTUT^zKSm@IF@s{%!jz(mJ^Plv&B
zZEM2M2mPFk;lWHs?Rf?sA!5%Iq&i)J{mvHVVlz_b%KBy+F;}AjkVHLAWb6bKR-UI+
zxdC$o#(N)VF!N5;RoI(ofmaX_;5qdAfh2oYzuc*{X;>AJQ7~IqAZvmRyY?((^4>vj
zXV`cMTM8YgLKIFW+c;h3uo!=3s_%DPB32CpD<LDy7((;eNjDQB2s~<B|K(BiH>~I6
zn`jnmMtcU5=()aAb8~Ybv8Bc0B-U%Xq{L=T?tdNJ*qpB_^P|3%P3<W-YLZb5nWPZ4
zM(<IUFGw1KB;WM)p}dufs1;zGaVFNf!oF6H$c<Gd6q{%<TxbDUYe@c0EPBt%eAD^5
zoswD(YT`oH=H@xb?n7CE)~12%8bhcbaL=WCH-7V5k$eIv4-hU*>vmHJSmyNk4aG~T
zE+<N0-P;3Uu)>I69A_`y8pymgR)YBJ<Apgnx|yMijqjH><aOAv55sj3xKWT~U+4Ke
zP1P&SADda)(&8}cM?%Xz?Vh4!;Xm8WF~_;#^((lLZv{7?EV-31UFTJHx#M05_?*=g
z=2oeYqxiv3WgY*^Xx7oL3h_e5vIYEy>3h$t+MN<BC@p;mQMEp3V}NKR;ctA$I*)hO
zd8MVL1p$MFy`-=O58T1oLtyA&_zfj>r1%qo$x9H&$EnoLJ*N498!um;a0;5-@*;Tp
zz;lA(*5kz>srN+2jPQzF0oUnJKZUZ<&^zg?;yGBI=Wqz>N;CoR(4e^f<tbAl{|L1I
i3ades-hX%_PDIUfr2{=l-D4>DG1N0TS9!+y?*9OFV3`I0

diff --git a/master/_images/sphx_glr_example_inspect_predictions_003.png b/master/_images/sphx_glr_example_inspect_predictions_003.png
index 21a97b7cbf60d7e71abdd65b44f25d30378a66d2..078fd17e20f6a9740462d7bd3756ba93c756cbf4 100644
GIT binary patch
literal 41262
zcmagGbySpX^e#N~&<vdd1JcqU9YcypiKKK(cS?8jBGM%#2uQcIfP#QXcPU6nN!Pi@
z-#KS}=dW)qUCtt&=e~Dd``UXOqoJmRk3)?Efk5z|D9dX>AgI9*2+ArH6MRMJ`{E|}
zN7O?>&*QnXwTHKvn-%1lnTLymvxkG7IfIv#o4cK}lK{66Hy;;+t%rw;yBH6T<Ny5u
zZf7?eo>{ZwPv9ZgF3S4u5D1|e@-NDJ=`uSAq#*l=ysWlQ&fb!@p0>?A=Apk>A4K4<
zF}lMnXY40$ix@r@zVDXx?4{Z7V#>4YvcLTaEYEH$Z?}83M@@6KEQ|74(Z*Jw@2}hF
zITk04IU1eMSf+2_%;!d2)UAP`5rG5Wm93ORiLo#u1e7uS0-t>s5{~>>?5K&N48B6N
zZu1_(jQrA(fHs8q-w#o7K(N7g5I1;a!T)|ZFcL=w`5hu@ISdBmccSwPQInC6=^xYh
z|9?z;jwgg~RfNFG$_h$A5GfBgql-fRa$w??@8rx36E`=3mfi5o$$N-HWL+IUGYd;y
zqAc=t6Y{REe8j}W1~;DC5MenN7hcF}Ps~R|3Dh(vk}hxS=*Qhb@_!OgNvyvXbWs;5
zc^fsauA7s;{-hN4=Po9mZ$$L>X}<`m>Iz64GGiLI@gd(8G4?hcR#$$it{7mgJFRJ8
zIXE0NutC54Sxhl*@<o=*<}m%y6^$IVv~{(&ne9Pif{@BnO_W)@bf@bIQj|a=2x@ob
zi1=atl1af#_nijEROJSdw9CRpRL$M2n8VGUQT+WzszwExz7o7xAO4>f-$tQsjn|xX
zZGACi-$6T#R%W~>*T>(pjJUq7;^R)mE#&BLs2^IovA=osaF^|y^<!}cU1b<;Wy{`}
zlZSoQN=MG!VIN`2fB%IB2SY5at=GFl&?=c?{xc|-z5e@MuYS*BS~HqHbKD(He3jUf
zb-of_TNm$1i^zIbt-bzMG>M6U183LdHQMyN!E{Q8V)}o0yS_)aQkV*uz3(5_Cw_+$
z$Dpr%x|Y0s2e~;Kq(7Q`uD<p&-DDtzNol-Di7=k#F@fKwJWoMQ>Z}ZuG!9B};VNso
zh>H}nC!hG-zi!N4EwPH!h-QN~8dz%If1#ATC7eATy<h8oOINw!6Y;VHb;|?1#ym?I
zKmN*Zn$JKH^-vqqw77DA=e-zof14p4B*o9qe+b^q#LG)`e>Kykr$ka^=%=dq<Ts2G
zUt2!4iG-9iR@_=5YNviM`F5kL$Nw6ZlB`o4mpB+N%^k8ft7#c@^-zRxP;BDa^X*X6
z4?=a|URgsjIc<9VK&=gmSj71}kX$uZ0(z<V??p<F164WF%>QnGpKbR(AnO`fbLqC^
z7xHC#_NAC#^+24)UiXHCb_*Ek1+4XD%mRx9YWd5M#`^#)r?N0l=6|AHdn>9gLq-*G
z`tcKrp%LA#@?(Wm@}!-cxJrwLXzc~8w;Y_il#diSgCA)d8Q~`lhHY<ev(P6cCMqf@
zd?xGqRn@#4s`J4ypopATxAaFi6JzIN;mVs>PttW8W@2oX^ZbzQQM8(!PmfKjfxKgq
z$*IyrsfwIiDY94Y#0ebczKD(D+pkp9A1CHm(iJJ7GI`W!>+22LQek%l4)$%~cGJ}Q
z9GO!l3=A}{ubWS&pxdinL1y<AX=iJ5!AqhPO9}fMTnSA#C=S-g6D5HWPr%zU&_{e$
z6(|!Q$e!7l#2mk^;TFZ1KKTi*q%4QRR!t@&hVF(2hr=0Czi8;|&uwm`rKLs3#I&x@
zyw~Ok4-XggWzv0IG{%*~y@=QT>}6R+wDb7RJ3gBMcQ)x-67fL&0Wku`M{wN=mti_}
z8L9`2Uk!WR(zHCVWJyC%T=a5~FXEgG>NJ6Hq^bP<{R7Le$-(Z1a?o|Q&I%WkfOd#C
z^FV_gb-Z;$zS8)N(eW#*Dq9*#hQR5Q76<Ff`#njY#dSXf-12WQJEc(L+S|jPs&>6W
zJD7qkQKitjC=@=LQtbiVjFdQ?e15T4KX!OJW5UG4Lr5t3mxzlT_t#kdm{``>Q0!AB
zWo3!F$KyD7W^(Wp^Z8_PSXNHR)WIi|>@VG7#Eg%Ksk4KB2fYlPm&2;vN^h*(5X@$I
zgd&7Np*MKw3*LY^k|#-X&~ZXw?DzZO-@Q4K!9=>gNPHS@aY}^K=$mSbf|;$N(zbA*
ziNOJn2Z=FEe|p-IR(igFiqaosELI?`fT-r4Q0^qgZml?7zv&zjcZ>CmC)dQ}Fnb(V
z78V?S1zB!Udf<yr!DAfWx7K}kd!3S*`9<u%u@vRvjJX3Z2fb7FMJLz&j|`m;CEzH{
z>i*OU*E1GlZJz%$t(8fum4>5MFLL%G?ua3pg<s?4w*tdL)W6xN$qfk%jpX}Um=;Ss
z`Spp2o0q|2P;b>BOzYP$;n%lRB5DDwTru}JzbC1pnq!M3<6oAbZxPRQE)=lg<E5?t
zK6=U-EKjP~u#1tmRm<v6vdh*#_N0V=g!U8*J#x_e^xC<{j7<N4XT7<OKeG1?Yu-Wy
zPHqc9s4W~V#amP2(XuTu6qZ2sschrsGPfq<^VCG}Md^{`08{yaUghQLBT9~^Kt!8u
zsEvg#Q!`-{U%ocZ5`-$tUBy;YFjPT^HIs))Gv;l`b$ITf5<H1myz2{*ET_Gd>`6@5
z?PE1TKi=*Jp)kf{uskd}+Y_DU%PM)TBE3mRJ{DWDT^cg}$P~^3W8{mD-41?c-)Tb^
zK?DhAVI%b(@gWe}a<2i$FL!<jYX~&L+~^l}{fj>NkhRZGRsu^Fm8pR@&K5%h5hW8n
z8-E~(hiSn_a4<J0hQkPfw7v{k7~lz3B?}g!za`_2$^}uRnhh>WH~kI-48gcrme3(I
zZ5Q$WREmNp4DyyB1UT*ciyt4)^!_epHGt!RAf0;sOvxdgt1|EaKg0fmo)UbMU#pyy
zKLJxXEX4DTwPrYGpMj$(T_5XrMk2V%t2E60A4+E5>VoPL5y~V?6PTv+AV(d=ox9jx
zJ59SnTUC{}@MvLA#3`Q{NL1kV*P8prWHg10Kahc|$dR<ml;8{`KC{<TVl2k4;>JPV
z|EFksjm4#wrX(ZL%40SQHdS9=aQ3IyZvuxXzRM>ECqwiSgn>Y#2ttVX+Dis4-@Ks>
z{i58w5M9Ie2I-~jEp$2#7yrC;o<=@EH<*Por;lKC!;J1{o5z<gHPPD)(HW&7g0!ed
zf6j}6B~oQ-NB(|oWATlLK4c^O-c<1>P9EO#DC|H+ubo@(^v`ZhCSD~I3>7k7+#u)K
zV>iuL{j^#u6nUBgWnhiw?2u1E?O8Z76Qv%qu;`@S2z<^ibGnFRPU<B)5tf9UC{OyX
zup)ngx<onXl{Zx=ZOWeJi|tUeD&j#1B(q8UnCERx;&%xTX!u2F?LJnowZMA!C|zX{
zk(<IA47=s^)eknY3r??(M#EwRNA70W>@{hzMH=6WSy#71NRYxYEhc#T>PRrGR5G}q
zb!1N*w`|NIbjccI&}2>=D72UEM_xop8!1iK6?%@cq8n_5DMh(pe8;rB+8PK&toU!(
zWu?f#BVkoEn3BM@&NkXJ3<fi~5~$2pB1x^^T$IPof6%wu9G^#0&JiS;!BYoa*imW5
zHAKP<R8%4yp<cXgo6LGVI^YsqJG@oV13kbv*5&HfCt$?HQG58q`bvzOSp(kmQN@89
zHNV27`!pC64PE=2@pc5!PcbkmEncRpvgd^{@Ry)INEgvLGh8F$1(>jk-DnW;Q95Ft
z+I5ki;m+RHoM$BK7`^WPR`kqyz845XK*f_mq#5@?dYtE*U9IrgEe2>uP}92?%FM)K
z7^(q4kDucM-D?7WvbwbAf|%s?EQ^8ePsS1LtI;j<-Sd_4GmMGS(1{6GpFv}!T2ooT
zDTrg_X|Mea3pGkmlX)pBr{5WUHLbk<={~IqZRP&wW72r797Og~SDY^ivRt3dD_W-)
z^5zoX6(<nv_q!9x(^W*a>d<1zZWBR|o-oayqK>)Grhba?Bt=v|g{PQmb=gJli$OoE
z_@kW)`Aw%p(~p1d?=xn+c|z01d_;?m9g7#PiT<s?ju?7m_rqU(@bX05jE@HCB>IZQ
z*5+5?QMJEK4mQS{jKG!XS%EHNqvc<>*2Ho0(V>5re`QM4CnE9+O)el4QQAMb+-!b&
zOxSY*pG}p81`|0lBk#|l_UH52Pw5|>G7}XD{<i3Mbn$uwp+gOlnhc?K4ERwp9oLfO
z5av+g!UP3Nc`OHm6Ukk+aKUT@Rq#TY1WMA6#MAswoUy4{@;R8^wCE>PYhxG$D(c-Z
z7&w1C(V!p^c0wrAohhQsjF&Rj)n`l!NY73<Mn2RK^P8+syOZAOGEEJ2MAwyUzf!d}
z3URk1#o1cX8T+j%tVcGv8(L*s8sbAk7A)@qqLlSnc0$KZql;L?ErZg~LBbzFyi}!c
zgIDTzkwxu<z^wu_T&@X3I$jg*QqB}VRS7)}JtV9|50C6Nu&|&D!{&g*Fkn=Kj!V5J
zo#wh`1vhEX)ANz|1@etTl(6}EmJm{q0y60Prql$XKImPnIvnXSMTDlD1*pl;>G6JU
zjq=bMcM{1E;v-EATbIEG8`CQG!GVubm|I6RiShyR%9ePvQ!he!s^6?J*67ga4KnD;
zVlcoO0Kk-W{lb_L{{p!4ggFSM0*a$b8rsA<UH9a)<N~|O{YvQl{k_`oy&oU48hkWW
z(Nk1P4qI*yFsQ-h$1Sy}N>6hWgt`LI6HmHKG}dVbJZa*J?>1_G^+^|1cVu|duA>13
zuIe?__^UUm|0aA|SQDoWfyQg}Y5xr383k2WC<QfX$VH3sY|XJf(uU3CNc_#GCs}K6
zuY=6a=K-!DN}<V-BKlPy0X#94Tc4IhRuD!_`FR{~ei;1zJKm;SnOO%<k(%ch_!`))
z)Cr1l;x^o&jZLV#(yu9;%2GX<;adC=s9LS4YH==x8AGHvnzX<WF=H*LqtihFC~slC
zvfy$xqBy&M*Q~7`5T(u07r%DSAW?;-^XZzE^0e1EO*Wjp#X^Y|b^%3LwdfpT$y&B8
zi@esnki`9zfFlE^=1_o$dFBL9J(VWD;Z#9X+h|AiWbhrq9B$7@_LgcwJZa$weWS=8
zwnH!8tiL|AOF=oF@%F^Sf<(?msL9b^3tG;g*fF1*Ows||7<z_FfIXNiKvCFP2c_w}
zNLCz5A^#p8*=HpfOyNS`e)gp|A4*JFcrt9q0Ls<xnTwuW931L}O^UOG9N~gdsw{&b
z7QzL~Ns#BCoY@9-7@U|gKe!DSfU_13KHvr8uS^9lRwmhj@og(V)Ee3I7N7ojL+bRn
zDuZZ+x9tb7D$PBSaL_5^nd9P!7y&Fs$_Q(tziA2RAdM1Wjbbg?E&E5{?V30`yUG9k
z7vb3zFbeX#G0D4xesJ_YJvZvEz)L=nq--F67fp6LQ_-MAaA&PZd$XzvVD~o!z=RqU
zn|ARN&rO|$tKJf=Fs6QS-OZ4OGdI!c+^y{L&gj%9%A>1tq{^1+^qKM3Cc3@I@>o+x
zPhq!o5xv>Ppa4&pwEmlySij}nZO#X>e6()DK0nU$!NsJ6dp|-fVVn9yIdra;=a>Y}
zWKQB41njqF?4)OTj?XL^RA1ap;U)v+y~4>Pe(pckzhgxoycOY3oR~~gh?_8gl>aXT
zxA5x|DiOj2Ip%blNh^@=dmXWgo~eX4D$ime?Xb%p6Qm>9CuK)kc;AA>Nftebg#Ejl
zZV#Pk&WQwJNB>eZwLMS)Y7xP7uVZcS3G_sv2FMUy0dsv4<38&Lc)$@hi5Q_$?ixIC
z?;l*MGJn}zn<4Q^N;u$(kan>x7Z?z4)~4koIgYn*4L<@m&bS3jws?NIM*C7%z!qpx
zJz1F<iF$Lsn&6iGxW5vG-4juhM~Pc=Bc?KoW^b0#SJ4+C84Xo|Mp3Kb!Gpba6YH+3
z-B_~JS+dVjwgtFfwgg}QX(!yDZwxc2QZz82HV(YB60w7{tK(>66uzEkt2DafLk1`6
z2t>pj3B7EvFztus*L1?EAwa_rSWI~hIYn)v>kB%qD_lRgvS~PhwRAS-`VS1#z(6UH
z#6H&ifY4M{^>zCD3oNO-3k!++o0H0-6diuSa$Hgzd@m<sBv7n+g&2#;5V>)xyuYBM
zo#$Rmqn{9F)?z2^VXT?PaZh?bP8iA-hPK8l3M#$lQbv$Odg%uyftL>l66xY5C$&{`
z#mLm9Zw24-l!?}LIv9+xvvEs7Qfon3i7<HKBaixP<YoBR=hw9W#$rg4ylv=A9tALA
z+=~~)hz>&_RS?p5p^l?vxZRw5)Co{rTwtsXZ~(*X{tpa$?}&NisA2`NR`<h;?KZ9S
zyMHwMt=AE%YA{m}&&d083n_-LE@t%TA^B6mfiEEB<mBtO=PQF~gc2^_#@t0{c{5d?
zEl0P6V=6F6RkMPsHw$^ZT!8HQms3-6L{aRe{b-`R|5(w*@D||U0sbJ^9oPL`1dF8X
zIH(j~VeFCLx8iesMRoc9P623sCSZ7M%=uWTq@;vWI#9y;TUNyNLDzlO%KiT-vY)D|
zI((y#YWS~qzuvTdYOQ-T7a8<fBKy4ubEwAN-Sl%?-v6cVS$1NoQcBH=tXPq}$5x1X
z6Ihe;$c%2Ig1R&$Lx0d#;2jr)g8OB3M@NT!XGv+PaA2UmK9yoTWk_WuSHRWpvYFev
zI0~M~@3syBoZM$0e<gQX{tK^vKh`YVZlz~kJWk^vYF76}<!>eauyQXzOMcaSpw8cp
zC;R725=562VLt5XxHRzTg+@4Lq{@9JCftT@RZ(715zfzFr=M?VU=RyqQY=u<m%2C}
z=3z`T?)vqjTw9}D`!P)_G*|@|tVYLbdRmv{vP(C1@l(-cJqXDroMuPT@hSaFOiKmj
z+mi34clC<#!v`Mr=&+1<aq>sdc9g9#L=(}+38jP#wd=(bWU^NlHDGI2Lxgqh?ARj-
z=@W~KnLAEK#ot#{AQTiZI5@)nNppQx3N8qs1hG9$CP8KZ(h3U+)>0sIcZpJy`E_0X
z_S+hIpdgLNh@Q%s`b0N{uTVO!v|fi7OjBn^AFRiL1v||*U`IqmAPaoJ^AXX|EL<qw
z48L*O(9rZHZWQVbuu@gF2brCB>utoUD1p6H0F7F9kWl-n5@T*Vw23aB<<Wyd0-O^D
z<vu~8`^($I->5Fj9loo+cd^d>UH{+c-W&%ws`)p-%-*0h!#lTJDg*DI(JuLCcrs7K
zVtP}@U;RKxLg9-s3?XX<dKGpU$wc`imD7w5AL8!n{)_Lbcu)U?&HW&MchQZxcVR{K
ziSkS{79!N`>LMQg_)I%HgKaj90Vjc4?ZZ765uinHz7BbPT03f~^<E$c>D>%bMKD13
zBR^!&Zg4*64xhU{0lXR~bYfv+h+yqB>}Fq<sF$29xc^{eTN2G?%rWY-xg{t-S(u60
zORc{tI?Vhl{%BW%-gq*vTcNe#Mf{$80>H5{75ics!|K}O&+N$GX4_Oz-IpY>fBPDm
zS2kDPAB68bLtkGi6@%+PV?I-n22ceLPE2fskqP^dISbn(;~ElK?H-lcoqfQ+AQ72K
z?BoEWgam%}%w0E#N~##I8d@x_>RmGM^{$VkSKTfKQ}aKj0=+6I6n0kTc)l5KQ2#?%
zFGC25cp=`aC;wvdO4h6|`pypb`cU32aJJqzph{_A^oCbFH>nIf_oU<q7q5;J%oi@=
zOtivvCM^}_gEhg;9w5O<@B1=9F6XnPDD&2@?2DvzVRExp`4RE+J2E)s9*Z=+zUSF9
zZjbb&bxXx;91Cqd#M`AzhaJljJqyS5rMTU|AOd3e_zzY(GEZ~e1DXYaE=vM9L>mb#
z!`uh|gN+UiVv>KQaHa)o<D@_H;c=lM2%W4+bfwCy{<xxw2I%dJPlZp63N_x>x}S&v
z^}*1~K{eX<N72WRu+r~bop&dqIVoOH-6bkfq{HzsID(xB^uIo_0jnE3YA7g@AQ{u{
zII(er-If)g-XrRvoC@T%R`o+S?7BtwpK&&&i&_j0F^cT<;7Iq3ANElGLb@qS<k<o;
zM<yl$(CiRPEz#1`Px$t4gP@5gHHmW3rs!Bv8>5+V8XtLjBYC8P2Y*VZ5(mh-_>kr_
zHij-ZmpxMru6iE|o;LIX$0z6Rh*kzX=bnt3H0EM9@4|EHG0+5@jX#rCKxsK{ym(aa
z@=v~oNs$d*Ohd33_k+kd1yHmyrVaP_i}a9{-h~6!KhQ$;I;2L=-o<5$8!3gH@41m?
zSAa^m{6JXiyumW<sw{B&n;$yZjD5iOjEst7D2yRE=$`<SSAg@kG#)sziF`sKXb;*$
zV=R7%Xh>;s|7A}36z6C+oxX3w3&5OeCXPlDuC38JAb!9`B>tgTmM3cDhP)M9k%l5g
z(N(1N{(JcbI$-@$OgeCwi%Tr|MV;u}r4i-}W>f7S`(8ecDhpR-Ntdh8?y(cDPn?jC
zdUIC;Jp|mA7FrlRD8NyZlQ}6#S5RbSHD}LuagbN3tTj?J+sU}DJu*2kyZ=>HxshxS
z{XK~t)r_+%TqDef(qz8&4E-qmAJTLqd)}6yq@X~s^J*Bt9QhiL!tdZB@JS<)h-WJS
zn4=r?tZbiIJ`8JQ9IMqPeh=3mL{@Dn2UtI3094=PC?;kH)l!%zr&P$ovnn2FD>dMj
z^*0`bv%nu9u>`r?qg~}qZ`4LJm#OS-ciLUHXYb*1X4OQ~$f}Eu%Wq``O%2FmhKTmP
zoWZ=yCcX4G;GLn5s~7ZC9|+NR1)dSo&LYv$v3N+#llPL|0CKVj{yq}%lF`GGy|Mad
zlrINsimNrSY+cWa4_1g7HmxHfKd&VMlQQxl&F;D;&gs_4xzS(Y4s%7ATC9cp5Ha&g
zpuc^EYzRU4QD0Ot5y7c(B5>+iB~Tl=QIY4?L1G|Ozz|4sQE&sIAzWD22f#BoQIVvB
zE*IcO$>c2)9Hx%X*tzp4rXS%U!c7B`u#qUo;o=|4sU&%Sr_7OtsQ%t<&0CW=q5K}4
z76mwv;E$3Q-76F!*XLY)Z7&Ys_mkX*0ev3^p58x=?!rw{?#=<{_I)MY6nLMoO2BxL
zjYKo>8ZX`owlzBRlp$uRE*xbbu}6p3@Z>FD(mbi?@M65Zm3rDif_ofDvhucWT=LU5
zUm+uTe<?{r81+u?vTLTY^^}ec!}uc$q)hboO$F^xC7{Hmimwx#^U+If*bxTX&DSU=
z>VM=lTqp;g83ezX%$rTQr4z2aBvK4}0I@}rV7z|_B3JFXC9{oMe_K#zLZA3~UE*ds
zYA9m0b`E{&qh>UK&neXT&Pe#o_;`Eo3-M;&kYbb?TA?<k>FiKDpVig#_Y)@Fol8&6
znzI?xCO%R!b_p?eivJq%A^*wh#duwIKmN!JQ;>aM!-yegS_c8nNJ<5jdkCbKyz&LA
z9kL<8#wnRFY7_~vojp(gZ=M7VYbEl(b%~q9{Wy`JGq#Q5aA=YWSu~O}$b1<yMSp`q
z5LaV8dDRwZ6v^P-?lRv41<bJa0xpI^=h9Uihy+f_qsAf0$$H@tgJD~~%FJ5ICH%$S
z+fn-N#t3<NxR>KS;b?vlJUMv<<c5q%XKLbW(2yd$k7sHptV!H_+)JWm`iM-&wI~)R
zqO~K-GhPuNP$`5AjGH*baIK@&*)5kdOKfK1CmVyld2-Rh#C=L*a<+Q3P0p8*W-5qq
zM_Y6Ck<2?hl_ZDHXC=}&ugHdUpA>7cAsT9?p<A0(ERraL1iyd<4$+r}fb<(padW19
z_)xaTj8b3?B|iTK`av%|I+Q~uOthH`ozV1Klyjfm$v&$wBlOF5D3hT#=o-D3a|_bL
ziKyqGD!`3w932o-{O}G~FpY+eJId668|0Z`pEC1*1LV9CW3nYU@~RHh^*Q0u5lB!m
zk{Nv<O8a3Yh(zi9fi>NiGW<Uh=%<&LAfisNuypHG+RzoHXtQ99U7o!X1bSKw7Q*WK
zT>efj;eb_T4{$NaJVUv|0TKL*TD+hi-Y`<LDU~7V$QVm_7faX0RkTv5^G+Kp)LnGV
zwN+0E1)^-lf0$E45Go_1X@CYE*pNKDA^)k7U}+v70wH}v+t*cp9}<QoJvdiCr*|@}
z#MbFU49VKVXAT!kGN6kWFO|dxY!R}p#og|_5^M30?UQ81@i8bZiem}nUT3<1Qun7;
zORf`W+TkOgukk117K)pwXsf}@u@hS>X(J@of8H%wls9k~Uz@B_tam#st>H{OifNDN
zR|I{NSJMvR{Uv6fSB%Z+23b$-R_M#l9<<%DHo-j21@X4Re-Meu&_LXCUCxU9{5{_Q
z6Ek2t@0CTzutkB`46DP1xv1pYR1ukilzBCy+jARU;kJF6n#MyS?2PMH??3bt+Ak}-
z{)`bHbVxPr8ZdD*@$$XJ1~|SYXLWeB5!WtdLgkcBxIAWUB1(a0smhm6cXZfCX0&6B
zTp~CWr)5VyI~1m#dc37GRU#XkJtq?Lo{16K911=9bgGJL4z?`vTsyn=HG$_zF5(W9
zZo9eTpP7Lj*8|?0@^sPy=(R3=hCxIs))Nl`QgNg_H*7?(Y@Qg+oLGD2zMmX0RLSH(
zER^$(I3vPnUte*rdN^X^)iGFlUaq9RS+Z8Ny41&`3S{jo4v&tTdez$|Dcx&Up)egU
z8m+8q1uUr5NX^BG;!T#!v?c^Hokh*ohd{4)ZfmNcOrh%q1qt(`mr)IB%eT%7$S6lz
zIStpVQmEp<p<`-eXnY>9kuRDnobz>wz|#~8ho@4OGRxD_df`s*n_$*=QpUU<d^a{8
zQpt1~VT92RTkK>40&`QhmRpd_1-?0OQ?;EwmkB;?9Xn?54z%BY0%@!`LQS6zLM>eG
z{d?&z3c|z}h3sd<cD87mf*?Y^?`MO>QV$vL+Z@G!15HZLt$?1&y9>1EP6U{Xd~wKU
z=G0T3Od0{nRSbQ%_`25vK8fj?DqJfy7~kdrLS+IqJ+`ZXTX#AAP4X-iEF~s+_*1)X
z7?H`ES9{Ic(5uOuVcD1Gg_L~PY8*H7lw&6@liExoERmk15PaX4m;#e#%a2@~B8!$$
z64WT02T{76HVVG`G+P3mDcM@g$}@ONwAkCRrE&P=UOWv%={1jeltU<#XrP=A5gxuF
zG3(+WS%E{q*fuPqYIsp$Dm;6gJ^vrB#vtyNOG0@A=}x%vG#|CTeLNIT7c{kp?A#G#
ztG6p;AU(Dwu6(?jb2!!b5EdkGTvhn^x$Lr2&X<*_kT4rY;ZC&dns6u{pqn$=smNOW
ziRKC)mp%%>(4vw&t!a?9!D>sQ>=jcidd}L6`|950_BVd?{)2De3&*`24ZEwip>G0l
zv0Edbel`I<qDH$E-unZKA(a!5Mz9C4Brm(y)Hb(nv{+@siL1%bw<*{#bkKb(F~w3z
za+K3kp-_E&{BVd&Wx5s$+5LP1&F9Iv^(sx>;<KC91NV8cscpVk)him`pFbBqnTXVb
zbzu8(v(r>N@8k<Omjr=I3AIxa?q`KEm%Ba9{sB*!m67u8@Ai0Y=d@ET{L0Z?^9C&x
z50#o^XwqZZlRiNrID$Qyk4<ze=r(10lD~!@iG22zSfy2C6YOpMY2e;O;JP32J&DLx
z1m&jt_EX2*>)TR@hY0P^$VmUu#-QupZB5WLMUs~qU0pxz8?#)h524meNBr+BsQpq!
zX>}&2-W0Z6kRH1=7d18EZLmnj7Zhk%w_=+)UVoNj6KjEG9g3(19NN7XY@pY%_1bb$
zQcmYanw_LSK~}@XZjd?6+M<(kF>w#+5JW>~Q^vgRJShF96cTVYX9J_4h}sxT0%Rom
zo=RB&C+8@Ff*Q?e`(ACNTstH>=BbkEOR<TC^S<N!WtpQ0T8}?s4c`PvdJWR<;KG0I
zU|$*N2A0k#CmF@2Q-)qlZf%+UMXagb4@tL@NJgGo(G_VrnqkA|qX0;Z-tu*!CCnVT
zX8jrQe7))2>F$ih#z2D4R<1X|n5dp#cUoS>r5_%mi{uLL&0TH;sb2|`QIp3t{OuHR
zzmmzed??Qb@aW-fMUd(4+7zYl3_ai7a#Z;R32*Zc?B$I}xQ597vmIP+c->QK;NJ0N
z&G6C>E(F0D6(i6j?S`-CXP0E%U%#p#Co%MD%`s3h{wC!?(Z)_h>}nAlX(o-$CLj@`
zVSejhL$j|DxHh!(ph-N8mLIZo1AkhUxV2x9F-jFarzv9%$4{-r+4a7MjBX4h>wC-6
zkPV9w+eL!dLNsYCLWi?RfCZ>5OPsYga57Op)vOdSaZ@xAs#5Idi@x}<_tMZALt@MV
z)b2iqJy_qHolRc0xD@J}Y`KW6x`<G(JkC_DdXrpZ>6uZxw@Rq2LLH3Kr}*;3UEJ#d
zx=g6tweFhbs26}HjNF|}mTnl_IpT@I5PV>uOTs<aX8q>Erb~JBYYkbvJ_B^nICU^n
z(GYxONl^DE2kB-NH>l_~#YptJWUHi+5eb4-p1?x#RGFuavVuN@+n@m&bUAENeJSee
z>pR=x%Cxh)3kEFK$zR_$FZ<zs?7Z6fCeMu8Ty#M?O6sJjSoUN)S7@~@pC|Q8%6CsT
z{x$*yg-iV$Nq3@$3S~s^zrT|!L)m`A2bJyVoqLRkemv)BUAb^AQ0YmgyeCfn3CpG~
zQQiPphlu?vBGN}W3)yAre-K^leIV#teh`qYA{n)_gotE2_jR-z*dSuraJwULDa?WJ
zW^Qr-5$LegM&^J1CqiY)gG+a%^R+85sP)~CU7hPQ+{PW0>XN708`mtU*fB9Nk8`tY
z23;RiO4O2jwZod1P9@%aeS2Rw*k3AicIV!1lss%`6NJfF2~tVrLKki6@rm-D0JX)R
z)_hN9x3E_#qjI?nm0Ntn5pLlnxHj-7$k6qhCFhI>V#?nFBVQIPQUC%}L^4>%R~8n_
z{s)pBcUQX!2kB{PW{WMZ7hv!wCDZp*w}vBar|w-!bp98EI4i6mnShr(syTuAx}B_K
zLH!gU6;?`j5I0h&F#_r)KF9s>w_UPUG*R*K*Zrm-hKcTcOSYd(hT77aDyTgiScK<H
zIVKIGwRW5a9?0th8cRz{EA>1ql=^DBxTJ)UQC8%mnDXv|+aDrs^8LPIJ3=A5Ed=zE
zK4~xkUuovu)re-l7ft+cTobZM62Vs2x40BUQbF9v40%!yn-EF^rBwy4J#0o(gflpe
z;#Kl^^=a1oYY%`rhd26<`-T})*<G%eco2#L%Gj-VV1gy19+p;i##r%O8e|rZVt%S!
zz5b__0|soN5dybpkoZAdzRhJj5Tn)?A3%atxpRWf{}Z4|o-RK<W4@ZIH&ao;@yW9R
zY;4La3sD_9{n$w_IX)Nk-0_E-=RSFZrOypD^XhTe?3P@IQ+eb{3mow~t-VKu!##d~
z6;mWR`)Uy0ZxY0*YgjI>8+#~dMz<-hYudNp+_W6@xC?MTL94zGXmGq_K8_7GEaZi<
zC=Y@;m(bo5iKe`{goz@<!l!%6Cw{MAzLu3z<Ou)`9L2Xguh{zJ0Nm>J@Z&UZ$K47S
zguLWxU=;dvdY1>Fg4E|f4d%al=s~ABZq8epFh*XaSpo_Rb=)E>xEN<aS9oYP5vrh3
zQ7+rb<1S9ODw$JI8nl@=64|8W52gs%q%~tMdDK9-#BPp@(wg%dGXse!zxG1|1OtO+
z9FPHj=_H~7rH5{%bi_wWw+LwzM&0-mM7(2aR@I$6UCS9n5h@O~xx#{EsO5alu=!&R
z9+lk_xq8Te%;5>b=n)Nn0_BlSAXkZ~DN!0z;a$Gn>a9X8?5#9z#>y$}^wpUqX~YAf
z=<nwecnNBH*E!0&{COj?(~F`=J^4Q1PQ}>KSnGh)(tLwLVLH3bbwB)y8z-PGPm#?z
z9cKaa%bo)m-4fYGsn$n+1UDZwM{VE?VX%CtS@W2ja(K<z<CN(^ciJ3JPiQQ$nmx($
zC)krQy<gBEr)4utrP(!s(k%jp<IRRDAB@95x5i*@D{1{E!YIRK7!8V>WCNxsWqD_2
z%$NSquX*Y5`TWX-CY0or-BpxsN}ys0-)?Mv*&p{(2ZcvWEUl#Cvp>VY6l_i$Gx$n8
z6}yc9vjU@Q9!xMvyhF{~;=BIyFnKr)%v1E{{0|uIajNJk`0Y-V!d%JHUc4`)dXK~?
zB&V%gydGyhKu7PTo`+o|_8Op6R(rr%nmo6NpM#cSNt1CBSyUiGMTHzQfp!p#e2svV
zUAfRqYW<oI)P%D6h5FzvAvG^MqH8?XXq#3g9=JFWl7{r!iT3-79D;<15Wz1HO+*XU
zS(%UrheVDN9gRGAlb+u_sh`9sB9a5<XENMG_w~G8FybGU!B+g46_-Tw&$|CpeA`CN
zhCrYT-GcERecOrfMx#NOLcBG*LZC8DLvtBb#4f!GU?re=eO7~Ya?*^vym<8o#;EMy
zSRMwB?0cE@rA)mSO%H4NBgbogmaBb{GQ9U?3A3sj80XJFGD&8)NR!vDq2e=&@VwSF
zXsMo3hoAw|3|V-t6WtN&xji1pbwvo9HYP{R6A$_x?jGTfZ_&$pq@&=U`2GG&)Qa<5
zR%$@vgmjnkS%x|V({WNWC_LpA5+AtdXO95+0=!Kr-DdH@BnqfpSz^rW+m~;=lyR=R
zIa@>p<Z5AR;JE-u`nch8xOp+XlXC_E1*e$bZ<|GT%S#NB7ln9~^U2F83XM3Els}^8
zZr|Wl<+vgSz^nbOsGMRF6K$-)D8TF?`g<%~k{4)jrg@9$+_CTbkJi&cBzYT3$0pf?
zZ^mxKdm59@pgR%c?e@C;*c;v=xBHQyQGWEdBlL^yAUG&NDn<o^nNTn$q&*=_@yFFq
zD$tYEXfhd*U2ApRp76!-!zZ0Kh9Ty^PS^Wuy7}ND*0uP4e%~sD+Aet!^*2+xyYp@5
zi6iCd(?N7wmH;LhV=A6+5&d|f_pCfum{l9avAz*$_|ONVVPu0kJwNNr&K4;OG@u7B
zs2Hmp{lCLV>2JtF%=j`cR)1*N+3up*QRQu{3Qo$BP<;_oA(?)M6OO%)?s7+fM?aGd
zV)j?u56Kgqj>d<e_nje-@!@E=kUk}2mR?R#-I6XUSy;D(r*{B2Jyr4N41anB)96G1
zLbawoZxLm%RsN<=yz(0odZmo+Wm~@Eflr(lZ3be>-3UvuCtwLKLY_rjYihYcQ*F3p
zqIL8>>S5bCdN%%?t+~chw!(?;*bPLFPyj;-@_z~-WkH*pR-n{M0EhBPL_7G!sEhLf
zPt3%$kN$rA!?Ws#mH9ar_+<+rR#EAQ61?8M_H-G=GU^z5zRU__3|W=n#%n!Qvn^T&
zlVDQ4S{|^q-{(mDa^%Apd21#HjEj<6hoK=CkbuK)Ue+-2wK_Hhj@NVBx%+dDZfGyR
zYEY+%88Kth1M`O8DDl%~eOw;qSTkXM%LzNYAO*88gnTnE<O-*-|LA;>Gl^OiF<V;W
zBXdwxeGD}I%l82-){v#+kIPUnTxAWcrme$aax&+0{6fC0sVQ_<IlMtEaiiw+ubOtw
zJZ`I2!X1gTtfiY6NP1ZNU*@=5e$&6L;CK)B0KKkKcp0%bH|cbq`|~s^m}abS`vbDz
z;%Bq8<iI*55pt=n3yYs87m_X{;yy=zQ<~}32%s)xr=mQM8iW<Vbo%ty4R_RWov)Ps
zd~vY#haO-gwSGl4imxGORKev)na5;W<p%!V(3bnT%%#mmK+HQUTuL7edGD!1i6mq?
zf_oE+j3#7o>x06QyJzV?rq02iHMma5)ZsR2J|b{}G*P=IL^;g(x#w7><Iv6zuP7@J
z<EtV!B=mgE>Gr(PoZ5GN+0rIQ-d?ao^~FA84OI9xM*@v;j9T*^QPN<`-G1n3qr0qz
z?!5;_l1U`S`mE{@bR3mw2f*MlW+2p%(O(O^2i4Q`DBjrM`Y$pn8}+;nvHAy3d;5O_
z{!mW&!?>=o%vnmKi$Q$!pGEo~d48F{{t~mOmbn2h0&}jmy?QU>yP3`C8X(lvGK5r9
zVL<hhG)qui=XOTuUuA_3r{UJV7QF@vk}}EY@o_yc1$e~Kyb$w^wOPi*g>q+{&GK0;
zwvj3-&{7hJ_PjDA?3FLJp0_T*CX9Tp?<;UBQZC-t<Yh~66BbW2Ezzih<~?=#V9d{`
zbUBvPjk9)>;^n~pGC*S0{2T&3w5PD|osU_h>!3JkHD`vfd+iq<<RMSKAa#B>>5G>u
z5b{6PzqMYVfxUZKevH5;e!g&aB(E2QCf`u;j7-n-!lt;1CN_jg5ezd#>$>D%8RB>q
z98U%gp$TqFo8iF!o7g55hrI~{K+s}S1LUODMW_EsCVnZF9Mf{?i(c1zSx?=p<+2lk
z+d)Ia?K!a4$WXfR-4nk3Ce}i19MES@&bH>8G5hh|llLt-oT~(PVC5cNJFDsUC|DHG
zkBcLE`xXhJgd%KUlg2)VOz{}m=>JqkHp}a8y>R9Nh7}ii+keECL<Pz#yKq^r+)6ML
zzw(srkK6Kf5O1dc+1ieB82$Udh?M5B5^`?&@sY+C8aoJ#^8>;AO@hCa^T-a6_NMmw
z+2A1PpJ&EJ2p5+&@7vjNK)&G58TnCPV(cROAO9QjIA!!X^^|VVXt+yJFuIn5vhQ|$
zyH{@zY#f|zI><$XTm=cfRK<qV=X?hJs2$F4&;+vON2UNssXVm*VXuXvY!JR7S~+@2
zX6^pada0GN0)h{Gb}{*Dtoy8&CEd#=KY8!bVsO$oJL(jsd+xZP@aT1Hmg7j5<)D~8
zNF6p0LsSthJ~009C9)v7>Cqj81M&oSav@$X@X%jg$-Aee+#Tcmsr=Asi;<dxSS!_w
zGD0R^F3hmf6--BM+FiZyMWXa2Ol9U|Ea+~50y3wJf}Ap7RdLM2P#SkgEt5{Roibm$
z4`9lABkEz}jB(#;PZ*!u<`4^gvmV#Q!pdD_y0PCE!*n>FrHM%7$(FZy?;}D6E{Qzk
zDnfB}6RlxlWeAvrO4;|f_*yh3SMAUq`60X^(#TgF2Cw-|4>6T+CC?O41oEb`xbw2a
z-yZ<)N8+uI{wmofGxiv(oF6+-BM>3T3J;V_U1pQk^;DqL;~)!$|2%6s8vahU>PvtG
z53MhKPDrp@i4#vupWb<C|Jsj2(8b@8l)~Xp^jIHaw^9wHwo<ySASZ|h{Ay7P%X<yk
z#<yND9x)=*C$(btw~UFiYvNCSxEI;Zojsu#xXG!hVd_$s=qeA%aLsiLKMuzBNi#ka
z#SK%$IhZ76)HSdE`1x1St|^yz+@Y5d-(8YAf=v0*%V_5{Q3CE%SjgI1Ws%?-ZzcmF
z0qG(+W8T)9mtO&%MopqyW(5wbx%om97Q^aKRG0o}*lbsjbg61)=vrTt#phVE<(so&
zd`F2-SXMs=`8<DFKnY;eXn2fWgD13gV}Im6f02YRNB_=>Nbz1|YGGwWbd>CvGYX>S
z+@-j4(9Ao566WQA_PZ!<S)Z~X(Zn!$RF<R{#+(qI2=%8=`#N7=yDT_17a|83E;f=r
zd16ms(sGNxrMtoo2{FrfeNL>5F>M%|yi&LDZPvp9?bX;8B76Kn(J!fBnMXoQtepp@
ze}KOT#RGq#R3|mRKoig(Y>EH<m!;o>!WeD)ANxAHr*vk3GISEKaZiYi-P7i|d)QB+
zZaUv!x7>zqXXXDDCq=i0fNnlX;hqr`FpQfO9AC%G7g9W3#JTVvy(I6%DX-jLU?b{{
zZET@5EAL(SvGbmvfsF;(Sk=yX-|vV!fNT`&uZb0VzPcvq4yM5E=IOd2HrT&^t6QSQ
zj;wM2?o8o|d+i-gTPbq_Be7R=itN#gH{6nlf(Y%W!+ST}L~Gb>NGir?Ls;MhVK$2d
zj&BZyJxC_z!^Lj<UY193sAO;+mLB_sf8GUiYa??3Gy_S0hBoJcC=Xh<@;Czn?rtu@
z@=V~y^-`N>AGl<4M#gs4)6{I!+8K+v9(p2JN}|5El#+jG^y=jF8}r?i#+vEhu&G+Y
zQi&#XH!(0xJ)4)~IPOKZ{5L7JtgdjMyW=*9I~y$)vIdMY*d0U1!zcXPa=)-M`s=k`
zjFy+xO`Tk=q5BH0Y!cZC+_`070V)s-6HE+kxIS)j(R|e~l*3E57Sn6b^%RN=riUrp
zFThAPx5bEPl<U|x3%U*dDc>zv8tjYM0NI=)tsu3qfT%5TH2ty5gG!Mm9PV`d`UAfQ
zI3+c0LnQYvOiaSQDNXlk2R;BrE#Fa3S)q3Q9KBpLQR|+t3`fYzhn0%IBnEpNo8U4O
zBVu2z;xQIdM0t@pPsCB}z4o5d1{m$K!Fx-6<93L}%yQP3{5=&Y8yx^@`!!?joJF)4
z?a0~cFy5Wf%g_!ZIjDX6u@^(Jg$?q^Y#fTFT41a|S36TTVpAH&dDFtzj7&?c;q`qN
zDDM6;W#gU~wQLfxjOhgvRq>|x#oX7-?bQCB^#lE23?p-__|7dx^vG=xljX;$OM^y`
zxA!e;l3QQ`<ofWT1Yz?Meu;t4b>yW$cJg!#PPa~@*S<l=*}T1o>pD6rDk`tTEN|B<
zt$h#inwvu{wlwOIPJe7m{;rELHFQ7>X}_T`zeZZ@e)N~2{u)E)vcS}EL9{_vm&_d)
z<%0nr5Y;kn(`~xm_7B=7I{1zOP;eWM&hE(TgWb=2S5hX}T!u|}=PN<d%K<0P;%P;R
zotkM{TU#qF8Z%#$ZMpJ0<Ik5R$o#5-0ToOCbepeb_a><rrB%?dM2#G2!)1i^+J9Av
zPJUq|g7&QLur$cU>9J2U7EWW$*+W#B+usS%zsW)JAxi_;lRd&{p;4FzB7M0@BCpi^
z4XtZoe_$g?I<f2S^kd6G=QXu-UJ7H!w}Rz?6N%}cRz;=1xU3iN1X0l<E^(qSnzD9?
z9VF{8y)5wI;%7aRgXuEWM9Wo_i2hZLQG&m;;{O4WPUf)7YxtWrd2x=F_m5mFA;Ws!
zR*!42>VVV7dR^S^M}|$N!*3sr@1CxTRDa38-8H#Kg)F)b(k*vv3ekzWp@aFcWR;Tb
zsSgw!9312IHtIL0AA=NhQO-1u$+P=HO8_TjYfJ2u-3}$jUNUBK9NciVa5onI>x)%@
zL!fXU^BnddzgLW$@tw`1N3ymRhx3d-^A}_Cn8N2&M)b@P^8X!XJ<qNfiH_Fd{k95n
z+LW%zXZw$zB0)@q3k&NIv0jYkieo{1ex~Ujo_F2Tv9q)POLHJDZ-8YeC~o_)6i!qh
zN>tzXi_R~WuTndfYw9mZ)hL`4x71R{S?wXaBK=7js#CK8mXCHgxf$S9SS(V1@fuwi
z`=9RE|J8_$#9nVbJs*j9811tSHZ{06n!vK*7mQ199Rm&E3|KM{9`a!Npc_6wXL63@
ziq9CX99NJ(G6?Uw`N;#AMKJg_@M4I<*at<xMi{T`4ve6sBqs;w2i;K-N*s2Rb)A8U
zyWS5UJ9n1GtK$C$8aoL8Z)ogrZw9;XKA}YtO0lWe3~9r#1y3Ty70-_h^e))2dP&xB
z(c3<&`v1Wg%UplJGF|iz*GMA8KPEP6f0FbgS+H6eiN3!6@|D+2gUcFXB>V9}R(@($
zR_yUHXwIJBlxuSomzKVJ|9-o|q#Z24Ave7e4zg8QWRt1|^ml7+Rtclyc0*!*99SXu
zMYE?S$9sR8Kk#^x%WHktr-4;ZWg{Og&703IGhH<X#!)hJLLJb(Vatwo5Z=7y$UC&~
z83CAW28zT1cT-X(h{gAZQPQi@_m_mm0Y?hXuZKUx#>dNBZk~R)l6rk5Q=(N^Pyh+T
zBE*EWyaZFGE+}7SM4psrl@}1}rJsJ7bNj;!=*Q3fc7$+2!M9*yybplWv$!bDyms82
z;h61s6iQ)x0-(U+G*L$#WlR2{7ro*#R$`jbDD?y2cnWE(ACe_Ze4?x&-Igv)Tse-p
zYmKPmPR$!X)TKZF`j-9ulfO3`>9fa=?`{u*&ZkSV9x*ec6G~k$AO~E*o-JHdlqv|g
z0t%$w`rE4$iRH~>k=mOfp5!N1Ew_It%A4u!gw)FLv(+5rPv<tEi7Ndy3x5<{T6h;Y
zh}Pr0e(uY-u$om{>c^dqH<B#B!3Q{_u9c_&wrrj__|weq>YVuF3w&0!9GS5^NtD$(
zDSxhWv|Hd-5ihw17hB!U>gSR(Gh?a@o8f}mK4qa<ALg|~UPJ{czG^R|6fqYB2Cei$
zm&{@}MaU)N6>O@U-pThyVTja0y$ZITq{`~f_cy6`VcFs_bs_XS)NSlzNgXC5UvKtP
zJP4AM&7kfUi#*5a-rI9<e$Dw&i;Wc7SvedPKL9xB$G<utcYIgfY~E2FLBhNYe+x}r
z*zbMY;jn!|h#(`59EoXW;Np()<U-t#gUC9ST}ZzAPC?CB7|~WQIpiW#8VaHu_lWfg
zU}`fGgWHQ={GoaK<_!dbTv^^&xPJKmlmw`!XH3Rc{5obQDMIgOD43A$DS5+1in2#v
z06p@)Iw>oB@&t^V#;O7?_(s;usjOnCy#LWjR6S7i7LG%PWa@2<6Yv4r!8NeBIJk0m
z_Bi94HADuiMU#WERN$8(!m%siOC@C~L4Igro|H+QVOX_b2Lm1O&wkbv9O{1}w964)
zuFr|4s>2wZw}{q168zkg`FPZuZHWR&e8m)M4%*5saJdw0P4?aquiEj*NsqJUvQIjY
z&cj~Zi}g63!!7B1sxV9;QucPQ{az$MZeL1#yzJ9wn0`z{J7(rN`tki^UDIj0xB{GU
z<N`+UlzD%t#jYtEhm5V8O?>OQ=KC-^Z3ae!B`1iI;%klie7ZUavPT^Y6OdJ}7KRtf
z7}w_0`|E{TwFlYTAWVX`#lTj6P-6M>GU32V(A9`eN*e4Vlvc$s#hv;;1&594;-n4w
zHDP8n=v~bQt}%Rd_YQM!hyAdEC@y-jGB@z6N1J9=a2DbbQfzMAi@HnoV%2pe_&&^8
zNLY6F2XO8MVHZPuXD@JRdwcuu;AO*7%jM5mC#R?VcKHJtuY37&q`OE+oO~f>Yeyg@
z!rVnFX9LP3o~q(Ru1L|#n!ze9V2CU;=TTwXb`Oy5V?)JR7>lTGee|KYW3pvlp8$-j
z*c3F|`v}14ELg1$0DUf>gHPvK0XJcUQZh*odaL>)`uANo(u+8|PaYui{ifTR&@J=2
zMoCxn9+UkVFAcRpi^L_rrP~hKljw|$u^h$esvlsb#duPoliYQObea$1Qws^W-f!l!
zn_zZW?w~@$B0z`iEwxj8{`^_5*(tY~d!zo<?x~*Z1#2-get&`H2H2*AZ+eN>UQX%&
zvzM)k>kj_%14hjmZ-^hB{MIhxj*(*XFX1{(hYo~*21I@J{DCO?!04Ns07!Q{mB1m`
zPG{m)^D*j;u-wR$WaAW1EEk}cbBX3ve*|0O2aNaSkU_==6{5LLihoo9*)>a1crs1%
z?i-h~pq%u4ioJc4i2&nItfUkaN>P#q-D`CgbA|atl$4Qk-d&LLH@D>sa01#vv)$ix
z??oC3-Abr$2h?RPRs3b0$XUZ1a(=(zEenDgX>_AZ!{j$gak@!Z)e7uqszEXbSgK!0
zx)~VFV|Vua`m!4?uu}O>?wW5~+LCQoq&Tg0i5txGWarH}@X^#<&WcgSsxt{bMBBJK
z6hRf-yM6*gbVQx_tP110Uqw-R;KW$b$&xzl%)Pf6%X|O+y^OXtso&Wic8omW&v~vT
zz`sd@?gHt@RR5Q?<#Qjc*ZA=J%fbxZY&$9iPmYXCFeZDs<oDVW2DY?sT?)gX#gZ!>
ze7v}{DoSYHW@*;H*Zw{Nmr?3{(_eU#<HchS$q4`0w8aeW<HkxRLD%OPtCuHRu*;R#
zkgc)&a}@>?>;D(;H}3|#w0k1ebaO=oHZj^fXhJF~;$}bH`rC#gQnIGo^;ax5zC2<Z
zn~O-E^`dkg`~<}TltlOiT+WKU?(q$lRs^lwgBC(rP0dOsqsSANmiKMS*Z0*STbILm
z3qjX>g*gyRr#gtLnwrziJ=i~%*U=&0?6MaEBO0|pEG+rENlg?#-yJ##w|V4+8lzo<
z2xO4?MLVHmkxtoug)kG%MyA?U<cqEcBUa?R+p5^fk0db~KO0OXRwnlmHf87IX+B2z
z3;<YW>+ZyZw|H%rbD@b=LD&Dex0`WrhH)^XR}?TP-(Podr#;ck_eKejm*1e1*1@F`
zD&u)4b`Rfq`hnp&I}v4}g3ll+y4PC9W~)3rrQjz10qhAF=Os;`uQ#M+D{o|Cm=CFO
zA^Q=Z?z69N1qxv{9v-F11LH=@1TfRj*X!(z;x9s|Lu-hiJ{&veElH?=>vRh{s?~p{
zffdQX%#2DWPKeNxM_3%$2+zoVh*PxG#@X`{Kt&DB3fM1sM+lkt!F~Gx+nFqw$nAv7
zQ)ssCNZQ4)YN*YhpdqH$Zmw9`6DO+cf>@J^O2e)Z8XFTSEcF2VHw~Mg&N13}cF{d;
zc~jSu-2GkrzNyh(`A{Ku<QHgNCJkY8t#q3-!U)lFEU~dT3%{uSPrfUwo!&N#pPc)8
zXLf=d6z-PYp)Y7)6pr0XLAFW8uV=k)4h>Lz8g)^HgoMUzZ`>oRdbGk!2=0RW!#gSY
zt@(j7kmZidMHWR(7ug)R+OF6^v@2wonXLrx)MzC;9<(>#L!KoQ%w^wWlxQVzFDVun
zj%n($-#{4=pn=8xixy!TO)&pn<h`Km%sm{^<#dbk_GByBcgrEXq_oNH>)ZP652hTP
zfER&+$po9%lAAja5mDA+j)!U=mqWmOD_pEu&bGbD@kA%h`vGgGVNy@rFGedLdP?rM
z7We+rG9<8AnJN_f<!_2VBdAAw)0W(;%jZLDeCQyiKUah;#>SMs-@pA6jep2+kWA)Y
zYHjiJ7vJ`*{xBEQY6PmwKs?p|qUx)oqI$z_hmxUZ=on%^0YOqp=~h8$kWN9myB)d}
z1nCd~X_b_2L?xs<q(MSD?>qj!d%t!6a4p4QX3jb9(|hj+Li<*D?MFIp@@C+ZbcT3b
zbXKCRJ@&&$!A+3rk0Rg#DyLoZei8LiKamIR*^NrG{_X)aZS|nJD^?_5J%Pn=m~qLI
zuI!%Wu>OtKqN<ZzLmx-48Sc`#YcAr<nXvN-FSjku@7`&sgc44%IxJirKLQnNY-O#L
zh}{vaCbn5(`;k${MTV6_NYEYz7T;m{q&-D-7b*V*y4>eyhh{py!+fGeQ7RxXaWp`Z
zqh8{4Jnqj+sbjM6#<Rv7L)ZBWq~W*jd%|6;pF6?aB9v844|D$*+zpdR4#kjdSSpk(
zlU>Y>j^%W+FWxlcBIgikH>)nH;9Kr(c#~S?b&ub1V@Me1HwkD!dm2+NJ###H`PZ+-
z{TEJ-_Oxo-fmr;)f=SWnO}cZTCI_EMD+CP*5%Q%Oa%!O04e+8#`X&V;QlCog6;nGv
zPN^Ddp?oxm95Vm%10eLC*Gz)awWnyxnm2%j9MU-v0NP31L<v&uF+6Po?oQ>BkRg*^
zj0aD54fkB0sOTTGIzYTXra=+(oUnLq(Nc3nRp>&DI4&VM8X<duYg%n>>%!j59?e(8
z+pH`;Kbi>AdX=1>Qhmy5B)i+;)1j-0Vd*qIKkGW3nNUpb^n6DZThOkUpHeHxwH^aP
zk6R-vQ%Id5()s~%?_qwp^3dQfoxm!}C<u;EY`>J%25lW3v)RGnZNH1IcY*`QzCwnX
zerGEwkf5X_a<Qw!Uje>fsUXK%2CaRKqt@AinD06i0Q`PR>QAH{O7XT76>aR5@7w$@
z!X%`{<okG;LrrVe*hR0LI*oAp0@MATyjOo;DBtEX797#3S@ddh4wDJ>#lU;<ku{6?
zkm-UFM9QZnSGdai&Vv?_f`Pwcar;`ayV1FOUr}q`*w)q5`@_+_*T1%ePy4LaPg<{>
zXud<^>`!T`AMq)Fx<j-PNia>!61d{wmLW<`RZFk1U%pIzb_sz5ai~T%xUe)3I}Y5=
z{$UZ`vD7+!UsBe-9=?btI`M;8>D6te!S@gd$G0+7oJz*ah3b{!6ytzN{rSgkAQe7u
zv@a^Z+Zi)S058gAb2kR<KEOM{#C%XxQPI`r&*^;p9eKEYpV;yaj&NoR)mc~ly5Nqf
z(tCdNwOZB&uX!U3Z|?uxo)61fb}UG=M@0nXej>+6Gr`6VZQXHe-*J-SNXc(jh+o)a
z-!{G$Fz}#zYHu8$dPtbsP$2U`8$gQPo{}$8GG*cRI46`OQ8(`0MVuXQmKq)0d(VLr
za;*>fB#MCmu5T)KGkn>&tUEV8AWSn?D!!g9k)IurDPyMg+8Dz%lcKn367liTABK35
z3YD;qB{}<d@L(i{)5F==KqC#yo=9`|^`U7XQUGZkE_kvJ3Z>klBIwh?&wbgT3(%RK
zLQ42Q)qW8-UhNOVk>AIR`-Umk4>^-;gb|eFSt1b_uU;?<G?5_a8>ID;;sHxfdWBuy
zOo3-4$w12Cn06#719x}dWPL^v_J;MlNH!|r76A8q_?;j#wC|tKx=`h5Jd`fiWUjJh
z{QaV$EzVIgsOl5>V-=WBTzYWh=N{D>SFWT&W?s`i*V5&A-bhV+JX+S*MY8VeV)h)z
zR~N?X!ltUr_BwA1?R@7w2^zlB;<#GU|9*@ULYBuwB8&|w;x6w<czBMi5l?S_C@4-^
zwLw(U<WNY*9ZrXzj6jU@-d^5oqZQs^AY|-%fpb&~QoH7pIbW~EP%2)>Ik8fMdeXmZ
z0{}N{<-E5beNVTtRX+uulzGQn*#8CD|IW|KavHY@Wn`Fh@rDJ?P75enM-JzNo$biP
z7=&!{@=e{w1WS+apjw@LNWLA!v+(6p0R%XVAv>fq#z<-Be#8$FS?J4}sL$RL`?KK@
z%1;$nV|-x<2Ap?!xfS|0yZ1WX*P~uHn)lAv!_@*)No-YlU*o!kf7&B|%5{VDVOSm2
z*>nH;t{%7JijADxc2MtAdStS|%m>0JD)Lk2wah9duQa<I64AcG-1!bE|5s4aVQclS
z6qftOZ4eGXNq%9D+I*7ERhP<rEn9x+NFx7IHmizkna{)QlA`cKwE(EJ82u075?6;B
z-{0NGHTFNE$nrma4*CmA1t~@{_VznREx{3BB*pox^wp0St@a>AC{Xvm;$hBnDc|G$
zBov7EPLnvLKR7P)^I+3B^}3AtweME`qBJk7AU!<uB7eK{X&tBQ&`UYlEsRLHsPoOo
zo$lxJgxlpT#S}24jOWJKi|5yPa<yAQrk;|wAiP>ykTyL);a0MA0Y;TGs7k09L3<#j
zvdqx(i>(w4zIw4b&l(fHy3t^E$P9U(Ke+wkI9#Ue)_}8Wm@OR9=RrDjwk_Lx<{<7f
z>q#*6jFwQ#iODOjW7qcXyNucEhNU+vx+jYs02|nwF^46)yB^*yOXYm2E8^p>mghO|
zgbFLh=EM7Li({gure^+?s#-7h{M-l93JQ62z(ChCa}DGgthV%T880!MoyB>b&zk_L
zMq5=!C(i$CO0Hr>mD27Y<BFwKrQ?aZ?mc&E!2<?tn=cDzVGe(L3~t>%rYD%#N~m4d
z!45+)e>7A~T4fAZ&Xv=AjN2_q1js`mstCu^t;PM3UOn_RDG@U5w{m>7EaoJ#9kDe*
z^sRUqaF-T~%0q^C<Qf}I2ZQsQ8MJT;>1f%*QM}iW^veIV<9H!F<lJ{c%;){WzhxM>
zwOk3VoBsgHNEct<TR0LwP%{-iTUN-Jts`l=;U$WEJ_|=U1wH>T|M9NC!bQrs6l)4|
z*)vj)%QdE%rzFQWR&*mAcLJ_2VdsY=J(3a(kmPT`vvhWL=G4?s_Vx8CJ$~E{`g>?4
ztqjRPBiNJ7xyt2sHS!dO?te|6q^mlHVbPCA^1QShL-kv{)(`Z+F<}bEW{b1$Zo4Ys
zwfk*=aHYA{$9|kJ#6`rn=$UJB!n$pE&VzQVFF!N6c8kfe7Jj~Vh9bFIU5R-9gvhX{
z1+Y?C9XrbE%HWF<QU_zze|7rPi!4&{QPw>`kgoo0dz)g^bu4rpvq!w6=FUL^Vk@EF
zl3FkMk}ZbzJ6BRTy=dP2RiKoYkLXiw!?+`YZ0V7tdxT%DIR0vtG`JR|9u5HBZO|)5
z?ygmNMkZ4!9gCmOx_eE3q*u(P;gcAmdVVZM?ygGH<VY>_*w~%>3H=FR!ao_<dL{nw
z$ZIQkS@%5qZ=I{e#oDckD+1tjZ)=z7g&H;an1Ydp*Fhg!mD`3oLe#CO>iL(6Q5tTu
z;~9WVqA1Nz(EjbD3#VitiSHRbtvZZ*S5#8cI(Z+rvS}SEW2QbRaTS>E)v@%N*J;du
zUGSU|=;XlPj77@30a=09N(Bqz2*ANn^{#=4d09KWi&z>o>10)w@)?htxQh2wbNv8Y
z-1(Jwbea#iQ^c!}ftB2+dh{IY`3+R4l%I{tbcgO%a$y{G+;8dUz53O(g`b7Yb)FM7
z3}m8uf8I?{nhs#hnO;Oq{|ORRCIHEM_w^1Q`L0NC#*JO;+i_3G8e;un0i6tUnow+k
zKoZ3|JdqmTQc5;<!Py&@j0n?TQ<N@yYd%sZygjED`9|c{3!1=wFoZ134#5A@QjtM<
zPd8|9&uM$|Gp<JA(=I@#UMTD56}WzTQFQtj%hc4l=wky^)(}TVhW5CzcLo2;ldZ^K
zci7ZikFkA7Hc9fE39X3j6d#``%iS&r0hvaq0uy`Ejs%TNZ?(4p_lOobiz;1ME6nhX
z<gy4c5Qy6GA!;4txxQs7hfONmL3?ug9P{=dYi9=l8SEWzL~PICC1(0B;~4*#{H#Pb
zJ2;*2*NaJmW5|1xT=*B3TCEF}N8>S#=m2$O+^;a%a+~S3_Oy^_RqR_*^t6OOJES?@
zF!Pl>lv&Dm3b=}&IP7A`m~y8iV_=B-BJs#RZRd4Pr4)JQJ=v~F4!Hu<^5HcB6yJ7=
zuKvqJyB~lH8{J)=br3QNymHsn)O--IeLU^h<Gb4wuE^+5{<vV6E(Q5srSp!PoSPd*
z9D^Qi?N;DU%(spXCY@sB8trjM>PJpzpOY6!R9V=qHC&j;_ulx&cHJaA;t}-ojqxjv
z5|J}ywkE|T7=x;phI+FfchWEovztNjrb_->FhG3GrOmQB@&L74?ZiNgfoy~B4h~A4
z4dX53wW_0zcTkCrVllr>o2qWE7)`k=8pdjEDBRcjVIX&bNHJf{o0kcS+<KKw9pO8d
zMf|3jL9(SP_s3_$VP6@l31yGVt;hl|hgM_wxCCkPT1nM$4~M4CWR=ULd=jPpkdOwo
zO52^r6v1^I92}&O5EYmb%$<6KnDM{M8MWD8Fwd4Pzz0yKFTtONf9alIl~9XcV@wQ-
z%5vCRuQobP>aa_`iQb5w!+<1Zi$qQw8&VhEK|N6Hn)KVgzBYGbyib(ax555R1QpVx
z51Vc#_~h`FQT<(M?(?sOq#)bH!zT`xxS%4GmtTKiN%|!+Qn)Dq%Znr&BVHl~QpT(8
z(%9wj!07mXW=gi=T0}%QRb6);KC6b*y>mU`AKIu7tPjrW6sj44oK+?`rYqw~v@nJW
z+rNG(N!dfoy<r;ANGO0g&sKU)s%02_Cd-{-`i&^*`aZ3(H*wz(M(*~Pv*_rE`_Q2U
zQz0LY6mnYcwXF?OjHM0TnXbLC0F>)z27b7Q>*5R#J+2}wPHq!Y3lV!f5x055!`w7K
z*gcTH6c5`k5=+O;T1%L7WCwtBH$J=sPA%cjfpjjl<q9W!znn%i@RV-H*7{w{2r-7H
z3oW<*G<14y54BGT!60A=5o<m9aB5JNlMG%HZmt^}^h+*OnvB9@vWJHvFvWr3^Sx`Y
zEJeEucI9&wvQ6AoZPU+!_|;U&sR8=@4(}$9|3+4H1FJ^ikyTm`G5`Mk8%^d<`GK7!
z>i)5oTV5$0N+X76N=z1YQl>Eb?y7QRREoY--X&gnYzT@J3i?|IC#4}s=^!p>Saf)*
zH;?4i(pmUDL|^IQ7|+ne4|#wP1<5lsE>^mt-m`7Tsok;~7~Ho5MU$|uw)pu5BLdpD
zL?3w)m2Ri3XtEd7IXW1J6_@`6-w3o+2_dH!^EUyL*^h3MdWkVO8b&2(dyY&FHu`<P
z3^yram$u)K?@i`-fBe%dBL?P@AJWe3Y;)lQ+$@Z8zLGS57`rVo@z%@zm+aB_!A3Mi
z!ilf)W_<?(E`?DwPcR_8T913Rnp?83hRZAGsr1KaS$=ciBuU%VNE<rFSLY66$37*J
zr(=A2)-=D?czE)%s?M~!4$hEklY!r!HvafU+;aK3LKh?a4WPmJ=#l@%zr#$bF=W~b
z<+z5z>yO`c<f`H_JFZm#JSk%3Q25)%P~Zs68dI}UQwz)5dfmX|kXpH*Acm|~JW_7>
zeZ=C;#ijq{mcYb8yP{L-cV!pDjXKOVhhA-C^KQ?9iuPflK>ICSl}ia0(EI;NE8%gd
zO2p7;C8W*W)XO|YQT@2^G3b3m7-txz{iYoWBxNETH-(P4?Jmx57QQLTDk)I8J&n2r
zDf)se;^D-F+$Q-ZWBnLkW@%Yh!QK`0ozvPsC)_26M#Y7I;{Wwc)-B>e9BHk>`lZX+
z8m2_0*817X^{pJjF)0cJbR$r)dEZZ#;O)YEwSj5k(DFeLD7BMalnT_@e-mxl;fIoo
zW|HQba(ws~G;(^D+W?RxGG6kU8fY{M`=38h*U;Dp{d?v+cW_Q-J+r}4ix^kiSBwbc
zJyZ%U*KtfL*9$I^!V%-kt)wBW)Z82jxSZ%7_*=L%VEH-U<OglWbOTY8l)!EN=N;dS
zEto`!|KRl0VXV9Fa(zE=W*{yMm_vPf<?S?0@AS|h0hc}3mMZ+qJsi>5*P-5lmJgjm
zmW4AW41(HUQK=cK<EG5UE;tdLTa1_;4#F&6$TQ!6LcL?my+6gZ0A>kT=u~?y;9`f>
z|E9rXCu%_QW=1~+*;xV#i!p>Qan@S#DNAP$J-U;^_e?Ufjyn^`wpLUfB}*=jTT#i<
zkfTlmuBh28mM3cGbAK-&-^VgU{`^&#dsXMGgb;=tFbUvm+x7euYDQJny?)+|wR;nE
zNim%}Yy<oCmF)Gcqaf}(lr`UEw$Jt;Z*RO=N)_ln`$6`$Bt4hX*R&UA63wu~Ujo1N
zs$;XWT;*0{!znc2{6s`J6Y(piWv!Z`)v@{PR(8PoT?OEna2fq7_m{<e4y6JDnrpdb
zBgQA}j&vK2{W~oP<t`4b0SFIS_FH`sH>axU$-Lwx27Px*tF8fIvUYTxkvL6(d4=_W
zk%h!16Y04SSCH26n+Sf<%D=IJ$eR?qLwzUx^z51yF~rk~$<V|KNfZiWmR9nQM;+vs
zV4W+M?)6%1gbh_(+vkffFKeFIJl?Ea>2IpL*m=1w^5H(Db;YU8<_pVvKdxg(r{yx8
zI2Jw;C>Ab(Rc;%|@qzBg@Gl-lhwsh^1MYE(<`9^N&BrH|KEKwK=tQx^aVr^_pF&^K
z1H#^u(D^$q$)u()i=j`RSi+z1W>9jH+Xy5{r|j;$!Np<zV90+<Oa#~W`kc#W<~|L^
zRLv}wHAi`nm{e%Q%~On*-#@4`v8zQtH56@o>CSwy<^LVObr@6pfu6hhJ}v}58S}X_
zU2vUA41ES>e0onI<8OkGkVCz@*@=n4nGZdUj`s_mq~z&T7|k^hB8BW-+lWhN&neK8
zZ;C+FDSOshh2CZ2P2#teh>ba-3yD*h))~&8aAt~soB=(gaB^(5ixI+(iIt-g(>>SX
zkE>r+RfRlqeA$k<`04Ak*z}!E6AB<Y<8kAME%obHE10rCq~tUq2oF<tx&0~pg(o~s
zDBnNzxBxx?B#!%2coe0jF`Ca-=}Mn}?gq#9ci6sI{OCnMSXnu#2BR+qKhAW38Va_%
z?9HHmTpnU*9H~Y!KOh>n(5Lm-t=|0>;^=7NW=ImAl@(XSZ3rV}4EOOCi7T58UlM0a
z;1Bt+na0XNXt+xq&XQfxS}qwFTNgkRv7QtHG34DYYU}Q>a`w}Z@QK^;yK)?RtnSN}
zr=iUu7|=*oU2x0o+hcj%M&jyV?mDp-7LMGC$KCuf=^0Mom0^v71k@0ZKQoczM!rNn
z@r>b<b_%<5P5Z-7cucn^s>W~3F!G<bF0BC~=HYmJE?#^a0G+frXsF3VoK9@2=-&KQ
zgeog_Ce+swn2Dlvb!~0b8XXy<P8fr#im0DP<3z!TWb!&-(}njN<|HVC;i9aQAWm)s
zzv(ruB049tS}3ZOVNPadS0_?2fiw_0#`6VIqZXVz>`f5bY&8MZBwquBv1LQ?Koc+I
zffzH3K&9Sns;~e&(5sHcEqBhAHQ@8-pe8-YIR^)%>xUVi+2qs-eBPD%Jf5}<P!Qfz
z*!551Ume?ZKFbW7`%=C9!HIzcpWZvd^+5F&j!ZgxsjLg7KUb-M!rq5h>yvyG&h)6{
z0o9i>D`&LW<J+j=V%d`{)Aixm3l?n08i-x8!XOIT^>dg^xYk@mVeX2_jjvrVfDJVx
zmeLV-#3}Jgj=QH*=dIL-!z*X|G8G`>1<#YVu;`fU+xoF+Z<|)cE5}BGsU&Tn^kv~-
zBTqIbnpm3Zthie|NXAVA$Hdy&`t!{&m_$0Rs-wP4pxHgnHjJ#?Uc{iIhISc7HCr<n
zNo9tiseH?Wj=#ar{N5%+mCSFFlM+bPouU4KIoQob%gtYPLNLNw^Ddn^Vyr7YN<E+W
zoAAv#Hyvi&?maV6xu`E}^O5gQ=vw(iI|}AvFLd<XYe4+Oi&L>g1O$yN?^T-Zxh6V*
zx|sdy*^_;J$QDB_?c3&;*uT;`e8h+ZYGn$u$jd;`<9f@t*<b0r!3Nz*iG-gF06p#m
z&P`_He$jS!@1uwj&`heb=p#jUQhfYnZwdg_5ztfK%rz=UD`$z(0PvB1twR^Ux&VD<
z67)|fC@8-3Cyto{pyNsZO~^Vo7Oz`3(CC24Gy?pS-BdkcdcYZiCCdx@d9AVueV-M`
zIj!j1*r47wj0;_uP8Y{J>tkhf$Gh|KEBz@Q0EdE>%ZmHg@=SS|xVdSQA>w)~nvzTA
z<j?F-v%l}<*+5{}sExdX1NR%{_wIH)q9K;Ha~(y#3<@oaXq+I*D{<7K^?+v%d(Fjk
zsh<bTB-()FZhyr4)yZab3Y8}xy?4G<VgBobit6<$`B-lvOO@ligrngH`Sy;Ep(=Al
zzrR0(Ds}HUeVsl1{pJ~rf+AGHzkF_F`2Q+Uw;ydq8mswPld(#ki^(vX|NB(6<0=*1
ziy!j|0=wLVO}0Q0Y;$LPiev0gaL?Ttc=>9SO%j3|Oq|nd%E+$UbSkb`SAPFCIX~P;
z@fYLZz#knQooi5zR=Q1!9tbPOGspwL06>`i0~yf?+DTcjkNO@o9*yY_+Em9n0#2ZP
zYT5&;0axAT9nRE}C_vfh0yHf}pi$j!S!=oycun$_V2y8alq&bhV>9|bN9wq&Fy*^5
zj~=alPB-PCIBx$QW@)#X^%hK=ihgaV8W{yl_QW%YN1dK}IU3%Q>afX|W3FG-2g+!m
z?+mc6=-U7sIa=8~RnVRmU{PJVN>lCuqM|~6)qYuB-9*V*-w#*OozsAo`P?xDv+YJ)
zWPuijj9>T9_NLR7&qA=V3P}ZVKgn2oJ6W?-B2#w3T;D=XAdM9D*D2v~=zLnY?G>Lx
z%aw=N$&Bmi*;+Od<X6rJkKMUgP+rq|&imKaa$`Ys>zt4PG~@YW;E{?-*I1c;)r%jr
zBcr1N9(~vA`SF!6f3f;q9gXXLu^ysm(JoMzE>QOtxWCT;h_5ZBH_*O!WJ6Z#Qvb|z
z$@y!Q>8uWrC^nyY-KA<{t<_RCxw)A`Xn&>|S!2P@&+PJ)8X)p4N5)yi^iFy7&^_3*
z)pXrT6AV%^vc-L`gv3M)pd0@1PaW2ixA*qPEFqA@#Kc(oZxvlO)tIf8C9*)X64B!=
zK=zVu`X^&&*VPv9>YO4R!O#W#9YB12ds69rRv6Jsi^6!*DK2UBUAe$gl?{3UiKy<5
zN`NjN2bA6wU=ox*jaCAvhvdh1Brdk^Hy#c@>Fw(a2HwBIsA&+Gg`;6!7tr1YgaOqV
z*8y>_je?XM?UMaTtL(+gv%jeiUXp$`_}yJ9*+A(|^_O?03ea<Kz_bLqj|tTHoiEf=
z2eR6V3bTXGX)iuM+InalfJx6k3{WQA$Y(dIDUPyEA$y9XKK^ntW@dRN#42)fawb43
z`(H2G<+l=mO}T9h<8bPf%Yh@L+r#ps)0yGk{P~^tE;y__UT>dbkS?*@3)8bZ-CzOS
ziU>R&Zpy<q4K0<)>$!9+Gi3q}U$I=0D+?cLB?<GHy0L*X(Qg|Vec$S*+|BjXKDZ*#
zx}J7yi3K37B>*!y9xKp$?>OIl3J^B$4AN)B0GT6uyg9!2FsnCDOCKD$G1)D(&w)H)
z5|@f7ud^Jcdgq@USPzC40q$AL|2YL1?at-v8ZAl5<n5hQ_}JgI1lr%J0El48n9Xa7
z>9IYrSk@1dR0l0l#<wl8oz3P&AemJZE~P9zJ$9_rNitt5;Zfgb0}3Ou9AtkCDb1@G
z8sxZue5$6-j*KpfaR=i;`VSRaDdRj<_M$}omsMq<XmS-<_tpMgYqxgs9eL=>W_p~j
zs#VTdORP^7vXSpV4|d*p(k=l9>bskke*il_+WIcq?H+0R@YVRYONZHp#U94M#_eUv
z%RMZh#oy>ovxGawU=aMR(LeQIf<W^UzJxWy%{F#<z_%I4(0!U>l(Y0K2OKA81l2K~
ziFp;CT3@?A=x&lowMKIW@id<@zxbXS{Ml;2?00V>BnUt+7l0(6Rn1L&M0C$`_opA(
z(#1n&!?Mq!mng`Qzo(9q5b}LdD-pD!gzuICbW^r`(e8||hzvDr)Nf!JEGU}l9#g`<
zz6FP~Trvcq@z!@WK9~Tox)<=L=x$-rsEyn19DTZwV|S%#2l@i??P`4P8D2=3@!v9O
z72d>9h?sCjgvaU!1|2QQ$9Ov+F~79W)yhNkUs9Z_P@eu3;MtxoslIcovy&hY1dtU$
z-Z!i&dDlMgyKHP9SNSYkDbb^|-oqDwbm{U2@HZy*{z~TVOwli=h_S7+*&Rq84|7<P
zg?`0c_Yg*Ed;@5o4?zd^P)cbs>(*!-KwiLYb4KvrPye%n2Hacl)0`jgY?VEMGkz*C
zVS60!*n`5&4il9_^Cf??0%B4b?pWUO9+FUh4-#T-4p?4Xts;hiE>Y`M%u^C3C_TYt
z!tppGRaoIv9D>2afcYGFj;QEpo(nIKxr1gQ=nlddfvSo2?gfojXb<mE6%{yXa=*9V
z`jnooRyd7~nk|()DLN&i`MDD&$GSaQeNMVq8+4b)nF;x*Lsa3jmp#tjNU_HH?8U*f
zXOquSI{-zY?=04g1JzsmOTZVhR`y%KYY>-<=(U%urcc%vL<W50TF;dC^zH0LL)aDS
z3T-i3jl7}$KAt8ZJ`oCVqe2P3CjYzx9F7(T1qUZ*Vp`hN@8y-1m4yCK@tV*t^d(&p
z1-A2w2JyMQ2=Ui|%h9Xi^pIY+<-m1sfp0PSPC^OF_aeby`dt)@DnS#Daw0tOSE%88
z%yx<Q2l_@O-_cT-UC~tLE8mg_uTQ?Dw3)R29gvLYGw=4=;+O06c$?K)Pv6gH_U9Ee
zRt0FV!Cc5ZjKxsHTAVO(mj~zZ@Y6mu7sHxi4!8d8HqmZ@(XAhYcy%~2Jv+h5tZ_>u
zduAA?06=K6KA5ZY^eNg&LLk7xW!>B!G+!O-I~s;XjC|0T1V|++GAe-fk97(R#zT&G
zdyAdA|7tO<VDFh&%MRs84aoKnF4HC1v9AJ-gN{TA2yAN<Rw&AS5q7x2G`a^_US0;^
zCA>?}km&}R7`fUde1X?~==2FcR5*>96kR`*cm$G!Pev5Si<1<`)L_AwwLg5sw=+2B
zb89*`nYjEa=UNhNkUec79mE|04Cp}EFvJl<Lj#atLl`N8AUQ+Uyk2`Ix9{9(+i=5#
zj8zzuKvF>)N{B~Zu64&{X^jXZc&s!m4%tthQKBNq0kZCs=N&?}IP-~o5+wWMx~3(i
z{(UhLrQ0C^G37iJ3Q8H3I_X&7YGpZJ|7WGHYd=jv9)b=ia?BV5e^svey#Mu;B7*9h
zDf`F$gHRIrLyn!Q=$I+gtL@l%o-HY3Vn)i^*odwE&dISYk<MI*ItRA5$p}f68^rvO
zHxk@u{t$3k?Tju|0>H+Sp>o5A>qkD=w{9_mZs>Haqax5dsIs3HL_dkvxV(JbnpN?b
z#tS66NmOMoqF;Hw<p~pife4#*#ia<8_@=4HaZ1w~_wCjC7+)w5>HWy=%q*dXKpRy#
z<PSj3fxJ{+P7eD$5U=2;xkbVdgXROpiywnM8RDD3fH5uvtXKdtzU{ptr7Q(|mNB+W
z?@b-Tqx5D)Ln-O8vY#cM#pC^LTV^iLaxq-`ukQh#Q>ELCjJXhn{+rn*1{8qy4!y*n
z*Sjw(BO;(l$;r9qJ#pycB`qBcdIs20Ya*AQuJoN8*glATqE-*~OkzVBOqRP|sMUA~
zn03d<wj{m+I6nJUqr7kri`65^#N6}8_hrmc;<t!l&(K~O$7H<R(E0Mz`HSs1@w<2L
zyZ}ZpSCeCZqcGcO43Tz9S@<iiU`sNHyNWDqhCdc<cSH+f4t5iJTp85@)F<t(s<-fb
zpHc5Kgf|mQR{^l20om>%P7{u_9r{jD27w$UxXTO!f~GK}ih1*g5z>T!am6KI`GC3{
zPW`_*(x8<!Si=J#fF`B)I=TxOS;SlcNK-Q29WmVPWwg+h+|RY{2QeusRkVol`p+18
zV(G?DZ+`vym6)9!KQ*O?3JZ&}d6e{EzJ>tT*?C?R4J*YCC-BAML;iLFDk8srt`rSO
zh9*7}4{LQY-K}DtTl+ML)WZAf#HTaDblV8{&dxC`*q#ex>pTGg+4Xk`7xOVj!|18#
zLf(G_WmP~oguX<UR1t3lUPGl{ZlvxF|4o2;wPU^%Ke9xpQNFh-sr6cmIBQrm<-{Su
z+T4)H-9=&I7K<AK(qe=LrI_L~T2zGU@lb5hToairLj&VfryMxuvR0gN()2vrn1j!~
zVY8yO=)9fdF(Q@UNDSOaG>c9cp^Xk)?Z14@2H!tI7d(2sFF%{j9z4r|;<ux8j6;;A
zpJ818)Xr0!j|6o|9jk02Tx~hjU$M%-;=lK}PXj2JE*(8eG!H16aF5VJ0mCv62ezfp
z!rUY1CE@<zL|{e$MU{P_oqA6Va-u#>wfI3{(|Jt6u?85ef-fX8e$vN-CyK<rz3w6#
z#hN4-Nyf~|P)kQQ7JygV6|>L79mdR?0Laly8NYD6gDOwT_W^I;;IzrI%VQxK&@5GK
zNy<>g+lUVm$C$q#n@TFdWYRuz@^r7RWFwhvd2cDxb^vK@>YvNOava-vp{~A66sJb6
z@zYrE3?>5Ob$~WOJu?vFSV@-*XXd4T!w{h7R$&L-w-W0yU9QPCmC%V$dnSqDZq2@>
zBo}ngM_hBLeNNx1yhz%qv@-oQ9fxw>76ufkz?HW;++SX+-bFUJvfheg^t|F>LP5g{
z%5TvHTtnM3`FR(+x(=6O4U&t$5N=|do;V34<G0`_s}y5YG23Ebfp-WftocA*A`}od
zg8Vh)YQl*2Ra})vgQ6qb^WnK9l0gZl!eh=6t%mIV(ka)o=m(;>g5kE4+77}8PSUPa
zYHykpPzZQEKjGp7BE5ajDQ>m`np%#KQMxZu`l8hMm@mn6G4*saI^o+!Zl*zYnI?(h
zV<2EQZgEc#=@L`vsG*Z2?9|^$2&C9z816tp-IR9KZ&G}>BdI4V+yL3E!;d;Gp+9@5
zR<)tisgf2V@v2y!?4?A?PzWTmQnB>iw@o-h;r|@9h-c*U8$KM6s|)~k{}h`I<^!RT
zBw~*)LZNWH+WyF+k{g`4jZTo|1<BBZbvJN$)*1%1#T{>Ll}Q8o6(^?ZCy`;%DdkhZ
zgEH5ShWC&82TeG<N)&lPjyCHTQ-*%n1)PD>BS5hWAp}A=vy4YR9KWpc*|-Y>G3mQX
zf{`aEBPC-%a}KbvdP)5Ak8W6^E2EV4y&sr&<OqRUb~o32hn@k7Z-7fKNPD7?CDLg$
z<*jrk(>6^j=+T6jqYU5s7%=q8slUGlRPg|74I@DgmlKoWjsa#ScGSPy3Y6dXw+3!d
z<-Q~_fr}z)<-)K<VVwZPf3*jzMkGT6wrNXmaS1PyU)!%X5Ln(7{=i?|>I+=qO<O)t
z#xUi-2|+;_&_OSdh)|Kld_4pghG5yQFOx&q-O(D1g&pT26}tl5;9dl&Y3f*JWL@J}
zupG=mM?iVKY+Pi)R%A21yow*%#6U8e)Q*^64KN~^SSiVjo?{j(PoytKdI=w*VV~&C
zD5i^5E*9oNkQ#Kb69F?NkqW|>YB3QUH$9s=NgVVgKaWMAcPG84smY!R`&O0?mvpD~
z(a`W%vJ~r)<)Atyz}I+(2XJ1%l#BB^BRAgs&v#kdVV8*d`&HX)q%k}<Hm?tYf@V%{
zv^FGXO9O$vcFreEI+1^mN#WeJv&6;fAm4HKo<RU}O}R@X4iP9dpMja){CJ&yVEczv
zlLFBVBAhe@`}P}66wKr#^cA~d6X_6(V*U-aUKv|iho&~Mdt>D@$X^6oRM}(JS9E%r
zd(j)5xtN2v`(oh~51U^B<>Mqps$b1W8L%|LYj5jAV29E3;Hm*h0<Fr*!uv4ihx!4^
z*jy%?F!DQb$pU=jAAuvtiFX#+IrY8@AH3tynxOvy?V9U}HMDR9xv!{mn=Z`Ina+g;
zSoNm;Q89ZT1j0I2jber=In&yp{W`IPzOIXl$2d78KsjN*zkwuiBv9;}n649@XVU^h
zS{3Is{VpE77%xt!^1f2XB4W2PSyMWl`UkLH5zdqRl2-lJkz_jfoneL`2$LoK!i&cJ
zZo1xY-(*ZonA}8XDI^O4ehd@~=zr(s<?VgHKhV14QENXaSh>@Vy!`Y>79F0@>4&g>
z6THRm%*xA7TiI&^hvQl)eF=!7VTOu7oKm%uvG`O8J_?LP-!RR|>g5?>xQS+_3fNMj
zt8Y+KEq#(a@e)H9+-7EGl~w~l4<%*!&)3~BncP+|?CE22-qH=UqqiBz?)(ihkz!)t
z;$bdC(@*WBN8{`xvIVi4&QHM4PHDjSotff#D!!@{6(L8D-ps)<A58pmiN1hi^B*q_
z2jX`$s-tuHH<nIl3Df`+jo$b3^7Dt#q_~Y|f$g9V5Uyk7jq|v`E5#u=tUjm3A`mZ9
zGpa9M47V5>P>i>)$o-XQnz$@{@stmn^^Q6Eq>-I1lfc=cjYzcI*uiJ6IQszB|Am{q
zd?eZPrmnB+%;2+qj{nR?gMt((dLHOY-yKUA9?swcWqXz-2NLb_Q+MKUEaLF|eg8Lo
zBb(3lzb#gpZGbj!))>|+Rv_sGE*EE1-m1#?#HLP}6?Ob48jq(%DwNR_BzaLmr|iHw
zBG-hK-nUB{sHrVUg(&$#FeQv2+5kdu_Zsq`*;q}@d9=cJtE`q*hbpvJRVOl;?m9B5
z^zOR2@EdeW<S1;fZ72B!BaEykCoZ}pXxa*SJ1eb{%(PB5F`C=I0q{svz*zG^V!QG4
z5}^io9>p=oN;ULP)*@&Pf|*|V+1u>w`HwV|JVu21LldXk1_m=9$896GKVMY3EVEJj
z-jIP}n;9lHkPk@J+ShA0m7D3FOd7Z*;0E&Av|)k&m9C=L%2kQ81YSe=j?X$NY8G>q
zl&{vFbOwK+J(PgYajwxe;Hiy`&F5wvWo1HQgz(^Rb1(#q4fGJ;NXF;Dh%?sn^`eL=
z2u(d$|EpSr@(+oI1rAnD)iQv~%8iTe33~Vfc{>hYg4uvOpJuZXhRzSvsz_+)-=Iya
z_7_pTOk)=#Q!6GfKR;=rkxeA8(}EP*C;+!IBFbjsFS=0%*1<8c$clBOi~M|Li4NjY
zulh*DxuVDnP^TPOYifpWl)fGn!6kF3N{K64Vt4jVU(SeAFqg;3qPqN;Ts46z3T|&!
zu`l9{Y`|p#A69uMDk@4<Uq3o8j~NW75f42GAqikel9R&}L;sMHhK45fp(i&O8`=X%
zh)Lh;JD`;o%%pQ#-sy1Pe52t6%^m{@zyPrOfPRoqO}8EKt|G!kC4#jtBG!eKS=qm|
zXA?Ny<3ln(nh=MSJn5Xpg2B8oPQe6Z5s;7^{BDMe=Kz(AkWm7`|Bldhth7y*L>v{2
zg)5Ukw2<Zdrwy<<yE$B<(QNC{?tuhR3sE1UZII?51H!>)@bD$Sl>tGFtW~OGN6Rv(
zGO!2pn<tZqRwGE-5~+f~)N@qJ7h?1}BO}IrYphy@iD$~ihw5>>rFlCRK;=>d46w@h
zJ?gNT7o~Q6Dn*|i6zn+{wxDb3n8Ehj=Vx-ZRCtWrza#EF22U%<>kKZHXqPPZ<I3ZP
z_`^YLb2SmUr*#M2(MjO!Z0ZSuRn@t@_T0@9a7zCHEf|$pGPk?qV;z2sK3aAH3fAqR
z#=6NV!ilN3=YZ>@2ntyOTJ~<sJ*u`WXg)S}i&_fIc%f3S@@fHmBiHM%Q=Sh)%q!;Z
zfOmG$9Ag%KI^kNV0?z1b)I4UloePT`=N}we4hT-Wjese7JTq<2%ij}Y2JAK)5Pv^%
zXS8LB+b&k6!l!B(SsM?SE%1GmZZ}7-c`2>%qYq*6Jg$9S<@CrYzRwr3AVW(L_L*kF
zjSdI-`0rUDPhR>=t8?M<R4OwSRr=MsI{p;5wgaPB{dAg!u!}V&lY1M9+dPO4(Fye`
z-qWgyU+KsgeJD3gd`4v>Y6Z%VcZ{&&M?(giuRw1*1^xTU=bv1PmRUeX$8BB(Ps9Yp
zvyOv|TSlFJD?}5*-vOE{J7*Xk8eAX5jxjA<&Ywv~2wizp^YHTFJ@*`t-V<Pjz?b^*
zIpdH_MyFgR27*GG!5~Z3(NGUh#WuCFBmO8#4c1@jHY{dCqaJsnE*SGW7*Jk#1Ahmz
zC1aa>$6C#+k)_6xdOIxGL~KD9F@0v>(RgLy%I&!oam+`0Fz=#s+=<nDtD^VxPcW8z
zW94oD;<X3a$tGO-!cKC~>9r8=U7A;a>dB;sh5$j0nHt@tdK21&=^!-1!mA8|oIB`@
z^44YlcTF|L^j$qp4&(TMziw)s{}M%x6v{WPu7LP@R)qG9RZa=*SEn!)?2w}jy3IXh
z;MA=kO1_KY%vFH}5Tr(t;*yev&X?@@tS5eowfjCx=-(!-9AE(lmz~|E6rl=|UAyTP
z;ty{`pvq6XbG*MyDfR*k)l+i)*o@?h95QJFYS$*k!+%ALK0TN~R4vatJ+wzVDxe}d
z`zjKKwD~k;Rf{Zj3<Njj{kzgpEme8E!1qPD6R07u*d%UIXnp82MSX_HW#hH%I_Z@8
z;g>UTG?+z+qiLpzOGEc9xF7WOQ9sz_#+E0zxy^!<G2`Hpm1?YH3xWhy$)TSTu@L#=
zyIa&o(l{oZ|C=))K+XV`0U`51OU)y5HaxA#cfolAK2|HqI%LC$6x#Ni#Ird336i;n
z3~egR5Y()3MR_}LEc^4<IAF0Y#WW8?d#JcA?>XqvlS=!(fxaXT?~1X%7;pFDph`>f
zpnh-zzSF<=>NnEEnLPy3tRP(dv*nu31+srPrOBaV{|GHl))y~9QlNKPh=RUU0R^wl
zqS46qG**0vbcsLe3#OH!_{d_rgy9x%TJ9L$MG`-Rz2AfR2d)nUQxoDt)Z=_=IjsW;
z>WjY@qPa=uiw%8A1W-~nyT$~|#c#key^aOn-7r@5Q{N;goHI8YQxJ61pOAYb;y?Hj
zM3Z*+`ky;60_wT@f2U4hSyMzH{DO6<u(m1pYsg2BKae#yrcz^8)~gLmQJQB;oGz)#
z3Ah9|4oMm<R8!cX%UI0r&Lkmy8kqe<8fQVpMY#0Td1XFv!tvfYF~xBzy^QXp_Y0E4
zN&Q?xf{ug>V2cSg>c(opK@(2krj#_##I&*afM9D#$+&^G056Sg!TSMvN1nGyGGgXl
zoukEqazGCiR79%P{&bjTW_(MLII3GfZVJAn^HV2VPU=ggehDO~ei+lxjTdAicMkP8
zxWIQI`lJ_g>V0&BbnkULL=2+$+F}E0g5AJu)W^(+Igg&Uty^WGICM`*wgF{M(UQxD
za0l#saDQ%73XLScPP$i>q3uA-0df4-;2LWThr2S$S=(WP$_R=%;5^~sNA~e#TS3*w
zH^%VtUJ~v61gC6-?j)xzoSgHLiwb1(tW?JVCTM_U0fkfxnl60SK*IOpngtpegTBS9
z!tuD1;X*m9tr94=pRILwqH|RVS~9Va*R7Vkvh=K%9TaA|&p%U_4u+jQo&5b!@(~>`
z?V)b%5ogRr$1)V3?3$B9MvrXG*?jA|x=7p)*6>H1cX0*uRsGCeA0V6M)-T4yo>E!h
zn^s(7NcqR&<SsUA2%i0XHhgQ${u*<c(eD;*=AO_9>98Ic3Do>CB<k4DhwjW$ks15a
z@Z*Rl!I&Rp^k79oeTMaHV+Q5$1HUTymr*o0;Q*78ji2J;U2X0G_?>|M9XFHyatdA`
zsOWWrk$QN}Inogi!}zQ{E=7a9T>VDrPal$p3WUg6P2dU5Vje%kkZzw>l#ZwcOnAG=
zdc^8494MnO^o>y<h5EZa9ST69e$iw^&SfIJER79(0#0T5RBToz$XWJ4%)vwI{H{!^
zvYb}e{h2NMJ0X|a;{>5L^OPC&pib8x|K*Mb+RM8Z;P_tv{iEdubgNqN^e2W8`d3u|
zz+ocoip<WkAOgWp2-Z03@1^(RAyZt}ZQ(PuVAJooW%(y5Dz3|icJ2ZWD8(YM2ln5G
zLLi`pq%U{_u~HFW4v%e(!2MIu9uI?m))`?I<_3Z?TcrKuFxfJKT(r8Sj327LhE`wb
z29nYju@284v{GUPvP>|Zy$J+pzXdvGPA`V3b6rxZJ5Z4n413nJp=Iu~`L(o|e1DPb
z32*mi+$S(CE(Yp~GSHVS;2oE&2r2GN)NCm3iG^c$&fGFoC;j(^t`89j9HH}{<3Kv>
zXWf{V`tOa3?luL{rw5?qaywh=Yguk{1NA!@CC;k#lAI_YleOGn=Y9kw)B;^gL%=4%
zXKc74K>yyrG_d+Z{aUH=A(K7-zc6z8E+8KIKKUIdCme<z*o=J?wa$A0T81h0WB8dQ
z0>QU{c3P*^j@W11^X)RQ9>{c}<AjA*3@z)ZM0fY&Puc%6nSOPxa3(l;(2)RTHZ2w|
z`m>%bdNVvc@XWLE75x9KT4pNM5UA9twL4-g7e5T+-U*d7%x@kG8FZTGc;~lOi0pg@
zU^qMVlSNbYSZVKX#{{fCDhdgm*NKJ=i)NgQVs}9UN9Rk0WQF8TRdQrEJ!GXh!S3^1
zuybfS&QazlYZm(WNzX*rfK)L&V4g=tX$$DI_R2ZoPn6b1y~q1P_q@r!L<?;1K}EpU
z_L(kvZ=t(GU?Bx{tpy*s&HOvHN6>m_v4PKVgLWQo=!5kZfma==gHir@s38T!v+&>)
zKsR{|%T~>Ioyl$1*QC84)aH0CuOyJ~cZ7eu=zZ1Y$|gm~!nS2I1C8D*qnZRyml@L`
zg-rjXJZ$k)4OBa?A}g8*--QJDgOhn*XH$L%OI4XO`GcsCTW<4_S$dW<6b@vegSozH
zw|ep8E?(i93~yCPalEODlZV19{|eMYBPat%q*A(WX(b3pab)3I@Gn+q)?kB+f+qAk
zbiAeSR|^znT*SgP1-~f6@wbfo5S1`yT2U%r`<(&(c;$`5bBchxYMQ)qN%D_7n^s9^
zI3Ez~fD#g-{@^$LL?bXnTvwU+_~16xv|#Ap{dKo_`dAw8C2RO7$a<T*qj=^*I7DfI
zS>s?8#-Q=K!*+4sM6rADT5xl;ZmZ<@Cl~3UcX^}te1Y5>U>cjju5=Qx*}JHS-49?J
zm6Vk`v&4P(dL*ymAWx4MJAl6GL>as*n?sdNO15e1h4kij@Le|7=S@(nK;aBJWq1IN
z1ma4+y>Lj}T|D%KbZnwT2Ty9|A{zcx!Wa;Aa!zq`38bCu4+3m#IM8(}fRzBfQ|vaI
z(vG)WlI>~*iX(f8)cdVXnx}9X2hdGQPzntkZ_TTf8aD((I=b%G$OeN3=J|`r@($|<
z%=eT})6F9t?~iMnZ>=&e@YBEr%*fA)`z+`qVM|!TBy4F^i7%RP8mJeaVepOX!8Xmo
zNs;7)oNQ0N0wOJNKyhzcT@yhN3nSvXPgGbnY!z7o+QGW}?H`h@=AT(yf8IkCoM~Hm
z%Ru#O8uZvfI<JWlRNWm0X8(+sxfHE}p8x_GD7ysKude-GS;+;o0<QCBpaE33@>YKj
zh`X4hfnS$1VBS%(f6R}Z0}$Z08Wl+ZJteGL7F>V*8cZcor%cgC@j>j)Gb>BNPswp-
zLWq155p)BhVdHq^5FmN-{}2`SlOOxQ2$9!jg#B_qKCtukohRiNXJBDkx5hK^QM9BD
zfB=|9mp_6+Mv{zL<5VPN@>I<h3|#>R4y{Jq8E#l4fur?hFI_4+GPgT+F4NzZ5b=M@
z(8NTeivZ9l=mek}X8>VZh!vW%9z!oCFVbv2VNCy$X}Crr)#6U@o0jO-yby`>5a74N
zWiEr7K2iF2TL8HrLx*qz{sQ5rW3)B~(D=QkG$|(>lPBa96uE%QBQGz%=OrfWu}uwO
z5*MeJmX_}Q@uLfT7N5h6Fj{2>$N@kgy%zixl<!ampFTVBfb1oQkf@_Fd{@!LOBn5S
zC6@-rk-|pUptvN0Mv*I$v^LNRwfox<^D-S^?g36&`?zu7>}kNind0Q+B(;Q}@LO)f
zsE;2x(C7LT^m=F|d4OUISVCvLRo58~B4O$Ei6mGu+ZoYACFLL5{Tl@vo|-w*W`6~0
zx924`aB{B842X$fJ)HdBrprbYs^#K<7On2HCl)w4y|>tb0jX#__(pM)5hb~#0(v%_
zEq-)E`8{SM9VZ(5icr(Xg!U$4VI6tTF|a%LJ*U(`Es1CPW2Mr3MaPlFW5(;SZxif3
zMs~&n4vEsr(u9uHSbE}6Tm!Na2={`K-j16Y`jtK-)Q7nWKT)vmds<qm3X#j$mGX|k
z+<K>1_}*D3eWQ|~-znn|F}>3Htzuqbq!PaV(2fb@OfXE>0tSqsrF9JJG5y?V_%$`?
zq0y(AIG28`AxG~dN$A<>{^&R-EjMOzNGAf8uF60$AO@vBWiN>D0>$6x>5VX%JD2-5
zAcUaK7sz4g&(TBut*ZhvZP=J3*;NlAK&ruHXhk?c3igESXJc`5cgP@gk?Nxm&^DDx
zi~Q?VdGIxm>5C4dMI2opxK`AIMf7vjK5bGIA;B%yM686@k4=?|v}#AwX-9u0BK3}U
zEVm_HvILqHqW~;5RX)pWvh)w=`<wsYv}BJ>d%`yMza0|y+QR&mo-|yuPcXb}fk1w0
zez)0CGGd7%&47ggZr&#2oUBt%9JR0|0)0D}ZQj~oJz(Cz(AC*d(5>5x-lRjvDFdK-
zqI>l>eDI5Hc<D>D%o!?y(?zc|0Vm4b;GMiNTiRVF_a7NpewW*af=bJVAa=M=fU*Is
z2bZCzikH4$2n0r<NFB?eEjWxTjv<JlgU@Yi1Zc;;;2Wt+tHz`*f*Xs-*9PI$1Eq^9
zk1jS<c{!6kyZxq5ET~he&!DH5q~rt9*2{^82as7qoF=^0C_O=L4_0@9DtWBz8uI|g
z6*hy+B5FdxZnO+C<KyTxfTkp+YcfuQ^`cPlE!r#+KML?fFMpmN8OwfDH=UT72HUB@
zJfTWwjF}ysE4^?Yd6+{?*%Kgdtbl&!s_St_jI5r$Gta6x87T6{YN?<M+1)ttAc!2F
zac!<~?w;!$I|!|36-4^YxNA-(r;LBVnnEy7LxHqsr9JyO=Lh|dl&^@Uc;{zs(CGyf
zVgiOs3xYI!OToVK``?maGTd+twUyFZW+pj=|Iwzczb7#-da2sP7grI4mKOEd9wSI*
zQ4w*oCPDm~h;WX3T4*hAVB6=k1Yb;>MR2KeB!W5z?B~9c)(n%z1RTIYCm5M$QBMAh
zfZA4z{sHq{h<jM3(=U7`WV2O&us~&oWcbjtdpRKTG3!y;Je9fgnsx9`1Iv19Mk_#N
zjA(sLq-2XLMW4>4lp2P)pz{z+@V0zP?+<OIpl}{vOD@sK_qaI)-$GOq9|Q}IG{6}a
zzwX$qzl%T3+%``)0nKiJnkO!f3BZ9OmsX2PWh=q6OXK7kl#hN7v0Fu$ROMMoF6q>)
z*h$YOc2Hk@e*=o|dH!ISOnHyx0~RJtgnV-K&#(n2=WUGA5@+Ot7+@%Jju(CQMdFo=
z=Gpqt;u%k&V^W#Rc->>ZzX+bRhdk(nRUPr%5w&}5fx7IDyUZ+_`%Hj-?YodDG<5>d
zSC!<{RpF+&fdZXt|5C%9N$RibG`&fW(G#~$TOf>}Di0L(tSM_d@aomYcv`zYee}r(
zsicJc_@sZ)%qP5MN=dBxsDJOrr?Z(GMfr@RHOJa{`^NR*Yyu~ErCgGSp>V{3@woZ+
zk`)ul-faFQIFIhR=j6i=(d!@Xp{ow&E;UG+ZQ0=29>N6pi=})IDtd*psZpS3NW(s2
zgZ?(}wS<AcKyb!U@-lRo4wXX2#&vTDPG<R>K%Wdq9N!g*`lQ9l`(o(6s!iv<39v1p
z_S_0G6|@^mNPG<>fIp9*Lwbk#0wfLWLyyVL!@k4$)_gm!Fc+T~-4T<~OYT_>l3|Ja
zilA)55lvLa7^3x%eF}^{;+2|lT&!eZ%+LS`hY`XI7U1p}r<UCNPvjdYjX1pTIsfzD
zcV|%^O+G#9A>rG6$Cf*`@y|18qle)HuxFlFU_(OaV}uZczds>*Rp36x>J*3F>9v21
z!KpB2N9PbRc;*go3m%hbX@Zu$L~(g{Y$bI54+DCGwm@_)<<|8Tw}RLi`vdsW1XKkI
zGK3+W$Pc(oYsXR?6wmt99GzsTII>1Z>5y4Bw5ZFtQQh$B`D^+PM&qEIBKChbQpFHB
z>u};7^mElOg9pQxgiA3YXm#lkXd0}oYLu>OHNC}jf`tT$Ui1U}`%BljC_U_+$mO5m
z1>b^l*!SIDEO=+RqqG&#diGxn7)n|KaAg?-Ajo}Z^z<A`76*JL8ld+eTFW>(r$}my
zuh$V+yIhy4fca61Ba7)2p*5NRMT7(rWk9fdEcF4_Q><6JC@dNcGJA9?sES?$hF(@3
z_UlF0N1l0wnWUHNqb+P`m)ssR;h>gwM7f^VHH0;2U>p3ir!WY#2T2p1Chc{PMyDe(
zg=v~_kI~}^AX+~*E(>SQQcE#Q!&Emm59a<}``@eKfW)kO242nZkiq}f5jrRx-0aNy
z8faQx<L(U*esgNMj~U{{#^o3ZdoQ{F$;^n<zAt-=)r(W(Yp`t&w^D_!P}$B>L%AMM
z4Qg1u*4DUBR7&`WFd4#QIHD?-5bcs}`XXzdap!4OLJW)3DsT45mOw?yn%EjbVomKu
zvxL=8S0$Hf01-V(7(fK2iE?dcuJBbDsmRmMG&Nj9O~xf<UkhYvM_+Bw<uFLX+stJO
z;$-D4;M`i|QUz~&5J4`>ew=i{&Ef@dLyN+gN|aCLM^OF`g<5UD>&8SuQ^2%8v1>3%
z8_8Z&F~X<JT)WFTS3cr?*3*h{57H;I;V!y8ea6w)&Y;kX;#2QIGJj%EQ$PETQB++1
zs;NYb`vy}=%X>T!BujrWc?AS078e(toScHv)2Hcc%ZiGET3RH@@o~WHX!P!xsK%QQ
z@ab%S1p1FknnLmN4B0EAaR1$A{CG8_my<I4+|7|)<IUFOpuo3_#)yciCmLEr2r<Kg
zIMde+pE!g+?p}pas5DVN!7c3?#<0yT9<QmNnUeT`sQoB|OETcg1KHYgqh9!oi{*Em
z)BJS&<we@#F`%SUA`ft2Spnx8g0+Sq^$i0mvv|epM%0Hs+rGQ*btH_zgMG{Y?pq=X
zrrP^^vf)UmWMPCD$Hc0*WT;~@)DuJh3s50x1afQC{1(0RwbNkq1}-X;f>YZ~ou5A(
z;MuNE=7Y8tF7^jU9{sydO1GLY@C*F1G)h|BX-|&cj(1++hx(n58+_u1s5t+b5kZ4`
zC+c0<lK9W&Uki}l!jhP|qRcoak(<bmSE3J(=rt42)NUV>Wk?AO4t$2m^+p6z_9!Lm
zmO}oJ$<GXtiAqx}pQFvtCn?<T8nYGh`M7m-bUZcX+T&CBB`!_K5Lmv;i_d?(M<0iw
z&2Qgw4iX2*c_k%q<QQ;rX8!0lQ&9@tew&a036*i~83&7Mlts18%$RQx(V~EcU9D%b
zDNwL=a(5SK2B`G!A`~FfVB#J-gk{DvGF%@K76Y-h=Z*a<p{m6z_wn3hwOBI55bwc*
z3Y$_87WKSZgJ4k#V^TBzpU$p3oXYqAKS*SYPPQCm&tw-Nk}^w3G7iT&cE~y^<wzv*
zk&%&>Y-MI-WJb#IG0F}lqa!2x_kKR#>w8_Fzka{#`nmkYbH@EV_kF+L_xt@Cg7*gp
z2Q|F<DvRCx1)W`7uv%Jzf?>a?BC{kz<>cfF?CN-cEL*sk^5_w%s!Dar4?Z#W9m|oZ
zca7m+?0^1<QP|Tknp4%&E6bKBY-#Vvg~^~O6y;fe^WtSQ)*z{17J2&{wAq!FDz$ZW
zU#6`<RqV8`?k(R9Xpo|?ZS-s#v9q}|8G*LhE)HbToN<q4d7hJ=#IcKCoH92M!X`4G
zcRQ33KCM8P{gr&Nzj-A8=NJ`}l|5Ck3fkPVwvK`37NFTOPK08R-UcF3RZXp2e7Uo$
ziwe25pV{4iRC~@!F#I8lbEOxThIc&Kj6}knzBGU&$7(*m=;q#={q}L#*|@aMke6!I
zt{aJoiO?bQvK3Hp+=q-0mO7D0qy+Ip<<q|&tRGnkxfkkJkG{i8njKZT>?Bx}GhrHZ
z;ajVT>iRDd+D>m=EpcjNp%=2LjmY~GOCm+G-OME(+*I{H9KgIvvf`7v#)-uaeR6wa
zI@~#X(0QbL?K;C5eBP_x#y+Ey_Y2=1hWTe`7avwZH~dQS-qo^R$)iIeSa<K$X_cJi
zz$EoM9V?Z7&a0dWUy{}H;qJbpPEhaoRIsP(WG!)2ns^teo)1#pjk>bugmTpoqn|!R
z!3|sUx+_B`KT`cggNY<hoZRr!nk&XQOdd0N{YMyEL#OpHt1XhM4xU_DP5N>O!x^Vs
zDwg=N`V&n!X(UhAIYXT^ywuQEoOx9?oi2rg-CFXz^`VT=ijNoE9_N5VjBVYtKL>r*
zJ{>ZPq6zU{8sjV_zY1@?&CXs<gI*c-?HFCQ9bQ}I@=q?aB;E)~i)huq)(?0Q!!Zpu
z6wQaq+jxCFvH;WY(Ey8%Sy}t)($dn*VhKu8>hQ=5+BVbVP=+MS)>Cdy{2wofV4ut|
zsXw0Ik7SG#;NO_r5s%?0CpFf&CF?bB7|ONk4t0k_JRp(O)8PC$!nc49YIFIE&x&7y
zdg~Qm|H{l$jBpdYPB#3Z?QBquZ;v`u@}tLyQ%`RpucCnImB6BTo4j@S^+OpU?RXY#
z#Amf6^P?=wr$^&jL9}kXyduWSU1@K4rn1W8nA!I0zSs?;y77f}gKPb(B3FPqS3pgm
zT2T?@#c%glOWVJ(*fLJoW{hb<PhG}WXPxdAM^?k_Twl6*rZl_prrjfY74$24co2rN
zRo)D?WqWUMFaIWNGL`6_KW`CB`0&Yey>AvgU^!Oy_btgQJC6+$XO-4_oNt#r56$cj
zSQtLOZ{NN%{I_6&zD&)(RNOQ+wpX<7MCD3-t1<HDKymZXP_@xHob8n>agmXTZC%JA
zS=r^(?7o8~OCOpej<IfaYtF0uE*bkEe6itxkg0_Qjh2>{c=sdsxRl8A*%u_cUlqwb
zj*Vp`c67JY|Lbyi?S0Oj86X3#)z=<~f{$<SEI~xDFkD4xq!1(ob^QDkFGwbn&2Z7-
zc!9O_>x14i4h^Z%J1zp%YrNjnLjPRLNA=&=#K3zLFfnSwpRrQS)V-ac7lYN*<YQrB
z@$vWf_>`HF5{*{CjvPVO*4DU`T-iW>xx0te-C=l61re8&Y=Z4%VtgC{b(OK5#h~8u
zK4D5tB1$Pe=JFvjCFQWKt*v~>iYqk0&&^qt`G7(Xz#*GPM+Fua7fE;TCaNcTx+?LY
zU22Bg)S$?htR8>*t=}{Qsu2pJwC~``oy}EU7}pz{Lh>i+&(F`N_r4AC9vm7HQc~hn
zQ&TfHP&~k>EtT8v&d$$|GGuD{*S>zW1(5yrf;Umu!9mEf*e<T6MGvJ+Eq;6Y*R#Ti
zzMrdr9hQ7!BV(MQdShiJ0BT1xhK7cZGsw&5)0LUkAEU5Sv}ghVeR;yL5_o^<&beX;
zKr#%Tj1iLubTbs8gruYx8#cPVpT>*w0_h_vD=UkvfSUkdgFF`#>Jrbu$VdT>)l@NY
zM+#&^M6W5uz7AdYYI*-&2r46}Iwib68#i}MS(##9Zu$9yi)W3DIEspjjuPGiQ2@#{
zRq@Os91UdeXcqp1XP=qF_h_#2`aRR#^#1*6SEV%LbFM>PqqPzU3FRGE71h+(AZ);z
zN|sM`_HcJUcp`L-)4O4(0%pXI-W=V@`mIF{aImM2RP~>^M}C0A!_9pHoPP7R5>+7~
zA;+OI=T_%BJGt%c?G{jJ7dhw56OWjudyUmIv$C@0Ia$3yF$LU;`+y@XUla@S!4Q{V
zl%nTctTj)ZVCUqFf;#Dwq9PW!g_kQlZ`c&JLfz?EJigR8>tx*Iq@#H9S<ow_uCK2T
z{`SDY)YS9{+>}c;Hg^FBKQ$6GcKg<?TV!8cY!;1-Te;1*nHk5)mYCLSXP+GerA)oR
zz>27jbIsHVi{D#fI0oxNRWiet4fArX-tek;DM{@jTq)Pxev83e$_iUP0O<e?9*>8)
zGGRj8<X{mz$rKb6ls9Ev0NDW*@^{gd*yA35j{}v<Cn(4OVAFu{7d6ILN3oIO;@(5a
zkByH{`WoNRCKR(ue5dr$(wVtIsW>nI06kJzzuq+wo<(baIN@LE!s2EXh+RBVaN3`k
zlmx=2Pg%~$7RJx1c;MWxG6<D;4k^OS<kO1agQ*J(pb5mX&CJcQt8M}67F*r+@(6-X
z*EWGrCSe1x)wAJ&JaKJ)NSI<BaMEIn0EUY%^YTss{fPWP=BC=xb&yI$5zGL10*gP#
zm_!SM??<m8H9-^#1qM+!r5wJsR+sOTsS(7&arE4~B$JadFy^ni>RtVrJ2O*LeUm-K
zpFf*BUA}~Ozv|-Rf)d+azIt^N_8(+ij<~@H!124VLPSJF0(5M;9NZgrUZ@Ia2ABBO
z!F~hqb+Zc>$Pd_MK96jNcJ=glQAkCg%?EnihkyL|!T;#<^4EpXQQ%LarN^k~Xs4fl
z7xHHqM}RuFy_1tuSn=L{_;l;qrPs(jHKZyN3kxG&ZBo+GEq^Fe>>M1o$YgS$M$t33
zj}eTFjJi<F?u4ApGM@^8S$;b<*620sLQyBVzM#06?$^{hF2XqP0*^s;K}_C`GBC2B
zqH0?k%${p-e?Ovh{P;v?vJ_PJ=f6L~w2>oF>=ar#=1nQS+7RE~eg<Wa=H%q?bAQ5)
zbcfpc`YHgZ3(rT`T^h5d_9PI%(c=sbhvRyIQ{-Z5Z*PCMJl*w-8BdLgw0sV*I>0T#
zmUz>ySfwYr|3CQ<s;CY2oCY|Y$mHbYy)mbf(o(kptcBO`T%}>sQHNjSjp%+-60s8k
zOPdCT>6AHl;npL6#dBgX`(d_%Jkt>gi370x4ptC&Wv|xXgR=Bl{*R8;RaGc(2nF72
zYa`S&G^GyB`w{JcfHy5aq=8^wmnNg3rbdMTe~OMsBuZdAbYMWR(Sukm>C)BkLn_{*
zN);UrV9^M3Na*O%{f&){^k6~3>s@=Jrbhikme$odS3=UJGH%XRK|z69J9ePdsV$04
zM?_4la#(<iOWe&Xue{$|hu(L(^AM$7_&$ATXz0ZBwCLp2)M*Qgq<|b52oEn>o*E{>
zlaKZZ48*XWSV>R~_?|#6mCm%xKLnFhr_hoS)~JH-q@EeN8Aw}KhixdnxS?>$#>OUp
zDM*^zrV^Ikr~FHD4jvv-IyyQnz>1Z)CXyr`#rEVFqcIx05;1af9PI4u*4&s#;N<C9
z-htCKJNtuV2RJzE>EWEM5I_e1@BdD91!fFHy`=fstJQwA47Z#9i0fdn{uJwXdU_af
zWo1mI*GLnX0LV?%Kv;kH@Zr0zE~6TIJ+i3Jz68=s4iGC%tt*a;V-5=o>n-HnTQb&L
zN54l$ClG}dAP81fS2y<caRA}A8LSMvJsScScxFDN(!kEDs;-VzRpo+~fKmnV&qOrW
zad*!7`T6nb>D#Ht@H}6Xc$h5<tZqQ3sOi%uc9^4oR{Bhq9^qmAp$?=DcziUf%ny1Q
z>W_N{i%i8!uK#w?-29&F6A+*Wi)JO{#J?gcGz3sCTHIE_mA4pJrh}E<g0RiNFX+eh
zS;|on!C4lj!i}@fZWb5Ifk1fL&Ngs8G`HqoTUGmsYHDhZPfgt__ZUbc4=`e+-@S3`
zODlwUa}&0Ejip+r&kCe=#ZpL3tn>2ob!~0=&?-Kx9PPXdjxb_~jxP%f-$RaXW@~$I
zXKP(pPL34~ppEtj!lxO`28_p+hY}V(GatC&RUlv65<CwLEMW}|UR22&IuCK*zq_Ku
z>oI`P<y)htnd`r2CctZ?x!QlW#JLNDM)bVAyzYw2Zv$mqdNN`_0~*8uk^Q%*c6WCN
zU*lQ9z##hRQ;`Cja)!M>ZCVH-Wm)hJHG1es63c!1bf2iGC_O8yH|2M1Y^;KFC+9t$
z;|!>wDx7ggug~(=ZZCIrR~VU?MCaxT8)s_9pd@j$F9MwKFY`Z7$6_X+Wg#vhp-`mZ
zGmd}<t3v47yz;sjmhmVYgyw;}OC8z%@#A3#lk7b`J%Gl2(obk^YH9(Ay32&EiD<#7
zxVSiRSy``&h^D9mSOo<$dLf7iX>j@^{Jj>fwVgAx27VMpzgAa|%hIqx1`w7$TM7||
zGU#-lJH^S-u>~}*(+K_>%g$8VKp@wL+S{v(`Ff_N3GfJ73Ce_Fk&KLtD7)grI`j&G
z^HwlRzq_<?A9sHkOnHC!%LkDtHlUhfpzCti_d0N+!203ur{OW}ew0`dJrOCm`7&^9
z^I8682=HEBgLOC;`hb%m8sG&41PT{nz}7&GuIjIVyAMZ2CO*}Yp^!Lz8WWR`mg^l|
zUAmT*mao~h>g?{fw(64y2AUfisSt>|UmeFF4|4!oJnX1Cy1F!ws@-7~9G%(N`-&<`
z!`^sX$H8_T^t{1T83S}7-G3TUP@E%FWWk(;C~<C+8wm4FC^On``s~BM{+i6ae^STA
zW#-DX2(}Ay2ola2#su-8rGoHfe0G)&(sw$6V8o^n1KVxT*LCBC^73-nu}=ZZ-qX`F
ztsG+Z&g4lrm;=(;GFDkbLt`q7pPxUyd||kp5uHA$MMg=9ysCzV6Yzq&>+=UQYl4<W
zJ*{(0xfo|FzybnHeh#$v^t8rsi1ofO5rhZyT-foTnV2K&RAnGlcmsNjV0c7m#eRck
zv3vz9zdD-1DMM;<nV%Mf=EQ;Y3i;<Yh9IneFVe2qz}h;sgSH1$*-7xyC&azoplbqX
z7n_x}4KzdJenvzHGa`e-!#c$SV-!_5nKzBrg$hN<$5XpFJ2wsual?8DO8$`&91?;B
z2ELn{8+<>n@+H!lh+qzS_KrJm-)cZ^S9<NICED2o-)lWv+l<n1$gX}-D5pSHysWIu
z=f`JJRJk2Gfk@=HZ@AjM-7N@Bt?;l=8Wf7ZMo{D}nixR(+6*ofCbsoq+;B%h>7AcN
zKy%PN6ISE=*J=*9zv7Spz)MR#24q2*WF~(0AI(O9d80;5lx~%=aWEY@Vf<LV-!mOX
zGllBzLjq}LXE!>xq5#?5`~Lh(6acrQC}VVcgJcY>vgw|c?Tn5d29prTXh7po0OHz0
zpe;6$1<pW;%+JhYuDlIIh04Go1hEUAp!FOqX&m(O@Q8(EOuNw1sq>)(RJ!PQVZY#!
z`=Jl?Z}5shDY7s*<AM1oEGE`gMTD#kR(CW=2<!vhGq|TOj2|mTq%1a^*4Ib#D^<U5
z^r+kp3KuVK*j0z%y%a0}u3elcP5{IwgyCodk`mqeG*tGK$@laO3_2i#0O`?#<mCQI
zRmkdFa&)6n+bxIYI}Rv(xXQ;No%Z`knyN20EIgc@oBM#4Rzu(kWp+bdp(r-<=p$Hc
z*G0f1Lj=Lf$Q1T3X&ij8{YwL)C0P=tJvfq}TsR1S&9F!$fhy%LR=e+<9_|_H4}}(X
zHf!LCLarAja3W-UiMXdx;XT%4(93>J=s(v1hTGJ}X4w5Q2vmI8mKA32IQ-vpr4a<s
k|6ff0KSK7u{ynotm0E4kxy5Kvi@=ZJX=A-29s8UA19fyaa{vGU

literal 39740
zcmcF~WmlEo7ww_DyHirSyHmP5mF^DdL$`n+AuS*+Ege#dNOwy~N=U=q{Qh^`Z*X5Y
zkTW>Xv-jF-%{f=>SPeA=3{+B72n2$mq$sNefxw1BATaC5h~N{xiRItmhme<?ftR+c
zjhC;b#~X;MrI(wNtCy3#6}8VB4^Mkn7an$AcAl5ic3xg?p28d)&j0TZ*j+trIevVU
z9S09VaZ@z(gg~$?pZ~#>OO)9|AlVg4vQj#JxrZy>e(AQ0{HLDFwwv+^@V>sjwe3=!
zrAsb_ummi3%|YEOUF|E!E-hcYOA(aa(Fm9GEVyDAdD7g2KAt4MwAZGdr@s<QJMbkJ
zO2U2jgM_qg=l#LEm?EZdY-wZ~Xd|)KEDZRGEv<N71e@~wSxIE!|NsA_Heh07!b?fo
z+cTph!qMZR90-CnAlq-qYC>0B6qJ=wOC~GL$BSW4QPbbQr)6VH^hSSPZwlhik7DuD
z)6<o~5-gC~&|V7&L{xkqqa5F1(0=&x+GT0C^>4F;G(#Cj5UzNz*&WK}(xr5=)Ptb{
zjiuGTK*+b@GF#~ytuIMht}B5j7Nbw%W5V)cn%H%h>L;66PpvuL=>Y^ce1Z*!L>>2M
z1>d`>`;Uj}%)TEurdP^!Bh>cXIGblkt0Gur@sOl#co|{mr}PS=(&!(4s!v|OlM9ir
zEsUViKRZXuJAFk$5fM5Er|N}&z;sE^O_DB*aX^fr_#4*_{r=u8S&hH%0LPM(spdMU
z=F-k6z777YHsk^A!-o$Vl{0&5-4Z@Wi{FHCEqg2!`|9D442P(He=gTCD)~nqi?J3m
zB8-ZHhgyt68K2>ufLP-2;@*YhfY@Y;SjTM<ZtTo}tq77U5h89gmK`}~7Km~8??~qA
zc!ewY-cLtQ?@N+=w&D0S`uyRWDr;<ynL#gN%-RORyOtwMibyC-<Op=CEZ$(qPSSF+
z*Sfwwsh2O(2qxLs?){Jh4#K+m)2+Vf*)Uxemo=iW_a7*h#67lAHDP|xKavBxk`HoZ
zxa(dz^+{60d~KE&5lr*9Xi$suD(sZg85$Of$3>%AE>exG3peA#Dr;WqFnH-Yf7BXQ
znpDN?>LtYiqEPXCtiY?|Z{X>F%o#J3xomzAJFRs+6rF5c23R&MkanaDQ>Tqm$KAhe
z!1Lg1G@^SajL|bl)oo1?>89qG5VkgG2t!;NJ@jMlL_0$YC3CD_7!eQ09HOVMPe+VS
z%f(gSdw_?Br)g#uJaOf}o9Ao6{o#*!IfBY-O{ONd_Ul_3+2IV*Id5`u-CYYdi6LRs
zjjUPxF;kd%<^06^h8JMf1^g8ojb9ChENC~;dK|fYk4>g$XJ@-}#RA`Z)8k6Z%iw;l
z+<#(5QOAvipBfZ}VyrzPg-cd0{N79_Weax^0B%sODt}$A_G}7;iuJ~`%M<!Is|Hau
zuzz=V_mq*><>lper?+u&sPr#hm|g_wun;83TSpM#3Vj_Y-5S{Fe?GS{^eBS?8(*2@
zaWn%3n%qryhm6J7;oY;n!t!tqX{2gs=B%O0rTAv9bt6=an`n}yTCuC0fpm<FWslzB
z)YOcONN`A)ud@>Ba9x$d#^eRUl>Bn09Z_F?^RJuJ5P-v_^m_>@JPMzB^Y%gf7s(s9
z`FqtzL10PPo+`tJ(-Fq}_3s?r1An)M{tn=&_w0LF4JMFO72_a!InZ^Ob-5Jbcr?-D
zGJi@6#y;qNY+YzhQ1nY>!AJJH*>7jmucYyN`=2TQY}9P;t|3D%5J{ZIG$73}O4Po~
z?ns|Mf97;u6;02~ELKRSG-z^|K_HS*{7#*!2#JLcr#6{_I7|EC5ahvYS2{VFeubC$
zw9+GtxslLk7rtZ_wt+j8hg`tDuc7;Wj?JP|6H}xb67^GCT`aPwh}N&u>~Oga@8NDr
zW$e95Wula`^SAA_6zrd>$G$wD)s7=8XZygZC-l7zUs5qAO4R&taq6ZA`IBvrQ1=Z#
z6ng)DA|;{t4zmvUxObZJn=D=jVUMoInC5ULI1rMyf;ot=DCUWoy7FS89nLUhZyI40
z6}9&%^VNu?6bEAB)uS1NaYXcLL9n@%SnD9(=VEZDkTsnyUbi9B>N8_s2s|V%Ra(wo
z4lVro^z<N$Ky9D8nK0reXor?1n^X~TEW5D5b}D>xv#mV8Sx?f~S5oHRlUc}BzEQ{b
z(5k=!QHn1~VZ}k`f4~)?8Ixnzr}q4%Q+0Q1J*$b#$UA8FxC5iGzp>*)PVUN!NRaFE
z2ToBjEp;UQ^zK^t{?+g$@yAs(j|{?g5B@Jn(t-ukb~qQOeB08NlUrUm=3DDM3loOb
zni%bgsFKpz3B4&X*`(QJ%Cz4x`j5XYbRMkl9ui5|2aBq~DNVI()Q~i)Fh`8{k0+%J
zmVY7u-+0%XlCXt@^(9G4QIR>_q2!7zIoDO<4Mk)MT6wX`sqqa9gfBNSLi<3+O{g3~
zUovu`JRCZv_fr^=hDLxTkQ!=%9Vp-mqMegLmkH<e)N?v=$ldJ^6G5(Cu-=bSZf2;>
zUb+mhS$&Zgms{Hyh?Gfs7jhvAIT1bT^<BpcLVGjRVjP!e;0#dign=QGsC0(>7t&`L
z6+!_4S}=-6FJom<V`%xtU*IT}&KBy9<IJBvWL6bpN8x|^>h9K&GWsX@_^hM`^4k`n
zu1y1cfj46jxhx`|DhUdm%)+Ew_co{xxibR>r$fbyo|5z{K8oXc!*(KwO%Pfl$3yx~
z(=_NkptS5C7{!#bz3;zahpVm`ygw}~W{zmL`PETDhBP!Tqy$M(et`}BWXfC$o6=}!
zx;5p#d1^V!g>27ll!p?VhiI{tyw;E;O<bOX*r&hsrosrNJ1769oh6OZ<tM6SwHabr
zv3-7{CRksipEmd5zxOQ{>PGMXnhy19dK1o~{v<tp$56)i?Nk6`U|^0HyGaQtIU_$D
z|G!vlQVk4?N0po`vQHPjet5g`NJ~Eri274vFfv)!4ZO7HMJDyKMme8}8r+6GWt~=S
z<Xo>Ij9q%Ki|gXfv<~DQp*n2_jpA;xB9w>N8YgFlpiP3qf+CWaj<PI_smSZp01&`*
znquj4V#E~5>$(|Wcg3{HgH21WOu!zBX*tt@EHCATENd%XQ$#jeR9`F%OS~GwAuSdd
zwQK=<QcU9}e&^=&1oLbV58POM?LpC)OY;zw2M%Lw_OdWHe#|5RLQ^Vy{Y?IS>*qu2
z_=rMMw_uNu(uh7WF~a%jwdKK=(yHPV8?8C?#ZjOGD2IBbU~7I3IiNcTI;o0aWr6)~
zej~<Rchh@bQp1PICqkXPdAQ86QDIbjL~jIY32TlS2D<yE|Db$d*KO(A@Wl-2jN4)b
za9yR55|0?^so-LqKD-K@s93r(y4}iTfYoD?mb@N(7_v2bGN-3}#YPmRn-1hAE2z|$
zFu>uN@k#webDfp2C!w2olZCO$J^yqofq+r1vesen9V_rJ<^ue4I)$rZt<x{X8VcnV
zi=aW_)8)lYYJ91Uv=;kCrtMUCe?*;rinO@Z5pj&HNeKf<pEZ<47(}|~9&TpRRy5{+
z8+!MhV3?ej;GpXf>Wx1MZTv+1s0S&<MTSXATD23|@*<JQgdt!-xEIexP*F^a7byLu
zC^$)rF}{b<rkrScAOHX&8#!}IGC|nS`MyfDQYFAoFb8RQx;h(#E_|q&b@-5sF@4FK
zua@&cg6*~x_qhZp^t^t#h_nI#f$T_QMu*K|=BJ<kxce~_UoQAo5@*r`Uzu<6E=l2M
zOMZQlv=MTo($I{k#edFO>))el+NuqsI2>Z%#07jtLUNKj(SZkBnlrRBrv=19cBoDT
zbZu-LHIsHUbSm;bAwv7Zh1(P=H4R%DKpwiltA->gWFCQ}+@wB=?>MLDGsTG}5|I<|
z$uoVz@}0avWHq0{hIN(UJ5-l^PaK$aas9+peKn&1*gaMTz7<Y{(y?VLSIbmV($#I;
z)q=yM(G$Rc=EK&xkRnSUP$zo<9^(&53wp}}LAj|Fd=5KQ;GvcdPD41Si)?~*K^zB6
z^iX_xJpC!jX$HJ|Jr)R=HRq^h8Ei_*r?T`)Z1l-lA{v2&VC)-XLK)<<rWf)xrbFfE
zHQ7?Emx%C(_bRK9>vlHN7T9^O?lfS!-vTAv2=~EVF5x(n%Z<W#xFVNBgZW*yK+DAd
z<)X`DSD}K+>z?0U`9gbm{fd7jZfRzc>`mR=B(Ta@Ux7NbEh<099$LrUb>Jt+5@%Um
za1yOpz$f4Dx@C)yqVSBZguwxpe-gPDVHrx=h%rV4eb%WDh@qT}sfwJ=Kl<lS#~85@
z{v?we|M?VlI=53&QD$nA_|?DX-s#nx9a2H-*L`*d5Z}YD7b=R;c`mScun^AwfkF-u
z2MBKy#?_xKFrhltvOQztsn7memXS{&323G)Z?Ok^@@3J5yW3X>gCS}tSz3U4Wa%j}
zxsAg+HR)Hq3xa})N(g0<;Ln?+5TgHMrAQquU(OF@&M49b?-Wb2xOb#}zkR?7GKUHs
ze+tXGI5oHy%4BkA-%#-vIv~EZLr6)n`J_d%?s0Z56N^1tRGen1^0ROD%_|Ueb73v^
z)NxE*dXO0g3Q=-^Rm9V7^Vu~7cmzNRBmtND^VtfV@5X5QV@tu=0A;I_-mK;@pH`NN
zTh0wLL>psan1lTdMe)bc^5~(_@ei^YYABfv`|K6~!iG=064m=->7s$Uhq7u_pr^@;
zhvwFbTLFvEMNU9@Y_C5c7IBl}tlN?2j+X2?AZV~;@Y`3)KKoXlpEl~DPw>w$?7sMG
zD7szIW_{3%!}GiM1{Qd+z3~B8rw6c!`ULAia!CJMtzK#L`0i_p2yAv$inm0gnA?2^
zo4Z5;ZQP`I=3@mj@^ts))ez-3S~1e!+xGA$5G*V$X($8sXc}EsNJ$eB7Zw)$j+*Ay
z&nG(cCj|jTO4)r~(uYD>+f2fKNNJ~>UorwTt<-sr2X{A*CQ?|1;yI)27<S0{*%>u5
ztoZF79l2Nle#qU@+UZ6#)n;j?nH5E%n_5|X#5Wa;LX@_w4U!;g?ecoZ2L`Ya-F0tL
z<Tl<4d}PEN8(r=qv>p^ZhV}DtiJ?~at(1(6Nc*Wz8*RHE`-|k_4HW2dc}UaTz5e5a
zjaiKTll}!U$Bip3DI9hHIDKTXvan&r6VXPU=$L_1Rqy1zVEnk6nwnx_V=XBn+-g8<
zo5ItS;GmY0?gtIg>@JOi*xpcK3Bz9BZA~04`2B}AW|b+T`aNpTfzR!Q{rwizQ;&7B
z*zDetf2Y~JQ!^q*$(vX{`reK5XA9Q=R6>^SiSBG$0$!xA_*tj-WXE>3y01K9Zg5-<
za4kl32Cz~<F}aTs%>7KRkSEZ#q(LL!zdN3azj*N?i_;R;rTvKiIn_!GzV}`AM`B2;
zcznUeR?RFy(MQv!f(guzc?gw_BUAV(Iq`2B`Fh{OR8WIB(areU9<e%Rvfp{fHdW49
ztKWX3Omdd&b=@5Q*wy<EC2<G59I-(DPwM%rdrPn1yqS6O1&O$P5_zmt1meb4vP9S6
zP`-#SJIEe&pKNo+<mD{T=rb$VFTWOO*@N9cjaiy7nZ=^4`?1U|4}-88l4RIrEV)6h
z*q8MVNN0$o%ce7VE~;9M9^{Q`Gc56Y1dF$CMK`y$YWJL9DJye&{8qd;ULng5c@!#{
z{LT^ZS%;-u=kr0LQeW=hZn_PlQXe1`fC-NTao6&0SuqdiqbcD6>XT$h3pqcy-Fo5R
z3j117AV3N@_uW60!&l#)r6Mh|3+np7n*K`i#)X6$E)jkCBL*>wytj_tCOpkxVP@`)
z!J$j}_%X?MEr>{0SC@%{L&0B&;+QdWj+q_9KiZC-oJM~$Y66QU<hvEJR50d68k#rv
z8@CyBVbwLnN28B=5N{I@{eau&vv&enfHO8Z&TU@oPYsWZNI5vLyh}?vFh_37g_h&b
zA5EjBq_7rFeMp%vq~d?P5IbNh!7<WChb1Qn&1S;8Z@O18(;~S%HMnU*sNx5@U0}@r
z0)$DKdOoHmKF6)}@HR2;gj;(x5U;rWgT7=~fpRzwtjfUlnFU<(?tEAVZ+<&c)&WRx
zLHMQgECFiIRkim~;Ar&aX;Up0v!d%Zmei>|rd2ZU;{bpucT&QLNMy3>$agh|@*xtC
z`v3&7_aRl~J#2OW=!HTiZ%uAx<STW&83{r-K1GopqJSq=!p+<>k>=C}bY*ltp@oB3
z9t=GcrI?FE;oq?>MMO#+cg2-X7Fu$zWetF=l*+!PR?$f1q|kgD7`(~Duir}h<%y(_
z+gP5(nY=_yZmq;8I(+g?XHsNI1!sc-KzQ;4AKuwBaMH|C3p0=mmZ&%xcO`h^pNXKR
z#a7x=Yc=7*5J~JF8=*cg7phz+Pu-*b3L@P%k`B{SMC&TB%+Gqj^jiR5o^IJD7*$dZ
z-+W5)wKB>5>ZcD~y?EeDZRiy`Ns^&ExwsxpC0Zm5iEkkXU0wIcYLjpA8SsAQGTQ)T
z?kjmlW>JeoKs$VAEyF;&{pLoGlFf2d28=`h@*Cr8N$a@&q!~wW08cZ_ENNhEs6b@%
zNFZ3Oh+2R*XBfdIi(CkV{ok+BlZ&{e0Da##JKgJBc(KmlXL`?j?>y2tXa{J#jPy&Y
z`49DTxBv0nDinYKzwus6$$#Gt7_<c{OS&e)q(*Y5fO+`lRsv)m5F>3~{jK$9!O(Os
zDx{G!?Phds2r~_M9LOQX^Z41J-u)_oWGQw0qQzLRQD6Aft_=t)$Cqh1VqctjwyD11
zrtO&!wXkd7R1uO;zx750=CPD!$)I8Q#0R(>+G*c^qVV@)(`jJEHqy;=DIvMRzsH4D
z0^!Rl6A&lZdS4o%8j>Ux=e1d86i;{v54s<~S@q$xDI-h_sV@<%?!X*iaO3MnKhn|y
zy1|yt9xHE-z`3}q6YxbkU}DQw`L@Q3ZC0KTP=ne-x&jeA@<PkPW)Qa574*Cy2TPEJ
zlF6$oO#FL;d|lDa!0J&PTWRx4RFk9;q{cUk)1bCB59AIM8+T{gK}_o_)=(5xmkmYf
zHvl%|CAi38zlen9q**ehu;FL9{@5o14Ih>gVgj~F1LbY6&y(-7uIvq?xHC|&PX!(W
z4&Une@XMp4i}g4~CjhViBIE=$Mn(7jz37zgH(XF^Dpm<LrLuCBXYKWzvH&{n!%LJ>
zHSPi1qm5Rxm9~lwo^M=7n$<KSuVy9<CX+zO%x!RY!;OXJI9mX!Z{fKxeAZ|zCWo|&
z+UWZXC8fk`&Z17+k>%CDZSf6W^(+Yr%B54<#@&2qtC?oqs6U@T)1Xb|9T@^iUkdBA
zQY9=zkb;nWv72x=E*Zw|tE-x{5p1k%xV<%HUzDqO*zic)t5*dYj?~xJOl}Zr)3vBd
zcQ?{BRLrCu7y$&K0|*qF=K12T&i09NdRDLR=nGw&u5m*C1-y!8EGXCDHYW@U4!rh}
zKi-rh_BpSv+QLSB_A+u+I8ESWhAWbY2lU?AR%NkzAe>Mc)oSDq_&eQ(#9Xaz=|bc<
z*N5uqa$q367At%~`Qs6iq4|_rP-uiw(5GG#KYUD2z5^yGOw-7~3~Iq*Nz@eSLgJTa
zH%HBTDTEB!n{s7Q(L~{tdr_qRWG@(C2bcTPX!x}^+8?1x38|3<(f67tC?fyC$GGIr
zIjV`aMVr1UUhGn{<j}l)Hhu4O)Ur_^=2+5RVIxET#lQo<Gbw!c2X<hn3E}kMLpT4?
zZ7*>~=tH0$1XA_hBlpKr7_?+IK?@hRtH@f*LS=NIdHcJef;|pCL-40C#S*R1;;-g9
zzR!3-qSPsc*v{5?{_el^#uSAz1(D~n^QciV*M#KB(n{^t(4~X~aL8)*gauWmX0PIq
zCWg6NN%u|t51fep11B7H7v5fHu7;%E7a6f2eR<}45wJ>LkyzQdj|(zxccFLGX(La@
z_cK9vUVFUop)q(vEgRFZd3y`CxJz|VC>KZ_t7rshcMsGJ6k`3JD@;G}8<Hj<97|6Q
zX_ziasFHb7h=FY2p`6ygWML$O{t}@hf@e-td`GS_dvs;`t^E4Ul~q<s%Beu3u0VML
z*L=MV+W`|!B47rsL{u0eD!ELcPRrgGvsutNKSMX_=($^o-JIOh@%^{tgxQrs4W^00
z^Itm&@`?l5{J+`LDd*v9<MVS0Q{^4wHa|mjvOt4_8&j=Dv-bqkxz}6lISbN(;7)e%
zli!jMyl=P+`R8}UIhS70x`!<K@vFJoUzB~n^^NYrlTb2)Z}jv<!`sTQ(R0IdT_5{}
zl?NtCZ%C2KVz_&p>l>0_Lc0;42u&sM^`rmkar1>Ul;HTZ5ZDC1aG^L8gsj_5xHyS@
zA@5}~la*uawTO}H4iliIsps>$WoouWiRGl@zFRNTzN-uDx4EDYO7PB&DF(sfcQfyS
z@JA0KmRkimlXukVH^f2TA1VPAD$@D`Sxjy`k)HBIdChGlC$VN!)5!oU$u)(EwgcpO
z2iDYvovGUlNohs970&x${1Y=c357^hd(w(qr{6^9Q-R5hq!FZ;WY343WBo0VCdPYA
zAerz;ekr|%c+sCbVG!TzfD5sBPz*h@fw0*q+e}hXTLVDq!GDmN>zzJKpX%my{=U}T
zjXM%O*frD}9pa5pFr|j@oUUzo7CP=gT7%T?k9K$6OY8(D@fM{Y*Z4U>Nmp1HMK}*-
z=Z|OA7A|gml+8u>wi9JytRpP$2XrYas!R%eVK!Jy;OI==sD9*ct3+9wP+Q3h*&zGq
z)IeG2IdGLv`FVL_NmjIB+ZZFT$P#q0b1Zr(AwgPKd<=fPQFL8NbT^WwO=dzh@OC3$
zIrpJ++#CBfG*F*}uL)W=>zDuaRsdl6yI?i$D}e%!XO7q}IY+}G2*g0yG4&Tq<5NC>
z!x9|jHdL-Cc-ez3_a28RlNxECT^&g1&{5B$7;9?EG^tT_s=Jmn?e|ZcK}?JF<Xj|q
zZd+Az)EN`Cl^PB2^CN_rKY;RGGu;ib$CJ)17X%_rjjRY$k)E8WrSvHLx_JV6VLj=K
zPouxyx`dRWY;JxJ(=*a>R9+zz9VX*qot2O>yy<38=d`7tGD?tjy60qve)0Qr#i3|6
zn)Q$xSoeAQTu)YZk1!&nM@u7azKR%73PIl4df469=nL!TS`}SV+aKZHr{+_U+E-Va
zib~GOt=AVjNpgw<DDZC$x;$~5N^LnW{vl!KWRD1I4S41XCDQn=_B}t*u|=k4&SNTM
z+}#EaS7fLmN%O>{5Q(JvlI9@`M2!bZH{0XD(<(e7JrH82e&iaJmpb=29(l5w17AB-
znS|<am`$TCep6hzX~WBMchAAasvv%NmwX<gpym=@ugw6GT9zc;%_NBgTZ&>cPA2I>
z<Rc$m;E?w}xiX2w^YI=-k(Blski&nilgEtjN*zbJ7S4R&v^cH4+*(Jb(J%XSORnc~
zi+FpjvwS+FPow|YK*52l`QtpQOinW}P6Gv7PtVc~bL-(Y=0Hi70XSX1T(k1&LWy|A
z?l$C`#C{(URMh5z0zo3~53qxH?kV3h6je{4n0bi(_!-E_Lon^tq#lv);v}SN!iKhb
zCYMjJ-9)%<ABXmRIN+K)tcl%*fcBmC%&(NyVW1+?3Ky4n?Li{IIzezVGmQKf0Xr1b
zWByQh`nxlP;C*&TvKl^Zx8#0N@bxS4&L3R-9jcy%&EmaDXL?~T((s%UKN2$ziHFa~
zC`jSd7TaB=<>En*{RhUP@j9t0x+cXUraog`1Q~8BHvPA1M6XHt1Q0TBeh3XTZSBIw
z@P<%uFFl|4aGt~g5@b6A>Df+u^ZrZ_YPMpY4l|<BSL$2ysfqI)8)Qrwq_3OyF*4$k
zA2k38V%&&mb~qAjF8v!wY*<_W)G7b!btOv#OIY$!GsNxpGy}1SO<j`ac(4N<uJB(^
zn3BClva4mBTeF7e`sq;%kyr;MqlyCFhMNItN1}46ocw{yu-yvAcfuzO|Dk4Do=V0C
zSi0Y+^A44DPE&Il4(M5;=W^wZg`FdcHLFR*#ukS?>UTs-6R;QE?)1-Ik*sfd-}nej
z%lh+*nvV5iO@M7jAZzYGE}2pc<ezTqJutKG$VkC!n{v;EZwDlD@j6U7glKGJ6G6%f
z&p8|>4iSv*@s%n=6cmBrZXyZSThdn^;@^H59}iw5<ri!w<0VZoP6S>{f_Q0*AH~U=
zi35T9laxpW1rdCNa8IklVmK_}5^o6eb|aD~<=wmRi05vAG_hkx%bJ~zF22>}Sx8~f
zkksD%8)B7D@J#6r$kN!mKPe;hY{QkS9gpXsPy|i<`ki5;bc%(j>5p_=w;%VHbB|~x
zdv@VuLDMPQ8y^z7b`+*l8VXILAn0B~Ys=1`9NzuMH-ceN5Q!hks$VmZ&F5I>wjo_*
z()P}qyiZXBd9$vXP`UKD=o=Me&x=8&&7KgG>PK(a%6E6SC!+zz*VO#%1h(<X(v^Ic
zzkZ2i3klI|&|ED`;k%e4YFF~N{#}@()@CwLpzUtBoaWzaSOVu-dpm$aEPU$uSiAF@
z3wQC{leLq!8<Au|Rl7>zEsTMtW|#t{KYTn9N0i>H=zqqajchRpj~-zb4ErZSuD8dt
zy^h35P1}>@{rBy21-l5$^2x*k5*6eRC_xX&F=_w`#R3HKN+1MI6Pxq6u&j_zt|3X5
zz1QcS_;&TjP=_f*(A!ERK%E#AHpT8fsP3(li2)yYjo_F%0WT05hRgwx(;NU9?BDCH
z<YM)Fai*6qJBqH(13FIwLOMwt8WysdXzVeCZ2uKpAF8Qz?JH#!)g12_Yble_m!6(G
zeZc+*9|2P}-%0#_b8DEGlcW>z%9<k9B&w{4d{B=(uZ5LKMujZ+qmDjhrSvU@jNwqZ
zzT0S`h|B&YJylM!>QyDL?N4$=`gii?XJ==p$szY}KJ65qyFWWWCqMlZIq7-LJ=0xM
z`uF^sQcbSJUjMzRM30tng2-{m_UCvB>EY7t+~siB;Cm`2-hk!nLr0zaNSHv?7wPr&
z%ef=yenBb2B`KpaA-x+${!V6l#WxFVqB$p0dOlMoiYVU+dKMaNt#2>(6T4LP^ay8b
z&6AIoT4YKc1kK8UDOz_w2A*zbnwwn&?f%S*H*Bz(d0Mb((8M9Oad6tcesiKeJW84r
zMkfzyABqEY2BURWTXVl;Us(K^;*cqH$v$Gn4OmLv@tAom-I-R}TRP`aUvc!6D&4i_
zKl2aHhh%)UNBv#*60)_mg+mo`!;pB={UrXl8}eB8I9^&>N-Pm9x-*`i{P@7JvZYX}
zWvapZ?vy%?b~bl8?ktY3b>p9H;oOuVy|i@ifw$$tEY*6`2d>Q;B+G<d>p1u}k0fnI
zPMd!`AkIZTv`~o2pEY=xcz!`fp}4k(99`WOYVfbfI~x#&>a&P-Zt$RWcGV#1`uOl`
zX>$alr9(Mt*CHYVgQ03f(Mu<RA4vJ|aAAv>EM~$o%0=su?1Ki2PKGV3=u$yqc}D{e
ztKsHFGU@*b-d+9F6K>vzh3Mhd*8=d4TrkGum+Tr-by=d&lI?jtAH<>a_r<15?t231
z`EVUjXoIR;phlxL0w_H6CN`Kw>kN(?vUzy3psGo(3FiRVkm{yNfoKUyyiwFOBbc-~
z<w=#*t3#~L*+b;~)rZ>i<A+L2(XcQg3Jt)6`}qKamyxa8HADpyr&FDIDQ??!IrRV)
zQ&`fouCF~c4idqV4;oMQ^e(bRKZ2+4OxYGV_Rp#1j&DZ~_9#R+L70TM2gd}wT)rf)
z6`@Mv8ZMhL5*|KMa&c9=FLiQwcs3-axDg;<*S#H*o*zX6<bH><#F7Sj{<O!FxUz3f
zJvTOGSM|KemgB+WUE(RSnS$daz*DEZ1VUjzUp8x0EHgT-Y_>o<dSs8!4g#iLpnPEM
z$Jg@d<jc@>7a6XzpMm=*y72A>p9^E+cK$q7Y)u;uh<WjV!2NuVn))j>AB`HaevbOf
zE~iGN7&Zk;AcbG|`<HU73192OYG#3s78<H_je_@#Vb$V2#s$iL>qhc9wf@-*VuVTe
z(NJ^Gd_5g_7E}^aL@iyjX{%H4a6yPpGLr{Z9PA45keebr1msg^={sk&aeR`RJ57M+
z;ul(J+W>`<B`<}Qwyn{wds|<?sY_Cgjo)kI8vVobCw>_E0DO&Smq;>mY@8PtQZ8&5
zG0lkg9Pzq35skB|?Jrajs8P_!(6DH`sfqAh{qVW!-#)BU)8_(kXT4&V^M0gxeUTHh
z+K>1pNj5tL8(s3$zwq9bjF&QZcqy(XBJfMngk%p{s8;{@pO%*)7~G4zN0>xF(l|=d
zgyvs8g%@D=P+WDQw}t1sYDo8uMe&GebOiE&@t+#U1L;m~I<q$$%zNC%AaM$P0BB5^
z_1y5Y%|Q4nNCD<C@IQKNMJcJaJF6p5Z@iz;Ow7Ofp%7pcf=<Gd>P;@)_rAIr^*ets
zLg`QF2+yu$>GnY;g}rcKvWwgm7oEXSIU_tm%p>HtL+~eOp`TT9R<>R2q4;?|bd}Ci
z&E{aT%FTY|!-`;_i@>L4Q9OA(^R+A}6s>za;>}VBwBZwa|4D9Axva4G#ANHwJUJCY
z3D(frB56CGG{Yggf-{2`^6LBF!BmG?G9<6Y{uM7MY3Sd>P!S%RLL#N(MUDeWxJiOy
zL5T|p#O!`Y4IeKNt-!>-y+E5^r4@w3ZyKa)1SI<=^g_&53B5KB@69jmd9)T^0E*(+
z!;)rjygkvsJyLGfbY>lT*C|0G;s}qJK>SS@lg85id;yMWxeHy>{^(hN(`?yIgZ}wa
zlJm~z2Q<k1d0eU=ql7{L(Eixiz{}2qv<zR%JJ?JKl<w^EFtEL6iZclEKx1J^+8A82
zl`2cNqKHkY+wz#T+TW>#(Hyu|ctoz|iu0!L%X*Ndlo!1r2EHjP6}tXcs&K0;q_Vh*
zbP&HwK4$b>pQadOxb%Jg50o$G#i$U2V2@j$UMSMvJLzKMVutrZ^bQ|M-f!X!lXH@Y
zA)R}_kFBFYhJ{rlnq~;z5uE!dhNLPi(fJS36C6N+EArR-XUAzHS=tc)!{~G3{K2`j
z%>8}<wb50CGAJ6;k|H{#F3pphX!qX^b^SSHS@3{Qcvs<%E>T_-J6!~rxJugH?wkLR
zkH!qq@b5rPJ0>0g$0JuJI{b0rqVgRGARERw+V@n(NR^i_T{;is%K7ejnO@M7gCxqJ
z{x7SY6)xB12mJz#zkCB9>sof_^}Hws^)0O>%wl?55<s^I#xo2d*6^)ruX}s=uY;Um
z6-qETVA$4y`EvdNZqw?E`+*q8uQ&u6{r~X){mHubwAFwZRzLC3oDb<nbY&|tmoq5Q
z7Jp=lLlzX%AZe^Bp2kCf!G?S=85CCTw4g?=X!H-}0J=6?^?1Qr&|erSPgqT=HtfM*
zh=$$?6gza0^ol?`tK^a-x{EFv1f}4P`#klO9d*WE&rgeHMyRT+T|XCRK2}SD&Ni^f
z=Xf@|j0sWI=jx8C8aBd9p+;Dj7KE0yJKS{T#Twe*%FZvBBR2C`v2$jcT;Ap+<j<h?
zl~oUUTH1~4-(_0+w{iMx9Ngz`sbKmNN&ZHtOj&65cTobx&Fu6Q=Wl(cYjiK#L>N{D
zBc9_iYQ-$#O-@&oN#WoC<Hu;DA^l3!K90aucl5vuDKJ)52QrjvqB0QelHVp+mao^5
z!H3WA-kN%pvt18TP*oe54H$fUr`{4WDj6>o)MqS?3|)(UBSO?;W)@@VtD>NyEPS9y
zpAZuc>0ZH9_2LC{e6pIdJ;&j~akY$GO}5T4<1v;L#bJF@F0UmPMW58bxu>9^!rXU+
zaKTFfIO;}kGZ_)arCG+4!*NWmR@g5zY5XIOUVPk<%&wGLspnaPy*SMaR_3OoBbhg#
zU-$w(JQP30gO>pdlKYQ3;nU7do)W-K=J;Y#iIW=`l6^R?*B9_Wp`)P9=zcS3+b1JV
z<96?-+Lfv)VO0IScqrbCU>b$-8Z64%KgaSg_Zx7_**>iS8+W~Q#7LW<Nd@q@4^3sh
zn<gV>DxEjX#Hw#8s$M`9Tv{l;^M}NTu0_;O*-8^pd>6(=@%-5bJFpT$1n_Y_iO}Ll
z!xaGG(1#z0^lm;dCAYZ<%==@WaKxaF4-J)A@9*nGE71Alw);>k$x><@2j!K=4mYUx
zMR7R8qAPxqy`UVOi^D)sH_L?6D#g_jGVfk|&_imhf7@hSA!-!4^u(D2+-Ouym!??t
z+<RpGpb`qr`@E?2a-jv=`HO;8&uJ>CP4;zO+BR<;9`h;Mtsx1FfnfgmlREXZiO-c4
z{Yaz^tgqoS>gQIbSuS=9))qR3$l4;E;c>aO!lUe?zu?&CWbx&ygj3^3KZ`DV4LR+3
zy$zxR;~ARxDV|n5qnNId?EDDYAG6U)qw-^s>AB~VsQOb#-a0o?jxGH#ed;b4#{v<w
z@}ja&oAR}Uom2A=;9wKJ+O<T4_T7Fks!4G*7AY&uN(?nop<eILmT}bZ_TVU*60w-X
z8`s`Y$c8e-of@<K`u3M^Zq|pFs@P`rd0g*3&x;tx?{f&)5Vv{Qm^ha%Gz;KMm0y}J
z@Xyq$WTSoXe%TINfOFq?2h!y~Vy`U4d~s1IZa1TB%_9;TJPql;7%0aVDs9xReXbou
zG=(+)_WtMsk=|?P05*o2_XNWNB+MR>pj7##_&~!r5VK|Vfff|e4-}zpLwd%MGws1q
zWzD3Zs$lgy@5-ZPOqO=4#@hIo{~q<TcYYYtkUyWfCxQHzKTd=Ek=H`pR!+;|yg3Ar
zo5}m{2R-2+K-p{9Tsn6VZX@M~PjvL~X>I4M0Fs1hyGoe^@E<6aM7saZm%3kRsZW~d
znZV1j-1{+iBLQ*KYqwBdujFJCF`aSZ2?bNXDZCEn3&n!o0b;_wx#1t-?m+7gSpbmR
zt+uhIk023?3~$s*FjazkY?XB?;qZyO3`?fWCylzK$<QM^@hAx|Jm9kMqGrQ0_Zi~S
z!R!H8c2iZ!BWvBT8{yROf4>(JoccMq4|Q@#6*_gT%0f?H5~Eujr8bodcA<UJbLM<Z
zm5{F9a^wqZd}*xa6xro`-h|brwTg;%G4j?_xkAV=0s8Ol8`#jIDVN5!?i&{f95}7`
zy7nrA*1N!xl8c^_9Lm=e(b(*Mo$OVsE$Hw2tYaMK(5626XUI3>umh+1RuFjo)&8KO
zI<l{P>v%Zj;lk$PcnYXev}q&Qm{{j7W@<@fv>t)Aw|Fv9=)L8d#iSjp`(B$*k0(!H
zF0w;I4BuXw4E0}efgLNu%C#VvSuh-;feJ;hB+RrE^&6VT{jQ1N+H4c3@aNfx0zNl0
zEtx+&_cwx$k>B0_v3ApsVDjWSF@y`J=Ja1qAqcJQeiX(>MOgLUrTCG}SNV8}84c&!
z@fVRGG#3pX?%Mxp91;qY27-r>xvdhiy~~WQZN1CfIPz_$-HbK-px7xBa3+(2<}dhv
z2I!xoPr@j)_e#x%Vhi@85!S`T+sn<y^o5kH$&Shp$(WYV?%O}CKXQfeqOqt36{v#X
zUpOa25gIyi<F6iLek4c{i#mt`Nam!AqvhW@k=^fA)v&uD_L$1!d~r5soS1U%Lpc^%
zZ_Nnaw9Z}~#g$mAtCZi5P@2^JwdCH5fP0DmN$4(xFqBRqY2as`2$%OCodjYoSdeUE
zZbvh!S{zkS2<m^unsV_cMYMkT{3aS?sVcE%N&<zJ{*?6CdLA2T>jW!`_>}J+x#EdZ
zEq}eIW*vPcYv$a1K|U|bg)4mK3nMGofYKQ#5OPosEV%omU~6k@J(NVz`tWaQqTSc^
z?e#`<VE{6&bXO=5+7)}K(VL#Ut8Ul|ok2hfV?R&MEiS_d%7r|U>laoyx3Pdb7LlFM
zi^9J(gM^xVoBKSQs-0u8mo*|hh~QKvQ0&@`S!HcfaN3B3wY<^+GI(sl&?2VUr<Mu3
zfj9!z(;<mR9D!9o>vGC)=y1}RG`Uz8zH@^@;nOJ&K+efS&|9fmI7d6&j#NAp#TNJ(
z45vjRhC;jX=KF~AuNLR;;5CN|XI$&e?$58`t|hgfx@ljwtNu&Jz@9J;34XlU<@}&>
z8S-Aa)bHuxYNEkb-D$N$NC!)%CISe(j24uP8AkVWiUv?jWPehF%3{a2p*#Arhp!z;
zDqn372aTg2H&-CHY5m(8`gJl(W;oOudS^mVWOVN9GrM~sWxehg`E<WY5}z+a!wRAD
zTMvux3I;>PU>02dA-~<{2y9Z1nvIQ32TQdkbZECPeC6DdyL3J2ZbzE7ke?7FnFg&$
zrAYkJpEaz0H5%jOz559aJvXLp2qS}P&T1rHhNji@uANbjvXa69&m5F`48A+Zs9QBe
zt{o;8prN5{T;1i|SsFK1mzJgYS`i30=7)+$+>jEgDQe@$!5k#HN-v~V|N2&HF&&gF
zO+?TyOh`jxS;-R3!v3*c`e1l}jMM9f1@QP8e_&$#abQq`9}w|mhA>%9P5ioxVWauO
z($=G6qC%|$P%0`JL4Xyn5^6=&WxS#hZop%ZxPeV4?Ya%d@#yp+{|?++;lDu2PVUlH
zs7rE(S=z5A7A;~uASS{-7bx`g<;Q5KfFx+BB~zgJgd{_+|FvEo@3sZLS5z!-Tk)w{
zZO3ogRiml5mRROI@E-nsL6;=Ua95E=e0N{<z1xz<lOTgmSOm<$qUq$7%P->H^Zj5K
z-5y+N;d^=}jGmarwrgay{Dn|8in5~Jo~~ynDsiaJFlaci^*gem_oax_^j%d@_fwQb
zyAHyRkYeI}n`e*aCz`_YJ7x4*#G3LHv0(aih8+p=LpOMUvF_$+o))!d`+Mfrn84aV
zp-gmvUM1Jt>(dw~EjDDSm*TDH;`wmTiV<JAD_9U4SD9e@W_Y2A7|hYbeOe7~ZG1VK
zifm74tnL~u-VSb45S0tN<KAN4=RJAq8WzNFogm|XC+m?!V}cN|o5Va9*Ic_F_~jEx
zBPFSyk)SS#z=J=N1kLv;bXu>I)9JAPc}w)q#FA#L5g<t9piM5nEJP7C4?kOvY_He6
zH8f$nC<_k3jm1?YAaVDksE{s%Et5#nf%Pax)uM%E;SgaRpciytVCMrCF*5w=51#RL
zj*uHPFg8_qJLmgnU~PHepuWq3DlV?sdNMJYjwQRB2NP;iYrz@H7d=<N?J7{DwPdwf
z+)yyNZ#YB?MNv0oRv~ZJ@+3^BZDtE$d(V=jXL*q>4ni_?27Ucp+6c1q`}gzoenv(H
zHhf|4TF}1*TS#d8NtZ}E)<SvE-M_+;5_qi&e6hfPvi|-;0T+8=-S^uZdy}QF8n*3N
z!Ym;1|2V?owU&d|zNdUAsG`Hc)iSF^tGr^5EZq~_T)G-;l;}cj3-9(6h4K<bx1o;i
z5g-3-L+lU-{ZYm3d)4U*A!m7Ikw+iYfqT`x8KJa=t>`Mq+pR=_)14gGB2Ye4R8!m8
zjOUS%pn{y<9@K7un=P!|JUo9xo~}ZOxvb#TK8nU>WyN2?|5t46ybNhxa2k7e<&brh
zVbF1<(D;)$L8|zgtW9S4vTn(-&Sq}h?E`Dd^czn3WgjAXtw+rFc$3S`o{P`bHRipg
zzb!1xakSxK)W4M<)j9}3F8?oQUiCNXLicbqRd5VP*EQ_bim!RO`0p2EA9#?SPM#jR
z<=1wu|88AYrf_uMbGx){6GFV{4z#qj|4yk(Y}5}?w(s3_o&OZ^1G|zY6Nb*_E+$P*
z!e#j!OmPxh0j{Ng{h1{`6+{AZFU07QILS-~u`Wp#S^tnx-QuD_y&=gxERC_b@Xqb=
ztD4}bqlF4Th+xeb0&#i<xgNs0@!kFcsWH!M&wqytS>izf#ja^!pzX^SZV1NseT5DS
z7|SYbZM}YXA_A_L{K&p^40(|=Hu!B%*t{nm)aJsCoPpHrMz2WkF7=VF0eSYkn+=jO
zmUFn4V_%B#aTIe{?O#3UDK#;&<mpYp^xwJpW&f}6?#U~I0F=nf=4k7ALWm?m-p=%;
z9R?j*2YKeIzbn3L8#DUm7&FlwCD+$}14HJIA$BLLorPKzZkq!r=p@|mkoBuOO#_3N
zgoIb-y80D58WlP;-^ox#M$z0ZuDKhXc}O3kF>FBseZcTcsM!%`)*==v4vFqqk1(Wg
z?MK+se)@)PxL;@Qn?z7=;UJI?;oZ)&qg%RJZe$?>X-9l&0|O0WEmU8gj46|ms6y_!
zW~+^(T~@wI71~|LK3vv@V9dyvaurX+3{i!U*LMCRot&IZNlpFF$04ChE?XxbM!&@r
z#bo%Cfhl`Zfwdd*Mk9WTeiRe}FrUD8ms!m4#JRcXH8d^J0UsF`poi-r6qxKRH~zQy
znUHRN*zYYiT2X8jIh?U6jwV?HIDb>dlDZHz_8fhB5K3V$WU+t?NGRIvVZF)b{de<6
z*fx`-yW@5&mlm>zyQ`tw%L5z=im{hd8E4lf26cBxvL7bh?t;L8-!B?J+^$?qR=pEL
zbcJn0dT+`Yviv|it*`S^!vvn0xiVf1A&KwR$=9Ml#j745z9u#VCdJ*4hvbHAqWu-a
z6fz)20jGKF&sYRnsFAQIU3cuOL06`;nl+>E-{A?md+-T2=R;Q`kiI0z4D<2KrCS`;
z<l~&Src4G24ldte6PdyyqcGMO@!%uUWwvZkl>MZAEX*+g`#y4dQN(>Ms;A|*n3z*m
zw=&`Q4&x3K<6$6L+Bam#V<P_%80q}ccI(l{HjRpFk;7NR+rZAMk%1Z>Zgjtu`4H9G
z2_{AkIka$aYnt5`ALWn{X;5K*vpnAZS{(mP4WN@q5>oJGI{LSsQ%k7KKOJdo_>P;M
z93U!&hleHQ<q?aDicZTU9`?gO{rbI`OG(zXct*vPVYKXV<*9R@T1STrJ4-&+J15bJ
z;Mt3l(DuZo5^!k9@`&7~2C9&Id>+Vv6Y740Dr82V@SvP7W#vH6<GtFWRVN<q7lNKl
z^I($v;9KU$tB@xm&~-wDI4yiZcox2*lvt|)2h|#4I<kv8Uc|O%0#<#E3OHm6tY!2a
z@b}_X;;pzdW&h|tdE2UdxS_>&XdZV#iN%-Zj44>c$O!8W_VM>O{`E#f9Z<!nk(12Z
z!EZ?7b+2pP`|$gNzwy3CSO@QpeY)S3$l`Z4dN}y%eGsKrZL|R{QI)`+;)Y9Ocaxi-
z=8s2$dOHa!g87jxvVFqRn|_e#X&46}#^&8_4KCEmux!;5z|(A0ect13Bmosyll0|s
z8R<+X@Yc`c5n^o}cAvP)PxsCbT~(tG8JiZI<fAbu*!@67i^76GN9K|fr*zhSut-dT
zx?0y2okAoMlu~N4X0*ci+6jAi>uB4SI1|3!!R#!NKxIFR4K)|t4Hts9==AN0YuH_z
zv?}hW4hZ+jdC%3zD?5qIa!Kn$d`?y=p7-_l@$N*x??K|}M&0XVRZIs9kIUois;Jcu
ziDB)3qeffU9&)4K)vO<C%&AjHIo<xS#i%-t0eyxuu4|I8_wtcc@?52M^_wv46^OFO
zoAHpr-y>=DE-S+5<OXxiufZKg!N7kn^{NbTK1ZVjh80l7|6>c+VHpBA2utFeiSskI
zC^Q$ZjY*^fA-fztq2o?!&-%K<P(iW8eaVq#ZXZ;xS~3i&!6lSJO^O6wtuI_OmEuuY
zR7q)!>YLzJBN6CAK_DRI;`#XYOsS;g)L*8&ZDyYJ^!vyBUtce=(Mg5g1A~!*4nY0C
zQ&OFe^AY8ZKA^YYRNffytQRz!DaIq~;X;nge;#-tSR10`OMksjfA>y0Q;gm6a-^Dx
zRU(1RB@i(HE~Mohve5W}yVcBUX&vD%O;->VXtevMFwd_U(3QOSvQ05`Iz(kd5g}V}
zzyiUd5QegwewAd&*!)!}li20l;<8e+TiExWWBptBiiC7!;vgt}|JY8XmxlD%R1RjF
zO7uHd$a>7t5c}V)n5a=umy}{7z`#V~WtTI}YAV8it%0orx9Xl}V5OyDfk1!(!_i>y
z)DH|Lg<Ej?iD41EZ_eD_GFbI!bOeAN^&g<efByWk@2jbV3_9XO$mp2juoEZ}(ZEYy
zF1)pyx^^o1d441Dkt5XWJ%>UZLmC)bnKOhC;^R|4cHccAfC~y-yaJ%tv3ZS{HFndw
z{d2cq;!EC@!thCKQ4B^|OW}nEdwKv~)JgzH!r!;eDqzivo=&V^F{uWG2VeYs`<UE<
zMqR|@F~@T76}2ZJA>sCkorR^ZKq`VdQ{X-&OcJux?Cf~jociO))2jEF3Y#J$Z0R_!
zf91(-@(Tj!QvxcA15~6uGT0jZF=|j!T?2U|wV@ole(J<ByH`^a<&_l7;r2GNUx!9c
z0%vJyUp@l0_(wy_tqSjg`yM5FuaPMJS$I&b$kz1kz^+%?Ry?6Zs7^iNrOh0}aBRcS
zRFdB@WM_DtDfv7}4Vavs9$a+^a&~3|zc0`>hqmO3gyEIx6<gma?TPtOgPe2v{ASsp
z(^cpPf&%tjGvg<*bJX9;;yEq2(vaFIm<nZt3YEc^G7G&jg7^Jz4dUYpUMYqy+DUV+
z*Roz~T_|>juvjgX@-V}WJ4-<O*J=k_eV>qNnAZ31`2@S{;B@3E=-mp+tzkZiis8E3
zPmE<BbXd2Q{rz?*=QjJ*GG@Z+_XP~`(|qk%N9RXs{wmgfDD$CwcT^RmW12`(RTaZa
z*tDE7>Qy+C#r}q_n_#8el9*@V^8`38Pyt|i#%9h+@bfrL#X0rz^>U6EPHYNi-K8IU
z!qGx=$imePQOwCa4Bc)gY@|G#tv6q8hOVG#2N7QzQvF5hQ}4X*Q;c4a2wz}xoJ<)`
zS{|zDTV6HxF}aPw&SA<qD&|fR8iRHsg)X7+S~HUbM)9v`q!C{Y8*~Y0*LNjq<?<^`
zxsE4#=i4`#_=B79gwqi7PO1ZQ@P-`!{y7iqZ-+b+3(U0X1rS%lI-kw;3Br4+F+m_n
zMO=&ZyLFONk`W%?QpU#SaRo0vG;Y7j9#}Q~zFs#6`urFR&UYpgZIh7)5)oz1h}8k;
zLs@rJpK4X5`(oHmh@-@L^A@&{L@5PXS-Ji+`D@tyY3A^DcmMV6TknEu;slS2?6GnF
z#&tM3z0X*b$x4R^s!Ke^pq{rac6<xDq(?30-UFDLs`M?4a3_w!Jc=5(Wg8gDi`*06
z>EmirWe#~VWYY0i=Jq8AGcWSv#K<5hz4KlAQ+~!OlEDAYZ|=8vy<~Rq6-_d&!sWyi
zUyc|Z(T=zK*poFw8J(CDk_@g4T#BP-@)J@>2R|~XWc_2Z+M3t>OMjMYt|{SdMZ{nI
zglkldqr~i7&$WxDqf2PnlVV&rLCo&g{m^B;>w?vNL|NXL+Ig#{NRKOr6P$ywXy7Ae
zEyhoVJ~ATM9wf4;nr)Ez3L7+}Wh)>MJ05$^+3t&48;(Z^b&w)fu3-~Ja2c6$Cn2h7
zKaUR=GsCprVxb1#Q)BgLIB6Cxddf)1N*5Jh5aexS;NNk9crFh^##~{|_^Ly+kSKw&
zgU8I9`}1)a2)GOHcigs18{AF=0TdaJ^-_v@{TqNf*fwcUR*%KrMUw!wk?cwCN182i
zyqegPD0e9pV-ZGCz;cEJ&$;ZkM4IOuM(w(x&}`7HUs;Z2yB2<bEpn%ypAvc@1?T>D
zgnKw$x$(~ukwY2|E(Es%zb4x#{a73|$b&rXZMDLM{Av-#aL}&jfVWfP6l-%|3VD3O
zM<0U~VREw&+}_a8MzygJGJ`@rNrucmv#n4XJuDg~NtU$DdLpgAXAyFKSJnN)63z-s
zn4fIM{~6P;CYlExHE9Gf)yxPx*4d4}Eh+u??H5-X=(*s?V#9-0sE7Ybv!wrmJ($!j
zG9EV{LL^0Zk5M0no^|i4RGiY$dpL=9P-c+L(Yc&zplAIEHnGcbx87(HsXK}1sNbC$
zFH~0v6=frhvx8I1W0w1MAagS0V)IV5aLVG){Nv#4(-pzLpnD!@LFib>Exx;RalKNZ
zw0|(4OQ1h2di_{8(~OHGLmXyikM2Pj*bKNg2KM8czSYzcxDi?AyK$jJb{tIIzR>XS
zUbe$vcgL~Oq8IiajE^v}vqz2ZI-dFhmPVWD=M~8rLrfciQm!0JD(wgVZ+2ualMN*i
zq`l5<=k=~vs0rjEa4P``FdDvL))V(nQYua?0arQU;XGhea|pNOldp7-g{*Rm<uAq(
z%?$;`tOnf->y>xo1<GLF0_pfUt}+SPhfe+#Y!2g0=L#pDolR+c?T9(Ziv1^$5X#J!
z>MaWLE)DMv$d0z{Mul)#)sNQ~A14u*oo{Xz<{4V?tBb$X*#*<=+`9|tIguvwAJueX
zWUd@5)HcU!8XKb(|NSGOHjLVHwH8A4oFjtkGD*o2kHkDSKYBj!*wzJJxq$j5Pn9iv
zJ_m${2g!PLELn=d@sS6C{tTR?vA{!U1kx-uKR-c4)M^qKqJ@DJF-g*<ZeEO-f+?zq
zzI!-qC<OO5AyTJC5_E=`N~F-cpRhPzlyE+(Ak)nCom^S~hux^I>Pd;V?MeBez~iHd
z?dGxVP=ZB@tm>r`(aoAd#OVq#^?pgkoq@}>fHA&0XA0RUhG`tyk2KQv<=4W7S1g@O
zQHN<;h2?ym@dJtKZRqnBd{?}Qx~@0TI!^o0hDS&HI{eRpKIC|U>Fa^o^ORrH(SY75
zCe?6nml!Lj`yZ-bHT@8XB=vf<NlmD%*4*b()%M`>r1*$~Y4%2{!2d(mS4Tx1ecui}
z^w23?0us^<0|?R}-60avNH+{1k}3!a(kMtHh;)Y{DIJnZDlJ3KyW{u!-uu1vT+2Un
zG0e>U+<Wdh`|Q1sU@-AEDWvqzB<@1Q1yKeLH|PLzd;^sM@#Rc@@M8h;dr7^e0{{>&
z!+3mXws*OvY54PJ!S`edmS#UL4g;4ZPMA`u!uS5KUDYO?`sT^<x<8&x-dF)nr=H>%
zvx*Q5VaVk;eX2;Qy0I^lIK}*{d!NvDuP0{Q#hHO|DU?9%Xh^u`J+)P^igLA62o^2W
z4#U*a(sKEaYSDP7M<W*GPs_eMIxwR1ROM4Z%#$lKe3m{*qOSfevJahh#|-awL@)2h
z>M$q&ksDyMu{gu(Hh98_`x0Lc_it51M1mo%rkU_$@S8Oyb;;WFw>=jndCJ+(6YhJq
z61;xVp^0$BBT;k|>fC%{|6K+}`kPx99$)rSzgV6SzgQ~~FhG_gFq;vy6%prHAlT5g
zF*RX~o0X;Aen{-`qb{uj#s90L7PFa$xbQ`i>E##W92w&jIYJ){$a0uo6obcbCtt))
zmnb&`v4^rZFfh2*un7*9_TQVu$oIJC1x=+RCl@YscrpIzT5)f-8D<l4!3g@{z!@Xu
z?Q9B8B8oJF`>asgA9LPs7xA_50grS&gFrTw^?NMt#!Z$2KbhWlwQaZ6p-p$?8woXO
z!k#8SZtppGo$An=6w<_h1Ia%c6t2&fWGedA7Cqe)zq-{OYbv}eW#~SeNFrRT^$|zb
z8fj&%bvepMdvO#~lVNPXUF7Mo&|=eLX>0d}7(!C5knp+QS@!&kT;R&6L|3|VXG$LF
z%8%E#O&eV9-XP&lZeMTrG#hbN6vcVqcukh-o7O$V5B|HCKqvBOWzH$DFB*ri#%VP7
z;?X#wo<ly5;ny1jl>rXrk_l^}J8Hn*fc+yM<6n9ZF{0%@5<~JwBfWWkoZMODR9p2X
zx(9z`@$if(+FjuWum3ZLB)`#^YM=dzJE_v+B>R&&GI$l}$eDILmU8vqSDDE2-jZ>J
z?y6@x-VuQRogIfnGMz2O?9~=|KUrnVx%Rl@pYdk=@YVP?frzZpbj7j<%I^&GPY31I
znJ2KgGk>yq9^R`6yQWPi#TgF@TzHgs<83`xrcv*jXSy6QwvV5$okyhAYDAEs5f}2P
zzAMhpk%!ky=*t<n86XNbNCzm~%6AaN7B+ElYLLzL)Vq8?m~}{v-LZqNTC=PSOb;)w
zZxs*=(XOwYg$RdEUJnlL;Woj|BD%03g$@=<Qw~CI7d+~ahfaOUY|mLt$8QBKOn2@2
zC{{|?kKIbm|5S)=r0N9w&Wn{`_Iob;1tM+eC`8WmVku<oU5Ud{_AyIF`~aO#W8br^
zz2+qxeVDDf27L7L503x9%3_^+spJ!StNqtcr`bc4C%zSNmZ$ujIb))X#(YR1HLuH%
zGO6QFd6Sch<5}5VkVBD$+z{dr`HaJu@et8ePpwl95;*!kHH~p64I0Rgi^ki0-mI+m
zeYZ&`wBScmXzcXwg6Db{9<z$8>r!Qo%zi5c5Is8a$jh?Y{hUm=#Uqs+@6y-`N$!ut
z3Nf)Cdbe-;(Wt;_KjE}*`J5@-)xj-{kA@Yfh`5x6R^odsaWbl^9<*AP!9nkFZDQ_k
zG8@8@4)6nw!%(Cg$$3xZdjLq{+vmo~DfEs*na0HA;akB2!+Y%H17~WfN^)K+$Bwfn
z{oI4}OO12KW`r8g$D}Sh=63uIfhrF45bQR_azXY94G;C@q;$pk6S5iiXjJ0hRW1_w
zh52BApk2*Ti8xqyJGzHK$3Db{BJQuZ!}XiH$WTn_fMk!qZXGmr;zGe~Olm2NN{=2r
z;*n=G8EK3Py4pr5o-X6IBCdX%-}&*gqZO}XIg!CLtoCMKyyBAar=ydQf476=<B6$i
zOzVpHyvk-wi}p)crSuckGw&y%(P>^E&o_(>E6ie_kKSNN_uqL?^$G^<Gz0u*pB{hS
z^SV$#xV)eiYUukZR3Z_v@aAjLx!%s?)30u`*NX2K6H@g3s*gji4<uWKdUvwrm?1gy
zMo)Wob=l<Eal_vG_kAT5FM6_%p+a{eS-+eVpXZi3koN~tc5>mt>-!z|(aF~MY`x0^
z6>6ZvNoYQFmq%V(_x8P)K(_eb!YwSVgE0KPqn7==k-f!Fwz$s5v)m|eSj}(T#W0Ug
zarNewy2}iX`vr;TAKfUJ&)ex=0*Li*E?QY<<N?HQs}Y6*9r;KsJ)0|hB@FU1#Qj4&
z2>x5a@WbC%uK6&^e7)QGdbN)~PJFWAn*TdgfjUNT#}#^c3olNYFJ3E(_iMZM@h^DB
zRzPwjz1e<!dmYmgPg)eP|AXcmb}m3=N|~O)Jih$YaBkpA&mD@H*^SmH=?l-Jv$y0=
z`ck<nM#FyF9J5Rdw_|{~v5mSP$-y&ap8i&v08PF<Yyx7Cl+*WcXcGajB5))`5yy5t
zJ@n0UnvSn3tM$NOCx91il9+9_WXX?9ZHchoF4oRhW(By=OI>+<2y)j372LIapMxAb
zrVF%oSuJ<RLonM>JC1I0A+Yi7->lEr=09o2Esj%N8Z*X}Ph)0`L@s(NL_*1W{)7>L
z&qCCGEX*VJq&G6nYu$UcW^dsPwh+OBt3}@<t)qU`{<tfs%9W2&nwXNFD#p;c9!Y_X
z@z<66i9Y>c#+Bt~GHkN4=INS+tAn7JrwI?d53yMvmiR2OTqbtt9jckXc;U<rniOk$
zxnk<thDGk#&Hng@uif=5%m|tcaAGxx!`x(vY7-)P;FPKWjKf!682LT$Vic8>{^bGt
zJ0#Xvz9Jb8Mj;Ozie0;#r(l1achghRvA@sUC}0AC=ovZpJB<o-f}Qp8sXvB3<t|q`
z_Rd|;Inxxi{e@`a$3wK-<=;6TmoTNizb((tx+oLzF<_%Dwc9#3b-n5F%BLAj$jGgQ
z)XyhwJPm#W>=#;Xc-J1Sqy!CdiB2(ywDOOuo#yER7SXl7xoV@~a;5|LbRbNW)Id|7
zeZ9D2f`&d|`QqCJceh_(e|w?M0lmQTvT%c8DuL>U_VjY+5kwObR!@CFyYy2*kAeU<
z`qT{A%us%9Hqb14(PeAECLK4tv%Md?N9`p<+q>Y|k&L#)qN1Xb3_9YdtE)TMDX(S|
z5uqTJJJI7|io<Wl(Gu7UT|HZ{Xp5f_?Aw0JfC1@8Jj{{thb#>&B{8jUKN1AWbHGzE
zzZSb{XrMivTu-}=obza90)*Hg%~z|7BUYy$BUmv&IhJ;+)jyU!p!n`#Kogu2I|u0q
zSS&e_jnh$)19&AmF};89648%S-HNYqf60lBle4HFNg+d^XY}x1Ck;Z_IyKztgK)4{
zNd2xe!S=ehFCntB@kcirtOQ8jXyC?kwH=$~5>1-cvmYULRzT2|l}Hr<84K!tSnR)-
z@S+vV%lw0|``?A9I`=zf<5>`hgvimvlI0Fkt6l!2zytYdZ@BC7rcEL5KK6@dzgT%;
zj6xP5>MAT;BZ}5Ms{O4+7pxFWQGe(bJp23Hi}h&ckBU~dXm9=fSQW0A0XOBJhj9MI
zcKSAMzy85(to&&)6hStOvgXfschj4jdmQyCs;&bZJd*^Vk8A2zNT2;m(rZ7)UC2&@
zcf&_s-~*xpas^iZ;BbK10feT&tHXg^o8`8UfQ_F&|J|xzOP(p@dz>psO9AA%Z1`3p
zYyz9{m21V-pD>OTH1iGbV+w^&Y?Fl$a?UAjqTK0v7jz5nU$Uy0Oxo|}b;PlXVJz+E
zE$&%#f|7TD#=Ub;f>~`o--Fm%&{VvuX&^d-+3jBFBmN7mKNA&mkbKz=`rVt4#NU_O
z=;XMcM8haod@olk{FH6#$g?93SCX49$9aw7bHipw9>@zm#Ie19AAi2y#kXr0M=5f7
zpLD7R){Ry_44Gq>j0BFakwV_?vU5al<jzIaWef=)y?#B$UH3X>A>Hzk$Oi8ePF}VO
zTh!hDr%8Y34qDn~&+_`Vg3l0-mj8aD&QK~aVs8o<z^yjf5*-`pxAm$XIS9crJ=+%N
z0`F^$b0Tbhk~4;{XeC46V2#sY9XmpIPm6AeSn$e@M*e&STOR4A_Yp+2v`S_Fok{$r
zrY4%!@UCc_t+|hzsy|}V4mhBg;pV3p-(uAS&92DcYy-{i9r>=`akT>dJXEM7AS@<j
zu!T>={#VB~H5uuOpViMuGL#Y{hlV!36&bvb6_BS7QLuz|d=DG97>qN$b;Tr3s6!t^
z07P+kGh&N}d;B)msO7^1SETIK6IJ{<RA12Xuhl3#CPlDlfPV*EA<#{$(9<as(@$<@
z0Lp^B!^BVsx4e-DjPw4LFAqT7jSyvJrR_cHV^;oPXy2_VC^{b_kdsgdEzyY=Xk3a8
zvk8s8g+hXRMuxRYIb>@M_+LpeS~f5^0VL<w;18|4Z|=Sr^OzT6ytX5573Vf5-iqFu
zzOA<-c0fQ~v0xDy#+O<+Fq@KzvenMFLTMHIYjuxCSSRk-i-0={aeF+{+z%3`c(>i$
zmta907!WMn&DsD<x`c@vxwvSF2C*aq4_FvU@PbcwV!`*bTm0+|C?#B5)It~#Kp;NX
zTvj&NjM`H07X^S~8rO!O^uOr8R4#$5liew;^Rbqay1;gwEB*H}sgn^^^bI6k7<a2M
z$$6@pqIppaV}5q*$D=j_P-}HHTS>MDV!7&2DnqeZOf(7-XOwQgEV|4hp}jm{dE{<u
z+2B!eAw;u4^Cy02C(gV&`U?-6xG`b*Tzr|t`}I}sg&xWhqA%-K`*&6V^XB<80>w3N
znS=F-v$W{|#tWkL^($P~Vi(h)QM+MX=EA9ZhVjQN*;Il|$R&?Jckvu?=6eHsL@JJ!
zlI&R-`3jJb`<yBZBq|}3pbMwwswRG%5JkxS-f+XD2ppo|WB<}<{jKkUBvv84h!f!^
z7c6D<VjpRK)0&fgXSoxu6SuAoBB%?#IFohihs9yq<8C+S!XncFJB@<zm7D@btPmoI
z*Y=?o{~)E0isL)QYMsi*n8UOCC@b=hYiU_j@gqkfi^CB4eSkJ^ZEf-0*b8M3&m8L^
zgzOp&Z4W<?3O%~e$|5)Tw};5Ukqll6I50M{UwbU5^v-tXAj_ncGJ-?2svHu!-*yJ4
zjDg*_-Ihxd$)>iyVjqPw-x>h?>f!|KP<j4r6sfa$iT&Q5Q|TD%oo+v>NK)AY#Zn*9
z*XYi~`DWe#Xo0RCpN~~qNL+2}MsyE8XE#%0XIubm7gZS(+liNgij^G~AE^*&2P^e&
zYS&si*zJ0GP8dXzK)>RRw)%d3o|G?cb?XtqpXOZw-~fpUF|M*q0M2@GM|M?T({De3
zL9uJXuY>oNT4<J90tUeU=viBj@KdZ>jFjC0&P~Mc<i#?9T#s$`RkW4q&8dxW=r0G*
zlB4DZIm|ol?6HQ&CTLT=MCv=yJ*yB1qii4y^kzX)_3MM7iOKq8*@G+*ClYku>y!J!
z8kz^N>eJ-4F-XSs8oE5qXcm7|v>^PmObR%P0Zs<|jrvHHW9gt-;$ii-Sk~9qoS5W6
z(nFcPAsE=8yq5&Jp{r5U73p-96=BKI43k9Bz2xqcTQcxdArnqeJOL@sfQd~Q&in5P
zfccRse*wkE$KP7IoM@D4vles{w*Xd(=f-<hn7-rQT0uuRgH#3E-&LTii7ZglPe_N)
znwr9^Kc)0`N*iWM7^NF9Mnr(`qou*`+CFuk+U@fIPXEp6Zja3XN)G@Kfi!acUA9N_
z$|-o0(yA(wsUAxzG-kUyZHKzK7NR8rj-Kcg@0n4#F{`xn#f$y3|0{1Q5p!^zSOk}f
zphZOFkU`Y_RCz1_`-9F(F+Orp;Oc5lW~{i__T*)%B5&fiFH8jnAxNUTCqPUmust~+
zij8%ZWPI7?s9>lhuzm*DA8fxUDlF79Humq@J=<SS1zkpP(Lp9F#Wnsbv;wH@qmi5h
z1Oj{G*Dw2-Dyz?j{S1H~H<|cI&hyV=Q)y*oWrS47UpMt*ig1A9sO~CYsbH+XQ(&qs
z%7waYyYdayAioEsNe1)dE=j)j%c)tpT^4uy?eg5fUjn?smvdU2P2@nSko;=de~4An
zA6RUz_0ib$>MC4ggIr4qR^zv8EPJ--`x$j}33`%{$p24?<v8@J%f5Cp;LFWQ3~7!3
zZ}ZUepP69)jHoV~lo%+L7{rQ-tfm4c=l2g0%6VyVb`ne7|4x&e^e6lb1|Ol|zsv>6
z#J@%3)TzCi3-0cK#y*EW%!TSq8e{~am%kf@eSc{cmz1#55*0ppuLH%6`3(B^hhO=s
z5n!<(hlhtxQ&T(k7Md&tAdx2l<<fvo6U~AUrN9@Aiey$0h+wWYvl4cm-~g0dR&9Vx
zRm>f*FLG`Q+5H$wE8lY4&sYo7sDM#^{xh@Xdzo)|Xq5xnDu%on$c>k}XkSZ4`A}Lw
zutZaIK%uXE3NvwE>5O`slG2YwDjNynyi(CPx;9W!Qo`WUeRn7F)z?j5hnp82tR=g8
zCk)EhEcgj3AR8PCzu-1J#HT&${%A=<D0?)&r$>9<X1n^KPBE|3Q{LC*4*3IQcsJS6
z@OFK@sE2@dZ(pC?)CZ$Lu%N=OlRU`l@F#y3={l~C2%7`;N2HnTI)8s!f6p1A9Ndwx
zCFh3jQl)XN)S1;@WuL(mp3q*TzUK~_>9#f|@FVD;NZaEy5-K@bct2%no>*+yp85im
zSFJx^S&UWND6}Hi5{a!*n+hnpmVo_!@cb!1#5ePXU7U37OqIWu6<<y4@#yb%+-HA0
zJG>>FPxJ4+Y8spcHC|mM*@-a6+!o6K+bhGyJ8Ktn%%X|n?Ytu((cJ$~XIVeI<{-Rh
z0RN=a7Igebt9j7uLjnc2{qe8if?!&BRYIRZ9L~^9%J_FDgF`iSgn5S(=U0>J&d-_F
z#cTW0K^o=bO`4#AE%3$IUfFWKWTC0nSbt4fgLheuOZgM#mX{>GXr#Pufu0__W4EEl
z7uKA)RwxD2TVgB9B>-+tJdKW|8d8W~IkoRD_btnGf(B}o+i7V7>fZ}j&Al9&NbCBw
zRtV7?fh{;+!Vx!Dgx07YQiXi5*ZzIJul><WJhOGQFP+-e`<PkD0EKydQ}Bk_u(p==
z_=Ex|8K16@dWAin-~X#zxqg=PwucEZ-n-wy4C|)q1c8kT-K8GMk&4kTx`zSr@$osh
zK3f)%L8+O4%6ZwuBbT0dTX(pmsU4xhV&R8le0`wHp__$}?*SAg3Skx3m(y(E9OOv*
z@lp#~Cp9jIGZ0eqN8T$^0*&Pwbi<ulUKX%A2dT`$0*L{A-)`rDFDp>Df^-Qn`-{Rt
zuhxSYNJ>SY99~%)L)EI@(3h@(tMxNS^)P-$)bdnEz5?`h4(gLKt2>rR(9y%4YPBqj
z_Dvc|-ckD_iqY@>Fjg|>;a@mj)ReZ}t(h$E%d}aY`0lJ<;MTO&6TCCN8sYhrz5vVX
zJ^L#N>&El~*2OSv3_z{%Jh*r(69DIWSpZ081;IRU0S!u3{<#Y|Olt`)`3oYka88bH
zY%C0Z8>A<X7FSZ9NOt=`wz^#(S^fTGoL`dWk&#FI?Tlr*&R+(6FU${ZAaQ%-gWq0@
zcmo<1p|y>HK{SBh;sD;(!~^&)LYog1+4^Sjm`ZGj5Gu5dOL2LqV``u?u<eaG8$S!A
zI7&C{%6f%vP*3H4EU>to2DikEccCKW7Ka^oZvn~)U;`JJR1;p7pYF|30A?`gjlg^`
zl%Q+xjhGelvZCxHetZLwi=+3`+l5qL1woY^?Z$kk4yKBA6<N;`oQq;put^inm?D2}
zwr0sz0qI^QRUM{Yi9tbOA*Pv`8L_pEu`zCw>;}aq6a#{dDS!2OFRb^|c8U@OH=qDP
z@oUi_{G(0(r0h#s!+YUs5vt_+l&l%CZZSd|Ibs5|>!;|8EhvGs%q8>;f(`$SBkQp!
z`R%O?4e&-gt^R0U8Ms}NDFo1FJvk!Q(@=SdcAF~!av4$}^-crsrkKVRNb<ixZtDF*
z)9^~{W&Sd0LeBA!FWYB%Oj|T_xC+M|qXPt2T5&hpyLa!70n#v}GYShYEi3COntB8F
z85Gj_yKZ%$WS_MN#c{%<aoqt;;0UkA8)`?%JTY?@I$FIn8*tHQ{7I$f+j`A#g2(dU
zphY@}X%uHC2q32x4X&}u@g%FD^6u-3!7CVd7INq(TJ)Q-($ixWCx`o65K{}F@ap21
zqJfLpzg1xctVX_$5atAjfujAvu8Q?~$@hFDir=3tQwyckxzZXS-7(mqa8pV^RlkJn
zJds8--?q2656(j&kr{j@m=KWjP=AeQyOZ8$8E0c0@w8Ut2I3&w(Kre069msr{L?GV
zqn?k-;)Jnv_&T${eqT>-brIkQ;=@ib<S2x((|B7^D>RTl7{tqOqoV`b5=#h>M*!4T
zk+Kih2mlF0Q_9b&U5}e$cm}WOD44T}D-(WQkdxn^_OK$lOppaNlmI0?{nFsGmh^pU
zp&wtDwk8B#f*t<hb_3h(3=~tOB2+L+_V}BK^HQ@IFCSmQl&d#Lx7_CI@zD{uKbZmC
zy~VNfdNqWx3$JXCzDWU#334TbEliUp6w=X`gfA$k+sEt>UXJyKHr!;^b#xAvIpRr3
z>wbQZBUXu25jYk`E;M@nW4qDN;FTE?yw-|iRgPPE8}OqGkSLSEd5zyGP$&E-;7Rc{
z)ng$2{V#Zd0T4_>@yV*cN0lsJQ3;1#*sV++R;;I?*3OBt+a^=8($UhLvclf@=P?U)
zDc-7Z;2%eXYbP{$>swM{qB1b5gRro(V{z-`K&N`rs2@!gOqI$u(L8wcpGewk;7{EV
z`nN+pSMv2U1s!CLz@TVowKWbFCM9op2w`Ss;5hGpR>uU=@?;xrK2Qh}HMHUS4Z1y4
zX1Innz{{5sWi_#%t5WrvO1A?cONsX<YF>eL<Exu&ddyB`6YcRDC?^U*L_|3X^2NL}
zk_9J|;~Cp7G}8#keBcIwj1EvU3?#@XfHO8G<VWbSw$|&rIAEWuy1DTKZ8!6Wc-A|S
zJkOAvXBJQl1uoem(U}D@L>eCu4jNZpQ@6iyMFXD?C)twt&Z+PUbXFU2k`~-z#N+m-
z#FSvHFJlMxD}d@?eE`*rSbnF?03eMnH*t1p={Mm~FZsBhVNzFJlBP>j+*=lT3(5fX
zr|44SdNpio;=8avwOsA`^Ok!0wgF;5b$qeMnE|fx5G77VW1fie`w|ll_FJF0ntXu2
zzCz^{P15Z-+TfmKbYc$gbt&t5wttg?(03T~Jkgu>9_0LY`Ds1`zb7<9#!lmBBYyq&
zwdmV#mX8xUXyotTN3+ATI`;OTdd?7HagRlL!P)v=UA^V8YNuq+4JVJ=csT8?&K~U{
zm@XYYT0Qk@R>%h~h`yXbQqe2qf`J9hdnJMU;KBNfjsl}pGSH)uMCS@HvqlGeH?>sT
zD?Z<X$-^-J@)E-M-$V1h;=S*WTRxWZjVjrfA2?$YgV)pgHn|~PwnNbR(6+b}Kv0E|
zV*)pjK0#)ng6Qu!VJp7d+&3zdIz`|p+Obd$S!oC=5Ibyvn^~_Zcl@aY8Kgb1W_jfH
z$(7&yMz@|WoUi$8zY86^v0DYWUxi&tsQB$CC`stp--@?n4RN8H0h(_}&N2h9!lr8Z
zZRu&@Dq5g4#rM_5e^qggz0MR&A7@IBrdX6#^7QR6D=2=QWEcz4V+Jc~G*~T}^y$_|
zblTc6gcbW{|3yjNxA=|p1F4<PvktBV_3Tk5#ja@^p~J&XVAP71N8v|lt4S34e8)`+
zzcF(z!%~?&OgHWk{mczE=0m~qxLT@?JGKy(6`BIWxKVFBpho{VpVoopbXL)`@mdgG
zGKD^vO*(}<_QMy<D*}9dy+NRmCCXDi*f7aGq6Z=IEoOHJX{Q&Zd<idp$nKGN;YKHy
zR*)+RRRlsM5-2Ba+Cxae%NLJF4GwAgdlg}KIXhmOwNizQH&_8nn#>LFz=H8FAWh%}
zhUjnlKE&|Z-TvB>4&EG$Ymk9=zb&7R{<+*9kF;YSi&%t$b?}2lgNAEfqqT*-@?!E8
z(;cBq^XEDA*1C~f62dM%L;eDYA%leby5Hi`kD2d+Zq|bNX9iBqh^~Y8O!2iweCS`{
zV{D|90RvMWiWoC=RJadYa4daC=#91W(Sm#mKyOKuh#=Od6kc6@@BEjWm5jydjKK74
zzXzFVgD}ullVVentdAs%VVk+;F#2gRUZyR*OKTDfjG=*SpfkFGS=l|Q@9DxH@YF%a
zLPe?bnPZ;P!KLu(lvLW+?%h_3ruyz9_N}o>V7-yafsLd7FjSBwJrgP9p@KYtU~iE_
zT`<w7j6Z_;0x1g6^8^TJ!zx7Fpz@!k4rD3F>h#$l&`&b`)gVYSr1*w~IF8!pE%nz_
z%O?-EF`#1vEcn8pN5q`j9*-co7P7rEMHeMAUyO}Xj1;aYzPx;wKdMJRLz~b#WxORR
zkL4^_xz>)CKX6B}oUr-@Z|ekkRgj#{H=84g;=lCC|Fpj7jgWopqyPLPb<raQEGx!k
zr{2NS)pP2aYI_>##v$8V>}%jm!MnFB`l3tr3r_Q0wxm7_qR-?m+NHw78senS%JnS-
z-^VC$$%Dg0060T^^+qW`O68oC7duDLY0~m8X*oX2Ca}S+$_DECMB^*MqCX*w0ER`!
znJQ%=SZ5s$D*CQ8?z&BmcH<d)PFn@z&IU+`=#9XBu%z#>nEW+i&QVWL%@hftirkQ%
zGR*6lXh3zL`F*V7Q6T6I`gtupuDt@?5E(BaTrI|7X%ZJ)CtgyYWlXLWd?4d*?7W*f
z$(8^f=L<K1AF0vqnnK&)ND2QtQcWZliDv~558*^%NBSxy@g@o)KEGu-r?$lFfhHCm
z?pf+|#wGoN1k4+C08jh_`n!C&5Lxi{@Fd|H&B4kLZw+Q9gVURLc#3^%39w%r5`;Sn
z)ubrD0kQYlF13(+3&jpk`)Iy#dZ2KzC-R!97^o>58U`ftkB|ad*FBbl8KDsl<ZuCP
zO6&^Db52YFZmxD|8iP0Am)5ZN1WJpLcP`xjm0(Ml@Ah&8_ktmqXyItR56Z`M^lf+g
zduxQ@JsWx8(c3W-Q?lHKr=0Cidp`ICY>_6Oh$;U+ArNUd3?KqtE^f~{XlcA+@N7G!
z#wDi5t~5>rLnsp4wY|NCxpZ@hz`p@xAd9*>hGJt5Gl;_Zmkucyjhuwf56#e-L;s#w
z2J-x6BTV5TZQP<a1=tK^)k`EzQTrL?AR<3`y!^&5X^3E8;mGdVVsZ0$_0xh!4@d+L
zx8LVkADngB!T{KH&%l5j;6rhMJ-HZxy4K^N3<SkG9S4bSA<x>Twv7@nH@%7x={X{N
z3p3z8n-ov4v`%;)5lV1@-Q2YNVoq}?fAtF2>$g<9MJG0v(x7DCgV5WcZ_y4Q4oD=>
zfvgoqbUYfSXO!_L+MTXcwzp?P%VPlJTsQErkLD0)Rg}w!Ivy_!Ys0YlO)tqr_3>%F
zcW?dMkv;Qw5^WXy$SJJD@ClON0NWtb@IKZ*7M-8JdC;yS-(tB{*09UpmVKO1^8>Rz
z2VhV|k7ugXwXRO%z5}TV7+&H9dJvlJc{97Q!;c;xOppUw>NlI40HEP6L$Y%u;den^
zU7hG6A3qkh`*REy6{vdZ>EB8>I@~JqW?Nn;|3%J6-cy6e0$atg4GId1StC`v&KnvU
z5<CyzZ#^c_;(BncX>N`@J!`TC-b~A=qll-T3luT$wH(pWDOpxWhWdKb9hei@Z?J!F
z%01V3lL->je#7uo4D7J*LCVmGwhu$%thGjg{yi&yZ2Uom2@J@R+d*m}<`UQByIJ_@
z)2ExCdvhPPF9KT+0~wJwn;rQ^sjC2fCPJV7#}9Ac>3=bL!~B9<r>pb!I>?W5$U?(;
zM6aj?Q|jnxA0XQ?K<HA=+#>?_A%1`p1n@eNx4*xaxDW@`jLAvuix)3yz^dU^mxJAh
zYSGjO3X_Jh)lHpvD#BOSw<__LBXeUez!8Er-D4oBQXc+WNh`A{RjSPewzQLqlI^Y4
zIycUY#WP3+&tajFD^e^9SMt9;JIu^4QpLPh6kAWHElMjYRJtQPTPm$usP7e#y_NPO
z_~M}Q!u&C3xb->4J%t18LUIfAz1oBe?zE|$>eY{?LaS4mEDyZS9CE!n9mSRD2F@t3
zY^Io=xa55S*9*TGMsHvIX0*O3>LHpA2iW)E=%87rjFJ$D^Y?eeC&6+c?T*)ZE4|u&
z`KKojb>#%+o2+M;Sa*RE)MAfdxx~=qZC{;Vp8Zu%1VTP;-@*=&mPm+fHNCjL9CLhR
zzXvEe^0K<K1ew%AhYk+lF0JsomM7fB^!m{ngghD&+(j+^E3o2-8HDV6p@R?a@>_f&
z(v>~L#wB5dzSW7xV+w?Lwx4sA4_eL}*3-l;WZD5`y0XK{B^@j4#@fH$*WZ10MiG)Q
zHDcg{;Z>Et2A%hF>jB@;t;D~#_~26JJ~YPtq#xmjfdsgoWgXs15FH(zX|28LF*bqn
zuCBZ4m#E~dqe)91vCsenH^S2L&+A(}7PUm|xv%k+**CNCrA8l^<9sg-2Rd>MC^U`l
z{LX{DDIJx8!|Z*ZZV9O-kH;y$@^?1mL3SaqF-e)7R#WeBDOht$<{5tVNGx1acWp4^
z(AYFzcP)U>lu|YF_^pv|0@Z!Hc}(1V%$h#)T2DH(`JUsiW6_z+G$9V`7}Iv7gE|#K
zQpC2Zy5pkXDH|hj!!T${0tJ(giA6!;&rAH|Pxe^pWveZL@Q!nfd`@7@GHTaXJdj)R
z4w!J35nmcBXXmCUNoWD5W5jFFp1>af`q5d9@+MD=2jL^}t61mqYIC%_5DS<-<>7I^
zAWH_Hutz_IU5IZ&+AEI^y@aeEVPmXfqIW*353N;HL`py3Uud}bdTWs;l8Upv<~IgZ
z4`bFI8XbvKjv{t<yBoLcBZEFunowNKv@btlIIGcA7uRr3h~BDNRbpj5(T<QahKTFk
z<sRq<zEE{_g5{?9dejUi@Z`Fx*^{HSaC~s627QmPg)}XPM;0->!hoR3c1lQ8;aVy}
zL%&b|$yV>#w4-|hvG?{kcoe*}l&qQYo^(|0U%pTsUdkfF4_nloXBc9*b-_k}`QV?2
z0AtcB?NWZpgteX!$kn{<t7i)28o4iAvnI%aQ%(BJpP1=VJ5UFNUvBVQ>SNG$$$H?S
z@dEQx`q&!>HFS#U0th;)S{k^aOcF<D?yBypw;{GR`HT>%2#zPA)C{8D`V|fcF_TJ$
z{?Y?^RW%%21Ze7Hbs()ClDv5a5UhV^k|=X3X|G**+~I1t2GH?d-5RR3DODLbQF)@C
z98de}0Up#QupN6Vpt)d-7YwpWZz`rvuv^4}<zou5ncbm67Xq|$*Y{5tXS2~+!hYb}
zlna!6M@I29gT95?v-H1K1zWQuDbb-wpBaZKLf@5ukWxnqeEg0D;@5He-F?R9#1Hte
zc4?E*)pnT?@-GCt6p|!qp+UNyX@u%Ws^!MsW<{%O20fu&EaG9O@+6fOMDb1Zoz;B8
zKz((^B=MIp!&&SUcd*ri(WZ_P7<3aPOS6qIlKtItyjq5Q$~5ju|GH0P3(?E)RDNIa
z1cXaMsz|bj4y#N#U#{t1=7wTiIJb9n;k&rFj!><=m&l9mF^H)bXe-d0x#dL@tptf=
z6~NEkFvYeMNDn>2kZlm(T3fT#x*xS*AFIMvz?Np&C_)vVzuF2pG1^m|`3vMkq93ku
zNGlD_a}9l@R>Pv7PYy{n;FRD=H~{!>H=-5>M|y`I^VaXoVk!R#C4#K@wm`0$F$3$G
zAym85g^#`m_kMCubYZ^2QnhGY?XCL>bQD`Y3ze363v8=h56ok8i|~L!M=82KTlyA2
znDy0w`Gp0@nP@G1AKv<)vyVmy6sw^7BKQaudVLsRb`%z_V8<<=;Q<MjTQ&+m$5;eH
zHaMa<flSyE4(9RCm3#(h+~fXNl4u$?oeb`)E6m;o07-mjA?to%PVB(De8-m2NJoY9
zNDilWAIPo`<#hIaN+CqPe18R^DqEosz&aY31x5&ylP7Q(+@^FT#<V5{L6C9jGm&*B
zh5IGx!1(t|y%sSVt|shvcz=4vX!7E@cnKgZAeDDCI0@QdZHB<%Tq`Gvjw>NQH7Hfo
z9Z$rT!9)|H2jJfL^6C7b<#*P}mdFWVG%N)t)VlYbQ^D4Xj$|b@pzP>x417tO{&;Kt
zL^@@OIV|a<D+ac})p8Ryg2F7`cPHmv^@=luVzWL!44drQJyNQ~N9_+Cpoq*_K)=zs
zXETz=cJX9jn+BUT=lojGJh=-|+G{+x+(^arp7`9`ul)4ZcZ%neZ460eE;ZCtwNEw7
zVY=>jD7k>+RkDxq8$0z$V35NDLT7^A4|M1h;=gX)*5qlxO={i^m!{N#a#v7sdd*cp
z!rNw9pV^0DZ8EKU#)^Lwb1fe#xFc_+FsICw$-Z?Ar=aNJkZ_U8p~}$X?gJ(<JZiCY
zaSz)GJLffI*2sH0AjNl9gpP+^?Xzaw7Fzq%0!Aw!AmA69qVfV(z2{ebWuE`26r4Y0
zPC!im-XTyzkN8@tS$sCh;CXNAg{tJ$x-KXMvnC9l=Y7NggC+XYA5P~HBzWQca?tBP
z4r?*4`4c6dyKkgX<i}U+y=o&Li*g!^=3i)lV&pM6>_>p<HgG5;^?M3&9*Vd*OY)?{
zQF?ERO0wn=+1^yfmeip!hhso*yP_1>1M`i~VmQQ@GW4Y>L4RzNI}gslfR$kQjCFK4
zTK@Td_G^x-pmP2#>3lr&DjG_SH+lk@KD}evUkAlhesE5K^RJzKU%R811)uOu*X#P<
znow>_V-dHE5?ZezqTY9p*yM|d>)$yu&eafQXNDzDlgbe^87Ka|U3<FRi2qiat>A&$
zlcrN@q?HHw#&i+X#U%&E=J}hUgtKNp;f8F>4l5O6;l2Iny{s`t*C|wUaX>%&l|U%G
z|0PHoyUAE-BjJ@%%ez$(RtDKRX{XyHtT08}O=@&-RN++}-u^C>_oX^Q#@&sdDx%oz
zd|2Fh9Y9&m=|vM?!Z4i6@L9NYhNb)cfDw!D<5W)-?JhE{&<&Sn@I{;^gDI}81wTB~
zQHEjf(eR?f?s6@A+K)CqSDm+UZ52dS^df(+Hboe7Q52y%+`F^Iu>|oq!5E$p4dP`Q
zez#YEsEQm3<V4<RIIm9U%|H9vYL#^|%kO%m1&u!u_I8%=;`m!>g}C)I9s}~*qEG3a
zA)CV>%OvP^0Ph`lH=?`oBudkzWO_SD`4t_u2a+ccBARH3ddX@B1MoI_FHXG&|I($M
zb<*pb_GP=2i?#Jd5(9~&p(sD!D;6vEa)oyV;;Kpo`aSN#Sde+$Ttq#_wb?z+J6WTv
zkvW(6ESS5{I|NjbjPsVQ%zZ60wnr}o$J}h-l78VWjSlz7@3l!Xh`rZYUuj$}c(NRh
z-D*wgwg@fx0eKKCy<3XtG&937!URD_wn4|a5Gij#bDyuQzP|VM;)@=h*-zLnokpVI
zeP<Gwtarx43+iq<MkRd3-xb?#9e?z)pCfP!Iz!oNXvY-QVPf4JP}Wbby!OiP45j>l
z1fVGW&wow5Gcx?@?|7B+cc({6AsbHt{CqV`D7^iJCV%4ST3b<F^GQ2W#s+z9Y|!ZO
z^vg(GTRjm|HgUpR8KtUs+};S%AH0Op{`;LEowJ;x-jejhbg%YilQsq}MCy7NgZLKi
zRN-dVWL)y%46bcx>$|~-b2$&U*;-0(<@h28b9j+w`Q$dB?P!Z|vTuA1;;S6v?gM^4
zrTKme(hrZ8=)V}RD?R~pE;v*eNeHPcku?1ZS48z(?W4}H2?Mk!AjG6F_);9Ot3u}1
zQenA4<&|TcM`4Uz`ah$~B$nkMKmo4OckkXdC0(8>8nrY1e6mj~62^5ROyj{v27i-R
zvd8c9DwHH_*G((<SZt8;r04}gLKwDjvpZ|Htpd)6j4kdE$wM`Dtbq0odbXUI^<G`g
zo4CEz#_n{*m=%1x;eBt1)31W7jdMb4q`HbU181-3SfBSgdiTJF&gc|1{UF|LZ0l}<
zRFO%@P$o1oQqL5M^<TK}&Kope+zH3k|BvF@R_&HVFxvPAk+l3XTB`@OH}92YXrGWK
z)d>&@EpTmtSDT<`-2JC$QBQny`%cxuANTLnFMnhWT#M7Onhcp44Tq5`LKJE1XFpCO
z$rvzv(Poj^RzkF$S0xeXNmJt?TZ*8(aj6aUpfRpSl4HEbV$XYx56&-mYf*ea>p7Ey
z!~l<EjT^PxU^%u)`Ws(seDFJ-vtSEH%q>5kb%9)@6yM>3o5EDdb^`>}hO6GH)BiWe
zPe|i3a+8nx^xGVL8<;duH3zn<O{vm@*C+50AiaNfz_OQ!uLn{d0aZUW0A-U-HRa>m
z-DR}GeVK|yowDm8KQsu5dd}rNHY)1iqd^o~>dFe&Z=wLW?J}$;RPxYatTzFa&=YG{
z%}cwzgHE~8;;;6@oe_u^jb90@*p}585%Wu9-txEtz>bhw$>;%g7;h*VnwyTmb=M+~
z(`rG3Ip&o5l+Qo<(tQi@XLZ>N1ZM78ejM;OZ&yy0y3X&|I%N>H1=j=bJbjWxLVe-q
z!Z_HV82EDi;_?n9y<FUKC<B;8P|qXA89Q6EK}b8_@cy?oTynGYNq2fDiZX2r`Wh1x
zWZ=cG+)!k|?G(6c*QCJJ6IcDDY9IAwLYu8<#dcks3w!dfaagrEwvLtPaVwK^jIMMs
zYK_!J7}16)TpvS61rMYi3wvCwgP9!@Y@kn^V>M*^<0HfUb*}vVr@(HStUKpp^P5X*
zA&@_v7fc_wX}i~X#LB(`5Y$s6Jd<uSt7v>u6b2X|0lqTZk7;0lZqbe%_xF}dOJ>@#
zm$9+ZeLx-VHlXOtw>AT*GSRY#iNHFVwF#uK`BScBR8)Rl%glg(gI4(fZ-o4D=P}?U
zb9KER;W?oNDyz=mza0GvCzJrU6&0=?isDqbrZCDDx1FqbFTD3-0YDKO63@N`Ie+H@
zPKWmr>}d@nApBF{!cps|gt2@xfrQC3qbCy)F-pRKoPHZ8q9b768mL`YvOU|C%+2Z1
z2o5ksp}hN=NLTLEaPW8|XHfNY`#S5Y_c%?Xr}+~X;&CAhi_@lo>EAy}=wwbL=iLs>
zN4TAr@Fc7Soj>F);Ws4g|F+5Zz!5U|1;Mhj{(mqW6ySx3BYlB!?<ddY6xBrP*jC6Q
zFu?el;?bARZG$kRrgN}|u8vO^wxPQW{V8vSwsot=&lp&f!BtwI3tSzF^hbF4p4g3H
z&wCJeQkFct#s09n8PZDqZzL4HJH;$8r}gr%A$cb&hIdzWgAvAw;0D%p<(Xj^U7I~y
zo4HZH*aqaGti(7u9;HN~sitTPn$1>xQOHv^mW+EN<5ko~5D!3aJl-4~IQlun5GGSS
zLs&gyfu5-3|D|zQE&v9|TlQbSNu}|zJGo@RtkJvWmZt4oBh7sZu7c4T46s|^a30eZ
z=HDU)puY0{EK@4}KP;20F4mc(e6S@Jy|Ao9ev2hBNK}nChW+o6w&2kVT=<JkZK6k+
zaLL`sQo0TsIk6!`JsyuU@gonkq2UT#*V|%f`vd|w3nU?_#pamCD^7`G%d4r>N9?l&
z2AF93f2KrPEs5=cZnhnoMtd@@hUUsx{{8(w&5^D2u+8sKnKu7b$T|UjyW1xw8S$_Q
zNzm$>8j_C(4r||zBM|cs|6>7qH@hI|Cn{t>|0g<P<Kz#l#_Ehl2#1M%=KGH+GBa~s
zYQV{seiFSH0ieS{(Ebd3!OR&NI2k(=3r*;zZuNx(Xq)ki_LssqIT2v>{1zgjPCIX!
zfBz@|6Q9{Q+?qaoC?JSoimY;#0HSii;@QVJWMVkjD>aZ-`o6wCz~}==%{{={1`TNl
z#C^U#%DQ^sIlGz=pvELID=>Y0Gz<q8IuJ#!msY;dx1X5SdYvo$^V&{KA)a~<!b4+;
zm&8|N{+KdN6ZCDQ-ebG_43<kZ%}AJdn+2{HsWo!u9?lDq5&@(iG>hpp;?LYiZisD9
z95LDmhnn9s^1}z-(BokVB=FAJSr!1j&Dhvjlj}gHRIEgk8-^n0K+L#V#O{bL6axTw
z#)8*GFqNdBaOz2}_RT1Aa)DMO>JvuQ@~r~fa(@;s@e0j%gox7zduY>}LLTdby8s4)
zbnaGngpYT0yaI$_>S`oSUEB|JsZemq0d*+=RN47ORnv<;8`y#;q+jlg2t)=%TH^#k
z82Pe{5!f1KIVgi(+oc5Y+t2+TNY#Cv<9apii5X0^)d!}s+IebrRe>e1T0%!CF7g_w
zf|OsE_cO|yshcI+FDWr2OZro1<)2UOBolO&kF`a3PLo0LA2Y=hmuH5PKxi-5+)9%|
zQ9f)OfOk81trcwI7bqI_$q)y|KqK|<D#Ql^Ghx^pZr60v%6QLZy$OOzfgLU|jsU3m
zA0V&uiv2<>shO-$2m%G@(ZS!$qPvS0O^o_W!DIe{ti!$T4REr)08cpx)ReT);(ht<
z@1f{rZg|PB%Xddpz`+ohKrQ++O!Hq8rD+KHkNE1*AyJMh>c!AZrpOUq?!p&aJWxXP
zuZC;Duqzhopbhb)tU*VHCrp#^;0`j+!8{f&B?me5XMQzK)C0vT#1I|8{QOoQ7{}@w
zz@cZKasWZwhz~YZ#Lb1;7E>>-U($pUf@>2@aO;xpcMSS1?@SbGQw2q0rt#4$=+4Pq
zIzc+ijuzgq8-TnFsPjBLLhhhHAo=T@Pa*iLk&R!w!$>=i13N0#gj!%!6hS`?Hp1>9
z(jKt`v@5d()-aI5<~TulQV4Fj&nprO`N@f3+>w4EPhE<aYSlxc#V!#Eu?KI)(9_<1
zw0(3v)y_XqJ^eGq@NzjvZJ`a}r9!Fl$8+)Wtw-sJ2580(dHAzk+;v##-riddTFuOY
zb*C>p>f7PWQv$^4y+xY*yIbJMA+GLyuBebIuNaF}qHvB0;(PHDu&kj{=pjrdDp1^7
z|6@3?KUDs@QMM(P*3U-X&7OyD5}*!Ufk5R19gfY{)iT}M1wU94qF<Dm5#q`Ks9P#e
z_T{s#ewCxwZ6L>P0!y#vPpm*L3O;2(@$~a;7LaZs((Zxbixp-^w~KkH`rA%0(}U;#
za3TV5eihm2L!>|C(27<7@3c%%si`fZ?`uW)pelf8rajCdmkH~%Co5P_J>H`IS~m#L
z72a$Snr>q>gZ&0|o8|93i<W-!GA4-{dK=IW$~6!>mYcA>3$W1&jsXt@UjjVLlZ2}?
z$P=R$+h?SyRyaRGsI32e2MENZ66f)&P~6w|Vh#m6pd~4&o;gGsVNm(w0?j0Vm(@jM
zVEAxmuvUz9$g<wafdMnh{ycQ6A%>}ydHnTj^gU1&Lkoq!%eUDtfC=33tHKG3g-t}e
z%h%%kd~Cx0PPYS`K<MHx7Q}%X01Sv@);T>|@Pdl~z{#Zq=EmGfE1yo9>IlVc(uqG5
z9E~y<h~UEoW9H7S06*fYk>A_GZLA;6*Zj;Ma{BXqB%*xqsshSN!-<V$`f*iFQYY_2
z%To61F0eKni0%@+{PKAMD_YU0_7_r}tAOVXH-9=D)Wl=cLSK80hdy#-UxDJxD3lT`
zAShc9sua#4$rRq`rR#qF-L?M>sL(B|2M{Kk@PePHcmdXpdCF|ve5-#1D&x*zh}HZf
zK>je39@2T}^d6KK9KgDj1CE1*QL!Y%oFejG7`eLrR72P4Rsw9Iw>5)$EB*|<jyV}R
zS#h-JukkpU*&B<^lK`p+zCBr4cuPR1BG?VZIETh&1%xqNV<{RYrdk-alSnd(sQ*Nn
z1Rfnt;%WYKN}rhL;EB3KVONtr6Z(v^R~wX{xQ3#xX^@W0GQB~`RpWrorV?q(64|i3
z*`x1;tIU*3^1%FIt{E#rCB<v@^}(fTso^AE8mzlOdqASVe#yL-m=STGw3VD%!4RLd
zfac(W#pvFzWFzumuq5Q~7=O7|od0~lQ&b2Dq?*8LXITdb6K<LKi5Yrxq0ip_kzxUU
zDm>#}7a9CBn5b6!=$h1vn%Z`-N5ob`(-EvfHwLHz((NX&&6#nv^RItE(lu@drmXnD
z8Nz4JPVrFbyYPF;Ad21depwlu?zCXlU$XTsJ*X$lCK+P<Zxm&|fvQW@<27Yd9vv%L
zuU*xLg+IwDU5Fqs&LH7iyCLZ9Y?jw4E>|;vM@m((gcXK~;*0!C&t2%y<4}HKj*d|S
z(cucy8=<5o-ehsIcd`17D}l)&!t1MrODt%M1w%+^XMt1nP~D6qnJRcH#ZB(Abr^uI
z3zG=si5vZ#2^O6-nLu(0MBim|KVxaKj6oD7tjLc6h@JsFaTlyrH(J3teRD(yn3CvF
zkRX}j%B^Lm2YFL|wKdOmH(L>X19T??Izq#4d;IR+9LnSD-X9TNFs%tc3c^u$@VS=$
z1gAjnu%p2h(9^^PeDWfJ$21fVtrPf~yi0~|Un&GLyPuK6tDX(H#oc`cxKyjX1{8{x
zvX$t|MRCc0|Gi${fCFIqD_5l~I%Ol$-mO<uMDr2_L!L#$#C5%HePSWHcVog^O$9cJ
z5bWT*=Fw7sS7uH*Sksl5e(>`q59LYzdpv6A3ld0D%Tx<_;<Rgn?|_Y7{MubMC+2H1
z10dW=-jjV2p@ND8`+?Lm14mAc6p&2oa*T|(dh+o_BcGz?w7^mh_ZnM8`GMnY1jpoR
zzaiN8FeqmMhq_p=PGF50IYtQ(y-_5L?v#eV1ei`57d845{ciNi%&^~L7qL|bF0BbL
zayv|M?n(WtMq_+)(tKqu)bY@gI4Kbq-7c**`iQ@I3QXpnNaAH=i?fB#LX!fKQBRi*
zSP(U20-IB4ZpA4W2kmW!-V*C<XPEyEo_052AvAP^5@;LTMX%@^L&>EIKVuoPj9!#R
zN#%UP*6g6b^X{U@8=UB-9IdD`tls<%Z~>zaTX^#ZNV?y_*`9a0Hi&Ds|JpbUKA8L7
zxc>3IXZ-T6-7W5i-M^0sKyR#lPo%=)h%Qe4g5{pPbPP}_sK4~FsMJ$AwlTzrTv&?|
zxR@7+3bc}!o%`#D{u^l3K6G<#zwgV*Oam#bXqu73oBc_Y?>?dHoP9!JIimQnZ9jtm
zP>%oGrKZr8^`2`C+bW0$3XH}_>Z;}+lH+}9wu%&#GIOI6Kue1Cb^=HP)CpHUe2C-C
zMU(HAj2jHUMr_+v9u~Rt-Hh^Dr`A|6qb(945S!3L)gsTc{Rk9;aXA&BEcblP;LB#E
zb8(s8H^W#+0Ltl$^0e}Fbko^9E&5H79;Zre3Yq@5DU4pCXM0NwH$w-zPxfYo(dr;1
zkhgWOnc?9pWT$vWpN_F6y*<SPDLt^4wQy8z`Gh7eRYcC<HQkk+e#;g{A_RKJ`7^tH
zRsl<~Xje#eUEPikXCLR>h^?VbS(ULZ`VB7Z{qiXO;<`Z(Iz{N}%f_#m%02*QtB2EY
zY3En|g4o-;;MAC*jTFffipMisTU+@qOo3CJD&V(z1Feg-QQZHSR{Psro{D@Ghz9iP
z-0=I`C9{pehx>*En4>+KGR$w%fO8=PI9d=e6{X}w3;FHh&lZVq|3~5BXf-}y-VrWG
z!YrSJLr4uNl?f$)czb)JDeSwWu$v!?O}@g9rz<WL7Tm;nG4EqlnG+h1qG;|Zp!n09
z50Fv!pb1)F*ah!;hbg}K8JtE}K`^Mx3WJuJITeXSriE6-64G@4_Mbro0iJQ-K^NZh
z?Ciw7y*<#W*C>g8W*(5HdC>G9(aTjCcYZRM7Yyo}=(jerP`nWh?u!$5VDw6bpV|Iw
zCZ+}Rz?G7QrpHaZ=<{dVE?XEL2_pmobo{u%Z8y<u)y>WH5KxT|4E-hM*s#Y8XMPoW
zD)53^H@4OY?LePKJ>t};>;X;n^#gIOko-Ika}7PN#Q<Zl0WEGD66~lz`HtqwVL%wf
z-8#WYTiOz~=O8ELKmX<Vl=&`E8{QeZ*AHK*02`4N#9V1U;Fx@{Tr+RIu5#ih9z}w8
z#Z2$qOJfrXl(vF_O@@aOpg?$9s9qki@c}q-ITGIBJlGuBlJG}U>|mG>9wBDpY)oKY
z0N669?wq#%ge_!^HFFEwKRKwSan%*UhgnIYX79TIUV#~mTDT~+d;-lRmGow*PC)bm
zC6agm8Fn5`&!3i;Xc*q5)50-I{y8`#AmR=85rFt!_ktajBmrWjsD00km=`sb{p30+
znEehikOQL`c;;Y)1g2*tmg|GSCo?tE*f1LAxd)06aJ4lR4|6wmw+2Y`k6d7Jh~5J?
zE-Nbxq1A;75_8HBa|;Id!u8W+!c_ay;4x1zxPblcTv(}TBksC&77nOtn5z)5ZV{6V
zumx{JDyN<+;Gx~@(+FoD^Xk_H>MuQ&@Y_JafUK3fV+iLpm#p!J=;>?cyw_jB1NWB3
z-XeaE>dgbOF}%mU(e7u!zQbgNQ~uq9<?3g!37=Ra(oXb4>{)kmm9#Q#VgfhQk<C9|
zb1)5nI)URABYKS|c$|!kW+@|6gj_I}UZ5{hbZWtA)1Fv-NHPA=GXOw;tQ&RDhPlT+
zY2yO+S+Api5AAy=a1UTBIg<f9?X1-3#{!%2?w7J2{r3fi_m6=m^yqpV50vOTLdQ4E
zdf_O5t=UBI2@3UlDMZikbQ$m9OyJ9{jbC3?RVF!xWh7V`^yv<x38Xi&*jI4GY8mJY
z$jdvPYYl(<M&$X_zl$+jl;fr1P?`#vK1`niPNr9HQ2co>JWnC_eb^e_7tW2n=KUEW
zk(^_><&XQ<NjcYnBXd@%U8U7!LHwyoDwEH1>qKMC0xTxwCV$Ml>yRW4P;}Q`>X=ts
zVm^JdjH;+|7oA?LGcWPDt5)fHK@a-EOgH9Oo*h+z2l?$^{M}CndT@vLr$~1S3=&D+
zT3koap?37=HMr(w9;7a|22qTRj2H~Cu(7eZZUBPC)_Box&l)?dSpwGvdZ1{Cj*ac3
z@ODMsTwi+m`-2~lumvn)KgPG+^9`iC)AAJ-mo*;0dzu?k4gDzKD}X}o_2dbn^;VwI
zkY;BkOXkqRl+|MyO-;?4=ZzGPXa4WcqhD+=SqbPWzjKEqqF`iXWZQJ&kt*O%a2#<y
z6QH|C8<ZiQKlccYZWXutJs<D)n0dM~%(U^Z-O{|8$!`Dueo-z42l(_8B5sg`r%#`r
z5HTo%f3nH-;OdR`4h;0n%+S1h_pYX{&h>6?W@bE)D0I)w(Q$Ecp=rzFo<A1Pl^V1(
z{jee@CJeORh~k4EMpLK2Z8HY`?0ve}12;?-kb8ir2EKa&KseR_BgtT30w35<L^U=_
zXlZHLy1OUm=d+ZtZzs*qn-krVM&oyzqj{OqDaL$Yel{=(TG`nlprD`tBg61CO33~E
zP=Cr7a}6yTJyff$+icz3h&(+#A38Z{-YPSih3vd;Zk9$M5G!EHyQ{w}Ff0a!rFo%8
z_yhz5H9kKz_-)!5nZuEG+}u+Fcc!#U4D2R~@0XUB_mn+&pO&7E0NN-P0fDNQFI_u2
zI~7GmMbWfxDJd!1yAPq5SXlb{`mBDUmHr1SO5bwj7<_+yQ?j;ZLYoY)B?@ZX(sWW$
zQ|lQY*WdYHm0f8#mFpT_C`n-{TQXZh<|&kf&5|W$t6j2J$rKrj&_=Cd8#1IwBtz{C
zwaF4&O6-!EB#C59D6$NZA!IoB*ExUA+2{N?F4wiL%UZr~eee4W_xrxjeLq6Feb{~j
z1&>Cf8G=rw?b#Vhh#)Ci<;LnJw{?_VFaY5A{h)&Pc=ryX&Yz8qt(fucVQVv8mZmBr
zD;t$(Q~J=Yt2%m9baRHO!8=3U{#&fX<YZ9HAEQX=RW2A@Dftf`2w>|lEy=R9=dq6|
ziEOYq5#ro~8u-<uxWq(RdyO$GTU*uC)Ko%_KI?LNJhX<ZHJkLLF&wK%NI8mrTha=_
z!R%wP6#UjR2R?q3p-4+>H7F$P*{-V_g>K_}l5+g~Ib-S5-#(XB*3|4~3DK1m71xDu
z@)#j=X^pV6vwN*9GXw+#Xi0lZ+dDfe8XI}eejTp{v!MCat02_xZ57}yE-f_|`^g<V
z8eS(Sr`U#%1T#ZaR8-BrV%3MCA#<^PBZsW4V1De-@(Rh$&TjSg6|`NGoOAD<v4sWi
z;?k1V$cNKA@EVtpAY=b$n>`2ZiZ82Gezdnt0z9Hv-6$&*Z@IKCvb|E8dOf%jUw?cK
z5=^lvDNtBnCnO>=@Z{^eN(`kvv=ROEn0`wOexzqPhwhO@M53=<6R<vVq_U}r504bT
zJ#BV_k))oUD8Vbec<~~>N*e|u->E@R>1eSfYb!k~Cx=@hA*aK!Ohxo-2T7mJ!NCE&
zZ^h?*=D&Q@{;UHDYLKqN4rEsqdk=^kpLX4qS2CQ@)^>m^xUP<d;YigB*wE3@!Ntv8
z0lCrr$BwDYw0!bqGE?YGP7>M!ZfLpt)5X%#of|f8l!~t>8YD*EG8uCB?0M+X@!J|c
zaXZ}SfN2}|*p5L6a@#M?O@eFsBb6FTtsnOG-@_tJe^$0B5$NmdyV89JY9pSVLs{ko
zp~f0iWl}*w!3rg9mQ`2M(bc6BFm%|-Da(1(>}@CA`S@{38JX2(%L|H1%F1jc3C1)(
zp>zz-NJ>hwpxxU=Kk}$7d4P?B)1^&h^XA*Q_Eu$OMk;^Nfg{ULFtu9qQ7U|VeCXz|
z&dkj0ih9<p>xlYiP>F?h1#UaXU=%<qyf)S1YkNmWIXuiKcf}lcxA^BU5=>o)F8j4F
z!#>rVUkHVUg-Pk_Kdqd-=c)VWNqv2NFeGbH<A~P4B$BJED;&UMI(;eA%~RRk&5gA9
znW1q(?s|jPiO17s?jwWa<Gj++(gHvHVE>^mjuf=85P0p{wP)4UAH8c^b(zdALYhp=
zz0D=@R8L7*P!NeQrg!h&UC~Upa%)h(pPRcLY~{98&*jf$zp$_8;prP6zdSwd+i0qh
z{l+cx?p??Ft1_>{-B)V|1>S)x!gvU(2pE@xQ~3+fr1SCdfpI&&NK)tG;xasV@Y<k0
zCkICeucwCJK65`2os*)R>xI5EQI{5G;MIm;X~6yb#@AIy$Vo^@z$m4al-2`xs;R4&
zgKBJf^ynqCcv&;J(l2znLvy-`r9uMoOL%njI&F`35(#Dy2m57#-J0Qq+mj4$=`&ts
znDWU}hC@-{IyXOGUR}MKkZ=a;He+*3nC6i*c?AV-dwctdiHV@=*EhYpu)@$gH%CTr
zi_@hyWS$q%@XO23S7ZCVzG-rxHjHevHQ~8)t<BM+TN;&%5?QjL*a(+#s0dwKQB_67
zyhg&*qA=FOM()|5&qjC;)T|W|5y2g#c3%>bLI}#boHfH@v8o^rA7*r$EGQ^g(c5eL
zJr0=P1(Ka|w4+O@s~2<(h#d(Z^z8W4WO}sYXCYx>ZdFxPLMz=2!Tz`rSMv=XIpjZn
zv*=AT|B=IDk<w^5&G;WF?(c8YU4d|s?XHean(x^IZ{B8WLZh}GTU4>5P?QD4#V_}s
z|Hfm>m`QL>D=gG}-q2uw;)I-y8YPf2v0jpYzg*7!`=`Dh<>26G@9G-JzB_l}SIWY#
z?W7HH7Er_A3OUxbii(Pej}$dEvcbw?q4VNkV@lEvA6X9%4?AaPPIyhKK7+2gJoDDJ
zvZZBXO-)UHVWF{?SH5%Z!-s+miOM1I@f%;fc=3L0EM}veg`P$eni;5_oAW|+BP4~a
zp9E?N3kz=>Knf*<yuhc&b;BX4{>MRu7b!Y?d3kxTqu`Vj5v1?j8-;Ff6t>97^g-S?
zw*EMNw|n<f3$F*2tt_AQ4+to)tK;J3<&B7pbe*0%fBwAx$ArUVF)<!YC7#d`d|qbG
z43<IsLPG5>HVnotG;%pJ*7+ufL5@4)z(J$xe(bBf7!eg!0b`82d6O4<<?r9y{qe>P
z+<ZqUu5DFRti&hc04_d#a>3t8&wvrgqN10;s-@`YY$CXCU%tE^JqtI3VxZCK(;6yA
zi-Tfg$;2_q;rqutwHxZ=<@kk#qX>HHbrl;UC<}vB8G<F&?k*<1h`YSJyx4z|GWarW
zT@gJ-d{-a{&m6_R!yEeh`|};DxzC{hn-m1<H`>A;dj@!dSbX`*m(z0#3)%sTaG1TR
z+xc(ayjj=KP>B;pR0*r({!)0S`3!=anx<wYs)!4aK0)GI!676)y6C3C1^vt7yP|N7
z#uDU}*49=#bQ4j2DMwM3jfn7VM+#g;@}BHBiHnVuR8mr^g7iU3Pg`xRDbWF^qrmq3
zyor#IPy{r5c8_3jtC*;2s}f44sH^Yy9qTM=ZU^JA$K@Ns=-M@F)--C6COh5l7tQG*
zZd3y!X*Iyl99XT{F%RWcR0K)XlPB}YMiJ;?!lk35bKB~kR+mRl$)P0V#e_XFp=1Y4
zgae{u%<RDMusj%Jje?qMrK*D1II0khw>6+XUJz-7nQTt&jMlkIj<a|iIHQ2k^IDj8
zJ$UGl>&x+k#Kiu^`6(i!!<3WL&Mq$o0F7?jc?M4j)0-Z7d%t-B@D9eWy;DwAl@A)L
zF)!!Pa9|NS6Mtl92WMx?lb%&pX4H3MufqXx6iKBTGW@|na{z@8s)ts7#R1jC<Ye$(
z*_Pb15NNfuv5861IhBrDZc-57vrh8R`}YJl@DWR$=wga20a}_c(j#%k1+D((JE_!6
z#&?Ijp%<V5vBrGIT3(<WBc_!j%mOGdBr>vIdPv{xFRYk3IXUa<>w|-Xf6P$#@tL}i
zS{OLvwl_taS)YXbIryulX5ntf`ucjPylmD!+08|SYin!bh%ILP&?wEO#MkwK5~4_q
zkGYYtG5^MmA-TCqU_W!i3;og5M#6M2g+&GSrk&`1a(0u6iAj5b<Hb8`=OG)UH2{WS
z4IrC0+8{VLvGa=S3Beg4Mqm+NU|$&yF@nC?u`cFGFG5K?J)JU*r`^{i4ZWwQPoG-<
z<zL|3)7x9eP6`hXN2`P@-9JGR!W*EKOKRH*Hcd@U*hny(;emXMK|xWGiKQhONCU{~
zjmKC7?d!xoNouBME|Y2R?jDUR4|Eo}6I%I5V9w{DVV3{sktzNKu-6+XicsA^89j-V
zbs)0IU_}%l9aBYk9kTu!(x<ogC7c_R!(P3zB#t79qfaz_g@#5ndZTPhkyBCOCD0<s
z`TH#`gDNVFw#dmzYirNOH4T8IQG?hqIrgZxA8QH!5#}Q=+<tf0H0#YeG+Qy;zn?!R
zCntGnE(th!0E;YQd&YbIi<2nYCE$AiuE#d)9G}p?uzU`Y?sk4Y8BZ?u(3WzQ7ptAu
zn@>L$#S~(m%L!5lh!~ETzhrM?V`GA>fu#fDY!ylzCLo>RqB6WNKi`KBZa+7t7S1ac
zrS4cGxpU_cO#^jx{@b^2TbKCCcXxM_1q4Eo_FTrZzMTT9r<s|RWAEkt@Ngu*^lrPB
zJNvx6y#qc>Sr>arfzKNby~$@?T_OZX^VKGk0D<2R4id@0cchI6j<Rod;KPT|*4Beq
z<2csEN?AW};YZ?l>%h@zwY<FS<>h5)Xh;$j6Wi(AOWneAgKR_-`z<2k!C^nBU#F&m
z@!USTbeWoY)}K$3pI^(IX_W!=3hWV@knqEuy*DU+TPQ5azgF#~X5Q6bn*JG)LFe1l
zXkk-25Dg*Aa^e;v9IsP~pCWGl>vQ6I{z1>50|b=#4!`bya*3K*3cvck5<SK)i%0k`
ki^{)e+<#kh{$FpeaOCUwa5URjrIYYuW@16jGqU^bA6X3RkpKVy

diff --git a/master/_images/sphx_glr_example_inspect_predictions_thumb.png b/master/_images/sphx_glr_example_inspect_predictions_thumb.png
index c177bae274445841baa91b925cece0903da17cf9..1371740118af9fb6cae5b28a4f92b96c93b50322 100644
GIT binary patch
literal 14304
zcmd73c{G-9_ctu1MCK`TQc@uznMJ0`SfnJ%JY~+5AyFAJM3T8E3MC}T6v;e<6q#p|
z%tGS%oZtJtf6w%;^{)4y_jy{&T3y%0c^&6*?7ct3-e<73rYaQ$8wCjo3DqezrL!a?
zTO#qZn|wRoaW;JHNkVcw<CKzuu1Df@uk{t((Z9=cGjW1(>daA*kul1*Zth74$|J8m
z#80tTZHj`4j*&^5Vn`?<<Q99!*=!1mvo9)5U+@Rr<lh@AwoO~<ZFZzjQloS8!mXoz
zC+S0@;}XqQe2m_h-H9tof8x?qG~?2*+A2nM#A9g>ParK-)KF8vt};#hcW7u2+1^0f
z%EM>(D3bG(?EQax!RIHpnk;+cLaqrJ9UYx|l2lB(Mgr-8Tk+~Z08MA<jpKuZgZH@Q
zlg|hh1rFClEX0<tlf^cC^7-AXnjn5{omGpRM~Rbv!ywq`!iCzWr=lMnHK*7?&9O&T
zHfeQXJdK}1C^j_3cP)5g!b<$w_p0UT?u+kT4@Mu*O848`2)uJ=zEzcb?&~E7It~uQ
zQO{oc;|t?0HG_@==59QJ<Q^fZkFD9T0`&xOs{F3?1i#*dZ?oNWwZRMr$an61VqETP
z=r#FFX1<nJS67#op8i&yS*p*U<!s~f<gGiXKfk<moPv__;Gsh%ixaIE1em{OUp%Rz
zvi;?y!tGbao0B+XJR)!1A}#y#OM0v&C8B$JS>Kx{(CQ1LvVxu-@lN~Xx-r=hQ#n&m
zuYHY4F3A<)#3UOXy$+Z1%uI<S&H9E0ry|1#v9YGYyM+^)o@a)1*;NlZdTjWrsgd8m
zf8Q!zMNyG7N$UE%zANr#cWY~_?+yEMjntaz>Zp{Ih=$Rj#mV;FVJhl;(UudZbQtZ6
zMwu-dA64K8UK_l8*K2=NRP^BC!)fo{y{l)R`SOAxO(l%Dm2G5XtB2pxDsJ1Q(Y`MA
zWu}lfa;zz#y~xFq%X?NKBs8>W>Q!;)`|F3Tt*zCQW#VFuymsC^m!T=*IB+60Esa&&
zZnw)=V}+I2(JLQFOiWF)WOI7$V~c2T^Cau%_o$@Zm)~4%dHU?x)%-52_H-&TGBVd6
zUxEh)u2hGys~Q<`;pT{qC+R^P7<!JzJkLB`6VCaUNrRjxkY;5`%qYK0Jyz(ljJ&Ff
ziq}SBRFvZB)29n+TqizPq}(q_&CImqQYGi9y67tySuG?e7-rt^&}poZ^}YK{bLsEC
z&?bv^>4&t6V&wA$@0n{NxW@<Ya;KlXOWpWLBp}6aBd&()hWeN>g1vl%{3$cH&m!H{
z?YpiFRRw?OdQu=_TJ1RTnMbJj=h3A0ojidX#P2+N_AL3OKFO_Hx3IAfVqzvAohIjD
zqdpjEu!~zRu_H@A-+jgi=jl21ve2=}(9m~v)Jv5?F=)HOo!2(4X{t28if81Keb=ma
zZQqKUm)Trj=vW@3R}2b`VXSr7Eo!lgSLNOwdHIyUz-^_&$cccxCogQCbaZ_E;4ZD=
zmlG69*cp8@^3Z%s!)tvdtam^9r|R}H_fNYGJ4%pKQ@?WSFG}*i@%z{16sOy?sS1&#
zTM{)7{P{goV_&FW)O$rZGA8CzIH&Y(N#}hkX}R@^iW=>&yQh)lL7{ERdV4&TpGk0N
zHcLmIP#D~}YoYt+&-MUC8=FH#x1Xt}tnTKEzQ4edY8!vr&0e9P#)d8M?)TRh1<SlC
z70Fw#7>bca#d;>XTlOkmwWmHU$j*C+Lhs1eEPmbd=c6v@R<XMq`MQx$eWP;jizrWj
zuPgPt++Qc6%)gcYoQ7i12bWexHJz(lG_LWbYxy#&r+=4U^{Kcruo!Hby*Z}ZW6z?J
z#&MgRhqOic+@2zHq2w;Zs&Dml>M9IudX!Jor@u2X;!e#+&6NUGzj$=5#=lxI*WP6|
zZk9${nX>N4X@a067m6n>dZWQ^<hiPR>*ZG)*~giO%+1X^LOxiDIgixR$*s?xv}jHY
zt~1LN54IOi*tv7(i)F8n)X$lESvHpTJ9(5Izq6u1b>i!B{77}IFE%!|rZG<R;E^La
z&c&(x4<0@oa+Fw^NQ<zM3Hx1tILEB|-VcUZMRK|ceDSKFM!eY7id$sQGYwwu)7ASl
z>1vXD!A5|#i~S6_Q*IRL;7@HmJG<;YkwDs1o%;)CZqq7yDF5rgpPmT(?=RM+g3GRT
zi=TgGc<RE1-R&7#2Sh~ZZqCm;KNPxLf24d}F`7?<D4(ZJohq5Fq(pV1O=zlqef6V_
z?DTnr&&ImjwkWLpa860(NL}<~PmyRSv!J4$UdvzE8_Oe;87X5?XM_f0^o_O*)ZXXa
z#d<XGK#F(N)sJ^NR!WXt%M!x+Hdd#}N4~r`r=+CxxxiL`Xn45mof9t;6H}+1xT2z>
z%%#UWcI+_C?;^44$Ye71UEQDJv&`z=<Fq4S2hFlf|3kq`oyG1)l$Di{L;sYZAj*+K
zQYQ(HN1MeoUy6LvoBI!$O?`c7_x(-qZAunWVbkhejTWLngn=&0lxLPL$*B3Hk%@`-
zQu#;5#-fsvrgJ(0R6>#c6#ImP!b?hIkcE;?LxGn2-L0(;K6&!wXJ>9e%whdJ$3exR
zp&^-36XD%KK|vozod?~DWm{tZsVx+%tE+as?>mfhvH6C3%9+&8o~8F$TO8c@Gn0{*
z7xm&?Mjp1lt*!0dOxYh|lc&EIm`=7md*jfbjDjKTx8Yr1Ute2SceUr86E;WphSA-R
zI@4+@Naeo^m;86{+Ro)Q!4u9cAKqcJ9}CUPJL<LkGq}UX{O{V51P8UD5DReXkvd<x
zdYt0bt2_YHs!yL*569Cgl5@{aS(l8z6_X$DF_S!jGt2z@*P<!DHT$Bdp9;Mq`5<pz
zq&86>{-<#HKl$PtsSd&_mPL_VX9vL$?ohE`$hV+6pB1CZ<r(<(<)y9K1_t#*pF&S*
zX&sP|c!d3pii)c2E4|@49CqBXxK}-1EEu$6O*ENSF-VGl?8)|w)$aa{>O@JG-7+4+
zp)5z1Ih=rf_U|Y8m<0{0I`oa3*d0pzGcyH@N*+niltk~~_Ky4%%5v`P*|XE00RaIz
zjGz8<zMTF~Gcp*Mn3Tam$Z2RSahiZ(1~#^k4x48HI+6zu9?-I|s46II(aeaJ^I5wh
z#+GaF(#+hPj+uG;hw{z84rPCrF99iS1(?B_mKq$$hTMupb~kek@Q2(=x9$k%*jUQ(
zoBR8_e_<%}$XOj7|5Sbiftbd}E5f>QypsUqxw$#q%$)+;Z)*e?6Pdu1jru=7KhMw4
zKTyN%S7nhu=v-9W)I`?Ir9+UC@f5#gK8<)<PR_7&Yl42<ySEF-X!SKaZhV!VfdLUo
zdAZo_liB>#moVaz1BF^cxa1ug?&bg*xuA6G#ze_zhqBc>Z{COzki>Q4r^xp=)*nke
z7v8PUHx&K(RZRTE9e3a&DeQG)2Uc|?e;ajPJ)DQj^;bBMrSenwt+l7Tn@mr5eL*1b
z?%noRm!;Q!&uD3D(-H|!eRpeNcUXrF*}(g$SHqcEuN2$GQd3jc|BdZF0?)16)05>d
z2{5;$*h;-$dD)GW6BEdzzx;nw1b-TtVF5lOAJx+{k37pfq?y5x-*vCz%50BCVYgFp
zFW4Z+z&htlDv(tfl_n%rx46Z=&@}ggOnHE@=<Z$n4jeEQPnhCOr%k;-XX~GRG5^le
z)Hl2pCt|vtQ`&9YiFPs6^12%{r8PlxydD!z?vgx`_mczXs;aJT%h2MkzQ;P;;=vO3
zyWcMbr>b|DR-;`_L*toV)_LrEnzcd9pO@zkQ?dx5^1YH9rr=@QiOS{XCZ46AGgk1;
zX!wKw){^D#wgQ*mGp4nYfB5|5Bq}W;{{~ldLP7(PF;vHUPN3Y{dK-m)Iy-#aTBrCf
zx9{%4?#rX%$~wqHg1WPcT6`{Y8MndFhj93i^Vp0Qk`j3<B^6&)rG)B?k|+cSKxIa7
zG{qm|#U}{N7E^+cJvn*zk+i#zzLDpCu=eE#pMWWqZ?ARp3kp8V%WE^v-FjBzWYGAd
zCX4LzhnYj4C7;{L!zNwc;=lfVb@*p|x}IeGyT8lbTZoj<@>^p#-tV1nxiS)O`|8P=
zSfP=Tk=bVHnJ<^ds#8&?5ok85bmhu=ddLb?V;*tw<y+iY)@(dHJlL`u?E2(%1VaTx
zj*O1pkBd76j)PJQ_8%D=tLp5WXIWX2mq#DlaJ$uN1ms#tRW&*;E;ut&?~WFjZ*X{c
zwMD+?or<c>)rIVf>mDnrc2C|?9wH_CELPjtI5HZ`VoH^3G&`)T!7G0x1+WwO;DPS%
zS;ZigItMa(7M7ZKPQ#BRUHB~W<0M_iUf=j_4@`L^=Ocw9vly+90W9WTsURI|jQco0
z+92t>E@5AoP#1lGfs^y{v}X%<1B+$EzP!A=xTQ{0*Qu{m8l|7pXsN&kCwt-?S>}bw
zdW4QMY>4@<uah36<q<OR<*a0rb`t=({9q?;dNl6cNYt6#Vpg;yX{r$xU!LorSY>9S
zE)ndf*Zu%MzZ#!|-wp`0>1&54U4Lp+)-yD785r^hY|#y)w2jz2+pBKxnMfWz>~6`V
zxQ!aebm8UsmJ~nP9cSCwXrk+stgN1w^OA7{RDAelcRMirvQ@%ouBDPRvFO%7ljp0A
z1tIek0tyO`?y0<CJ6$#EtjGENETe90=yb;KR>PPlN<2aqOH#o?>mR4xT=w)=g_rxB
zG8-}avPpgQV0~A>eMFG9q2bw&yvGh5=G0?$*@&;WvS<IZV+slt$@Dsm+InH4mW;oy
zwS7r(I78n0;gtWjn~xvwG4i{<Gc)t6$NOLJuTM!hkEn%NR>Yvd)6<_F9u{DxBnq95
zj*ecq;)s9%&o(vxZTD7IRt#o}fA`&(5(%aqGT})$NWl|08^Ra_IL=Y?H>?87SOq!A
z`l-qnNXv41?1a!+Y4=}dzj}(qtlI*6dJMPIa5a9Ukr&8XzfH@c9)3bd5gJ0~caMvD
z+|k<zjQQW?n@QKY-{yKQSd^wA=IIVbhJN<nx&!3o%F68Ehc3}~)#02CpiHise1TT#
zPjLw+CnrWm#>kYEnIges$6`iC%s7d3VUOEdL|$Om6;<Z*=cv<=5>!?V{wv|VA`F`N
zO*)=7mF_-ik=ozY7C(!Rcc|>T--h9D2}T0-wiLcU&`JrHKf67^pWXf)@uM)qkC?@6
z{SW-}{}ZA3G(#4ZkZ=ZxR?t1|Chh6z**U#@sE~z;>3(eN_{;@xwjZ@X6Kl4H*dv5K
zvzQ;L`}w^ftKKK;{L7)S1P7vmE%e|v+(3ycfnK%sF^B&n5w9J|ZfsN+x9j|nXWk&;
zI6&f1y2Pw+=yZEL#2q5H(n@T;;Yc}&<r(v?JafX8fO+6JQ~LWT6f8(<mQ0SJ2&@OT
z?d8mBuC8J@{WrGXsEd0wJv}9_UcJg6LXc@0zwP``&#9@ms5zx*nVD7E(>3bq>*pD3
zRk?kta4@K0OZsNO|MGx-iYhCy)XYPU%M$=R0yHwL<1<d|+8!`2PNJtLTexKxyZAlZ
zoYIFsf7<nV3{&}iVpb$~^8a@+8XsR+Xpvt?*<8O1@qQLw2$J5q`>i1L-GHbl`qO95
zKx`b?w=b}^Rt?2d1-u0X@wxMDnS`c0O%_NO{euJ%S)0A&5A<+uk^L>~{>u<06+)c{
z3K{CR_3gDwa%&U38;Amv!tRT2tmq=-HzaUVrgc&KJEz_Fw35|bU0n;tW!-goaq!0V
zMLcQ!by+*~JZ~9O87IjbE?vsY&yPbn&ne&Vy37l53lIx~F3L5lqX+-@oys}XCsS^3
zsxV%^HInBbv{&Fr^rip*k^R$CcM_f-Lr^L|dy4Yw_eZq=`QWE$Yik$$5z|Nmbi4r%
zku#y7i+p-QDeb(r_J?1+#aT>2tyX5gfAyjcx8@kj_t?9~{5ht<D<@|}p%f$)NlF)H
zkfE70f$N6-E?sDrCh%umezORu<>VA8NlW+IhV3i8;!3!2#0fEt;gz{z)JMpU%wE5t
zs-3_}NJ|m|dSYYmc2D#1@!i33gXLQ717NKn4W4IbmkgX0W(!-N{YW!{umDQyu~%k-
z?}2hix&2fFR$Vrsq?<+LWM<C)$jZs{+w?tj_;8$r)821JH(ri!l!JOg^=!>(ue(;*
zTfOwnxP?JW#=m`XpQL0QIJT(czz3U_!~KVj$Px1FNZHC>S+B)+1ADa@Z-<0Xkig~&
zNVjHLlt^n;13m&O#wvDjFG}UpvTOyx@bZ!}_WOGb*~A5|0&?-d_Eph#)P7A(YPc(s
zC|SI`ya8mC%)+h{+CBEiV-1~m!cKdt8o{NPJkzt|!)@Q2K_7Gs4XS6^Rp%Cm^QT6a
z`V2OxEuHgzj1~l1HR9H~y1Ib94D|Gh&^wM29E3lI3<M#Eh(Hp+M}ad46-7|nI8jTF
z-@Wdu*bI-wR#jP5j?I-C`OXqgQK<agyN+IW5eY7QA9S9&qE2IAjnjA!BN^j~(${G;
zB<d>PyMEI%b1f9RP&nnzUGm2!!%lE?lQ`2ZhU0Q{do<8P_bKNtGodp`kcUMeBL@Wp
z_8mP+e^Vz-#R+;P3f@Ypa_BnKC9xCC?yNiBt_Q@HZLHklPU9bNeZL`{adnJT^)-q1
z<wHI*NfFgu8+3|a8Xd@@5);*Rbm-*QXSc#F^HWn(+jHQ+-Su^E2o`uq2;69}C{UW$
zU3MH2jyG3UJl;C=6Z%m?qI1F9BA*p5nVMQqV}POA_vK=;GYwfm<TR?oisMJs3AR~M
zVu*~s)jhq&nPW8&^|0<Nm*P5&@8L*zsYL0AZEzY!LcT@Q&_G4N`pR&G>-6^+=&Xh1
zZU+WN#;UXkxsW=up>{3#(au~`@TY|@=d`XC7-Y%L+}6Ak^5K5s12yNfLo>cTznJSf
zZ#5bEUc4m#6w$vV)`z;?R_I{d(gJy#2A`D#wqx3NrBy6!omJJ)Ot&K}Gt;YPmR+NL
z#L95u^!e1#o95yPB5`ap@^kPsV4rF!C@2*CLHqzX@#Wtx?0soaU;<KnR=6^UZDzAL
zkOXzP3Yc!wLum{DkdcK2PAc)@^J4|qnZ8oF;jgumu3lfW6a8-B<%{RfpU-a~awH`s
zOIF7cWY#9q0s{keCEMD>!XT?aEexmqSn|6Jhd;lj?IW^1{*wvM(W80acRWkw-^C_Y
zo6kP8>S9?i-FB%TvA`<tD<c$}Ro@ltxGb9}?K1a`QMtmtaM-zM1-mH{r+%!g+{IEM
z`NpmjzP@shtm!V}&Awi$LniRm*$SY8+uvWeEj9Ic`R?7LwmfGezrWWjD=URo_R;CJ
zBj~qX6kw52f5uM{(VSTKV$t;=?H~gzy9qS~-WR86U&e1YC8rv57rdaO=sVu8Z^guU
z@6#U3qZIvExuWW;5R+H=-Ywx>`d5YHR+f&&x}3bwryiIU?*%iiXW^jbW?I7}he06q
zoU$tqu>JEnHKkpwoEeVa%_^<d%LTm{m7ip1-y3a+J!fcGwQ4(bOylFI^VOqqHvGH8
zs;c(=dN-`4H>L_j+t#Pp?pC_K{;KMeB77@O<>S}t10ueZof_$sJ}~60XRMq<_-)T1
z5IQ^p={Y#m?d};;E3CwVPElP@^j?~pAaB)H^C#zVJ=cBFQYe-(phA1NxjRoHZ;-f^
zi6Nl>0Ps9Jm$o@kGN8$Vx5^^RIXy1jdKZU8r2P6I%@V8L7vj4IkEi(L#t9l<OG^*9
zv$(hmk<@aSmQ2}yro8`*v-8opIj4^EU7NPU7Rwo&ua>>zw{KnhZ9XyZL;on#1`nDY
zvy^)663e4E3=KR2N5%xRt{pJ7v#q-Fy5rr0PhEd)?x~nkWi2S^H8>S#F#{_`z66`D
z&yPBNQP?V6E398iEO+I0!)UclTdj;+;CePgqTj*Td>4+>F$pa`KS6)csPJ}Uk6-wu
zulKV0^25uTEWJ`oFWrh~xk1<ijW0Cvt*LuFtnW_V*U4u1@`PrJqy~yaea_c+;OeRT
z0^YtnTi13yc4<l0TV8DYY(?K1_~_A(SdOnZJJWTX7j+^Cn&unbmTCU-_qFbMGE`*G
zu;_{S_U2bf#xBCwjbnHOG}?Ddx$^7x`xJ7wh4i1nzC8(*{`D?Eeq%1;Sl_s-lx^Zx
z|NI5c)T{q&f2G;c4zG%wJFh#+R{Aq@+UN+w?dE^DFu?u46XSng7`Ie9OC|l6{jU*2
zBr%`@Z3lvbv`p`y;ZoAmlX9KdQ)~~_`V2Mit$lC(YQoKfH$M)3m~_5Mx%0*Qw`beQ
zbGrz&QqAVZV;=-mfEXeykbk)zyN}y*9Pe@5>gXo&nea3RKi<vlo;LLRyVT<~^>Y31
zjGrZ>EyN8WLmwPGmnoa`AcQuR=eeuI>g2ewW3i!Lh@#E_>Qc$y<`A0%HMU*bHIt;2
zHJhzpy&IwsoBqbGOwd4bst6b{dY*~t)`7RbTlt(i!`61-aX-ZfOw^zK<tdQkPG}ey
zdCw`|q2@#nhfi8M!J+)`L-#(<yV=<S$=>s)?(AZ%BwQorBSz||xeR>j0}n6RL!L5A
z+z=afD<-VuB{ov{Wyaw%y^Ylw_OIL?Y`JnY{+-Car)vR>j{BsXj=9jA^yyr?!BU`Z
zJ}zpT=~Ml}K4#XUDfr?%6P?De&#L&3DhD#>wMA=ShH#uZcpy9}Yzee=i;A^)LfE56
zyFp5@Y@i%HJ^g2~u-REh*iABI^K&)a`A{M-F>Oj}@1f<8(SCb$#K7PR<x8(OSF{=P
zC@IytufM6TIau#^{PvfStX>wzZAsGCEA7PfCz(5@`SwY%M(^*uJM;J#&Z(lJ0-ol;
z&!q$u4U^nNAD!sB{DiThO=>EtPv}{=O66V<NZypQL`UWxtLPo<5(@qdVdXk0y!z=U
z`?7c`OFZ_f@NPBv>mge5e_6&OjvwUjJUordR(`B(_5G_xrT9^?2vr=uTt~Q~fQ2ZJ
z0+sCoZTU0n)>N|5(NRL=5KTN(Sj}XaQ?NyP-(Nqe!zj18?t;FU_1A1x!Ao!V$;c$a
zfum<;z5^*KBGS6xJw7I9#?8T_bdr{XLrKxG>C!91uj02GzRT&h1pi3suf7#8W?c#U
z|6;M5AdrMW5yHnqcN50<qhq$r1gTOD=eP<L4};R8CAkS`*qLWeNzm<ziVm7@W*a97
z@$spF#;jNFQ~rN(>i0%;L=yl9^B+!a6aQsX-YBVOe}3!#){kxDr&Riis#&#}&L3hd
ze}w((sZT5)(CdkhuaDM}o4j~(=*T}E(D-!hB<ZhR!pb_@dQ8eXGBd}tx6RM|`Zgn|
z?dLu})8?LY{b#30`L!(_6U~**#l0MI-ktesttr6DxVX6WX$QaI7a7Tw({6wM9-E3r
z*@GExzP>WIZ~y+)(#WakF1xY(7CGMI$78!4$~iOIU%2l}(P6yZ)^;`-<tN!^x$EYi
z$YTjj$uA>7VRvh9mW#%zL+ypa))Kk?<BQj-J1aeX<+E6$wer(dS$C{mRf#Sl;QdSY
z)G^zy(BmFb*MI0b<XlXFe1pw-CPq2Ie<;@2mj?%k#?}wcPveIU9SRB!O|$*>%1d=F
z^y$-mLgnku#7}A3iXkkzix)3$nVZRLWm5AIO=yCBqA|1Ew;gSaTeobCG`c%K)zp|4
z7%YVk9`7x7|0aE_-}Ptb{nxLBOHEglb#)o1y9>Wb-_a6#&m$n4$Zkum<iF9k$MMJ(
z+KKYw6Fv@~_w=XeU;RfU*PcedmUc;%;wdwPFATRS&W}fQ^~@nGl0~~<^~?WVjBbA6
zoq<HHZ`r?p|B~}pBOeLBsy&g-Wk|u)H*dZyEI<KD`mDVP5BAOg4qu=8s-vs>)SWRU
ziu8r@`SZ*XGV|mFcgxBtqj-L9CnxoebLr|=mw$FaH6XzVBYiP^RP^xq>;Y&9<#w^<
zU%d@mX=F|)DERl~dYe=R?j+2?<dpHI1XZ9jzGobUp$vq_`uakPU0+EYc=h(inWihk
zuOTtGJO;Kf3taesSY8Yt0S_sd=*;CsofH3fm&Ovd-{MSv3OdEX5ay+69T&~b_hTE!
ziFOzq=BMfD+jSLiP!G|fSzDYW9A}S_`>KR>jn!0+Iy?RNa*(j2VXD_PHtu?q7FiLn
zL)3f4kzn{PEwVMY$SAG4-@YV%+ooWCBwA~JZ7PR>j;<ofy{}HNu&45Sfi3$I3kjOC
zXni|7J4?7u&}L>5PO3T|17t4mfdi^D9@j0z6IcNqsO&@NdGYb`o}5vWi%uc<<`zyV
z*F8|#iHV6fFYaRn2kt3yvF^&-kDkCN>gE=g`vs^ev-rFD4_j#*KQZag_T88vYmAJ@
z%p?dq`VG;@KH1+rAAvZ80x|ZUyWKZe!Z&ro<~}}wPMG+X!>uJV7ijt9G9+^vw^m`7
zU60@9Uy{Wi9ijWrJe&p7(bm<i!jJ;n-_$yUov6p`6_tCtE6_;3SmGf<EE0-z#_GB_
z`tOX&gIL1;g9n4D*u~MOI4CH%UrbCFH+mL57OdoCj!~($UJQEVaP$4^%=Su1ywhPZ
z2dl9Cnmzd4Hs={eDUy!~5vJ()_*7!^4QDn7K^R;*jL*I+zCm?FZ$=UG4%bhQ$pgl{
zXDfCzCLGtA;lG}mX%j;=0>KoLI*ZAU+r_`$hXPVK+gFWnc@W+)lnomzTS{+})>P#l
z*LDr4Zj~8+qpWMaHL$f?aKyx{fvcL{r%$;tr+i*a|F{g9xc|_h%;fejG=4KZ7mb_!
zXrOg&qJt>DmR1x0Jd;h>@^Ht#<HrkiS{P1Dhh98$=1g5fL)N%hS^J8-;E%Xgt6f1f
zUGY}&%-YeD<ES)SR;f6xzI>G448=#C{g+HP+rD~4Pr-jAYYhCk3YP@;PKc3=c2;3a
zfxq$T)>{g?XIDv=es8yo;JPdn+b2@O=d!1~0$~b#1)YgcZ+&BZ4(2T?4zHx-MQpe8
z%IsCS)e#0HFyz~tYuyjQsUX5np^M))_anQcBoRcXWrvoxx6GwN`)BSmePL^Vt>deS
zjR3|eK{!u#<sXPocUoU@f(Tp;*SCb*V1lg&guf~pV}k`PiTatFn~Q#H>2Sv>Wo5GN
zLWct=yh4}XJ9e4J>Rh;>hBG?3GQ@Y|>S`O*oab1aCBJ|`75v}AZclg&MQcB@mcq|l
z#>B*^KN9vwK1V%z6x1-fS4c=`PaiwsgLtlv#>$>NACmgos&#L?__a*;H&U0%yd@Kk
zJEY@y2yepmS5F)qlu6Ep1B42LNYubu;g|6qJ{*eH?e;1*uHxQi7!L|qB#?Hbc&5x0
zGZyG8E6rGq*Y10N1Z}rav=EHECil0jOcm92RyR+(MJFaUB5D~TO~<Fc<%~m%S`K{h
zPwp0A4kb#|R8ILA9E}5q4u#a$pXSp{4Ei`o9`w7=yn$IIoTH9=V|EA0SaVW?-^Kzr
zJlJfPy!Q=tbx)o@4<c+4lAqm$5ln~Ai#;Ws3t?0f1PkhXdyTz(eV!7W1!E|XfNApI
zxSE@<C(3w6-RD(pdvbCI2lWnQp(PgYrMeLO-~prW+T;NokZ@e{27^OiNzr&q*hP%$
z9M*sJR6oavAceEDvo}{wqS4gED2$dtfsHci;awW;F*uuFr5|dcN34DEVjaXQbpHU*
zdJw($R_XIeb+qRN|J@USD~V$Usy&jrZgXPSLL$4Wdy!#s{wrw2iMD5<Wfk5dP)1i)
z=SR_Ox)J^wV7q#Accql`2s!!*e4?V!ot+ncVIt;cQM}oyQ#-l+)<xh_SD76ZfeXHU
z*Y3ye0x9V7c*8vLUZ1PMAVXWOscLluxABzo7S39ZAo;`urHHUFs^^2rsi}MWe12cn
zOp$8>PS?X(V%DB%!Fi$o^{;uQEiP;mrV+B$qAfBE!m6C*%cx68d&;$?sSMzjwCDUj
z!ZZR4de^68!|SrA3=<;NsEBH3&VZO<L&2WTpD?s&A9bFCfOe_dA1iEDOH0_iXO5Is
zfpaZ|k1<S6PRjiGnGZ+kN?|uKi{##WUH=5*(^P(9>;X-i9m+bG?%Mh2Og9X+c?8h8
zix*E}iBJ#75?VUCCkSWTwx>IB9u}WMnBLfR(d+32M?{RbojKAWZIGnq|HTA;0$Rd^
z6#`GeWAHZ13i`ODChBit2M-OIv~)?syf=p&mCBe00D0bp33kV6xLOBJA{NRje1!x-
z{{>eAzUSJ06C#vry0w_?SAi=O`}Xc7jjkiC3vvpI%0WkTi5O|gOvKnuojz^S{q`E7
zl;-Y71$Y+EPLRf=TSLUO9BQpDF<A6QCkw1g<l1*Jz+Fvqk~C600+?8DB?h)HoAqd`
zIgbG<1}oY=7#IT+DhdiBC#J&8V=t%U7|t=?Mh+!&$$HT;GCp<ZNVzW=X@Jed-W{1m
z8b9Hu5Hxsm1w>+t6Ekz7V6ETZDPwS$_FU7Nn};G94rmHH^vNJ4P(NUa1np$t+g2r0
zo8VAZ2SWpebP%vi6BI4g+?8)hlT*6P2FFezV7rLN?7+>gl{|2n@#Guf&~St<Ln`-w
zX7;T|HL!}mgI-ja3gF~`R<csNSp0E^@Gd+1MQ+7@-#>y$((YrMf6F()Rraf-(Zx0>
zgZ;ew^?nLv@&LRfX6lCzA4c83uYi*d4h~N4-l_o3cj>Ji&d%tA*-@wQB=@6UixW3j
zU(2FlaOR=FAl7^pO)xY$%bxgjdI`dQqmf^aEbQ|N1qLV14cCZTeSUmLSpGotM42N7
z%;3Li7Q0Pfc<1<JoMp<bxM*|Z5BuNaavAMt5n>Ou&z>D(?F+tlk2rT!H_FgYm8c;r
zi|tw>Y+=dQ-fT7B?*;#qut{$$|9s-mUq*mF#&2d3dUK7UEfbtMnsAn&Ya9s0fUUcj
zZvoGs@n=^)G5_%O`M9F4-Y5(^yPkLViD5PZb5Vh3VIB`T7ron9n_3E=U^*DNprouU
zg6zR|5*cS%n+fK%r)+R(x|>B$Zzm1SI0gWgd&(@D*#0W6sv%^+*q=Y=fA#*yp_Uh~
zFtZot%l*t{W8-%@12c0qZ-k5@Ix{51j<YT4O*gM1HHhIKv|Ca!my>+N0rVowmpTun
zme@g~QZJFKmD^fa-h%a^7nJ29HFL^8vBvBuW+<u&F$~k174v8$?8oi*9XWzU&PA+k
z&*lYcpto`Xa3dljf=-Nu1s{IH434%}O;)rE%ceo8*D=5KVQyK<pXakLR6N&7BOO}%
z&KDhAdgJ%}aD+SxiK3FyHthc}W;5Gy6u6oDNl8s1OoxffqRoZz84;(UDiy!A^`dAg
zVzLxp343tQ(W6o5!8x~U9G~QDu|2!zSZA4!G-9O!Bp68y<VRr`_TRzv8JCTTjm^LM
zar^wj!Yb&f%r{<428HysdJjz(_FHsh>TyUq2ciPw1Xx*Ffwc-VR&rMhyNM}{Qm>`G
z=-^@46uVYIBoBHDONeB++K?PD*QinWFBI4?;d7#GSb~G4vhH}RPwX0s9zxu%^Ys>G
zoeSl@vKk5E1fB4zSpcX}A#a+)j`zWydWyEsKU1WNC)7tUbu|aJhmFtx8Ad-uPgCFi
z=X=-5%BJJ}jia%~za%SdzP-A9qS24U%)(+G1HAN{oEmtCNMw|_10X2GZgS$bHvay(
zA*YV-)QI4ct%YQH`t+#|iy9`E;M(>$Hwb>{N*je{5izME59g5FOGd9|@W!eF^DtVN
zp#9fiB7q!e$fT81*Gj&4@j~Fz+j}5Ln7CyXG1;aiyY!$h_eVCE5|mmP*q`&_gbr>V
znn(<{MHS>338vz*`S*AGvokEzTrz4{X{%UR)~olYL5;r_yE`O%j<FISps80>R7AK!
zU}bMxH-DL!n4s{Qpyh~Kn(NRnNAQSrYfN()&3G6^w|m$+K@SsZr$v5KePOI=75)h3
zRP|vN-R*1rkqr<4`Elyd(<fxhtFkp4)XR^^Q39X{{ZBM8@ypni$hf#VVyJ@FG9FwH
zY)b+R4rP*<drel(@pp`O#*Scg^Al<mE=b431Ot&GDCxoBLx)z7qr~JxUV~5?mQFQU
zT4!BSwur{EW7Dy3pVKwi3DSeZsH?603|ax<12&71;l1ej!ROaZlz0M+Uqwldle&J0
z+k18k(e6QekOMifMQ2Y)sz-PpsSY-zp!YU=?GC%7i!aY(oX2?dZS_VUsp&vB4L);}
zN_u0}dJ97ww;?jbew=x-VSqAwyw8IdcY;QZD4LB>D8zRG!XRsJ0dxrC3zdnHmc@`h
z^~rsnYy)(@bda2&N{}8ECb_~kU!LFWR4E2E#%zqJeUC&-OPJ@adG7<7iF?Gv7z?|h
zD{aAbD3RLKK|EYM>q>8r&OO7aB!Z8i7mx--%kQ2B;@*X2La>b%ac+O@a1w|Q3$)R<
z<o-Gt<h2)E1Koj|OAe*C|7p8zFYG4hD#$w#&~Z%`h|x+=8WlK(I(jjH)g6rwe}RAr
z7`}^y?7G3svizf?VZo9o&;jhA(j+%BGV%i^)WDulqSV#ZPcQ~yY8CNaGwF&t{8>!!
zv790A-o4~hROUc*jC-S<1P6-p`EP!ulY35HD1Y+wX&}*-#dI(<Z*G?z)zxq6DJv+e
zo^zjAibfw^n&>N)COBg?yTcdhhb;sn1@$1BfXFjUm9|Pp(Ne|fpS<AA4AR)%9tZOY
zsXaV9_>pJ;VNU%A==o1K>6!9Pxo?-<IpKgPsi@>xxAT>nt~R2P{7BT28ZerQnynnl
zd_N}UnY#cFX*ig%+e}{*F(HIWP(T;3ga~x}Rbazt<rQHf@^C+hAurYW7YM0ace%WR
za6!UIYZy&^{aPD6{%ss_TVE(c?i9Pvq~j|vboSP+tMX-GUnT+tZ7no~0`!f2Vt`C;
zKRYX+u$N$2)Ya7yLz{TBY<>PrOPR4Ed4H5-WdCdXUMWyYSuqXMk=pw`zJI2-<WKHN
z?0vN#z>NX+=*Y;&;4^6RTtV00d%4T9s;Y|U`ip;W`Vsj2Kpu4!i>}^&d3kP$;-;IK
znIIgkR#i3y^x#1*mG3>lg1gH7<e5XsFzSU*><bD04lR=AULu8h$;iwMiiX{Io=K4?
zJ0*aBS2S<ZR??Vp1|30oiTbR$wuq1OMehVT0ZS3JXnYJg;H8=!jWnQTWj%!%x$Oa4
zFkRF=O&!!8eeT>jqL~imku3P*vJfVn@54Y~U|>jdSK=WJ;qXsXZ5M-_)Mg~R`>_Ik
z&?!yL>Pc66po#E}LrKNttv9!9wdyaEMYq&-vQ2sB!gYimU<`D}^X&^Qb@eJdPVkCq
zZ{YL$iPck-NcL9&Wsr2~S}B@NP9ns-9ty8i#sWr&DnNPeHH?~obi!l>us(6mmXdat
z99mp-P4*btNj%U1=MREocGnpS9&f_kLSXaq@c8@wUGpsKNPqT>P=hGZPEtud-V04|
znUU%kqw}hnkHd_-3x?wfL~(Q!#)9FQs{AKLZsWlR`@%G9Hn-v=q=XZEl?hzRPMp0d
z;rKYP>;pLR!-p#UWj;iJFo)hoV+ZCW;u)8zcF{c;ILk@I8l)zs8~F&KgE0<6T&OWw
zR{F~FxwCpPKqDtW2z<n<NS>4Mutc#6LNXA>4PaU&f~)BtRg-ara?sS7E$4`J{GIm6
zr;i_#U@!zxY>_XyhtCA(1YK&0+LFwZ4Ez?c>l7|cTMYyOJb&I9BoPxuz~gI+`o0X6
zEoSt+v0#nOnCAM&aR##nG4S449ml2w<1Qd>1}?T^a1NEHT{32dXiO4%Ow6G#5q>?S
zEW2>`XY$ral%2gtj?lz5jIMS;bwx*~neOf1NkM^e6Esgw+uI8fJZkS=#hD9o(aFib
zSZ&1E!eHo}JAXdi<q!{P1jn~;sX$XyhV9Ba=nr@=b?D11G>Q@n!ITs+WrGk<N8n=2
z2-|iE7iur}^E2`sWkUZ#5h#Ue6y(_y=b50JkZ<LK87aO)(5?!2^rU32`sRD3A5bH*
zmV8CM_8>hdpM*HdDO(GHd}d&0Z$RcB`nHvBsIZPB=tvUE8OD=DUB=Y8d=_<p`;uOZ
z&mm<}0S9n)S#L&0(t=xgVx8ADN2U9HeshBL$jntS5aYp^It)iOEH&Lp=lJx5O|E4N
z7d3~(!0IYGL}1Tlts?hlAn4uP-j2UHAKSpB&0=U!?&}j%+4;trAucXXijRJ;a2<!5
zAn=WN3k(6`gD88LlSGs;pkV^$>P!%fGM^iSdx|bN5q_YhcKu~Z&}`6j!~;PHHLn`_
zy#uwl8f;$hJMTZ!g%}<s`bh9i{cjrFTLwRmijO}Ho?SE+Z%b;TYIGB9hgryQLb9Hi
z(SwWn4Nt95#o3MjIs=MI%nDUjDgqh+NW7}y+p3h_r(0(m6oep^plIvxEL0bU=_#27
zK1;`Ip{^f=O^Et>3t5W66BHx{F0NPZVmzd$2fUM(bCFoN9&;vTn;SRahr*X6gEIYh
z6ryrh!*FCpTM%S-=W|<g@axShpwK^ii&<`lu^;ym655X>gSpBf<r=hQ>-Nu>k0EF(
zs8cPZA-raW_;flhE=`Qi6BZsuK#+zy2%>zr$HZ*WU20B``MLv>-wO_7v;WyKBB&)L
zS(SC*><WTs4Ff<kpm?yXop^Wwt}Y=u2&0)GRN>OUKyq6CJ*s7B_>YGRDYT`a%_!)O
zB44X5X8Ql-e?CCOVHS4+)b)g2`tAvUpnV~`o?aK8YP$CQ&CZ~r!#wM1Zo^C%pJ%!l
q{LixuL<jnR?PdNybUD}k%niPuzwdYd0G`|<Ii;+r^h)u{t^WpJ5w3gy

literal 14594
zcmb`ucR1Jo|1YjVMna*Gol&HW$Y_vRC_+|bW<~Z$WMz+{WS5=DN*YFGmYoV&Ss^4V
z8F6mU_xpU$xvq1a&+nY;`kg<z-tXk~d_A9!$K!s#-_|qqg1QnlB{L-n2?@1|vb-h<
z$(9)W*+))>pEwyl^duoU{$531R?8!Krr*;;YwYjF+;YOPgqXnm?c`4a>jH0uXgHc^
z&=<FJ(9_EY=4sI%veDK$eTXuPKUcSm{MsW*O$obdE&8*Zu3IR`Id4sGm)qT#s$wb9
zBw2mxkMObWf~uCs-~aNw9DnJ(s^#}oDyik<1(|VAdbQ490md}D<;Xku4{pdb(oWFJ
zk#h>sXma2`>s+K<fwb?A{O7;$R&FyCjBl!Y<vMj;D#c5QPA)|8$e}}FTegv1Il6tr
zL)O}6?`}p$MiIAZUBzg&$gZi+0=H@9#B9RSALbomzqv@q?z=85o;1Vf+n?5D<)p+x
za>>Y(XQUyzOPcsSub!SB*&9C}GAdhG@JxLv$naeFzBoI`t{{2YDKWeFTCe!;nO-Gt
zZ<(8Gzw1nkzq!3$nQuBW=T+I~)_YC2;EVIg!u_7)oVA<T*35Z%WHT!PIvhgr;Y^~I
zw0rk9gzn)x?lelihgX&4&B)yKKfg>WjE>-2%<xAwNj#F2it&P$R$z}Eue5X$KJ!W1
z;i?#O0D~T`s6ejow_nmJFg=}%QOL0I(V5@}CvQrb)knOwDKiyMViGpp&cMJB7#v*c
zxnMc%COvp_nRT{mO^A+;E`seuqQC!E{c?Ni(<&+y+S=Om3=Hq-Lh!s>Wd&kBe$-6Q
z$k-AR5<;TRelJ<_u(p9g)2z?3^mF_E52Np6ALr-S+KKOBU<k^#mRD5N)X*3bh{?|v
z6u<U0u-`uC`EzmI;;Dm2j*R1X4IS$koEEtQX;1Bbwly>&BJ<TNW|H#W$ZYGtprDnX
z<@yB`>+?<1J*C0%MqZ%<l{ZF~yd0%c@Q)f7FJ8R*<+P<GZ$d&s-2M9^&flA_)cx)d
zi_AI~KR*=A{xM%qlqCCnn!1;lSG%zRIcK8(Lk8vGU93kFR0R#5M{L`%E9TCfx2-AC
zl2~K+0Z+a(-wnyo&``UclA!M`NwG08r}5h!3oQ~9G&FoCPo|EJj(W9CIlcVl{9{gG
zrOkJvW^m)T#DK>LwUpa*V=%i<<3ii1wh#9NHoIJoa|XT;+&**GC8?!8U$2z^(4q9p
zuiaErWthV2<jl<v@WzJz_+cY%*L_+`i@x+~TfMo6@U^dE@7}$;{<^>7*E^~Qs;ey`
zMz7rtXDCK;@$islbTsB_WMI*+jJKw|ve+#bQhy*ft^Hc>^D&9)mB`ugpZ%{pD{r`y
z)JHH~oL$by$nae3Pzqq0^yfKn;QDAI-C*12&!4q)bau0_JTd0y478FrKb*g?59ff`
z`i1W;W`9?gR1?KQ*-m)H+`XIbE{5Ns&ph3l;X3uDV(E*)@%_7W3iRs^m-oHH6M64F
zL&;e?_Ebg4!s+od+peE?oB!aZvX!Gw2hz&@QJ@^%Y8Yl+gS!1>sFPDfNl8`Q`$Y0Y
zgrk?>84dcgXV2=Em{X|dln(3*ZWwtxbiEzd^jfw~RK(*k9kcnQSD3kll7eDRUP*y;
z<}W#CV5(5kR!JYhXvgJs|JvH^^Z~=Y9r7CVefAQhq@?49Wt!UBefG<qm&jUAJxl7F
zE}ZJi=BIRfo#Mt?_^nK0lYiT+Tu6*d4}FN`sXD4IEvL(cZv2X7{fCR`k1`MEZb}xe
z=bjcE_cqe*yin;Ht|ZJK$I$ibrnPL!Qgt9LSDc|Zsp+MHe15f&Y?JUe=EMHt+MnJ$
z<kw{&edb*F*x}p$Z7N~!9;ZmJOwjmdR!3dxq~er6dRaWp-d>x#@Y($l2Daw!RHw3B
zzLVFQmKps@jSfq{p8xKgW7#EjV+u~WFBTU$67w`Qy~7U1ve#WX`YAy#i=Wc6^yPqA
zYlOw<*jPuIZI|<Xzq~vTnN+W6XQ>pI*?|-MEIs%dv+F(d$~HEK(r*6lk!;}JziWrV
zbGx@QZe0f{aB4RGuC06vpdmJzH}<jmoK>QtoZQwEUW>WQ$zkb#e#@dKULe-xy*U>r
z=TP$U%YJ)t`@WExp<L&+1Sb6yX?M$-+tf>JIo8bh)VJ&;d&;wxI2+B_kd=duj*iV~
zr#35=Z4-ITnB>SE86{<Xw`9r$(x#t0NXr>W>vLx-ogDel-v85IC{}|4>+0%s3v{@+
zxu^P!E?zu?p4yUlEJLr<vZ^;PFYlLECHl#i8xgcP{M=Tg>DJ8W;)RwDZxONhKrbR9
zLXw-Cd!;p5vU|oY@5vL-KSROq8ymm%TusZ$&yUE=Jg~Yv+xAd4z@OnfIj8(K(t^|K
z$;aO&9<$Z0@DP6OHZz>!J}~s-NnyQ}m}zN`$8>Rn^4YV)DCu#Ox+oJ9)2C0L+MirH
zLbhvv<2_7)*70M^k#EBU18Il3wwIkA9v%*fjLfp_dU`xX-NwcyBq9QBFg7l3`<v(2
zdX>+g-|^64u!^3Z{?`YALgG5SYKfx;ila3&r_8aw-7~+hC@^CE==^m!6f-)?8(w5v
zTP5C}%<o@eJYuBk=_z?$W^OwPI{21{c%HeD22soAeTfndyWX(6_sMQ<{5>crINBmH
zxKqat%f6T2iu7)BvZ}T9!5zEy6HTMGq2Z#DQERbTgR+XsP_*wlx3v3DbF9~c7E8<(
zPfyQhRvrE`z2#=&NyWMwuW3Bx<+o!TcvKqoWU7#JGW(%_nwoOa$owKf$KsWdNghwV
z5!KgcNX)pV`-k<)Y#3BjRAl}xb}s(-98R%^x5#_NX?}jbNS_#2)JZKmvz~4HKckr2
zFI~u)+r=jH;BCN;H6L|#_3vK{97wOA1L2YL^7zC2-(I<bPL`W@NNFE>$o7d>kK)q#
zmn^kaFfQdgInuuTU(Nsjj12zY{!8|ilb&vF;!W{K_@tx~-^cRpVP~hg)j4TC`7vv{
zw>-MGR-Vmkf^&Fegcvpsl}n6l(ZR-le_YDU%-k-TiUUnX{*_J{n@lv~?jn-|yu9-`
zCl9TeN${IdtkT38$62lSC#52JAbNGzi5tH&jjMf{Q%|tubn4A|*1UUnfu2XVIEYrx
z@Zm#Nzx^k?=#yF$2di#Q_PrE6e{$u(uHbF>L8)cSJ~1m=k^+PB?-;;w@$tl=_%&2h
ztSgwWUlz8$?)~(|FeORV-`@sqvjb{+dQ8WT9eW#g7{`1o=?r=8q>E{>u6lLI&|*_=
zp8mLdNoncIV(0mU<BnYT#eF-;nk_~1@&GX$Um8{omi5@x*4DOHcsdU*P+;HBw@01c
zy=TwC!-wmDXy!*%A{4oKV=3aBg6d378lqT<K~wQ(IQ*G)Cs}ZfU&F{yZRj3(1qBT~
zJrzU4<k^*vot-M@&*xNBB+piFp8T)pWl-*Yer>WNW5>m8)kh^IcO6ISiycQ4ot&MC
z`&u6KJ%#BzhEmJCn4@lPY3Vec;y&N!+-4%i%yje$38wy1YS*O8YP}|g*94yxsi_#V
z=jvq6`<50SE-wGLIQkDCKA?shY|7^2tJg_!zr8cRTUuXjZmhlho@jeP``)u>&#ox!
zkt3%gkxi+$6;GP%Db1N3fAC=6U|MHqXWPmHw2v!_0J3x(BFtFL|1a43|MqbHpRQ&~
z`91cMj!ubvzZBYuIy#94eVzLt&8hX%>gu~5KYr{o^@aKt=D^*gr0}jOqQx!#?3aEp
zhqV$rjSV#v+KXPqAt4c8{ddM;ee%5Te1A^fAxTNe3ak(Dr>Ykq<MQw6o-;V#!bvTd
zvl$)x@*4>(FV{{v+IWW>7*qJ>s*BG;(qNI#UoRV5+w;}AZ{LK_aqp9`-_P6WOXL5p
zrbg0d{W#I7?E7D}F5HWYBaqE5R;j?QDeC`fdpSjC&z%d6iX!9S;@W!rrj}`${>|!<
zIgcMQmebwEKrsYP6>XELNmmFbr=r@%As`U0mLzUc-g{Hw03PWSFm7X`GDgT2U@@!2
z?*iKf!;@59)m5^xvbJ1rIo5U5`_Fat6e)!y@oOZF&CLfyMPmVqEnAbB^1F#Ay3kIU
zc_w(*16AJphYjlew(V$aY$Q7H{*&Gh=(rWH|N15??l8~{Oy1bqntiFzDBq|uvAbIr
zb3)P3aR0xjam?wFwSJK)6@cVvS(a_ZJ$B8B$HD>Uzn0lHpmI^KsEaK{4udE2bp-lf
zxtgv0o*}VHtj|_UB4FarzL&1*?x;B{L;6$IEmLmf7j$(S0j2exS=}+B!+ch07mEbN
zD7ii)XD8l!{%_v=MfcQbLuLN|a5#5=Dl3v7iAhR2r=g*dQr6g-`}i@zCPWf`s6~q>
zsG6IbH`*QNtd&&p3k>=2;VhQOwbj;~SOcK!fACmdUI{pnl<N<9Rn?t&d0$piii(b=
zdi^+vo-lR70R!i-!SkK#EOc^WOqv{=!;f=PhbKBR<du}V)`}#UBWvsHQx_Zw3I^N(
z@{(5SZFFO%%&Xisu^xj>_oWlh%-1riz9gK?gSfzHZS6w>0tu4Nyp>){dCPvjPcL4&
z`od61m3I$W&CB4L)+Ix<*R|Q|w6L(S+rfL~#Psl~cUEE`g}3WWU*<Mko}S%?0oiQX
zbe=JmmTU37U_R}T21CV%m&+F@#I!Z8+lKix(P280N^YgtVpJH<VxZM|Xm-!hu{U}F
zL#N(fAyJe5I~a7~&BUCO?SjlZhH_uOAX-jY$^+YRA3NFje7&!o<Sm`_kJ?GE9_^r|
ztxacNY9fBQdxS(xo9B?Y-*=5W)XIkgt$t*lmOG6F`pbTkm6|mCvy;iW%ZDC%X|vs!
z7WP?sVtgm@ept4VZjhCg?qSAZgXdPd*E!{5c}aJ3y>hF1SJ+UesMKDiCLs0v+TUtl
zFhn1DvW_EWd3iuld53N^Q*jEV(u3lWS1bD&*6$?S1`JYYD~o!lKP|M)*n5sNn1=nS
z^Uz>IS&@^IlgDi3vK59)Wm%CRBNgxqx0KYXrV^u^*cMHW_${yd)%EoBDg_#jshKs#
za01i;@&7o{YFhkBfPpjc_!-Jkk?UQT-6vUP$?5#fcgh137-#;&9C#*3H4P0_!8$8@
zQ`Gr*czBMdv`Up<($Y#^@U9@BAJ4zg@4nzraRb{3sGcRjhwD#+VpCH0e%HMY6?3;|
ze}ixWVAaMQy!dCzc#}Eb3Wd3q!zy5cPd)hxP^F+z#RJTQxmM}f>r0<6nTauP*(Ywd
zPch>TDtAlHIg7dBI?%NQw0>f8+rJ#U@70@~>WMobze)PPg?nHl=$}8=L{GcA9@`Zh
z@adBlploP#bk4(vZ!U4L`~2Z#5x09|U%o@21kVvp`)sQ)b7Yk*xtQ9SPBGQ&S4VQJ
z6AMqb+CBENq{8!0*_C=Vu)4Qb`}$H+QF(%h1~7of){;`{`yJrpyQr`K0llUBr8D2$
z+#CjZY<#>@ywK(AFAZ^u(9Y1g-&&T^?b%a<mH^E62(S?({7>o;jI@vt@(QC_&*fPH
zb53qAAf~eCH@_VZPTUY58XA&W?sFdf{=MXC8`~?F$#(>~ZD99}0HWDgUy<2dUr<ev
zx)(&Xci2UW0NSD9;ZK$;oGhN9?skQ_Ju*o6^7-3Lk+1jakLgbBQ&~ktF+HVMORYN6
zZ!6-_V<30z;ZrvsdKc7f_dIP?do1Ut410eo>0gh_yplXTfwXGT-cD$Jzrz3POyYq2
z|NqOm_TSs+hBb*L!}&Ez`JB?0TTr0TN$~C;{UA7JZtonbXK|>Y7Y-s<(d0;12@ef@
z<UTvFL|JPkHV+;UTT&wOuTWIrT%BzV#5}C2$iOwK>=(jYfGOho{NzS4dLP;ej#D>}
zI#2DGZ2f8Z#(57J4;i*fGjh6;O#D@oek|FHTQ1~H%thUFm0w#E-G*2s1SnrrGTRs4
zUfpEB@s|Fmk+(djkTmY1qqp3dYPM2%hqh~){cWl%V*4aLesQ$w^HsO>`d?bwnBB3(
zibV0}E8pFh`dyL+6M|ms$NeZ44!4WjmyJ1DM5lhMEaGY>>2$r6&QEC{%`zqzeK10C
z*WaSxyluuA{Vu(v>~rLfl6SY#MH<<A?pvW)cI*rsKk|O+=Bv@m>I@9d3p}33I1`^j
z=lWk!?(A!auBVrYna?FC7@@@G83>VJF8I{C9{>YJFB8C~d!M7y(UDMI`N{{v5o905
zNN3W?)B<0~fP#H!YD$oBV2kBdGsR^pfO!eQhd)dXCFDHbYJrJ8IzBGyFd)Mys7ny3
zeSmAA3Ebl1zmwZo7#Vp*ZG}WZ^MP<5q^H00+Pgh9I-MU=UP~LuZP=gct(Dl_*w`8%
zT$bH>Ax#9@z(nb`6JPy3a}Gn<bN)S_<MI9Y0VO3RCStEcbqY55%hSu})YPcozklz&
zu`+JgUF2uHF&}S;22(Y@_Pf`ir}Bn`g@pw`-FAWJR)J(RY^vapMLl+V*Y|;S|M>g_
z7}o;8X0M@4Vz#i;sPe~;AJ-!VYAwKAQ8gIfB>Q&e9Ffq_W;lM+uPdRqb-l1@Y&euJ
z^$loaN=nO<!m1Bxo9kBxgKxK3a`W<%sdais)5Q~H;3H;TxKCe4h7vd*K?D3~cmm0d
z{J<JC_4UJmngI~@MJi&<K~0>Puu4|vgZe|icdyyeeZV~2GBW#bp|b&G5h#5t5H;V4
z6DjlaPE`PFSLzr!Ye(h`f87bD9UjIeeSLoITr}H>2O<`f_wL<`V3CN5WRV~ib<cU3
zw@Y2dZ6ndAqB<uD9Q5ngb<4I?=%`~@@b&4drwB##A8h@BB<BT(Whqmv3NJGyouKTR
zZ+wT<=wwbBFK{Eek*205zkoo-jpd*GVeblk|Jnk)W072DzB1gp@$+Ty@UST;_Qa>h
z7mLzEX`O(fz7!Z-axe{gZ*GbMp{1oo%g%1>peCr+InkLz)BoDNI?nM(Shn?x{-4f1
zCLhJ9qZ<Nx>`soGQIpkp&hFv5;J>=mY$x7XdiDHZo3OAjcnV1l74L}?m94YJ&?O0N
z@@>QcxrnB*^!)tkWYL9JZZo)EvXpDyz|F*L4FiK42f0>T?8_Ixe=g$wXHP~q)I7eJ
zgEm(u*{~fy@)!!*J4&v){oeZ2ZZ!z1{S#jbk_LqRxkN<nLa>D%Jk@7^bYEjz+tt;@
zNszfD%a+6y-rQt5oa`}9fAx(ZibveI>Y9Mg1>Fm}o9d)H$wnr}s(oA-&q1QPci(Z^
z<0SjdbqW&`ooDRE){m1<zSFF9J*RMnyP3IuqGLHGCWbay;((BV2)hnBD-d(hL}XWo
z){!9-oqU}I-_6no!bv+HM3AXwZ*E)_<FBn3ySjKHqbSHgm^n^rIVpmynb$03oU&9u
zgq-GTUD$H>{gi^`UPewi=zQ&WGKT3zMUoBK%SpqJP5WQnmuGau&H31!J9qY0hy<kb
zFIM%)$~iCjZk9wHI&`S}m8$??+u}sW8J$$CT8#U4jV5u{qXVbdKifV0olkRCL8C}l
zaOvb}28Bm2azna|ZbXjfoN`-TeO&6L^jXVL=u%4WF)_`;vXA-w+Hbw2sekvo4-%vS
zXHnAi2R*Bl%eS$3<EmV3^4ctJcmq-SLv+t3Z>~wsI(urJT>WD<t*z~P7IQi*JY285
zJUOm#7ag0<aw8^y{xhq$F=^EHnoS*DU%rUA|1>1vL;IsMn4%^h;tuBP7W)$&z^h@;
zI0q@0JolJum!^*QUGPFLCC<D&k>RJR(+n0`m%db<$2<|_)hgVZQ9L<a;IaP4QB2Kj
zqW#gE$+2+Ejut4yS*Mu~LJ(tqjC#TMWV%N@JSZfTjRtlp3MCd77Z-Eyo`R<4ZpQE+
z$ggy4Y^^AE_y$n#z@RX0J0&Yga0PkVm)em(n8zh&4So-tty=Fmv{f;{PoHLeIQ2$x
z?+kjt?%|Qi?qcp})r7D-2uk&D{I(?=HF+nQCdwtVv$XhN$_v}C1oD!7WWzkQDH83l
zQM@}gYvZw4Vb{6a+GEY9BmMoqoE?cN-TAtJlG>`5H~CNQp3VMFqks>a6-_LI>yHN(
zJPnPWuJ!cqe6?w9KKvz%6H8A{8QpMhh-Q#ZCo#OoW@!4cDOmtrfb8QNrCYmrp$#OO
zh$P)Qq;<nHWldZAjP_Jv$+V+q`-k}p$EY*J8hQS`M0k&U;f<>rH@n2_o?maRs?7qm
zPfYybCE84{U@}nQxj5CCHdXj~SNVALxtCa)z{CEHxw&42Io2_KQWP1oA4W6@^>MqJ
zm;MjWS%VUngW#?OzA02!3i-#MsIhdHnX&xPyBIgSPb>t28gSVQa)K>ij$k^GNLLe_
zws6)NRR6tsUirtOY-<=c=NL5B|1qB~!-GJPX1nj~ASro}{x5L;PavgHP!Kdu9$k3Q
zdx09zj+i58)<0Mu%Mxopyqo{v!6CH%D<I#5Cv)ZP*6*z;O=Vq1Pu#_-Ntft#poJ1v
z)aGhe^^a?x^)6nlxteyfKHk_TwCje~WtQ?@U|QQhYlEBW*REYVnnA%?%Ts38BeJov
zT-}+i#zJU17#xHb;hpBIrtv>nyO2;6px8eS!R}`8m%YZ`bfrCn-iTIgF+ciVP`Bt7
zXeVSHm|e{egfj;2<*Hc+?1%c_l(E9dtf+mgq-yo{08^K+-Jc^+Lb?MP^9_sr%~nXC
zStmtg`=y`0xEe_HpY5o-X=!cH0rf!M6ciK)V0njIsl;_khtNIc<+o=jMQ1;Jcnj@*
zTMo@8gkZQSHXkDo2a;2d@kL*JArM~L6(zw>hWXw7YnFN+ea)t^tZtpFm4KqQ#X#I0
z51B(MDxroZOY`5K&B-e$4J~<DGj>E6>=-!fp?mym^!<Zhv?3?EJ??&7Gdn7tdMEfH
z4`w%yloazV%%yLO8lj(UvS3U-`5gB4${WklNBoqq9CQU4)6>%d7@!rPxp|Y`y?3t;
zK)a}S#>wj(`J>ktcesCR^<u5jyQ-i?pKe&TSayHS{vv%4o3#*7PqACcBWq?O2Oc;<
zncdv`=(i=R)Z{V=`?!<0_1NC>Jn<q{YWRoCF_SyR@xwEhW1zjlHo_2pGg(XxBrr9A
z12_a#qi-(@OL*^0{iF*xt0Ewacv8!<H%u|eTmS<fniE9-p$$UH4H2?hE-zD^53?28
z94hd2$*!Go68&9OyuaJ`)~9D$P*-|<ubcv?$=7`uD0KN1Nsh#4k;Dl8PlMwFHf6iN
z<m(+B{HTD9beU-168pZk#G>i=^=~9l#;5yU#?}PWh?+Ojf&sM-*M$@981yO&kocLy
zV5-ERHvtWxf2ydfKV7a6ihnY23slO`2864-o9Yy^RnMV>?rx9OrK#YWNseO$(rMoF
z6bDmY$2{of3Q-B*vnnmJ6K8>XnVy>);x_YD*mK?-%oC#q<Py@0incZbhKtO~Xe{tU
zY1_$jy1M)9%aisRd2ES&JeNGRM=s>{KG9XSFNMYpUB<q3b7l=5?sKO0Oh<2JGOCrz
zko@melHqjg#QGiV-fwIRD-{7s!4Ag=&;q|9w@}h?SRU@G?p5V)``5qLYg7$Wq3#~?
zQ@Rh{l;Bp3cr)how~%I$SC$e4SBNwAJ-tZIxP`F7|FlKodG$E19&vz;oqke5!++JY
zXH>nGX(t&d_5)!vGI+vEW%LTNonm5YHCK)X_AT-K<4nW28L-VIAFF=NUR{3S@9S9K
zc)uG3*n$jxAGMOz!NQJD*_akazkl|JH;qxQ+TxOg+wpp~stE_;j6JmwK&pf9nn8Uu
z`+0m>JJpJ&NKy-db%7F{!H$OAUN1$_9}P>tQr^4LKkL(g-|mfBi#e9mvW})Q4h+n9
z`0(M(4d1IG3Ed?YR4|=P&@SRf2eVo%GXUN<pU4YETUTI*fs)Vt_ByQp?4d=|{mh&k
z<C3n5IQ{2#Ds8DJZrVl#-NmZXF)%3T>n9FqY;7I79upT=t9|KmmV4YoWp(vzbRyVY
zmuLC6TrKZC$j?t6>Km{2^t8}1G^#?Q%Ec%;u}ha-8FZd#&+e1$D6u$)v%G{KR^C|t
zA}!aRnUzJlv{m{<U48w>0t1Okv+eS+(m`3lqT`8YG;*sYI5G#*D$27T(^Z|Sey8k5
zbnz#r%}?h4+tiblV1hS9vv)}=UdYY+fHrhfv`2=G269`cr1rPt`=g{zUDDRRd{^&A
z{JWm3JLRl@Ys<=RaeLjr>}+Jj#&<sTld#4&txK0my_RgyDHfx4U(n<j*SK)uP3-xT
z?+GQcdh_pWOyb!Ir<V`JtaqbdXFPjGkCvX~Q1#e7|CU~rw`5`UhSa}_0=Aor8bBS3
zQ4F+527HT@82I&f{o}bKLCw2T2pfwKMKI*^jjL00QoRnDZwx@qk#?K@)}h310lUX;
zu<G)x@)k}~CpK$qYr)H};?R6cZ9X4D?KKD#R=k5IxES^2N?BPMWZop(r!Q{<(*k1c
z(Ika?0aNVki@ACh&&jNK4{#^zj=*KZKf=JMASE8Zd#=L&ABAYo=}a$j{o82}qvmgK
z13c&I_9lp4jdIAT$`*oi1n^sA)xm`z!AqnL2y+r)3ys)~tsJ;5Od$*$sKo(!d3_!u
zkq}a|)l<%DXxtf?V+Le*Uha1nMd=xN43W++ED*;O(rnzldv_8O!*KR;bNS7M?mp8m
zAOJPYDN*za*ic7P%lO<c!bcU_n2YxPHIWhR1m&j`et(SQSVrEVL7%n5RE&Zge0;$>
z*gS5j#PSwHh^|_nKUX=jqV{x~WOQ=!{_kyR=b?KgoV@vJR_-;p?y;4HNw{P(^Nrjj
zi<6zf5Jj9O+Bvb&$GsM<Vymr0pf5wG5kYf=<JAZQM_ElR3x(jNWLtzF3fcwDiA7p4
zO$=wUjE)e5B=op@kF1&Y%($_6eA@zMfQ29gE0{B&bS5SyYOnYO4js~k;r4fJIR^x=
z%GMA&gt7HmGG)`8dDasp_mKZ=B~{1C)d_Awa0P*LIm;;X^5x5aMES*!=e(4xX`wO+
za*VsZ)}Rk6AMgr?Z-P3^GJEyr3B1&|{-n*xlBLC$4QOc>#@oinPi;ybJb2JcX|8wW
z>Dr&4uV=f#f=B8jJ2|yk;~O-2<3FspQAtCaC@wCJKi-l>H+@g~$i31*@Gi0|%Rl?o
zHYVmEPEjZHoE?YA*O%Eq@ojx?>DL)I94aBl#5dNfX^im?I;?hv()2iYh#k%(>uj?W
zNsyaok!sI8Kf8WJF+<pVb0Bdw{&{wiAs<^bpE;R%8zUQ=x|350cvbz3TX1tk7JDZ?
zs>5>9Aa|qk?IiF+Lx%g#>i7E^&;Rm}NtW^4UDy?#Z4E;-7GgF~_O-9iI;4#Pru$#V
zpFj0C4#^DE_vKeECXgRq^xH!nT8!$t0(%QVlhu<R=v?pXBUxS!-DXK#T<a8z#1W1{
z!l$F^=8dP-8(v~+BQP{vetc4y^`Lf}?iMD@G~g$RewRI%QG^WuLy(J$3s>cUPzbFO
zgan2WAE9iIjg7$!uK!i1_T?RAW-Q-%5!)|<Kn$==;4Bkz@6*>mr7AOy>0Y~bnCJlC
zlN_=^5+E<#y?a+~xN?hK|EqhKjBkE+|8Q1=o;a}H-W3BK%y#eZ@(^7iB4Q_K{td6i
ziML?WOXwpveoup_8gW1az_SBKJ&<;{S2nB}agoA7m%=Tgiin^PoA)nS``0t#l^H3V
z0TX`DESea>YFp4XhNU;XERQ~ayx>&{@9s7Vh~q$8iZA*>-;LSIKBuO`RlS0Y;ZdPm
zaf()G!Wj>9iHqO=`c<E}B$%CYMKv{xcuYdV`xDE3+up#Z=z&Uykj^Jby&5EyU^%>H
z-~6LQ^4wvtb$+ag<)R-EGLg|QvmsE|ahC~N)J1OIS@7`KG+UaS1}T(0;m#a8#t`46
z0PgMa>+O!EC`MY?5~e6&><>lk<l?D0s7k9FeQ{+rpTnv*mI{d?W|elM#g~d0+o5!7
zp?s<$3IwUpVs@ai&$a8aR|91TWEJQqdX-+s3iQi{JI|lWGHZzPYOpyf8RW`_GLzZ(
zGXlS2Hyxc{&CnH0Mo!1X@n%n#=!TIbR|X~~V)!jlGjb3<=DGWVj{udbxbp}9{&eSb
zq_hPQ$Txg>7tlQB!Gp##G$-gqUaRP6YtOysOZzobJ_y$Z4`=@7))w3gd0|%*B1^zN
z=K()4?u32T|11~@J>15$jpDXwzyvHMPS;mC>FNH}>tjby0e9iZz_WW0<K>HPi?kQO
z6*Le|QR+nmCM<yUIGsPunL$|30E}Cl>Ai&E8y_Fv0MO?FkgycxbW((Dw=w`J;vZ=U
z0njTvg10KO&@9SX+>#9mYGphv2P;A(-O$)><7fwA9-W=F+TsXG?6cN8IDGv$Qh9xj
z4Xn_Vgbq+@;<UCTHDIEJr1L`<j_LXO90!{iQ8*~LL)>cWi5t6FSqt2sk=YJZ{6kSz
z=Endr*|fE-drEj9I)0JX4^Z|?Za@Vn01hJ}f;`PpD2V9fO)<y$rmi2)&Cg$K8GtuW
z)plYzx%KhZe<B`4c@n@42v}z6)5V?pj^4ttpJFX|i`Lo$gp;G5BKMC{Kqo?qn}~$+
zwDjXM@FxsVv~Nxg1UK%Cj+#Scuf-=Jt@T~6q1NPhbhcA$Ct1Mkp^j|tvWXr%2!5O3
zIZrRYp2uD~jYLZPU=?@>FWjiG%bmeF*D&Nuh7yfp)PBPB1|$+fs(>%m!ykpFiH*ns
z!bD&V$Qxf?IkG(>!Y?@(hYXY$X#*Ms4XduFtbh?zp?bR4UvX^CZf-;)vlx5-KBFe7
z)8d4Nk=G;-ZW<%L&ugkc^}$ij1NE{4K5N&I{87eZzF26LK4Yrczy!)yfng6$>O56Y
zj)DFrM(<C_t5=DjiFYuLyq15qAUD&~EwmmdyLh@?4D1%B2AU?S59A27g(yJMdT=`^
zz!}V-|4va4D`wC|H=IEmKpG(iV_jZbdoU;fLM|?f1wuUGcRw2?f{Cz9i01+Zz`Qwm
zn}(fDg#9^ijw#Fo@4ri3J6XhuFpR`=8iQ(-?A+TO?7v{@v?B`#Y}-~)KJay4{kS>p
z30lIk)zM+ZdA5K#T2-x!s^N~7`8ZlVhs=|fcHPkR3)<Ra(v^Gm?8(3Z>&2Esf5C)Z
zKw4&7M#qGKqO{gi`BU(q2}2BP)91a=LiCI2X*(m&F$RctS?(2N5MGEoBSd46o>b58
zOqd`K#IG@9PT0q?QE6!3gFIR4Hlx35|M53397g~M#sX+g5+P3n6MnFEAkum@G%|AR
zd9QuVjN8KcLfe|U-sexB0&68I01BD3wL`<ghBww`m;0-L-2TduDr0Rg&<DXqep>lk
ze^9Hdq@;wmI_e{n?`mfnvPh%o8-yuMvXfaXS7p}5y&fA<1235gt{3=F<!qCIs;X)%
zdq93}?l)}dqw{GeaWLL42La_rLVZo?TscoUDhR#`#=C+1qo#quIpi-%ps8eKW(Go_
znqn#PgUCT_8D`Ca0|$U@2)_-UY{Y&^O-w^4%Z0*w0+&PpjRr~Gke9iNunT8#^>fq`
zA{^_#YG?=xl$BKtcEHN#LZ2*bM5%rOJtJcY!bJB4^^gpDs1VL5^XFTDn$PNbf(k2<
zh=l_Q&EM42^u^eHgFm>rr|vS<Dt^glu;EmDg=()HEyAoq%n2^$cL3qfXcAb@BFh$L
zqy>lo%gr@9q#sq#o{C*2HFffJ-y%7uqN-Xu;}#2nB9LOwB4&qrPAodm$DH%*AWhv>
zH>ForRx1Af?uAvThsPV9^T1l0|C=+wHDqCPAF2Aog>!DBdg<C)*7N80rGLH%0EqyV
zWu{jc0#oHiwv@&Gh3u>cvT~BaC>ar#i3?~3kRBmwJ<%zUGvzpfX^E_9z`J*HV9{tT
zuGs<-u1Eom0u7{tAYpc(o!qYdv*Qf_JfYPCY8bkH);7O-)fwrFVvjjffQD{7)nr#;
z+H4I{9R;xLFI>0)ms~qmC<gLBk_4cw$n_outxDuPbGyiQ)8}B!g^y1!OS5^7(gAJv
z6r1h9Dqv*f>)6H)?ih^#DkhW~B+UR_b~1?s0_gu>ee%|8>4yR?Ltw-mG;Eo8bbuMB
zwVCqZ+T+R?q@JU9lqLXrUc1kJ&wKI|t(7>s#640=^u58%AU+Au1wtxn^NAlm<j2v?
zpSySOMxIOtCj}m8>CH7a0zinVWf+v(do?U1sg9h8CZvd~t@}}2r*WoM;1GMvlyxD5
z(T-i0+`kKC;sd0$CJ>evKVMBDgkkyhnYEV^RY=>Z&141v9U`GY41O#mjv0-dX?8-k
zHF#4Bt{j`3+<dIdaOCyGG5YL-<KyF{NHhSIL#;jx*o$~(jIK)0(Dfp~CD61Nu(U8}
z&Y5{Rh(t=n5Wvg+)g&)a3NgNsc!Iq(#VWvoqY2v;+?ar*2!SX?vwfD<d9x0*M6hZC
zv;RZ8>-HLF0xuq*go-_=p~G*3<VWS_AeQ$#>;i&<gF6e2tEI}1EiE%L^Hl|KaB^-b
z?^Q#$Bo;uo)KV>WfO4q;{S}>nM^ZB3#f#&xYg%GB&wx%I#Z*EOfs-XbdH^P*2&Ik>
zw6Yte|KbTn8of~gXOSQs=!zyaw}Wa$zG8d435<6WjVLW?o@A^iitWT%oNa_Pi8vXG
z#i@o`BCz%#8>hegeG!UGl!Jd*7^Tl#xbXxl&o{IS%;5dNIhkn8RTd}Z9_`;kczqy^
zXNF(Va0<}@!yqPiV8C;n6{|!+L4iaf5d%AW_9ON)p~dp|>|h(zk{82a0+1h(Uk+@(
z`xAT7oFW}RR1>@=BsATb6Fu8T*sWV7X^_Cib70r&r<UM{;)0mmWC4E512<@pJ#QG1
z?R*xqdOb2KDjU&LyRXkE@Y5bUBB;Z;#1>?=6AN-a7a2DIIHG_MO_P_GH`6^P%2BoQ
zoyBK!ZBW#-mYf)kNUSqSx#SKMZ{<C5BpSXjOhEo4M{;0_c`5Bq7hag6zu){$iWtrS
zPHjhBe`rCnY0A@n(*43Q10Q=g=`ufXV^-Lzodd8uBsA0n`whhR0BCUz0k%1O>H?tg
z`pN=yk#_0(w46#n#YDsjFjWcd`5%Ez7us(&lmyTpB=m6jjwZE$FHf=_d`R$HpTASa
zomuDhBCas!A%iHv6zjQ=7;ZC;aY9VN#m*`ykOI;Y6HOySaT4bh>;(n&4ennKdVgWn
zY8WyCy0f}l2(`It^@ji$-Jh=NjWL7ebL4coB4qP>bPA1H03U%0nwwRBRxS_3=(*6U
zAL-K6(vnA=T7S;FH{iIn9;Jg;b#imv8Kv$q@hCWP+SFCfkjsfxnFIJ9ZRWaH!WM1q
z4l(MC4rLFI<B)yE1F08~%L@W-zf!D=hlBBflLe$d?6-rh9o!CK8-hbIN8(?!v`N+5
zq{6|25-4cpw)+|G*i8=Cg#heT7OiT$dyokN;7s)xrU3WOaIYm(QB&iA1N*N6JTmuf
z$?J@%mlM))L=aX`u(@2m5-D~ou@VUPxa-9zetG3P+%ctS-_Q)h(!V*qY;0=Ez@`Bi
zVM@tas%42L(NY6qBFUzqtE)_qfvC?MY|%Rqdwg#`fB53yD_2`Xp?C)eT12Yeiw&H<
zKwno*vx=SeT%T(oR2JjS)x8MWN##)O#5*qVi{bp5U?uUa_m3Dqb<Zc0E@hnqI3h^g
zk?M63oCRbd2i^5G$?1+mgDSS}<OP9J02{hFXyfecOprl<Lfj<!q7d{|@g(Y;&VMv0
zsPn|D2(Swf*+L!#3UAy%aUi**2$>TbeBGSR&}?fERLC|ppD!5~;Ituz_7E>9RIGeY
z!Y#qieTo0WR~$wENK9mgECn4McIpRrt%HH&`<b>$(gfGMg#q8fm*!2pB>@P3jDWET
zOL1D-7Qov3WjtaQ7hV3LpxmUYCMKM4!$$B49%11-DD{6RO``u{`CG(-!ShA;LxL6a
z3Mn)I(KDtTUq<GBf#L{z{JJuDFcooUXJl+9SPye(X%o{K9LA(rHzYdRVP3kLg3eqd
zgiRSyCPBTDK*q!R?U;iEGywX2ga<?b&IG<om90MfDBPY@4rOxrpifdul0@h8YX&$8
zh|{<=3ZCJXH{*h^g)kFX`e56?0vGEZ(x}fi@E!@!?dT{B90k)sA2s9*f;9LT2`put
zc2F#-oZVcDjgEgkiCnez6CF4ZfMLF9G{^*=!v{>t^b?>Qq6`mV`wsB%5HFKdnvtqj
z`}`$Z|BIRDy)c;@#+*8EUzKJQbWY|wNBM|oG6WBL1&w`DE>_Cd?sy@QJJ@^x@<{It
zuei9*xR&3-tnX$LCJypJcrPPvdIwkS`B>o?BDpbwlmWq%*1t<sKSClmU<VCo3<tvR
z9uU^HAPfP$WPWw3a4PxGbtE2DfjEexNJM|IZ+IEy$3J}rD+4S&{k5!Vc>zk<AAG`9
zmHv2Mo~EwuMADZZa~^UwHCb&Z*Yed(@(s!fziQ%va!C6DG^jf<F>mm87r&4Ya{Gj7
zWMBQ)4Ns1aydCl;A<F^iMB*3{84O}c@b&{X6zo3~dwnM@?Yy~p=Knazd3p0hcEzD;
zg$xb{&T&gd^hsvTy%{$l9^eI_cF;1+@9Y@eJ9{5KdlvW7d3-gh1VrsU=G9UZkGvUg
z)5{%qfl+-oH;8bxo?d8P-t=JgsRLvm5r(S^V<2cfx(o1m_w#Ekcx|aeTK_b{2S=fn
zA~JLfsR*b?#G5)$;_w!Z+}R25lwC!DdPV3iRck+`fi{eNR(OEg2ps~IL=aB!hemM8
z)^|ZvXKV^1h$OBz-T<=eeNKG~n&+x`{BF7JQS^<T#M@Mm9ox$tjO(LVn~2Y$a)`=L
z%qB#5CnpiWwF9S4r6I8<i^ElF*077s=Z}fz#ft<mAe45zsNvPHn}|InD<K67;sL+t
z7ewee46BZG1*-qP0|1i`0X3jV6r0vj;_IHEY9_yy37{e|(s|XB8}Z)KnIC6wa57Ud
z$)BRZ$s^>?QcH6S3)JC@jn%0$rowd>#$BY~XFz_3?uK^^C_;+`IQ^A%Mi_`!OJ452
mj9A=1j{pBfta^Tv#8qC4g7Og81m0~VQBhEre=2vy|Gxq8G$#1~

diff --git a/master/_images/sphx_glr_example_pandas_train_test_001.png b/master/_images/sphx_glr_example_pandas_train_test_001.png
index a970657437366ac15fc1eca7308b4f2dd0058507..02ffe04388b46adb4ea1d6117a2ab4da53a5d6bd 100644
GIT binary patch
literal 44871
zcmdqJcRbgB|3CWH&WMsSinOe35gDbBk)5o_rtFasX;H~2p=5=uBxI$m%!~*j*@Vj8
zn{$7CuIpUi-|d|9JHOxe{CWPku1}v&yx*_!d_Erc^>{u~RhFY7XCx<)NL2C{&a0D1
zn{JaxWW77K;lG@0?EHm)h&sz?J8RgPJG&Y=nvpIVIon&?Ia^y9AG%@Y=wxMQE68(_
zhyTPOOJ`?$Cox`LoB#X(kDa3h?+>HtLVU<h`wKcwBoeg|@jtR;sU#~B>0P$``LmjC
zk)vJi`kE~(B~uN2qqVnp#IfH!9Cvv9y5%*iWV6BOdAgG)xgDy7vh=fNBZk5fLkFjG
zZiQR@JUsqr-}tAYz(a{2j;sjOb@s4)DgW~BP_6hx+jRpz*LsNsaaSHe>fz3t?-bPb
zZe_hsBfY~&pibc<ne-0)OaGtXEvm#HO3N>7V*lqajgOM;CH{_&KbS;L{QL`gd6Ex)
zjaAA2=<R=g>$vQmGyi-N?U9|Yh`+i2=;{C8KjyqB9RtH7g@pdB2FcnieM?Jr&QQN^
z-@YX^^RfE9c>a9<y=_#Sf6RBNCE`Jr9|_~R;-`=hXI2}gqJR0qg&oaFD&EtwJB|A1
z$2zKJ|1`T#_MVj&k$LEN7LSs7YHgusU~sS^_S}OHo=elTEp&VKl=?FW`X!r5zj)`h
zGS6>P>3@K>gknEW2L3*&hn=;C4{znm;lqdT5`XQz?x}y}%F_DEgyKi}K`morV|kIj
z6h##kHg4{;*AZ&hB*yuT%*=dzd}Ma#$v&LEh;Mzrh#yZ;jorjsw0OACW3GmutZpna
zzs_k~D@}K6$*Z&X7W*Ibl2lVPedQ9%%g=XvEnE>77teVAzO=3Fd4uE%rHmT+_tqnF
zfi(2=4<A3K5Ed5xo~(9Hq}%OCgUeo-zON!Z<xNe&G2ZK<4ez&8Grx8ku0JIvW_<m+
z;-?#9J18gwIi2SIUh2NJV&m!Mm6@3tYh2+sJKh!X<;#V5$5gE}=>+vmiw0)m2RJXi
zIiGW_g=E=NBr!MAHayy%<^1~_nVz2B8?CgONKq&9Y!<C|`qfSi5^@O&p}5F6gF=rJ
zdbua6-s|OB$6on#<J75BuNo~ay9-1&7Q8o}bC18f@<}~bdh6)usO7g;WRLf83lwx0
zdENY(XBU$;e)itBzQI8=x9GV8bv_jP`^UxtLPO~+8lsPJa|fhoBsFDQX&g9sFf%JF
z-lijWw7c-MUZF=}w9MmueG?NwIXNfay?f_8-WiU^L;Cja%4tSn>lY>kMMXthwrv~A
zqtN;2%yH}1Eo~j0zVY$;FHesIc>KJ3?_OqJ9vcS-N5{vt<+*sr!J6OBVrB2S^>V-E
z+UVgr?d|P*EA|`ynke3=&n+Gu8{0`uopinJ9d2!Mc_e-KN70|=BtfInEu?GTlZ_Vj
zYQ*cACARa;xUGz&7t>342;rhmiHh=jF4{LWHMvdplAS$!Rx?@k4mlm?WcSSpvz1TH
zBXcM8KhQ`?Nxe-=b8vDpS)A-EFE9V~Ny>+vqD~QYBjk9G_lAUo$7~5+?ya@uOzU=m
zT`ZDHGech~uCz_|mE}7OY&(5p^y=4_CuP$){7P$ii|fX_J~0TG?g(H!Y21-(qx6FB
z&PZE&*458<#Yc*~R)l#ACO6^Q3O3ePaDBTO7{V>olbZjy4FrfRPF1nAe{eX0Rcids
zY-4eN`lIt7>E^_X;lehI=H}+U=SLc(HeTbwlMYCFzQAsn`guKFElv02;Uh<Ma;+K0
zI`W9+^5f@E#+H@&4!eQBe`~&{XasoN#5VcenJ<W2d8HB~`PQ&F`VlQ#|HOpKbDk^H
zo%?IUxXT(Ef}R`RVsLVDy0tPULafAD4+*?ep%A^CYjk!6Q`_(sm0k+%WMg9sWRa?G
zP1F0+Q!GWo{r4t3H;7XR<-|VrtqJ9F8t)XnwXx<}>Cf=Y;<RIy<Di1Ny1LuUSCLHY
zsP^{o4<AkyEe$a(&y7@N7?nK?3L5UPD?Y}@_vGAzU5&|VN6MDZ`BECSyt(B3r^yGE
zA~-s_9($|$&gSiwt#27NZ=<~WBf|(=!51IF)Oy-;k#9e*f$8+GYWG~b<=H=VFZk~T
z9y~qWs%PiWSGu*#fo0!5Umu^Xcw&Q(d5f;%{bBQUWZ0^!?<roJXF@@fPSe_)Vg8kI
z+pYr_u3kM9b8DGHUS1x%%3s*#r~T@pL;O)u35lqZk~2TEt;#H7yu!xZs=vp}QS%rS
ze4&<FExEgu!rsAwtfZyoWWx{H07jX?n$U25lWp5~9q7f9tgNnngs)dC^l&l%`cejU
z5+x!5Yii5ZtqiRPB|T4}3K&oJe)eaUc=_<*j<6FtrBee{R4XzM1upKEX7`I@mkT_g
zMl!1kJcvL08Khr+b15|uPrNd@6HAWnOnc$YyvAw#>Q+^h2P<u1$6ptYB}rvnj1cCy
zJ&mQs#68iH`aDiOC8YK_?{Ir-0FwykZAUD6{aVvpi}B8UKdsbD%kyKYKYsjR$I`78
zWG*yFUtC;dzjNmfzEC7YQ9(f++b!1l@AqoU_W1dUo^(>p#`=O(+51qg%bd5H5-&0x
zlaT02_C(ctAF6Wk;w8?@@7~UiwhK0_J<(&@w0U#r%ahj`qov#?&p6oItLo-jryONt
zV+#H1&o3cyB}FaXxBt(d6k-=@AI~Zr`<YXl`xS4UGdeojcQ|~-FusvJ&#p(9Ga@?r
z-Rs0eM()Rt9;tP9c7~u7mt~9wNv*4r%w~T#w&y-?PedVRZ*6U@DSt#8V89~bk!@^Y
zk#zLvQBysUW^5HYMMcFypZx}^&SFtUOTot!_ny(oH1;vlyOgT+b^g{{y*y>Ikiwe_
zoOs?-Q&WD=o;?$={=paTC}P*GiGmmqMt$IvDp|-vkGE9YyxB?Das92TXjAVChj0&+
z2TpyP6rXXsIZJQb^)ZKd?7NT1(uXA^C2?V83wo$*tj?%xRV~_BzZpMf-SzQg)u42U
zjKG!?<>!d1V{2<ml>IOCxwuqbRKAmp-JImCtgOi`*JOj3ZL(LpTzh2H_U2l*2M;eR
zD=U*$(c~<1eu~#7CL}a|a<wD2ptQ8KTc6MV=@Z`l<vV$<d_2FgzLxJg!R$KGJ&09v
z^!V}V)tQDY*9ik`b`si}?_9Pn_riErJUS!llf&qbOlucvCgHko8X=*Dg%TS}4c?Ru
z+}zw&jd5EqqYY)H1?92|3x|DjooGVwf2W@>ZEbzxNn_T27M6yUiDEQ}oW|wwvgxWI
z7B=+oOp_|c`H6TlX1RuFiL5!zl$L8JxVi86$5yT^OgxK@R?ag2>ObG%OT~nCtP~;4
ztydUs<LK=CEA+BK-vgFge3T}CoyBggPRSW~&0RvLaJ1>Dk2)={sAz&~eG$I4;=+fw
zB*aJ+Az~+Nor*iXfB*iICr_@jFAM}pspnj~gdM!ubaWWY<6F9+RGEX*!uXZ%*G{7O
z)Hys{Nz3c5xw&2P4(mBz`a)Oy$Mb4?eJL4?wG*{cty<rPH)hT^t4Q_z{(T`yIm*A^
z@@F>3{QSIVCF^+|0wbDw8ll^U?K&t%vwQdF!P@8fo=XDwlwYWqqd#*}FXXub33PUK
zVg2*z8syntDH|c}+_|%}z`c3$gVXS?Cl1}0@o0$J!NbE-FkQp-RzE-Z`}gnG9WNzr
z3P%Xr7^7v1Mt_=5R(#Gw#%JGq27oAlxyN&cDkUYwx13W{G#qW4vEIbl!QpdDODJxX
zzQ*)YyllXMJe@S%9BX+kA96Z!=b}s6+Vlrc-zcxI--V)1z^dMR)Feuq9n{p@`4|s-
zX`ZbS59~Asyf{h4C{*(E=Sx32u6{I?p|24E^LlP>r^d&}Pn|wJoFDM&)uFh!xW$<U
zsXPAV<G5E!dcH#qa{@tvhK7dmgJEG|-{0yqPl}!CDe{t6Rvv0I^xi5E`1Q#X8jpo>
z8@CC8ps3;Kn>-zvSzaqQY<r4qvJavYmF2&xi1nEJE8B4J;6bv<$;mId;v<<i{xlsk
zHZe)jOi|B0wP(+s_Z4{dM~6w|XDdbR8XFtm;`!%XZ!^EShV>^h8#&m@r1`z@J}Jg)
zKD<##fT$&pDO%GF8&Fvm$MSl9<k~P1{S9qlpybZx9~~V`tve|xO-EYO>cRyd0LNO4
zkloZYpXlz)cM7PNspZUCSBsa;L|>~6JgA5tb8vP(pP){3V}`%goY(EHXo=2|Z(1^-
zp=D;S$1_oh5_5SW<t@p~%&e`ei%Tv=yKuBFd#0YEDVocbw;+}DqIhNe^foHS^3?RA
zn)^FxlQmQJT+4aco@K$br?GSgC1v^7uK^Q9D}hAsA^wR*+l<X5-4+^wy$S?Z@zG`E
zH##!W(W<MfXU96)RF6A(r|V=9P1z!~tg~~GtE#4^CV4nBdGLn*V7B<k%{Of$#B+R?
zMf&AB9!q0(2gK*e)80#OG^_1fY_+5|Rt1?vY<~gRlvh^z0z8pSBZ5v`x06ro43<$*
zQMobRDevi7pvK#~G&4+<eonSQd1#W+ROPsC&NUgdw8dY~R2GXjR-_EP*IZiPU8(&R
zdoG~HpPv8ja|4e@-Gw)^90n?_4a;@hZ{FMt&^mqb5b<X48p1{FqK$ME6lydg-Y(?z
z+&XgdWT>I{T1#@?a3kx-x)=44qUW%f9p7?Y4=?C{ry46Ac&=Hra3f!*_?E=U>#e_r
zW4wjMWSI~CotvLGM?dQ<_LgceIFqA$5k*i3n1*Qb*iJPrA0$%i21ZMLD2O$hAr9F4
z@Zm%2l~_Qf6S~<C0au86Q&3P~+43e-?Isx@y5{R@Ht}{oPa|A+mea8ErcIk1*OFu&
zzE}3W?NhCY6^^eTB5={=bc1&0&!0b+S5-0j7KuwpsHL`yc719~j!sZ~Zk~F2W;Iog
zHV!o-uKADld(%ga(_1O_y;4u1yj;}!R_7UT3?nlc8QGc2OIbnl(q{9mgF=*@nI>|?
z!^A#g;JbPzTEYZfzM`We0$io;w`aOic-glnPu;|S239=YzhBYLQdxjD@~zFofB-YV
z64^lmx4x|h4jg!adIAhCn{M{5<F!03TKi`_gsehCosot4`JvUN8QGP-U)9gr6Tfg+
zf1+j<|FXWe5>PWY+D<E@<`}88+PEfpwS)w4Ne=*nA|^YmIh1Hp6^KPXy)+!d)Utr>
zwTFo*VoYHQ;5Yc?%bLF}sqG(KxV)BU`8_1~_-F!T%iii_If{#%9FgRD+qoX*cP>WK
z>)Rjax=2yMj_ZH^{Bac7qsDWA`W2sh@%l0sI&s13bggJn_nS*xWv{;hR8=%KJ|$L?
z+mx{;CtY`UH_-`2qxXLYaUr2sUId`SR%$c28Gu&Y7<+EJ<m%)(U}mcy8LS<7cEiBE
zMra<s(IK604GLf4y)+fxbS)?>d=wO9u{1qsDa82k#@NdNvGk&4POi)Ej<U1wz;$0X
z^a?c*6%-UCz$~Ep-Iy4bd)uiKueH3<FTNGk^~p89(UK@IB)2~a&$YA6sR9E7zo%Xf
zT3cBl)(ofC+ZPrqk83D+3vMSQ9P9e%B6w61*eppsLE*fd98-;z5aYfR+9lXBR^W2@
ztt4q5@<?&_;McE@YNZ<lgoiT#xyS3}*$Q3z{_yj??HUORboDV(4FC{db8R|E1gjwG
z4gee_gTQ^<>&)PD!~Wr6IemToJ3rT&6nUpmo|<vD2A)$g056NvwY({p)1%dzy6~K~
zQPP)S9oOTD5cC184+Y^F(c2Cjux9W0;9&B%ExkcSYVABO-m2-fe9@S_i;EflKtOP`
z<r#dO6CgN50tJ-UtYc3vz7&c@z4Bb_dq4m<stxn;VQ2^ZSVu2~tV*w^7y72Aa;d4Q
zS$7wN-@bj@dSiC4=1@-)TK^Z~F9GB6U9~fZpI5a2Z}?KP)PwSfI`nTLnAN_0`}99L
zKf=y_1e(;CdO7`Y&pR5GOFQ-5Pn=I_O4Adm`1w;;UP-Agrw+7#F9ih!frEhXYNAfN
zhD1x;<jl>@RX}?T=_`A{Wl<klElj24;qg)6+IRMfmX=FuDVnkxrP#|{69qGY{o~{B
zw6(R_B_$<mh~AKx_T?*TatLreM@#DE%WvPl4FNCW;4PdDcKr2a_qp)5%_XKe&VSUm
zs#XTEq_ZoGH~%~!bxZ$GWBdd4Ok<gfni_?v*;!4Hn0TOtTGivnkIOQMyJhtE_s5|z
zap6tYpwc=;g&j$<y8bigdeY+3l9}H6?`+QerMWi4P_z&N^!h!2{(SmRQVb`T*4r-*
zHBV4EFS!qeTkK_GVv+%h=GM(ty|vi4Jzl_jZSHwfiUwQ7=g()s@iiEQEMJ{4DEN_F
z=i1|ygo>z<nVqf4c=Fn7>yBLRHmj@5T^oME{Ao4VwtLY5h)cHW{E#_~W;;ypL_8}X
z;RNL<F~5X_1fYdCU~w{{9<!FY9>aCv@g(9Gr@QQQxOLv22TlnzHZqF+ke|<CS2(AY
zn_@5Eg{49M>N~b~+wzTx?li~<oY**i0Fuxq(kK~)zWiQ1aLWD_;QZy}8|VWXDY`jt
zjKPp+*Z0l3p`^du#O^xY$*&u&r&qMTd_BRcIngg5AfUgu_toOc3fDy8LbT&RWq@>!
z-ovH~3G7=7#%|L{KUBAAjFTytsXu*QR@VH}CaoQ;@Bh&ZYsh5OEKw@xxU@pt|K6ur
z3U-1_7)Pn7r1T3o#td*dQqt?C;jN{F*AYkWvu}RA9ZF+O2E-kd)H~0f(b^TS3CDO&
z%gc+%?@-(JAN|pk`F|=dez{blawfgaX`qq{g*y7L7bhW1PA%@0VU?6*c6D{Ft5eG2
zRoYX+W{)3%Sf%ihfsiNNDBIp$iAVE2*5$v4b!RUvgcOp%u0sK`8n0ch=ji6z_1J#c
zZsu-E&VRRDC;El7{~?DkwnUmEdnN19=kP(4Jp+js@VQS!M^hUu{f?1k=j33;gmg&h
zYCPP1w0^pLukn^gS>pUY^nFtULW9kDN>06c|0t{X<^ONWs^hL#G&<~l(jFdSVo~qD
z{QZ6{=S4)sv*2L$*K=XQi8M(Z<vaFRC<A^E4GkUls(-<52vxQ}HalA=E<RpfL4mW$
zgH9$E0+Ifw8)^F9?k}VxakUpB^D}dDAb)Y9)kylZ?o=zluif>RFFbPiuxe_{hfkkO
ztgLiQ4Lwn_?%&@60jYNGe*C|`Z#W*05ncYY>o`FAouD9<{BGV-A-tC@429kvh0N5{
z)YdjOkF#wH4ODb>8L^)RmlezQ_6zyYpY5-E6c9i~6q-VhmW)6eGoA3o+U`%TyeLa$
zU0nx56)F-hMoeE}6A}tJe(~j1Gcy^$6m9U1TmeJRsjUVE2HJXh&F{?=QAagr1NNIK
zxUavbl@@1M@}wD*4?p#imX?<2O!#4$J2i6W&rd>}Y(h0wjy(1GN2Upbm@`4o1w)oV
z>KR0A4?rOrz<(FA{_zC*1i<qZeIw$1NQaP{9VQ!XN}yozUgbwS;#FPZEnW@AUgopw
z7XRCtrU(uKJ=FN~-L1X?Gs6uQkX0y|#a{q&PoYsjzpMq~_v4~uEk6e^S6W@o(I5bU
zHwo}pB|$>vu#nJR+%KA5bTTt9FYi7ceeI(*H5YZ-lcXQ^U*>5bBS04@hLG+)dipQ8
z$UFWsg@McxHLvBLQKm<PgH4&pK}U$WUI^|&G=%7?iV89Yjh$e!hp&4ZTe44iuioj9
zYRh-;J(GX%#<!8VgyWUYH*Yp|<gtyla@W;8UTlz7<5Ou4j($QKT=c6cEe~g=-Rvj*
zb8aL*tg)gZ)-uOE%>Ad>++UeBF&Aw<r$p9D^XO=Oi5rv=_qR0dp=J3ro?+-+QQw;T
zr9bD5R)9;7XBk)2X{NA%@t&BzLXW}XfFzf?m*KGjg7+Q-d3y8q=d$}5FQ&AO%y6f-
zjlgS2(|W9%#?zF$_j%n!V%%yVX8<{qg?kUnRy7{l64228p&|3^I^cKgu;6FU9zA+=
zHmP|~n*dn(t*39!TbtLtcxtl*qVagPlx+CmX@~ZFTDe1glYNI%^Mp9L`TwSBp0#S+
z<Yj&0qwxjVmfP<vEuM3Pk~1br^6%ce^@@_xxS6`Ts`E$jT5%f%53x71&dxd>E8lOY
z`fOch8zplV7Q7d-*O`MuMN>s<`TmypHH(=^wI^P>*Mf9H$z{y8nZL0vPD)5JF*5Us
zt8ohqZ!a!Rty8TN@V*>tdOU=a4xnr6PiO1|@Csnp=d-i3#=EjD$9`nSzmAyH+$&Tc
z7d@VOg7f*#s}A8pR(nue1;v9NKQ1dPySLxKjSSVtv_t-e(~-uu=Kwewsao81-v_g|
zt0wWD<iD1d=HA<T`=hZ8-lgv!zfCDA98in*D_THwvKa5L@cd*>B{I?ZIm!H@B)v#r
zbl1mBwiEu!5*`_INimDP13{v$Q-`V6R<(;PcNgVcBeS2LC}xmy7m!nl1}jrbHNO~E
zAEns*;GB=qc4~p8XDX*WrjPQ}*BAV=GZtBI+_*tn6uMs73vFcYEuV|55h>cA!?$Kh
z#$L)Yr(_Wc?DAfhr&3V~<JGrJ)<|o-txz#^MDE3Gqn!RHhkF)!c|)~JGgHaBbR@yk
zr!|L$UR-`-`p~7qwI{Q;miLjrUukpMz24Y!4E%2PcjlY_WLdboPMhvJAmNu5K-FdF
z`81!!>;78VDPMo)ulIgvGcK)USyzy&YiPKq$x-j~oZboYOvxbeQYz>`U&9f(#r{td
zqz`V3e3#R&#A^gGcNV$rW|Na!&%41rp&VZNhPSA<@K5v7a3z(P)P^|MeqQr6yJCK)
zaed=~ez(HQypj>3&EKB=YJSTQ?@MJeJK48tH>T?w$ncEYw|byT(4}-s-~p-n5S^@3
z4nvf>X<@!|BQx!CER7$|Za#N~lrC{#ZgO_Ax9RB-Pa#*HV%zr8r&8;D2To`0JJH*5
z^7;p-31$~h&qdQ|Q-j}=2k<J-_k0YndpxxI!Fa1;@sjVksvwP2S*oHJf)BL5Mht&y
zNH_RY`9l8L57*x`R#sMSTIu!PNy-U|Re`TcdWIXOKG%h>x=bGOi~IJ<_DWbo__>SJ
z)U$v7tS&s}U7Syf`EYYXvgqV>0jE)I1Iy-Zp>YF$b4I?8b%t+zjjC<C-sZSCYb6;Y
z;6nyhT|>pR=s9=oaJpetb5~TJfp=72dSTeja}+;lQ(FIIG>upV2j30#*=nAf`=b44
z?E9RGy4G!idL8~fHrA?ia~DcYny%%5C)@+JF;Xj!u}P3O8mK$)S<ozDU<bD`B+wxc
zGTB;?3?JF}&y|(ia&!Z9)9N#oK)RgL+G=XZxwulTYq?LH@SCNsjl0BU{Ze>P^}-Gw
z2o`9sX<B;MjFsf&StSo|dhAD+>2|8{UL5<;>oKc5JBT*+!A+C#!6qt~4(()$liT~)
zVmwP1q&DiXqXq^B%D{jagp9Xby4F%!Aib58=rzwUv!A;Z`VVwRKF`uErKP1L`q@q=
zA=}Ouyv1w#a&*7K1jul3DvMJ;R$5l(o?<8^xxQ$k^-iY@V0g!_T|-eplE%1p1}Z+m
zftkVDrNx0D_u;iio{;h>nNRvzM7#TnK&kH@juCX|Ki^G%I;?nOohekoM>n38Lo{5a
zc=7Ir*M<ZZaKYMq2jz}k!r6*zd}P^fFGx@)7wQ&A1tD}xqg;sIT0K?fOLa~oUtsh0
zU3E?@#~@9rZ>(qAcD*cnK<%_LZ{xe`V9jYkb9(a~d|A2X3qxOjh>7j_XhHq-=~MJ9
z`nGE=gxH|H+HzSx&@b-B*w4C`Cn><rplju$H{A9~<=UsL?E52ZeZqTnu~`Kmz8M1k
zG9^nOAR4P)9zX`2$NZ>#hjP^Kj=XdusW%4?WF%`oS}*gRGLDh-oqyo#D*&Dn?bWyS
z1m~s0?WK3mmiacm)e+n)W7J*XUa%UoVGnHzfbEp$EPE*DCF0@18e_mZ@JYR%>-F*K
z=S}-=EgduTnxljKZeHO>H@kzpA1)~Xu+e(P^Xoq&MLp-+EugwjLhTf|`s)IZVUfSP
zq<sAkk?u1TqiZuN&^%AwoIk-Bu<fxz=!aWtVhszw&#ugmEtxFXLiRK=GD`01%6xa}
zF?58C-`*PoP?X}(u-`7u>v;q*fBiF1HI<@6`}l#kq6Vr%eO<WF8$%W@Ma9;!uG*^F
zuuS_QCC1j&sz-trZYygGnQfMf4XNpO4X8aN6k1t(<f6Iq%j26z8mlTCysP)WkdYf^
z_*7H#*XZtxL8Aq}vqIvd%shF5oX>0uA5gc@DxJCVy+<SA(%AtaAFrn|j^U(Syi(|X
z@n&$2Ub&34k^tcdj}dRI0+&i|>M1|*Gv{${Z*NkUzM8_Z>n`6TK5}o|l^GqCyYoYL
z->8Zpq#&Iv^GtB)vZ*Us<`>tOM|g{t|8jME&RjfnMBcjX-9ZwZLrKE#g}?#4MN3Cl
zL8xjPNxMj3Yrf0yH^0J{u|4J8T2=J`zFH){pD_5!^ethbnh|DTMFqcE&EBHbX~mZ!
zcFGgog=8;IIUG*Y%M)H^Q%r4nZ(6+v4#v(BXc{Scc{~qx9gG9(Y&VXAUkA1uFBim2
zB7EX3i-zfz%Z7syr1hSSH5Oda>V)@(;H*B6&inhQ7E>T<jLO@=vrMrfVBdkESIMz>
zmE($Eh&mky`@u7U<arM3X+~)%N0*y$?4VQLhBliC+lnYTQ0M3Vw#36jgGam4vxScW
zT+iCtnglu5;@6j_*ba|TxyU9vof;TIH{hZ=I64|bYA%><QvC4o<EGwUze0wkav+m3
z(6TB2%6U*|LP%s#_RT?z@awgzwyTR%1u)A=;P;#1oMLfWpE7sE)0M#oO)V^Jhp4MW
zz+7B2blBEOGvM|y{SIM^aee89(Y~1k#U;R5RVnc?^U)6uv%AOz-ipk$Tn^8>O6IjZ
zx4A^*&}~K6@eJeU<Gw`;(Xo+n?Cc&A4%E!OSM7rS_BqXZeEYM$CLt*LeWWolAy(l+
zNv@e<U0i%4+c5^IJ+ysBO>bW(#wlLp{;5GvSr`~}JnV+(@d#A(-)V0%q+A1!CXs7f
z9uNqwKOvu3Q}af@XH4eTJEfzyAztYhdA3iA!=k{=_l}M#;b9V3rY+rY*n4Azg>*#L
z|7vycv4Vx}n*~#qOshcO@Q4DxMu;%|r3Va0-Fl5xnQhs47dCqEix*#dH|=8Esm6#8
z4u@iM&iDxjNJnJ=6H~N^+bN+Ps!(f4fa@3WO$3g|n=^xx>^%B|kh7TfJb%Wf;`sMF
z2VoJFEr+w!feWbS*>+jaSsd30H(b_v-c{t)@#F~zRkL_4pD{#cdq>B92oddjN)Tf}
zO(&dalvTkKAnH|(jr=`pC`Y1j=q)d-lb~4wPV0ZL|NN~8ehXt%L&ry#>rkgZ+!$ko
zO1@;v_AD(e&4|Uka$;~0>#MNw{c2(qnDiio&3@SPrlt3H`VQifjN!fB_w{u|nRXUC
z2<P;?j11j}00#$$&JPavZr>)w4?@F)bk$_3V_T(iYiIB8#(4LUw7efhUZ=qiUSm7@
zegwfe3ufaq`iMgbX2*eX-_nC@oS~cpt7mc;Lw#iKcnrI6aJ*nUre?OBt<QE>tEq{y
z`fBbLevX}9{qeQy+)3ig%VlCA+1dUzIl8l%nZHd;W^Ar&jh6(R>=|5*3)hL3Hw}~s
z%b(Mde0$E!CW_ygM<GYxDT!Is@jloXeY8dN0}&W15U6CFowFOa8^Xr_QQ$5Jv1b=z
z6LBEcMDv3r!B_)<{W%yUpUYoO+wKB+*egHq_;M-E-T(7GQ`KoczGRhR=v;amEzAC(
zbF99%DLDK3CI?`hq%2Q%<|}DwY2APDU~^^J!SIbS3xl~+UdztQ3lk0af%f-i6d-jb
z@IRd=x}(5f37HE|EX(E-sF`n%(;KODn|)g5kpR;ov-IYi0hdiAUw%*4d~@LWV2%`C
z0c25-_DLv0OohU@BZyp8C{B~Jv*k54et;>@!ou12W>$IeqG~BXAO%e^M-!=wmWXhl
z8RCcBb&h)YS<TDu&wU5Zn^`^j*|f(+viSg=Z(K{9igG>cBhRDmb*$A&Di?a!=SEKH
z>cpC+oRe3sPu5SDi+OT;y|0LC59g&y_a`^nX{)M4Mbu~dAD=3kHEzsNA(<T|v+gM>
z>~ma;{N?;S2)xG$bwzxHYS*qXR>P-Up$eCJV}BI<p7UtpZuz16a{rRE2=+)Bt~@!J
zy^O|AD}3kGTi4p&UBnY|AC7YU-ImS*UH0|Og>fOPru(S3>LYzykH~PqvPXo0-eZMF
z|3e9ADrKRcPJVtqkwgJpb6cBhqbN*+Z}R?H3+EnMTB@DrV2MFUT3lXctcils`s<hJ
zIY-Cd5;7<VQ<IY<fjz8I<JAom+z(y5-5-DoJE5{#7hXiyJ#pn@2xR<C_G3R!j&~Pc
z*F0XFdrgQDR^LtpfW&S9fBAcu$Amn1u$?F+<qvlwP4Ng0p&%YMR5zlHG+xVT4<0D^
z8P@!VEIfMl6i1rozJ1BuCj@w+Igh4k)mA^DS!EGz({23ojV<Ywih__bGt0%pPNUUz
z!ETaG@sHwe$C<81hn!bi&Cy<2qC3mouLcQVuk^3F&!6ugPXKXi;#J!S^e`%<kna2V
z#KLkpn5(3$Y>urw`?uvW#DPh4L;J-^BVeJOrjei8pu{1t1KA<D@BBJ8YldlcFx$0H
zP-3y>`=K9Kb#_Mf_Fl#22p!qKicA6#z(E5bQbFIpYX%+=l~z*nZ){l-b;lDY(k!Tz
zrqDQ1svZED5aMsV!3H!R7_Fr+93J`mo5H2tvVA+#9#`lJPz)4IO^*WdSXd0R{oJ0f
zlka#8?xz`$W3nbEw%^X%Xn*iw0i1{!2T}&I`uYcGk1Ks?jUR$Hh8MGA$Bq)LAaZKn
z9W~c*w=kEf?Ck7t!{zYD2nG3i>syZCQ-r67SWNPjPs05C{OF|}PkOg+-%i*I1~<n}
zLese)AAbb7j#p^*SSUk3bKfQ%`SnXakX9p3JyorvBQ`g$w%DGG!|!fzaInPFy5RG1
zT&uI9TB55O?9q{M*g_lxE$Yhw&0+630d+=3Mp`z#-pS%QdA6v`8D0ez^6tHR%K<eI
zDv1S-@Ll3z!V-Z%y*OD$Z3QLPFRpiNENZ}q!p96U4KfZy8VAiP2_Y@dm9f11Fct_!
zP_yfzPCwsCd;F?}#e?GF;=WCxtGPKg9l}WGoC8Q9WHc1tx)`aR<W9WHpr9aXxJV};
z$IOnj1tDd{WZ=Ef<+_ucJPxup(i*h8cTauj8@58jM`+97){<HLAW^OGK8f!wSQ?7J
zCfouI4M~@Z#i@aSn)(QlAozTlo=c7hvEfyk=8h3Tmmfd0iL4#b10h2#qFO;|;0U%7
zKn26T{e{ra+4#n9`A!z7lCVp@K*;(GP-F>!ho^*?4MY7(ZK{pCd+@+{aRv~Nn#yg0
zOianznXe;`OvkEEGwRW~`$33+MTA((;i&C20w2fDNlX9MB~QG-!IQvI84|J=zDF{a
z;X5{0DIO?(SU8LUiRBY+>X9`bkq8I%!&<S1&)+6sCQ*4e_>_}YWDIRZf{<7JzNFGh
zoyeXZGnu8e+I{~-MnwtDSZxT6yw?6!XM7M+7a`6wHEH<gua7-h^+(w6vwQtIaEA4K
zxQKE2j%Cw)Ct0nnQMN~u+P9qVCak1j!6aJu^A1-I9l6iWNJLX>#4Ri=<WDJ-#5<|r
z-*z!<`sb_uw{xrgDMW+|84>rea7iyQ^fV97vE9-;G>E`LT<2qdf9NC*{pF7ir(I4^
zA9No%#d@DTjY!8nbiAOpx2d^V5$f_obp>Q8Yfcp!G{A8|R%aXXB82FOl;UO@Mn<GS
zjUhe@U2mmCSTWjthyg|Ss`du<Gm&J$76LngPdSJTpl=)SC-2W(o5#>n;ffQB0f?n+
z1nQ_KkQLH1Phg|rrBK~ko4br~QW;lz!CmZe1QGb&2x7fOOioRGMkE1}CpLGO#aki(
z0XhML=cSm--we18o5td?yGh^RQ^6P{ax|(@&=^ow2%`+CKjJ1Jdy14(eS-z2$r;+`
zZZO@^hGL#=*Tamv;M7b$gXV}jK!P|a022e=?|oz*ZGL7$o-6Zp5$qX3y5k%83(THH
zFp6n$$@?_Nm39Yl3kz$Jg%GPQ{WLfC#rABgXTf1%TEASL@58b~V;RI_RB&{hE~c3C
zkdeRW*8hM<RQXZ9*OkU&Rdm(|3yV~35CRMe#oe&O<GTbr7fR%;fE*MoEtB8hT(I5x
zPlO7{i_?2!Ew3rpMsG0OBAWm0#PD#^>af&?8a%ip{Q@^NojltpM`VDI7#f@&to4hE
zih2(+I@n+FSV3F+MH3#8*|%3`edi=3B<Qf(%>Fbb)I2>RM}(Hx83SZds<YnPc8Lwb
zAg%#Rd-FX_?{cG*zYQ3PAHudxV8Ld{cKq`@`?z)KbFQO+D*$^^H4w=LLaMI;$P7W7
zW@i$%R*j5`3PsRr@KX?g*K1M@E;RJ_Z_7a-->q+uAk6oB@ZiCF1e1;l2&kh>2E9k|
zkV$IhK>kaMMA+C0KxhD*W1byP+lnM_L0~d7ODLFY)vJN55rP8C{&+vH7CtE+={YPg
zT?bd!%Lqh;K%>xOfPX|pV$YojpK!qz55el-dd_R81)%hNc<L!rlyVYcXzXH9JUuUJ
z-j`N}4-B8@xcpMm;<1L&z=`^aiTBblNe=Doea3N-e`xV@^vhT5%|a?0PhYC`-bi}=
zny9=r@Vx!Qs2K##|7lB)VdeOd@08S+1H!;{n^kN3_6PdUl94iEw<}soY<v%ARf8XL
z#oGy_$B1{w?-_1TeCrBX2!}=@eQSHMckw26q+9&rOpo5aO+b=!;m8nj+;&93T1M?X
z+if`CH5+R^8w@#SLv`T_$feA5sqEgpoBS0l#!gH3TP}f&Cl%bs^M@%Pjl}QV{K~rX
z1Aot$Ma^}?g^@OuojZ1XnaTm!tSBv&CIZ$}wTRB{h1X;TkIgS`HG*XOGrL;#c=5my
zkzmpg%%>*BJtd{jZjw=``Cj=K$-!DR%_b5Q|DrjT-%2KLLI@#!K=c)ks0HW%<w3C!
z<TJiO`A7CAATLkk3BRfPe|U~x901CrhYw$Y)3*P0b#mHV($gc=@gC;2Mm`1-prgYL
z&yQ95FS-NDRmvXy|HybUOIA`dTy`-EJp}db8yQg?u8;cjGlvV7jK|wCGWYQju)S}f
zxCh!OsHx>rTlUb<T-7!=GCB*mf68?{Qmq|bjNumV?<oulNT3=&Wunkq?5(7|9FNO<
z93(f;dplf*O!-qz2Dg0|Xe5a+u93}TEHsDp<L2fj8;bPlAZKXJH=ZvqY!kdU)`_1$
z(crWBDTEY<BK$-73m2|Jalk4S>&S&`y>9o+Dfz!Jei9sV+4Pi_zs9Lq^3RT+KYu>M
ztacx=H`TDjWkIlid~g^ru@Ox7vH|1Kc9E>9udkoHd8_IHO{p_Xx%0M~oEh1*#rn5q
zJu<$TeZTZ%O9l<#<^u%1VYO#pZ{z70`CnsdSU`tuL`-M14~jaLf9r6CvCk~-cJ=qA
zx|*+FsjyIx{wqZZH27_!qpQ2=Rr4S6D@n5Pv35O01X*I1@c5Q<{qlHEF{}X!IB2pi
z`_9|0d_({pUiV`LL32~RDu^1$WdbXMs%$xv+Jb<p5fE9#NbdA$6xKvVq6hJznZQ49
zI6YBt&mcVsO$F(sJ$THr@%%?A@)0ompTOA?;3TV-H+0D3U>WW`efo5Lv_vg3Hb~x7
z)8=%Qmfl6rnL>8O#MV|B?ikidjqqxO70?o14x^l$oS5sl(zR=xNaL~|Kkf@uGnk)-
z;$~!GQU+R;qLoJ2g$Vx-=?%l*Nl{NAhtgMBSLcs5IRGQ*MfuO#+Q*QXcGJ+@#`CA+
z)cle|wKO(@7mGlB1ESDhAMQSq0Wk;*136`;{rmTKdyH@&S9)<fD@%~rMc9u=j~(MT
z{qjUVFT7`p-@doRv-=q9dG}*z=w_WrtvMiuF+bX_pYOO8(LN$sv7N>9$;6XX-LxT;
zSfq{}Mq0U%0s*Uh5WD{k@ebOv9w5p51i_Iwc5)OGyt9Qz9X9_Bbm<)}aWxzpbP1u-
zrQh*EFT`9p0Y{Gs2?_Ox9Yo+r&|(*$HNw`-1@2CGTo&{-F(P)%fbk~C*=VNehJN|-
zMZe=oPZ4^$Fp<r<M=eR<Y_!pDuuPHn-d+MV#t?B;#J~?pN=9RQX5u@<pDBM?S;Pzv
zPiP#wg1kuF+yB5mZ6U^sPg~Pku3>k^B1tPdtRBktKO(<vcw;iUc?RihJUqMLRb@Dl
zYS}d&HWNm-ph(4%sG7b)^RTr8r*&k-(%KcRke*`ef2M+-9w7#X%({_i$jHlUm}*2&
zCaHColvJ^P$5R*1gMd+1gaaN|$HZ^E1(~Pnshqf%1__7IhQ3|P5iJT6-5&>^cH4hY
z9@-+X){=n>_kEfKA4)Qc|IS>}L&t0A;RVZoQ)P55z?U$$gX*?d-iv@f-_kmfH2TM*
zT|Doi6eAf0_)I5%(~*{eK^}n@zfrG!(jk(#5;bL{Lmr|zhdK)?X#Q*d2N_cFma%ub
zIpl0?N!GR0ERuB*=v(Jqf>@-Yv6v;)4(bff2Hgd@9d3w80#lI5J;-|g&*2&3^F+nO
z!klT#k3fo#`4Cul`te>iNa7I@OhZG1GhBzcxv3NtYlCflX^ts)kbKw&5CIG-E|vnm
z)O@S`6x$PO+KV)cB&Zq$QiK9W4UB?pc^#zlfWKni>*4@p1R8`qLde2^d;cW+R<KoK
z!RC<QBIE*9RT|`hx$Re)o11s;+Ev-uxEq@*Q7Qax9J_CuhK2@%)Gcw`ctw4EU>LXF
z9!AD%ZfmHM&Lgb>$UYJj3K=hiu#pfU9C;G(;cdWPJoTkqHmK5#@p2j|>c;>r35xxI
znnk!^9TAaOt<-j?NdFGeLS1qgs$)15-{ye8KPnf(;l%9G+Ujqb49r~tUiCv@D{XBJ
zBcuuMja!gem4I{*&lv!`V-*+g%&3n)R}$XwWqERXdefPtX4C>2B5VRUjmaCN7QZCF
zL#3EOaFf23ii+y1!wxqPHgw8OTwGjlG*e0u*HJ-Or*H1=0mFQ?y*^-i8kkN1inVuj
zwLpemcA^^95~2sen!)X-kYwBU#s;O+|KRC;cy5UE-|;ueL%<P_7a@FlL=j^Hudvn#
z5CdC2u&%70`Tcqt8Li*@UgufNDl%=&2wZo&4Nw8?$zfq(9_33;Ag<HI!lDW!D;)KK
z#cOsqoB@(Runz)ubO7PI5In>~{E7(Ji{W_|akmm;_K7f!oPX0`xMmoIB)>b1J5M+~
zBEsqSH&)=!a@1ZBQ9H_Xg1X}o5^+hSjed815`noj)Rdq|@;@MoT(#!>cvmfMerdK@
z1w%<yu&k+=gh}>jZb+a61jlp|im&yLj61kar_mq$;3YEP1Pc>ABAg<aK!RXb)mQca
z6*mDW-7-g)pgu@KJd}-R6%%_2P>WU-Gcs>pA1RLo$fg|m3Rq+op$GhypjFdR=p$Hv
zU?syJZw(3KONq!hM!@>pEn>DJ1f+oppIgg+*oanzaULS1t(|9k@CkopXOapF`jK@q
zu`H@{1Mo1K(P80M)K@=aBSwOHZmn>GG-s;i7X+pk5O)DrXcA_G1B@z+w%jKqfjOkN
zi7=St#;OzEOckEH+3&5~N*ze%L%SNlBKn25LkQD|m)YY9!^6@2ZF2eH`?sF+6BKG&
zf}5XNEqQatEZ?A>mwK4yV{@lQR!)wH^9M0YQ0B4LE>|oi<97+jL-O?WM7!My*b=9a
zq)Zs9lauz+=gy_*=L?V!liCSnSCdn8OfigLZm91*|5WBcq^R~m$Hv6X1?YY=0iW@z
z{*{%ToSPd&B{-p-aoecOcQ?EqC_0b){7R9>`Eo%QS6hSs5PtIdKA;_pqcqdFf}c~l
zSadVLL)cO@`q@G<;3KwtC6L>1<N`=gRxL1V^N*sjGXBZ;Jt(K(Xav>(@j8TUJ1-!n
zYX48}n_=F>U>h~F^#!rx%8?fUyTb*{{9>P9gGT~m@>JWLU|uXcxx~%+a;Ol9kdl)S
zJ4EEeh+Pl-7U`D;BDvk1L=F84eeTK2moF{!&?dovwkw96IExHJlJY^@&JUllAH&g7
z&CMCScDf%lbjK~m!L`JT*^Q5Ub2(iZs1@OWsl&Jm<ZurXf5Uexh~lR(3kwTKfr<76
zh9DAwaR8s`0%Y$YLW+c7Nm*HZN!2?Vv+!AzYu}R~^@x2kdwY8Wp??KmL}aBp#}hos
zKl*Ztx(^!jW`;sj5O>)}XNI2J+bTzkSA#iW|H#ERs^S5mXl}nMWF7t-#V!_M<_2ty
zujq~l_983C7^N(SS4cHn)ZWnnwU+tZ8qBgGT-xzNhi;>EI{f{fvdKn-t*@^S8wy4R
zW$lR*CoC~@0{kw>N%$0H6TrBT-^{r*XmomLMMOkOPCWu$ICAWm&xa2m#7BsI@&oas
z;?-$lVAFSyu%(EcbW01}&6_ugU?4^o$XhDGq8jmm@e3UIa`2yZz`p|~WCH_SycqvE
zj&TaeiNxp+-nu1ysRWclh)=K!CDb}z3fo-E?#5=@Nkx@_B8f@}r5J@~m+B<3Ka|kV
zzW4;>Q*B>i@$B~iY=OdsAx3(SBhwwWXEFV9WAw*o@I=B00bd;|qn7&B`#G}11|r0)
zeu8>z+XzKLq1>^fN53JxD!DeRfzJStK^WA;+*}siiuqIb+qu-1YY<8Obkl@AxSe)j
zi^jsEc<(-Aj=&!gkVTrF8EUOZ|Ckh&PS3NtqsNoX+zVBDycP~p>^pG?Sv7*Pgr!4C
zflMrjc$jXUZNR;IoAGeI&MmsB2i^2^cXK1EB0)RQlHH*7g`alixj8?|uV5Z?ABsZb
z44$5W$Dd=zRia@F3er_%a-Fwg118N!f7{CJ_w_M$a+ml)J&KM^$8P&1sHZ@bJ%p!@
z2Zq6~YLG^XD?#b$ywH&bP}O~-li%)_`h~FAag;7HpqIM3It<_nQYO)TkFWlM0S>4@
zX8|WsD4-ca8%P8e%x}fI`3w+E57hPS_(fx2OMk|bHTDTf%1K<RdPsIt4L>s5m4ZMd
z<~bv6ib^3!tD^f6S$1OkKu6k&q0jdv9iHl#OR~W8^qgzCoPDjO=66m;1|L!4i4n=N
z_i{J$$eUV;K5bOCZLbXf(bsr!c){2NIfoyWVQ2z8wh^yRHt>Kv3L9)5q4o83cpHG{
zCiI(l{Bq5hKcTEtwYTSMuM7LIfBp4ve;nm$x5+~oh3bb`1}}|BY2!^CJa<@~9v@PQ
zs6XzVFbBbuUWmI3+-{uI)gmNQB=ia07MTg?w4Z_H>BSK{qJ0J91W&gL&LrZsEZPNb
z#Gp0VE3D!IxaZ&l#8(i%f_Y0~E&zVewyUVk2>oIjVh3J$B{n$K)J+c$z?DUeapz-L
z2@BCZmq1H--5!r2GH1AqXc27+W1SZi6pX(-VU;zLTK;_mUsHkRPA_ChgP|th)OUpi
z1sA}1@t!*iZ$5wh`t_!<$E6?LXO1JLIGodFa0Jt!xCF<Up^H#(Hk(E++(3RBw^NCU
z2Gm!D+}GWpQRpAUgeUBl$6BdhbE7aP5&?uy*jGdj9%;wYY^!Ftv-(euDucfEgIx`R
z9)TD6V2EZH9CA!#OifQ4_m<otTwTb#&>nW68&U0lFK*i*5QH531A>G?eWUOiC07e0
z_%uUeqva0-DuH|xaBoV{;^A=37<jI-QzBLaaS}K83jYNp^SeD=Mox}P;0Q9Qw4kWU
zN%44-TefWBv+X<u;h#K`9#8*iWTX;mH@&FiQM?!amnYnza1n21{L@7L$Ve?PVGZB{
zRiT|&R3LOYbqtOXOcYg=7_cS*w3>mV5@W@Pk_@_^v`NHZW-)r*GqTOwzG1op7_tt}
zFByX}xsLdbSd>`|ojn6W`1);Nu}-jokC3D1#~!DqrZ!e=m-zz-)%nroG12lMmzKrj
zen^zqE(DLyvol2Wp&Iqk-(xcwggsyETJ+R1m>L-w8P**sx;dNx$_K=pAL4HX@X^E&
z-QxB>c~MbO!U#Z7Cx+4Cp#H*Ji^o7bi5Ry<pkWH5%XqN|@q4ny#{IN8U%R@ZAdfZM
zk1nsQ2%=z0AN(X8V#b-k=cK&j2ux<sZA_UwmW}6V`iO7@zSkKhT!;I~K^ZGQel!HU
zQlR6f*Wb^45J(T?9{px%MFaf-908{g2!QLlhjG_B><dA8I(n3dWbCG;)p*7Axc0Dy
z8*l8S!@l9+^9~#rW3nnzGVm`8t~#ukH)EwlM|cQP8i^9Z<j0nxp`m$VH_djRCeE6p
zc;n|zR&s7h$!_~i>zFnm2J+F2YE4?`=;+LZTx#QkN|j~Q_bQk~D1F8hY0QY^!pscU
zpP`*b!9;}Ih-UBJ^Qx+<Prnb2kJ~zPWkwo!|MZKy2&#kpXy{06W86IED-b2FfA;42
zu#r5gf*`?1D@x(V5oVmTpnpiEaQG%bU}Tbrg!sb~zYMkiTL)UGDTB3qvA6pfx96Wz
zA!?jBLraOX2>>s+gFh;8{fkm+H+@Au{fGdSKh5BG7|nchrzL!SeL<2QNO$F?fw*Br
zCHYHmE(#OC{^g@CJ6X@G;Vg}Zjx~*qaBEa;IsQqB4-GAO-~agWqomJ1I%))KLqjjg
zi=5k`Mp;uwGr8Z;^9~N60A0_<*emi9gSuL&Ez^y1EPXh&0J1!gJ2AE=YnG@I6R40N
z-~qgx^r4I<?n|0sF@a~0hk-eC7%t{fi25Bg_MlY^a4RtqzzjU0`A$Eau)JDa)>qvj
zVBf=M!g)3U>VVG(yc{wz-+;k0ptEn5Lf_wAi2!s)_0j8{PfALnJ#fGu)*vdJQn)}x
z@~{9HA-bssvfl`8S+>7F78@Ju+m>!nsDR7_cH$;PPcgr-d(R$e(Dj32&ZJkE@WD3$
zDu-C;;s49w_-SWbKJ9a26%Yvi_DvPcmN?}B4-`;w60~SHMhppW1Z*1kQOK@468wg-
zW$9mB5+pqqW+IU(8W<=`FJ9Mu<wZwJTLzUb5OX%bMTBFDNyf4f%p+4{vb3zdodLQS
zAWtPQKUF%fUM?kUQc!`Zxj7=AB0HC=$w@$Ee5*7@nF&};B0v^Kdk6|nfF+YkO1MT8
z-_0QbXdd5F0t)gJQW-zsKRSslYyk{d{(|u@8Ww}@MdaZK@B^Sij0U37AQ0dX?}!rz
zh(Rc3%&4?A?|>Nraacux^QD}l<DJ@l=0AkNuOKy`Zrj--*Mh2zK^nPtE|}tC#61wi
z5vK$!LWx53d^-uH^S?nUK`4?+_=%6gBq~gbax80Nm<m2qK;InvZUVVu0!P&4?_Siw
zL{xRm9__r1>P!r}L1hy~uO|{zcrVa_(G$0kV2Pb^b{0aZugT)Z|HlDoP-47#GQYR4
z&&X{Ja@SR?yH)g8A<J(B8pY&U4a6{vAfCdDO$3f!>eg^Wt;h8}K$&a=;Uk7CA*;qr
z6ulHMBZr*}y_h@lv}+*#+1KB{>5R1WCUlAc$Y-m_eGwyMgoaXRKwSb*0O6Gq&5#FL
zi4jC-VgLYLn8}84bO|vDF#&$J$*b#_SHFArE<Ju4-!O=dWQ@1(IbF@RlzVxHnjeSR
z8Qf3^Zl5uF#96GOtLsX*#LXLlgQ=P+aVK=%ms{$942EFIc1?Qz+Z>8XzJeSA`74MZ
zl4=s7nBvyJCVTxlA%V`lD}gB9qM|yYH|O65M@Cv}Cn_pw)-SiZ|E0cSY<#D~YxcJq
zChm_M$@(YaGKdLresR0DcTvPS1n~m!SuTT+CByx(ZhtSsFCuao+?pzV;z#B=vJm3*
z1B`wBpZ18p+FqG(QK!V|QVL$vQQjzJFdnsc28C}(HGoXhHr`;XAwYSY9);Z**04$y
zFz!JoqZR_Op7UCcZvL&ckBuG(P80}$(Zgk{=;<+p+3FEEhuokZSqKCy&yL`ni|D-7
znb9z|b%<|p7Q6CpWMyfI1FK(Hd;ybR82G#Dwu>FnFJW<&3m0&l0bb=&E|LJYHNBtj
zIeCuZ!K<U3hhSR|E-ni;h{qEP$WjO26N-hxm}3Lj=DUDXML7Psp6jhA;^#0*8jm^O
zV?Fa>H*^M0JdBiNQ~1<0E2<ZXQ3Z!v-p}b0injB?w*e1pf-_Sl04qxD#waoI>7);7
zi|+dM>+xpc=Y(~xV39-Xzt;M;8dW5E!h1|QL>Ptl7id#ePfrXMeir&WkhCLr&#SYW
z<A*UW%Rl?t@Pj=WLIF_itlbJB>AKqz)-duFOkiWMhOwC9(fvtCMeS$z{eR<)pqV1#
z1bj~RhE7HJ#RQx}ZGo<G>D8EIAYs`^;OSjOtO!Qsen{Y8Wull%D~dU$)3Y750h;i;
zBaAwQ@C%7~acH4aI2I<;u1B)$Iw8@ZS`m}&AI1T5K~ng1_tDUl;Lp?kQ@CnRg#Tq%
zq_Q3eA<Wq$Ou?#<r0%`@3${(;A7=nC25vCX0RSk`ceEIM!5`d_>Yc(MG7#@7)Nv#y
z&V{3h62T)J3G^s1&>UMj+omH7#^o=M=4l917y+Pa80hc6?eDZJM|bR*@3o^@MRD))
zgy1+{%lYZgYFOCKXPT(3)93VSlC~|4oqTHN&Qm`+n2w3%hvmH&$Yzy-0t`Mx+!aAH
zVfHQ|JpwCH$}f)IxW?4hHag8VO8Ef}_n>65@RFEWs9T%<J+rvzMzFZ`m7Wba0a&L$
z+S@_mlOdJd>37Qx(8+^r`T$M>9B>>*5kH5A+c+{u;Ab2S@(3F*%cN?TITE%QR+q&D
zTJHQlO<SAyXr>-(IJUwJ?Ub0T2S$NyLd^SKxOnlT?ndJ>#={z&LKS2@JUohQC@CqA
zB94iyFhMy%PyX>nkQt)dm>s%Q6~Drow9r{uS($;&i`3!^L>uq|8nj{(63k-fuUkDk
zS%0_u>5^+)%eAb7K?#?vmLs{9eFCxyCW7v=&+Y6TKx5sFRuqqAgPsdkh)GL6$RT_o
zaK9h-EujArrUMuP?DK;-lcqYe8&<R+_ONv=xTfu*G5fXfVT(|U1Xfuz2`p_TR9Zv~
zWDJVK1E?P9tdmu1czVW;EG#S#TLS<QQt@VJk)SKYNhy;-3Yd(;FW_{OTAaKAEUTHS
z6#xq3urzIkticu{@<Y0ytbCk@ClK9$gp}GN<XLc94}d!r;!xI?%W)t^yv^WIgxx@6
z=|ej}OaN(dxCeUvJ>?{f#N9<y|9Ho8_z0BvrebezB4~)5*5cpP^d02n0-j%RA`3|{
z>n@es3n)VUUqjlPL!Ugk4KOoB-4hS&{fwBNh9=ZK>2d!279s-u7PC(Xdh)-SXk4a~
zsn$9!f?NU-S48)2Mj8Mls1nvS_<0{dNFM|t9KXcc;Q_3SHebd56v7WR{-ynA7=xjw
zr#KPph!k8*t{aGb`<S=4NB=+LMtOBtwrDU)TEzI?t>h?dGdo0vgDpV%o#0@_6X{a{
z#y2%JARZ8Ro(2b(q8UTN9!3jILLtbvt-{$m=+jP60rYE$%pLYN12Jbts076DHF$%m
zseAK(4R;tBslEPJOatj4)KuTXLa*HB?a?7|?3fxD=<QW~#avQ)3KZRYb!s~Zc^tOJ
z0dcn{#3?8^zz7oMLn3DQLr6rV3MTq*?UVRJY3U|RjT2Z6^^1tk;T(};aIgrXc1E`0
zIc7$>yO|eVe@sWYb{hd{5m*TL(F8s+Fasg!;TD$WgAw?}ZrI=h$|4f)1UH997Y;Fo
zF~Ar3v~PIBkQ=J8RWO;$k5Sk@oQwk-Jq7^!DG^+R%B%6|h7D>op^yP)LAN6OV0amD
z2|%9cz>LpP=A&W_L3G3v9}9{!%qgmkUbA-*O9so!%h+<#_+d;$6I!B-n*V$j)()__
z>`E7Q<?nojgc_~mY|u^Pq2B=JJ_Rbm!%Rk09@hzQnPuN6jRTjyV3I`|LKXpWaG(e=
z@__`$G)~iKgiA;?A>tG;sK1o}j9EEf;Xv5eDZ{`b&Z`16g_80J-^`{GO&k|y4w=(E
zH!quOw8vaQfgA`|d}TCCX8nIK@wP5mq!w_^?R<r)d#o8V)MQ@{Zs=-g@`s__G`Jx4
zOb}R%KM{@1-OL!<8momk)TB)S>s++{kr@_=wYPSKj;oa#qw)7Y&L|;Jr^q`sG*!dT
zDTdGOKZdqLfF{@x=aDHxqPh+_IN}u$ps+*O$s_vYIKH3X`8N^ck;fY+-XSqZil=DV
zW4dktC6y3(Fm-^g%Ze@x{TsC3KR@67PLbxlyLT_h%g2KIqY9&altJ>M9Dzyahu8fA
za*X}jvI{(399uz0d}>`CCBj>Xx-DWSgTD~@ObF;#akLnuW<m2hD5jEEA=*HA$b`)y
z-a(wOMjVL+UI0`5_~FB)G_K|m0t@Xpzl1nkz!EMOL3bigJq3^OT)3%RrE@k=0v6N1
zk60kix<dydMgg9fBIg%{q4MX%(I2z3&RB3&1f#>dgAsHUGX})y3o(FA3{4>EG>u6G
zv+HuLr#?CVDJ9kbF%tnm4}pQOP#~^ke+e2+O3Uz$2wzDQ@-(N@(y^WyV7H7+%;24#
zb50j^7e^_*M2z9%A#nh<ULrP3MRm1&zWYof&e0;~c?MA_L_(zOdv6mO-Z61;{q|3;
z>BN_FoV@m(@8h0H^sGxD7om_5j-5E6^apc~#6W&$k=Lh^vEbC(F-lf`nWxgnxkf{M
z6RI(5pFpZXuy=-AFKplhoMbh%qxGE_=}vgAzg6E2NQ>qbG7&`j74($QxQGxc^iHb%
zz7YEu))vB0#sD+&9S6xl9<U!nk(i<HBIk4}{SMsM_xG;?w2M;+c85pj7(QgnaTt&#
z63QsHs0740Y{=ZBp(%c}LMYKsHT(jJ)kNS&r>|#LHy!B}WUEMMd=1O!M0%PRY2!L^
z2%7~^ExsLmK^o2h1mo_12^BB)9B0FmOMtZ+)bYJaov|aN1q*Ev%9Nw$nAFA+H74H&
zHEf^K#1(J~5{bgaty{MuzN&fDCI8bWCw#29q4!Uy9PgLG#OS4Nc?PD_&H0#1?`S?w
z{J1vi8Ad|WU;saD`jPk+ZO`<@T{&F0MtjkH#eYVb_r5&(-#d{rE)tRICw=SfG6<0E
z2~`Lk*7xvfEb_XEFo_<aKHO(N1+j!Zi7*SHeM22N#`&LdX7U1MhocD<PyH!a%bz`Y
z^a?+R6TcKc?u_%lpr}{|W%zXjhVA_z(>xTg#qPkd0w#bm7~jF-I0DGj?h%FK6&r{#
z2WSiSH*O@tsDbO=mHZBgLli|OB7b-}of!fc(wCV~<Mi{qVG9!ni6AorRS~<;0TGZ*
zc~a=?;G4Jt-~*$Y5Dv^AyGi_C#f5~%wc`H|(!y@eP;wwHQ`6V1^{CG3C^>{Ze4A{u
ze}CPW7lEq4afVQ9`M)pyUaMsZ4GRON;{&J%+h;^U0?@N@TSGmEICua)3~^(I#gfGv
z>!14OP(OQzhl!vOqMhyI!=nge1N$LPERFdDIJBDx5u*{w#j;xmKrQj;+;cDOP!lF(
zR-U~wXCvNl1<QHpey%mYcx#&W;K7T<x7LW#gK*XX$SEKIN7}@Q(@>(K4kD#sifjyl
z&cT5;lcJQXz?S^d(zx+Azra=qd!^241!0EX{Nd<$00BavL^6lK5}>;pz4XLK<>!lH
zEW((WdWw;XLC&Fa(gU|lI02y1kzy`>AUVy5B_q_kr34ba{KboA;9tZs5I7BGFM1_T
zQou0%X3~Bh{X<YGz;_1Q!|?xv(?kfHXy^#@K^CB|SgK<2rrEQ{c&~-QL>cnZ&>g5j
z^@YDLL6rT~9*9;SiJB4$sell>iE$vDMMT2v4skvr+=wCsED6=!8^IWyq(DMmJOO7H
z70my<PFxW_PtdF;#5!p+QD2F~?m$%#{>xrS+Q@YZ!kPSqCjVAD^DaINQGbe6gMY`2
z5AXkD<x#|(HJoC%@Kt6kBSU?EvNbLp$MWF8Z^p|ft*)&({loJB__p0fdK{Le5jrZt
z1R%4~Ff%K|=|?r+*85CFG8T$Oerp~k=NLt7>yVqlU0}|R$QBR=5XAiy6sID7LN7r+
z6>1ma48%>KPw~NmgJY-te{gr+@m&6I-~ULHBs(J_G9xmwRmjTT8QCkFj0lC0y~)bV
zCfT9vT~<`eCVM56aUXAef7kE2?(4c9_y70zpKt2p{G8|ee2-(jo(I(aLmC&t>DbpE
z@45c3$z5>ru*tVaoE;oGE3f_|p#vv&<doLb9tcGtjdVMsuf(tg2VND*`Jh}zJwDhA
z2tzQG5C<6($W)^f4r=!l8Q+1;gbEJ8ce%baR>1iIW(FLiDNvn>Q3P8Gy#UzJL)^tu
zAFc4Cvi7D;h<<KByahtmnqXZ3(hq>6;SfqH0i7YN65PRuwBnyH34<WO;Q!@7Pk`+s
zCdrWiRk2Eu{Sb;0s`T2JmIgpd2n3+Po<il)#d^XRl&2s@v<6gCG6$AAs&xQ&6AY#h
z$UVTmf&gdKC~Y?=3I)v~Ao+z@d_1T!0BFbZ@geTz>%H9<cqs6k7X~?-wMMpSMNthr
z%n-=7q7rDt5OH0GpTSkRIz0Uwo*Mk`3s-T$O<o$112d+JbPBfA;O||qKK@p7x-Mx#
zK~of@mm_c;Wv_qN8C(fm1fLPi>51{Q_uXuVht=IzaSyx_v<pq)_*wc|83t8LNOXZL
z35evy8@WM_J7W&Et2#aX@nvC1FZe*D9*HN@!4;8`dI#1sL@i+V+8Eg~Ez}1O0U|_A
z=gqFbdb|X`2{|R?n*w1D`KE8(HZQ>WPGl=;B?yt7R`5zf41I&?1&X&{LO8(hyB?$t
z0R(m*0x(OX)lnjzzfsL9s!a+Bxf{`JY-SdBwH4KHxc~g|VATZT#Ltai+zRNw4JC>W
z;4k9{5x*d-)AMu7qrhc#_JLAaz`empL5|=Z#HE4rV+=(seno8n8Acu-Y(Y>(gSi3B
z+LT44V?ui;2=5XS5)545T(cKRETFz&hk;B!EOLlnN#+dLgWwLT=wDb}L`C%AE+DNp
z1fXI$^n1b1JF8J)y1d&N1`ct>hSroE(9}3vN54OVeQ!q7>@fsgAdZ}c$}{2KVSYOv
z9(W=p5_0$!&p!g25lNS{QK=^Edu0rLVXdyemb751nOx}QSbD_l%Op6>aW8;t2|`my
zUkVw>^H6;OI~J%;T)@(|dASIwd?Ij1QAs~6egaa`R<LZag8j|QC?MMfWi&bZ4Uee+
zXbaXAO7Ov@aFf=+65)gd8PaVbyefn)P?$2@{@@F2<ML2P1NV3xqIYQj2A{kPQr#Fh
z+PmNi@Xp2|-iwqw=NoczKtUJ;KrFbQ6~I7-+oH`Axo)5vhukB)Alm>a5IUd|d>9OM
zC>b|4AJL#sKvWm@?s4#D5mfW*H2_be8X1&Ugk5SEX5A~OR{>E4RL>qk`M^<s4PlSl
z3KL+-VYz~4=5x6f84BmX%|jSD5eiyh!`cz51g()cEay9Lq(G?Yl8{gmSWpt|#|YAs
zKs*w_B?tQwRK7#xR|bYX0RrSmO$=phenql{>SY!KSHSl|*(At-@aUXK`g#HbeiKr2
z4a_4>7f3CGDfxium~(8K3T(0eE4e5qdkjVVA(o475~BJuNSut+3<lD0V#0Bj3oaUr
zS2k=GplT!bhQ*9R`ro|{Jz<98M0W?%h*ASc4@5h_`{L2@jwCmO*e-%3L0X8XIbrM%
z+N>e41;k8pz#V^f|3N)7KK>4jcSP`oRmceG42bMBiF|`o5`IRL0I|^^F8v|gjt^!(
z9CxJ;m$eaL3{^dW>9~=6Ea#s>Rfv}&+$gy9h%t^wHU_{YvKZxfen#I2psawY1e6|7
zKoVPg11Ao?2m#!d-xhA9bGO7tM&5!T2CRLijr@yU8+O_t`a^;aFr8f=J(|-3l(v@Q
zU&;`3?}xxi<^NLl-wV2!3s5>w0J{nt5Ga8N0U=r_V;GxOI1L-R^5esDqFcn;5Y&M6
zkcm7Ms6U4?F;tBwq^=8xm!wLi(C>htAePBbaml=}W3mqO3iYnloxepK3wtE(S$H`R
zm|=;m0r?=BE-A&L)>&D1K{(jHWy37N)ARiNt!W9^Xv9me@$uzAw3Ol7*~{<=NMiZd
zM#F7LhtuAteI0I<?t8vE;=?JW6ofZA{7!qBnK{?@9vujLxI}3>H;3LdI<=oKId9zU
zGY6WKDvqMULhGU^8nRf(g+e?yIy%}xG)vMg?5G-cQz<AtJUY4=wq_8IcW#EV+=?cF
zlF@ZVNLOQ>mA<b8m4t8^NV5g8<+6py(aIk|qeg)xOD@9b%e08-PPAZeg1--lOXd2?
z1(vfQYy^7>MM%50K1MJ%TOc@90ugNY^SeYdL16g{LCG-L{HZ$a!YjhU7p<%o9rjHu
zaqJTjz6!?xr*pOA;*t1@?hdM?{r?o@pFSu(JRc}6$2xBseedscrVoC$TcSxxE_)GU
zK)~=V+99`_-W%09{TIqqw3QYNF*wh@5O{*n0x|)?Fc48nMuX-mp!`OFbmgtMhoyi(
zopOH<b6ltRF%TY!a=`|S!gzqRLrWXjbKvm`wrxOiViu);)+bxQjf{me2g#=Z3Vtx+
zj7EQQerM&KTDp<St4Z@7FKxG2C&$h#kXggtnF{rJD4B$Oe{d-R!8AvNy3#$!8ty>&
z;Rk3lz}1}w3U4}`i4ZVZEnP%i8pPLSrS6(pnBTw{rEI_da!a@IDod)2=dt+M*jC8-
zA(z|$4oY$`IgJkw_W6sV5)%<Ju8#{MSy16p+*YK5Jjs^TucDW$QsX7@$OPSa3f19p
zD-rB{xBAgGyFl_e@Lw`!z*0+sOkbNbl(;6qACMptjxeNBfXoefvOJiTw<HY{5)&<5
zsmaJr87DbH1crvSe$@Dr#^6@$m7~pLCh!Pu=BE6~@NH^*>#&!|(*C1fN{Qmz`NK1{
zdLScW{gZJc=J)Y6dxRN<kwJBSVDh5Vl3Hjd@R<wv-96yVzrG66DnU>bDX6;GKqOy@
z?Z178y5T6Xg`{B~xtcXe^eZD@9oEs`Cqk|>GtYBsYt%>oKK`+#<pk8Gz<Upl9fPtC
zD2SY^LOEavB+x;L1PUvFEE<&s!+u0lEAI_aJ9Gp$ZT<zrA&@i#jlw7mLm^U);2m(B
z1N{)T*et)lPipv=Di@t~A;U-|-fo9O>_c$MU2~1E<?j|rd3(DFgjCcQh{QnP0d;dI
z7U&8^RaoExppgaOhtP0&MFrw#LeU@ZgAp;v45Yl0hiDQ+a9&@nEGc2C5G|N2sC{z{
zf!Np-<Y*<sZ*)byK0o!kZcbWuc%ZQ>VQx_kDke#4LZ7s1ec5gqP=qh^27?Dg5@ImP
zBq2wsS!G8HekP*Ectidk^;H1l4+|LuNI(nKk%-?3ZWBm+koS(GlLN&Kv?-whwSfhK
zk(}QJUzV*JF?glh!9p2<QW&I9gfj|aN>>o~A0oB>&u@UMa_#z`?I*JkLxzI5Zg7D>
zB!HS2fZKBt9nQnduu}!(7#)N@|M3&&qKB@<%8X8b59b-sY##^yPOq~oFE{t#4>yQQ
z0`g^vXz1t{CbzdkG#?APVSBIp)H94y3@9zk9d$Ah-`)CLfs1jn`VpF1Yjmcv)#)#4
zFo4kh#)e2O$LLM#Y(t%#FAU$PqUa#<f}Tl0LMgtMCu6%<q0Dm7u9S6xG=UTqkpOIP
z&`<Ae)64chnu#wsf4ikd+Jtue)BHpsM(@yeb?n`z6jl2-%<wDJ|NM%oUsB-g11G9q
z>a%Z$>eZ$B+UWl>0RwWb=;oLm{CbW#YjuyvOt6d4Cc@&Mjw}j2zcUW)W^CwsAI6@w
zxCclvqt{4$dYZObB_e&O)O?r$^WA1Oq)siNnnt4fpH3_(L96F<7zB7vxZJTi80aj6
z-|GMSTjHh975X+>R=P03^FN-J`_=f~Gyd|gfBN&Q__Sf<n?a}246`VpBvM5&K3#>c
z!jV%VRvy@*&&-YrkCeY<DX@<3btThMT1uKxq|54aiYQxZbgW+gOH6o5k!gRQF7Oo%
z)wB)eJ}qD_Y~Ql(9_l|-{rmkyX-qgB+PW3K-QjPu%&({9WD~_jx3-+V&RfJU38qBG
zUeh!?EziY5dFFY$pzxqK+`1rFiH}LS@WM?!&1#<oPI81{KCP5Dzl9TM@%hmJ7a^Z@
z>~Z#Q;qMrPYL*NfayV4Ko1S!8oz-gm-YGO?ki0%HG4Y|><edL78&`^%mZ{OpR=F(q
zbkEJqH}l%JT%PsU-+si_eO2bo)t0O`Et)NT6xYilJ;x+YqW6}5A#?F(*x4`<0jk6%
zyIiHP(T1*|$b_J09f^KQ8>Y6_p|caKyriZ&wGI)pAr#}rW-+&KnS~~2xg{LZ4O^D0
zzBw6&Q=zFD?ctzG1xs?>6%BR(oAQ7&|Jg9pJ&*9{&b~#K-qkAcNcGgKZn7|@D+Ug6
z=|f}|wQPfVUQrcPgj`LItd2`Gv5u}k%by9`=6Se(hh2uYgY3tbH+l4Ux1v?hiPtG4
zp(NF}5ct6%Mcby%Dxv8me_aZ^uSE8S=DT=G1Sf1>l>3qYQYUpg8#a4Q=eHExzDa0I
zwN`!L3$U%CVORqQGlp5|Eu8&n`>LWxi>QZYM=r-XyQ*na5shw`nC13HrfH4eHwi7w
zBFA4D%?zz13$2LdPdfZ%n?85nIy5<kpVXl|<9s^tk<QzEUO+f7w<egn^4XcsPn(HD
zzo`*;P18h{%hY$=yCoxV#$ujHBEB-0>x1p}1>Lf1XF~)u%@jsn>%P06Z2$O$z_ZP-
z=rK{jn)V%kNe~mEKxcgT8@A0#a=f_NFSVvZIuwT`&)6MBikAlM<{!NLYGRYkrNdxp
zYZB2`ZSBTrTM@gPI(7xN5-<m4>be76rb7HB%NT9_eXl;7=<7btezne=L#~92?>fEy
zLv>4KBExkjm)k~B18e4=e0y)WMQwilr$17FA>@0dEd*8=GwrUQk=J=778L4{eeiKM
z{%sc7`mS|F>_QTUWV?ImE%uHVvlArQ57lQi6AvFf{@T!#5n|%1dYV>-AEuP6uYB<Z
z27P|{iWgJUwP8BCiHTc#ypA{AeguBWQue9KTL1XU{)wppcicUjrmTmtmeqJsAl*U%
z+4jl0Ek-Su=~TKx73=IqnXH^APeTl-D}$dCusM)Z_wc)wl`F>W2D<p2&!*K)l^1_-
zd~P*$Zgk%<slh7+cbg+T^Dc{%ZWHA>Gw~*ZmX9(=F=l*d&XPUF2b~LgdZ@{%J)LYS
zo^&sp`cPa%#Z*?qW+vvUr<GRda~K^tXX)lvP3Rwyt>};UKb6f~P=22IIEyI<=R8tr
ztt&ow{ps47wfT9D@%y^m9f=hItH`Win7P@BeLFJla~T;qS)g9W0l%4<!86+v`U~m>
zRy*?#mUVtuf6*Etkr-&X469@ErQkaGVWGemO=2xnEoc&v&NM3b)<SB7%j>(`+XSQC
zpid;ZDy8R~lNXiS_@~Txo>|7{PJ2xCBMm-v4M5bKoJ&|HVquRQ02o(5@bhk|jg{oN
zz5UIkFn)yxmJdQ_gsFtiO6y&m6BsPiBqO7Y%Ney?&pi6_Zt8b3N6RTU`X@1aiw;-K
zwJF1H31uhd&z*_+-RloGc<J4olNJ?gZ&_2ZU6e_iNq+_Vc=PSbsp}#b23XzEr%H26
z_dMrp8<iX31f(IO(V?f~J(e7~FE7G7%R6XaWOPHLIMnf?CI3*PvbEN+t7HmKks<GN
z2LS2kO~T=MWoouotKIRR+1t_|2<Br&93Ct@SeAqPm29dS;u)Ip5I*X<h5wEBEY`g^
zJ6YfJbY=o|q510?c+z^aB-Na`ocyuHmKr2&+;_eVTZg6Q)4)BCTTZc5>Jh`^Ma(};
zLe7*-n5r&IoDn-^@35fPp%9r7dt_2dQelC1R(7_^-35U|!I_}qG7jh}5`2&Q(<;km
zIL>ACsy`+luRbk)T8V42f3nK%@Cj|7pvxf@0Y06SMt@mMIcMejoF`#@(X8bxgN-uD
z`(=Ox6A77p&Lws9iY{)i(VY)L8_p!CKU^YLjp^OC50Z^m-5a;4x&*cO$&>RnattpD
zT@+G!k(=t!CE<onHnAY}J32D0USEoSA?4s?$2^I!lD(275pycc??24jj%=$ob2e2s
zE;}xe@ZIC%*H<9Mg>^GN#_A1|`{PJ^D(vWX{Z1A-kL^B(Pn-0X#zm(@7ZxOR22&(b
zNV@dCn$kIShqaB+y!x8j#$62@iKMr;ch<j-Hkh`2zucwBul6kH`+fD-wi*l^c+S{)
zY3gwexyQIZ96tIy%bnNr^?g5e0r&lx@a?9?bKRX|2jGXY(_Zz#<s=LCw{J3Pk{&h_
z)&C%4EwPMK9+IuQo;&`uygHaWAd=zCeP*XQ?PFx;Mvt-aD|l96>(r?9_e_q~o6E1%
zv|bmsnwZfq)xncB*1;;ccGZ)BN}8!TG0!7X9M5!?*LAEWbYbLTiF2oLscILAJJcMx
zM6FUJ=erynxD~bNca;{jDjPW!=_TVhuQ~i|eCvDYMn=czpc;E?$b;2eW@2);Vcl7(
zb?FP1)Z-g%QC|p0Ux;m~-{1V{kJ~@+b8mmnnR--)&Ul_VZzxwxjDtoe@3Uw*FL93G
zncS-L&z?qNUZrhGNy&N-?%biaKRhT>I2HY@YtjLdc7vy?N_kf$<G%{Yom|H2FtU+0
zsW6Gvdc<IzM#V69O9;<h;)Ce=;f25Vx^iW#r_{rwwzG9^k$*TVNH@k|E3A}hj2?^4
zZ`3o%ZtL}Bfqm|fnul$T1*ej}9XG(8n6(Jd!Y!RuY#=)xj8*Bbkfj?Upf)%1NlJdH
zrd4##NOQY`cHv~zz{X7Ch-K%m<tHo6S&FwwlGmF%u#bs$JUSua^D~-?uz4vkot;0H
zQlnR8NbRYXsd55O^`MB*9VTJ^sR-9U9Z!^UH>EE*t`>63vn<~%;wGL6SI=P(;!SE;
z-Rr5yw0<yB>hm0*J@VDvZguD&!f}UViq$@OQSp7vtut?Q0@5#^xv!PRd+Jb6v4PR|
zFe2i#<dUC?)pcqSpO2<&s<yfX!lLTDvu{@Pa2qPrK5skTm}s+c{8db;BjE5nJ6YIb
z=k2LkQj_ZjRToo(;y(8TGxmPpElyK5`T0g`#M?7V==nmqcBr4PVzre<w?krpTHHAc
zH^vC^Fkh-zVP)zQ#1}J#+Fga1WgkuXUJKKcimAS8Q%a^N!S7@3_e?kezbm0@PypZw
z7Ex8_9p7IIegqSTU9>7#J=Ns#+h=AYnu%JID(~mUjEfB~dL((2fS^C26;~-3UW}|N
z%d0Jq)~Vl0r|%s%^YWAZ%0)e-ZA|~9yq0*;?+gVQr9gvLDrFqa`CkJJHu@oR@3L9<
z>b6f7K5|vbx5(n+JL|wG5|f?x?9Ba^WAxq<($?pANw}yLZYf>5x<2MxA|^eP(M9u1
zZh?zEi1sS&6ah{0>THjXI!1$Z8bFB$O(S9FOy)*dEf|}=(rIi*uA01XU|?a#Il*vG
z>lE%O57r$+ax3Y%*5vj4={tCB6q@lJ)l~>+t*NeNjpKQ6YP->3Y|d`-)Xv=bf@OU6
z`t#Q)_WpRgt>g973_GL`H=F7RIYsTiC_3*Y>bE?$*OM!H@ba!L&8Ve0)Q?FlKlEVH
z%e(ax6D!Fw)TGDvD6KPUC?+NGPT17tu70l0d6mh4I%hmaN}jxe+utxX+)_1h>#pe+
zdZ?udmvO^2XRp1n&gzhQE0K`NgmXy0XH<o=eXILK@0j3+<>@4a)K|tT%v0&XY)w^a
z%lXjkE5-iBn?#2f+y?H`L?s3JKk&Yh-u#0<KF)RH?XeX-Z+*!|npCmH%9rzL!5sC5
zX<RB3U-NKyC^aT`LYY7%*xAEcT5$ODbosriz6CW~swCM%r&BJIDu=?S)-1N+A!gOd
z@oR%l=t<-Zz2>9$Ez&LXNLX;@*O4RIwpB2y7mm`>Q*lcBLCeECwcgD6Wjd8jM4-E{
z;yowxv6W{}lQ)|ttgYr_*d(><Vp;KvF5pe+ReAZiT6<KyW>Iyu6bfPQUUUAHm45eq
zm=Ud<AHDaC-&0Dbq;uA-@B3m(yqKHEyqBX4R&~rph_yr8w>`?+OboKJTV#1&k&XAh
z*Kg$a8oxq@|825!U66Z0HMx~TwXMI#iQ2}!uCdP7_$`}_3EL~pd}B)0l=$5DUpS}M
z>0j$Qsd%S!j8e~ckUNXCO7P8!J@UpKfRfu3VV^f_4?bkp={VhP-F1hvkK!BtwnMi`
z|4)n0o!aqQie0&+&a9tWDxVZj<Rld{iocKUwKZPfDc~X3I9W-BmGjuQ<io^wh8;)h
zrF+Lh^q-kY`Z)I7Qp+D^&7A+a8+*Nyk8#{t-{WhI<M^DNjbEDsJ$+=E*S9o_N{XE~
zcDGGp`|hO=El3TRjQZcb({VT3LzENGY+qN^R6A15QuS91S7ZoX`rYAlwKtiWIcX~M
zK?UFYvqz>>bGbrot?(4L)))IM%6`h2?k4=Y=F=?{#s&Jey|IZ03=E;sBej|@ER?bs
zl&8tQX^xQNTZ<-2%{nQC8hmbgX)XJQQDke%f~qX=i+Gc9=aEsTF;z6{TAz~amrUEy
zwe&fEUXutnK~N>2EnAtG;g9|crttq1JZAo~`R$Fib!#=VZtB`7aUC`L1OEon#;wcb
zU8<&vt0bo{;I4!?=&I#t+_+#Lrmrx1Y1-+{gkJTDY%%LX@?~W<yM8lzSH2<lhhNQi
z+k_5{U+*xzVq3|msP_MI1*g1+r@OEyBR<CXh3|O`J1@KB+`^D^k*wOYy!hOpgVbia
zKS@_xz@1|9_)8s-!oTlkW_r@O_GWV2xTU5fovYiay*kb*r7}-RI?XcSSE@`wjR}kP
zdk&KV4aqg{3#$g(#wu{wwHMA*N@Yps@fWIe=N4Sg(0@WgcGYRA^wWmeKJmffHCgxg
zB@?0Uf?@+x8xJq}T@I;$oZ_0>Tfy`kUDt}9hsa9v$)1k7bS4c%>hJRJwoc6^1T!1=
zKJaFJX~z()s^RreVB6@usB@La>ez%rNIAdme*T)nWtY8t0g=9+>>cYCNh1z1vVYQZ
z3)NKa*oLZ=tLjlvcqg}JpMKI^Mf)e~6dC`cRc@~_!KTlD%&JIfVhWS|5>AN7*YBtm
zgy~CoQRt3H&rBUYHN8FIM=sGHD+dvM2taQ_sdn_A!*^YelCnf3>w3-G1&qx9M2x8r
zy*}$U_eO}CfrmOqxmkFUwBj&j(R{|c+V7&eJxf3>CW0*J;k^@?>clD4F8Et6`f*=z
z-DK2)0awUvQuv)^x8k%k=C1P0?~jiBF52~Ng@v21wmy8&9OLhK?jo<EuOApg%CY{g
z5?_~cI4!xDRee^%(Xgg2hClh_U{;4CwXbzmcX`>ibiHNAYASUOEbMiw#n6PB@g||5
zbi1nu6gev?lilC0C)GP9I&Oa;S6KV4==^M9r}p>M>ZAg8+{qn*KZd!z4k?4BG+gZ5
zc3V9+@bgbw^ciNoZtQI=a=L;i(M&waM#QNtylWTBdo{%R1s}o40Xdr!^RB*=kyF<8
zWs~b4a_-7@yV@-Ve$`Oz#%OpqjvmwoGdpv_nfym1tt-hexmbN#fl!Qy=tZW<;#eQd
znv;KuM{g_DDET}*%RTm9v1{4Y;l+G`_m8o(-*z9*o!CsTn^9`X`{hJA6wb<MTjtoq
ztW0=+^~>|5?~cjddgW_ByWBecPch@)@Z9D$TS?)h_rzDGF=m}JpD~yExu?6It6wT2
z|4H7BZ$Zzs^}{Vas>|ZlnaVG(mmPb==h`pMCw#T*1drBSn1?ID-7_-7^QayWu~T$(
zMT+0Ct8sp>JNvS1Vr?VXXgr3)SfrPb{?Yg2ndZ(*0xmD~<#iZ;HeSrgcYEQkJ{}|a
z*lBCkr_Mxx>K^u=II$(3ovEo{$CEG4JyExPIz^=8wtE_K5YD1zqhIJJ=065x+$$m^
zN^W0$uAuzwLjviyx+?1!=%M#}#X^3J|E?<g?P_quOzGX(w=Bl{G5Dv#8u8Q?ozD|R
ztk-s9F^TtaRUUUVD{*F@k2TIPFCPtflde=(q>%C>RyhmSHARWNc&oMC0AboNPPLlx
znTX>hN$`AUa2wh<7>mmGSMSkh9y~f>Yt8pa&q6gOj<RNIBFVXzjn$mzSR#H$mDqH#
zpWUC;VBfWIjq%wy`xEvAN~}1`hDLoiEbZ1R495z2o4?se432FF?)^Du_U)H_vk@*c
z>EZXZZz{I_4Td*8>-h1A9X&`K*^0-!7x}!c!-mg$y0<2?491TCP~v_M)llt4H&cT*
z#k?qv&v`vNT*NK*OtpV;XI)kLm}V&KE2Xba?YU11yimB~$Tu<8p^`7~8Le?Wi)s1f
z&fTQTnHB9jm$z2&;#^GWpJ*ID`9UPn-+Ri`G^yhC^qUxx94a<`qX?OTatg}>U!xdB
zZC%d<3QoLNUinv@SGil>PI3%Z;=VM{CJg;;n3UI2SK6V_%nHH!B6$s0&x?B>X^NKo
zdA_|o6ezK0nU75%md14XJg=PGooRM{EbR8wYC8(AD|wmCt*-&6nA;&odJ88$fn+0(
zH3r8+z(mL<#eZbc&*omDNb+VN-;Lz+ExEo-@{Hp4S9$*I-@mGAx$s4;rNO~@Z$HXD
ztgfqNzjw6G=($4<jmCgphUHf$?(*o#YFD|eCzd_WUmAk9|9S8M{>_GW+rkrHZffom
z?wMU4sf~5&lJlqcHW?J<l-_&Dl?XF(Zm-8{HPFH;_sYj&E)G72oV!=a*l_ICtW#M2
zm<eq_oKR_9DXGS>)2_62_s^W;Tb5y!sle1S8q><|AAcU`@9FHn?pYWsDQn!_@1xqA
z!Lcv3c(DKNm<tW#m!H4hm%Fcze$_-#(#bxmlYYiOG2eyoSZl>TUa>1s@4tHejL8+O
zJ%BC8Ls2PI^JPMU0UG<nL&4j$_T!00KP0z=A{PV!8p<aF8Gb9&b9tj`AtV2F4_j1t
zBkEPJIps>YO*Am#A&{uApfhJ@9BfeSr!7(YCB`)?Lh#~Zje*gwlRu}CvLP4yN7tPj
zj@xG;H1N3-bUPEPlVJ^3tufVGx(idV3&dA{D>nQ!jsLD&PNELuV#!n3TxzYsU%E-b
z$1Z&T+fvh@g<67ov8&M>JtICY020-zMc$T@9Bj+9TFYuU_5L7_#@z_gg<?i1yS{_u
z%#cEohVl$82<St03Q)#AfMPOXJ@kUc$4Co{YMKzG9@52J<{vLXZUy1~qBl1;k)+Rm
zZgw`y$OS??(sHu&TB+SVI~NXW2yeH~oBw(?c3C%-)B1@FUXKv#$Dsxsc+v^;7ZG9>
z(&8`n>*p09UQ9Pd$L~2e=DVuBSN>pzt%nUFHu9~+Z1_?2mcNEl%#T^fS~+UhO@`=y
z&089~(3{m%d86U*N*rj4J8*|A26F{S6`1EqAky2+;@&iVwPaSU%4OGpu!?!A@wU<f
znLDjfVbJX9I;7HpFLCxAJZ%T0k)ccJ38;pBT(JS-#mk_c2aOIk6pBbdWgO%%(v<lB
zY7unyu!#GE23R%S9V}X-#zZ=HEKc%@w9fQm&v7*FJy$4qlYGmq=P&+Ytc7L_r@rOk
z(}$<btn@BS-g$a%XlF4;b~Gqp>Z2_|(*EMfLcOi*hmL%ze!?G=AQ{);xZxUq{FnE=
z90L=Loew8peJaSjr)~qetba5_NCov@HAHiOD4D2Tt)GH+W{8muAV4U3JE}A6{6|uR
zusPpqcwi=GL$PUf1p|~g4i0*)h5!QVl<_FFOn%Sfl&@3$^Uf0L8cGhY9nv4S{NB;{
z*;brR15+%c)MGlxP?$euGcKyzQ+>o!qpvA8Zp=b#r&N&YH$$Xvyz0Xup{@?mqQ06t
zcUgwnboqWg9=*{;@ek^$Oo&;9Q(Mh@m@oZcou|8q=BCy@SB=QnOV{!qmmvUsUio#0
zQ-wK5`L#;v98~U5B3=~~(4dmOrVS`u2!l#G<Hez_pP+=g4y9?(We6%aAw7!fy_yyw
z)ChrIENF-cIyBVZ6QRYS7V^f8X%C;9@^g5%TEojA`+m+!h`KKQjzX&wV`bF)p&F-A
zUZs%Vf&LE9pQb<Ek&>Nkx~k+vZe5mJ(9*|8;`mkB(;P5S?LOLZx4Kwy?eE@rE^qRF
zg5v$C!vmenub{(Ehw>JL8S0BJl7E_$HMort2?Z%Bh$C8{Hp~KasX#siAPFb*+>1i>
zY#*~^$!_odPx6=lAtxBLhxw;3hEK5*OWih==dA?JhYLTaIj^*2+3##iu~St^Y~01|
zU(uc%d&v?a;CA0a<@GZ+rB<!Pgnzm@LEobg&N|lLAx(<<qqiL6jDIHA%xb+Yu{RAJ
zPk61Iim9opN>k6TcJA6aISGo7&aJ&VxY27+y$x+4aJ&H)1qnTbo<)u2;%DrzEur`x
zkWW2OhV5t#Dq<3K_0Pouz;5_g3Xv)3!bMi?kM?Oac&YjFBfn*J8rHmt1P>+}eeS*S
zsJ`NiJ&G%eZ%A<6Z<|o=^4~Isjm+*Aa|0~T;dg<Dx4*e!7O3hg9rX(D+sEd5D#-<{
z)a$jNKJbt;e+@{zCN8M9ha_jPzi*iglwQS7=hA#XE?7yh3PK(z(*rh~3_-b0^l=UR
zQ+niI<Jm9o**lff^6Sa|*d~LRjzLj!gMNPx5$q3&mbm5u=W*iq0vnmPguO8cm<GfG
zV66tjc3uKH3Pi8gIC1L42_;Y)LwOsX7z`7VtAYgu@l%BaE<Y%iBPV<hf|3EiQGlok
zbuWiP<lBJHDAsQve6B)_2S!nkWRdHq;`RjLVg9%LPaX{u?Kn@Ic}r1s{$m-byp4oQ
zZ#Luxl|DrW9xwSHYuKTaI`$<Nmd;mgwg80(Q3T3;LWS)=d?ch7_2>FWYdi&IAJBOK
z={3($-SE!OAN9|EtN7wY-<lh~$36AgkF%$Hm#XEbb8E$utMBapQnNTP#4A`fz90rM
z_jzWbGnd4mh6?l`{Pm>3>iP-ANl8#h4Km|5I*a$9C{u}7A=>cMFH)hemAnVbd4v@c
zTp9{vQ-d^zb`$UJCN`)({k@!*&YpK~7EGY?Cf**4XQGr|zP`v2@}RvdgK}cY=8mPZ
zfLCE{^Ovx51ojIyMRU(kg~QC<gn2CD^jEHfj@hfvdiW_xdtQzet&OV_(#k5lthknO
zqlicyQ`ehf>*7x$yN~kYP}I^m5IE4-nw0I6vg7hfeNy8Gu0asQ<F9ncSxfGl<tx`#
zQ{JhRbTAy|anknWv{CYt*H7AtBmLX8tmUFv4R1)p;~Mdu(yu~fsv7Lwgp<rs#}(tb
zWhouCU5X_J*!42$Nk0_&fb<%=f%Ac${N3o=y@QUwt4t{QM#e$?hkCrYz;j_H%Wcoh
zcOC9hh^*4q`P>p>nAs+h3%FJCh8JQDt_M_*l$zzmIoJDX#(hKSXn*w0Au>ZZ3HI^F
z)n3PCowI*<b*}7<Sf|iv#^J;)2bn<aqgRK-B}k^q_-3c$eUs!%Z}Jw<)XuLaA%O1B
zG_xZW`9GYquXVGF|4k%|2ZDHL5Z7+5JojquiUxshP0BuH)8}h*Oxrf}08RE~szjN@
z(@K+6Nri6ARahTR47K#PCVkbI(&AgZcBevI#y0JIix|Y)p}6#kOji2Jb+sRS<qUcU
zBz-xo$YNR)74EEC?5AWBS31fzi{~2ZzYi+1<x86ych~%!=*Y;D!MC<Hd)f(oCL$cz
z&9vgV8Oph9)>;XV2eg$lUgEkZH@+OB(>xrzJX%eP7G!!i3_JRq$(*@F9KnGTg)+XC
zAnuBu?EU?v%T#;II8eloM?7hNl>1IsRe`6pN$I21EGuP&AWlS2#2)vb(=O?me2M{W
zv?viD!1Z3;NP3QMakEiTv|#yam5Z=X>cJyl>;_-5%|a?aCa=E(VG8TXBeA;AOIsGK
zC*O398#Oa5i>5TQ>c%?Kd<p?vsP+^la%jj?2A`&df!&_pBR4ed4khU*MT8p<LMAe;
zoEg3e(aH#;#P6+Wt?zeP`fRmjJrZkQGTA-&s`km#A^wGiA<~URTcQ+^FVnz-rK@fR
z0W`XjJ4X4X^s}afq9251M*O=nbEIMT`HyM>#H$s{B3Spc%bP1rOo&g&F9l4tDV!s^
z$c`o#%2~mh@h?w4aq-mfR7)l0{W<fFr}w1y_C6~e{GKP&hS6&L`x<9LHL%I_UVMoZ
zC}e$ljL}7Rsx9GJmHqrpX}-Gw$~XUBDH5*q0l%P4_B~#y<7>KK@?4a2G)AjoWPRB%
zBd%Hhz42k8MUBj{H?f1bua4m)ec#8oGNX?EE@tpIO|P_~^*MgPKuotL#3rDwjKtnW
zWoLhLpn1=lq0XoAZ`nelv1+kK#A6lh$?XSf7{UxTEzw1$;odQhiP$si09M^OT>j>!
zqq=n<x;EThQcO7cgs$pk0A*U0uSG(9>}(|N&z7wU*Cy`8*eg=(78cdhb(VZ(Z{!q<
z6KC`SX9!uUqM2EWj;AvPcrUGv(Hxi(l3*V<olP3vTwhza6Y1u!*MDs4zwI%bbX{CL
zlSRFP^asJ0V;}r0{d+5pSK3S)Z8l~E=z8LusC*u*ORW@gA6rUXJZ~_~4|n0@OqyEd
zZ;UrlL+1US;2S!>r)nZtFXugefC&tcDt|*N_VJ_Xg*AVgB+IXt1TWK!c{-dG^>ilO
z)xo1bVRUqvVYrO{orkx+GEyGH$gjvs{DjNwCFlEO%Atz{K--?hkJj;gC+p|&&G_ya
zn@vys@uF!8V?7kplYd#-bvaM@=yNNR-`|q@OPoyCRlpa+P7<&?VP)DpvA6DI6ep=_
zGJ+u_Ao$kQFY_#owukdaA!abo>e=Xv6C*X#Hy-aifp0r|bd^rUhTQA^cb88$=!K;9
zVFK+nN=)6UdN0<QF7*dcpP>|^qf38QQquo&aoqI6&xaVDh0nBn`IxV<`~JzPsqdDT
zjv2+>QDy#nt(SU6W@nSX5FFGAzwE=9oy#)l`FVq$L5Js|g*Zv;$?$-HS4nJa&Of+H
zm`^qgV1-qGo%&UH>%rmb>d3*^Sj3B#Au%qRqf4>8DvBfWm!A(f=UJE5XBi%CM}nTN
zdxAS40EXcAy}zHjX_hOIb8z>XvBRk<UOO?phqMpbMc>0!aF70-+UMToJ7*L6u`Z@=
zJyFkpUwwIat>>q{{qK(?MJFRW4AJ}e8Q?^!ef*cz6S8b4F>Cx&({scI!*gEuzrG#|
zcf39}GDGm~8QvuV0wogrO}l2?D8V$9!`6=r+&8?~E?Q-`(XJ1x!nLFg(6v<koCK~4
z{Im{!x%$u-zIu9&bmAF@g5hH1RD18Z<)ws^yxzbV9(q^qgRSSvh)F)h?epo@Q``pK
zwnTFK8`YwI`98)@N4K6AsGcM7?z-NE-!^v!*Bx%XU#8o8vp`&bxXn3Q6JfTrA=Ps9
ziq)=bxSTqB5GPu5_5n_^A8DNYs2EJBz0x?i6xGqCEWOop{Jd-x&OZMUP!p1@7_Jfg
zO+U5KC6#K%^3u?aqg&!>&q@(KxzCCTYxsCpK=vwoYWKUudHbFHIningDIWi$5x^U;
z<S{>Vx_mb5wD^8@FH<*1Yl<!5`f_7NyG(;M+|r4oTOyY(ta*50EnD=7Kld}gEK;v4
zruTcgpYB}cC0A}763Y`gx~pKR{Gl|Mx@VuC^?kId&D6=!`k^nU)Z#tC=-XH;xY*01
zizRq#J1<=u#}jp;oW;Ynx~SZ1Oy@p(lj_ermj-rjXDJpa*|`WP3Qt<6J_wi~@G<|G
zP8x(GQ{u8r{vavUyPLb{r1H?w%P}@g`u1Rb@3^QFzBQICT<ODi@t*}xZ~Sgzh_>Si
z&%>a(PDhm4XZi7W;K>UjSI^_!t4XUWIsWHUYD;U_WQ09K+sI5r!qF-=ZcNg#?(@3!
z=%;Jx2{vn|DhGC4&a!i~26Gbdg;{@Z%O3h_)PJM6WUS4|ZMk(J=(&vD{o91R-Ot-X
zh~ZN$e?LVO8uod-dVP6qFd(+%b-CD=UAE?Kk$|0LGDT$_!=;V(;P|`PJE@0zr=a$<
zq_p%Vs7dUAZ~E-J7sMJ6x(`~af`-<I@4G9un_E_cRcZPkG%CCylGv$5E2k4aok_9Q
zQBtPgaUS5`v$YW_I!u7K)h-t=fqj99=qb1!v@9$*Ak#tZ9}qembZH`M2Z;L*HYN&X
z0N8{O3N-ma*SrSuM0NL6n8w-7?^`H79F$#k58d#AgQH0AtzUT={>0j+O{Q7POZ#|b
z<``hshJh;e4A3tmK{|u{HDDeXfSeB>!U+3+)}b8sGN=RL(V#9*66jvJB=tcnQ2_z%
zgCL!Lv!p{qJ0S2r-`YBw7<ZC4we2_lxy!-{`97+=xJG1aE;u+i=V@rd5Xrq2OskZ~
zzb&AH@wm@dB((-@7Y_(r0-%!tFk8-nXhH(YDzmoFV0L$R+dq8Z8}04vJcc1AB~33W
zS&MM^F5A$M_q~`|>9(Xde~dUDIo2%q9ra(@KY!nmrEk%1s&|%_YDfx!XMCw!eDXZg
z<*6oBC-#_wLnZ1fW0>rmoQ$ff3s7Bul}8UKVxVSQ@c9ZY1n8i$`sU4>=Lv{ONG3oL
zz~gaTC&BX=s@cB==;A-Epd&vcC|DaL$r=Y5Ew!%4cLOQ7j&<kKTehFRmQ(FsP<E1>
zu^=MOG%!Oj(qP{F+11L&zA<GRqNMDrI<vDI_tfO-u?m8r=?8QfqC{j%G(|v9A{T!R
z3v?3V&=(|YTM&e4Aky!!6J%w@1yu`-w|`WWidwXdqNZj5eDO;F&H^2~px^Ta4v+Ei
zcopfvaDw~t4B=Q$9M8;;)UL2?*kTPocJ|h{yS=kzb3EBki)vqumLyqrpV8IQbOY}`
zh$=uS5f055ECE%I2g034<GdgTIgN|kBzmxpj}e2K*zl;Rrx0)<ha%U)K@qHX5F3Ni
z>Ftnv+S=-0pWks_`KtPg_VT+VnRNePjALSLZ1T5>mF3{A)TNeX)>x^L$K+(8V`3j7
zL|#p4l|LuVPJio|+q_%E6y-$;J%Rl|pSuT>%D*%+k`Vm?prc*bl^s1je(LuG66JHS
zF#%Fpzof;*jcp!WEF52mdzRDo#<wto!6tM1{{FM|8zszVTKqZ>rQ$nHH(0$fNc96i
zTS74UQO63zpG``lLd<du;<^D$2;(^0jfDX?AU?#vfMc9Gad{kCRKs6<_;~u@n+Tlg
z(dmJyloaK8A^PbGQ{5B!@Dv1dVObe-r)dVQ3%ZBI+;6~Qc?PsMsA4q$hEEj;ltJc-
zq*|2nlU2~zJeQ;abUd&oVC9X24#*d{RDd0N27I%IQMBHb3Gba^J${(MlXybs_EMs_
zapvr}mm_|LXR+kzrQr*tu>wYMQatT!6k@1x-5pog9c(XBO3Hf+70_@9YHC4Wq&Dvf
zRs*vVktkmkU$LM5r{vc8DsHOJZu7mC<5Dz(R+A5wHeUJ1?{T7IOtt(06<eRx8;;+h
zR@j1Otj5_hlU|#Xr7rvA#H7TOwL*DVM?1Z0p1mNSuiT<<Z1N)3F{(ckT!y9>#XCRC
zfuS-Kj82w_pwA=s(&t{OyZ$P#C@D4xd7)uzG}U%y$7I=pc-E0eG56>!Ab9&_P_q4v
z*p5>n{%O0t9Aohh0s*25W*2I?^G+Up)oILqYqP){$(^*ZoriXIrIMe@zm9Mpo_-uE
z67u9N9hfR=0q6?}V*1;MNnmGrsmAAmd1qVN5s=_~z9YxxLN!+&%h9ENh=5t>y3rE*
zL-eIE6;0Vqb$9A|U034LF?t6cMXaM6Qc4i=pl({T@G;*P$Hu%1Og(IdbG*tUGT&$R
z=)nktqhmr&)I6(rw`sfcKDPqzwJjKz=@}#YTg=mAvzBT6M>dG~On(U&g%Zcp?}qtf
zyhVRyUvihEVo#~EneKd*{`ae@wLWW(2TpvyaIFJh-ln3xx4muqdDujo8e>H2J#%<1
zNq;}*$FuhE{aS$7n@644A2B(<aty%+)7iQDZ6r&HzXTnZ=`y8nKGB++JDznPpy_pf
z5aC#cxp&}5CF-$zYQv$5!o2h7N`04e#|lXO?;KCxA8=8}Ci1F#!ZR>scQC>O@6F`M
zBo+_rZj^b}rf(kl{t(AoR#<S_l;r+m|5%At%b95tN1l&^f4`7Lr8w}qzx?v)!|%w*
z%)!r|ctLJ2CEzeLlGUfn7OabbSHk?<FYBD{N#5I0zns^m@}k<Mdfxw1@OtKX)ykZu
z>gQm++2uEj^Gd>gc^8gO>O4|sN<o8>+Kcw)U2gPCZ+(-x_i$F9k|h>VAHBd=Uw*Ka
z{}4)J{_@j2U`dAC`&aL$Aw`m-LpR;hOuvibSNTYH0#3?t5G{#kavl9rfa;N{LlDt<
zbmw%+<VWU_I7<_<UIV25+SLTPXOC`yF6PLx`0u~GR_jgU6M(_Ie*HSnV+`83wD<H#
zsH>9#>@DBpFk1{AEvkh7vB|!2mET{r-e0xnF~r_E4dJ8{n6<UFYYq%(l_^2I=}B2x
zS!vfboRC-5`<~(q3=Dv$@BxSnz?NcG{@KrwUMeXmDbHyv`__NIKEZnsD2%VHoGVmM
zzSCrXVvPW#0H9YX(LTh3h@OO~1(x%F+VkNEOf%#ks@cfN>8*%EPyW~3JbH2>6Km@b
zunM64CuME4Wd_=(S^j-wy<#*|o0Pr;_YT8Gf{d#Fb)5b^$dciyskTdmmdzkwQBqbm
zT;j?jnN!vpyq20IYogTT0c~IhMi`g~Dc@1N0WlyFNdpgd2R3*F^uZ{OM!A4G0HE!S
z(6J*Npk_-Bc_;pLfthZGw9wsAK-D1#0BW3#`VNBD55r{39^!|09}gkDA>bK=bhW_B
zLKtn}gV;^A;=wR6O*jN=6zSXqW&kH>NI^GpQcFwA9JrUIU3m`Z{(gHnu*Y1F^LC$g
zghTK2nKSPJANCY@p5r;Gj~hXFX}>-xVP5_~H_unGysM`t2B-tTnC1#z28J03y-nVE
zW5Es%&z-rDCuv~t5>%Rjk{J=8aDauu&K?U5@nj_>{q3LU-%9E_J4LF|5-pMn3dAH?
zzwWF6smG7&S!T;#y^;b2{FUOj8$kw3AiUUUCQ^`6R1BY<o1-ihn7NOwi5(Oa)I2bt
zLdv1{-~uZfTQtC5)VV1_K!&R#z9JOSqh)BA4ifQpP^)lI^4SoBCicwf)8Sysv;&#s
zEC=*fr|Yg(?bQhyO<>IWFb7F362UmcS!!T)V_)sfo!))>0$QCx54gg?5zr5{w6;o$
z{8_qAz<C%9jhjn?0Jz)^>cSZSI7G(8FwoIKL$rRtt)Icik9+%8sL8YJ^=o6m0#B~4
z+7>vlF0Ukk`iRb9uH_-9qSE`!n}>%rV%c?B0F@mFa9mk9B{?|ienv**7sW#(!cMTN
zj6u~2jSZOS=$M(AgI;viI;IFX%ZRTOP%&GuORVf|n18D&2Ow?7HPB)5Nc7A-<%*4t
zX0l-i(nJ;{^SFTQ)?^vpKdYSSB=318t7i@@RR$Io8D87(Y@i$>A&6L7dO*%&84aC0
zG#nfqC8mo_ZK3ns<jjo88}}Gry!#~53JP%`sq8Qc6TSk*pE4-A-2kMQSyG}Xjv9z&
zfh>rSjLd8#J2O)pBsrJB0!AGPjAHa?O=j3wSO^IsK<|fCGG|vSwt&5J85VFXP(CR$
zmT;;;bP2j&)a(fX3E+DffHkD7sCbF4GokC$%+Kv@zG>A)o$ptn$rHTaHz55PL`}4c
z0(Oa+1&w56g6L*9K&gaaEFkC%L5MN`;fumq-7{k|+}pcs%*+w+HlTZI5+E*R?CeTl
zuTVr#Zr8Fp;0K_#RUp;uD+y^&ZaJl)q4BxKIlr`1<1Dqb)(KGTKL=$fC>$Y_j=cjN
zk2qmOL6L3*iv!~gf$}Nnc7@u1a&d8CFyJMDAY0tPfEKpL5b~Eg)UlgDw!*@IY!x3I
z8<a%#Zs`5A0Qh)VB60`g-@vF%gN73TJqNx@8>H?K_}19ebP`AnG?@GFDojP=OlG><
z+YyHm3j<+tzQyXd!@v{geDv$V^z`(7!20+@4+cAE;?-9$r?PsJT;e#`?mU00g@8j2
z;@B{N{!f7h7RCAjCz$&OVINEnN7)qQLCaQ_q5zb4Se8@JJ>eZBBT8Ubf?N>1T5M1c
zw?KN4fPw-C6j=zc4f7BbUKk9txdDXwB|g64cINI>3sITwW^k(z^$#^o1bG_pOj;q2
zr=+NO-*wXc(IX?+=te56DNz&VrV2MVH+W=#<8!O;&ipD#`ppR3?WcehHlW<!0Sxr`
z!U7J`*$wLHp_UjRI9`8$iY5&r_dxdX2EZ#MC-68bK!%2hM&0i0G_jBqOu=A)2ir#d
zsJkSjx+hD|-kuW!jBY}NbcddRYjBB6KuE%T1ZWiMPyy*S&ab;3nv&pO7OLaI%Wu*u
znY|GQd2Mr`V(sni0dms^Ug>?fdDOrc5#d4jbrGHqJb~+rbF*VJKy(6K^$Ez4pZyv?
z<*@=(5fBN&B{1HQ<uL?$9OA)%z#B>T(6>Sw3h_qIew~?_d6|}mf%PJpr*kv7?s)F;
zj*booz?3#w?<<!V7Z(@kRQF#?iHl<_uKSe;lSly&Lo*p+B9VabYKI3$d{iDi7l|0Y
zobO;@&%lp}=#E+>IzLQdu6@J~M+4etpr0=jbi)aU6r_T9w->tvbqd(}3ESU>spgb9
z1O@K_T?)aEV!&aRfd;#Xlo<iIQiO5Ogxwe>Lyx;iSEnm%UCM386Q^_*@UuX?9tUPT
z^mLAcx!DCHOwYt51q!+=kn$vn5ZvwFs@6(L0`?7SIlPaYgKbG@@(EruxqN3I&D0J+
zc>`-}YeAnc1-rTN!QSrVj~`6HW{U;?0A%3_urYkJTjE!vrhF&EvM9&3|MU-#B!H_#
z*8t?0fW58e-j0q45QBoZ9RM^DX?P@+>nAjnB^(z3t<?pY(<~sJOioYJL(dSU@?mW)
z6B8kioe~h_%L59472x*C040!xwDJsaUx*1JfGEyl$llet3%#442egBZzsD4I+7DW=
zBLT0iAM#>=xgsTq0Ign-dOkErxQyUz64Ub7ChVJ9eZf#`5Vg1G0S!7{f9}mX;9=zE
zR!wnVV6JvuPdD&@e>H^d0emCadb_*j$`WKFg62j`44wuApf<}7W;egLEdlZg0qvpR
z%>JqrK?FEQl)V^s*vP=Qm3DQl659Vc;QwbJ!w7pP05rAibXhcS>FD6qkwU|5m_p2(
zs#}jLIZa|Z01qVt#IPbDQ4DJ)hfUrj-)fNCf>;O}bQgPTd(SZy?Xm0YoTCQ-`s=o2
zmpJHTc9@gf1JNT~C<1`qXj27VrLY~lM7Nm;YgFIp+y?AZ{!#oti}}%zfVvS{8`QzY
z;9)vvxGQ&s0F(vKQ|GP*bX6u;3P8JMq4V-&;RF{HHr!3muZdF(9&+TA?-~GWbj58p
z!E74h2{G~U7vNh7A_NthVG01mlfxNHexGOy+J45t5lq<{$EHOGP#hlZM-r+gy}%uR
z9x&=Mn^>yGJyo}{xoHAqGIT#35EA)JmXX(;+i?6TvuUAJFXY{&0G8hjTrb!<5QzhI
z4z#zk3rDYyE<H4uZZ>t(=J)VRm;o}7^8_3Yb5`r!yJ~a9oOQc<K*XPZ&QX8u?ZY1=
zWMo}HR}s(XYlj<<5T6D(Pc%ekb{h`=G$?Oh1gMlj_uvNe8HHoax22`eAeAS81mNVt
z0rLVp);3rvK$W(J^N8-!B@7S%#7#}J4Fvb!S0Q9YisvsaIAWUG+D@rt%lSZ4?TG&9
zLIDQ>etxWXo@PblvL@HRTS0M3mEFWK*!;0U{Td}WP$b~+<;$02-qI$8Qu6x=X}_@-
zFJHbK0gOgS<U^|mgxc4Hx)5=35^}7~<gQb~rd!3gVGFzo!46IUtt(Gur`I?f%f6lp
z>>kAMg$fHl;F(g?S86Ea4^4J3rZyuqA)F@I7{H<Uqi~ym00n5hB_BLsL!23)2BDVr
z{>!ftFVtf5sKN6>qk2Y&mvIGq{+`e22fCIUbjOwu@7xne1QP(7dxii|;kkTH6d>Mf
zB`u52@=O=VN$v}-z_q%7DuI}>WWZ5?S!~loqIa`<cRcAg$?<)^&#R_+`l(s|!E|B2
z;)NeB#e1x`kRfs<U<V=rqco&5L6Z41NkT$m1E!BGa0t;`2QCgSx`P>5RQ{>YqwBc-
z%*5#jL97Ga2K)!M{-EZvrK!Qv$wAebxw*^y{MCad*Py+2XmIdZR6B4%@F5#e9Tb(7
z9mWKs;V&T+mj+M^Byxg!0^Na`xEB-->Q^!jE^fmvfzWX8p`i$%Lau!V{~|U<C3Wj&
z93~fdb9ZBOX(2Qw^u>r5=m#*c&p>Sr9f66r;!kw5q9YNSNr7p!4#v#{?6C+(n&LWn
z23iA+tyw5)v6jK)X$CGwM11^dKrI1}cG!4#!DQ*&ATdm#E+9?O1A)`ur601ZabP2M
zyf*2Qx%hMqrffT~mb4b){Q$G)wp>1J3@8^mItPH7OUuaQeLg>v=AWb%qvXYFATJ*}
zT4kSMdOI51QJrlL22P@xJB(5Jwq!(}SNHcC=dO}QW0}G5&6VKaLv@c7DgfLa4OUaN
z-NciCXubWe&dx}nMm4$oUTE+Pu#C;4d<WjZ{MX825JcYrp8-!lyKuP7OaZmPoH)x6
zA&5VyRc0y;GaPTYSyo0S5;mjxZ?!IW?=X?U>{tl?AhaomewclG`wah3xRNjbR&@C0
zO>!7w%>|L)rZ%w6_5(##tKMVR(^jTPzkwaFr^NdDV-)$xD>+|IUK0_~15bb+4(F)M
zO!{i4`AF2`VgsD8E`YQ^NDWY30EIOH;&4{Ijt7j*<jM*=yxi---yU5QQhDj0WZBNt
zm?YDg3lp$AH!}&ps|t!8PT5=sbb;GKZq79DW$5Vqhr|dXpi$7U)HhY$Azp95n8&;Y
zz<*Z#=Sg^a<%Y+eK#U4EV1rWv5KG2z^!k^=q>+IoLx1&Zp`;aB5URH&31(H5<#~*W
z4dfYS*lZ;u8svZ9aNEdqb#v<gR*nh0pvj2|TG&1uLq6R#QPIVIC3<cG08Clb?Q|}^
zv70}CM#I5M(F#3+Xo1NR2{j?@(COkT#8^?srgpf<-Cx=&bPMfA55%0F^!EYjIy>M{
z+g|uyR|`iyYO4tzR++X{R*;e!w&o|lz^yh$!06nAzHNB2>%X6BbLiI4gJ}@~EF{$E
zNcQ5eLC|2h3Ld!xgkk@_teEJsbn;d3f+PSQN(!w9;61sOf0B;L_;Jz>#u)#tfAT^1
z(4!C9U=~Dz$MU@WX|aYA?3htPP4amEy2!t-3E-KB4=W&GT-&|a9=XoELHHHG{ovFR
zAjB2~uRHBd>_N!fpgDj599r;Uu-MpS^iQsXRS5)@V|H`y67Y)J;KiR6tt>6&%*~I?
zr_DJETxY^4K;$~oLH&*-Y(fC-KLKn{9!met59WntJg6gBOH<Qrh@~>Yrs>ZS_X&=V
z6uG1ueQUbD?%?kMvWywXa?s>g)1p5!5Rhb)^;KU6U!m&<UCT)!bjYL4IPJK=jQ*nk
zDRUkcG)Ig9$GrpUt|X>=TdWn5ZYV9($ZkKrS^gdNNCf1E4y2cWxI_qXP9g@4Ad0`&
zU}64y%}Dt1a?D*Ny!+s1bK8t+0!|FIGC(V0elRx*?DiuUcr0)?ZHgTnXg28H3+KpX
zr~oaP=wy}rZzDVu9?@x9fj15#WDEs`S?6Sg9i{`25rzC$SR7HXmvw+kG6RsdSTGqo
zVLIV$*TJ&dThu?~N~(Vhm@R(rS3>4?1TFtgF<3MIUP7vzro$M$+J|=~gD=p+?SNm+
z2pJcStxe$8APdCT*Y`pz4lb?<u%cjO$eQ-Oc4x>ezP!nW!7e?S&msEHG8t64^>>{J
zGoW>oM>TNcL~asHq!Q>&4AgjZVgbhfCb%Auayh$=s_ejWFtE0^zEB3f1=NW_k5a=M
z&zYH+PD2G^Geiq4h1$lHZo_3!DgmY+$|cOe#odp$u86TN0G%)f_(WhXf|J$K({mTv
zwJ9%%fXNXJ9zZ9wm@&NYJTx>Gz*E3-cE1~Y&*GmkaN~3QaJ1h3`GYDy48!+FrPAO>
zx$VxxF|e|BT?4c=KU#PIkn92?nepe>CTLaxjP=#L1MA)Y>20yJEm4RG?+>-WYGQ?%
zvhCS+-^1fAFq!mHJT{~O=oAA-!{7xI5w8Db*e-;$A2v)^k}m>n1+bx|?pv0yU!8))
ziUh!nF&LQIuhjBm?Qsj}j-CbMc&xJn9_3&E|GC{=g9Xn26L5{>lWzFH*<%0(K9m|Z
zLR%tey{(LZaJ9`WM?2}!5`$R_xe~OYs$#Y3dKU-)sD&8pYfyW=1|XYJoQm~-Y$Std
z(?^M|`HAj&EFr-E0~1&qT4)#p|G1Ss=gYqqSmZ0TyiiO}c{hH@8ABtUGhhHYM=1q`
zHrSy<bmhzc<IVm}5<$R_g{Dm8C`uZ@(z~RAN*MxP86Jg_|DW;Ff5!pEjW-Y@4tncc
zs69(Z*HUurA(nWi4MoW8OVKcBJO?J^{UxhUVH7+#Fqk9I`&()85?vk+L~q+)A+Wnw
zSYY6L=uH$JA8&ra(az5M>C>m2GMDL;*&ZL46PHv2lwT7KG`TeAPE<BJckbM_=kd{u
zJNO8^1wB1|p@Htgr^4p|ra24!bS^*#6F^?|0d^buH5CpvY(Zcw10pWKUKfKE4A3=y
zumqYuK*eboG+21|=@ThXxm>|$i-0s2>IH!o18hLJ;4DMH05&wtJLs;94R~b)e+OeQ
z0J_9O#LplVwMs!b9}-seG(e@mR{$?=SAm#}%m_FruK;NZtN1C(_ox;^%mlDy62Nh}
zZ)Noy{M!5AFRF>cP|2W?U@O$ACM>raMx91Dp<otK%K)=}6DBq?Z6W#y&NnR|9~mY*
zAt5j)#T^I2N5IFs1_;8?H*dIs08IQYIi?-#5ftqN0<<{Ts&L{RCwPC)w$2aui<3}x
zd=vWh!kD8@85qPguix8{g{A>C$lyXjKj<5nrdkN56bfOWOM}xn4utKtvJWE9A-V-I
z%2PCtcW5C(e4dO<!?G6AW(XbwoWdq>8v=j<I0btm&su}5co3tw_!TI+uRjlL_CKxA
zCitie!^A^~$}%u9wcFK#!wqW=IKti6V!?C5OhfPq(%_D>a?p~Ly?hK#D_o4UoLn?u
z_^~d5)#G~Kdi#@NEO=|uVq)GC4^>s8;hjjUtH&A?Ky(R3)mmCw{7ccP4`Ic#m00aS
zXVSJDgS8w9<ePnKJaKWaXKfh52rWyP9uun&!hqKgXv+|T0tn+{yurIXhJmf!9}F?k
zHP=Qko1h<M2E+^y{}Sce;a$94Ys9s)w|6tY2<OcXAT)sl<_i%<$QQrw=*S)P1Z>{;
zj~|+Oyvg#X3f1pf%IW9pL8qLRTIkf#_w3VcaaHQQ)A0e_k-#I&{Ynv%Js5wmGzke%
zjM&46yeLZu{zdF>Ffho4)|-hShBN{MJ=j=LHwXg=PM|6RfJo_^;+;B2^$ra^bk3WE
zx!wj{N3J<9s1g$stG-g-kFQs!OJ6{hJgV=<#%KVthoVXlyHH;U16roH;N20xTZKYY
zcnZvflGo7R9fjZE@Mi*=b@_+=Kif?M14NMel>iPb3MfJNt+dnf0DO(r+U2P!6_Y=V
zbXzDXmbV6+a%8=uXgbU{G$ydMp_md}OA@%>P>dS_&d?MbY={U8F*sB6Di|s|XmJCK
zew22H%nQQ#K~&!t+J!=E=Ls<1Ik|(?3}Ephw+_KOA(H`!`STEIRIBi!Q|5q-PbLm<
z|Bz<_y^aZ>`4BAa`vB^My_*!O3?R$02D3N=EZ4PPza#;9hghF*ik-r&^~-X+1UCu-
z*w|1s`+o*p$!iDfb;#%M`N#$w3<4dK{O10CdEk_IXxL|<20>ZbUSKqU?t6*YoW={h
zw;>c90GB!>et!Dp;!+|HEHi)$jX!CB=vWa|z=}EtwH_EuDNPk6e1f6s>p<(hNGWH*
lvoj=3Y9STU6l?zcpW*49DOQ=~w=XgPfv2mV%Q~loCII7GMMVGr

literal 45025
zcmeFZbyQc;7BvcXVS!RA7J`J*C1HVpNDC5zG}7H=fV4^@B_JZwCDJGY5=xh(qzFnX
z4R0Rr{k|9Djq$xV#{2W#ajzHU7w7D=_F8kzIoI})mA*l?hh`562??2m_;q;_k}V!2
zB&378cHm#Qn|nv_fBZJ0cWe~Q^=<66tn^53Y1vqqnA@1#*FI^dXJvig+>G-A*9A_N
zlZG}n7S;k6FPi?(7hEv6GPu|fd%Y1KvfDyj#hQfVfEMvT(l;W>_en@Hr6sOkQFMr&
z>T`6F>RhduZPd-V80;zjSp2m1-t1`#)`u<Tzp5IQLX=o;zg0E5Ute%r?YwDz6=k|n
z=mk3?<0+%3tk1c^4jTW}<FIyjzb+=GbT~*Vc5yN^yy5Owp~je~$Qu6f*i?B6(JQ-7
zJwA9fghpTVE-CTTu2~m}EsVqua&_YW{}2DmMey<U_P(4XKf630H(t~iE95MjqBzx^
zJHD(a<$ZPR*RL;wA0GJ?(OeB-3?kVx8$o|F?4^8)VaqFt2ld}d964K)WiC1-J!Tx;
ze~hnUVWQP>X7Gxfvgoehle=FYe5#VGPuAbxe~jPiaZ^*%me731d42tc2sv?X@`MEe
zi$M}j{f5sxq$Lisr|ZL5oi{=cCu|GeChINrv9gjMzn&laQrX@f%5B_1VH+x?boJ_0
zIpyz#H*elN!_A%lCRNTv=ohz^o}Q<tr|79pv0bg_@THMyD!Drh3=A~fmi3(ePDPq-
ziD-=|^k=tcR!SqeH#7LbWqCB5<c%UzrHt3($J_cH$BzpN31#QxB&g;axoj+tGx`Q)
z7M-|c=U-8Axh{;=FCc(SSXj7ZV`Z}NcOv5+^OD&eVPRowE0dXpX3k49S^{=c-ie84
z-o1O5XV&*(aPVHD>Bjn^X1~)U>)YF@UU<7+T}ZrgrpEQ_*Qrbkw5#?gG97VV{5_gp
z(4U}QXr-8T=a2`F={MO_C05*_Vb42Oe3s%H*$0int1O!1TFY~zB}+qoA<WAAnVFf5
zI?`%el5Xbq*<s1>jWj!V?&P8&3!vd1dP3*+`LO-?*BDDXyXvoB13f%Qj$TkVyfg6n
z7VWpZ`@u<5A0F)-{`Je$A;xjv*u(96Xa4-Ds;l#T_Uxd%g2M7l<$?Q6FG&Lf1G~Dr
zo8R7MwwvycNm4J>X#D88H|1_IPe(_``=X+jY@Hi{$1m-;JlT<+qL3{8ZdNH(DWIaF
zqD1=C%`lb^0|RupOhG|GDsH0?xtnUTvIqbE{d@M@IsNVe^MSJW#l?xO6Z`7lnRH)G
zRZ7pw&8__K;X$~jTcq{)S3Q1By!NrDXLF&o!Leh<ehiemzxxv<>?%OcAib@krlw}5
zAjC38NO%X^u+!@5bU8Jr{=STi3_gB-cI{8Qa2E(b(ycr9f9me0MiBJ7F7~mw+dGxU
zZr^)!aC-V_em=LEnVCNo`@_}6>2XJ4PsZI1ennfZQZN{Iq{ZW@+)-6+DKOUzyKuLz
zIq}B*wl~e>;oSIWv)=a~aMyS8jY69RMCBYtLzTNqoq4Z4*`H+AS7g%jF2IMJ!D@ZQ
zcC0Zj&bYg|x!LAN6REPYa=-JmP;_*(x3~9a?v&OE8zPX_7FuOYy0VD7=C3i`T$#Km
zVEgm>qa9=sJf<}I`ubkijW;*exy*X0JQ8m_wHmA-)o}jB&7_nje*OBko8cFdZbk7t
ze*E~Jv9Yn@Z#DB$&(u_AI=7{GIpfCl)y0Xf>@$7E_R03s{cCge7o}cs`FzVaZpE`w
zZ>&P-%q}jDhbn7)=<SUb*;sm1(C@St&-jg8;=Z2rJJ}CAHQkT9ySty0iS-W-#<EA>
z!-HO!>^$Ic{E}T$(oM>`@QaOI**cc?_Ty{I^J~Z){>LxzbUKabB2v#XFw9ren%~2&
z^z+-Bot^En`m12!S5Nr&$|vNMlr{DPt~c)8J9#7USUlFvYyHXo<8{7An18gTkXL$<
zond7Sz#UlsXgWp1WAfR9bSssx3!gz_Y?4yCs@Hnp38A{Ao8b{0dXgFcq5}Ej<=tYt
zlI5`|^%93>i#}`y9WAXZal)=kxb^P5`{boge>kS1A|gK1xvzadGLuVfCrMMy_r3CH
z#~~`J&&hFalG@s*iAB|`4~yor{KzOFjReES_AHE5Z4NhQP^iISgPnkv%nm;l;c(w@
zL0Z`Bk(z3pksWYCh{oRD9!c#({Iw@bV^QW7R#rn554I}4RrEpn%SOf<Ylys-u9{E9
zXYo>N;LsVdETjOhebgT#*mZUtIAAb4R7HHPg~nsXW{Ojf6BvE1rn=a-%i2wK#WVUQ
z{P^RPq+R7(hZPTa*Hh|Tu!XVMews)1#EBCbBz1G6^<?#80W?Xviw5s3hsCzb3U4gw
zB#sI>|5a3E3O!6i^TPNpuSwTS5+a<fd(R~bTaPu!<*BB(B+JlMRac93zcW?Kn8mkL
zH8#rZ-nUP2!KqZM_|vEBRW&s??hlleHs{nF7j%#%4cT0ucK6b&55K5PLaddfs_Kj8
zzv*fPUvpd~C4Gl}|IS$W(R`+J{+Yt@EnByS)`c?DYKS<@T(-2ZkbS{*KjADrJ+*1n
zAsxJLXm~i$!O`*j;`BhHU6lXIJF`B4diI@UG_UZl)EWlS?g3(n*KgdQHH``jd;RLw
zD=OAv_sz1Qv9Z_0{_)atU;h)ng;A&4zt*b6UqA@olKt+T(z#nNbo1OABG`RnVqy-{
z(Z%Tv?88p<O-f2?iNCf}G`W>J?4m}=9nz5L^38sWv^__dbR9)H($$hl>dIVKDMaN$
z-oNLos;!lv7It~xY{zTfuXybE@qn-cG+g4OA&dRa7i(7ZcnrQC+b)X|m76v{Hz&V?
zj7HZX#<En%jf~g0LrN}1vF`KpGi3EGEiEC4Y?{(!q#9)sy)Q4g#cH~XY>nU5doudG
z+Iz?QNk^K>qo^oa)1G(RTZ&Cr7bYsPQ3Hw!_Ee62i9(?IQ*%}uE?=51d6d9t(w#%!
zxgHZ8O|mrKDAH_{UuYKf;)P_UMk(?4Zq(xA0ydv*SA!X3td{5WR{l)ZAPE#&kMFnt
z-FxcPsR!X2PCl-yf8#qRNwc!E*{71*jS8I?FWi2s_@TDe$9-d&;bwQ`lLNeY6ngsR
z=8c={1Dn(uYDkCgEp;-~3xkaB(q&$glG=xli?^NZK;`m`h@d)Is(10ig+MIL0c4}Z
zb<UzunP|Q%4i0?7)d8H*12qMXG(t`bpIhXUlv1<v^0u`U(@ggj-%wShA*1Et5-^|a
z%)F5#&u;oHk7ZQ9t+u)O`R3ZV$lB8IQ!g{A80i`rk@ZihnPqsQcLQyE3auGZQd0bB
zdG~YX-q75M>%EcKoOQqTTxGV?pPyA&HlL?Yc}*5uWklrSuWna;XR`ZM-1yD|Cz_Dd
zD{WC?j<*?c-+IAS`_6Cm3-3U~`nIDLP4U+P`il`fFB$hBp>wk+q_%UJbT$1@UB&*l
zw6S@GbrZAXGHfDoc3w3ZKA7<8`8L^IE#+*T8Fz2?yffW>-~a;`S4dM65$7MVCA#vB
zmE%uL54dkuRaS0AQc4`<Hfm!_+v)xE={PRZE3T)&eCxz;nbX3L4|vLqYVTM!mmA&h
zl)3PqKYv~&klXpM36ELt=ZocQLBG24Ywni0P<tR{egFB>Xkq?W!9Y+@khXXl{#-)Y
zCL5*#81O?CIB-Ic%_KPT=~J0^HEDOUwZqQVcX#t^4yQkS^yq$P#&J74JJd>iN*#*l
zfqQSR-MK^MAxAkeJeT{W_4E0$P-f-B2M>CnOaiog0QN$qp^hoY%3?!7x3ILV{X>Q7
zdX|GD2x)rjgijxCmWu7(<&0H5b%XY|w>eeK?%$UftqUcaBwbmUxF#=uxO{Wt(la+x
zv;LAujZ!D-{$hXQ@hVEq2CH8t17$Hkei$YWM_?0FKNQ{@%5<B<>DP}RyF8MU4Z=nv
zGxL#c$jQmC6<Uqvd^&^;O~3z?<Z0@yjN2x3`V$fo?iAax8h2)V$#KK(@(Rn1R(-Um
zIP~Bkb_DMWvFiHzdJF3hSi_G*az^s*dS^<TS-v5FI2grK@9OEH{fnHZiwfuRx7#2O
zfpl|k*D#0bB9gWa*HPA7|G67aPkcd|rRFs|hA*jXZax$$S!JAAs?W%f+mLP>sn^N0
zYsB%I6jt(L>B8amRAqzy5=RFoC+++9Yw@fjMcm!q*)lSQN`{)2S~+dz`Z@mj*}m_D
zV1%-!Yba2Ie1<wdf-={jk>_lJ%xN~ZgS@=FN|_p8_FgQh_}QL{LYs^tC!?+;{es8T
zx1fL*mz$_r?zT2tLx-f=j7(n1Lp|<{^#{5#IA7V@yTn{mS6BCDG&^h5=B{3W;JowO
z&dKx28R~(#BqQ8^WxazgAHIA_TxJGY+Tgu0Z?2~MdMGyJOuMqCmX6M&++5DfVauO6
z3c_?hpI2F!o6`oY{8Uv%oxerny~T&D6D?|X97UZcjX9e0?knK3>KNi&J{}NRyA0r>
zf~SIF&iPG$<nwczi55>h_iU|7FBZ*Dk;^M9TV6IB&PkngTOOrE?)=nSXiZih6dWAa
zqg(2}HoFt?zqYYDEtW3y)V^8olg-~LmA8s1L8dB^1+}bZ1_t9M#jkUnH?2|M%Vi6#
z^xW3}C?^_^%w}k4YHGeS?WrzxUbHgZu9(^m(DL*Mv(n5&a$Jt(aP_g`9?tsrUoU8s
zh!X)?ub`lSjoWB-!fEmM7eHK}fPhiJfsZ{sQLk@B(X>;(By!vfm#ZI&zQ=`RW?sBx
z|NE7w-%<XJWKr)R=^_u&l<<&{PlzYU@C$c;bmtm`#l3uaa%gC1-(5`&4UISLMt^_x
z7$|-XqQ99F62rdnbN2BYMBWb@fuk2SO!8bd)|QDrBqkHP?lqpUSKLunRgww&FT(wM
zLYq~?@A6s{+x;S9T&=)Nufl`0@SQ(5k`V4wcyrA<aaMzxqbUBA(5yqr3BRMPPUq6&
zyKBWQI66<AJLff8A5Lr9iK+(}9U<h*Q(IdrHp%ffU9%kBj4;cctdsStp@lB^tPASz
z@7bJ?ju8me^NnUmUj$O)cbIwEY=^vkSuVtcf!uwq<Ny_$2kICnpfHMeK(*r7D_k+q
z@*MUuRp?@${lFn|^59YFSz*_eXF)+95m+d8%y<|)s&rIT-eQSbD0kg~x4yW1ja8I(
zzu9<oy1%q}!p3Ex`G!gdN_1_j_+yD@M=#6$T*vcAu+7#nXX+N0-EV#EB@vJ6_y}zl
z;GA}If+*I_#np8g`#kqJW!uXuTS=<@sjmT88s)#})Nl^SxLe!;U?aRVxb^hu({B`#
z_aots7Yw-R^uD+B3JZx&Lq$U~+mx=xix2Yg^V8|h(KF<t5wQOq{YyBrY=sG_$6&DH
z0sgHFki(>alaq6}$GkiVEzi>GDyQz}gG`EVD*jA%cB2oV3Gf1z-s*ATZc&y|I}<=d
zic0P|)E!;yn~?Jg=Q>v?!}cE+xQFbBqCw`<n4?$EsGM<3PEOA27y7sP@kWELEG;gh
zHgC7}mG4ti4Z!Q-Qc}7I)riQ*Y?}%BBD-IbeMPp^9J;&kaK{m`(U<H40Uk(a(2R1L
z_X`4cQ482G0`j>0X;;QolGD;gWjfho3ljSqfO{6ZhWx~duSh+j$YudSzp-OD?p2VK
zEe-9xpjjqW6L`FN^3T*jc^rOmn1VvL>m7D)+uPfIaqb(DG~7nA<!<YOlSBm>sSS=~
z*V%!tU<B>N_n|637L}aG0Hb;mr%}~qkj<9{9~}^vkYNA0@rADu0GbHmwoJ`9JR&mm
zMyJo5As{-dYToC$xY)J2CAAg2!QoeaXI56$OC<Kf{*k#n)dI6G-iOW*LChc%`wA8M
z&Ye3Jz&LIj%LZt0#T#RV_O8FbDyS@KF`j*}N)iiv1Ym?7sHets$AJ@Rcs(OviyQ!+
z*+5x3VCuE({QTR|F)^XQ<BaGpzHpF90he+(Ef`j{x8DYc6I1wz5Ik>QzW#3L`}cU_
zBi!8F>Z19r{a>elt`*r_OGI0voPyM(t*4iS?p%7JCCQJ#?v*@|$q#BxYJ=$ayp)Py
zyCN47ot)2aIrO-t)Y(2sDw4xDG&FQ*bTnz<cb{5!zHzu#6t))gHLrbg=qyrEm6#tq
zcu?oLm(nUa>`d`}W2`_ro~D7y#!?;=|ML3cKqz|qs&C&^XaD|H5?)_W7)3d6w5xjt
z#Buk2e@UUXj!q&XHwkI4?y1OzVcr+~COyx5K!ZTzQ&UqR_y9d@B(a_}jZ)PV^}-yK
ztCY5MHD#h92yCqoblW&W@2VP#Cgwbn8F9(-nHoZ)jUt;3JY#uzTzX%l`6W@wj)0FL
z)@-6JCD>|hX_jTw4W)F?%Z3-+%#HJ5-H3}&R59kGN00hOMMce_N79en!%(@C&gCVE
zB!a|>#eDI1pX9xJFNc2ocymWhjk&C>tPZVOKt>}V&v{@L8O6q@!Ykq_vhfeK(H}2t
zj4e7v@tD5Y!U#%%M>SUAZrSn}Z4zGX8xRmMG&uOOE#t15BeGH`Ew7oNT&X8xprUV3
zW6gfhg^V2tn$OtEx4_7RUApIa@HD3ciRZzO+ZcVX#t7IlD!le@P+)lcIoO|+;;Q#y
z7Jz50UJT$gdM9c(G_(mBK?f7IAy?tkJ|Mx&%+3-Z8m+XSjPUZSH>qbIGm@UgyNR#U
z(>;OL9lYu-FNT`?fBE6}mJ1~4z5{4^Ei5b?G8$8e_k3zqpzuT|0?ft6$sA*<6-08h
z0xy|aSYY{SL@=I1=haJ{LeX9vwdMAGeAZ+8;NT$1E6vL^0k6G{gL%`H>QQ>g8uPif
zzosG=SEu(mU%l-$b@;zOJ97ii^Y9#%k&&rfbhv+Z_dQZwn^hyO7l(sMzJ2?a&}=C6
z^V-KY?JYj3XW6SFD!7I@{{88s!5%vHT`ehhSdF-X1PU;9|9%J|z%cPIA<x-8uIeT*
z?zw;e{tug?)9Fu?Qmeo+jEsz&&b+C`QT1Mm{T$nu_j?WptEN}?_Le9<r6zwZibXjg
zdFvKE6O-?!PoEMp_V^|Y{{H<HK<d`dy)_j7{dy)tDbQje7d21ZyjgD`d+Ai{MLe{V
zuJY3}GZuhP=!a2_j!uNfYGg793!iXyb`E~=s_lQg-RWBdh3F}6?oiYLB*4J9xVYMA
zayciE#oH#*-JMELU_DGsP5p`l&<le|LO;oqa*;;Czn(&8ORInCN%ropu6tTqAAo0#
z8LxQ<rCndUJ>66A6r~0U;(=t6HaZ^1-+*x^es*R)4Glel)=I0&_lU~7Wn^9K)aq~F
zUZBgc04n;JS;8nMHR>n5OKxdo{z^p9VM}g?vu9<KrH_|^v~e+|Ly`u-kwSOD?>KiB
zAFYG(0G{bf))%5BTACe>1X;AXHh;kOXWJ>z6+eFbFgW4`;``L~Y}wO?D^Sp90T5e(
z&|OS7f$P2x4_5+grN|}jxN_yn!c=$g^87gMyGUO16X>H1K0iCvmZlPn(zi!VO|5Km
zW0m;0p?Dyu!I6=WNgi;$$tWR>Ny6K+ag)fN1l#+d4tJ*i;o){q`Q+%35NoE#4u7Sr
z_F?<(M8gr&oC6xQI)HZPtEY<sZfksKPkGXKIXM%79*zk*u+>;8$jcKYq9iaNV0%mQ
z(I%sOPo>SrLeAn{xyem+^)EXoEsO19qMq!Jlu{PcstVCJmr;%4dDUO!8R^Krlq*Zo
zppgHns^*bT?}Ob`LoFf>B)(UF{+Z-Ad+Qe<+H7RqH|A(Okjp^2vC$C2UYV7aUBDav
zWWP@Mv&%;EEehes*H)*yy1yM&Em1ncob$WKd|9R@&`VEe<&Vzw#@OXcVq?Zl4f`4z
zWK$X{q(V1}BUM)NBWG8KGwC)oq8WX)my|O*MP@e;K%#QE_BZW6KW6kR)GuwEZ<vlx
z_?xD~k!1ZR&-VlG-S5SY^XI@miq-e__Id{d(4Rl=gS-zmG^cM2)to=<61qNp-Ic+{
zEHs=~OD!w(YNYFa_o^6<>W%ICL?^T&Rd<2O^H6Y^i(^Q2Ky%Hfm(1rhRgEqyFL1U{
z2ynlYk=iDD#eUN*uxBDTSk9(My}+?QB#$d3gd$NOh4jEBwFP$e?MyvY4f}$Eo9P)S
zv%;SK0GRt}G(W>oKAtC=aU~SwbjXv<v8C0@SL~g?a=NQ$uC0VSxy91bQc^A|#=m;~
zItZ8-WQq9Ao6c`JVv|t*DQu7I>C5#^;gedt-6q9B>Y@blumKTDmZ_toGrO=r)NDhp
z7gsMy2Sw3Ja2r#oI#MqP2*t<8E2hYA=d_&_x7zbIOS?r?ZNB0Iscj@j1m&YADa?Cb
zy*jI-rx#<e^`%Cp!Pf*)mja_ByygSdnO1uJbRw}Q?iT0p@<sgZE6)8h9_P3@{FL8)
z;;`-7y2?9?Q|)gRNe9~st)(J3JcpB|<K<J{$QN?l9BYXGP?_;@u<}XdTrI;f5nJD!
z3m4Xs6&Qp!2DhRvlFS%)Jv+6&V!2#*gvadLb!tk=G55LMw`h5EvFE1x%jxJYUJPTs
zb9r6BRzgCey5Y)JrXx{-$93i!BGpSRA5p&G9!_<fpGlJ?CE>huNpWl}BAnU4`_F27
zY9uz{_3L6oqZ1NMSz7+bIL>O0FGz{+nwE<E7(ZN-ou3`wQ-6`uU`%R?uA1GZf$GGk
zPqGDereSWILTHLgI!Z6ii!5(Lx%CR9uH4Y6>2jZ<(#+5xt4K<`;p1H*`Q$`jq3!J9
zuhM~Q=D$??BLzZ_Xu8!Fw<cTvswNXCTR2=#$vPA+vN4#*Z|&_9kmTMN>$1^&!z=7;
zxfdC=kPDxVnE#-=aVLk#R2K`oR3zJj0}Rq5T|VSeG2DK^28{{rMkxxZN8=i!B}zR=
zuBnh@3Qhf)nOhi{%hqX#=6o_xVDi07)15jaMrD3{zVGer-ZJO#pKfk09C~{mTrBfW
zs7Pw6Z!_x-iYr?WqaSc>d``Y8TUGN!G|uH}tJ9y$nNOaiYpb~JJZ;`P_)8_1Z>dq_
z)$641qu)O;4A->IdQy_?+6^&fZFw7=NMHivHpR@ZymanTdNMK#PC}dOE5>D;qm;jU
z%VqNGLYEDD`2_Mzc2|^lW-27=+}PkX*gvqgKz>6ePQQ6!V&9MEN2A~cK2Dc24E?-a
z!lb0z1`KW0bdrtp%<a^wmT%vhj8f{vS#lj?3aiLZ^7B*E($Z=;eD~D0rOZ-F)dNlj
zniAIBZ6Ps}rc?Lh1M?R|z705W(1~L8$djI8i65$~cjT)EsHT6lqqVnRJXW8dlk>5z
zj;Y>i?_R&FrI%S>@QAkN-H#6qzmnQ+r8}=*XZPDgp6;;cAq6Wd4$JS~?}VIxR8@Jv
zPE=03MR=3pWY?WHaulAQy6!Y^zYr{bC$c<3-m$jAtjVh86&FXh`uCU4%;)DJneyAM
z4Gc(LyKeB$AHDDiEY9;YH#oG1u7d<3rCJ;EYfDvrnpw8GKVS7rZ7@T&rFCV3I{(Ly
zAAhtITm((dUay=!1pKgV`}Wzt4UsIGWrC$!^z~h`O{jDAzi5ABh+_i@?&W{t76dv+
z*B`QDZnN8i&)pjOV>!%cJ+`mtJzvAIDG4dbq1KD#Bksl>=z3})Q&f4pJbA6O?Ev{@
zzbCr~UvL{oKz3x&D2V`dXF2p~x7Pu>!0Nsjptw<|z{7_R|7gjiLz6Q6qnY4>IZe9Q
z`eWHE*{*p+=5rVGnttqPelFy^c=*tv56@0rvq-=59N6ozsHXcZL*29aM(x_*a7Z9n
zS^fD=>EqbVap(Q}mmP-UQj*%Wn;YO2H>K3oj}b!2qW2+?V%d(9O5i!KK0P61*j^^&
zy0X9?eIx(T)}y0!&E`uM8o&E`WH-U9v!PqkfVSgjH7cHyb#G?oJGjDwqH+<J>@Q@l
z4QyJXGeOVI=d{L$-c1LZh|N^jA%uZvs&SaL09Nj(%bz3Y$G8EWhk@IZS1TT%liQD0
ze4)UMCL<0I`<RG`2+_RmC|&14nv57{Q@<)lAxqI8$+bJZkl)#Ey83Iim-VSrhEu<?
z(-eKsD>Z-r(|NzIgulnV&*M?)@tp_g#LNwze5>`T8gQwkOjXKWN__f8E%6+)+j69a
zPW4QqdHL|<?OKbpl)XpUoIh^+>9%l4Nl9sKX)?1xJ<G#!x?HC3XFI>+_bpdilXbSP
z4Y+Rv_rG1{V)bA;qI!~5;Kdu2T(+k&6%F5LIMrwsJ9r{g)Dm;Ds~f71a?%<rc63Cj
z^yn6|xY8Lnm~i}cUNR?r5=ZgZUe0nRO`9Wset`@7#iS#RI*QkvNA;af&3**L4v(ZH
zdNLZWduZWGR(_^lUfFzX|LdEmtZdVKMh4^SpgaL1xA&~mWP4RUH(XBpGT-?k5?li$
z->yRI<nJqZejM(b>%7G35<=~>*t3upXgokqKP4o_krCar^mHG8f8C)^K8l$d$4Rix
z*Or0VUb}B@c+KXb#dDdhq9l}D2u(DqSHK4}p!p;rXjYXXoAc0F6QJdi5UQH~mlu|k
z9juiWyMedS=G<?2wVQ-lIpY<2U_$c->@o1Mh4R`Eb1D3HU-8z#;o)oI;=6Elwz0bh
z3APN|d~<bRb02HCALYf@c#1A76UkoRPeNgK*<4@j`u1&Tdb+V8k~5Oe@-&_6!Xc=$
zfVtYn#!-{*6r9>>xFmE-Bm|nfovK7!&eX481or@5Ko4!z)58@~CMG6e{7Fx;u=vX*
zim@AauzV{pKTzR!vCJ@Uop-?H5p+ICpSa-**e+WLv}b%54X!o7>L)OrJ1Rhn)!_zI
zvhI<BXyP#8)%v)N5ZFl|d_5>H7eR=68IM`MebSfnTsgBP@y3ClT!ti<Ltk){^|+Ke
zaR*9EujOey`d;5v?r2@PvVFj{=G9`qPDN!{Q{nM%xtk*rVH?XkHQko!l@2sEzMmbQ
znD0=V_%h}sbli(<cz)AOaynK>cEM?lhIec(UCpK3^30<r`|0WVsc-JxTlu%HGVCVB
zX!EqwcF|^9RSwd`Ki?*Kn?yxN9W>&?>OOvaw(muEm!RZ6(-Y?cd3Yk-8pV&9%CfLF
zsTW<3j-=VUbM?j~-MJ{wNZmcHa;GyMFD*G-y>^W(vl1=@VDAeTE)Z%SGNm>2-HpZa
zO;R?i5pi&M4aL)C^^=+9U-3k%ifkvlAhO5%&komc8aADTfGG~xPORDP-MjbI8rTON
zth!=o$mrwa<Mrv}H6tS?BmmBm{SIB+yJP|BND>kgK_?s@u6asOL4-(Y_v>48s`ABa
zLsmcA&P&G$BO4m;TQ%7C^1y)u$%rzQx<>KWMu6FRDmFG;uv(CSN^gB5cczjD?Nbd}
zQi4~KAR#||c-P*&IzO7?_t6RWW(l&glHIsbBiPuA4gfq;;Hl-UhRq4vp_3yFX?`lB
z4N06W!MB|e7LH9-&IDNzI&R5<aHRRFCANkyBB*oLdgE^?(140;CO)E9Y53$rPF8Q;
zS5yVUt1H*w0C*^HorlTEvEdr*j{^%9S`P03+ry^XpCGyy4T@eKOY$T`@;0_gh@lSi
zUwFIATrd6lmJikYgjbGqte^xe0y(BWwK9Div2CqyB*+_12)|%OL!QN~Om%|G_mGYl
z?ce(M(ru%4&yUjjnF_0Re3!eaFP8A;EQ8MekWH8UM<r?*YbAaJ2Q@R$C!A$S-i(cY
z92DX&e14*NW#nwgr?U&SksS;oz30xKU#Id&XiqTA{}Gqq+;aT=pZSZ9=7w6&Q%2br
zKear&uCEeSc$>48LrqPB<L0x&vljq;{lSvaXc+ic(sCIdgu~<{6I0!fM~yW#PavGk
z{_Qc(wfKG&!u@AZk2k?^Xu->{VzK(yRCsOX8YB@35PG0Vjl*cKU7yg?i9e^Sf3Kfb
z_|dR_Z+b*H0+9sHkn5gLbjCUEsHhw|dUOYsy}hoBrGD>re%BQnLgfQj-Ib%axApZc
z9ju(qWJdtrCgxA|0ZW>5tXMwjMxRA@Xj~6_oT6AXWFpQyDt$(2dmViT!4&XZ-52VA
zx!^HDt$8Gt_&v(Jj9rx%;rqM4v>r{#?A#m>i!V~$^A|5v9j@rk<xP)fX;tomYJ#sM
zBpeXmoc6!f#h#upHT=%<ra{70OOxTP6z`iiB%GxuLOr+h+)h3Ejq0+foUFWpTu@LG
zLlT3(q*ajI!zBmpWkw&VO<q>^2R}ZihFDs7goG?SUeF4Plrs|x@=@~JI3L{LlAHCp
z&3BnCyTQ;iFQ=}v@4J)e)0%WcGn4Lz3gH=ic8<&4?FAuBhj!lsCuFsLwc+=9I_Qbt
z5s!sov=kI)@0Oju!4$ejtv+Fyyw-fnom~A}X=!QR{{CDxv-)ArNyd_l&@QB_hT<tj
zp|{C#S+>G<7F%EWgRL81S4k_jM*)f@Kp$t`-@7Fa&%x|csUG39wzfu+c!ZqaRPMgX
zAgqM)0M%_f-sSc96dxZINMo3Q^t*nyr6>^kcTg}|AuV(Br^(3-&-z8vT7}{)<H0dP
zQ;z-)iC-ls01l!s7L}{ujY;1-{b2%c$^NbS<HysTMMhjSV1#z<-_KKI2Pl{O>!G`9
zY;3ImG5#O|;5p9e!m>~ew6uHIu6w{9-Tj5DdA9YF0*>D{Jp^xJ1wcyNQHlSmW7ZID
zn6D~r(k~TrtokY~ePkqgz5f8G>P1e5Kx`KBV1G3h&U{^AqjeX>u=skxGqo;t#p#?~
zTYbhGb<|SrVtafUluv`Wx&D`s7udA65ch!efeXNXzh+^<MiX$Z(X1XF+&SrJf}x$A
zo&64V`<ReZ$O$1Qk_nwfs78`72SF*SgjoqDBsd;Epp5QSwBNgTFT&{yrKZb+!a{zq
z6<ns@xN0sg!S2JR-#`U=oiMqeDLRHNh#s1@FD5Q#V||qqFC#Q1IME7SS9y>|^uaib
zrEfqq-gjI8dFk~*z(weMyY}qiu%UEvcD9Cv3k+UXK|xJbRnq10lMDp$5AGHQ9a1QL
zU!o>>b&&0J@X&y<k&hA}Wqqoy<^W`6VKqW7ocflJs;EkMUgj%lA96&YR%35ec6Np%
ziiluBCcED@s(BFgAQAcj5o);VT<1lzYY*utC_W{K`Vp*mgrFk_3j8qv8*g9-1(-&X
zGt|sFC93aqv{W<Jx`@@tQqY&~?P&^HzN5W7V^2@NEN85H!A9VCvP|5mlP`JR<r@hl
zs)n1vt0(u-6~YSYIR{7u;G%~p@(?S=K^`~eR$WV1vR?dscgOx?pTQ3i@-u)iZ-YJc
z!Gjh3rOwNs;SJm09sv_KIT`gUV6RMFM@N`M@VVwpO_2=Ea%n8X((mGFt6n!$DwG8d
zcp~tiy<=jI=Nor^#f_(@vx2+Qnd-`hrm+i*5+QY$IL`aObDl^FpP8Q@_N3G}Nl)*I
z{l<Y*R=Kj9{2Ut+4-9S+kd428|1Lx8mhty@pPzK{Fa!p6Rbf2c*D`Tm;UTJ_H7~h>
z?Fn*07qSvq_78?I2aI#LtvMhD#XR>MAskm$(d;_a2g%7R;i0&O3JE9S$l`!|O=DxG
ztZF^hEm16B2vJE0KXy}w4yP3&IrIR4%Omgzu`pukN(u#LcPrnQ_x=ovdZ^AYG?Ajn
zq!<=<@4k-!-LTa>T4JC4cVaqMbCoD~ndr-1Z=Lq*>bipfMU_9_=13BE1lSNN@Px@#
zcUjLPL|=zHTUuJ8@LMiL+X{!2;T->+(pQXMEy<t>Be7(Gv}gF1$ycLz@HLh46&)tC
zICAQd-1XYfyy<C!4_q5>J^!(MWrWd{aTcjrkBV+_NhVo3#<$4nTq**LKXwtk{Sn(g
zj}I0;11al$*c_ze!q<fn|Ax6sed2bxyS;l|L1ZH*=@tK0={tGmG2<~r%_&JiN=ixz
zKHrLAYZ?4)|HPJ~{~@veC!77h_(6~`T2`EYx^qA#PB;efFC4xCO7i|0cr!nah+e-A
z|0o|)ReV^gd`j$6vn8Cm$No8W&xy-Xz%=6vx2d<bB$A6HADD5NQIR^(NqfN~Hv^bK
zk_GY+7D`B?#7;!;wUjtoL)A6x+C-MkGwnGF!v!QY^bvzdsa}7GjsSb<Bf4>ryo6AA
z_3A^dQV8ipvrD)I`}<E^ym--VbKM^L&=>cOc{-qIWKlLI&PwdW+4=cOaJOJGGc-zH
z7^e-mt+K+;QGtJhvXQG>8{`smhK`Po1b=%XnUsLqzs$r5vKewP+;W8L5E4p5tWeKE
z-plAOE$r-GTaDH^cAUwyK7Q<&6v|dsetuPVcO<^z_dxl9%aW3NU~Y<s01Z8)HAn9z
ztZznLyu+paDD}9p3e@Cs7%Ms_ukDiaJwKFE_30C%2U4#>+wI!$wl+mQq#8*hBO^KG
zq$Ua_DNt3C;^N-m?iLK=H8L5`&o*w}$W*owh_Tw-aD_2yRAmu5(VY~H(xM#ONhNv~
z7Aci7mqJ3?!V|awjT6Q*_-5q(PIV`uTa~`w{x+0Y_{t6Zr;cz7ogZ(EQ-%QgDow3G
z8o>D#tZXo#y@i|fJYeX*Uv1mDlNqR20jiuy*Fb&!O-oBlxw4g?%%d<S)&YE^5UQ0<
zO<)}&Cj?-W34HWh`9v`?dREq3h46sQL1{h@ElCn}BNWe$5k*;nhS$tQF%v@Zd641k
zB~UE$<$<X6kRhN$D~BNQF$2aEPQy9D8!#Wrk?5hoN-8RbSfNRB3FSxrs%mXjfJHS9
zghD(7P|fA<Pv{6s7%VE=Wy@B7Um67-s{0bfJD+yLY<vaif)K?BJ&Ostq~h0Dp;Q*t
zJaH7Bcery7)G3&MuECj=YxTRAU$5!q72n+4+*4uxivlN4UIx-FCRV2x>T&1f8D1GH
zvM`G60q1)feY7tN_r$47tZuj+Jxj{s5oAYN`{er9ob`J$KFT5YUyN@CAC8;8_!iA<
zRY!*sjM5&>b1=9tfi&hmyB#hCuCyL>B3r1mZ7nQ5v-{8p*u0s8DU*t|r`)}q5YYZD
z{c`!{sz~L^n>UQxWnW9j7~Qc0oxfcc>D^46gp}~dc$t2U6Uij0gCpRGXzdG<m6a8m
zpv-;fL!Y;x=T>igp1>&S;P8G6BW$8<s<FzEsJKcRF0-PdnVkk|7ha~pZ7?7qYBpX3
z=dUD+WX@emI=W0kOoiW;o#p1OTekseLf~|kk_)-9dk~wAtY^g5bwuoNMcP?X7dQOx
z?}$pp)sTDWF<~NZDRXsB97Q<?ht)WPg{+uy2E{_^9T@mpK1tFav@nuFHZ*vo^&4=<
z!Q%{b9Sf4tf0`#|--i&01O=m^W$T_F`0|C!1FtktIePrK=??*JUfyK%$X=#zl`{iT
zF%i<qNXKHcXV0EJc=TvB7KN||YB>T8qEasl-IY-vgVq6~q!8>kiHKan@B*b@uzl7;
zz3~sa#tF1y?^a<_k8d}+-<;s773t}@9ctbf+_PdmJTRKgIJ;#YICXu3$dUZnS)PmL
z6kjZl&}&9j*VJq!+--0g+{^olJr7-c;WR7DzWUotN1h4W9c)Qfo9N7>LjutTv_~Qd
zK>2NfsR1SpYIf~y{~VH2P+5iETJemI+chwR9vX-B01keiW`TUw^@cj)8KES>U+5Wg
z$-br7-W=<(*rxNt>QW~U50A|4v0?Dy`^v{{U3Z<5Bt*B*K9bwR>>SM(Sw=WX5eMf$
z8(~JH1zVP7lJjsy34w0_d*{&qv=V0Oi<;5}W_@q)MUdn{`H%X%jiZA|2=vy_^v!;D
zNc4oD!#%^>4Gnl4Xl)4^ZgXQjZs`tq5E4j<3>Op<o15TsB0OaPlLY$1zOAgP+776g
z$T5E=+4?x33$!?NJz`<5@N~dzcoqm8d7UIrl@$%89;_eyCzvydlu77**LRN%4puzg
zP40j~2-8EJNp}#sSs?3XLoP4fQw&I&Qqg?RF%*H-ti;0dIV~{46hm?k^AA9zAxQ6D
zWVBuYJ-9C*okvxEl$39NBRlz@5PEU3Y-tM`goh6w8iE4~`#rD^cIA6;r}7C15GHzg
zd3ip&DJpni4IrBro=8Qu@9OHxLi$C==#8R*2y+bMhp_`iL<Y_mHds)`<J_zJ`uf;>
z@=?EFgQX-TB}MyI7jiyW=lGmJCj_?Fpbbj?bf}>*I1IG_HvU`K2n5J8FIzqbOE-)S
zo`{LrZiBeza7H7jtp)WpI%{5Dc{Jw`{`lbQf7SOrB==o#Y3Rcqfg}oJa{{a>B@hN3
z?F!$v8xb8MoX&-|lSg3(P6C7ErSt~20zkHUi2Ij;5oIa1gVn@CQy6sW0ZP2za+YXW
zk&aP5jtsai-VN+IK+Cfa@DOBeB~og3Zf+tjzwn)h9vd(3I5?_%_>Ifb_a}Kq?Y^L2
zM^s)vdbBmp^)Ge!MNQx{g6-fj@Aro;w@;NAA;4M^o^xV=0ub=n<g?Dg4Ap!l!v6#8
zOYCMWa?>jb@|C~8_ExxWu4)2?ZQZf&By6mBnmMLDJdif61B43x1xgA9ecb;E)m=$!
zrm(f0d)014m~OFUcxd8o2OUw8g2U?ZlPAN&!;kXwZBKQ~Oh8D65941jRZ>J~t3!&9
z&{y9ckn-MDTK8GwUk2rtvhuj?txKu@FxJ6kVtyszQxv!HQA|N}Z`k~7n<#9r92gL(
zang;Iys_()?+e0`5aMv=M&K*h`@1uk(<W}u>41he%>QQ6sn_6%(Wj~hX9>jDh|NJs
z8J>_K7HfZZ@2TsSr|wn|Q0`!oKYsuJ<cEki5OL>!re+>Biad1a5Ik4tV^kPN^O2D>
zpIUJqLGJ@2*~`m%QvZQ^M<rdYP(2^0rsJ7$HZJg8^-dsi+Bu`|A3b45O~A~~D*%W`
zl1W4O=*B<q39MLKd%k`Sg@Y(j6mO=`24sN<-4gm-#cop$mWL69OdG}%ngRh*H3&0>
zR7hf?j-2up!Rq$*{0co?L6s2|`P6opA7OY}m3pe62m#-(J=evPaZd}e(HDUmWsB`h
z!Gyxyu@5}S8SL1b_0mY3R2U(dE?WgWxPTAfL@Sb*154ax<WqP6UISiF^%jz&6GUyY
z7Tf^a0jCQ&9?<*3LUw2Yzk3UVr&p4bk`7T&JcdIJdqMAKTMDG%J8|Hqkzd}(#&3a%
zh3kF`WbMCH9h)HiZGL?_f&mp|)|tgc13W-@QJUbqAwlecm4Q8!uW<#&ccR?|gn5E&
zz^e9c1kJtIAL?{pT=(gdCrPkxK;<^#n#d-f?4!O4dp6PKOX~48KD8%JRb+y7%^39D
zLv&*Rtb?djkG;HFFb{+;Hta9q*V5ARay>D<<tv}q7@fJfISG0qFck3q?}Xlk99|8~
z6HupGXhmPX{x?c%05_o9v%SQAS_7ENAAAB~MJrwG$K=MBEMbvOWoZggEo^u&rn~<;
z2E?iF-|S3m{r}EcA@jJvnIofEfc)9-wrYxGP4;vr%a45K>k`Bk;FqQFyly&Vy;&Q*
zHbU4!&?3&k;%2?E=7{LIhRkoxq62~h$kGz+yYT81r~cRITFmEoEa-lR)d9FD44wuM
ziUTtrcG8VFt)Z`)8S=YGSeWp#YKU~mTS|dX_3Bfk$HvA8x*mQj4&>Uh^}lMBE4Z}%
zCj>vE6oO~tLhemKX;&v)*nkJm@mz6Dgw_@3wsIb^0K0FF+qx48VJ-!zl$#3<Mwc3n
zM3-wdDv7BwI(S&G!k>t1=lo^{^?}d?aES`9rKmUN<2GLc8{m2DJ<4)9vH9SUBch6m
z6y)cmKF>N3tUqw>2<&!HEMky?iBTs)Vb9h9AK>}&<;!D>jeKHw!+!*I^wkceFcrzb
zd$J4zJIJW-l?Oydko^omH(qwKw}SH$B{VVKADyWIe>E<1qPLKfm?1+4#d5bO1REj{
z{Xmytac@RuFCaoJnBw_AotoBAw_r>c#}*=++zI*sulfw0<^z7ejAula{!&Js7gYzk
zOC21Rdq%$<NjMvP;od2$XCYzNgXvxfE)JjgZBJ!vyHS)33pacOiA3qBd8W59<f4B0
zG6|@Ytrd-pPf<*BjXN(80ggvT@YksP*rHo!Y8m31u>QyD)k+-L(N5@q0`Ss<M@tBF
zn0k?o4+3|L`Vlxb^c|QkkV{wPhFh%?fD7o2@A>oR&@`hze-;P!^jLVM2R}9O4``VW
zQBzBxLEC2ut8}-lUG?yHY`*Yc5hXSBXU}HAgsGHvN6U~Na|P6w?4}06|LfpcdYLBQ
ziV}nT3KW6WqrmtSVBKY`GKL4PC&^QTal^!qwzvQkmFv(@BLr-jVDTft;8fVDYhEMo
z-K*ItMB@QcfDPub%v%Yk5e5VH^{gx{Spb(KU~k~_vO!a!6%GjrQJ*Z*tv!q_rGhnJ
z3iU<nX+2y`Mo2c81N4qmaZ<Z#)~Rd;kL7<2<UrlBDf0B#xBDuf_1q~i1A#q7%ujjD
zI!6x_EiW(6!YX~3me%h-X5H;&cAkv2ueQs2A5H^VU$W406Slf2bXVCX-5lss<A(nN
zeTZd(F_Qti@~mHSY+~Ym$Sz%%-Qy#Oq^bVW9<~c#`uk&Xbz7-y)y7_<M<YDXlSr$C
zP>OT|=JQH&t6aVj(DPJZai{DCAlwq%`oPND5evg0^Ogn7?4VD>Uz2Cp{4uNH<Hv_Y
zHby>3G)1}8U+P2e3=UVX`<w_zIRi{K9+rov$Hv60tgK>(XZzhZU5VWck`nWn0b0w@
z0wpkRXJKnwg^L3!^7{PwGsD|q1o8cjw8-YwHs})oeT8<vRNrNL%>D@_?)rI1$aQ2L
zB7|spOxSF`#|k<g&R4yK@Y$<a0N#=6moimn#zFeXJd!=nx6)eE?q9s5ezxX|Id7I~
zx;`4!V|*4Kzj_LC4VzCB>h6BJuVwu}i(;98^G>B3DU&pGbTZIBmoXrio5rkJwx$ps
zRp<hp9-<Nkg5cmTB6&haD~3!$V19%gScNEp_yWDQg>KXF^IEL#S@v$$`d_Z95Z(y+
z2fGGf=tGtZ_AH392Y5BQH%@}M<us5E?7VSB=+~doiHTZlxNZ#C5P^-^RBh}S;(0Je
zvhKQc=~7~6;|jL&dtyHK7#^VstBK!8mkh^#*(7=RNcXDd8GaxFFFc$Qi{mh)LrjH2
zI<kT!dF`g+SQK53Lb;o3N2cae0N1fGnMKPv2M#Mt`mL?4$)>iWpF0pEeFI?xs-`Y`
zAD9}kz+<%-rHTYAiejt-+b*QfEfd4E^PpBKkz)YW$?pDCO%Ef8-pI@u09m5$h|}$L
zoM@r1%sY?vVOd!J3g*|Ip(~ul#)92KA<_23hYwul{n7Ai)@M!4cBHSh<wH$FnuCdJ
z-|qXMulMiYkHOq**&!&W@qi<!*P+P#a-~kTSeI*9O2Q_M<!RckPG!sNZ>;PcBvs_$
z$@pPEop1x1A_CeY(GYXOy~tH5W9G)_5uw%Yqv7&~m5%(_u?FzbB>qL69JJg<6etXd
zXp#wj9Md+Cm8xNCn2RA16%!*!O%#@YsTNEz)g?O?5{SJ??wcF?j`2N%_!tTgWZ*IW
zkDzJ@Lp2G=A04o-fTM88Z)OoZ^BOUcpvbhbHZKE?>z0d)ONK4L_Ap8$A%>gu6@`L=
zhXq3#<B)&=+~4c~;s|OR>iArTnz`IN6P3!loxpsbYHEnY4B&H3TU)#AYgKR&oNXvA
z!}*;W#N;F-eXIGgTlgNbS5KW0PgJ;=VK6*WF#_B-<)<-LKGB+7h_P>|8?e6U^M57m
zA?TB-g`Bu3FQ^X^PJJ{@Xy^OcsxwX^>wX|uUjQA90_{OqRfEh!YgbZQN)6?WnCXGh
z_bl3a0%4iyI({oUQvE;h1bqbZ*+7}A&F|hwR0a;cI*PHcF~;v7e3!tC22ssiTm(`4
zw83S26?<5};QRtn2RaP06Nc(uq8}k9&KMYyT^M!{<4-W|A^mU~wJ}x)9yjJvnIq<d
z_$*FNm(1@63or}>GD;w|4qgnf0lDJ8giZbxG88QG+dMFL$fi@hj~ESx70{|Qil>t3
zLkNZ(Bd<iCj7ko#CmE()Sb73bBoLj4jvxPm1g39jX#$`?bO7Kq!0*qYiyNI(7fjhp
zdBFmk4JH-|Fd}3kxT7uiK6xJ^hF76F5-kr}Hsdy$U2HB(7FY-|P-vF6w)A{_eD#*Z
zFayv4ztzYtaA;bfiy@e_&Cor5h#abo;Rs~jhv@F?e)rabHoXF(oa7GN2SiIoOtQnn
z18q4PW64|E)7diao|5DUojdD#>!Y}d2@4uoSbH(|<69))*kY8AJBa)mbIE`@RYCs$
z*f}Y%obzs<{Nq)8#BZ*ik&iEe(35Crd^0o6AJj8nusre!%{Vb3kER*wcI{d1y1?_*
zx0@n-d&aNtST=FDqW`tJs`Bi?fvdY7VVvpSe?%LsCQ)gj(NIXvF~F(3#1^pGvHR2>
zQFZlbEs_m_z-fDP2ne@FA@m3}^?j}?uhLQ>40V7ogdsZCfhp+Yv%4`In-rokl8*=#
z!Qt!r`nq3{)7213bQq7>Fx`6+p}`W8155#a`&Bas5E$YL;Lyt_CN6pE=fcuT36!TX
zpWiJ<XE54=2l6-gZIonq(n0zFqGdDq2(vWP?YC5{%s(yly4xiLosov{bSP}OA0Lbq
zr_B1t`=2LTDl~A{|5rbpGs1BPUIwj4!@RwBDKg#8P)HH<fu78UVfL%%Q{T`1$G9tc
zm+aJIw$rT@yR;JEjKA}9Jk@zE+jO}ZZNg9B4nv6^%nl=0Fo)z|?Xn26-p9(uhUF(y
z1Pe^B6E+c1<Lifh{`|Re%xJ*`ooGVF!-GTA=h^R85S9*P41&Z)_#s6F4ZlOaj^s2L
znw<1QQ$xfW>Ze!_TFB#QTox37wGEDY9rXN-A{~#dhi+XKAeo@^a6CfTg+byD&=p~}
z|E{{ax`n;H0s0Lwy}R!%9w1z_O-*ls=_3jmP!U11IjkXrwWKI8fYZ0Mw$_0n2pcrR
zz#kdRc{4KZx^d$M#xClCR{~Toz>Q4EnMaruD<Db&O%ZBC<q9N(aKh+4HRTVs2aGon
z!C2LA?@af4U|0@cg@$%qH5`BDqB+7k8ZN1Wf4GP0i?mCeb^ySUm6J1s(c^P);t<1s
z*n@~JLxdhd+A_`0Ubu%olduYaE+Z5HpkM;kYL5SexVbPrKnFSoJuK8S!dK??2T*}1
z8t>n~zh_`@1KXM3WhqN*ofz`R(GD#wLC81+7n(Q+fB}0CPA$+t?IR$-mKjVy$e|*`
z5QXs%4scT^Sy@>%ZI521kfBf5J_FB>Ho>J}l8ESBae%@>%*z8K5MzX4zmnlv1*%g4
z5=P!OK!PW%2iO9#Xt65outZ{X9m~rDC5%EirA0#0>37ZT?yeE%9`MazUg{0vT*Yvg
z-k7`s)yHqQ4$?0OD2LEv39Lr!NdQ2gr5nkuvmhblfn8Bs3DX|(taL_=JC1q)$95XG
zt%FfINLlP?GKpb6q9SHeAraOB5fB*-&}%q=xBn#=VkD_!9_a26OT=Uy#s(ntK0t$E
zh1m{35u92<0eX3=r{Dlk4_-wm!>z5s=#g6CYeF5yDBFi~PpymL?10*F9dIRr-}>CM
z{)d!@4v|(kPnXt!qmu?u#4#LWSR&9kK7;`cbtlW|j}gYW>adD%`p5^?5Cc>MmxaL+
zEubho;y8pFB)D+(NnYEZiuiuQ+yPF39s|a&o^%=_r9hZ`ebsXhF(rw1h^X)|$&;dZ
zZAFi&2P=M_ac3a>Tft~8WV5wJuO;l8*yP`scHgXr04pvjsf`sUkONllAr4l6tpwCd
z326nglT~`=!2~J=Ne>h_W~ieaB^?tszE7S!fs8z5QgBhTtR%-_Ry$D-da?4#;<UMg
z5LDSDv<KVYDoaWEZ((F7Jm%w$5}@``THnAie$@8%o?CG4hlIRBd0+&s>ZMoocpLpI
z)j#7Ki8sY%{|_>h>NM<6%pPEop6m62j|?HE!M!?h%Svdv?&saYHHAQkKJS|VLeoSJ
zjBzL;s&WPOmC++dt8#~^90k3gpgL&?h)$fJ!1zzCz#L>E2T0JjB*fb1woS~wCx)s(
zD_sMQ^D@Ooe?|NSBVx-Ch>zr}rmL5`^}UJ4)Vu@$Xbuz|c@Ri?NHyKDakzPA4=88r
zCY=V!7=|ZKXUJtW0PJ$C9+{rbgbwsMs~_?fhJVYxbHBvQSJjMwu&@Sc2*Kj06d;!{
zVcx6Ot`Em;q$9;Mg3kM5=L#Yy6lx1C>tuUscnc0eNi0GN`^uQ{?>e}R!<yYdtIC59
zjx%uei(1>bJ9@Ub(P!v?SB!jJDJ@_vt9eRC3=MnfLG2;J;nB_m`o%Jk6>*4*SUqyc
z5=f|ku)AP1K)5q0XGr00;L~DRBYEuL1^%k?=Ec(MfVogFD1p9*k%0Kk=nfvdw@~D=
zJO@ASRtxa<=s(e2@}`-CThFKwvHeMb!zrqf@n?eBLr$}>uxPS_D5pSTcj!aUfeAx@
zsQDlVL0P|5N;k6c$ABnJ9UVA0LW_kr>>glWG!A^xta1H!LV(k8I=V&>!-TmK6djxx
zX#180MQ(DG9{{Naxv+8-fOEhl7zDjM#CoDG;54FpckJgzB(VL$b(n{Uz7EPua}FSY
z1>UZ%BZy1oH2BITun7`|;Xk8Cf|D$d8Do82<Ho`2qO5`7qh__YwjKrt*8Q6t*$>+X
z7H~>vN_t4W1h&V0RPsO<B?f&I(eWS_;fcJ0hr|cpMc@He83~C~^z?Z87`YOsv%Yh9
z$4!H_?6Ym$<$IXVr7QU*FoGt&79Ods=Q)s_xIJSySUxRnP>a$`EXFE9K82WA2hDO1
z84N+?eL#501CR2xy_U$k^1?#LusL8<r_%l=(`ZN6>S^ln(b-N&s)Po$38$jUPo*-K
zWtJNG9G}Tq0exXHU@=U{qFqW*N)4#XDJZy+**TBfXf1PfMqT}cJV?x0V%C#z-KhMG
z(&TTi1=jv|3PmL#8<DW`joLqgV1rL27^e{A^$m}vv>PRMq6X&W<q1wwQBjHG{0B4v
z`p{Sj)_@Qr@XTvw;`)2*9M%MN7h?@RMPy%qxe$E-F`>jVs-mhY-0a9|)P2`o=Jaq0
zPy|b-j$BaV-{gzsT+DxW54PTUZ*d9_D;Y==6&0i*F$s(ESgI9;^h51-VArmf$P73^
zBpb(xc<n#_!n6>TUmt?MN*Jo-aW%D`j|WDmyP>_vDaSV*;W`I-rnj#TM@BH<K&fX&
ze?o?aIM!*x>%E>5L<b-Tr`^<@EnBt_l>)E@Q;(JIits3i-@3(J^7YPZxIY2{h~rs^
zX*?YNP}%hj@x0|3hytuMd<!ImZU{mJ9$8|x9RaLT-SpE&AW?_<36%lg24q|RcaSFN
z&=}&SgX7~*q?6B}jUkA+^9(v704JcoZ(!&atjG=&Hg1$yE|<?ZScQZ$cg@xT!@<-;
zU!FC$9HFGdtU0!+Bb-D(T9WoYi)HbA@}w2pt`g%(v5<=s=T2PIlnY_$8TkY)5i<(S
zPzs5Ic98f9N{VnlLps4kR5{<JO9tJIz#qi^q_+Y?2QYH6vuOLaZ5EiwhrjT7QiK0w
z{HxmOf%ak0?Gh*}=&`Ei&QUF!G>Y6Wy+FGA9@+-57QIEb^@huM%Kk-fIjn7M32y?<
z2_g(g0Lu>3rHgMMeFfH7LSX`60Yg3Z{tM_%Y7w`R@3vS-uzJL4N&#Z?kDm*<<3S+V
zg8+0!enCUwnUJL|DJ@NuT4V<UYGGLljQU-;upc&%#NW*|$Aw)Y$~RX0FwvlZqZWw%
z9Q*r?TEPX3?E`Z6TA03wiqe4|X_HR;pr!QyMo5v(<pX#qgph}MJK|iJYxS|9(P;v9
zV6<>hu3!}S=?f4X6~EQl|7vQ80JoJ12B6Teq|lMs_b5(akZ1s{27x{Xv8%MOBG?H#
z1H`Vun?OjnsAmMzLX2b}4(3q~<WOe_b2UMY!<8hCjuOTPVKCUAaJClS33v;M0OUhB
zHY8lNI^L%kaH-LPjzM@%guBqs&+og41DXf1@k|_RMu1i9Y;6OB8a!!PimD%iQIU~a
zrlzKOap1$6jd3=GUepvJA0ya?r>BWRLHKLXb|S$NhZW(pFe)_KaV~Sa3C14@*}=h~
z39&^Gj_`<!;Xd~9J|K>yfG1R_r%10v8RQ+OIxPg#iiQS%xRNYkgAiW%k*GeY);qE|
z&BTBX4(NX7<KBxDevLl+=DIgV=H}yA6_lsQFN5Bp;W8u?-F*tGNRj9SFa{u-v8t-3
zrhubepy^`BoH*o)-;5Tl8IIq8i1*o&7ODfxojEKxw1iR4-G)BBY+0@;0%;P;2q$N*
zO96$@P3k@cYTG&=2>h_&89<zZpAR1AD)a@8I3A8*&2y*X`X(DJQP`e?I%-Duf+E>~
z@$Vns@|iJ-DmH#@coFU4VemMwB!YkZ$B&5Q(~R_j2cPRwt54YpeN;kP`XB4rZUnNR
zo;Vt4H3P?Nfj5u%?~!5XK5k;ICh)<i*0@)~`-p1RWXOflk>@C0Iye&yY&H={1dC7j
zxB=fVnK|P08`g`@07`FQvIVjN<zwJ;iD0@N*Z*|lY?D_I_K9Hy^az5D&UT$DZW%&Y
zqW_+ZK(Je&lG5%LcRbStUlU{B@Aw3$_E}h%7{ZJgTxG|wCc=tE{rxi_&1iHdSZ+FK
z`d=Z=Mwkk5^rs@W=74!QtjDAXEe6~2jYg>ea5?i)cV?N><?hP-_>^l&@<)0an}$>~
zA37+pM}Bj3Vru*o7Mz?1lC`R?PEsXX+Y?hi2_>`Dw7$W?!32c_Fh`sjPzO|P(6s?H
zUxxq<1wp2VTvLY&cmpdPRp0KVua~A7fl%deEK?|$n<QW#N$j;yOvSo%roQSEJ!|hf
zn)bf0Pr0f1$liLCjdT)8Lqo&(5M}itCaKp{!3of2ttc&OPuY`-I9HWzKll>-3aGm%
z%xMse7J)4wY8bs|$MH8UKisVmy})&zF#m#PkU^<aAr6O2e93W63Q6+<j-?|U1)$PG
zf2gc_GQRo0<0D|NRBO9U4iB8S<n=APCW1eVwT(tCr;v~!U7kZl=$z}oB>SD)<R5jR
z+7k^92>tVr@&U~_1<8gazz@npQ8?E5<niMX0hPI2*<+I--3T)kP?hf{H>Ni?p29=*
zlfP4Zm)2WJDKzg6e+l0wLUV5t3&0;?PR*0b>?Eh7dkWSA9g(M1^uvyc**~K9K>>m?
zKGkDO|6Ppp)b-j^pdN-Iq0A>k#S^RN82j65$R%pE^@^P>lk`8SWSG?x96@NB+I!*<
z&QsSPz-jUS?l?{#8bODw@5|ku^rug6L1vm%calkfr4AfihHTy8tGk}N!>Ra>tN4r!
z2*04H(SN#ZxJaH&yT*#judoDJV)A}h<Sx}+<fA`++-bXgkp4PgF$iIT?)YSVmeu2G
zh(-C@++J)n030uLXH6M*FChSP@&ur8d%-z61B)w;b9q=?d<l##FViG|AfD-A94<i|
z!-(vQJZ#eYp1o!`PRKb7D1=D9ILru{Xb`_$&TxlM2tW#|jy&j7xUA5Sc>RHKnIPgp
z0}^dJCg=QPakyIfW($suijIi^)dJB!4o)-jLx--QvnGXMPz;A;Im`^IPjWcS+#@U&
zxE6dnkPzI<b4LuMrz;_psQg!`jc=%l)7*FuBOvT2j2Pep>_={52PUmtK_y{>gow(Q
z1%Vq4AaRDD7aH8mn*SW$2v#*$u(2Bl;lKjo{b1`(uoMnxph<|PfpBiX7ess-q7mz5
zm7aPqK{trd1sHuF@`x_>x^pmyIQsbdCIEF2=e0$PxP!O*+k|a#to{xz5y1TzsxoX}
z{-A<#F#JUb&%~)K$gIR~2uTK5_hjcVfQ|?B#plF~AkNkxWJPR^ALzi!bcm1o|6IJK
z0FT;`e*u5Kf*?q6n!xlBD&a(E;QWe2&{FQcX)^ExK7}Tjr3l|JW3J=;SYgL7^3fo=
zM?z)AafHvo6Au3RMfOTWIuq3o*2i45|LEwK(24u0rtcWUh%}A>i6RcHK`+7v>j5I<
zg&to8K_J8k*z#EpZNocZ#`H&uCB~|V86ZM&1#YAga<VBKxJhhE7&}Hzvfggw)mao?
z94^bw;x^iQyX;+nukUFbsEEO>;nC6Umyt27dxcM)^neCO3Z(hJQTLVsS+!fgFJK@T
z2#A!5ph!qaHz<NgcXxM5H%dr@ARw(Y(jeVk0@9!$-2&1famK{wUHjd8t$ofu=i6By
z^pSOQ-}9dHy2dr)KYpVPwDv$B!G{!!A%qt|-9*~@!5>CFn2V4!0s|W|Py4q!6X_Xi
zDuNY?P!f=&2lyo0K@E@=fx{^UTB{XuiJ*?B4WSis!=SQ?XukCH9)dAJL6$l^GDHUt
zRX;)aDYyhrn$>~b^n46F;-^PU_Q18<$_WJZ1Z>drE~97eGu2R_&)e_<fGX^l7y1j2
z7#VS(T@TU#{vv6N?=cE&7A26P2VBDSMK+tS5IfJNMQi9l@QQ*c(4d?F=1X@F?Fb70
zTW~-SXn@uz2s~usRz8q2ZcUz@f(KZ6&b%@=Psp#lN-A>M81+{2MbJ>+Kww^A8`7cM
zk4Ci==`vU>cYFHOvu_o${kV7UUQaCxcXD<nBBw$k^xXrHU+X{-2WgCQ2le2Mr%-RL
z)dA@j74=YxcNwfDL_Poo0a)=6T*4I&EfK)M%JQBMvYEVqZA40F;GeO>I;ai#^KdJm
z81|;*4)2^k<Z%)24xD2Kcx|cWq{Vr33o@|}y06@z3<|%h{UrzBRK7N`pTJ*H)=`7g
zl#~~|MKFz!;sC#Z&yI%QZ$?-J1bnY6(g-bS%+&vHYm>b3YkU94#mVdH@r00HeZ3Qv
z4H{-7lmvG>20}@YWYG@hb&b#1+SokL7$^jJ6{sWM&$|IL#c5`C2xvf1yn_5_BN)u(
zi7GZ&rn8Xx1;uyz6aL1Yo}MK;$ay_@5rjzRo;oYq{F78)Hzf654*RDXNs}|^WO;Cn
zyMz5{qzK#wTBMx~kf@|k^@9A~w122FXaGRHjTjFr8=DJLP@kz-|M!Yyk@Nq%ili@>
zDpQ&3Jmgv#yY?ztvPM=RBJAFDBj9s)E!AHT{7<sQ2e1dBq&Pp?q5;BYC}IXjT|!5f
zg|M1NcxRRYu*6bnB6um~pkX&K=vRK|o%OEHgjA#ef?p=sh+qm{z$Su_<}uK3(!oau
zG#pfdA?1sRj3Ie~P>9pTV$ie(1>?q`_*Cw&aSz17NKFda<?sl=r4QfmrT~r~l5&Ma
zj{EVZ5F|&MA(8cZXVVnuehB;u$#B5Rv6*YfMpS^%YKB4p96UIZkQf!1k#E5c6VL2#
zZSey`8m!hUXl;YK`F0Lc<b8`p@`#M=4dk{VKe-*W4et%rYOc2Z1~^&ZKt(<miilnc
z+1({Kkloux;xPxu8lvEUScwq8szA;)ny6%<ul|h&Mgz%$_(masMYP|rpFt)@2l2!M
z*u%}>dmusxL^8Ac0qI2t_Xhgctsw^lNvE$6PR~EO3Boa`v%J@W!VcUHgfR>`C{Rwy
zBE8#SZ=Gj~fFy$y5+O4I_mu%!&YRThIVMo1rl+4lRhTm&1Qsu%B!l$m+lUC+TGe3i
zrCvh63K#(c$RI+UY4`~(M*V-T2%cA`EhmtAFUYYzLb?_)RKOuak%Ev0p~!-x`xYDs
z@EM$j4M98x1xj!={Ij#Or!T*UO$37OT{i@Z3l=y9HaZjl4U9jz#Q&p-1RpaM%G4kx
zD**Cr6Nh@8I}g+Wk%Jh4yFd+fB+P0A_8cTF3&HOOi_(H%qah~Ujz^<_g$ZZ$D=3~r
z)F_Ami6P}Uy+BOrZ7fYC*+23y2nm#ymH<q44Nyaw*%1`Ef7@Fia*8U{BA16WZ+yHM
zq<;W1*D_<#Hq(D{8#){ygNSkcddtX2S^Yd-bU9|kIf>}snp>W=LJP_@(9U>*BMR9F
zNChxL1_rE~G-XE;Ae3gBm`_zHyeV%%k}HUa1yQ{)fj<OV?q;Zs?#bzcJ4_@V7tT=f
z5)yJCy#!tI?N@6_c@i-miHZqu{zU5DO-%ozafLMTJ*YHDul>6#OtU>mCdg1qQ^m<s
zd+G$?y<Ui<C{>QJhWAY}Hl#-;5d-P_drw9f;H;($jxQ!0V!j#g!P%7#1ECY+JC|hd
zn;h?Y(X~aOS>}z9WINn1nweP8Z)OMeqRUncD7?4mzeIC!*=C9Z$P$Tx;84cR&b9S=
zfixZ3Mj%xa5%cDP7!#dr5QZ32UPiF6^ev>1HYB7K)RH3<pRY*EKpC$2`TF`gitob$
zN~!>UScBk}3C8-}X8jkm=pzI2{=9I*mf=cfNWnHj-3Te*B!;dy4?_$Mb^~d|dXlyN
z=wH8epFvqk>19{a6Ds(&n*=2)>T@JJkT!($7Zi)c7rOBO>&oEzg#9w~i7s&cb@tC&
zlECf%f5guJ|MN9|Iuznq48l;~^z8tD7tzqR^t}5){k`W$@4m-dd^>qD8Gm6Eh$=+E
zg+|n)kdFju;lnob5*1o-@tQ%m0?!&ruR~$=s$4n}A|-|nGV^fd#bHOs`^pSPGDwq?
zp&o+}3yu(OME?Y)@Dq507cXDNxqCNZZ4V@a)6k-30QDj$ry#B9AiQsdTtZ@{7+euQ
zHwm%C&nb-nOhp&bBepEL!661=c+2?zE$7LsOtZuU54<~G$crJ3NeE9H5lcXskk>8-
z(F!~6$qek!jM-fBWYhTF@SR`1nT1ijg%?#!Tznq9Sg0`c6nup&4`gYq!C5iehg3uZ
z(8P>lz<q;?*1l{s6<nI1;YRfANI7;yP%Hm*i7y_|wjOm=XFd*#Mh=OAWjL9cRMr<3
zT0x6hG(iDU5^$_HAwL9hau=k64nV;J$Vf@`N2wH1;1&9gK>ys+qYuw>OAjWq-b=TI
zON|@@uI%XyweVlo81iRsIdbu*tBff0^50TkoQcga`q;=z!R5OWft#;)@Ov6zK}3r9
z0OS*@GE`Rk>!758RQHRhc}U8&4G)(-)#sB0Aq(XS^({`=ln?X5o6osm=X5Z?)p}Yy
zHl0Bunp?5avWd!K(EQh%Iv$=nO5q8mJ3(l+1Yt(XJk&cICnk7)xT&eDkCj_LfD8h(
z|KLLmjz~cuzjQpPk0zGL9Rr6lu?7hJW|o(QfshV~AW7mGn-q|o7wV#ezuwl^2wG}c
zO89nR<svxeR^r@NP~^c{N}@X;erYfmC|?-FbM{r|98MG9hGqopNigZaCdNl1EfmNJ
zLzV<o%kNifAv^?)7%CWZ$ghAG=J2}n%YS3l;Oq)1_OHZ|{I4)Ma=tvfv*WJd4|<&(
zN`3jl<fq@QPh(o-!h2=4srE4L*TdRC^f}0}1j2g|i+jS%1#mvl>fw9_oF5t_=r2$G
zo^FC_TLLU&grE;Co*^3fi_quQ?eKT;@xt?SJyfJbu`Jz;<nl=J!g+_Xw><C`Nti;D
z7Lq6ep+XmIsq1XkVDWVf_Lh7Xz>0zs)CjE$P{PdsKNN@xn23ny?^0ZxU7S_Cp6lU8
ztcbQFs7@Y%G&DnH_~V@_!~I`lx;t<=!?3|_W&Yh&E>y;x&sQf^u~g5#s0gDZmFVzc
zo$D~fMta-w2vyEBOlc%f95w#-M+&@qDLhc9S7yDM<jDJ9KWaBRu^D)|I=PLgoROVh
z^`7~zCXve7XZWGbzkX=fmWRDtu$wRWpf@>BBI_TyN8;nzLXZ9<#an;ht=T5B?`Scv
zFlxAtjO?GcBTgBvD}A+SDg6(TueZB01^J3K#3HMj|N5ceQ9_-z=a*AF7KC~T6I0Z3
zxSS*73K{qE*}<PitXi+y!f>6mxmzJ)^l)9Wp2i>^@;bBS9Zc&E6`hEFKjD6yxE%7m
z={g4C<t~zwJaSf!;1A7m$t0OhX9slelW%;ynU<Cn<Las?t(fvH(!1Q&+2=j8+tDbW
z@yX*?$RR4gkU@!wkyR;#q0--^N6SGwB0#r3eN_JbO9pol<~KY8$`;RnK{?q#3=tW%
z4ckoWT#pt{q;X+0h)Tq{OGqQvA{3`8C)e(X-^jPPL@uSU^r^0HNJ?-^(<JS^Q;U#C
z1B_f(@iIb5$bP4>Dk&@7I-Y9rWEJm9oNzH3D7}7rAsrj{eNK4}?O-{l5}qB)#q+DG
zHL^Q0!!7x<Np+_tk4*26{*0{M*rQj4!YVR$M`$ArGUic<sbe1u<EnW?R{BS{zAJHy
z!QZNx+kC%N0phfQ=2$VRqfL$RQ9+c@^tUZ*!iRd>S)AS|MBpRyEML)2w1bz!h4~R3
zt!!YtwWzvI-h`=%-?iT28wS+Za?L_wa5+q`-Z9^nn9KRv4Z~P`9Auzn<NVigo`Dx2
zYZr=Brbo-*9~xfv;ceHe6@y?&;pbG|7INs>I}f`_Q@RF4#D7hw4ra~-7bir1(wERR
zKh3<fK5gpq*hY8?POLY`r7S)o7nd}n_MVWq&mi8hXW%=k*4mba*Z7p(^lGtLl@L=|
zrT#sV=!%k_pPhHM+bX5u0E65|y3OKB|H&h7N(N*V&Z)*qB)J<Jv$9bU*k$wbrnNUs
z%J%Pga@@gkU*Y@^msRx8GWJe*Sq>Z9hpvgY{8y09l=sxmHT?W0DzzK@Zi8gewY;ZU
z8MsaYK5x9F-$bcr#~n5?SSyzm7fOk;eZI+mD-8LS`vJ*eIYWj&R-z6UE~N}th|NyV
zgkLq}#O4yU3(2*JZBv<GocyI_i0b07_OYw$b-y~b_f5f<&_kS9Jjh(rr=BrJ{t636
zw%oBLNYjFZf>lsREq&LABf&2vcyrt`L^&P5!<pi<SIHr+I7N62U@OIlHbCpt4#~@A
z-2P>}nFx{%-fm+twP6aDhsN7ns&$0pHF0Z1B{9@(3Z*o<lh=N5t5}+o&cY2&W=1r!
z6i;SDZF1OeN@ouE>p4(-u1O<QY@+YWalaeCmRlU;UtA2h3vg&3OUJzwFuSOY)p#Wo
zgW}gS0!5uXE1pwM5njHhy8(&Id=iEN3D`k7Ok0E_<vXyk{ai&vh6)#?Qb`ZtCM_N`
z`&6gjDlB6aN)XjFIvOOyUsmgV-r9*l8j$<pJwc;xV+D>)OW2&cla&uK1{-FYYfW;m
zLnCD((gid(Q0TGo?(@KGUZKH8mAM`#HmN`sQ1BfhHeSs);5F(rmTygoI49%V_;3k3
zyc+Lq`F@E+h=iI;p>?DKAO5SFSwrc={7KSxW^<yd)9)>!^tIk8xw_iE5AGoeL08Nh
z;v}IX$s)?ar`|Uo3FR;Oi}*2${ALWVEgvkuvu{&u$0V}ln&6uCONr3XbTK0@P$64T
zH@#tKSl;<rA~w=&m~laS`E0;kI!z3FL;!52Ml5e|wzcAQ42>82=5XhPjfKb7b<-JB
z-nd<Iz3aAVz11+nz|*m9z0>fs|CB@MKBgD)O`E)$<$BaYQTtWF=p~nn_NCp##??0J
z?kXpJK*sup*=;YPGA=G=Hg}WNq$3+H$5O~LXHcP+18tKAX%^(YQ~3&C@HO@PFY5s%
zd^<bun~nFM#nqR5F+H}P$0D0~aK+=MgpwG+$ZUR!Vo?&?VC(p0%i#T#hrLuhec4M=
z44S$-0v!}!sSmXX&K4g})IoEUHQ!;aw5Ud_dRZBcWtycG<tp!us2j@@C0MxCL50ri
z_Uj+A5-s>0w25H5k&(Q4KG1f?61}cT+RT(niP`7h=iNd;;WW)KJuX@sJ?p+l?eaM~
zTLLeGTF(Bn*~_C&rRaAY&jm}7U&fMV5LT;*@QlHXRfqy`tw6)_4yQ%)<V(vw7c-N{
zi{|_xf1F002voKiD=m4}xDQ<xCvDxcipLoqZ;8W3+6wDCQLReG4Po7yruJ^<%ast-
zQ%mi8*a7SHCN}Z206Ov?ZiCb;i8m6J(-k&{9d%8$<8}#D){1XD|8O69kk4Q2=LenP
z^3Oe_qV4!C0F2#iwcx4%iH_65$tZ&;M=oc>7?l>$Ssx#j(_*SbTI6?pAYFd3APs&k
zc`#5s*lke#k5%~G?+;dR<xfff2t|#>{xFDnI5yIUkL7|jXuz0NO-TfX{9Rzdk4=HL
zG%bCL7jF3(NkbY}q+jW|?<g-+S>!jXP22ac!Rjl0#hS>sF|EYIrpi`<{rnaeaCV<h
zXIS96uS{>_;aWT&G`7mdxG{fKE)N`%pDd5}e%5)%j15ej<}=Dp%V%HXwd;%u!1_{0
zpHYW(y(XcsRO<Pt1^XGg(wJ8Xt`ZmTamR7hgHiU^6=Jc}0@mq@JL6*IG2|UrU2eWG
zGcgPK6Io0ft#1<9DG_O1XzvtKf}gAf9pAndn!$_5uQSC&P57Cg%H43`#t@~`sFMGP
z!+%=)+C^U|c0ZHfNprycrGwRI%HmnF_?sm3=UZOKYi1Va@wXP<(->%2Y+EIKUl_ai
zP=M-vy5y|NI1AlGP4myW8l(lmulx$w1sm6$7)(68yVOJ_bBCNqohiNE9bp#>2tF%y
z6P6q{4VN^#SnWO;XUC`<+|gGc${f8y&59CuOo7RnfMB9zAfYz6l3|J}79>nQw?RB!
zjnkBG6CzuY_gGvrH>kL3LN-z{K5cpDPGvX)s;%;sLmr+(8%nIx2~v(nCw8yIhDmO{
z6!ksO^Yhs9{(<H=_7!JWKUzpwJH$UlH{JZ6g<%?M$+1&lKlyB*AuRD1vf0PPf9O@*
z(z)~>+&eL`FbIkEQjE%yM{H1Bm#l~Pa=4|RPJ2@?1{Pn&+x+U+N6Y?GoAXu&#C~-c
zWeW8?T$%eG-*&{UHam*daur17z7ww4{igY*CcVq)XuZ?>IG*ojOJqbD2A|v8SlZ5k
z%s?#rko18yrv>M0)*Z%1Io(<E4cPWMFE#tDD8D3r*bZ^#Np00nB6!5p?(Vdp$4FkY
z+M$*dd#cek=~{_GvzGM4`s<Ev8Y*EtT|m}F^Ea?r2-)=10_&n?Q+jiTh8<0-3j@lM
z^}C}EIQ~q}&Us~cWpE#fy%z~biMmxv_ig2_5i3#8DH{6i=Q|wiTP%0fFSAsp=Wm<s
ze9hL`FcbWN%96PaqDYHt4`z1m_N>)4v^240{m8@F<&<meubk6VLBD5z#=1@Gb+k_I
z$BjbGGPs)bC&a)Y6ela=)5xdby0*mJy1sN}9-4lft!Q?G<J8uS=m?^uQ(Tvm<rVG7
z=lvfPw~W@VfJs_>=7;7Oc6`5voPJ=C8abnwNEgp;aVT*Tl+5YdPDhnJ4y6-sgFM^v
zFM8ZZSEaY=(nN`EL^FS2t?c+bt{T8?C{Vw-rFs-<h--BF;*h5M0xdh;J^Sio1r#<q
zzNz~BPve-O^P>w_;Co1<M7w2<b<Sp0#RgtzI`)w7Y}~Z$$5&XpLKJG~qQ%5A%Kpx1
z9&;V_cdDjb<(D0W<~el9D7bzj+<`7(f8jHKrvL&ugXLVmP;y@r(d(zu`%iw5pxM)L
zgob|k;w57h@y>B)srW0c6Jg!r-cm8+uzrzPuFO~=qrUm8+GpZId)u2B(nP{&N-|k;
zQ-5@hdUI{8ev7i#(79Dl4K+>ux*$F&VwqdyS~|PUeH_2EX>lPhzP!63Si4z(qR?`U
zTw<+KWI)kvGj~H}W}QG{c!hm$!wp5n`A5^4J1ie9DePhDq``${+SXD@Cm|gb6JDH_
zUr{{A!lIW2&#dCmUkJvb%A;6VcGujXwmWU*T(GqX_6dF%))FL<Kl`Hbddfxm*?mWI
z*S{uPJUpI3p|`K|i{(??8RdwJ-)@y3Hf4)WrD09CvsvwMm6P01ebl%v`LR#R==ib{
zUloNiBA{B0ZmN<{?m2Vwdf?%UYs2Ml-qj=1qxyV%y;d$$<<&wT*KO-02q)KxOkFL`
zE(z~HlSTdf<?+gA9+!nEt&P~zZRRf>9FKx3y<OBWIG)X8>PzwQN-(CB$mCxVL>J!g
z_ek`O5#+7PPADRIU`*<DmZZ>{(#x1G3l517m~b#b+q+9lO@q`Dk(+%7dViLn!rUN?
zW+ZZFd?2D;VWYl-WNJ#MoCNjHBL8pvOw-+F(vo4O+&3!bFLH<Rvc5MbL~xOD6mC=Z
zVHcXkJD_cR*Ejm!9nE;J#RJo4s$ge*azVcHBPTDtoA$};SGof3N=g}L?B)8(X(J{W
z{$oKfDlfBm^}E9Ron`x1X4bA@ZrQpEc@~1vi2{#)3ZFR<nPgYDaqx=Wk-xpiPVU_>
zAyq&hSKBkbwH%`Q=3yE+HaHOFKUimbeq<La3d{CG#sW;$3;&-@To$xL=^j1aloAh7
znFXWlzd5oqyWXWU6Op`_f3Nk|PH6R&C6958ZlMeZ5<YD<<yZ;HSl&NvcL}AluPiMI
zMlQ^A6wYhX-7A0A98-1dvPb7GKAtr>M$EPNW|ARiXyxTln#0%%W%Z`7$J-ZPB3+{M
z>*3XG>?gjyo*K&&alNlu#RC=k!sDZYB=k9Fqqgv(V+cn%GtRiF1YZw&ztz33OY&1b
z;T@$xwL3d6Bgb|f)<{M(8s?HL>8AaWBrh|nZL}dTk+y-w_7mYj?#k@%OLh{g@d4ep
zcnLS;#YC=+tZOpGK0Flp8p%!Kw&xhWxH>0#=?~@>Ji!2r$YjdH16KSoD1@}YI43|V
zFcXPOD^fTph9+a3q7C`*N$I;9MQ^`dcqWx?m`8hilgjuPr$->4gka%DzqyRvm}GGn
zv0*}hLdR{nfpm`pQ)zaK$X4xDW|=_O%&2=zU0ffOHHFM=@Je=BzUHJz#Tak@l=qYQ
z+R@n&wxoYAMNy5!4WxMk8))lhDh{Y|I%>U~={UP5CqcJPJeXRxr^l|Shi;4&YhR-O
zuFmiFO8khKk$BQOf$3uh@i#5M>DJe&a(6w{{T6x$*bY|tnW>B=6L<9OHFJMgsZ-xG
zHuxgJn2{_YK5*+QnoKu`UfO(W<c$fN!&)gR<#8c%x9cU{a9R1EN^3(z5>7@Rk_sHb
zvX{&aw_eNz^X?Zes)BO$W9kW3<C2G*@uroT(~<ZC>*df7vNUEDD?<r*l@dAEnJ5@`
zrSTIamANV}zF(a8+>*1U8~tb|C|&Nnckk>NTjFc(4wcX^^_<N8A}rorb?KOSKHNd*
ztN{%VlqmH6)HQPnmvMM(+#z-7wVrmrg1_$b`~Lj1sI{EIVL7|J7uWGUOv(pSeFOP4
zMnXRxt>x}ecvRahRoHjU<0%U32Uz2oh*dJb^!Iv?j#ZUTVlPUeZ>w*pVzAT`;xss)
zz~CeKI<x*eYD@n+C0h2#T+iGX<1ctics`*2m4SARuL0Uu=Le{zt(Cd>R~<gNCaAGR
z+^#q51$tFWM6tE6!mP0Q@$F9@xy{RSBaW%w(p0uAB$TE`nane6;%u%fH7}z*w+>v4
zFiS);44$)n9<iBkY>2yaI*YlvFEd)Vf`-4Rk)iCM$(_BJum8+6T`Tu8#J4!MoAFgH
zAvW-p_j;6jCOCv_Wo)C|zIK1jT2YHs(Knga940wqJU!q%)Hywf9$zRMXnQ|I{Cg*Y
zlaHKMB}YzSRCzQ;EE#4cD1K_yYiKgfof%k5iaMgC`dY5NAiVHKO!P7k3Yn-byLh&@
zziE*R`hc2;!qLkqOBPvP-nidXe_iU+GwG33xrmDew?qb(Ic{0U@kt66?v9J=gj1t2
z8Ee}|-OtWtp1qhbB(xOod@+yha{hhFNpd%fQhvR3))WI#tYe}IYm*WA)3H<fvInyV
zZVkfJL$;sb%XTW*w%bo$G)E(Kc+P&;Ej9sz>u_a1zx=ib?J9$uft-C#kO!fMNo1~I
zZc$0*iW<uX*NqL|X9v7bBTvhDNJY?WM4!-KwbxIw@GvhTeJvJP*VK2yWqeEO?lunh
zCl>=_&by!{`v|SBAF{<44c_D6jZ9lkt&ko}W<SxoAZMmD$U<EL4k|fhbDF02l^j}v
z2yOOW(NFhXzoEaD>sV3BrM|+P@hnZ)>W}MDbHCwmL&<{zi~H;@&j+M?^>mD{K4!Vj
z<uW_j>7^U@fW2yi^Bw8FAGb%SN$^D0pI4XivdP$87HN=g2Y-Rj*XnSe_gb6itY9}I
zOP2g2p0Vt8nUm!O>_3M3M|JYwU(2SwL8lTB4rA~Mb>p}tWSPv;N6Es)cQ{}aKqZ(x
z|H2#3PMq=)rMC5<ohM7p07>p)ouYPq{?u4`rl~wN95szQ&OW$`8{2JJl&neqWSn$9
zJ3VN$boNoUIr9>*0+~0dyg0YZxyoXHuSADFisIc<KXQG8oTi}-E&DNpBbSW=-k4v%
zPLWslvhxIm?p|$2#REl-)!HOnJo_-6O{1WQDjnatpLZsknTeFfrt_7i9Jx>}NG^rk
z&S5O0Pe1#iW!2D$4OuO}$O}7Xgzifh&Q7mqS2mpGz7-J<tTadqdFc7*h4d{5U8j2o
zq>r6eF0<*}>sHMuWX<q>PW^zBEmdBgjb>|Q*xJe~I-Ngza~W5ZL<Ma>uRhoVGcAwv
zeq&?2je<ga$(OnKNOF7{TH%t)t3!27EVVe|XKNxpqn`(Zo!@=$6PKzw_ODxeWMYg<
znN2^M(Bbqd=c4?Dq%{V}XnZLwdV9LsA5rgjQc#J$b9SjZ7SFC*1Wy=6+%uUnJ~Ut!
zF_qi4x8CjFD+wP{jTxv-Css%692orJpYWrs|Fg(gbwu4)MoAjCU8|!BUrFgUQz{1z
ze2?H_3`OLxFz?leE5#CZGapOS)71q{oE&E~=|hSfHf*K~fb=uku*Q45`i*#hfm7}C
zYTsjo925kw79EXIhh^u%OLhz!64?BH;qCo5j&i+7G`)V;U`5m?HB<>okK_zOP;;@$
z^>GNFgwW>VGO=1kf|inn?x%$`qWLA9sb_H}c}l9>@vh3M_5Q)#gHuD?z1<)n=46O1
zb{N=b?k;Cydqn=ItCV$tb<is7wTH4tlX?Y&xgWBEa0fZ*AFFZOMm+FPBRzG;tSXJM
z*~;Mw*4&&)fAMw(U+K857O9>z_uoD`xyZK%j<B|P?@sEs40*UPIZ6hMDUdpJfU2|{
zI%ttzI4J5cTf}eI_d&lE!W;NU?T27D0DYmPn!z^*Wb$Q@LNpmALG{)e#H)rxLeB)p
zg`a%cQ7<PnRI)I(Xbl|d>`Rky(#X-cq1v^$Km1d&Hi>twbHXOR$|j`Pq4yO9)hd$`
z?N^%21-0%cyCR6a_@%_58S2Z(BrBcb<aA6F1==LjJCjHB4*6}x`pX<Soue?0?9RPf
zMygzxxJ`MR_1nIX(hrhG`-WJJW#LbTKZCC)Yk`XK0ERCzK+7_MbwlPMf^-;Yv2&m{
zf+0zCU3+}Tzb}C1`yOa8c|hZ~2;FLcJ0XD1H)yZ8_T%u+pJGt<G@06L&scNH*ixk~
z-LnhUs7q?zvtb*vUw=d@9U9Ji^(tPsTNjQ}S1_x5m$qes#u+n?zj(KbMJ#LZLpL1j
zEb9-CCGkxmSM=@Ar}!F|VJpS;-Gf!-QrYZ5u>M3iJZD9xa(15s1W~Kxx@m_7v38Qd
z^oCgaPwga{rjD~1HpNFAu+RuaoE^+8naF<BMV?~_joDNQiPi*4dG%fQrQpx|vB`On
z9gpu3O5TIWrw!^DNKXQarrRzawCB%&S`J~ObQPU|h7M7rfad!`J%}HmA^a8exuvXu
zP#F4WJoZqzKW+57_J+%f7Ug~OY|_PzI=gUebHgC1%i}QVcuV;8O7vaJ;IdD|WeQ^|
zayD7M!H317%{gl{MtuXJM6?r!YjGDIzEHM0RASbqu#nBRyF_Ha*V@V(S^;@C{##hL
zhrb!J95bWPSw64H<IB@XdEFzZ&J-d43%4FLf*~Y)J)pP0-UTH*BbX8i0>Q8LN$`yj
z!5T=c_(8Cb^j#w@9SBeL$3JaDPz?rL0wL6JfV~#K=qifzy39*=shOD_b>CFUEuETO
zW$atFbF!54?)(wiHrbLCt8XGRkoJg0&16dJaMZh7gfxibE4pbgQ8Z`AmsQmy4_5X0
z?>^}9r>jJZSOU^zU<=A}JuCDXk_cFc2^ZI$;+qyb{%fcYGMMQ745QcjWXkET-B&ux
zl#a+tM)bozCO&+`&pY@OFDd>?y5tI245Ygmc?QG3`e<hm+7Ae?y8AyzPlIGn3uG2B
z`3aPo&{VAlL%e)<p|gw>Bp|TAu<`KVkYg%$Kj8$BLQ4bbpA1PoHjA76Rq^KLSNB}U
zzwQLw<5`xCf0*%F0qaIfDubvGrX49CcIp%INoic0K1_2n=eylGJM2ekI;zhCJDKkp
z+=o=Cc$#F2baV9kP>v7@{T}S3LIJLY(_d2*ksm(1gk}#w-lc=$h6jdd1tMeWKz<HL
zazg0*M{qEp3B2?X{_tJ9fQAMrHw;iY0dz40RDlMn>;K+4@kv1K>Hp}O7;w5_%F00H
z?DK85jAPI6qV>K^xyLZkdil2lJWJ}KS5$+BM0}6a4>VpIx9SHl{Oiv&af`NK%c@yy
zo1j+xnQso1<9Sm-_p1AI{=LfCX?^??`e_NDXrX`#`~uQJ4_fBm&}E3=c(e?nN~qca
zP7d7@$b@2Dbr9@BYiwuHIMg)-PWmFh94)lXG7On%yfz8$v@zIV2{HGdI_(xHb;IR9
zqz+-;kiO(V(ngkJei}35a_PSYTb+-$x*bj1=K3?aSm@fqb@Y^;1(&M(VvF27H9C`2
z2LCF6Mr+5wL>Xw|dZFGJLk^?Kg|`N=NWv0iLmfa}22doJGgTXALKhJGy)meAjhbO9
z#Q-b>zq$4cIiqW<d^|nF`MtlV6m)+2fLWl#3Fy3&_7jD2ZcZ}w@gA)w3W9J4DHaTd
z2m_%|u~q{b=wz=sT*3m?KGOUTGF%j-4-^burec2Lr>l-22>`^zP}mg-P)kCk8|g*>
zNf9y-9F!MeFK#Qx&x{omm3&lY(=;i`tAEGWRAKcbpa2*0Rd#P<bGeOnim*y!br{=a
z5KDCf4Dl!c^FY(zjR<_e&}Ko$3k;HML6C=_&`h~q2RLM)vBEfQOppY%!JH8IV3>*%
zn$Cx`tssgq(2jz_s2>2N5L9?)XRq3XN^+Cjpc}4cF1T!`xivS{`s9?yCc4-q-)8fh
z_)2NMz=VCGl1j869|qJBlb?!wEmH@`Gw7Md@_rN>hW43)b_R50tw^sajU|Zve3W^l
z-@MP#l8lzND|g@IMR&hWel*D*a!S}U<~CwH-tv}+wll>&+>rb#yd!y#YnQ)FHb0A@
zbL~jjji%$i^6z1{;yQnl&=uJ)U^hw#EV66JArEC)MC|Ivo4s#D(i-zW5l608zOp#G
z9%6oDu_0HW)ZOiNNd=)Gi*69wlHHzb8PkUuK4bCIm~%)QcYx{qPp^Zr&riKVG?c!#
zEBf9{=Rb}0K^a^yxcDXa<!<_V{3v!uN5Z?*w3Wbe1!~))<h-V0tpMv79WfMY8JvLw
z!yk3=d{$nq>W^iM1)aX?`N5SZLKy;lU|op_MQT~TZ-J=?gJjvrAY=%c(qUW!p!<80
zvZ#<f*N{?<c1&b5w}0i{xQN~c7CuMKA0k?3g@&#=F=<8VE7B<G+s5Pqq(ucXy4w1(
zdTk>nB_DoJvX?8x4SJ`g6CgVwbY|-Y<@5Q<dG@l6%R*<@7Tx90Jm@%m48`$Pb~5>6
zF!u!2<w=`m!|Yw*zE;R&54?SOJSile^`jQYzywe6`TFKSo6u;@0-ArJ;D`#Hw<9M~
zA@y#@f2?{Hd55}X@!z={0m)C>Pl<C9ox^lh$zF;M;o0H&+&p}URHXeV;N%o-V>1~r
z=XdLL;`k`uXMIriI48#5PgKYvc}X9WoHT&zya>&W7fLO$6F(TXmp{_{b1dNC57n;&
zUT;lap4Q-R=z{7zqylv3b??nVIRom?to9jaLFvqHC5*QfOp?!nb(<E?3U=d|n#U!8
zN`OgP1Wg>US}32`h_YPgbe2u2K!1By-gsDNd2_GVyisz3^rN{Tazb_v+pdS-PEL(L
zM)}?JX;sh_{&70A%#_9uX;S%dgerdADRM{_%i4Kg-j-7~{)wtzxDXUW;H7zDms~1C
zvP7f%?5s8m9&kMCI!&T0MwX!DM3kZVRq19J@eQN=m}ciSFZZtWcwZE<o(p}<T^`xK
zkY<=E)$M<q=h!NDPHIs8R<d09`RDh(kNF;?LN7OUX26=_E<_LOO|9<0)2=wQXw_p;
z*fC57dGX-8zc2Pl3+Cw7*lnqcM1RYwF}>BNs>Xg(aH&vraCJ(sW)He#Wbo}+d?@jN
z-~<<9lSZZc{OzaHoezU3`<)@MbUBU;*-d%5+Yc{DGrs$Bvlpq(f=11|D)hkw#W01H
zBQ3LjuCSl*eEwO2;l!^LwJOKD%xns8VFc#|^;Kv|kWPEX;fmJz%6T75^o_-yAxHN(
z540psedC`sm=YGHg6iU=Vi3Y0?;8)c)cG#8bWIEwvK)V%5_>0AHq{xf%OLJb%`dN#
z8nK780^0A+-@U~**o&NgC%4s9mUtGLz?^FumnTM)PNttptOp51Kf)EhDot>{B&2_p
zzb~7em$q}%H&gGFE1)1}2%gg9Kh$p326NCQ(EqoH3|BzgOGLMG`Vre4U9JtD^slvt
zU<jhF+nkXeO^zn#CEg@Q>g@1B7A7V<7FEY@9tN_=Dj9DK%KS#)p6C{Ix2qT1fxaR$
zn^M640o!ZD&S2Bzb{lVr%<uT8dQ>&GJZG!0J-DrXHlD8WM`UnS3PaTNPz9Ha@BASM
z_(R_Q<ZNxiMd`i%PxRj#O!0H~GyQM^KHj+{z~LzJ<rVFwRMfIbR-k^KIk`b;O=O&3
znvb%=utCWW7{8F|M7{kbZ(#<bZ?O5neatsk0+K!B7xqu5xwb`S2niBhYLiyCDnG=<
zWK5R17#wull%t1o69;{Y;B0#Q5u3Uz_~j|bg{v9Anm;w05mbaR_t6%jJ|rVkH(8hs
zc$V**a(4CZ)WWo1cq5~khT}!ipW^%rpSGjuLNhgHYhRI(_0x(Ee0%iu0x#)0`$3oS
zFAW>CMG1WE-Hbm|xHBQ`4W~vk^JdG8{wsG`Jp}|(U06#md-MJZF5Pz_a}vMo{v+Hc
z0=;*7@YA-@nJ0cF?=Vs6b!O3MVL8hJ<SO);txBKwYNK9p%DhxcFWKUMa)^A&oga3W
zwcTn+Bct-dGTo;*kav!M(Vg&QswISA$Aq0*g+*X2QsiN7XMM~W=L2)ed-&mxvfD0)
z`T4ycA}e6<tsNu8;v>Ge_+l@UhCyqf`hn7E!@YRnI-}(+KN{cb_p!05#F>nY9M0Sv
zsY{vEE{JHGTGq$x@E<?1!D4lFyXkXC9!2*qv*-L)WowyQUQj$^<Yq}@dPBBxfqBYt
zOPMCI@+y}?YVw!V#(r{~AwuNa>&eGY-1+7rchmPI7uZ<lL7kWn_VW0A1ZC}~riKTe
zC=k)AlS$|5UU(@WpwQ7&MVjbHel4_i5IZ=Q>(iVJhr6=ttx3trV7OzO{Ma%?(XUC#
z$g<OniCno@eH{KgV&9oGdwAlAc8NpqV0q)u^xg4`JiZ4_H9mo9T@OE7o1ysowXckg
z5`6olD0<P}JY=qf!dlDb5hGTDP~-H&i@U%1M_%Vn|60(3=VfvY3l%zzPDs!xxONC;
zLb0v)o9L4X1w5bok0~+hQ-pi;JQXBV41dIO{-BQTvSA7BxbR-|(hPjcq^bBmT0v)I
z#g-=Ehkr^{&(k1ozObEO<i7S@A77lj;^A6@dlZD>D~gM*Mn~CFSI6P4zd`Zf&%q_p
zXd&AX#~a8i#f>gsf342l_h|6KpR)*m9#hGgUmc=K45swQjXdAn3P{Ge<7w0%_1)D*
zZkR2x*_!#ctR(t3WHn(sx|}!24IZqGaL2QUh2HQm^6xou>#HB@>DR}ZxP1=|B<PhF
zB=Yu_Sc>i@#*X9VzKN?>q8dHh5Tl7NI{!2gx`*PN^~j1@qAM@6YN(HH`aOCe#k~3I
zg|YM{oaj4k`x_6Zn@_pnb-P68_iExo(%$-SixGmirStcf+ica^X<K_Wk9}{O6Mt$8
zLvj3KVzi9opV!0#OF>}2T}$_AyBF>642Q}G$w^pdX4l!>cF>}lH(JvV-&$3C9E4>f
z-*Em~mgb**w<pjKKb>6A7_DHu+vc!1(;mX~a>C~Jw3`YkcPG024b~{S2tOBi!hR;P
z*;NmcTfel0v2ivO2NP3w294G{x)7NDtA2WoL+xcOO$hQJq-i(J)_fbza9`KzKA-yj
z1vP*!YMSrr4PD`53%$Fv1-4FlEsj?{`1xhtHjce~Cei(^QdjoQXuQu{31NG0b&x#C
z8Rle&I5x*9RxnQjt@<P7^EEDDSKr#>`8!j=2!Gjsf=?e=*zNEPWYK#`=RVzD+A=a_
z21rW39NXVFzrj&CI@^$Ofx%To@xb2dZ676j0moIaNA<NHN&Na8_~Gf_&o^#68j%a<
zY@hWf$K%K+<z=?djIA$~(u%gsIM|b#Zz#iKC<#Em*SkwjWu|YmJyzfF36yz7Wb<Hd
zixjMrpcm>4Ys?wkQmN#YxkdaG`s1!L5inoe|J~T`ur9PH<fd(E_*7y=kS6Y3Q1E2x
zRR7^)*x|RD0-pk~X9eTW<<ceAg6rS+cN%--m)R?~uZ6623V#2|?aeb2>{rWMU|G*M
zVY^{ljQ$(gD{nzF3XRI1&`nQDMYWv1e+JD{2+0+C{$P|mG13|y`B<Ls*~Bs{Jh7EG
zqO+GB=f<?IB@7a3pax4RTQnz@P!w8qN~9j|GY7#lTYpDKR1Y#qWB?lAahia$fDH^R
zhyi9ml{)m(V{@hjnU#c%kN*~0LS{A3j_@J8_+3_avX)mSEFd5^5_?##^DTRN>5Lz}
z+=)FowcwiS?Yk}E4iRBaOVw#z5AF^KUO4CzBGCqDHT01&Ll4U4e6(g`M+XMzBy;yp
zL0JhfN*<(5@Zji(jG7t~J}JNG?S=&M0eTp*>7T?up!>GjJ@5U5YGqZ_QZXQ?bh6BQ
zyt98o*-lkt>!133_f+6bQ4u3tGdeV6FoqIdyUe}S{s6i_gi1mIHPQ^N9)MKzhM<Ft
zhvzN;YnWgH)`oK&ij;~<K!3l?_F7{jpoJo%qf@G@j}u=FUy==W&fK-n`hcND$0N;v
z`_X_7kMR=Ul0R$0c%j1tRjOHcLx<O*)yg*x_7dBt(<d!f<}-TBIHm&Ty^)DY7g*AM
zvUvjLGBYPf0Qya`gV-#8-hpWVS$ho7U-Z=7o!5Eq7Y6n-LBUHf>haPx=~UI&BNV{h
z&90SA)>^niENk_uzc<hV@*z=lpIZO>g^)9HpmjT&lsgWhD!y8ivhBQwF%aHYAWleh
z6!omQ(DpS;O6kI9=)zVg&8FC`MXe@=B*S^T2ZznVM~KUSDbWJ3;3t67fhLQeyY>{~
zvED$58gi{FD7Xno6~BOhH`&>y;_o%GJG;6Ht!H0A!za4Ey@~5YSU?QQLO(7Z6ROFA
z3<EBT-aX#89`wi8=Y3cyxff7$Yadtg{3J3bi{&4H6W2k}#3fHm=y#5CjU@`?sW1Wc
zJGZc)K-k;e?URvV2u~+CSP)2zr@-F1fWm3J<e8jI#tpNE9anS9)=lD2kZIHC=x<kB
z*M9x7-8D!qbGCf@-!FkS!Xb<+R`B+N`V?IIGe?~0-e?^-dD#yFA3NW;?rBKOCK~NH
zVq%&n7kgGEzd?@%TvH)vphH;q+aLOjnju0&x<yS$WYyHJd3`crCXUPg(<C^TSBk7}
zO=F^)qa_?KYHj!`Z<t+IMJJw1q-@4w+SfW6L6`oZsk4lem6)~ff`U4Qz<`;0xxJvd
zECqV5kj_RaIXQ0tnq5GF4j&)rdAbkNqEVh$O+9smfjL*Ou-Hh@U>0!;<cqq4r*e@G
zOl7m*V@1r_MX(8ZL(A;dJm^^+hVN?Wu|Cz*^kvo2-??MAnUQ1?Lzj3-l4tkU*4EY&
zpbSD@_cA0CfV7%$NEQMK8~L$WHkf^46zdCf0T8h1@87=>CG;jV8voow?88pyST|}-
zir2SIEZRbIW$L1kTT>&~Gprn4ZXh1h8p-cB4e38gE~cua242320{lf7ZzH?50S$CA
z^QFM2fPUqh`1lSZKftBxuS%ySp4r6EW_S91r}Qi=;0;sD=+OzoaVh!Lch8TfacA;=
zToqjUxs;ALV98t=h<m~8<6wn*=g5*Z+@1S1{DRW<A*v%f%Zs@%6lU{fon^+X5bHB~
zFur&Ng$y5`=SLk3eEEa@GXw8hDG?(UzE($-@aXZH_APaqn%x&K2@y}kFUQX-;x5_S
z16_gfliz_xm$9e5mEM8l7q^2i>8X=)TAFC;-vhWEvjXhZ<Tfe6vC75R!^-63FE~{S
zN%b@33G9`t*P1E8P@7<%+w{lwNgfZ5H+LnLm>v<(iQj>9^w!zw!p*3fq@Z83_%!PG
zhHx}4@`JUSXZG@~r%N=#U3q<S*3q13M`67;N1pSc1m7H-ixXi!W|Q=!jp_{5b8k!l
z1x3|`*mh+x@Z(GRhbJFGzzceL#)3tNzSRfLl<mE#z{@?yL@ao}Z$1Gky8QH<NvOty
zAJbVK>V2Qxa0wn?`opHLGyVSV=vk7O+_|4O@c?lN1>|V=Z_^R5^Ii)L#uCJ(H++9v
z{=RjGwd<J#PXGS=GjIHWCA@R*qEGzri^E%os#av7RVif5@>?(ryB`v(Tx8`B<-1&;
zBL7@8`lZ8PA7Cl!8tdgR9v?5A?npQY>}<ad&PB}#Rj&^}m>6Z=xJid^MQY45eGfSX
zE6w*pN|$)sv0xcB_CzkT*K*+t*(W|*`B{8rV*qnnnzz8FpwKqP_Ubs|W(hca9h$(F
z*8Q&APMk0`)_NUJ|HUtjh}S;g!~WhwA_&`>R>W=cTaPs~WxIBobY<r*G1^G_bt%`v
zYiG(<7e!|s9;~)IfuqM(|GXaF(N>aK+{+`wz-Xe7>V{bl|JL_wbA}qNpC$Ji%5-G~
zZtOkRl{AEyU_|Tu0$3XVcMis<ljpSGcKe2nrW$=xS1<g%zbSXre{D_;yk7d2eF<6a
zL6bexW!=i`)zIcd^b}P@B)-tm5U25;vCp@+HKuh&kDJb~-b4^Sl$(?8R*X=O?vEn5
z_^!&Ju0>0S3W5uCnpA?u!h*TOM!BOX040;o-Le?D*mOF4X|Xs^O4)if|6O)wX8oS|
zkDFsD=;Px~>ssc2FTJEk>)W?4X<g<MW!2Sp_NZPkZyN(gH<3w+xcu5RSUB$@&wW_|
zJ^k}{#{S2r5>PmgN&ese-|qOod2V8(Jk{3z<ny?KXMTR3l8){L{EKliH&f@4XT^>{
zi<oC@h1}PRnK9X=Ofjbxai<}wur&9lpag@00{Hi<im+b<99H1p(27%ROL+T}KTBEN
ze!CGf85b8Zv~hT+8%NPCMqysPI=!|Q{Pn9Cqz6?h<B2}~hfnao0-c!`fznEFZP<zR
z&OXY|3tkPLK*v$34B}c?S^4}m&fEXL9~;Ptp7;H$UarfjSWfX8@E^#Ebj}TkU(Pk+
zr;lf*raS?}1OHR1d{30}zkZ}>i`edUdNc@>$ch;MuHgTJn@g?pv)T#p&jq%GEO}qP
zfG<8+4jse|M$B-TL~RxQSUr*Mne2+5mD_0)W+-;+(B|qJ8j6LCI|KJdG+I^{ff_^6
zoRpK32v8jS1_qB%mX59?qdI)SJ4EK;=;`Y#!(D3npWaOf-i?}#4IhY?jk1zFKEQKN
z{`e6Ez~E7I*)LoHL_@%8Mo{UnaRa*U5o!<&)MWQK<-T_L!cM8}JoJ(O{`MSe8d}9#
zZF+g$Ha7CZjEv5H{h=#x#a4Mc&r2%M1{3KgaQf4r;YWm7JOcWCV%h*6@fhaC-A?H<
zYUo@Sx3(@65EK*!5~`CKI7P6?Nuf&!2fI(cc5CaEe_$Z4_`pLhE>!^m>c@}g!F!XD
zlF}NRS;t-!;ggHYhFj1R8S*|VikeH)x%IMChKQ&ru+jw}?@a-37XipvV9OW7G6Cdi
zH5s54$zJUWrc>bI;X$MD1ArgKIOSAVt3j(fd2D(I6m<mzG%H}-6*ag#TW~HSLz^B_
zQoe-oUk?FE6M<mqk9{-C*zDk2$|)#NvnfJbu&hDeD;t|`zHDa2kp&s!TDgYHQs^*Y
z1t~ki2ZrW#K&z<W05c5gsX>sIMbJAO!(o#S-9pF|sI<-+b0`TRt+?&rXr~qwNC^l4
z_%_w!>=YU=A3(<>!z2)eq~Q7<!RC#k-Y->)TLupn-n{!yaW5}Z?+{=rNP%S<0XzqB
zpq1-^bA3DSXgEAP9Quhh6!JBYDKvQ4?|%I<Q|}!riGj)*8JXff&9Yk_G|N-KM2aF9
z7O<_Pp^=!k__<Q-MT+#kO^JTjZR4?`G~lDE!>tt*II|&_VTq>l!rV?HV)firdiLx!
z_x`v|vT-cXk~IYK&TN^1EeylEoFWS$il~5Ummc`5b_%Nr?Y+GSpi~sx@l>d0^~30t
z!!FHJGO!h7WXRXV;@OdkZ2AEo4EzQ<U`iFa{jnE~bbrp)sb?};TvVh7b&3VajYohs
zgOGy^8rY&J|3rbiGY3rmTWoA>%ae6=5HTQAq25PE`up)BUj%O0f|zr{;|Lgy*T*al
z(MEO<G)Z<<#yyT8*6-};8J<^5;IMgd&*c|`@wbn!Vd_p{C%OyvNzYL3t7W@dxt74E
zD_Pam+z3qWrKDsV;B5+_zZnTFAQCAm>UXV!mbqEDhKnfM{tZx7@bdduUOq}zGARbs
z7ob|-Na^zfYS<J|O&h`QLP!_D)4NhXBt}R`s05^ROiWA^6nHb}*Q2DTf18kyK>k2&
z(IDnFy~Z^ZV1Ece%QF<-FHD`~&#nHf;VLRF)_yWr7d-<X+y_IB0Az)*UXiy#hHj{p
zTcIL_ZWJx3E>^~76YEn`Qr5##MnQqu7fI_2D=N?)--@(*gKu5E-NUu82}(JjJjk%w
z*yXM;`d^7<v&^fi5P?QYxBZE?5SJkXLJyWwLbY^tW#@U}nGOuCVwTsa0c4{xjy`hp
z%${d{dKwiN8At;s(msl?GGJyVOk&251ftTRq4)p_%JTBE6^tc-CjQmZr5{#DJI26r
z(S|m7sM32w?>Q`SU)a?)c>r<%{239r%?Oj$)cqH<(!pnY!3$JAADc|jL>Nl|Npfm^
z-NHS{7+A5h0OJKz=vzUPx)^Y*r(z5fks$`a8bOCiC9o_S+uGWWFRyQGghG@0vO%6j
z)H3w%T}C0eb{QD6&Bp{hU-3}3(Vqi(_2BSO8*Uv!y@P#-f&zmEkV!2t=^-Vs0?i^(
zcz_vx3;3sTe8(%=c`zj+6#D-FOK=D5hElC72SBufVMk&^hi5Z1vI4x;4<@-yROflB
zE>>DiqnVkT_mx}gZtLzm(G)5Eb>rpdkmb3bGc!8dqhPq{U=2mUTJZN%D>JRh%zO#!
zLl1iHabTP*g!IfL{huX^ii#fc@JzV=uzgtt#WF^ZQ`hYSxJP;b!-URFDu^k=03Ia;
zM9T+&M*}+$&YAzmz=ACTx?}wThuRJ{iJX!$vvbIOx4YylAMhcMp}R90j>t|xdLf&Q
zEIGddr@{##h^aeh)_jfhYX|-sg0~8fjirW790@pxPR-LTd|^pRWZsCVpr8-1wwDa@
z1T5m0p%yIPgE82o79(K;9j7Tqv5578NnNHfPqg2_)NtA3lM{rnMfQp%R6|3tEW#A9
z5Ayd?-X6-_C9Zd#+gG$B1n!=avvVbklajgTIE8AXYiPm@#&&)-l#ewD2p<byD`}5!
zTiBMwW}WWr*cy+Q7vAA^loB{S>rRC`hm6i6grRA`2NnmDznVVv!zv?24DOyD+?jru
zZw(OgNVri2!=^R3Y@Lzu@iai}Mh55|0!|J=uEN28ANrGP{rrHdBkf>gmm&rSEe`fO
zn0B>dlGoYUnOa`1gj`%lXD9p^kT8J_S~tZ9<^Z)*<)>zzxx=Vf*0}>Y4iRJHEIk8*
z6@xsn4Q%8IL3*VNqOtB-m`H~W%T&2Y4<Z)Wzn^~gM7Ad^GA86FzI!K~zu2WxfSU6b
z(4zOrMmFYE;bnj)P%!>2Mqde>m`(s1KZ21|c~$%yVN}ss4NS^kjNSn}LsVM2?|L-K
z!HMolU{wb*EE440SzqYfK_8t1`u=+;j;rr0@pQ}E#Q?17E2MGuR42Yhwhv`50nv;~
zP*89SismojP;PE&O8M@DlLbe>t3NS2tcv*b8qAtswVgx3)r&&CaJd~u)#tgGxg|V)
zSJjnv{tTa91xixd&gzIHlF|@<`7#Xb4R*y58A7#X=i*XSzB8~x#_4fd4OUkgj;%R}
zM;QS<ylt_4RAApfvEU4jh&~KoSlEP2MLY0~G&4#jUy2(5mEoPA+T^HeW5U8w4diPG
zVb0(+f+tVmm|AU#)nke2us!nm`RW89axlf79)M`bi4Sm=@aAR#064t!IG!aTOE!3f
zeKKDewd^(vXZ6hDA_W{zpBXKTSrSB3-M>bc%nIJOLmW^bIs!0?22PK6adClQECDQp
zN={CW7&FKn{Qk<+3&FtEg|--p_vQfQx_phm6Z%wuLVFD%!62v3cpJ_-%Lv$sO)#T6
zV6LxVVm1Q3Zwij&Pt1x#FCGD48>M``v%J!_tis(A5<cJ=)`NG24$#LyjT%Mbm^*jx
zqCRD-5K|0@m1quyL;)-yWFilg@9~h)NkGCNB@;0~+hI~1;z1YRAcl4I4Ep#HdKdL<
zl#FTxN>>m535lq~$Ki);Y>GgvLzvAAg@})a-2Z%!lWadfCx&k;F#oCD3yYT$a9ImB
z`72&79mw8*@t<K}nxXUk9nd0cBur}Y@R<C^)#(N@4ge0?4tlFOe~QGLFg(E7CNC=>
zMqr&g+Bh`8SbEoVC_4X#3`0bF4G%gX^yw#Ew}ZJ)4`ydpR;WS6lULPf`D{qdHZnJj
zSZXMSmmKw}u`v~F)Xy{wn=q0j1-w%P697b2FF<G$1h<;--34U<7Wie)?d@4%qO*!2
z;b&Rd3b2-1rlx`5|7*cHbGtiD=h4(pn;71fBuI(yogO>@%oI$=__SK_IW)%8rAxn4
zVFHH~NF|>mj<UhA|G1ForL3&11axual~GaYtBd!gJZz*{XeM5NHUZ;|%pQf+{3J9j
zHT9*DQ6}Op%lCM&+b)G2Ph)w6@trJ7E)*UgyArZl#5;}QV`puZJdKi(t%#yi2V@%P
zT*H7?TX|LLI^4WLHM#*jf7ofSfME4sKRf;-=*C_r(jrF2a(1WMb~zlzp60EijeCBs
zSZW<b87_c?P6rHDq5(z`GP_L80cYbuC^s2A17M)s*@l7bF^Jc;UU%lc?feS(qW#5&
zlED1bOc?ffz><&+P!>BC4?DT950e*cq~LRq<XyxSn`K)32_oWpqA%e3fxDFGID^sA
z-JQF~3s?c2Cj@%1TIBbZXw@s_s;jFX($n{H@BLw!qY-RTqc*^iW{QR<KKK>xIQBxn
zDqSTHH+sS4nA*!Q-NfAp)LP)s0HntLBfw(_K%WD_(l-E$!LWy0umXJFzC{ND06Gd1
zJHnaubu##hTFp@|uM`8kEeeYbq&%gNPk?kIytjDN`cwIB2tWeTjd2JGset;9@LFJA
zZUk7h4oGJT%&k=5fMW3-Of=~PY6Tsf7Vp3}Xop#^%9g`!M?84&p27m%|Kr*}h+0-G
z4UeMyOqrx#QB_f)Mt&Vo<}dB-Oa0Vp9Fb{H{V>L5tkgUX{w)RhI}IR_(}VxK1?R<x
z-7;I64jnwz%ovCW|NBklVD)5q$igyoybYn29ssY&w$4n+($dpKfZnJNt6Kh^BE$@q
zc^26r0T3uZOojyWf4$Gwv@$i@KjpXf%~fFR1+Y8=VH(LSI7UpNn$?cjaI+f0`bsG%
zTt>Na<;qitNl@#nMQQ-d18@1&lr*N5nAu3H53f%ic{WdxFN}8nm^>>`+~$*Go0O}4
zz{%4FxThnmWo_6#;D3CA0OSLo2QNxJ$aY{D$x|Rd0fXLgVpg2~J&-#;0o8wLAOj1f
z%cCU}xn>CN;amTW^Pfcjd?ihrxvC!GTnt)L0E}isAR5yK00uIk>n!+qh-(I!>o60w
z7%_K()Q6mu^cg&T3ZTkt!O1kJ{~zz9x+qpf=0YKAwGTy-s|-x?r6E&PWW~ouSm)qh
zK-xA6R#JeUngh8Uj1NYlEd*T$qg`xECiRVlsgYIgha@8YxxeR^**!}-WeQVh@Bl+7
z1U`%t4a~O##0Lrr(C|^}0gB!Sv_0SiUjpFfMJu~~M+mbuJWlLjC~TwF6D^>QC>82l
zg267TYYxi8c*wH)XQ`gAS}nSo`D9L=<r`mk+q=3V!2k4zHA_T9jf}26ql4+fz#^W5
zgZ4ed)gs`Dr2+n-2=dXuEsBJ=W!nuG7Z(`Fk`VVBy}mUfi)@I0_8Ia4!rm1-QGqK_
zLu{T9P9gKqmw%MygX+x>9B(KgC?0mk!LfG;v)*XH1%qfy_jDx=JW^1MzKmQ05!$fv
z30v}iT)0C#*_J4R0x)`o3P#`D#(oE@P8GRa#Qv|lr=_JWB!R373A+KFK}V*j`F=O}
zKQ3UovQd=IU5X}WC#Mc5URm;D?0uh%e*gX@*eb{{2*F`A59hy~%4)*-j{l#(pukRK
z;L2^{deaZQw6w}a0^maH!x1U(o)j5Lj%;kmvLnHz9#p8JD7QnhKKyF~{(at!YR+AP
zcF%DS>*i~+0ziZY4u~#Do)NnQK_pPNUPHKgV{B~fU{&+XWEp28;NR9X{yPlf`pK-!
zOi##m!`8Y9wJ~H;DcD!QEHwl978?f#DggiKSNrT}sxB0{ulA<}!b$~1RS}TFkqnO9
zxxGApj(<&&HXRvu;svE#P{+djJZuUI3S_z+t-3_8^uIrNzq}SpG|=P*Ux%Ib2F@G^
zdRqJE2MPXt5hj~%-a%5aWRL&WhX22$`=o33e--!YfWR<{&dcv5VQCbdr(aH7x-r{{
z8%)Ol-RO1EeC0EE%?$WMi2EnFdY|kw24vFOlOdS;%180=p%w&uFwZvA=_||$1%Oo)
zWw75%W$ue&0=HonnDOc2!op?ofzLq|<>hJ)SIU8chy;`xPV;!o7UKkP=(q8m9ZKH9
z!_x-W&}g)P6aon#ihZX(%LPDvGMFtZYwKr_0fM<j!Ej5qH==1(0m{x<-`9r?2F7vj
zoe)xq$&$^`fkRgb2t<$?PB%uzU7#Si?58l}4DfH?U<wV?!;-73$E&;H#fUoxW>OFc
z`%{2sg}DeAFopXAOyGc71bqOcuKu%LC2umsd|lIZ^Aem3Sb$ImvEgMvkYyiY+$YO!
zhNQ2pF?=S%W`vmR324=;4`u=hIBh9GZeIxFTOc8ZB=cHYT2MS8e*6fOX($W{y?Vt6
zzT?~RakXnOw6=bT=N(i<+mPHF{H+TgK)~(zMBrk6%^ndFM7y6Hky22g1A@yhDCi;z
zs6lVS{b&M}E+POu220Qk#5=XLVFawDUUeICM@>Lzd+Twss_6)|sCSW(J}^M%65?{f
zyvHC2`sAHJTTO$kG9dKn!jLA%qvZ^kfjP~2P$w$TiA$F3JoA4wcdjo@hG7_gRaUXs
zU=0-wXD)PL83lRj0Fhf^QZy(<0!fE`5C?u(;$(DCI{3jsMIUB?bbw$R>PpaD;jBC&
zR@7!=wOD}|Cd{p<NnpQc5JZ1L`?QybvAuim=XtLCx}W>H*jZJjr9(>-z1D>+>!&W=
z>`-p|V&0eFqX&srR#xU*mBlh>-LBIemZnG_DUpzfZAwo++30mR9H|Ec`$p=alCG2Z
zn9Za8kwZc+8sF0!iB^chvhwmuh-IrA1cQ&yHXgu&hYkLS&GsZwB3ZFWa5j6XX~t0A
z3_o0_y?ZLc!JYe6;EvsE%_dVC(taqqkL?S9;v^p%X^Movrf45_?eE=LRHQH(jcu6z
zzQ(xY+j1zA%b%Zf*RH}UejKzzYuUJOtRtV33^SHeQ!`)q0|7(>%aQBt8NQ)&?0Dns
z30Ea6dB685KlE0hLT}kCP9j*Q=jQrm%p$rSeLG6ByQK>_7@A0Y-$CTuJdbzAB(Qq}
zWl_X^Rq!7MwEu*%^`JSF!G1AZuRsgR2MZ+(V1I#(g)r3$CTVxv?yvTy;%-M%zSe0L
z6Rr?KeM^463^EO;*Z$bcU@*()FK|Nz3g77!FwwJcSc4I+B%vL8541f~g07<yk!|6d
zu!u1y4g@VQY}qa=IA#8WQZ9FTmetq3H&7(0u7_<9tS}^(9pNU7SQSFa7p+HQZEY7Z
z6QKp6g3v`n>=J+1RsiGsmc;6+kVq7c*0FfpSXXC<sJ*@CtXy<u;BXKH)b(th=n*|A
z$mM4OXDV1*m@cB;XbS1HOXJKVcL#odJo(G$cQa`SZDkuKL$6>K1v@h`(m+rPn3<ZE
zwu{Oq=h-(Bx$A}VWd>(n4crjs;}>}w0lQ72E3Z<kI}!M%kuir9r?@(f&VDyjKtu4n
zAh(Gk43$?_HcacZK2W9hL}RPem2!^08E6=npOa}+ej39mXHo<BQ#zRE+Igj|RjKk9
t{5zJcC9)bPD?w8;$rDb0ee^#Vj*qsRA5YilWdjngO2aXI-@(%zzX4bL2EYIS

diff --git a/master/_images/sphx_glr_example_pandas_train_test_thumb.png b/master/_images/sphx_glr_example_pandas_train_test_thumb.png
index 8304e9fbe92e6cfed0a1350e2f58c3b3628fe540..f40be88ce2880c8f394c212cfc310b3d61ad2746 100644
GIT binary patch
literal 37306
zcmc$`cQn_1{64OdR7e>i5*4CUNM?$x>}-irX0l~QgUk>~b|JeY30X-<HkItXviHdN
zUa$M}Ip=rI@9*z9-}{{I`_6d3U+?GZF|NmTU5}Tas<JHQ9;Q7cBqWsba+j}?kZcRV
z9|m$Ve8o=xl?w^UvxoARFKIZ(OvF05tN+^aiFf1wd~oOISI@8h=uVw@^XS93hdc5f
zJbK_y!`ypW?c$}&`i681Bior%$$c$Dice%TzL*kkp6HzacU0ihqdk8@_l&p{Hw#p=
zOuO0Hcbj?yv3vE331!6|JK%TuDvw&dDQE6yO*#B+u=ygB?`8I9a`k)7kKmi^!l#oS
z{`ZZd5HfuESm?au|C1lk?|&W<aVxK*!Qz@yOZrfZ!i$qPf2C_`cH0Xv7cF<YC8eZj
z*9afm3i{FGXlr}6B}Jv`HlK^yvm`-w%E1F15)$!o;%;-z3a2*)K51oyr>UoCDR6wS
z>Jqs(@keb)JffyzxyMt*%4(|gLdjsy&Ag6C!Ml3l(~?=OZg`PT?9=`VFDVZRHs;&j
zTAJ=rQB&K0^XAQ$iX2zf)y<m{cBZG>^?%#FzP|pwu5PI2bm_6PXP<d_k$%26>EC8)
z((&o;=%Ooc;RIix%z*jGd%d?4odscYb9Vg$1Ln<1yJZshKQ1H5SDN_dNAuCL?SO=Y
zgoftgMBd#FL4DqXv7$~KXU<S7wFub_l5?uWB`L>>I86N6FRg|*d-L%lZDE%(K5}+;
zwh%LwA_tAPx3~ViHsh6l|FoNi97Kd2MyYjmb$@1P2QN)_xLx4k<uzHJ>Pk(&GXDE5
z&Vec_Ir~->m485hyqw%j(lvd3=7$d-($dq1Mn@~m57)^nD_7n)6A~Av-1XVfFDS@l
zYjfjycz9ES^j;<bvtHp={I-y7KdIyR?>ELZFOD5Q{_N$;(WXM{o(oKZmU3ohW^@}>
zK`a9+vjYNFoiCG>qOa=f*H(vdnztlVl9=Q~9+eGpes>3#G&nrm{MqqNLSkZe$0S}y
zt<>W_6^qE_Xu-Rw*Ye(t{4u=0NJqTQ^}>4&e}DS1Nj>SEo>tM)qNfk=#%cQd`;Y#9
zd&t<>m^j{*6=$aNRufJ6cUpPp=I71ZJ{&)L_UwqQ&lL;N=8L>HJ=VrJD+3wJgC#b<
zPnT}h=e8QuG$qQqEVY^H7CE0SDJd!XH&rNL-t1NR^w2H7NS$KWb2!?M<}FXu5@q^q
zglvSEFU!dAnbtGA-@h+j<y4?AHebu1l$4~3>;C%n>r27AOe8Xi&x(q~4m<a5BcovG
ztzK~uVHp}45?dQhL_81J6|Afb1iOsHI&qyi(f2m58}Vj8+Q>#m$@C4epqpbt*&sGg
zA?~)6grl|m^Zudt-h*~S;o70kpA*O8ys_9SK1>=C8u~R;H)m#fd8D`Oap%2`yMc$r
zqn<yftqf*WxpCvjwr$%MrhDR}tv>s4#yLNFylY=`s@hR@c6R5ff)VHS*(w*j<mk^d
znnUNU26GxN3}j@L7~yOzdrD$I-R<CWU$LP&EcV^0=f00Y7N@XqWL}R)^zXOVa-3(4
zx?RUlMV-6zr0m9x{%?M@QRf0)OZnKfD4l;65mDu}n`UJ+F*u#!`v>D1XS|7=oLujp
z>|0|0CUS^_^f~P6d%9%h(f99iT5t6#s#ra1Lys#s&)1#TExdPX@$avJr{}wGEKRgs
zmX@~dxSQ96@L)fB)X(34jzLaFhV1z9<2aOU(RDJ9$tW4bT?HgI7HA|kSB_d2FVbsY
zcchb-mnVX*u~C(epI=@<VW34t;y5pF5F@|I`<vxte}84jT3B%BnA9a8t`7FTd!euD
zMLjYfvR)<Zv7Vpb_vDZlwdBd`pEVR;p1DNM_HC-GumZp6RbjMWl6-cdMI~T$eq?XP
zFYeDy)4X^KZ~U}=zN)@cm*dwy8(#!vQ=yY}wo!#Y_Xrm+FMU^`(?Q&~^tsI>`LJ&T
z!4kYv`2$q5vsJ8n*}PB}`q##@4&tKwaR(0Mb-QM8ojge^EiIkkxwY9JZC!$Q?Q;<E
z#@k=Zv!W|@Tc$^}8&!uMPq<YTl<@w&Dz1?ytB9R!-{F(YR6gJ9>*>W^7l{1V6fle8
zz?a_ZoHSaU8|weTlSRuLp`aKirmGku)RYt3GDe%Bokfo%r+M?{P26&fatZ_AZQnly
z_9@hozVRMw@}0I-ERy7GQ>9xTw$sJSarbz`s*iCfMIB34jAXAh$r)|SNzuu*2sb?x
zQFEOBw86=#&*QvoUexiPn=au48c8h(&We$EPhKzT{(sY@2XTYK8wHd)?=N@#_jz%m
zUv}*CZMg7H5n1HhUF;T+lypo@Pmeq9{*vxlt4<*`EiDecJgZYLc(1#*7vJQI{Ps)R
z@DysqF)wPwg_OSIFU^F^OhxyZvRw&=r5?mRNR$os|MN#D+jY?b*Mi6`)XBGbEvKwZ
zfB)YS9V!#?agF}|Of#DJlbsSPdo@Go^y!duAqvkI$6Lc*yr?X4ng3o}d%GoB2}LoX
zWlWWJr?eW0SVL4Yo7a~iyKOcrGdJ5m-3>6vlFe#ud3RkjG&IyVFz|Ny6H11sGFAmc
z^i##l261jv!he6it@-}!Xu;awjHdC$+UWDa9oD6>Zc_!jY!2S$uD8bvQL$dAeZ23&
zsBl(ksf4hwuwtx8_{4-I1=$YN9-Wi)=m&S(KeC-XS({jZl7O~?rh~+<GRe^_*j!j}
zAl~Q4j~`-6hA4{VKKtyqHkRE!wTwA1i#dnX$BNR>(hj<0v9oZes8?*c7MENJ`{Pf?
z^YXm);mys>7$Mt7yw^XgYG`<MS*0MvG?%8k3eU58tgxk}rL7}4DOf~4`cE}CU&VWj
zHYM&Ni4}3giBsgX)#C214n-I;oVh_#U0tmh$>+WCZ#gV3u718jVqg9)C8B33wcuJ)
z>__TTx(<16zT-g!6zeI(N42%J{b=!P&jZ)Rzc&S=iA+WP!)xqJY4NKtN(((48nDa!
zNCcO});d4R|C`Lrz~0^)l2TH-r5@tAhOZX6uAge^>yuHNABcu!jVe;^-8+E8*8DWy
zp36h!vm>W%qS{L)h`D;EuUt3JibXwHaaKUH3&98MFpmqj8-JVU8kCsM&Bn&2@3A^m
zG+!^8ug|NYDV^2If9Lnf7-i1A2YHX3KOb?;s-S{jFV${dW2m;s&OD8eIH3Le_Y=Rt
zEM)V*kA^dWYfoQaAG3r<(T`7vqmfoa&$n|16~f$nZF~4LRx;URPtxxNN->=qtX>#v
z@fmAL35$&M!%0Yz&&|!5{dxa-Bp<IpR7sSWu&^+xEQ8|onAX<TgOAHDqUWLCA8u(K
zb6A*Y3k$xm=8Mj7fQClO!9jpiIp&C<;0u)NhzAcyRJO56NU+8#FQ0cJ_gr-u`uH}G
zdb?5!ITaNL3L&7=zQm@L;E`wRPS42fc7Jf1Hc*Zet6pE77r#5M6eV!8?9omNW+7@I
zq}oV+Kg9LWk8pk43Od9#n%qk7<9(w&r4smj;(1;74&|6OR9Xt+qsW(fY@`~NUSJWu
zN7@<r7wrk3iEZc3oup)B{UakjZI&EtY!9mry9E|becoGa(9rl=D!a|{&*Z%)l!q$&
zD*__L-2|T=I?sW2(^u|8!@^ScBb-OK#9bIwDmgWkTw2XWBv#Xcfy$?xUqF}q;V%pE
zQ#k?t{{0b#9(}~0lz0#J^XG9|Uwro;b^e={qN+*f>FEj3@<mhodIi$jc{tM4to0p>
zVDwYb>7p>eqxs=zYd+_h8yD7Rq#iwfylt6_!>4@PgR&d142oSXJM)QVj>@Kpvay?z
zawt9Hol$4LEzxn2o9{mV<u<x?t6;YB5Ru~OF%*udD+!`pT)wF1Q7C<JM8ap4todJh
zOx8C(9*~uF8j&xyvC!ha(&s}$;5M|nvyOi?))vRB2#^HWGTD(=Wt1KBf|ta@!^7*z
zleucn_%oJ&&KMaPiJ6Z&h-~rQoxCRsBqeyaotlQG96e46r!S|Vkch6fFy1<C*p(OB
z?!#Q^w)^VUt6vfF+HY=sLV6Qn&gK90kaf`|1B1gj24J$LA{Tq0lm>|{F&7t?OvBPR
z;2(bLZZSSSJ}w@ff%1J4Ngf+(n)Ni))ECh$+KOEGfl6-=*G77Ic{LWhIqBtENGHlX
zeaUZ1B^!MBO;J(I&GIMv>FH%GEl&bzT|hr0UiQnEi^$)CvE=A{G>SiXnH^*ly?s8c
zl?@HS-@aWY0F&@E5>vS$PJG+U>}+3hwAI}0DLSHI;8O+ncy7hw9A|+r`W-~XbG?Xi
z2lRwjC2kJ7D}d_^I*pr*ocP@LqsKTn>`-D4yN<G1eIDZ=!4b;>ns0Nm|BW<z-)5PV
zoXqDkcl%wzFcS%S+Sd>^sg;f8o)afd%nCfqHEViAe@gF+P45?e4^J9S<w0DIQO%1}
zD2dMjtG>>+u{d>>i7$7SBfNIz@8TiK!MCa^pbI9za@^;Y)zz~`-^jOb->xiLVvscw
zX&P5%k^8}8!xdj*xZo}Xv<SS_oT9=?vVF&nf}t1J#KxNPtPMpRMiGO(lX*SJ*?RkW
zf4$7Xeo0~V-Ge9po8qc_FdFIRTF{}gE(Usq$?=l>Se-a_=g0y1Q$`q*6r%;70&B@d
z`6T^5`rlX48IO1g4J0Z0C6NhdG5HEmW7vBZ6Y)_0YT|nU8DYD@?~Jl>>n!Q%TGx)0
z&A$;eOFbpR{iM<LnvY^Yx=8QB>*9lYQg>DMN~@W5ea<!gz2zCJ@Z%y*hyY0hwpUJ>
zXU)^5-HGwU!{b6zQ<K4l?d{vQsaaSi3xrP#_$Tz*2w{Rb&cRW6$NY6VH%FKs@DfmI
zQK;_EPan@jDgs`+o*K~541e@!yN2d*G(8>>4NW4TK78QIN|C;KlkEZL6TO{ej~~`W
zpSSOFY$c92Yc#`(_rwl0&2UkGtC*@pR$)7@Z{NN#U3I+2EaKohTo>)8x#MTh$1~A&
zgCip}hYpQvviR&{O$OXH_;n*&2N$)tcBZYfQy+PN8f?&VBd6{Ak00zf`#Ve|D9@_2
zLxH)0si|C9ts&`Q_}C9xB=y#5=mXyTsI|0KJ$tie$j-%4E33|A-c|2`=+&$|M-;ZK
zI}h5Q(f5{|FA^Up5?t3R+lXD)RCw5BRhlX}U?`@|E^LbSE*L$T8FhnJy38b}&#nO1
zXmIC~8tu-2ibpMD3sFJ-$x2EsMRD|68NV{j)n1od<t3*7H5b9pS5;Mw53{E+COy<o
z(0Qw6XKxRXq1i9D!4j2BWS_aYIc3H_H!o^bU{_7qyYhSgd!YxWe`ezy(dh8r&z?QA
zGn<o^d&^JFM!IX?;X(9tV$xb!S>cZFuoTpKr|+knG63xO%}k)*t^jC{4s~a9<6rzI
z=a4}MggIu7k)I}S&6xzFDJ3LInKYIfgkqKP+7~WdFlbSYZvHyv;CjFLb(fW_8f}B*
z&G2dru|-AtOyBXg(SB17B5lQP=K(iQ^mzQc8{QzGbZc#4j3~XD8CtU_&!8^4oO*^Z
zE8K2<cfDw(KfshiigHSM!JefeiQ#GQ&>=%L+A~aQ>Di1Ih3TGb#HjviKf6s%tv<E9
z&%qAumV4U#OZ9>Sv+3EtQ~Kx$8)>y4YHzb2Kex!boZU8gj8($Jz+6WIGb;bvgIQct
zJ`uv|-v;(@vy-$6geRVH3oC6=JG7w5nfs2+=nO;t#!|<A$&lFC>q9w+7?1-Ek9O__
zln-P&=l}JqB*vtwYu3fm=*vWZL@mKcNosTGs+QIZ)2-1_GZmkOo&>I52i1~$EPr21
z(N1<x{<>LY;VL3s=%n31$wivnsJgu)x7?9?Vy$hGLzS5-Cgtmk_`R-)`xjhogc@U%
zf9_G&(iF&`Q}UJgWLsd<Zkd|?w_91Q(Na)ANy=Ivp;chv!F#=~<n)$@UG>$^Y2%_q
z!ma6fNsABP%v0CW`w<o__LWCHxiMeuDDC62oQ=pQc0B%%>dhRiZ{_qk6m-hG=9*eo
zp3JeWp|=z~w6V6FGV^VB5R3Sg^4=TEC6W)6ecrcqj;Pb>3m!`DH=?rcBEAv0ZJX!q
z^^`UBpX%qhZ{{VAE@tW#zIgS3gm=0<yeLt&y1MT8kDi{p>oQ04^~nQvO4CsHdg{5x
zEwM0A+&J@~#=FL3ho<z%__(rt3Qa_EfnmHw^G|%%(i69x6%#{WYYSUfm`dKr?Ks<E
zIkn6kt#oc98A)I%cq^x^sGT(_Jw1h{%P3oe(e_Qt7;c9t#f??NH@_L>cABo)Mk{h)
zW^WpEKt+DtI$m8<Bc-XS^6^)1Z!h=cd&~pw=&>L<vcbTh%gQM!EqWzwDrnGESX8`s
z7yjFHi*_fOv*TuQz0vDGt$D5#--f>ZPNUgWit7EKr$)E4cq%@a$Lk^|Y5JtY4qJ=u
zDdP)$e*Ous(z$tlXN~msl{<`eMt<W^`t$MbA!VP(>j~`dSOt7veYvQ$u_@}dG?^$r
zP)RY`x=-CjSRvwc)y+J9vo9ony?t`rH(V_RMH$<4DW-bj#dlGKM8r2JN%=gg_Ils3
zIu&sGd+m?@%GXk}i?f$(3mUH|*wWL@w)oQ}y-8`wxOTq%qvhD{LW9@8-k7f!8SRX0
zYg?G^hyYI|Xx*c1E5Ex6WL9fB-F4o=K$QzC4>DtzKTxqIj}Ju*WSGCq$<Zw|r`Zz{
zH}pHtxQgmWoe8d)eB$2vgZ?ni)tha6KNLAMVuRjc#Gs+8IGv$&GpD0qDLup2FKD<+
z*q%D7@~xCl^Xt3UD5!S5tYa2)e;Iz~cjFOm|A*R!Pt%HGPKn+_ndRa=W&G#ML(*71
zsgnNBw9+>>f><njas$ZDp4E5zrD?`{5C_6p`aZ;?tFR^ATwQBxQ?xKFbhI;>vP;B~
z{lvNO0Y=8b!4Z4ie@k5HKho=>{rm%OU!3m7q(pjzcBckiRhp^tuO+%zDmNElU7vkn
zZ_K$oJW@+axJpW<+V2nj)16%xepC2&T%spHgO5#G?Y-W~=h2EWa`KnVqW+DzK74u^
zp|W`^S1*IXx5?}nC#Mf!&))Pjj<j5|gw}DDLj?eK<JBuHRJ(X1YWQudEpiVqF!;7v
zzH8ofHAR_)7|=+D8{$<kA2ygzlnC<%-SP=;Om)n#$WNb_A}>QVHE;i5yclS*|Iwr7
zs}$R|F-S@Z>6#78DKm;JPAgq*A;mbu8lSG@%BvKjDiI>nXUeO)0V2ur%9RAuECcts
zjpc<&%oNi_<v?q-eKedaw=4FL?PL13Ze2Q5^6aP@iEOcJ?e+H$o&c0EQ53DsR@KG`
zhmDLhnD%~onCdFNtQc4Oy(zIREcWSQUQa%Nh>))8MgO+IUk$<9602V3Sw)X{xqKh2
zIG1cro)5LR8*O~E*ZQ>Hx0Wyt&9PhRwBDdo<2QWvwY<J1Yv4RB-tD$^zr>(44!Fr?
z{BGXaf8$wcEeSg*ujM`pp+4ZY*nipfY=G9=gvoI2IgkVP%h#Qr4cN?lnQySKy6%*1
zJ8iHadXE5?-T$UUhZ~}011{X>;yuw9u5IwzcBKC2-hG@2N{9RvuDo2Dttt{XrE2o?
z8>J+RTAHYiOO$OZxwWhA-$LJ;eHUJB<+grs?GW^*BwJf74COSbA57znAGERHX5Cz%
zQ4wF%2*0{Q@PN3&f`29dE|P!rB~Mm~53;jZye#W16s;0B`1p+iqksdKwdW!o+fyc6
z0ZyNT$ttFQ`5RQtvfr=#n;vUT+1KGTbxeXkEZeE)X}Oi`zm`2u=@|18xKafVZId`%
zZaQt?rMxR@cxh+xwsT)tV?9Z(TqH>^Wwp3?@lEUI==OEB?b{ny|0S;dOI%rb#Sq9P
zBlX5~M|zW-t&4C$`jwpFI`LdCYml-*30#0SF-y7t+=F5vQNx&S3ED0+^Z;lG)IIR1
zXTjfr9dQ8@kgfL?h*UkxK)hj(m71DboUveoSId~#Nc!6Y4CG~(MmImkecQ>&MSfpL
zC)ZIgjW?q3d1_1cnhslwhw|438~R~~bln6#dY=H!HBP?tk;s<Of2^KKAOId@W}dt^
z-6jP<T>#!x*rt~xXQcL~^F<QZV%&kCGiHrP{1p^djVtzmD+@|UXBKntE&3`+3`_$-
zqU015w+AYpQk1TmfK9CAjCX%j*LQ@EFLZXGDzSCEVDeLkoPtW#jWb7Ry`>diP&3g#
zsW9rhd}U8Hr*dPGxq7g2N=tQRkLO5aN<0V1cTi&=?krmYSLfqbi2(jyQ)k=}#6l3d
zAtBVf<DHR?<E;Y&1CNu_ae`+|Y8k-2C|<ov&~7}S04%$UBIE9_-4I}!-M1&fEzhcG
zbvU~7j3vWkAF83j>adB9()j)a_N7Uv0rI<*q6Lo&2s~e!Y=36gPYNo9-~qm;a!nQg
zvwD}mZf8<7OV;JO@ak@~V5fb9I5QxrcCwOGszvw*w;MKRWDE_dSj4`6dAQ@O?n!3h
zdx0tv8$Z>1UHd8yGKyF-5SMf>`f1N6%h&Q@9QvZS$<GHdl=UB1_+vYB<e}quXk@&~
zJ^9^N9+NTHE>CkkP|I@KsuZYe(SLKAn|+~Pl>9@-eyUwxO{8-4_M7nQh5PKRv#ivO
zQ5$lsxc7Bzp)+UHfy-6g-(Z&r{uQi1F2E+Yo5I1<>ddRPwFENKS7iB=-#1D-Vl=kY
z8@5d?J8s)H74^(uy4x`2pWD%^v_Bs_Xj$Xf;tifP;dfS2<)I{SqnEcgP-i7rJYpEY
zfih5#5w-8a*1EmByZcKq7ryj#KHL7w7yeD~Lw#`BTph`9atKN;U^PhU-*rNt*Gbgf
z@s)WTYt^LoH#>z9GSlpUCv_I(4O<J(WL>)qw;CPNf4q9QqhP814iMP0XJx=yJM**Z
z#%!n$9jXC?L2yGVVzZ<JRl%I>?2nMuZdyZosi-g(q80N-1fb)p$mbaeYl3C^Wl_=4
z(9mUS$1T+&0O%_5q{o2<+RM^U1D-^#J%7^*=tu-Je)0ckuoT=eRc(R6AdYD!EzB+M
zexOWnNn;3g@TVV2yky03)+MV?z{Hu<$L#(x-jdQ%a^KC0_3R^rq*1mGRD+aoElRS-
zrhJ(dmP-_6?-;z_aQPPoUuse1Huk`nRO`8X9h3YwjPZ4#0)2yn_W|tRIzm?5Z69lr
z!<bn1piw|lnJ$j`p6Fe@yy0iEhcC5!Ob2nbhn}udbL@g<po}{W&2FLG8^49snoJYS
z$qQNPj4tKqP`^+gHp=e(phHd1ll;j2ik;o=JybF|`IK@xZ_NaG-Zk^ZxU5dKYK^mU
znB;8C#&uEoKQa<XSXuiV-Clntoq|M3ijw@u-|&dpyG{L_S>x=lwV1vywi-6zkKs`9
zi&NLOPn}C-bsc@p9lw)|jL)R@<=*6JBdILz(O;f|9Bj0-N)^kN7Tn?a$asP|0=xZf
z$ZkI?>-5+_x{l}uqycw?!OyqXB9Nm84jdpTMo`U%K?KRD(aJn!8jAeF={0jZDJmHX
zOyigXG^4r7L)NM~I&zS+%t0~XgkqJeK>6r=G}{lJro7MQjQh$Acb_VzIVm41V-OY4
z1Q>kEFZoch=zMk{28~k%1qFm?wEw6~NK8y^^BV>8&iu1wWo2<^Chiq>-O{03)8~yC
zvBy+I#oX(C(4ZHS$Z+v;ccRmDw}8b@Ur02nSFcJcD??6${`Gxpqi0KvD6>fwm;eTB
zgq-oYpeYEQ?{+@wTHJlhH7Yqa+w=6PQ%{lV&p$hkA7Exy4C7ED6fATh+PbK7iFaDx
z5#)llfs1!-ts*g6fGTYJU>uMVpK9GxlBt*Lzwg2d2`eirQ8U3-Y_5_~f4%M|#H#$1
zPhEpfI!ksVnr-c?(Q5(G=$Xo=4eJFpLx-JJXEV(;4Mk&9EX0S*hD<)r*?HAfGKxy<
zywR8#94J;g{#h<R>1#nltZ3`?<E2aOIo+V2Y!*h<8V<bazDQmMCDjaUcy8TjT|)yn
zZ*AS*OYy0rcSPb{1gnBsN5Hyz`S`RX$<q)_=I=~KC|(%+8{glQg4%k?#YOn-wLA_E
z4l<>dKcDU%s;H>=Uty47XP_DSQa}3mQ9$pb*~9AF?4qL4xb4qkVgk$iUcGwtz^d!>
zO#voh_mwk<iOQ-fAE;TOFJ8!8zDzpVo+~<V><2WKu!soRk^0!(pnwZT;u38KD!)}$
zhK7V(gt)1k5(cv1sMjlSW3+U1O;Eyu4vSw2pg$$n8vgR797qKW0AS&sMMYiNSX&g&
zrHH7J2G2uipZM4y;QGhM6~JX-qP6V$Ovgq_+^dftpMxh;*U|YN#3D)%pP=bq<BK5g
zqt06gRT#0IJGb~T-yCYI<G2>;9b{wglv5u6mL_Qr9wcM~Fdt$)>cks8eR}-#>3sne
z{b$?z`;BmB57X1Zj{nFB%&s$$QhUx{znIEwmd$a3mv<@Y@p)mFxnuqbjjzhaa@W8D
zyGCi>VYCwu@swox&Z=*=BaN88LB<%K*PQy3{a8AH-bM1kd9%g`XKz);2ucitVH=*S
zIh_KDDQ=nDg0Y*bX;wRo8&aA&R6k#Xtg0~wySj6&jd|8ROf77bgZp@o???+*(m7Dw
zGwT%gyWSg8tBTHDHlppB3V2~MU;ntw6V+IBH<?avQ*pYxc&zdDjvr|*49_B*)dP}D
zU?nKHoyL$ltKuxG_<gl^GGsVZ;j!MqAB6MZ*cK%$2fW<%0`lWVQTfN8{p7&kH|<9k
z*ZA-KHo1K}M(`V`h<UJnq0gS#YE@l;EDoh)D?c_p9Vq|Gl`DS=ou=D%N67K=;Wz6K
zOYa<7mtir!efvF5e$SpgHU;{i{KJsvfZF2cyntLjS++^q+40TDl1Ctuorj+yUTr4+
zoc(0o6I3$<VUq^(3AIgd{g0ZOqhO>FV{RO4L&L+an-b8z0eVCS))>)Iz~j$%yG`5n
ze%bb=SZoZrFTPFqoYDX~yJg!4ZwM#gP^yX2j%rz-&t!m#aqW|39|~Pu{S?ZR%M;EW
zw@pkw;8Gj(+Ci1#FRsJh8sBUc<mH94K1Rq{>}m7k-R84%BH6QBnTAX7keTVIh9KF4
zzS^Uq_oln+fSn{YG-#o^37_w@*@2p3)cEQWfk&XsHbahtWOo^)BssF9<loc)1mB~_
zk3-Ya$u`;<EWX45@$w)tG#S={VgnWYB$1^@WddJ;&BeJuTVoA@9swQm<(Zq?MPjKy
zv4gZH>Mj0phX@2!vEaw&UpSAL7-k3^ppw3U)Ge?dIl#y$55*N7*%Y<evi+mq%a;sS
z6cnmKufe22NNz7*>bxo60Ky{Vba1u@n3w=;C|3r<REP@!YNcQj@D*|VJM(q#X?(IB
zA98Xc&)uP~i8wt1X2^J^w`}3>uK=iH`xzL%gF`1MW(x<QcbDyW<#s&#5~OmYEzF8$
z<&eggX^QOk?<2~8oxT(-Tc6+aAL=woM2R85-y4O91HciWk?RpBt36wrvPjkw+}sJM
z%MO$6;mFOB^*IG-t$`rL2;F?C!<rLhKQt%;`hc#jhwd@j{AMqzOM*fKFPOJEd<^Ix
zRk+eRD9K{;A_y08cplQUvkv1Un4*nAZ)r?yF%xKhd+j{r&N@gVE|a;fgxv!q*hG77
zz;T5ay(mI9HfJEb6jx0IF$;gig?h9*g4#fc6YojLWGFy%JX*vZQ_ZuwE}%5<Cc>b<
z{Z7`<yTYk{Xs5>LCde7y(serymAKjj>3~1EcMcL^14@4o!d(Ozd)hgi+Qp1RN0R3y
zWX`-L53_B5uch?3u{7w7roO%+uGcRhVE&t$UcL<*Toa&FZszKlcNYmm*=Og`%It+7
zW)R5llBOm)I>F=Do$on@V|w-Kh>Bt+wEzE+FqboHM@B|ub#xf~6PzX8x%@$Re!I>0
z3}u^86Znkx5CvZ8&Jcq(v`_&Qu>5=>2zo%oO1?os!_a$-raB7*?zTUNShHcwc^{_^
z6GqfX6gr6X#fv0tDMq)`(yzF<6qpJ+LAp2_sYtqOm)n!R;h~}Zbac}0?qX2YF3ZYp
zO*$QWrZ0!;afIvZhR|MtXd3tW$xno;L&zQYRn+^R#a|lJHXr%x4lyz^La=U3l--AV
zp^P_rtC!2p&mRU97ZMTSQ+8wKWOJf96Meu=Hd5Y*y&Y*gQ<Y<c%4AvZWA-|B?3fqy
zX;7yalrr9Z_D61@%^ZSfB_ng^)l0>$i-b0U&prUTi4e-a-Z*nZFgjc3BvDZym^OVd
zRzM5;@#@lJ2>vFxAs-Q;KLTd}8qkbnV6ACG;^5x=uU`dfwE~WGv?K2{VX28qo*mCB
zox@y`ZCpdIB9;LHKL_pz6;@U0_fmdIf=5J~j++K2KOASJ?-$tt30-u@E~-KN_b3lf
z0E)j=@uC(9Cd>j%NVAKJ$Hm2CbMLgy&6MvW`WgZH!Q#Trfe8RYHY&-HW5-B=u&0XV
zl>y2qnFM+vEfbl8d%2&M_OiVEZpgD`Jw5S&^x+UOD}z`RP%m@~?2ZF2#6?{+CUkXB
zb2@<dkeNrO3a3rbtRc9@!0dtf;!nQq(WvCJm@lFP@BV<b=0{fPrfWO9*`-UDZkw9s
zzz%TOedbboo^=DN@zeNt){Z=D*%pC!#Y=it`F&*Qx@CZ)=3Sr9LBdQjZHOcJc(>yP
zAz0$??>)CX9@TRuyuEx?>bs84)`LLr#a#!loy-Kk@t@-K_5RN^bsvHlUrZ0y%E-N9
z-wm?~s+^>zrhrl+bQ)B)O`fM)m!(@`^Gs`MbaXtA+`1`$EcfMKe{ap3m+8uAuk*I*
z?vl$ilgR+~`2yPf_m0`jgY3`rlaP@59}qYi<7+>|6Bn->i`f2Q-+#?>#2ZS^poF5d
z9G%~{L&VRiW@_<3w?=>tW-hmVG^asU%oS<T4xJYLoP)|K>NLgZGrCDbcT;PV5=afS
z=Qa=v_1XxH;Ei*utE=v?M47TIG~MP?Gccg~f-y?>!|m$QX8r3ilf65tD;^6_SMP+7
zb=ANi2sR*;WI|yODc=&yE5lU}`;oJOK{f6vxgXb1Q&ZEBKXv}1t}bIKs5|<AiX09X
zV@l5;ZDYR3_Z^deECo3fUtYX=wb}fiX0KgAy4l<BGd!is04aQKOO|NAsFQJ~8WGe9
zKOEa8?=_qQ9^<p=y+|uP%MkL*V)11p>;iT{1pNe6oIIyZQBjcsqt}wViD0yNxgSqp
zM*rF-Ramu#@u8F!c}SdeQg6?^{ZaKv=Op}d3$EfggAwD-&@J8%Q#`_V2>Czm->G?2
z*Ex`BuUf|$1uYpewB8yuC&?FWZT<t40Od>=KA`myb3L@@NEo<mU<f?yxo!)h`78qk
z38C*|!k~G8<7+`9<{oJtUmP47BD7vKu!8B5fB7gcx}Pi=2?G-eU~Mw`Z4riBT3T9a
zR#p{Ax_?UUi$Xo-KHud)gTjA5e1L<KbN+8emQhaIC@>bGN)qA-5IA&UlvU?G?}L9n
z81Hpknz#gW3YtG*doysKl_tD7$R8nKM^qT1_adif;Ao;be7GJ;Kg49F)5Qx1AnyPX
zoq?`Kn9}eKP)LL%L<rf001JEzc*!E-a06A5Fnz$SKok!oy2^#sV*uI4fKLLJe?or#
z)FA8^VC-~<KI2A43fpr;Mn+;(K%Z}bth%k+WrPL42Nd)M=A9z+cVsF-RwXOP{y-&L
zK}GTN_3iyU{%)Q$Ux0z46CQ_;);*gs(=^qd``}phklTE?yj&8TwaGesfZVBCT`faU
zQzuqy>aOGI^BmpIm0cxUn{J;S#=fGbZqs>xV;jabSQ{}H>=T<Ixm6k12kOEdmbgy$
z`8vXw0n-&@PFr9C7m62wuI7en7)9@$EW9__4{us8T4YwU1Kb|4qQN8f5W=K^+vGV6
zFJ&Mg3J@PSFTzNOE(=1=dGVJv+-qN3TO-gFV<HS({V+Ae-=l=lGEpV|0x|CGrr~gA
zUwh)^HPat(dJca+xUbp~S~dC{VPk+Ji?gk*jU?&K8^RNXTE}qSiWymLggf94N1zx+
z>%qanS1n`CKn-$NuJjXKA5oGFWJ8#^pnCGdwn&t^wWY~`1g=0v{wF{dYA|LZG<D#;
z!}Z)k7b=3fMch<)2@O0}jhmaBb<pKuoRWS^`yrYAzziCPGNdgqIkze)2xVQwX)5rS
zh4E~EMQZww!s(J4ytdAV+q)qxK7aoF415H{97~ukVGjIL>M3CjW-R{xQlw?`XHpCX
zuWjXSJrQ7hbm`ps<UL9I^W_YIntAqSR8d}nt2G;k-6qZwd=a6}KiIb84AWC`Hu61t
z67F_<Ld`j_fB&U>_XIJ%!D|HEKn?ITY0iN}15K}PU{H@30wqeg^HS<MFb04Nr2+Y1
z9x4nF4{r+OVRE1HGBncB&H{ee2}}Vf5TI@eOL68@m`h>*z^E<*+=ZEM02mc%Mwooz
z+d&E(!88vzt3vjcYY$+tQj4#@Kl|CU`$08GN^XbylmIQbROq^5BlR^k1IUFw&^K%s
zF8oIt4b?=nd^A^AF>}(?6pys`F2J-yeefUwi9kdRBVv-EeM7GSX=*DH8&|*^PMHLD
zjbIdO<0azJD2bNpMeQ)%%|g%uaPr6#RuyvcSLiqdaP;*h2XekFCr1vK(5s@NP8uUm
zR1Q<|c$v8#875yZLC229WUJ_$cC8%q)C;CPj>+as(ANp4CIImqpqn#aslV>e-ni-s
zpfwB{1nwq`FaXEI$wEn1`BH%<{A3?%P<r~G$)y&)6JTUO1h##4<U#vJCc&wMG)Mr6
zfm?uS-KS<K0av&%+VmLVPn1p+VhGk4!(F0kzr@N{RaDHP{GM?bRVA_z$oWr>={^k0
ziGbCa2E|c`Tcf;=>iYUhDB|W(DJow~a$GogwHEGc=zB_UL4O$@ONl2+>PM?CCXmIr
zkE8Im!5-O*AeLzn2$VUBcSH|Rad-fivTM{&LCZfE!7IRZ1NFYarQxRpGDH!q(I_ls
zG#rXY;02qRu_;*{KIFPQ^%MjJhK`elB{A==7tY~Dn}G;}O>A~OrGOpf#Wkxkq&*6R
z`QZri=8MNnpC+5@T-RC`{-?^}pYR^g>vxtRVb|R!zCeBFcV^j*GmVIb23Kaf*G+Jk
z-6RSNtQos0C@yEU((K>=u+1_VSxdN&iqJZcZ=isbqsd1xULb@CYKw3X#t1#bU5|79
zd*Z_CAPqCsE)1_J+K`Db7NFVN5~Ka6yVtBrR!uIfjUEMOqddMN5|bS!zkY%ocyh>!
z_+VyHPheLf+7FnQWY4Wl<nw!YTMRrm9EnoW{?Yt#0vF2h;lw6i5EkYGg5!v_M)--)
z1I;@=okE78<7LB;4NGsiW9#**muGLigwG2x`k$!)Gvb|h1y6{w*p_430=Y-!)Br{m
zBUk_k;}bNlTW1)=M)o0~(OFU`R~?0zPwKpXfHBi`Vf0d-RhQY%)cr7_nZWhZ={)e1
zf|1V)jRy^cV0d6?CC98jpL9P(IW~Xs_cg-!j3R=8GF6Y$?25%{KKkDhz+oF@-oD<N
z`^!GPe;=}QFFjGVfNxbg>c<?Q@xh=w`|aidNn-fizJ1f<dYI=|OF_4O#@%cfZ)Ux;
zvuYEAFI<6Pj{zZ6D$-MWsYSqx+Vs95(a)}4?JxK=Gc&^-U=e$o?&#m2{3mx4Z}~9L
z8)U~3k=P(KT-Z>Y$KFJ2Hj8l^>73I2-~}v?l%Vo?47+HR`d0}<VuIM(T$Vt?1f>Np
z%3+U{KJGV8Y@`HoV-)Q?eGDY2gGfXB#ow8_<uJFw4gxC%j<Os*ZY!6nuxm4-jQ04!
z(+I{~f5ZromyK4(Uk8!H_ZONhS{u_fc|c;X=5=T=eD}OdzXBBfclVpq&(^OEUY7J5
z6kE-x4m?o(=F$I}hYsR(4q_Xj9!5QkR;I-&pj1TfO$cS&RZvjSEq2Y-ZtaI@4;fN)
zWptn4pwVR}-$5|UOy)mRn}i)dSz(@mL9BaTFROKH$ifTJO9bhN*j_#IGR)Ts1_s@X
z5{BqNuz}``<ff;4reUh=@9$5qJ)U-A!F9v)9jlDptSZlR5lnMAHhq$W4<Dfnvi`w8
z*%COdz<7wl+2_*xa2Kj`KVaQ3uwoT569vk?obtT$>K<NYT1{!;;5X{aXjNam_u2|E
zQ?Oo$!JI)uM^^!FB0MX{QR`tmwy9!qQdL#`@;~-_SXw}YAWgEWY&`x=p2r>8B`GP%
zfI~1({D@Hp_Yd`24ZncMC#X>{Be&tOB<aKwfPi&(3^0=y2KOc$JqlDMu%x`69uq+t
zw_bOO9eD}6H__s;JYWxc3Nc@Hx^(>!c1*rJ*akLp7O;$1n}B&=3N@Zb``zuq@6X)V
z#!{ZYco7;CqlgU2$CRd`5Kky%I3O=-0uc~4Q~0g}LHjggv4hx)0Lx4CdK5nb9bke!
zd$(NxAqayHkyHe04Fco|6$=sPx4HJqKUE*6NJ>rh1u%_;YKJ5|Y*H6h0XzEk{1I{>
zN1qew$u{ukA*QNfIV^)4xuD;lhcFDoCpkMux4z8P^TGSWP>mXery&ouWUqtp+0l#_
zncn5fjhfo~8{c*?E=QwWAU~4amZ!|3Dy>U5Dk;x*e!*GI%+KG#XC=_eVR6^@H_Enu
zEiWe~_5yfZLBvIy)&&5@(4W-#g1HX7$6lh8;VR$R@uEEemJ^0e5N<{AZ`Qs%`xt*E
zVDH2|QA~~^*iyNTUxJg4GFDj@j$^Z)5^*S&iKxJuL5vIxas=VLdv_n`9Af~nlgL@8
zZWkZ8%?LLv`af2d`cM{(0EdZ&1>0)?Y8XKj0j?8kCu8U;*fGkqTPJ=ILwHM$sR|U#
z70^n$z+?oT2BiiJLX=!$dTnlLiM()M1Rr7;yg)y^lSEx6mN0O(E!b5eeh^DWDx$%c
zuXg<u*@4Q%hA|ubNE7J!IQKbuuyeS@@QYz8b(tR~XCrn>V8jJsK8Ik+HgDk|RuoV*
z)Q4V1CBII$2+#f4V6xq~;&#TK^jj=b^RGV)|81NZ2~Ep+PkpDE?RlwLXj)*hdH$w=
zuHqldm5meJyfJ%<Qn`qz0fY#}Y>E1VI(dow5@Hl!l3=w65d}L8&tSL*nvo^w39J*O
zoR-RJ%|XKmVwI>rXy{HT0(^YU{I47BPbqVP$4|kafSD1V+}g6oyV~-sSqX=@VDwYC
zOF^ocBgF`4jqx2d{yWW;H|#&%hn&^HR1@<#7$^)c_tzHs{CVC1Fl5^eT_IFpD=S{C
zcLB*XoFm)0)5Ro7Rz`*(oAoL$VANDuBeIo1Q;6awezqyh5^?R07-@k;2+9ozjf{%f
z7oZReHQy`+bI{uZWp2H_7QCVF;WQ#uVV#@YU@s`b9bft(C?H!#XPn}jQMPjw54TCK
z7U|DBXkp+o1~&gqSM~I;J}!IEWi<i{?{Z2D!7CfwU(^I`F&%mHI_e0Llk3zeV{km3
zXsFCWHocjxr%+_$?ib5ny0i_TEaOwOD)}W$!Qh{yN<A*e)VX#ewF>J04D%{i*om~b
z5S4~bTfYpYjtD3J1au$bB^cBMqwDy`-7vYs<3jM^4NOefp2M-B<~k!}#ch+hvoc=8
zVbw7ShQVfi#TG;ssvL3r2?@@q-Wb?kB_)x<d5s2j^TUS^F9pm2)b~FqBx7b_xo+(H
zuI$E_J{wg%y*;YNZ$mNnez0ug+gvW)nk8m##2^*|P=Vue`78yaWwR=b%B=F-`sKC}
z6fDoX{71y759=nOx}nJx%>CfO$VJYEjVd+}QRtgq)NzcXup<HxBP1Bl8j%7`CqqjL
z2?^o%U_B>YVf}H=@u6tT$BxO{#>NOity?27u142AX|n{Jbo+IYo6Hx>A6iY2o`@7W
zd$x(E-Qtm+=@EPod;?ZUkh2w#m0QQLy9R4KHm`!IAEBJW=Qpg&BdJCU?lb*l`;IAR
zDfJX(X-&=H`d9_f+G4hW^<)7Rg!+J=ewU`ODeFZ&(UBL7g(%HyTw$=jqQjSrG^V`X
z{qTLqthoBF6+;LAJ5iOmTSUBr<bq)xIO7504T>cxDJdpD3`OklBVadg4%$u*vKS;g
z<^Y^DK^zlY0>((9q5#}qxW8tD?J5_KB)W6n0JKm@rLbW@Q2v-!I$g)#7A-U@V89~Q
z^Dr^ibQai)XQ%0XvXqB4o1mr%lLdo_gIV~%5@zN|epA=&Jjk4Fji@pN|Ax^eu+HSD
zhzJvCKXgPC>v{de^mK$3H8u5vCbP`uvQ3mPJE)A{s|bGM`e%omaBBu0Iv<35B1X$6
z6pUZ7HbpII({q1W4Se>aM~?vB@ET`sekHxIxpJ#ybL|eQzRN<B?CkP#=?zA6@4h&5
z8!!0x!O4%%3XbTj(Z0FDMPtn*H>AoD?uac&OUnt}5%H~^7S~Un#8`k%_`SONbyAX-
z+jQ|kfEf%HFNJK`A!?n4xXF3^xHn#k*y<wSIQ)>989=uNVQh)XPU8I!ItF?hXX_v<
zD<Dr|5UMrdCq@WM!3HjN*mx1{*pL9}0G1OY_5e^7yPzO_M9t9ILa-?S=m@HmUr!LW
zS4(S#(wjN8{@cgl6cS=EG&F>Ci1Lh~6iB#H;OY802mp66GBs#~000;*$#?HA>$7>q
z<<HrF7kjF3UXZM;t+~EyvdDGabBV5XWyBeI{S*!%VsEj>2Nbk~M9(c5n>8m;WZ<ge
zKAzy=NyMayB!&&WVN7%_PvhHB(XfZ0ItiBY*BMaN5LR*s7WC00EcLxgP5qwpwGH#E
zJYY7l>49nYfvA`i<SkUvgPMyoMr92R`*8Og#`o+<*Z-HcQiKcywszjL54|An+-Yv8
zrX{A=0|@>sj8nzc=%zS`3{3K~f#)(LOmBF4yiMnZKx7I*5on@{W-kv1#}m^ym*gKy
zbWcd8t~yE6{E5%Xo^$-Sx|(f2qD(AtzyaGL0G^B3a#T5_lnjHO+jPUFeH$Aa83K$L
zNMcRg56CNB9j74e)a?GD5U#b^S6NvJ;pdw1zx@1sei?=E^<^`d0kVph+Iy_jv(JRW
ziH|+Xi!A~woPP@hsyFCB_ZwlNiMy4a?%Ey`G=G=f99t#)CbcunQkuUknod4*eyj&}
zs=DoJftTU+QHRf-EC)8HgRY15%^VNi>lfq4`~P394&Cej%YxV>{$JiU`_?WDyP_F)
z=jY~Dz|u9i#;F+Bn#>zgWq)k59HA}gg5zAI?RaPao%xZ_y-$+O;n{A?GA!l3PqFor
z?LL%s5BR^hJL1j9G;H<_>0>8Rx8d76v29aGI}T{mfgF6Eo9;vgHQ0iF7`R6_jpTod
z4PjKb&AGKrtF6+G<hj6cf3Cd;{LY8Oj)vyQXlgn<v2V!K$}*?}?4$ew_gUNfn>&aJ
zb}#+OBhWw+64;{z%n5Ny8M<Fq(cB&g0Y^bMLVfOq-ov<avErcN`+Jjm1d&Am<y)JJ
zuHvB6#FjhlL1PilETTc-Pr6zD#D^MUJrqUwHERCCWJkoW3~g-ZVtc|l{6P$;pG^IN
zPz>2<G|r>1eVmuOoCq%(R70_)Kew>qF^9GaKSn-8!9N6Cz{V(A8l=AuKVnF*#)gKY
zgklUJHi*02i{6e!y?tz?d7XB9f$IjaCw%nWxp34Z0zL^^b$-XZB@IRheolPj7mbSL
zSFhcmHR9);jy_vI1QAK-)6od=(%8#eD#J&V!<`*LXo(2uEeE}B8%oJCq?&0njc>}(
zK1~eKSTbY1zpx)L4}g=Jk+B*FR}ZF42!;ou1QECZ_WMDw39mCVYx(yLLRnr1FB-@y
z5r>&)9^xLcPO2R#kj}LSG@)Yr1;PwSP^$ofsLWIZ(TV&QwjVxoZ$jtZR3`z(2_yzk
zL@We}ZLQA}+Cmd694V;pyZ7xgC58wbJF!CupiJ!j63K;D|CH%mKL&JmET|*5L&L(L
z#Qru9lL&?$1?UToH`(-RKZ+{BQ4)(xV-9n03&B3?g4c6_R=m185|3RCKy$(wMv#M8
zJ*vgFII+({An*B+1_>YQ>+Gw`kTZG@nZ2@le9>X8^gLD+cq3>w;uB^+$1rVP@?@!Y
z!UiYdA0nU`mcwD)+b>DY%v=j>a|yQ&1`ZA&0+{k$mnP&v>;T2j3G{)PCAQ4500kzi
z$|AQp_l~xJQby>dLVaHFROY|Z>FyP>VEx{8feT6>H0#}H3&s)GWuSoP|4q-pC`Sh5
zAIL0xpx7WdD{GoD+Dk*T!b||PFvgVQ2zE@hhp{LFQPj`RkC<tRy@gJ@np4P9XOmZ3
zk#Gapp^p+QDv>U@shHLmrn)+Eh+rfJ9%Ac%tl67T2Qzi~3AO_tiEtEQN?u=|Z+KZ}
zf+q}oDBYMk78-@8ICzoOeKl*GnE0yrCsr$P&E7ZujX9)Ee&Zjs1X_mpI9NS%oj1ai
zBe%<{3d{=<?1Sxq4AI{)`=Oh2MJ3}w2!6rAoZQ_0@X)v{t=f+n0EEu;UZP?ax`_Uj
zp5BZp1E{J1WD0yMi(WhhP!<@(oX-%m5nwk?_kZ>o`^;b4U;>78O9JW3NHE%^-;Wcp
zF>bdHVNpO}6KV?4?vcGfdR5Nc*i|MNMMCaJwXmt6)5a}D+EYS-`!?nP0fvU4%z%AY
zv8FZFmcxl<z+0&VJ%zn6v4ut{q1bx8gZH)V%#ToBw)+hRtL5EIF`q(s$+R+c$`GJv
zJvV=*s$qO5gV#Y(F_5!f9f`b~ipm68Ou!yMYLZxGh`2|9rvQ$yi}i(gRshIfED+$@
zMn5)=VsQ(5vol~ALt|qJ*Pa}p9YH*RXD5Ayv4D^gka0J1^~Qc>5bg?-?;&i&poZL=
z{U}{c?3}@6To2Z;UTU8>5skrObL4mu1Hvj44i!ST!0}<0Jq;;15fdEo5COs>ft%_U
zq^YO(BSzRBp!fCv?)8DsKI=54zd60N8BD0C5HccL4cwl}9(M5py-8?hzu#R~0D((F
z2tdb=SJ3Fp5z30!(la?;+Fah3MoR3DPi&TXQzYwTPqziqr|Z0akNF-h4nFJdq82_G
zpYn&0=n*;qAu6Y2$S)zGHto{d_{Ub}@34`gtb!p>c?;#qtzWeO)0EQ<zQ}aaQ(`UW
zKI`)WkOHX34WJ3{r>3^4oZOP{iBF56@4if?zP^J|QH?##r2bxNT6y0MteoOm8*nJN
zS6<D9i8|dMrW~B|5pnS@8pT;u^kvd+u|q6?2wZzW@fvTgug2Z*hnX@i)Wh2)=ESY7
zjFPa5fP=&+2@V5<*~1;X_{<t#m3D5%2d$bG9~_XB4&5UAj|l^XkGEP6?%2%+hoNci
zWI;3dQ3Q+jsj92A1pL#i$vzg^{v_t1gWuI`*Peo_*RB-iUXs1rsCsQBBPxWtKkzZj
zv4gulklFbifSe61jlyBNG4PIoa%gg9Kskj*sipo;x5w+$Q|4-CLno)#?(>q;y}q@{
zywnYbk?YD+>GLm~4-yN{bcwIHe|K(ja{rF02QEeQ8(io6i2c{F>dX52G1?9qne*2V
zH78)i!dGx7i_)RW@72^q+0IHxH2Ni@d#zB2Snb80nDEoo<}rs{BXK;kfUrSmJHl0E
zuUs1S%1HNy^>`}7jo!N~p?d=+c07wY*UjCx8QK0SQPAu`d!m$`IXmV5r!W({f&|@%
zN5t3}pe*ANG1V??|JQy%`t7+8#1wYS>sNG<3P1yz-8_ixSFd^{<-+8SMF<&daZaz_
zd;W2IPTm{maJ{_4P-pf@tcQS7Ve=v$MFEj=Hwg+gHhZPu>PANebLxZ2_5oO<+jHx_
z7d6FUv6pasI<Io$@+)fRcC;0hZjws1%er0j*G_Xh7XlYA8sb6BL*=PziOz7MLNp?T
zH-Z+y4o3rwVJ`UUd%d^rm^sAIhzXtxffg1Ud%yD5xWi|A1IEd$->xMg8pHx+k<#5p
zwL=1OEZt2vJ{7(!6^s-utlWQXWZ{kLma<P?OX_FhZ36=-h$mA(_8}OKq9SI@hyU{!
z4^UpnU2OJ(Dq+Fk3EkHW8^S0P1f2mE94z%Vr?*3}lR<O`NW!KrGAH=N20#v7blp9@
zur}|@<PYmi67%y{GFwk)?MA6>ZNYqs!$!RYP3U^}((%GB-j;;WAnap0h{U!!=%;>c
z9HZT|cB87`*eey~gJE5ANL@3@JAo@@3^@Wil7^XC|7d&BC9DpjS|*!+Va<3uwrCXz
z;H4NP;76FLz>y(htD1N|(Mi5a_teP5U{tz3w-<|_XmBC$UP8R1@)^infCvMDxVpAB
z^V1zMZ7*u{wua9w-Nfpk_^JQ4yxf`-`wa>jLD59gW3f9N5C>le2ClmZMhmZ4TU*Q2
z^om>6?B3keUjZu%)`#6%u)qA8$H<4L<@9cLX>BS)@&Vz80aBw-&{}Pr;uzNEF>n)P
z69_cIk%ezVBLTM{2L=zGl$&_&jliAXfzV^|iEHs6^v9I^{Ff*i4YSxhcA3boA|WEh
z*Ut~5S6}8L20z%dBrzAIZ|F7bgct}UU#z$DQ^zFp1@}MW=4u)mm$4y>6@tYmGSt*$
za|4wCR8DMg$}1`^qz54kQTlO`SfJF?yMl58^)VyczNm0X=;5}_z{#E<k!?@-t-_u?
zdzip=SvI~~q8SQVfCXd!|4<a2m}Ln$9nS=spU7$WfM@d%o>wCyc0<pNbEPn15OWo#
zZOEC2T?Y4?G{ilvGZ_HqM8FPHu}|fFLL;YDS_HB@`y|O#%qydnbaL5FSp)>Nzk2;?
zZ06oS52X}oEx&LT!MGcW(<`L{!FyOd!|9`vU2^GOBc{o=p8-h$r`&`K3m#H{M+j|m
z1_O^rwm5{JehvDM==}%vQ!cX?IOx0=_P)r<a^A^m^zMn=CtFE$28vV|kSruP_tv@R
zh{v8Sj;8Ka=c(y3xHS9?6=-gs`^P@g{bd&-){BBAj{aALHlHL48=q{8EL-tVn*HUL
zweJHS1Vz-Z&FySAHrsDUNs~2Y>E5@lNpt+Ke1(?wUNqTV%KY8LGM;0F;r6{%)7#b|
zm*-ez)H?TB516VpjTJ?gtsQb)`J16#-cibuXrCyupq#1wwyIc%cr@0!Y?IfQz5{<^
zw`Y`lV}nBJhNIc@j+mFq;y*G|HuE-T<En?h4T}}pT-fscLQ_)SBeAdOrrB5VreBuX
z)Ss*LPBC3MNhNrEWWK2@;<fsnhYGEdm>iCcJXF?f)Rc<rZzeS<+?&(o*EaR_>TGFZ
zzvuF}T?cE!kRaz!%BR!3c~>nzB#ha0PItrz>2Hrs=)I*n+f>LZQ%q&}_+LT4_<cux
zl2@;8d=kPc@gXWw19Ia>kGiJ5SlF{u{7Ni)v-|Kw*q5QF4Ps|?iDMGx)Q|cy{F95n
zqnf`Et7nzNKi`X<vs9*xZt<?@`+D0kV!bAlYI!Rz>cQiy87=b#U2k5`$f`eo^eFwo
zU%xLC9)>C0D}NnHSDDC{e$k9(q<FkP{rSue{}#EhKMdRNJ5DvYFRY9@GAwYD#Xh>E
zxwT@LV$_&6no)2$_L;{4@awmqFFz@Wx}xy?M}301k$skV{!bk8YQt;w@SJ)QZ<3Vc
zzGhNorS;i~148F01ImZ$|GZ7%yLatPV9$JM;caWje4Ci|hQ6O%c{+`Exhm4kg|2T;
zQRo$-6Kl}UIy$3aa(Fa@vvJfTXwPh-x!1pqBZlJJWQq$XyH#gZXU~n!@Q==<Ewry3
z+}NAkY&2(QZyrN!U)gY)Hi|=)cKV5it@-xT6=6lUpTQmZi<~<l>QO~sLg&=6qdkfy
z@4T6Fsl$HtnK#c0O#kUrx=5!a`9i(!5`AIIMyXV}rBlH_N`Z&L1{DpX^&dAVMAn~*
zEZ<3_l6V|P!YaPF{r=2y(|)s`G`$&BseAnA?ajjfc#$Lqk+5#0`2>+99)7GWt{ITp
zwva7)Zi08Z>g$1&!x5jKC{SYR&#6pRKO5s%-P09$$1$IO#(#_FFn!9nhG~{)%lTOD
z+4!`ayack|pQNs_kNhu6_l~IhDr=q@?L2p2BQ2S6;Di+U+&Sv~6H>2&A6_&pP8s9;
zE~^rfRi-X==C6j+TbY!Y^sfBiMA1c|qrAIzh5ax)Av+?+dlOMg;`GGYX?F}SrH?{+
z-+`34t(DT$TFK^WyP(f}Cvr8cI-H!NTz_Oz{kY3J!RvHeL&;Y?mwY*utzkZGS!3t=
z)2ZQ@BR&4fEIanIm4q7<_UxT$8pzPjejai2Y<Sa0`2ls7Z=Mdc|1{>$)R%r8c1Q`h
zY<J=)9_u!oQXIK%cVt&|`Op~2Ct+@p$6Z;Iai&f?9;$hVknG)d0f(CsbyxXZ-#4$-
z12-td7UK?kvO4jreo0F{;(1GYzL9n7A5*=!z-8TRNrRHuN=Fq#Wes*=L(=WhEiw;t
z9~_mImE)FU&k!50?b!1$xZ9`jT9=Zg>cmbhJ+}}*Dt^yN+mloWBPzUIMBE-L=ML>!
zW9ppNbT#CRNn%z(rxqOjDIv2rE#2;t&t}JUCS|1;o5#1Pa>^C2nD04p{KT<{V`2I+
zwN3qrgOT&81J``Wx5&S`mkYIYouU>x$I1IPYPfk_7U0d#$ll_mrSjpwt<T3r<GRQX
z7uLzdsa5oF?f4;Y!8*m(m~#H6h2+uJ7=<|D;GZ8fsOXsoFK%w*tKn}tA+nGjM;%Wc
z<$dnf?eS*<d+*5P8w%Ylxw$gxUSxB8i-mYz0Bv;Fov*vr;))r~B#-Rf3~l#MPkk9{
zGIwe<DLmQsRWFIWw><t&&4}mqPXqr;=Rz1>$c=ojKjy{JcqDZj<@Q6=!-t0Nu5QT_
zFZNuotLER6wL`_Hr1AVf{iuM<xsyTktj7zuBBfS+gtTk|A8qUux~|MbcJ7!U*Nb;e
z*Ug#QpC{i`GN=AS@^~sMA~8JX5FXK{cID(ffo-on2v8o+=f*f@JeCyQEYGVvtlGy$
zb?|WI;F<jsG>X0|<b7*4pIgV3vrLN3Xght6+@jt_Cfu07K^)vwqkBh4y>@<Yewnn?
z`!j}CZ|{kZ^mdsm47=v`*}U&P;{98aIci6W&DEZ)3#YzbJM($ctY2y%ZA3eGZFA8!
zybv4J$1ZSpcSP7$AE6v%Rdw9+cW2Z!8@3aIqSXFZ%wA0#d+xleutvEcF=pU+l{~j%
zzg*w`=9=c}vm+ta)to;$pL@M0ZEv+!2@B!)=KXE|!2`1I&#nYk@i>LwYMnT^!p(a(
z>ESx1B+X#*3w4|P2M)5_E5EjpYLqt?zvuGYE?E@eG7=%`t2NgYzs|C-d1GPMV+n>G
zoeicIoWB|Or)Or%)w8gXaw>7kpOEjnT)o?=RN9sgp@qBaOYVE*tHh&Ei|x(qMLX;M
z3LO3Ve_DI*u%6%d{~yubTcSNs3GGCC(bQ7Wq@}b=X4<8xM2d#?QbbcjWtX&+q#+GS
zSt0bjUtXWj_xSz(`Tg~KAIIxB-fwTOdOjbI>pHLVI?wz4cJ5opm({&d$3oAY>ZR&m
z%0dmN)e~2pdBsJ}Ot~A!k8Pq?Gg)9RmZFOGyEs}<`$xz1LZGGGO|50|i}$(ydOW)4
zI&%0o<LNBgNM>fK<(lY&ED>fxKPWF4N&i?MY9t+!wa?@j6}!m!6G_u&E{0|=uWe%S
zzrWzp#58k8&1KIw+xX5r@nDx8n#Xny#ths63xDr8+U%@*Z;N%RXd8Tm;wP1Y#cBGe
z_bu5xi~T=eD%{-pXVbUBlfizD>zt^+&<H)^em7N`l)0UJ_9l~uM>U*^d!MO)S)_io
zv^8^tHiu#S^NPUTQf0P)A;Y7MF0ETB^xoX)?&2us9nRF64)u}ovSg~s7;sV8#Cvpu
zjBmbrI+eiRZQfT}9ez$a$cL7fJ<+IPAG6G4c%OH!@ss#Ij!T-K<Y>(8X3uH#FMr#+
zz>{FK$k+MK#=*GoS<ytz&Z4;`GnFEx<=anBw!YN+Jg6W;CE5_P&SUL0Ke<b0shXqr
z_s~jgNFCO%syp6J`S4|<p<jsHj9%WDwkDM}8yj<&>88sjslPqQ3W!}78gP2Ca#}~X
zRPIFqTc#D2^4`dPA%OxmTRGMYrU~0iyCim97Tx^ew%^HZrF#tow+UQ&N+~P4Ywyit
zQU$DrYh$-wW42R~nO|q}vCDm5!}G3~2T@SnJm-_;NbmI&3}kzCV8Nw<=J-_F-^#D|
z-^|OEzPxy_AYjhno>_>e&tT-&`}T`|^2Hnfx*nrrSGuUqr1x#rHE=iHZuZI$ckuh=
z)9X~H{oKdn&kafB&bU9cuI8c+&2G17tD11)l~j7NG%<NIwIInlLqEAZ+W4?|Rcn>-
zz<y`dytOh69{O1R>v<@eBoCX0>Eb<yi!xu&KB?q7!o{;3`b_24QPdN~%dW1qyco(E
z*19~;z+aTGqv?c)c3g$Y@s-*2`={pxma3_h=%YnyE9?wx8OP+9pH$O)>s7OFANpif
zciFt>;!XK_zO&N`Ec;y-O0K*w`&#sM*o)F;mXFC}sCe}$vq+t*$Vo4aZBN%lamYk9
z=v6tc;)L8MG*aM9@4G4PQ*lN@_l+$apEmw3+57c=cF=02D{V<{h}Yn6zSbPYM>0oK
z9G>v!3rE~+6Gre6E!{)6uZgd9Z)^BUWH0uYyvV3#+#+Y{R4U6n*MSFDB1bRps5`Gq
zp}2!R+KXwhJL-4dtw}bkZ|ja<*|7OjWZ=EbgPpefq^_t<4ku@swGZS!{Q0x|jm#$w
zjV}u(KPb6@nd8g4U}$L|pvk_t^{Ie(%NhrRt`K><ET+!6IZaEm7sb>4@)I(3;?KIp
zgcpq5%G}28)LmoGy4$-NWMwGZWWGyerP#%9CwqQq)3ZOrZa$j|axFr{udl|hvmQEr
zOX$&??n%EAw=O5$&~q|Rf11thFX7HM?rFE|IZ(L!g5kNjrrWEBe$Aa;GU+}itsPSy
zF6hDHcGOLnL%6EhexF;u8&*4GIt3mALd)cE=U@sB1!q;^Z+|}A7_dt|l9>=YKYt&2
z+vgL-uh}UZ*B$O`yS~HW1J$i>Dz2yK{OdQVZ#pR#P%D%0?#hpocA{c*lIAw5>>IL<
zwDQm>4&@7<K62m#!$t+0^?vO$g1eiOlKv=n*MGE9Q!AXqsb;1{(I1*SY<yAiN<c9$
z!>Dd?Wu(W0UKe?0)wSupTIj~-XyqKSse>k`Ob+hptg%b2)B+`eC6>QG>4{CzXDt-A
ze^{HfX5YCMz9MH`!&CY1Yd-6r(L5IP{Go%k$@6H_FB^X~r|K83^e*Smv{;$sEn!Ji
zx)zk$PSPkG6@GG~Vm8r5%e>0k)2fWKjrIYv%Vrnt^USG=Paermr+7Z0+`tnhQ{d)$
zuwHKd!^T~kKKccI2yHN<Ftc1ceY$6b-+8`-E><RKPVna1uOYlrn>L(kzVPNo+k#Q{
zbMIZ77^bH$h+qELJ>h*!faV*8EBAXl*|#j~o=d&hXztjc#d<!rO4qn?XZoH?5h+$G
z0uC_!RO~C)$k|phJ-#=!<(+i0X~+efNa8;E_KQ!or!N^;oe%BIVQ}L#z0k}3{7~hj
z=*Y*hb>m%kn!=eEzIeE9Q88z&uD`i8v&4M;^_>(!bu#+;`-=~X{kE+fOZ+ReG(DJh
z{czx|)H;KwZsjs9<B6?|84oV>xOG^v)Y?|)TZn8^i7A=3@V8&o*!_cL$|h<r4s@2D
zbZr5jAG3d?XHKhfXNz;u-rmL`^;gP)W8LR3Bg=LsKlUBB@#B8-d_5Oe=xw!cuFO1r
zZtIdg|Ng0xGQAmUCVw`^yMyAUTw_9$$9)0k{wU!Yt}%}j9aAwupGO`!$i&gKev<H5
z*>IK9TGB`;;2dL21;3IVTb$9W8xIl{9^ZAoWOBVD^a90v*Bb-FrJcVuIF#n-`@WPK
zl`e+fp*dlcm!!~k(P6!0Ba7pTTDMr&ub1H(gy)!X&+F{KO5gedM>ZHB+C|z=8ilh#
zzwO#np!GaU!@K0f!$ZZ14^8sz9$KUtnHW+g_XM<>PI%?sPXFV!KI`!tGnd_U@p`Tc
zQ3k3Ow9cZIo7>J*h%-9|)&A+XU$p+%H#n7B@suX&Oo>hy6565jZZ86ZWqB7TG<bAn
zP3^01FIjDQ`>AU}^)5}$hezL<{w98BNbu8nS(x;9&+#8DQhcXM%)T%^H=zw5G?@w7
zukiYU>En?zsg*&_J@s}@zhB$DvbZ17bTUd$QZMJGMA+xgdwAAszPx*6(=VC&>$PJF
zJR2UeOxRV;&lSHcuc4<+XywT3p8a)P%BK9vtj#8WTV@8@b6?ZF->Y#<b-Ko;=?Y%U
z)cC;|)$!)xq_0`r0rrT@|1C+l?7P`yZ6ckPnshCFKib))YGD&{{aJM5Q8Ww3o^~x;
z#J!B8apPZQz5V9#ue_fU`;wk_I&E;x$kuT@TtM%xUD+P8{?|;X#S1U$o5wGF<ZBxY
zYKXeIEu@4!jeV+b@x?*U$88at<IS4%OG<pi2cn!#mMi={crNuQtxn!$mW{Qcp&4PC
z>B;Z8skc8l|BTnzV)Cl&j=wT~NBlx-9vxOm|9#JA$mEWn%!Syqm)$kL1ZB_oGaENs
zC!Ug`e$G@FGuQuOg=X2c>8MfNHKvAy()&(!>k+e=${4wndt(pzNMB2%^7VGRaJi>-
zPk;eV@ACak+08SL;y%1T`%YAC>2XQkDPwQNXPFZFC!L4%7}N5r?n$Hx`-vLn?Yn>9
zM5ZiVwSlq6Hit>|oJ^>;4G*iWO!W08`aaK%b9<vcNGP>Oyx$UN(|j~QB<7u|PHp3W
z|CE|v&hNTgW}9tycB)j)OpA@w+&hwJxvKa#uTP(XbU#*?&#mg9;Rt_=I7-~-8z=(M
zQ-$Dc{eN9b6oDtQf<_%+1S6Ua@Z*5HD{$|1M9yU{bZnFG^wmQ?{MQ0>K8wq|j-g{3
z>nXMFsnJvDahOg@{bn;(_V&U$8y3YW#rYiyEej)udAIM49CB-&^+l4e-Eb^y#;#Iw
zNiF{IzGR(GhAOMB-{#gBoI3QXYER0EK~C0dc2Qjvo7lDpIGO7S_ZvnjzY&@`du43F
zvfejVc~_f=Vfk)S+Z_eni#WrGnj760NOk3L;$3>&jxGiv1&i<4;R4Z_|A&Ei_(6!?
zg0x3b7F|w%D924-G&huJ9yH-R8X7co>Hf9P&gTLPI2@|$7k3u<ALr3OZYF)|Fl_@P
zP3CYhi<*mGRn3)))%3Jw8#Jjj{~gwH8`agV)){DAGo3biAhztHVDw=ZP2AVXk_NWN
zhEhkLGKu{dk1xM7vqkIm_Dy4awtqgHM%&(+<pTU$l^>GaeCiYYetPz&M)DZM4ai@x
zla}C__3Gl0{&?}rmB0U6$)!wzg&b8}^gRnvh(%|b75WdzbRY*I2+h+=9b71N!on8;
zt}<|#=v-ApQ4QsM7o^;K9&VSEBpLuBAtQgxut{5ApVn~t?Zbnn#<!HpZ|2`LwLR9B
zo!VPo5v$C_rSNP6uX4t@BZK11ecy_oX4i+gaty|LH+!G<-;4Xeizl+N1|96NZTs$W
zU1*z7M^)0dbCYrZb{W1IozH=6$^*RbAKg7}{k_=8sxdZ~%e8CH%?7**_|qXat#Vf!
z{b_8H1R@>yVl@IChK08+Zfe2JFM5e4%W%5pjxl=#daLMsLJO@Gl`D>_EsDw@?N%r<
z)E|4Xw82r%P}}av0quq;pQ8KMtFv;&p;%c9v02HCG&enr&M$9`ckH%l9@rR-mZZIR
zLx|R~LrWqMEu+&60}1{MAe8t-u;6h58H!f)Nl05~N1K!b=4$sZq6-JhC0t_)Xp5qo
zjl%Aj@1HX^#mpCsMN$jv_iBZNh#FiI&CWcyd9u>P7z$HXDVdxcnvW+ECBg|^fGEPc
zwYgOJ`}o)m(@O(*Ut6?S*qN$*%6?G85aV%>%e*kV?r&BIh|_LvTXS-9R_7E}BV$vt
z3f!Pi50UKkSKx&#KKNgi8r5o9QmuBc$?ctOEYL6+N#@~Y)QR_CPG`{Z?BjkjoK79c
zITq{sVfGZta@U?ydS@`AgWTMFE`IhiWla~6&^_wz27MtZBO?~#`!y(R;}H`h5sY1U
z5JZ-Yb}2-n#^(J_^l2VkaCt%)$8nyBmC?VqdNuUWvwG<G%$M6V^xqa<UV-DPuhMR{
ztYjvgi8g;|DtX>Ix$}B{-8()XpP_7pIk_J}gZB4cZ<Ua^3AYgWdYtCwJu>l&i+)+{
zmS?bB(ar^b5dN0J-r4>;Yp@<rHLje)8+>*AE?u6$9j5kxg077hw5%mK*|ik&5@5us
zG;A;Ta?c8ds+6?X;dzRJyamb=D1>E!C|Y_Xl=uQ((f?#~(C?DAy+>t}HcItCbI@;+
zL&d#&&&BRp|Lb4U3*9mXo4vc|Jq(mT1X~Gw`#s}#YIjCZjPr*?F{?dJ%98ckJr1fu
z>omG|jA(f3#trA9iB)8vv~9uUgl}u_#127*)hx;rg(d-zECc67O5OmY0!=Ii;t-l1
zXoW|D&k0?Ih^#D!xj8!9Q0x)#H5l{Mw6vcuzDlW$InLTPw)9cqzH;Knefdd^+Y=f8
zIJxzF9O0^dxyj6|X<7Mr03}COdO%;nogxM$;dD{F(H2j^rKNMv${47r`T1Mk{Km>C
zIZbH{JzjHJfbPhLdNKbJqX(P;3bzC<9p3)()jiS8&8igB)zvp(7Lnv+q?#{Mta1tM
z9d8-;t$uy4&x1$C-~s3y?XTU&{Q=Y*`d}GoxI!Zk?aHfO=)u$fC|c!lzFN5qv)=Qu
z$;v{fRV!BI!ReU<+xg4AQIxA3w6u(cVQuTUSXcr|+!R{t4(=T8bi50*xWaIq)K*&q
zG*~`=$<ho_l$1C|H}&2Z2y4HA=R&j#kCw#(r<1GYZ!A1ui$1YD_~5NbN;8NPBD(gQ
z?UZacpXyps%nW4YyW3N5KFMPo@SQ!KJ>`a-s`@2`N;mI%USW|#uD!zzZYHMInqFR8
z4A{C3>#S_wRQ+d~{p_C^euEu}W3f3)sM;%vNmU$@OyTL6>ihf6cx<jmps`}oQJ3zm
zDus%d*ArQ%;iNN-sU4M{_)ofc$!<@zeMSoUQa+A01LMCQDlxyU*n-=#WB&!=C;N|&
zUKaVu1TBCdge3OaD~szXRbE{lNj2p(O0Qvg{9o@NINFV~lOZukIOpIQi!&MRl{a{A
zHd+m63%kG{r72i^A+b5sOI}pk)z*e~*8z#MPE<PgrfmH$jR~)AfygW7-Nkm{NS{A>
zo85*LFTCa4GBLQix!uh9+g%Ag$$!H$e(t|`Hl+T@+wfMlVUtv+V@n^D73x(UzOELo
zS^w_`@Wr_CJh`Q#{9-;<eBg9oGzZ`2>@wp4ZK(j#5=gn-hKl;rTfvlxCv@~G8svPP
z_sgl*KA6&#Qh9g4A01gkO_A_j^HmA+zeUnySA(wfQD|uf^_j>kJ$n68n2Y-m>DNwQ
zY%Q9^6Q;lVA|YkZ?hk*{wFlJV?j6;Wk|urK$0ym+n4bhN-77TsM0woWKXZz2zEH-F
z=YQNDpDU$*%te3V)E#`yV=onW>&Lgv!k$ms_N&E;a4`>2kv@y8VU)I+I`zj2*Uqk~
z;^RMlQEy%OJR7L74*RuH(g?oke@~Os){w!x^2JGGUJbt4^L6cA6Y_!H-WsCnm->_A
zSjgkJIUEyB_r6|dxVPd2g`Cvh&g$jwCYc`J@jPY!HGBG~L-G;~2RnV<2z{)|;FUOe
zhKg#y<<b4*IO>1@zIO3j@hdrr{@H)b8HRp)?%MbK^>z;-E*kPZ_2Tp7i~Ji!6CBu-
z$8ISe`rK*wz4Cd>)Yq<z?<sR?<RjME_jz)+Cq)f6cnH1S>@@VdiT!BWDVxjh_MrTa
zl_sv#r2lYjJok53oy@@2l#R+m&2pO!68e61|7(Z`M}#nhaqr66oK@l3{%IfY^uxMS
z;;A*stL)_W#|*yMXf|xe@$_3_;Cg`{f$t5!SL}^>pK?ZMbpv_Xtd(qA*dC41e4O1Q
zolw5t=+@VdD(ucz8YTCCX8PB-s##}l?mAd0Ft~336G47>4H$keQ;HGQaqqYOw^*#T
zf)DVAb=FK(w=NyGyWIA&?$jeJ_T=Gz$Y0YBFJ{e$zk9YYBnD`jaNNIEhp5x*tvIDy
zJ{(SLJ*FNsakTb}KLTxcmeDAm(>pr*x?yWlO|msi-}rw^P0S<Ynw)CKV-jW<w!Q5d
zevD$9!pG_p?XQunvZxa$AKcAGD+hXIU|3Q^`l9~ZB<uO3NBklpQ|n4nb%nW7ZarE<
zx%buOia~jY<w~(bzrK5YeyQIi`5~oXKvF#4jfV_#fR6pavD-VL1w^*A;e79`GX$4#
zbx!&#5kxl8$`Ii{fu6uX1@FWcj1u_Q*w-`2E}<<C>16JexELRou=J+v4c_-<M$3-e
z-oYoI7qtwKkb*ZvkXuAj4O=0BRKp~vQ`igXIRQS9PXHq{w2+Q9)<2t9SN0Kx3QWU9
zj-wGtFQB9}?03MR7c^Oc@WKJpy6>MY#bZs6HryU66fG(nvf**xZ*4ZCkp8KvK*r2_
zYoS&|!V_ES%{kxx)V&WK5cYiL_&QNc;pD9=mn{d>7;3$Fx-_+{GiQ~5`Y9czhi?Pm
zfLus^z}8^l;);5H3?ZKx?Q~oMAU>3ZCJf{P!nB5)A+R9}$15BDU#M{i%9vXhJ^+YO
zv*Ap@xAKFLh}Z~t86O^+B`(tcYZUl0XE5Vi2MN)VyEW(=5ZWKR4U|XIWs|<kcOY@N
z{&n$K!66Ha$dVE`i0z~|$F6g))o)v)GW+_8)gPao<1xn8&Ycg9KBeh4!$YvEPWdMt
z9bM_o2w*ZHn+9c%uA;P{D1Z`<FO{ne1$8IZgkkK>?jE9B#AqIv)d<H06!0}8`(|QY
zaO-_qf&n%EFTFhE4*AosRxyD?0306hbu3@YUf-@x4e7rQT=@_q9O=b}Rt%iKVo78T
zsEde;i}ORV?{T#NJ#BbxphoQ6K1~&54FN?8NKweG8{PS#<3=6>sjyTnV@+D@fX-w~
zB)y!3(naxro-@Htxf2?O{V$|4IkilPXP)qu)WWtAZZ(i0E?LXJx$wovSQY5AhZ=%*
z0IP5lA~a(7Lwg@8E6DV=Xn@!x%Fj=x98|&7jrlYOP^W^_^eeBC2|MK2Ywsx*m&E3Z
zyJWEt?a1(bQD*gEnc#yup1O42tIXTCjH(xCeS8VSMf4}$SOdGnTwxU8ni3f)9No?x
z)%e;;x3by%-A)&Mx)bM&`pL>hDzoeC?_lCKz-oc6`WWQX*f_i0`^3pu8faVK+9b*q
z+!a5+a!T9@%mNw4;*_YOt*aY-Z-)km6ZV;m@kzhV`Y&WW%6WTx=<}IfTTgwNRczeK
zvb4*nhz6P68lrH6ks$cUGDNxe&#bH?#78g#%#N-Qjbr!qGXf&|!YS6#?^(r5)Z@HN
z+GXgD_NF-fmUfZ<SmCZ4^*3JHc5L`jM^}gWjT9Vn2iHff_&lGH`&wVKkqYXBkI%B4
zMbd+92DAv{Letlzw`{+#{GOqhr24ZGXHLo6oK;A8Gxzy!S=_pRr)aL2!bsxZJnJ48
z&E1TiT@x$vK`AHiyn0>z_shS?nZ7<YH!1ieend}9(yzbx5lq*b_jh^)gxCz9<NWs{
zQUp<j{-bj@_wC{5iq#_5J{?;S{9sje@8}<~|6;#hMqt8Nr<gJeCr1k+>#N3h+;X?e
z;xp;~ouMN8?XT>xd^?zOHt6oy%y-`V*J>_(eV{dZ`P+^EPL_9iGiEn7-IzZR9Hx3U
z#D!@S7vF<p$u-I6axDyn5gSwJ3{;DlEZ@9Isg(;Dy7P{ERxiSThA*zii0NExp$Idy
zEY7E?_x?@yk_(wVQirtW_r{LHn_>*9af5w0bms?m>;gT5th)VTkI%O=<wabZ{Ko?Z
zSpOX?k8dn8?R>f1_5PKFVfg+QyRuySuT6Juzu3PrCxRsA#NyJWL+{=S0R8g+E%HjY
zW$HnySd1&qefp!GR$ZJ$96UrB;%#R(vlwXI$GN#iO(TL)OYzIC@?r08GqLp4P~7B*
zHyzeSsllOgD##)U0E(YKJF@?etU5JZ&P#BJ%rUheyPf=REJpL7%o<#*z}UNBSg1k0
z)UDjN29XZJNH`RleE-ryO$am-nb#$WJD0#e2^{2wM|v?cMjHVzq-nWY79+E|x*+~M
zE@zW==f#I_|AV8KJ=6bO<S0L6{1YG+Ov0V!5^xC!?8nZI(3WDr2IP8D=b!YlsNaHv
z6g%pD4Il2eP(2YnVP!#tVWE(aP#8W8nemm`MaSob$OCl~<m}ndz#e%jZEh|Ck{!06
zFI23-B9bjxJGVesicubpj=7fpc#&{QV8#k0z8$4*$*HO9Rf%9|1~n0gzp!PSa?7=u
z)ib*%Qv{TiZHTsY^s<0<;ucBF>9LfBtvUvt%TwlI%<5oQ9U|C}BmKs<wh|!cd@yHF
zrzR-tB?Q@%Hj%(ygjzKQS6Rx=k2cw0!c2cD4@UCD?YPfuclKxgy`!1ir>8b>S%%p%
zjQJ}yS&LA<kYRue?ZNnBNy2av=a?bWiF+c-yo~geVV1Jzl_lLZ8dmLUe%k3mHqEg-
zogXgPRr%{<mQn-Pe?t{gFc3@@pMhBGP$F!~I0QdEE?kWnoMQ_Vh>2us%aLAOU&zHB
zH^MI&l-|o^1{o_C*R=U&j4wiHo*S(EB+CE;EDj&yZLD-#v*!#kU>)iANyxOM`QOQP
zPgR}SE)%KG|8s_2CV7*m;ef!CI%)n8?4qaOMa`KrK%hN^5fNhY@`I&HL}W{xXD7|W
zpgM)!6LTs?Jv21d5^BocA7d1iS$rd5;zSb&?d*sCwVQ!6Amn7|*Mbego=!R9-}w!f
zIy85cvx{=oOzX+mFfukqNh#yqJ7-9I4PiLi&~I#E(Yd>Oj|OCYusRCv-TU{#_2vr@
zFB0*7FMQ<$=mbw^NUO4lfhZfixdb;z#(JTsK^O9u7A!BQkn`YAhxCoiG{H~vQ`7l{
zMM)&yX68o$Q#i7zB^&t@2pc$kh)9oxr8RZyL8%`_h2+cyIMW+I5<g7qV&UNlQR}vA
zCGH0#X=7VlcSDMf6R9aTKO^Z3|FrtiNlZEsA|pD$1So**5?0ucWiBPFHrqtsfy4tZ
zD#^rdoc>G$42?4+UR;`C7+GNZ2|9>cxbjZI&rZC)SOE?8A}1Q&BrvzJ?aYSwOTxA`
ziO7;+Nj}1ib%MRTHi(_fDcj}tQW!_ab5wc`f!ea62fI1)=@h0{B!T&e0_RUG+hCw|
zh5($}(6a^<SXe}%6Rd_ci69TicrwiB`w0u}jw=bVGb||QLp}`L6A|;l=ZxZ(9Ww&}
zt6x@;LJ+hIQl%DGnUJRb78~0XO2TpYcPxU_LrX`80XXkqa{1@ihE}o~%ZH~Q&zRI?
z9|{4a51F-|u`x5IP94LTEy%;8VJ=vND+4M&vpgPhvO+X^6^=6ECWEjZK94oTXAIdQ
z!bQ<ehB3KYvCn)U#=~S5SPKZ154vs1^uvK0;X&eoTw1n-f*;8bB+O)753vW{zmE)S
z{L`mySet)>g;nA#EQ<%>*dEP7ij0Jj1h2>C9d=!Sz8PR{0(_n!@$x$|4v3DP{$fDb
zF5Q`2aBvXfBOVU0qWt@QFH&Hlg?0cukxP9Cjh1?J_yoeoGCV2>gqV=F5G=1ja|$7>
zU0%<caArcyCj$QI>!*l_A3#e&lZ(xj2BwDC;f-1+O(#-z6OHZ?;CQ~6dk2dI$=UI$
z@tRctG}{1+8}KHNySpn*Hrt(E+5q6HJ?tlBatBa$-k?=1LhtQ|FiJd&$QdzGp?D_)
zE(lJuFY^ie2HqQj_abwIh$k4xD{P(ij*kA|<`=;C=;`Ydb`wGnsXgO)Jg4G2t(q=%
zm;fPI)ccCAT9Cy>m~)tl1$!hxQo!!c4we*RGP63$V;*K2@CYOJNLb&A#v_0Z^kEd=
zSzThU2k8TXRDcI~`lYdG5a)=05F@b=6o||p4s=yiO(E^oF-hf-JqXhwfdRlXC;#^J
zZgSD_x<`mei%@F5L<0`?8YSGFOq_UH{wNP&50a69c;k(~KR+c3W{{&2;2VS^ck$b|
z@}9Tuo>M8;?!tzpasBKsHzc&U1RRAqoxtC4e}em74~g=lr%&?$^^h&eLwTM27=p{W
zdpu_7IL=4nD11(V=}!Wpq6M8Y1WZ6&d2pTwejLVWmxne4F&u$Q9hbJ<64q!n>_v|Q
z6slkWkiAZ3c;FkKUio-WQ&aN=3@bnf{lI%8q=yZ&=}#~wlNjY8>~F+VC({W*+xdbj
zAgI4YD(k-Iydy}gu(&*!@`d?>p5RG96KRL-@F(hqknXF+FYiALdk7J>KFd<(0;Lr-
zSvbsM#U%0cbq^d6R#2$zlz{_<ECsm7TQQgmQ}{&TfCo{50vdP-!*S$^AgUGhTwVN+
zO#_=0v8L4GxkBwt1`WY0lZs3hRB6~H%;MdOYBF8uQ&-mqHXMd=9#iEq3S#gEa9LEj
zfmLin&YA0jItjcikk{7e8W~kuZ03y9LhTfP7en`e3_CSpt(axFgnT;&+%0N>aM(qO
z+o}GFa!>_E-}D#2!x{V%<Mw=gefK<kTkDgDlcWY{rgE5TZsC$f*v4%Ce*XUbrE-Xx
zxb;N5@#0edoy^G?rGaUG<k0$e?4ynu0|l@pW>oF;dw*aF!7vW*!@VCoHz^qdHuc(1
zmynv!3WO7DVF#dEVn`dd!`j)73syvdn<FeIj@Vt-KUFCy9JEIERt3b@65e18RzeK8
zLrZukv`~$cXhLofCRp~KJmKDj5l^RB$<L3y^8-fY4!i+OR3xu#0w^JNJU9e{7|?B6
z^P1|5PB+^~aH7yJfN6q+_o;d4&{e#H3H%}sn6tRZSB6=L75}}`f*}pJS3y}>bL0bY
z2)l>Cy+4fkhk#fTcD7_Vvm^fg1de%DxohQ}{<i#GPwAwrDKHQ<32wB^T7;=~jGM&g
zfkIuc%;ugU^|)LrVp|M6Y|&gtg4HlwGDq9L>~MnxA&pI7{c=Gp-GDJzY(RxAIX?qF
zj_@2Dai%#UVhbuN)>-e~Di98@SP?gdzy13i*w#YGuQ2V21+&X4s^0($*ntM!Ruq#<
zzWwIFZ&4#Z-w`LaBWzP($;839O+b*ALtc376JSKKWE7a6vc_8_v>_kd;bRbQtT3=o
zo%}<>UYpmwIOH{CGua_5>4Jozg?(|)8pk@!D;>D`cEH8Zj#w4Z@6Le`pA9S@2O0WE
zW<<hf$F;ihcC{UtC5dq6{_C7b8l4}r`oKXM2X`}}GsK`yvJ+&)UfLZn={5d%yZYDi
z<1|;B=S{$yI6$o^3`Rc&bD45p-Es(n(cd)fBm4!BdEWu3u?ctC1C>hfd^rMnG6CD?
zS8gY*#TQ6ms^BQ}fgvys!6*g>qgZfsEnnApphmZCH*$a_gy|T#_G2*$E)F|Z3yq_b
z`r}9pbHR_8fE$wK?ddP^crm4c4A^S_9w)#L7`(|b0e%rPI4kTp`mbL*hX<Ss28$zH
zUyevM`;U~cwRv*l5UP9Cbm|^~qjVtH3*#D3#%!unn^;+i<B}|dQK?&C`zUbV*UY@Y
zw@RKQ!b&^#>^8!StinkQuFIHS99~--7SH&_0@K5|U_@gIl?FO(nuNkKard<s8(f*-
z?b%TafO8omjsc(QwY~ZE9`@?ByGxOZ+L84JLWuQ<_X7r!wqC!3M+TngLF(}?+r-Vv
z79X*(bTaN^yjTa^>Fq8z&WI8mCT>czdUOIKBJ<c~u?Uowr=AxQ20<*+C&E5i0s}4<
zuM*xNVbn$Tye_k<QXe_D`iHoeWm(gH!FD)-;d1P7z_G!Z5y#x-Y0R;Ffq7wg7OeB_
zgua%XmS+Bur}0_7CqG&I2;dXQhY+VwbP@w-2rK>HAHsLTE_clUX@c!(<d~SmXDM-{
zv37X>Yn<2QwU94_6VV4pL>vfBh>Qrh6@_Nz4%={YkO85U7L(|HaRFk(4(1js#)H1l
zV2Z(cgFr=sAsM|y-Vv@%|1K{G-{g&&pS{(u=I&mC#3=%?m+RQ=Lj-Yyy?F@H0cSn`
z_23B_fpBFCz!tCKaO41NtN3M~#pe09c=*;e`p606kgSuBb}D1n4WyAv$f-zpgu$;J
z5V?|XBbcd^AA^J>ngmV^fF<BHpLdT~@IBzut-_I?%iNabu*2qVnYIjmFEWu1vR1I?
zHSyudXIF_R#H+I)j6FO*I6y^V(GCLYe17YaB^N>?d36YnH2Co(ed^w|mMm@HQLZ2O
zjkYj5o|7YTIr2|`iO|-qdSpVU;Eo+e2=oqBR{z$g=e;);9SX?&q{xpD2(RVNDES+^
zUp{jMRw%Mt$t+APcxbtJ3E-K$ENX$7-a@955^&XFBuARd3MM2lJI3W~$s~uw?|w$e
zJC}_1B3~%LrXc3vH3aYi>k<-K06>uc;UL`zCPYmH<y#o7N$Cm4IdLq`58Wn4EMf!(
z78i1bHdh(+G%w(HrKJJdtc&|DYORJfu4|T!XK`>R!{q%1=Yc8QFWA*T@io`t7?j5T
z$E^l#Zk(88z%>v~nCNj(k_UBAGC~X=QjVmt_*wa+eEQ?ZB0l>X>?`@;_gB!a<hg3z
zh*S}~bpwH9u^H`AL)L-kPIx-&M%*yOSO-Z3V^p5D_80WR2z!uRDag@E)C<Tn&^4UK
zWJgMZuw8zpf?g~7-uUXX*mO0s=#CPW5Me8Ub+9;KbArq;Mi-l8#z^AmxuwtJ*9nG$
z2?2~KfePVfR>{c#C@%t}H0F+;{rl4eyu%A%z=Bc*n<+DHaX7bdx5j(^8q(xaTMd8)
zCj%G5^z5FRSODH*a2vvM#RSF#Fr0EwyPd$NWkDJHGJbD5_fAL6nxF(ruubA{4741a
zyMg8Usy?6;rOlgU>Tx?cvp_1}3S#G>k0SF=pe|`YXdQyHpo2iiFnpn%L_pAI+VLiF
z%@EQAcs<??J?RNRj#x>&#fHUII1ZwQSWsAepzw$VtjhE^(hqW?m+vadsYlI*6>pTN
ztZ-|KO%~FDXz)16T71AEio<)41JXX0*z-GOkav*laU4QEjlhWPb$rWMB;P)`zsBG^
zvi)~ZV>o1PV`F2dKgvbXC%?Rp!QaSn%ixxEPbJX+;7(Q&-jD*bl1rNyf?5?%FMoJL
z<X-R(9BvP31_s{cR`k7zui^-|*96`f2dcU3z(ti5ikh)uKg8nap3KNW`Q>o_)kT=?
zaxn&ioI;A_Hx>ooH#Uy#Q@l7h<!gS3%LzW;>nUXD0o(ut2!eur8W^c$q9(vf3GQ}r
zwbg{64sPqmLq8tR`m<tIA8`*W1#AOD0qbJ-+223hw>wx4@Q16Kup<$Dr$+~`C;kpf
zSY)&#J`J(@dNq8!>(0k&qbLcBYtV=fm3Cgw?rTaC5@b>#S>lBM2VdPrSn;7i=`YFq
z?066w22>SzV17oc9jz1OazJ?g^WTH#julEuP5mV~zSbzgXkX+G?QzxYl>rnBSPuvZ
zZjjvxn&%Z5$i85rhsBEZm$U#v)&6Hmg$<RS9(M0Ya=nKw5~F-T7F4fsKff<`uXwCF
z-o!?X?Jg-nx=4gE;EKQy&4OXGi2p{Yn&WHe)VLRAasu0oiF&jsn4w}Gn`>5lrqZhF
z!hhI1hl<*XSoeOP9-r%7lN@ChrNOvawj<*e3-@N)M`Dwfwsz<0y`WEOCh4AhEzK>}
zPkqha2}cV=b6Xc>uVYkkDxQg^(rl$*6s5Ed73Y6kcT<-km9bCwsBnfG=hdl;iq0_)
zMGi<jmR!4*D$v2wu;J;J%4Mb_XA_&E3dOD;Hg4FrDzzItnC9_cZwfz%=7yT6qPAIc
z_bv+Af-#zoDGEx;%Eb_vz)q*7saalLZU)wByo9y2wK$LjA*W6i*VhjVpBQql$-vf7
z*(NC|A}E+|s<*QIX`1s)fYRpt!oq`oe!Kho`mP-BF_KE}CF+Q@v@}hbTZ%bwWv@K<
z^z_8i)K*bZF*|(Nv#<Q|a>JPwxueI9k<TEQTVJoVpeP_%?>j}yfxd@&PBt<=Kbzoq
z$Y)M{Mey>oY|s~G-L`Gp#M#@Lf;l*o=N1+UP`kR~JW1KpVsqBQf)h}lm}pAQXsW@J
z(Ub#+a2wRp&rj*PvF}9Rn*+-YA>X!P%!0Q~X!D=%HcEl>+{-^vm7GOERX9z57UC_@
zn4G%0tLO;MqSc0vzOaNm!TZK4t{tg2Zgfwn=bM^1IpyhHPox^jYo-KInz7C{0g+QR
z-?a6~6WWD^g^^!s`}fn$efU5!F2Y;>u>0tNkS5|0_6`W>KpP^b-`&B<DGbB{pQx*6
z|9rEI?r_AFYZ3{wbO`+#f}p!|@1E0mTZHkpG-lce@1US*Z9eQ^E%0!V_xa#2T>AQb
z7dJgYjf~IC5K_HsjkT4H&4ULI!ZvN7e-E221t3IOJx*&Ve`#}YaXI1mSp51GF2t@b
zMlUa~k+-%DfefT2iG$7i!+_*M`75hvYH}dW&(BNahU6-fv6@C#Wkp|QbrG5E1P(>*
zKY!I|Bi6xlN{|u=V)oM9_8uN9Q5s%@m39wNF+gEB{F?j;9u=-NFZ%kzameC&-O|}f
zxkftVt1D0tKIep3W4x!Hkn}`NFqvD0GEF}M3>HwKT3)~2L>aa=Ffj1km6)Q^QVo1t
z^5G^s85V>01fM*q!fg3<Fa4GuTP4!;H{=*oqdAAq1_%&vR7|_B(u0)%&OSy}@xAj^
zR8ms1cXNxuo^~2-JgI$#`q#=zUEqc{Z{9FduO|Y&?0$DlP>Dr(PfhevX!mqpULM83
zzyM%DEqF)qoMEQL;Wag~s9r&56oGs{4;J>UtSsCdC{QKa5TU-d&X%#*jCM<66tD}R
ze6|z?MKit6y?2kxw~iMj9rX3iTbFmDY{$aJMuEQ^uN{_<z(^gTjtY>FwDb#Cz>Fvn
zwibV$mXVj&5NB5xOy*Zs-gNiw-A9iewP4DRK(a1oKAx|rkRrhaiKMnlhsuxB+t_LS
z{r&AhfFQu;9MfVHS=*-6OZ$E6qT}KoK(BZaE(QDZ7NrrY{Hj=vKThw(Gr?8e-qSOl
z<Ulre=}n8-@vM`cPT5MKo^p{hN;@yKu1+y{&lW7+oo9as+ulFD5#?+kL)8#0!qo+a
zU%z%Zae^t|ltx}&p77ywUag8Pba$X@i)b6Vu;7JpN9z2{_PEztVG{OD)2+unJq6OA
zj^jJvY6i>`k4R*5ngHYB173^oJP|+j@=bw1z5@UqikEo>1)^RJhb=9+&H%h}2Y^P?
z(td5ekS9~V`bI{DWIpvBYssx?`ec+t`Jvm92wJnaHh#sRjU<cGe4M*I$ZF#6>+TJ1
zk%Uxgcz9U1a_}GsB@vO4rWzU)hwq=h3K)=qZ(Sydbbx)_v?z_=E8ElGuLT4`Dty#D
z@sVcRI1mJl!Ho{^_kV#4x<1zQ`-$?RBI(DGp`D%D*o1lTVGHtHDIhxmF~!)$MeO5f
z(=}XMbW8h9j~x@0mvuORSci;J3kws!rntHJ2C`E_jGV~EGEf52SwvVkNjonGQ&+a_
z*debc1)8%4DWy+g>~BwtVmfd*rl1l91O}Rdl!ZUNgxVJJ51fFRJ$`DMW;QnaQI0S(
zF;Uah<l1rMCN5RIsCw>ajS0t~W^E1*j=tBg_k*~Rk0DaeXMFD5z55d7ZgEV7jHc|W
zdv;yr8kji6w{KrUb|j7t5PaO28uJn=1A~;fsJ=Yei2l)0#_ij;2Zw|ZqQ!Zvkio57
zYoLqU5URQ+LOm3KD?*nU`|=ZSN=`<`9<|Gvxw$sv5!j^B@SS_5qIyq@Ju9dVNy25R
z!|r`rzH3HY+pyuuxGLYC!9gnYZxF?>H@<%VPR3*qD*TVr_f2#*uX*w66*bs+;%WM3
z78Y%YvM34Q9Yq+66ESYKMX8#m6)p*gdGXij?CrPV%1UopVXNlpDY-{QWhb&G=v0si
zQi48*Pqr@cp$1wAsONfqZQ#ks$XG;`ORcbX9CdHR1}i9-5j^;rshRPq<FzTEctpL0
zf7`Yv&UIvFIN-^Y$-3)<f`R}!^CLL3aXDsmtc7G``h=G)5X|xClPGk%o;aa{Qajkb
zwAlPyTwHs8zPpt4*!fmfl^PaEV^dT7mMAL;LhJS#sSO2%rM17GBaUlFq^4kKUY-y*
zbT8h#VSq*f#d84(iFt?S2TtuOx485fD(prYf|x!GRz5^JvKH(@oUAxOd8DMIK<;fp
z$XUH5^z;>F(22f*0Rl<x#MC2tDk>^Mj3FQ4)~%L<Mk*1=P%|VVTW!C7{mKu<8}>sm
zN{fVbZ)R>zJ`Iiv6k%kCPo}IswU?Tkn@dVdi+}=AK}ksvJxtb^^_MSSre<K!I)1!N
zZ#oHy0OAL^g=FqM!$Wi`srn+wo$Q^QC=lrJ>&UwmjdSq6XNmBRo(zHYC@3onVW5<x
zq~xvY>Ndjg$3S+6#~oPBOQ0|1VSFP`JSFbscs(<pKIvd%qNabMntLG;4JKyy8XlzQ
zuI}z_xcZLG&c@uoPX=H0-!#yta{7TmXp5*BjG=54ZdmvL+oEN135$lXH1VRYNkm4y
zO4{<0>hAfGN}w!iYHIN6s1Rq^X*c}(^$Ri<h2!$MEGz(ajm^!);!3?{&9c#Metvr^
z$2I5C9T@{~_tCRw;Rv6m#hOM&OsGpnehaKcHTu=ikWcwQzA4nI5*GtseRzwbAKdrz
zC~P|bM}6zgopv0pWApQS%DxN{_88(cnhOpf=M4?f?B2aQuj1SvjEjqEZ`VZ0j2cNM
z{vA!^SwK^cPfyDq_jxqsi_^pj^>c4u-*q>-b)1}pu(Aw>5B)letOK)~TFLy7u`zoo
zE;M~Qfdz^Zi3Ij<)HhOLS+=acp<%0xOe$f~Ugi}VM;ToJZI)QftEr!6yf->}gv4ds
z2d>?j>bSuY*YOosrd|M)iz_O;+9!S>FMEtzAK;PNW&2DJVGY2yg)I2usMlnrg5P_3
z1K_=hbm`km9vUQzVpyvXUGJJ5e{?Rtq$F0<FjE&0(s%WAyH{T(xwmTUTvuFFq<Z-9
zW*q%^mx`Ho&v8nk^d*mHxQOC{1FDchtb~Vaa`0dn(tDO8M~(=wQm{DSR*nk?>gqD?
zbg9ki#bzk#a1n71ut_~kB7c>msnNX%)6Z{1xe-#WD-MZkxnFHezjlodUm2IPKuzDZ
z^vC($;_3?o)kR$FGW_+wt?pKckOROG+F+hP=idv>^<=f-gxvj8Wbe7i=og4x%E}Ea
z@nKUFH*l{ZX@(!A(74*;2pqts#k@-{8m0Z{`zOv!d<{cPeU@jV${p#zeTm0(?MeH4
zOS|YFNHfV<kUzU3;IJY5rwv)+PEM1$QLWO6sZ{s5I3N5rM<!kCKg-(ZE(>r-%k&o(
zR@QZ5Vq$q;TjxG~dWa;p%Il3%-i_}rSE6XP^FI?36Bb^N4x$<uSVXgodk8ZNG6014
zlG0Kn3@-)-A~B>XyP$v)6+8`RXJMT03!_bOy@f^(Mn|^*0TQOl--2_%#Lg}emoBSA
zhsXnVb#qh0v5ie$Qc{9l9R@7-tJkk5u?O(6A7aNR;I@a1Y8+);LU^~e#SC-v4OqjG
zk(AhPn`&Bx9?)}r|JA<7*Y@yX%p#^^Vq!vehxzbSNaB!iKqE!qpNNTTi_KzjfFL@J
zp)p1nbJ&#Vip8TJpHoxAft<S5dwd<i!Xvj(&+8$nW>;4iQWg?Gop){AvuDqX(a{8C
zn9M9_+xkH7hinVYwj6wO{5lLCS-S=lzcEA|1VvRt!vwvGs-xToorIX*Cm+T9^u3P9
zjvYhb;TJ^fkkQA-M_OK9Cm=u;P_c1LXGL2`Lzs0O;gw{IxF{i&h&cCqk`GJK_4sk5
zzw3E<c}a(nY~b(Tzq@*RM(7Hw+t_Sscib6Gjg%9mXB8m7O%5H3!7EV}XGbY;BZ3+V
z+8wA`qHUUpNS{|xK~R-KnL8tYdRVP>_-hdzt(>nC+yoH+`KwnqT6$yKC#R&WvmYy&
zAn-zPJKMmXO+=*-N(II{)pGJt7NuIyn&8N9om|4`C#t3+OnYxIZAqlU`@u79efEqV
z!uj{_-}4KiEmdT;O)4Q-q3XHx>Q)QYD>?&X#jayFxER)sS$DCqU@e{xJ@^q90&rxA
z<_ymliQXE#PH<c%0sOgDQxkFN5@i0oxFrx`3~`v?ilh*<z>6F!=<Sa`Z_lhQyUFlC
z_ODiKmWx17TB^^UKaZ*r5PQH)rySF@r#j=>HHV`|H{bw8Zh#b-TofFuxJO4v`2}U<
z<l4aPJaguZtB1z}&{wm~wh6GtAi2k5V>)%}6wW{OatkL!TkpU?9IPZOfE@vOFhjC#
zINBI(FJH2vj8Kca=*Y+j`JhOb9FST|@7_JRkWOZ_%Exo!6mV?n=YHQ@Wg$xc<<=`S
zZQ(P}E>jeL@SCaZy!5jNh&RYhVzY!Cwy@(5u&XuNGPaW4{UC~i85{nXF5Vh;d1Uzi
d5C8Ko-Kh36ok@AWp*8rAo|d6zg}VKP{{x?DtUCYz

literal 37631
zcmc$`cQ}@R{5GBz6;c#gWh7F{$X>~+P%?^8Quf|kMTBG}$&5lqLN-ZuRzmj5CL^M(
zJm>ZKevjk#`~Ugl`5d2)L*l-#>w3T6=j(i(=lQz*uc|0e?4{qkZQC}Ai;A+>wrwK`
z!5=!Z-T02JPJ+v}ZDBPRWzVZS$NY?Sa#yeJ^ouV!T{0CRa4?L@mGj#(y(*bQ`vNFR
zSOZu!)!!#2CLBxb+Ep@{E_FywmWt~2pQ{)4PL){<mM#6uzI*)PrHg`OOYV}<O>+Zf
zM&kn|OJ)2<%}3<ODlS@IyA`I$ez4xe-<I#zHvxMt_JjViwvNYk$+CtjvY&WFL*<XJ
zVrZ^O|Mw5q6iM*ygY_3#{-69nvu&dxKc(MDVZMe&-g$G+&E?KhH}f+F+*DG+SC=Nj
zqoaF|39_fugiaLa<b=`Qau4n4>ESpPI%+Sww@ypR#Dr^eZFwZhC~mdW!a|VYiuf;C
zZvj^B>+kNiB`Yrv*p>9Vug@0H(9*^){yU!HTRD7t?f2Aj^COiMv(E=Fs$3t{usNLT
zx@;O29^PGKcjC~YLq3llsouPK+0jvOwkPjGQfo+5lz;u*#KOWCEt2a`zGvPHfBEuK
zmR{NW8>h^>-jl?tC<HT0oTVdI=ziaAZaz?YB)41X(xttpZ{}CsKAX@vLlXS>o}uBl
zj~;GH9MtqdHS1**41z<BB~t@sVh<iXNKO0l(PLfMev)c<c-Vfn=a8H#Cl60@aWTDL
z<@tN}j{E!j3&*OMnwfE(Ig^67`n$f~$sO4+cBdsyf`*2M>d+y%NFLqNcnwX>9ETZA
z0eT7(6B8dlzofL2(SjC<cxdhI>hm_2FI{3idp2Zo(KaA3(4;s2^pPV+CYN1p?d<%Y
zJYnbIc^WI`bWKg|LW$$7@jz)&=l2IYNb4FK+4=awR%Uy1-Tqi=7g)}oo`1?Drl_cR
zOhDj+-s*lv;ZGIbq!(3Gk~Pvb)ipJ3C%>|YI{qZPUwub^@T13ot3`it#E&0$zJ2@l
zI54ns*!n$Q=jA!Gnm7r!$*yd+(9qBU53hf#6SDI1--_&viXCQZ>O$Fm{rQuUkgzSi
zBcR(nIWh6!K$)AGnp)*}Q}o4)7g<C^x^LFt>0Z^=zO=qF>mM8}<<G5dTg<j=^whZD
zQP0~Oe}B1eZT{K#ReO}2N%Yy~`k$r0Bf(1@*Kg?y(7TKT2^n|i9IvmhZ~y*1$9>iI
zOGCpHuBBkWm0Kl+I;HKRk`lSycw>dJfa{-m#}b#H*|l|Fz9eep8e(VgdaT)RBj3H_
z`0?X?HnmJT_N{UQzE{#a?g&OliM#L_<{l~NZCal%Z<)y|SJBWol=X4#q-g9@3AdHA
z?tiW4O6DasGj3EE1WU-5du|#Jew0YRo@F@v`B4m>!q0)SXgpp%gBtQIVUoM|?(v#^
zKee&BdGX4XiRtvL@%&y(Hg4{~lKGF(pFT-%-@d(Y;q$(t9|7FX3l*gJp%l5m!?Fqr
z{0`IV%+5n3ac*-W!~>l#-)dGk>QOZj@73IJro6Exh+(SzEww?-6BY@HSgVh#yRn7k
z<#%V5t<YoZxG#5HFI}5X|6tO#yG*ae<4=EWhE^`!Ia8YJIySp--%2sUp-zJy1#$N`
z@L0wF^x3^mPPQHTM8fQ~VT0ALllaqDnG)}L3D2?N`RS3(KZ6r^8siby^@a-uJ=sp4
z42aZo_3zF#qGn)dz^W)~Yt!@c@>bQ<xb9*<b0#<~?WAUwUQoYfip$b?pJT7Z#zN(O
zVnvC|B0s|C(_i8g9uuQvY|Od3x;i>Kx-jf_H0;^4D&p>@JJS6F0$9bwqOoA1&!3Nb
zr)&RE^PxJi6tYn(;=GVh_%o-$hf2o%PruWhA65P`WGwlX{U_yvnNMeErc=p0#-ma%
zT^*19Y?$jW>9O{BW5b=rhiV@!4^5<wt-rNs>}PYqK$0E1AIE$BD~b#H^5s%%%9W$p
zhD|SB=cN|w*nT`9Wv1%Xb&7Sy6+fbq=`t6znR*i$fQt@nk=oLE!X&2oEHW}MAYiY}
z@FUuQGvyoC$XI+D#tv$xYh>Ii*CVGpE8X7S{y3}rpEmG713$cv#m9NBXrk)z0XCnu
zDVj%*9wk~0eoRbANYFc?u5m~uMeRnWX7=6sgElxz`wDtbCo4s<5}R$_ktg!<saqd)
ze$VyaRKI@RZP2br=gw$d=$)nUCfvH`-(galeuvIIEIx&sE4|gqLjCs7GJATof7P)C
zKQ?Jg-V<2Oy0$!X;`!;D<NY<Wj)B<Fm!2>PsgSW$)gJZIv66DAHNkcL7zvg-EJOBd
zwqP)T<$^DnhDJ4oP~S1htqnUoJ*Pe!zrT}7!Lpi~&*{$IzTZ{mR*?Gc-Qi%Ve}#oJ
z?@DBqm4g&7U#1?!Zd2`)x3I`89D2B~3Mhe<8XX(+$4fqb`I7$EO0QKQ5NBUX8`sMG
z;F*Ke)YXe)4S32rZnJrJaE@5n*e;OM3n(%QTTA1915d;`%k;TS#42=$#%RVoJ6y+H
zr}ZUGBfYM^UVos}WprX9*RY8m5YTFx#~JW2k#pOCk%q<r8Q?QXSBcY^SWyR=l(tuG
zQ<Hr~{3TBF92^|J9qHE%&e9c5r==H8B}cyybvOwe>$7b*tz%uWYtmj6I8~e1sn-xB
zG(hHTd3XBf&;F9gGPgf``k!}Z>3Vs21q1{*|L{M-dFqsGR;QYt-qQ}#>RHFOYy*YQ
zpFfx0vji-N#!`mY`;?bUIhK^^ZQ9w{Md5AEojW)1>P&ja7XV@rmaDASBv1QWTN_bH
z&YwR&@MrpIWoYf&Ij>li>sE?Cao`Pck0==7QBn28Ly(i(Md{~_-|k%hcb{a3^F*vu
zU32qz<SkGA_Egp7FS@$A%F4<V9_uT+5^fe)Vi(@dzCp)(XQ$NWZ)#dy`vWK%O3{Kr
zs1;$+(GB5OB=<OTC=*3XLxTWqw?A_Vx(r?toeDp2-4p$TgYmClvy0kKZj)A5r#T$>
zC?<xHkDotE(1M;Pa$kCf_Pn!VGvn*~F>g-kmA=p`b5%tJvbH|^>eZ_Z?YscX{$l;y
z?$X{vhYx>Cl-u{It*%o<t@88dt^%uDWwWl!(~sJws_tYvGrYQKS@ek4pys{p*rjCU
z=y9gB;$dHgx~8VF+!jfLpIzAlo5lk4GBPrg-MJ@lu0$;h7=q-zRZ<MTXJ{4b@EqpR
zQM9q)y{_vJ;gtV1v{Fq$a9m;FG*i9UeCY*ydjaBCP#AxFT%Wt3O2f*^D(t*)yUZdQ
zm5eyP^JSt84ngLzFFHIlonkUHxzBw3%c_>Fq(AiOLFwNgf!Jju^Yg(}55vL^;=j`M
zpvR9B3~HYeFhg_bP;0r@maAisJezA-+{1_4v8l@vvpP(tI<t<>e_ZDQ^sq}x#uXL{
z<I$lwjbdLNJ9g}on>Q+8fM)upAGc5W=0=@?8@mjdy2f!~;TL%5<1O*Q6}SIh7Yp0s
zxhCB#lW#TH{OW?A^HP&gnZXRo=iQ2jyT};?$x*Kwqxiux^rybPiE*1Nq9fmpZRiZv
zR7BL5-`_450e7P$+yt?JK*QbpnU!#8CVTR@hfLIsnqJDu$ZWrN@7~VcyYKKtffW(u
z0NBTG*(Yo-%9JD@^x3jt#9Ek6!d2jw$MT_EqZSSkk*Ft6$ZMV)xuB!NcsY{iUVYec
zz*G!Ckw&g!%(i1jAEal#{VvN+!RXo48XiHx=kcCv{=}c7o_`;JQ~Y+7=+v>J*e|tl
zmOf}VPSr~xtL6}v3onGMn1dx&=&(m?EDP+wC)?g!rqy#9J0$;vu7X%lY_OrI>QQ@9
zlCLw%vT}02P_NalUHg3fmWN#DJKa4L6!+>vSo$0~b#*+JJAkad<VR%g++oEr5uw$w
zCK;&>J|Qk%s#oOw@F9Uy&WrUYidK5eRyUSfj!3!-%O@R3YE`^;?O?Fv8s~iJ66H1t
z*Wbr*2LPjoKrcJ6q@Dt+E1+%fwKEl%Xt#ye%c7VabzdM!QV8adlpJs)*X8LPWWMQF
z`JnpFD{Pgj+S(-mC?EEnt*tGv(qLhzBE`P{l8V%*z(8&YYNqAK)p6IS-8kKCxkguL
z<c8lw>LtE@y|bjh`S(;CLHV>ZpPz_#55q3h%+!{%wdF$*U_-$LvJSXRFoBq9wx~!;
zZv&iV=$1Uc;77@8H?Bk+o3@nwOrrJ)AQ;#yeqdAvg3(8?p8ERw>>M16-)~x}s;X81
zSINl9*#!iialb2kc$ufjC(o?YwpEVSX1?qX_J(nP@j0MKXs0E(E3v(|d3t*4*i;`H
zs408OU|IM84`&Hgg`i3tSK^*Dj4iGlT$$@{Kp`Nbpg4{f{xj%x6s+W^*T!%x%Q=&8
z>|mu$1RPGpxfOy6ZKD%+J`2#^E?%kCmU5-mvLHM*w&@v{S|QFM$qN7vo5Isk)SZO1
zv@c?^-{0wq_2ik=zSpmO>hh1MnC}XI>;=cKO;t@IXjNIfu2a85pYuo%h{w(aB~52+
zD6mld>VQMwk6;}>^>=@~G<(N#?ATE1=8U~4v)fNzb3ul5%}mx~$L<4^Zl9%_1mD_0
zO1bkG7<l@-;xBIEyD8|aL?&Xq^|B^VD%iNV>dkWB(Q=FAUVZ97YReeY9*LGr3a92q
zzWEWmiRR^IpKwL@`{?MxB&hsH1@3g`+*68|jKz6X2=E*Ips*s%9@a*d(&kU~<Hh-?
zgBRIsuSrmG9YgEKD#Aq~l5VI!hFS;KV9+dXq9{PG?0uYhBdc5T`!lUuJyrq5oyt=Z
z48d+{dREFVeOs0%mwvgQIeof`VR`G;M8?4otb8`Q3(&jA4P&MPVlFN&;A`Ill^1r&
zkS#7PMYGR`$S1wRcJr=~R#bf4;?-_h5S^IF`s2qBchl>tG(R21fDzcn>!kzqLB7Bd
zfDBHq<e>v{s&x$wNuX=!N8)?F8w(J1Cr{htdq&ii;1axQX({8U=o7QfIIhILBsP}W
zyP{;?z)Ak*lO(Oi0y<cuAjzX1*j67bdh=B+VxHzNEiLUKCl@#Ul95Fea*Li1zlwA<
zG(Z85y?y&uvVbBaH1vGn;nOeK4d&2*VQ=QO%x=e;UenUjIwSHXi}UnpIkNNW>g{_x
z&Ik(T*SnerRcHzAo5|ujO}`_$>3ga~OnK$3<57O&=IdSN`L3!zKUPiI3r5W*XE>?#
z7Id4}mc=PuFv!#1vYRM<`&7fgMsr!3ec7g@ziW17_koz=$JWtJzSVbr6+IS-RZ-Wd
zHWyrPf0xL4<J93kOQ-Pm5}w!q{Og*GyG5_L;D~L}Q<IyjG#eWmcSXbbX}FFVaS28T
z0SN6yb-h<jpUAV}u})7<@9rM-Kjw4n)Tt*p*73cLCH+3WzJeFaNN|q9UX|}iojOIH
z-mz97;a1Qa)H@4CBd?+?xkF@jr!3idS69)gsVV!J??KO=J#&rkoppT9ql+Rl25hUn
zeHOGx3mX!JVBeY2h16<qb_F%`%D{{R2M$o=Un%K-DxdU==kVC&=|#RMrDTrmS=*wQ
zuLp*Tj$rYWZ6mGnw0|uvt=C3};K~~fb7?~OL)w~2f7ljv6+3VP+F7>#cx_#u=hx71
z`BhE3ispfM!>rQ9y)tA2&Lb2bxBj{1n|FV1yCUV$D9}D+lh~<TZl+GY`x%F|@k#B>
z+7+92RgoCqGW~rC&dkaW5{IlsMeHZ9mKA6WUA5$Cmj8WGL*sh*uSoKbwV&s#Lu?j?
zF78s*P)F}-^*xaG>)qN-XWzzYzHLGR3WmBX=T4ZFy#3O9BgRpJ3b=!_8c6jJJJ&V-
z5G`%4-Gs{FQ#bB68IAw}-;|cdL~vhUM5}g?mbR8?Y&10wqtfmaf2e-->d>YqY8|SX
zm0V3=;{7&%yIJk_#YO712d1`mZ@Y5dTulCDy12+Wm_Iux_*zil*{9}Mo1-cv1IJc$
zdDP`99G-~j$5w0;WfG1}<)M99T%sDZyTkCpp&u<4eU>_YZv6D#Bs9A^EEO~~S_A{O
z(^GsNUU;0WxsQ@`*ZK<oa*@%h?Xb+(DJn|Ra~&Px0wxDrl%wTs0}pe$mN~5(DBEA9
z*_Bx_)Lmm)JLZ_wHX|XJr!Bk@JJ^|7nVI(csKDd2pXRnJvq>7chH>+INwiOYw*1qj
zM$_jgmKPE6Q@1?MXXrLL35l1U8YO93S|S#iQNeGAT#7-rRrvuLD(M!zGS2Su_{1M~
z?rd&Ke6~7!ADhoo!KUczOQkr2l-9&z@z$RKnlloboO@rgxhj61-7R-s<Oe-ep@DNV
zni?OrtSQoEj!nk}pUYJwvx<R|^Xc|kfv&t{&H$&$M$^1CcY3S-yAjmcnH{FtIDAUC
zS8t7dpbimzx;*kOGqc)QAhE-gH%bWv{nDjNUyTI@-X>L4R4ms#Oixc=9rE6v`u44U
zr7r~_3MFH6K9@y+o+g^2ygIPeaKf2BctqAL@7g_qwGGP)WDmX@CM^#7+Vt`K>@OZ3
z%sFluO%ars&=#kyrJE*Q*hfZFb)e)L&8{H%%9eD^(2q`MCLHEfW}ctc%GKY&lM?<d
zU@s4oWK6;*>F1ljNP8WBCfu{-qorV0%It1*y8L{|UYO3aRZdn;_Wj+ClRqg*U3Tqy
zQn(<KKieaB(S+;z6Z%KZhH}D;?(0)2qlaj$S5sB#HaD*JS+-A`>Kr<>v9Q!KIy(NX
z+{sEb-)dgDMDMz0L0Zn=iTL_6r$#@Y_f}L^zBgMFq^_x9b5<<6nMu@IFPPbVVWoGJ
z_K=P7go^9(Pk*V~T&G40&Y4}k-&Z+yv%rMfF!0f3zx||sJE*U^Oj2#F|57kDHET7_
z`#n7n`?u?cr*=MhNoB=XmJ5%A14n{H?5NVdbg3o@r>LFyt#SgSaK69c+ad19!9n-_
zKBB#L^Jc@DXhHpg-nSX=#=`>7aC8kGEl6{%QX&%w_`Jp>)>UXg+BEW`FxAUTeQ15G
zfh#lZ&ebfv4^|xkw}i}Z&%{aC&$s&z^cQD~aR)YTxbMAuX|nsxKKjh(w@`R_IDB7C
z#0Sf#S6k+~c%E^P_*(h!4dY?X$X;a|6Rvw28V;nz19?e|BW13PhfhvdO}sxF)nv%%
zD8|a&z;@fK|6Qo?w-zHQcLO_duwk@oFU^ac46J+hK2}K)Ggaz@G{oSES_||9m}+QX
zzY**=`^NOD^ykV-qF37nVTEJQo;@dMxD)a%0<DWwQb}_RYALVv<}Iz3x)<~+18Ob4
zefy_X>XVRBOz$DRZ7qa~GU8^}b#-q|J6qNtn!DaMZTH{zJt@I?t}A#5B-N;CuK)MU
zVEq1%o6pbhAYJ@1lVt~Hl_SI-zTfj-)PYLBrSYxJqe6qW`}Ph!3Ai)&a2GrKasBog
zmw%goqD1UCuHc%$e`M$9TPT9XKW7#UHm@YIo+{t?d^}R;^Q&)se5Vhab5OT<Es(Uu
zyHQ$Pr6S?ta{o1xb(iLlO|#3yubxCX@wGa(Wq0#{#06kpw%xp5^Hfhn%6N;>znfOS
zs?#*?8XEGgpkLUmru8W>Q;_kaoMdG+=D)p-w4p)kcf%Qb(Jl8l&zaMfbV5&AQ^LjK
z+{0#Sf=2$#_O@E|$i_>0yx3|IDiyI}p50CPIwvgv#O~?nT)%(7<734J3MbRkUi|yF
zUMgrrX_I*qIAzk2b+mV7-e@FQ`2&-1h|z#&^gchzd#k<_FO)gf_S1dZ<TpHf?+eX}
zq0v(FeyN4J3;Q>_xv3r!{rE&~OW|-0L&dB-1?6~k6o2dFvi)4w0Y9pY{ty06QfrT2
zUGQC4+gSBcDGc1~))HTuP8XMS;TzM^axc5P$9rw%-bhd0GsC-Q|Mi&J6zgy@h7Kw+
z{<FIwE1+^B`Am70lo;vpvs6L#qaK_bZ1myf6Wzu#zoa&zT77;@_$79}nq1}-W%@Gw
zN?$HZp>>9<V=6WKhIkT%Hj}YO=ST}Tr!?95C2DdVi`K;IBZo+BwfEM6>2?>|uxV&$
zKrCd4ZmOB<FFC?*NDld|bou*DXie9yU3(l5u=sB1FayJkuHn*+jehk?FBv<&{-%Uu
z_WJt99(_Ahr_KF*HfHka3wrH3Q#*1tq($0j=WAX|iJZvjm>RDL?6JN)zf)&=_*FRD
z-<9?GvfQM-d-rw~TQGoG|L|q7yp1-gF^=DlmxKgxx-#EeyD(fyoaB^vujDK7lAQDu
zQ*J9)uj#3vHSE0QC4oL+V{InuNzvk^o)0FKZu4ce)s?;ns2M8zod$zcQbtfTIZg<5
zQXkLcJhirFCbPeV{!8N*RyO|d=b73r4O6y7e>a1roPV||n4pR1EwD0b_@23Q=T2*=
zm3AE(*Hy3SH2Nw>6Ur`_i2eNcph3-dRV^+1HIi7+JpUT0&2^gBbDSS&>dH32vDAmQ
z|C>RW%Ddve*HtPH*FXQslR2$PM|e$X{3x03)HcVe!-g@KP&t{FX55i>2%3cbR9{p^
zqY@<9?mRPE-t6o?`$~$K-Xv0jo=CKa1w*~Hg}$6)H7FSuEy?t%v-7egO<%?lPZE-!
z{pIl}FZ!)h!Xfc3kf|SU4KvTEU6E=!eqFB~7iF`)o_l9dPdned{_AC);V->b6X|k+
zT_240pXcQreIBW%esdo$@8rz8vSRy_Y1_mrwF9U)J39X6Sq0xR=1l8<@V?u+H$yl*
z?P{i`Pm}L^ZHWs}f(&UL(@ZCSQ>sGU@;hpr$0>T+Qe$X$&i0M-tqGGK5|+t=YVQ6f
zd)DeR{+^RdzHRu=rm7Ohyr6<*3LQB!1AUN@^SqTjbNiP|!7B0!@n0up)vArp9Zq?3
z|8+_V_2I+vH*U~jF;dW>`uO<3>>!{I%^5EmJEbT-q5u*CGBj0LSQ%F&SJ^%NYQv(*
zHI|?7t{0NB_E`qcR_bne$!bVkp!k$G!C|ZMM=fx?!{nTB%u|Ckj}P<|m3|ZklV4wv
zv7kv90jXMkrbVWsasq+{A)gR35$RF)5JD86oHVv9nH8K*b4OhVZo>-Ui408m^vSjI
zP+FR+0q!ik)+n*fB~trbl+u%;AA5_P=I3V1nEewujcKYGju{qom*#GqI=pWv%7Ny<
zfd>s^YfCDNk6#@suN;Gs<imW})+#QovUkcs^y9|jWq2n9sBYuea}VXwEw-LL*ZJOK
z-4QafQ2v<WZ}6%w86Bp}%=5+V^8ty@^tKwA6N8TfgT|VK25vXKJSV<3nWXgMEQypi
zNp#2bA*wQ!iOiI0^O}TCHuZ>*%4O@~e*5Wfk1qb{_dB_BcNw!^;*gO*;;P?_GWSlF
zN<V5>=ZcK-u?60PGRI<6s`StQlI)!w==^M%lJ<J1o!qVp78bu<b#vX{Efu<E$^4j{
z91gt;JLpsS)LNLz(Qm@eZ)Blz?!dW{{`Kgq<hyC4qy%>FxD?wmd#2OB(o$h$KG>p$
z{K@P~j-|?}PbOT)pxJZbk4f1dXczmPzd!j1WS;goOh;Gu_t%Jr=0|>tSo!lQBFkI}
zUP7mbi)Y@AKnzYs?}l1;=~6&vtww?48eFBna1990iJjdC&eYn<Y{SOt5<Db%RT|5(
zWlcs(zk>7c%oL#@PuYXAaq;ks6G{fy^@k4vkk6nn8Ru!gD;lRid|JB-FNm(@$q|0v
zsw!D4kv#3w5S;FeH%76aI5A=`iXPkUFZ%kw<y-4gaK5@rUHA#vh;%>m-MPLZ0jq(S
zgoI;O^ee@SZ_}nvqv2Wc<nbZ!O?<gsO~+*{qA^}79&-Oz2$1;}z4GvM<mBX1(=N??
z&s-f1zXGt8(a^x2fpr4OD$afJA|D?gotWb((b#tb*}~7g&YU>m2bC(Mu1@ixx3>up
z3bkh(mIahb!QayzfvQRJzxwQ2LgV9Gnq$R?wnwYLGPd)IfbkIc;o?54@=ZaUCm1p{
zxd&fUZOv5*$7xynd2sIab7Pig+p^c_Nl$wZ)!MU4L_B%5n1Q>4!tzj`Wy^wIf6&0N
z(Xf4@c1E0DXh(qQ((k8#10Q#0as{jSv2ctJ&N%G0(|ltWFXr?v<+7x`aZ*n=+A!RT
z@{b+e*#@K#(f)NZE5c<Z?j|Qc<%?AH14eP@5J<$p+4JysxVX4p;YMCG87?j^!Wz5{
z-xYs><yY#zN65TO8V4U{+XvH*0Ni(2Ow1F_^m{M|KbUnsh0%b*V*EYh$gW+x@btcX
z`7+sGBDCAPVx37!N($08ev?I1^d+9l(cIb(P$0nFCn0yEpeV(PMvRW$g<N`3S@}2@
z*B%Xx?hhu^n&}ZREY);$l(0hG<z7-i`lavb4m00jq^LpQHa9=j-Q6u<_Pw#U!MddX
zgVi9RtwIRC01Zw)DHQt%P4<m)PYD?^6srpk4(D(c_~l%aHV$ZH@)OOmZ&Oo`NlC?1
zGKoT{sBdnTJbCnGS{mG&IQT&jlqQ%02T`XZA|mYP`l0|#^Dd@P^#NLFtMMykGB>rf
z>ahXPhu8z%^s*M_Ik~yvHZyjNh-u&^Q`6E+52n~#sveLbJE(G18WQy+lb%FV$jpNc
zhHKLQsyes-*q5O>vuD>q%bpMZ&}UpyUJ7D6qtqRB7>M&&(Q;k+S^tPu&$)4vj0HEU
z9M(uB{=mmew+wcg6uX23Q>=<#!`P1>foBZACY-0R?1hxhPqo2h^!onjxa&@KmSLZd
zPRV=fOhkN+hsJO^T+lxugM%!TnNPLd%WKO(-cD-a(~d2=F<l{_@9NvHow%*=+~(7L
zk{G*Y5qK$74`1ZEr^<a3M!+B@uU)zQuh73hv+%8cC0YO1EF!blq7XaqA1!}CLsfsI
zI{V=VdNJ$E$tNRYGKpU}0Kpy`>gmfL7TGsmz*>dkRGSx%eW$9fZr|sao|e{?bMJCn
zE%QZo<@?AqTtR&>Ko;OxM8vRTNqtk(aW*z@>|a+mGHkAln*~pyIDZ;p9DvK;n{TmH
z69i)^G$NuJY9BNi(AKcHI7(;3R}s&iK^hv~WE8(BF!QgavGFdt!de|MXgy1KR(xLn
z+#P3ob~ziaVZ8^UL)0`h@b_w83Wn6w)Y}lz#aFrvpi`~9;BH5pAltGZ3~6xQVR61Y
zw^mHzi1=laEc{|gvt$2<zT(1=VsT3-PATrp>|wsLqOHe=K0HiXX<yMNyI;O%y@yam
zpbKYbXS>7D)^Y5<Ck~U5{HSD%ZiyqWX*=g`a=K4!!QwJ7`^on08ySzb8czzAWaZ}G
zC#U*%bt1lK>)-kjAxlO)isa<v-7GsHkln)_!H0Yf_81-+S=(z>uB@thfJw~pE*w#S
zE3|dOmu^jzgP*+AZ4}Q71`TViY|Or-AC857o;LjE+Pi$bPIEeU?%W|vB!H|Tjy}9%
zUl?GJy|_3z6Fz)+21x|jhjmxVKhTqZbr}S=7CV^9T)1E|P%1*)8H1oXp^XotpfNKy
zt6h<hgDTB+>eOAF$WvN5K0*V|-Ury^zm|J>d2akt*uG;2<Qyt^(Q-*jwGoOcyCHIR
z{U+J?`&Ivd@E%6P!i6J~=l1k9CrSQ8Q~$nq7TJOv$60N7H)6mFm!SuwUDyKXo0xTr
z?O!2Y0N<%K{rZ`c+}x`_b6W_z3qm=x`^RQkWzXTXTb6G+;^vEB96OIiTo;C<4>9@7
z<hM8bklzsh{q@2NB@WBN;oUHiXJhrmutA`DqA=Wnt3hA_c%Msy31nbk8?f~W{kZ@f
zOTxr}HTDVC>R)hUUizS&q@;GRioZdWCOBr9n-EMSk+gxlygck`8))>zb*4eCz`DA?
zpg|dx)zFA@89(An$@HbrX2co(?<#T+tyA{De|Cp+sU^QRZe=C(vU`!bHSm9k;``>l
zq6)+7`-%yb_oxQ@qZwv=R5i#fD0ymm)wB`f2z&lq9_B?jhcXHqVCQkGan)Tu7;(n1
z5UxmV3P^2j@W*?s9Kp&Enj?xh@RE2kWPA6r!hISVx{VElmIRLl_hD0{1B(sW7hpDE
zK_#WqFxREiEGskf__=ctsNrWVdIUtB=G@h_W^%-L0BGAvUF_ge6plT=Nf3G)8yg}4
zPzsBMC>rR2+zhV%qc82ob3(T<m)VwCFmOp)`o2)XGx?+nbHTO@E&gp-vA1{&Vc{+@
zih|d#eY(1Cpf^%fQevOmnOlB|J;<MiYR^tNH|CUpeorNmA{aS0N*qt4Mj?CCQ?hZ&
z**8dsC$h7k6>g)TWuGEMPIw>$?1Sdxr5@^K{oxrx%tkZ4gF$w}+WDY}p`qa-&H>a+
zvORlPczKak(uM_g!Iyk^sKVR+XICi7qN~}1_5+6wRijKy!Bi&xaj8X0A*<6M`v&|V
z0;y6{If<R|Z*#o}JB{n)$vex_9azxZIDQGdKdgkW$Q%focinF_hX3L#VmJQvt=b9g
z55^S<=e_?zt5wyAxaGO}=&f2x4)DiHG96Kd_lP-V?N8<DekE%wdMtJ_On#!rEbCs^
z`~P|~WQjt;!ak!DA$jcYKm3I|tIAk_aCt#LYRdi$Tt?X=*nL_W5t_jQ!>xo0?;8?A
zg|!{23A&Kp@eK>dQ}+AX!-o%@M?)2`9}?5j0uUn@hT=dpUSwoL&0_Pf%{t$m(k*_5
zT4{A<Nbz0q^vNG1BU*W;hj1)dg!-&^z&-haCkEyE$HYV}5-d-t9%3_xynOiuFSEMb
zc`I0Z=V?yP$DcpTJU?@**=G+maU2Q5jU(JYh(RbZ?Ihr4ET@6T0H8!90A9a&)0$^?
z9bpDi=@gaNLck(H7sQ<A*e3`1dM>d`%8q|F-FcTMV@sYVqp?xx-`|n_R>i>!3)VcP
z3!lUFF50dxj^QaKv>IR4hyZl9Azc&R)9H_T&PzwGd@z{RPQPB~#=&~8xbXT%8ba=x
zkxH$zj!9|LdxWYSrg~ykPVhuFlrSsCo{?h}Q48J9Dq;reV)4G@I!o1{;Y@^Lv-<1(
z|6fXPR7yqdO=zSdj;eb=lc53P4eA;iOn;n!jFpLUf&U!DGBq#j>$AWgk{GAy_e$bv
z`hs32p!1idU^Gk)67i?s5joko<-z~v%^UaMtqQfVzq9l659fB16Ps&wvEj^CtAGCs
zMlGlMZa<=kBEkmphSja~+kfKh*{>z?e{UHu;2a34z#Tv)PR#M=f~RHnji+2xIQy^|
zh{uGM$0>G!tzn_lT?1ex%t#5lhx&xd-9K~v3LaUM1--gi84eU5yUkYus*K`2YOs}g
zqnqHpx)?d^VVUk|7`ApckU&*9B4|#BUTIeyFrpf&?)Thv$V|U*sGzP9>=I06AO{x7
zy#0@2zK)Lh_|(kI%z~hXF~o8@soO<cMQsNam?z4`&V=#PcK?YG5FPZ|lGs`w-1<6Y
z|AblM<-LaIM7V`euR^1v1N$vc-~bWMEU3jV;6Jg4VYCx<0`}Cv+X{kygD}AOhr?>v
zjE=%JABP%ADDhua)HH_=+rm?~tJp=mfDFtc42vCS?^KZyeGNPkX)UcIV94am;xDi(
z7jZJ}st;Ya2Fic$wL<6$%Ombs5(~!US08?233VVBEhIWu5PtFXpSc7YpkO0_g51nD
zsf~pl1Onr<u|d_q4o7yt>(b@RgdRa`esJ=}7?JR)wv?qt{uYEcy7Da;h<*plX@sT-
z0ud1jBho48($8b>z@>YqTashm%@^mnW`fi}3tU`yHAGoM<+Md4Cp08P8Y^GWXS)xP
z63;SMXr+eDF&8~OCGma)+xgj(7YLb$2@VN-Q^JD1BC$-wln8gdiu%fXFlWMHMsk9P
zUm;F+2e$x*M?BGD`>Aivw9iIAoZtJ{XY1ZZa6x|gBUL9`?)h#VLCZ@e{taO~m5*jp
zTgo?AZUQ~T<%8&VeR{ZSA5LLkpAJzdMxKh%NqLIHxhlo6Cb~9qM&S_dtkQ#l)yVX~
zS476^OW5(tMA(R^{``iGLrAZ@74k{s+zU00$h2WE>p2haBJAL3tB>_)UZ29)LD)h%
zr7at|#W=5jE_VzJP^XE~Y82<<i|AR>^jWI`@!gb+b-3{t!Zxf#<`S5P8b~}(A`<uf
zl-6fp8PWI=VTEWRtK|3ZsqP~RGgmyrgVIS{)Qu0u2M}5v4LtMl_Gh0x;8?FZOkI!<
zLsUwrMSO{z2p+(fC!Ee_&!6)dH8T<=A83D-<`JQ80h3RjJgIo)iV>XY#%Q77b7tQc
z0Pnmgt~oh5&MVzU=wkGNA&5t%4p?8;<rso;NJUHQC!aJt7I|wNdBIl*@L-F4T$|>`
z7PKn;wHxskzFS_K=TOaoJp|ozoa?K{eS8As0*Xn`l>I(<)Z=$3V~^;*knOB2BJ5G2
zrxg-AV|2D9s5yM&`*7THSkh4V8_q=wkwQclTR52*h`cNr3sE1`QdA(FA7o-u#%Ap3
z=s=h#0wslLA20ADISJB^<{A&JX4u-g{&*zgp#so*A_sc>c(vGvu&AhuIAU%qKP#r}
z`H&lcq|^L@*WiVi6Aw1$Kt@?@0LcNFtJki{%ny|9p`;|5Zme4Z@-<*%BqSvE)9wB^
zNG=Te3y-5T)8#ZYu*ihShyXIE4GI&%RKZdg-Wk1v9KppkjIe>hx7Qb6zka=d2bY7~
zopzq-nDXqqR^!qjyV$~UG#1aE$wC`L=no>ZWTm+v5gJB@1O<R$#`g7nq8(<iDh(AH
zbPvCOf7`qAf37}Mh-r{+L*I+GQkDp!kdY-|LCm{z9=&)$hjUA)ZBbEC0O)y_Bj-%p
zFIKlH{&PdO^p!V<jfqVxZqXCCOaJEMO=S%YU^t?g`*D7mzy#hD{7S!k|HdC(L`~2D
zUw(c{9~l?(zQRzr={pO0XO#=-=;^=U$t^4`B_$`1r^;VSY9-{X?^(C5poK@ZN25d}
zHnHavFJ2_9eh51N_^IB4Gr)Rwetv2MNmW!R(W@*cP92hC;o|b&+MM?y_-^LS0^9C;
zk)d&M%?MxA;R!YXXXqeZ1N~6M841_W)nhyCYKFoa6YHtzUa?-tb}b<E@(lzG`@y<M
zCqt(wjL>56jgNnS?1N{1`)o*fxOdwWXfQ$-nc8`G{%x%O>@5gJNROI{DX9MLXOxy%
z<_0NMG~B@M3rKDg=d`-IT2@w;SU<kopSH!QoS5w`FalpE1Q~)kp=H52F8B%XJ$~xc
z&!a7Yqm8nAAHP;rrFjG`A=|X$Bw7(7c5);8&MrJ1L~2|*M^bN+jt;oZkD>1-m?F9d
zl#15YR#Ys4(Luj#0@(|O1OnPMApNPh#MH~n%Xk0LW<+Wg(1anDdK5n|4#w+_{0T%6
z<kc^1{o5q?K5F_fI*()=8Uo}>*XK&0Coqofh{7Eqll23w8o_yzaMh`40j<WPxQ+x=
zJR+Hh^P7FAdaq&Qi;AzWE)bFDfW)teMj_3NO8%5d;tmrtvkEF55RNz$IXQgLA<$&+
z!}mtQF7xOkSlt*W5e@ojjXuMGFNz4&xQs^`y+Ak&t#=Yq;Oxhb^CI}A=l)9$<(#OZ
zC?<OAIB2y>SrFv3-rpvHfcAT~H<CyyK~*zsiEGNa*HE~z*l<L|mL1v*QCA73AIheX
zqhGljv2^qhzb9M0|I5zk;b{%9Psy==RQO-Q7T)qmBeEhN3gNDOhY37JOZv}oJtO1P
zPsb9WN-MBQGBV+M2W3i7Y$WYNoK~L2$Hxnmt%X9Qeq`4o5!srg;O6)0<-=n8Q?&-=
z@81)FDyx3ftLf8e4vBR-H~46u%i8t;Ho8$!p)mpf$B$MMngGaXczisTwAwV!w%4M-
zWq#qxbMQxO0|^{LNYH3K=D?tnEXjC^Cs}cER5C->!YZ7h1{Z$Vi-J4W=FJ8i;_iZC
zgxqIjL-nTzJC5+)k!6d#YMacuOV(EV025PFl|OZ`+09JNhSB=)vaDDYBk#N0=f1w$
zfw_;3PMBqkWDD_^UJ5(@ykT|H!+%sg^np(p|3guQ53;Pkm@w>Nw9`R#_Mm^269D5c
zjbCNOw>Ov3AxX^5-@SVm6*}3gUnQyas;+JmBDK->H|oc15h)IHeJjNt#(w}^cKn^Q
zbjI+aY{WcBh=>%z-0N9H1eM6MVs~B<-m6Vk0TQHg@nSu%)U85mmO$Dw|8ABjyn6&_
zRJ?R)vHQyt(O8ulUB#C^ZQuMq+Kk8((M!M@0U)fb+ZQ(Ku4igj0~Lvs9%93!02897
z6JdV3AbCRUKqvdX>Rbq8Jw!n*ya#Zah<+lA0pMPRk%aLH4rs0fr$S|YihYc5_9J9z
z0}{`={x*IhpLE?Cb0*h`C<&q*g|qp+PhY-y(_v~O8cSrV@C!A;N3Wog6Xg^!$!$7b
zo686mJ>M_+_dWt0I6kDVf99INWaWW(L@toXhC9V0pz99#7647jC@c+;Q!D509wcsY
z<m116yM`#uC3kmmB0mW6`_1J@WQGff@cK2#k|2YGE{wiQeEP5X;K$fU`)FBxhM=P-
z!CfK9J0T@NY2x+VaE20H^ZC&}#2KC(m5L*JXv`(xAn*2mfJ3v`s&JHmJ5Z9z!SZsY
z<E$q3Dq<uMEpH=o*#}3)yf;4xfh$6GCW)hafc}wa!dHHFS0dl%0(RO3ogCQ4Lxg@Y
zm!hoWy~RB4Fq5j4vs2D%qwFEu0Tt5o7km?>{^$tHUs^1_Uw{TPA?;-OzcubiUB@Hf
zG7t@HaDT)L6J!{K?eXKs=(2f+lA|pXo#qF?{{W8AzE0hEe}4}($9tD0L-g~_F(P);
z-QSD_eh!w$gW9MeS$O$O*?;MNBH@3oQ^y-)Jj9*g`4KY+STrVdLe9>@P;`dhmCPlg
zpf9f+p`%koeS+IJjCZ^<_9em{;XML00aBsi;TMpAB+4TZvL#@*A@Yn-U!e_AWZu6&
zYgsgQn4oWn89S)|{6x%iz?_~$Y7tnrF$kX>e3W6o#BVB4XhbRrTm>g0LdVs#x9;Lu
zx&tKpytXz8E*iye^bM&@$Ob5}yFnsOFeV*?sG1Uc|L(ne2WV(y0rCG)M7*G8`XPkB
zn+z{#X!IaEK=lwr?RG&gk^KeBz6a}wpn(HZ5e@)(`J}fK#uny&W@^F?!3CCHW&92i
zkp{hGq+^fuUr^_m6&Q$4Fxm%PJ41hw>_H;u6Go9R-0)K{XXIfY#K#|%Awx2ZPQaMz
ze-oaF&J*M`Sta%o;&Hg)_|8c5A4DNQ{M&tPDh0(XF)<PKc=-o^pQR8XC?bv8HS0(a
zM~L=>bAXeN_$@V-0I4U0?})Vh@0reE=zMN!Z`<Fzl;RtFhBJxHwR6`lP`V?y0@%yQ
z3NK-m-@~m%U$bZ5K0vJ73>^*4XpmyCzdy^FGxMpw9z^jRA+o)>_ZklKlyBJLIqmy|
zm}qI)40t@D^+6^>b4`?4Sb^x6z56YVP^0L`ecLQeZ7yuhO>bxMSay?qJ}dgPb0w;a
zO?-A|m1&-<ANO?P&6$JKdB&~(g>?aJM0i^-t9ID>EE4=5v`-UJa42vo2rdth%>XX+
zJXi1G{Rr;_eT@OQTTrp(dGJ!ATV7?Ak|D!{K{qIcjRnKeqrn}fa#GG`TT_bT3K||&
zn#-#`bFI}RI}d^?0?I~U8WA85NQAI}!Z*OE>d-oheh@i!Z%|`E%aHB`h-k#F!4?J?
zkRy8lqw!rji;r|tD@qF*EJ8j(m=W~_DM)q*tT<p)VA=4m{{8#c2L~Qf#sV5)e3Qc2
z^%F$40e6jUu4T-afP&?5|BpPpE-dEiR7yNCDDW2uOb9P!Gj|Y)24MnKi>6I~*QlT(
zGS`t#OQ3~S+45n~kVFt9f?5Lhjw0en^$^7rW@&v(ivrnjo>le@xE_dE+0Ev4e#^Wm
z1lI?0nTTm@M9rsJM<OtpZ=+7L3*&F+&y$FVh)@_7%+cElGh*At?^En&#}M1%NWR5c
zI$|sfZxy5^9&g*TM-f5&v9YnjeoOd#tuxC=CV*7}5#bFpu<V@a%t}d4t|`$(%Qw&`
zuE0Z21HGC0iueQhB>n%j0%05-{WEQKq7q-20KUUt^AVjjG}|rL0yRxdU+e;7@gkZz
zh2aUmO=~vK*Go74(@5(e38Ec^PECkx5Ixapa2z|f4XAYWDixvwgaY+nKM#3)dfF7j
zTW~Fjg0`Qz5dts~HAZAE=~lTXkvSg_Q;lpODi+btBXEtW0K)GSb(pR~&qXjUB4>U2
zbkKS4-9%4jWtoO6Zp*qQMRc(5KCYRfM;^1FqmlbXN!@W=lbn)LKB@I9{1P-O)o4fp
z8^&09craojtEfoE5ng{&AxQoLz67-4fVkF9#ySn?1Ch`S5`k!<>&8gYXJnz}OifP`
zeFYJt4pm@2Pgc<+G^lj<E;~GTy_<j~T0&XJY?KJDT`5ld2%I8$E9@9<?BFfu+}E8m
zCh!%!lbcj&-sqjAzO}C|^sUm<&V2bK2&UPpz09}ldd5E67U>YB7y;t6v@~MO##)#W
zqLClKI~W=8cyH4EVeoDs{C>HQK^8R1Q2No=eZ>?Zhf0id<8Jlw&ui4T#y{t1XJQa&
zWMV=A_TJ&ZT|m};ONIYtdc-AI%9sPIWh^au2n_)5hY^4X_<wzt1=oLq{8GtK(bG>$
z)HrCFD077Wc{e!g*tz7dp{|ZIMn*|lV=REMus<XuP?s`246dOyhhV92_3G6!QIt1y
z<S^cljA-W$ZZb?tO|9z<1iKy(?J*aeSH7ytwAItqGwX<3B&4RLVcQ~vj+giDWb|l$
z`az#B3Xw#}*#wB*zI}T~Osf|dZ&=ZL6>ii`V!i~DEWi`W#U(kpM^2A&37!FlKnOPA
z8ABIauBxk}4edJ8AtA1IHcDxEQFe8GeMkGuX1j-;re+Ola;eaMRHK1&XQp9TY!`n>
zSRlk0CtwcoV#Le+AP+I;8ZCG3{+X3^j-CS5#{grO*p{`?`?#S()iS$f46G<TM?9E{
znp(~3q?ImHa7ID>U1G+pbg5ZvfF<?sUuTRYK1ebQpRzAb8dx{`R*8|BT2@ZZNzUPL
zMf&hI73oh)BjUDizp;50!X|~P{k>O`i)-pd_qF3&5}3{<#E_@|@n7%#l8@8U2Fyz-
zY$AI>#n3>^Iz0b((|1@jzUmjT0|-2x|AUZwB?TLV-SH2QHb<S`mQ>QpY|I8cJb^&7
z#MrZitq=2TZj{h#j~wym5n&4{de44vknR6B27Lv=K6fZ5|0BB!TJ%(*dHwqz=f&>$
z3sd6C$>!QS*&CT>DlH2-11VO8rhl-3E*~N2@~gCy*z=J73gd!@2Yus+Nd~BHzp{GG
zur=T|p@syD9V}5ftHVS4XTa?YTAx~)|0Dt6f8v1WOn#fOfZpKcemBSuL;fe?0FPT1
zF2VOEvgVo0qS^drrgKi!-3<s!mWFQ_)C}IfiuQ0D(N@F1BNQEd&{vEffkP2qJ<*O4
zULD$2Vhj_M6ht?;{_bM)!?oXSR|o|j1~}}%{^+?RN_6w67AP-oQd1v;h_#hCS!ur0
z(aIvnJj)QqIax4`iidL6`0Fu<@G2G-r!iFeHZml)xcDX5Eiq0E1Simf2tz{3IF4!J
zi{TvEUR$14pX-Pr5qOS-8+|*s+j;VpA2E6C<b`y3r=D97lpmx$2xP>l=zi>_3uF}<
z*LCGkCW*a(`{00o<j3s#7z%GJgwz;_Zm{5#p;hAUS9W&^%8C1o{*;HTxFb5Cdu4{j
zel2_^q&Vgz6RE(7H#g~i(6LEFyE=e*79*JOaH$47{^)|yV1Khg6((hNzVA<cA_rV#
zz-!Bcz-la63vm$`SQMhXi$YIKmlM_09FxvaN8`{UB<Ov8oyZAccoB<n4LKHqJ7OXn
zL&d{S#pXU)7NBKgoGV>YfnVPUYD<7HiZUT9Vm@av)mIoUn&;*k5VYcSLEmRI81?WQ
zijeRlBzu$+Cs@tLAg|z+<DfCXfB`jYe0gprb)W*mz7~uuyaysSRnZcE2|^62g4-iR
z6UZb)Tn)4Ar@+xr7D9ld7`kaB+Nn}!UP88q(NT)IFGOFqUN^_f+UB5TyiMLt-7QX)
zIXAkqYWCW%{d@~fY^qKJE+>e%*Ufx$eN2^tY57nQE+&Rm4<0#k5!A1!sEBa%P?{2x
zlBT|AMmv!Ke6x*Om_hG|_6Yh9gn*<p)3V)wLU<}>5K8V>4ohojz&5|4r^jseJ)>gS
z`fHNHeqy>9BVNv{W0Bx?um>&Ar?hQasOQ1<!>m2$ULuPIuMY&5QcjidE1`Lw(kTqV
zIPyz~2YVr|S?yfAh1xDb*h5f%lA(?eV{ox5wJ0Q|z37F%;$o5F6@kx-Ilp5dudgsS
zLx|xEL+8=!-oR5b%#C7}aE)m%3RhIwg~R@`U{8`uwA%hpORxFf#veGZp#ck+7&js&
z-f%hZ3M!mR`UhS=r1aZ?s*8GAzV%PcdpE@$h$|4tAVS*+GeS|Mg2IL}`>5h}LUHk0
zYKa>PGm^rfoRFjT;xAx);A(_>PXwNdq01EcfOlA9<hT_Vj>-UM0xQoLu~$C5<C>lx
zr8I&t+1i=^8P!dOR;$b|qX(c?Ro}EKt0pQsCO*OCxWye%rFn`*4-#q=B3%R*g*K6K
zBl{sr06a~?i-nYuh=3>2j)TqbPidPv)Q%7iF?$DvV@8>iEA%ft*Re2pIaM0Zf2&3q
zunI(vK@8V(Yi2-6KXCNueACi1R46p~*lVNs=mve9qyI@AfGI(hvxdoX7Bd^59I%T=
zaEeyOS<^0e&ch4D=C4J1QW0t~HHT6HCd3Kt9_`h3Yd8$N)<e>;IlNJ9y|(_r^5eA~
zRYdn+jg|InPrQrIkKiS^HKG3#dN=Y3IJ>k-t##1(717H>7(JifVcVuG)QUQSE{BMo
z5aWadKlaeLew{eFsMmy*kD-2cSdc|_6Bq!DKMUdIUFqT_f`M@;zm&z3qJx=5G>OnK
zw*8(TjE5k_>^h0L0t#f*L{DDumi^Wh<5*9_Kpmy@>xz4n(t&?5r+-_uXI{O5p0}^a
zE=tHMPG~}lNx`uG^y$;pW412gTs5_|oedY4t;YV&%oIQ4M8s&oAhm6()Fl;DLuE&<
zJABsYZ2N1*)Nd&SIUNLU{vYgEG6-PfysHI^n-RqyRye}nUi=;%QcrV_PtN~!tH?l}
zDhNF=;!S@3bNH-`$C(!<K8r)Tg2oe#RuOCfF89o1$pPDq<-9~rqBVy|(-y77i;s=~
z29uTP-Lk6tS+{IDy{OI9FaRP|4PFb+4X}e~Obr6UTaN2qt{xc%uQ<P`emv%1MN8y-
zG|>p~-}~}xq$i)pj(?-uPMB|fw*DL3Y@|q>zO0j6+2J*lSs+;V5%X;nM--i(BNN!d
zS!>oG>#WujnI9}q_Mmb&q%4V_rhA9T<vp+z5)$s0mX_{bw-IBc6rc}k@sPUTxxba>
zZR?C-_-6S08{W4!U!-%<n?yG`Dh*y?!X&bsD*fq_?AIot+ijm;6WW(%*)*&6#xm*(
z#yG95-?#9OK(yMsZ=c7<gctwKX92oQI?^IqW->YVl2S4rL#+S`$Df(izIs&}*$I^W
z+7FqWm%>H<9uEqp3CRW1Z%Z&!Yv1+bT(v7LK4t=czw1=2T-2-(JDdFeR)J$c#qEV6
zi;Hb+9>>CVzj^z8)c8xDw2GgDV)M${<(EXr<-bWjc)>(y>NTtud}>QrxXj+igkh|*
z_y7I}qJpjyuSB)$tI559wYGfHYxsZ-W#uMbDQ-mku;&Ndz>*O}Sxq#?C7h~FHg&dh
z;1L=7p3HM8+_q;~+J}mud(ck~vg{-TmHuLf9pc;rW&hS@h`2f?jfqo+55D*dbIlp6
zN+m;BrD$zla=d$lLB;THoynne-H1N%(Vr^!yzr7}8;J(=1-}s`fgwv1Ew!_bJD~Yj
zB8f)~GZz6|Ur2dLf~-P8kHl$?gqpPgp<VUH4d+G=vpntij3?jKXY4w=Ez&v$U(Y5F
zn`gSIbvqVn4BU*a)R3UQNG_EvSK9m`;>Z=4^C&o&WyME*wB}pf#Ank$>&4olVMBih
zPmOqMcu1UQ&S18*4Ik<P7huo+{VAYGFlj!unRO#_C_@IzjfgeE4A)aNC5J16b1Hd@
zp$}q)#rm2|=FJ`rvInik=3&PzW*yg6FGPR*bSE1u2%m~ziSNyr3yq)Kbi(Pyb9ev9
z`YMqV%{(k{ch|l=ZHmG9Q^(WaKRn@ArKYAE<QkKY*efd1(9n=P&zpH+O!B37AOW#3
zammPm!Uld%u=pYwA$1|I1ntX(;Y+c_BPwG7FUJg+Ys7FrE|ky&A?vE$xbe7c%KU>U
z+HLIlnTSeSs7%#&Sfr!|C$=ZH8sidcO!IzN3%d`56!%-=5#dufB=h#gML$E_g_s4E
zl_ox}hKHx7rKKe$p9DVvD_F8z3lmYf-A{nc2??Ki4Fx#pcV(zVxRj5dJ;t?7%IwJM
zu6O^n^#r$JXMY`fd0GsrB~-8185vK|@Z~@s1R~>~6OxmCajEdszQ9i;yi=k`L(l~G
z$b~ZlnJqOV15@(@CFNhgUY#r7^4tb%`wE&OSSUf-q|g>nQQdit$&g_%UIl#5!?fTj
zM0>2JKgK7DEFv4^P<DyzXpsdzaRHObn4r#VvGAU#d4;g0at?E8!&oAuFJfr5mxr4>
z37@nB#9M0|-?C_`y~Nx3Cx9(qi>j5^=EUz!w1E7kdxp^EfCUwyopqjmo3ild&jjkI
zQSDPE0#9*6OHd9Mz{S2Iz(f>3*hKId;khtih#TS}+5Y`f;~!&HQm&}PQEk`Ih<;7V
z>LX3o%5IS9uxz<^dfGF;+c#Na&Ztc!E3d?Py?~rY;>1(UribsvuDI8|=jwS>Gx~v=
z>OS}aaq#A6(3c_HBC~C-aD)=s`Fz(C$cRxAqa>1NQg9&MZseqgY@B2e`g2h}^R&tI
zkC?+7C*CY5U9-kVO05|<6uKC7oUgGrQ&D<P66rclej#>m&hh5}Ca>Yz*j>-syNW(u
z4@~CyV)8$qI<-n_t|DZWDPDJPLO^P`Fqog|KiFE_?_S>DwqB^RBKh=x0BwWfuwS&(
z^`m!xi~P#@V?(aepmhV7^Otv;W`m-KBX_n_HecD5g3@DTV8_;nr2C&wiU}=WG*uVd
z(jX8r<$%onlRGi#aCp1Mp+fZs-E)aO(}`cE6MDv9_l)D~*FDp${8{9)htp5IXo(z_
z+75kgPk&w;KmWJde0+=)H2a~U9{KFkJ5P0M)t#H*_$ID|izDu0l!kXwVDen8Wkm+1
zP@W^LVM(C2-$b{$lbhaA&-N~z_O6)rF7bG`TfMsNJ>tD&B%Y!a^4^D&R037ADpEV|
z2pHRF$NKCPZJjwkR5JBu?_LeZLmfY*F_GJs-!xY>W<T;>Nag<fw9j%TyiaB#+K1L9
z)d;j_+@V2cu5zfPA*gApW==i3Nd2dZkYGwc_wW>nx>oEn^_#-#H+)$<#3D3FI}V@F
z(8QHd7f=~ry(+MmUR)$bV6QRFE;*7+`HH~=wyM``Rd<h_>{3iOmSEwKsNj%z#nH0$
zK%U7{J4hkJo}S!qOUU@#&)$;gXRi<cCEdBUAw1ht6q|3+li0B?7{w|rr^OYa5y74!
zU}?%)PxQ>T^0&_#w;t7FQ8^T+e>C#u(6x@rmhCAwm&tc^$Ea(^`0e_{e@gvcCiPBI
z!xDy)%hh9=Hq8Ix+D?mS?cX{>HQ+mVE9?B3zN+<oFa2JSJ>A1mhfAC^`K#>8#NmC|
z%lmNnz|v2vuXp&2f3?LHZ%HTJ?~#3Ak+#=T<139GRl!N)#8W53DhWoBPbOaUywCD{
zW+ndzH8*GHlpal?jgcd3Z*zo&)vrFa>C|kbq&-~HyKZh%cvh_M#!pLima|7b?EGu;
zOK6hp?Bv3qwu8679vDhFezH!{_Sy4eQWEFo`-lDFq@|cD{C2U_rLxss@zYB=f4<@l
zZ+Jw+ML$XtWxJT2l8oYOffmgwy$AQhZKLo#pFWh5F2!9lrI{_-cO%9!W2Lf%Qb;pp
zUtpz2p~Aw1^;P9qyRLB?s_VTm?LE4wY0@bjx4)j8{6PO(n?f}6;raCQDIE!I_ouDf
z#II6Oo3mH_o;Z88?m%#{@8A+s^^)*L`HIyB-Injelmn59&4SqY{|=|ztRB8*!|3T)
ztifpcpmOZU-n1N7-v=l63;ipSyZ0hiPB_9m?Npra;G4A6L>@El)W=e{_C0v`us$KP
zvrfT}&V-3tHud6_kc$eb_0kFPlu3Ht@p^%(6E?@znH)LRr}B)|Y1dcgk}dpHB&AfY
zi8V_qX;Y_#j2PjT^?ve16kIoa&tFdOr&c|_DE!1iC75lst}|_$m-lzEMrw0``)w-5
z3895osWX$j$@r$p1n~bP*}{_e_D%5%awMJ(^Fob$l`4^+=E9VVMn`Sl@-SMGzWIKQ
zFP*PG^~=!JWtN!F{<fpy`XqVJwKoJjmB?O7-?@6H?NG5l;&i`q%2R!{q~=EwKhPay
z-PprBUDiBwJ3Q`oc<p{F_rD%xn%<hKUm2PI)i*bSx!;FU?V;z$#S}*U>WiM7o&|GJ
zUeW0H(J-~x3n~tjS{+(1Vhuk((U*O}N+#r+)HXS-dmEy>jrLUs<r2!qDI#?q*+F{A
zgqHSm=}d8~y3TWXZ}a-&l5(ql+#PfIYL=%{8a3r#6@@%WBj-CQ5p3#qPCMvuth}G9
z5j)G-2bQh<biIzJPCvX!^@CRCJqw-JX7&!Uz{P;8k1m;8n;U4%N+BEXA33=H*|$8o
z$%~}Rf@9{#5Y&tRzQ4fQ!0}~v;VGJ^z`E4NA?Yy^N=>HMjoQB5CdoBg-c%2L5_Vpb
zxkeYo7giM_`!ed~0VXQtXH!*?w;lHmW>~u7)6!(OOWOnm_3h=3bg~#|mkU*V>VLD4
zE1xD@H$vIbLD22;N|;9IBbThG(jIZF0xt<~)$Xc;v~&u8Cg<39p9rIgk#)Z<E+3oy
zHr=`gA0D_}T0T3<G(Ft%qQ0^&O}&;%P?9g}<pXl#tczw#A~zgD8mE=8j@iPecTVrv
zBe3(wiH*nSs%<@&mM&5#A7|lwAbsH&N9>+5`N8%MVQxSk$?Xf8BER%G4&MDH?7!c-
zkV#!E?X69F$gSv>2VZs!yh_3Iv>dK^E6Qt3{(;Q$eEL^Sa%K8UGFP`BNG4-dIv|`L
z)Be3ZEF|V@dxwZw==N`9#vk@-_j{SWjfhykG#Tys!=M5mvN5b4+okKcw{h29^?TOu
zr;l}Obkp1n;L43qH9WoJ7_O^oNY=G_g_V~x^lyUmj~yfadmgX$dPD|X`)OH{mBuQv
z8{-%x8i$@Sr~RMS-a8)a_Kp9yl37;B9!1N_$clubvPUw?sEm-Ul1OHvq=oEClp;ce
z%#^HDNR*M0Y-Psxb>5%v_rKqtzx&bS(Oq4x>wUf7=Qz&ec)gxSoqJJw=Z-3MwnAmY
zo;B_Z)+qx*2Tfwlmp5(>2^EqP3w|SA87{RzZ*1x&npIsLs*!D2y8h|sA6!hAF54=u
zw5~2c6e+w!=hhB@lFHoGUKMLodkNuUu@2e2MPhe<|8i}a>rCXc&WYAB+0s|<k+G6F
z!qap6j<AI823;|=Z33BJA~gc+={E-^{5$I9`XG3P+CA(2P-?+;C5y|C6u7?Ww<ZZ~
zqg@G6eO}jQZ(q>;&Zc;6(ouH10JCvV$cBIYBdzKiOwuL`629dBG~9aNN}NI|mBqPD
z+b$T&34N;ZWlxh0k)gSkZp-jTl~;3hMO}!`#F;rfP1>VAq<L^LwwZ3JrbT(X<eKjX
zec4k*yWhM09Q-_(uMi}$DQwf=J@33%t|M~FK38KY0x7l&Z)b9sx%cUYY2tc$@lL6G
zwwltHX2b@rX)rXXr^z~v1ZeNH-oSOpjqkS=m5Lh0U1rOTVcyTeTRBr+J;{kK)qOfz
z=1;OkK0zJSMykh-c>CWi_&Lny_gvz~eMUBC@8e8MVirAaCgw^qc9gZJ)&%cm*bcfH
z1m5Rtsd1+bknhadpuVHZl5*3rpbE$5VqfjF+kUBrq_<r$UmN!B(R8G#6b*gL;bjq7
zy~?qrl0|wUpyEl}`3vd?SA%vrf7#6P(etZNi7A))-Ax7UvSp)W2Ac8-JyDK66tMMl
zR)Xb03EJe#bRr|%9dQ-(fzAS*_eX-AGX7=KPcu6IaskVQ<3`XvqgyqaM=r7SuvzK{
zS7h2QbLEXYEcj9Sl<<|tTsb{n{(V;J6C3Bi^#hBJGIY^@#a~*<shIQG_pEKWS+OH^
zq9$Hvv|r^!XnXF9QA6f`1M=IJxip7)g(}9P#s@tNezvamES#~CW0Vo#yv>tm7Et^4
zxVJ@UPL<KvV&$VZvz0X$Ycs!yw(?LftP577V4_U1lbal?S>p|P!_9Q5%BgTQ^p>fH
z$0qJ8Z&-dWeqf>Ck=VviVY7R(@^a0%P^CgDM&ye#FO*CJBUQZx7GExMBr>+82}?@H
zcD}K|>hrqY-ioGJNK86>j(^n0se>}N!uih2+j0`d`}A+64YfM%_9<KUaczB)day+5
z{>G}-0ZyG?S8})w&6pDh7W5KA=H~bk_IDr8cU@JAIKlP%^xGNh(v1cp6}|%o*%Ht4
z3yM4IZ`B^jGu^k`CSJi_S8zYug-(lxX20p+bfAIn<?eMk?X#7?W|$aDe+y8)Fb^qb
zdunF;L~-Tf!rF|z!DPcYO)-0JqyGKROaXGe^M=U+(^HIU-^P@_b<5CRIWf|=wy{!y
z<^rVdYwD@jSJqPNhhJXO>s?;VigXS3vQom*c2%No@YCSANA>2Xux(tiD#60W_ITr#
z4^r1|wQ_`S*&w`CF03S2aplkI!q`gNwznzV{l+#<Nhh^gbWK34JSh{tuVXBI*6hY2
z?~06|TBSh7S2ngl)<Z7d((450CqhK^m|j0(?pZIMH=4D}{B@GPb+@##%{?8hmFZl$
ztj{jf?K{M8-S?=evb0}eE^Jw3u-tg%#=^kr#EEH>xdmtaA2&aasBYAkdDHk)E8jG}
zEI>t~`$0(i&}f}vZc0k*%U^jcW&#42o`k%=uYN)$SK!3;jOjrKBe@iT=nTrA9+TRc
zvva3^T~zn2c3|2{p`K{gCG?nW?sA&!xeJEM<p<Bnh1b;h)5eZH>!+D+Ur0AA?6g+M
z;WIPsdeV^l?$n7owGE~MLRre|qZ>o4d<Tqe1lF{i81_0`*j+Sk(NEn^zbA~{o$2iZ
zS^0B|TO2_u)8CX!X)HCR>a|BiK}yKE`-q&<iT!uS(q?b}5bW#uHfr8=)U@$t^173^
zk1d?Bdfw*DLKzaWPyBM_sgI`|So4!bJGq{vTN`=4T2;H8#1aCF!G3@D>4~En8c&?s
zTzs}&-PfjQ#34Sivo&!;e!8vrQ0t)-uiqL23L@s}zow^Bg|228npj=Ye^PUD_o7&n
zW>4O&4Cq>Jn|%%~t1h4Xr*zV{WS6oK1sxq_$mN=2rZ%xPn$9nSHe@Qf?5%N+wan|N
zHfq=2G&!+J`wFts<vru9JmqVBcEvR<OO>mFQ&~Fv$IO2=cnN;j3(nZ5eO{@QieG+$
z=TPp$TUL!SzpiZw-||u9@j5QKu_OC!z5A~uvI}w>lsxOR4-Ea%dNN*g@M-0vgA6*F
zA787;O*W@QMmwK59({U%H_uMe=hdCx_u^J8`05Q*7!?f-HCzfP5+jx~R~{ytC8x;K
z+INTByK=?8%dV>Ue(jXry7kmFeKsO=46`-k#@7>dpF-HVq5P41iMFj@!n<e9A_dCv
zUI!H{9Vqc@^?*2im1p5q-7PP+P)7(e_@CtExSO15I<ClCUGV8i%NO0pDil<qIU&Ea
zO8CVq1lL{vUd#R4xpVF7y|w7I+CCp+6D!Sg=N9@xZ6h3?@h|@Vcj6CIxspqj-L&fc
zq0H2tX3?D&YU3ZzSLDdEoTPg*!gb_snwDjF(H12aiYBT=yVj-i&(`HqKGo0T3$6}K
z*}dZXU0hd3UxCM8c_gss);8`ZPdt(&Y@;6<y~v~Q8+8`@bLLjcKmt}-+5?nyZkNki
zr2oEr9oJL2OK75OqptMp*ygqbwejb}+1&?QcZkug4ZIc`_iO01&{}%_DssPKmcy>>
zy&3zgDvi7!bDFy!*Y(IqnVZ?`D%GWLXGR-tdpp*WI=$ZTYuKMmO|65tP?2xk`j1;P
z&Id@}_vW8_HGJz=O49Aawzfj)ap_rwsqxj!W$pg$&m2@zll;1@iHG*fxRv1AUDJ8H
z6EOpk!=kJH_r_W-6#P5Ic|Z4kyDXbzXxFujk+cl9_g}6oP#azsR%AA~))PJv+?O}C
z#2Hk~m}wQ(6Mi7)5!FlWR!=7DmDi5k#>N}B3o%C?-ZXf}%A)Ye&GDT<;!mn|gVzk7
zy*kH~Eb`ZLta)SFh>XP_W}nJU_GKQbnwu{1g#Fz+SEwI#YVw!r>cTl{mAh}_{p|S6
zJR>?S@}q1{t>5-oP0LwZVS-<N3r+n^&ba#EP$Lp>?e`F2U2O|mZ4IL+W>~&Nb9l24
zci{U!{*PnDqM7_k_!U3Jj|9q5_!a-Fbvsj-60yqhPdz88ynD|_Rk5o`s$ERlc&yL%
zf^v4YgbepZ{<jYd&CCz^>M^5e6I1XgFC%QFY3dKZfRUh@$<>AvH>f^R@a5!99lpTY
zIDaeer{Ow{3su_u>?M;SP8Y-?GCN#?<wm?VpL`Y?Gg*~kr7FY9v-eNX+<Uu&&p2Bt
z+(UDWpIr{5l~`tC=>AG;nEXA0`t1B8X`zUWdxARC-OcB7_UD)!=8cKFan-tEC(WsK
zX#%k+Ytd&ljnW>`m0B~W?@nzEVvFG{7UpK<pr)*7aTt=jEfiv(WJ||rnzW&~k%qE~
zr$y%pV_(kozS%RX&mTs9(qw4Ry=im0@A{QfD<uUt+#+^F(Mh+8D&1G1eUO%|s`Y$g
ze2VVP37v__^S36h%sj1W{v?X(LcF~<rPAjJMT__vcW=oumyJe0ida~<4|{k}HC^fF
zZrvHH|4it8F?0WqB89@p9ghx5sk_|Tz2DgQO`xj($p}Y*y5uX@;ubcFeu__AXl2<}
z9oXr|F1NLxdyjUEhVfbM<tWEHTdBfZX66KXp4ArzCVB%c0s-m*{Opi5oSqqcS6f#{
zDIh*hD$4Mru<rK&V}p22jtBkoC5mv2YZR{l+<WFUwVQs^%+6%j_s?vno2mY8`b^cj
zNB*YDa&fxDtnJE11G=vAa_-Qf)b7ECFD0T@CKHb{K3LYLU@l2?y`Pxb`$^fw<M9Tj
ztC!7X#$PFV?HaG?X5(eiEDVypcCcA>l-_OQmKS!O8&2%X-Kx4hD$RYEh3%4MiY1)G
zy3xJTe?=Eolqs~bbW^(g9F?+ccA8jPzK3zM{!_LI(PI%+H1GlYjX02#M44zq%ieiD
z<Vp!i9Cb&)C<Tb96q3%Ut4qt<+#%yQ;JwnUCa&MxyQ*;PbHb;urV6Ky-Z<*4ZMLqd
zMsMy&EJYBt`mO3>?HxH&Th2Y%_C%;~2Z#3TJVORu*p7pn5C+|{ID`XZyF!Kpm)xYm
zAt#3H^z1FPVgg5%989ik_BG%+)_wTcw6^m1*Ci=_feyO7+}DF{Ox8b==V_wn{~nIy
zZHnQ_VNK*(6F(@Ndt|uhv*%HcpQ-WT3L<Kiy=(zhFHcl+b=dCO@n-MJZT4PmljN-W
zexR;?|8|9*l^EfONCQq}6!f4)f=ZNdJ&AIU$jRp?uj^-tqI6Em=9_k!OP~J0eIS-^
zwC4n5U&Pmm`aAh+NM(`5_IS1WYB@X2*1M}g@rK4}GX=|-J72_@$5-}h>gD7a-MM@C
zoy>iI7-n<c=LCJLWC@{jpwScwZ<VazZ=-DEetBTU=h8O)bw}f(sl(Rmk92Oo+-jef
zCib>z>0v-uxmfC7YJ0k~*=cQz-zMs<yv>a`P+LLO5t5zdi?0meFo5@9{f7IUGH2+J
zMTfo%1=@wzz1XFVj4tQoaK}`f@O%yvFGT5S?Tba*A<u^Z&{ZMHajE2aY?%UE^ZBct
zo95Hk#yww@?u|5YP<=*eTYlUC<|TT1$Jokg<U>K(JU;1Z{5jjVrw$q%ILfw)iP0hh
zj9LKH#bAcQ+3*&1_482S!drFSQD2Qv9U-BI_7j=u{>e!L)bQQs)vqTr2-6wuKj+mN
zzs<>zbwx8GLtuG+Tzu=+b-5FhlOH%ZLZ@nvl!Wn`J<htHf`YdL!&GMDi236hAHHuE
z&9SE<Ja)ZoJFmkhQKq2j*qR}5H)yaHXg1`{VfB=g`!vj)-99=;V-y~~^hq)p;9&cF
zIo<Lx8d8abktFpiEjR+AaJ>!}<Z;x^E*o*w6~DEi8&Tyu?7>q1Jw>?ICU$2==3A~g
zDG20N{<S={dycd;q|@6`VRU(UGdX#@{7Fu3F2UKszB2pC#GubdnLvi1ZU%7&_Q%)N
z9w#TOo#Z~kjZ6VQkS<EOc6KpI4u=M^gtw_S4ELA`(;(#^8g@bCyJ&b=>)&q{MN%!N
z=3nf;>ZdX)vnyetC_To$5;757GDP;PfdjPlKfR3L(m6TkIn?K}4-j2cFZGPi215b{
zD=%saAVol;G}Ak`$fIqPI9M`V?^C1Dzn9I}VAsSpC>@6PY>s*7EG`7~5HWF-O_VYM
zkc6sqBhl8wMg>Q62CR}q|4b)50@n&8YjAcl!t-owZT(@SEz|kS@6PrmxE=p*emzi9
z2(FZ}Szm`44b}ro)^UfrdvO@q9z8kiCYm3pXd~sW;q!5N_?NQ4Mde#lT>neJt3NL;
zX!45}@GowUfAxb#DZZ9kUz1=ek0i=Pp$3d8iEAW2We93euZN~c5z-&%#o$lJ%f4M&
z9`AL(*mF%nH&CRXs^rJehQZ0_C)=+dTRHpJ+_+4B00&(9825YbFU_;kKJST%H_Ylt
zYm<)=wdxKnqLkA+_pQa{;_Vm1azhy0Q;`tfU!C`bKe{*o2Bxt<T9g+GD#v|wezAA#
zxYc<l`Cad^`tZ7hi+jCgt61x46SPkjbhWz3kNi<$HvGu0xoL-a^zA^(q>Jv|xuyBt
zuD}1zm%C9EbODjSzDH7DSnQ!)z-=T_JT2jm@idZS&`lQ6brNj8+*D`EX{oOiYnbV@
z?fEu1oS|eJnT6fg&&2qbh?w%XtTU%}Zi8EN$If+488u0@QO7O&>yMB2mvt1~{Q7b;
zxxG5Kjx<U-Uj!q%&dHbO>)bWxynm;D(os=UMP8c<JlmREs$)?6QAVO`xX8_K=lYGM
zPT!m7izxjjUw%}Ty!>L<dfp24h72x+kIo+6%FHgYsg*~{gq>B9EahSPyw#&Dv0it?
zFFR)U-bJ5c=9Gb0DqVij$g}+Kdws)PkKL@nZ$5&FS}yZWJ|7Q?IMa~of+JPLpTm<`
zTXTa{x7%cNR;;i*;o`UL$hmld$v`AOF!acIF=xX;H&jSy`vqNpygqE{wRpvb-uZNY
z)}o5VOtbzW?=wlIPB&%gQ2ltf`H0043Qc#{mdLU)+g#t7rVmX|j~AwXWw0{g((|Gf
zipbk?vmm(SOZqjJu7qn6TV1X1Yu7)rFJ;nAvzb(666<2p(cf_XCoYSgCpX-@CsX%O
z^vU5IYXymS!hw2}d&9T3&)k0;^g%!(j*pRi;IZ)D$3O2gTHg4&H&H@;IBzAh=fPTv
z!S%j(wM9%Q@#wB<>Ig<^dEH^-IQ?MFEcex1#j_C~o{Av1ot2WEdS>|#v!lfpuM0sZ
zr_>mAgvq`$RNvaiis$Fz>52=^>|>QGkuSe=ebSD<Khz++D++l*`v|)J?$G?<>-!b{
z-K!(Fls8eUd)B_q>0ywTmmE^ZY*Adf`kE0Ho)3ItPcF|`e_1_2JNjgHR7oVEc7t9)
zZ`9zNHmP(gF__%A?0xWI>(DoDh1Gj?H~szEUh8r1J+KAZq`d5CaV5PQYUpn=Qyh9=
zS#X2a@6h&%p|#WQr$2KoXXoZow^BTz&{?f%5Wssa9<{d$-|c1ZxLe_Fo}H)IdIiq5
z@eEGID?Pi7298}r=2gqfdK*vJr->$=-z@q|lZ&p4KUv)Sgd&ZN%j$IwE+310?Rx_q
zRYI#CWH#Q=k_^b(Y4lF)PEcXQ<niu9tC!zDl1!|9i;@Ihqit9cH(0TH?R7ExrB!9z
zOWsDcBUu@Sv?3`I|9eHuFHWtkw4QBl{I5oe<aRJfT>SS@?90DNp)WiKO}Gp`HjsxR
zqO{etcwd`Fv|P*8$mKnp1*30w{dL=xF%~A_)FaDNo?4V0z5Rcq^d>F#kB`s%cfCti
zy&M^IF<xV7^86fk$`RwY4UZ&K9$ZGkJWhv)&mk=(MKs>o+g&c{()*!dfB(2vh5HjV
zHzN-*xnMBx(2Y{#3&b4Ikqx4~e&%gMl#Rz%>x;|pIn6H>T~G9SCDNY|{QELWeh0*h
zPSJ@@j#pRZK6G7Bi@#zOJ*ujB`R~8}>G+)@tB!Gw`XX@)G|*+8*XNM<8d2cwt7a9{
ze0n~S3KK`{WkBJP=ovwH?BG}0uB<L$(`=AtR(#s^SYplD)QPQuz5;s>JYebAQAi#$
z<$p$bIeN<UOLP3~SKH?O>!jYi+4%hIz}<QOcl&DBSjec-jc3~|*icuTCF65ZJm;|6
zMR^C&GnMadT#j6$9AvqJ+bQAHt8lHvd^peFh85r6er-vll-W%KIVUIInd9?lGDB{<
zTdR|N%D&p>xZf6Wjir+ig)4vOjoijRp2d~*@F}A(8l^nQ%bR-K0{ptHY&_QZP?p~L
z*_jzA)MxCrZ)MYUuw!a%cR08Eu6k$8<4~bH6~)nYvqeVtZF(?K_^O<d!!)~1)W;A+
z!s@XZZf2+w*T4DcB@^)16+|hxEeNTIs3*@&+pDcEI6#vVxHLx;g--pJxc}ioTi(~E
zjeRB0ocB^gf6~z+Y`om1$8Uw2KabmcQw+Po4H!EJJpkSz;$tIveY^)0PmmT6`UPoM
zg76P^cq;6j5F=I-v<wt2=nKIP$u}CsG?y8enPHcr6MkKOM>IexrYf%JV%YOve!HMH
zAD#ZUQ}0x9_E;{TyGBxVh18fpJX>|1iptHG<+oy@WfxuQPqURS#kVfD6mF}?39u<Q
z`1H(+Hm3*sHYh!?i$nE{{R%D+_yCA=0ZK?l$&z5A;WslQ%}yYw#*P6&F!@(i)%&2R
zJP(E{Wg5yz;GLRKMklx%l=rJaW47dnyMXwVd1B5Uqzhvq7$=xoQ@^fjxt@t0JJ7fq
z8O62mLZ+Gh^yx{6@t|b*;e#V-wjKVUCnAK96Milx+A9XA{agCKvUYkIlj?TVXdhyo
zpH%$oMmcxyG4*n?l3%Z~iT(4(;NHhI<L2pc2JqTIvO#5d{P?AB!{6Xjy?t97LOkAr
zb85iS0zCM`?VFj|x_!?(s8gbSVH;6}WglWGqE^%9vY`nEicD?HC&+v!z9e}dVCg{;
zM$E{ts?JZ8$#hl(s^EX0AKh>QT;~64R8XtzMhJ|Ul#PXDpN~%=z8b84IO)LfnQ1Ui
zh3|r>9sv0>Y?P#D3<PvE>lW2?``}<8e0E5e)C#hlIOK+18d@Yz_iS8#XEXdj;o!mR
zah|osjGCEdhCMO<0vy`Hd-iY%)4*KyHn178a6~jQU;i#7B9;V(gfuQ9kVdF>31STf
zUr6C2V`88rZrDjQYX~IKxu_17e5mfgB?=D>rI}m>``<~9O28B5uWeU9KK4K2yF6cK
z)AhZ}{p!REfTh(g>e(bG_1`=5@~haBJ#yoJ9+4wbk+5NUd#h8mp6AoGRU!5A{OG~v
zII_i@bKRHu^%*aldYgBzX@`!3CrlzxUP33d2XP>2lt2JE7!H_UCD1&)0!M%-jjm>R
zz{J1?vK?`)!*)Or)NmcB3kwVD*)T_GeQcoH6v|s|zw&D2i`<5BNynP`|Dq@UrCP)p
zRFuLMW1Ev>?)y<uNkfnbYvxGcUWlXteW!T?Asm_noGTwk%K`Xii=?8rL3Jg|_BD*X
zL-`gpZ**6QV)Q@WXvN_W&cZSN=w_j#TCPm`hsic_Gagfw?0q+twO=@#mmI5XIXOSM
zZjL+Qh?n>)5qr)5#OwDx7ZQAY1n+2#rL-YfO*+wQAg&;POQ0q`QlG*7SAVk1{bK_B
zL{1j+Lx-LE`I*MY0i{BrKY|L1U-S=@o8Eo5iIta%C^H7Hi7PAJi}UwC-4HXr{e6yo
zTQqNHi6UcJ-RDo*%Lbx|O%P9LdJXH~^5a$m)U56d<vH32?~T~nmUB?<>d=R1j<cvk
z;mQjbk~I!zY0y9Li#|Q}Li6d8y`aj`p*HW6yPLYh_xMh)n*0|mc$p()N}GQu@oMg}
zJ5BiJ#p$NxUFRLePpunzB>6SyKONlQQ!!PAV8*lKZ(3DU?6bY5{#?_0y7-G?D5vrS
ztI-pa)?b4+?($WWCTMBrlccfBd)S)DBwhN@|E69<IXP9wNw(dn`_7#&X-*s>`>hui
zQ_Gu1x~#l$M-~|^lf_x`wJmAN%zqX*Xg?E|8DLFj)~3v=iK==mDzi2AvDSC-d{vwO
z>F<K7xYM$_>m){0OO*5@EYJ4sAT0YPO^UdtmgCbD<Q5_n8DaYBFw{1~*b1$VQe%08
z<M}^Zac^r`t7jFv|J#NCzl40v%HpL`cW$}nY*icktx)GxbP#IC7J+5Qg7E<x#(z44
zH;dNnU>pFfR76~SeQ|e>lOM!au$PSZRP23?gDDvMkgZ*mgp=$BNGIrF2LBWESoig_
ziYF?d;kwXr-sHyJyMVyC+$<d)9?onZ1(!js*ak=IE@DfIclltziGE}|cI+oAOGWo?
z5E)*-R)w~i=T&IM02%{;RWJny(2bOo-p5?u-rwFi3NP4&@`^TBK1Mj1q0%E3RD<!5
z;kWs2-3pMRV#2$dURL9T@cYkLM>~>Ase8l-(H0O9Smc&o_ad7BrClqEdpH{!ntgUG
z!U6+F=>znnK{uP_xe}=Q&;|o1-Atf_N~|jmH$H<y5jFuJ`(B>$|5mg~yUOJyoc+NE
zB-1yt?!W&*b7mtsSAhnw@C~o!DDLq~Y8n3`1Go4zSn~2w;}EaGTDu^YpPZa5f%X&M
zwiJc2^uwLY33=EVwr<FcS+E)C*$05I(GUBBGg~*(B?%;JmRRd2>S=80ahkaynU+o%
zmUm=YQoH?>lZia6mtH{EBzE%4=ko2?#?=I$sj?;O56`9;@ay^JUKuX-d$h>0dS;t=
zS7!rDVvsHAV&L-m{qP1gC?=70Ai^_)@h?Wh)KQU<5Q%2Bzaw(A_D3--Zt#N>?4q9D
zpI2_da7@5lo>;q{I2N!&CVf~B(Dvc|`$JmXIu`fg-!8hB#VKyj$|k`rdH~)EZdgtI
zfBMlZ9g{^~MB)#!A%ndM9^MR&{;<gocz5UIU~{dUg4@8pxI56L8%{_-&Y*T1GAZ)~
z2^Koc@9;eR@a~4{*&fzcwPHp?7?9WgeP7FmLo>Czxus<ou9Gu2&+yyg_5tihQ_wYk
zxast%Qv|wWY+*6t-MzLB$2^+v7ia(L>Oz{P?B4xx_%B>ejccp(>cm|NSL!cW)^{I1
zAi?dP4z)6vpsAgzH!+fA(Kn44D#(c+`(2=YN1w+OzQQ!&*I^kZFGFew@9-X1PHec0
zoV%tH#z{{Zuy+Cmj$Hh&a~m>)Re<c&!j^Ik{TC9(*On{S24Jav2rCSTR7)Qwg6Ovk
ziEBVN54}3<yE5lzPLjl^tAbVL*QbmuCT6#rlJ>}wzYY9jMtE&7*96*^VVlDRfGCoH
z4@aj$8ZZrGT+m}7fchoDD1r1)gR$5Iv3fJ01Ran2M}3sbebr$?fCZ5_ok>#~Aa8Tp
zeurfB`m+9m@)AMD4Bnog31=W5gb%*?K*2a7nB3eyUpFu_qpid&KzK7~orRE!w@KmL
zX~o-j?^5T13kLU33vvMvB02(_flGnHJiroHg+L6+nCb_*@eBA_55?6>$#Vr@0IY%P
z2$%YoGrOJx+6BWoz!${1jhU^swlaGlf6_zo9HhSq=r2HmORPf3^TU!BhDImwp%5Zo
zui!;ZSc7U{^Dcqwi9l3Ie-v=g>>wDn_HL2~Hx7KEu5F$afQ_N06DH#aTnd@rkkZ4S
zOo31kxCr#@0%x(P%B`%&om1*Kc@2^41MNX<*TlF0j|MRk5X1yPu{_`W_~C+bK)*6L
zx>))6phXo$r#KiXh=*7WmvUgXG7z?HLf1f)`K<h}CBkfoW51yT47xq0w>|U4^n(Eb
z9Pzz_{LdKBEC}Os+tq3~Cf@;6rjI9erJHj<7DS@hBm_oE;$H-{7M&iUe{LW=HOT6_
ze-y1X>QF9nK^{vilK_zo0^CT2t_D+>ZQdYB5DNm5)BDWL&B<F%!G>vQVbOre2CL5*
zV$nytpdpNyhe&|J(UIOU#(VK1eAu$^gi@f--`oDWXc89^kOBeq;Qo?sSg<B39AfFl
zBsm7@ELK8ZTr>f)0aQqcn8=qwxL@en1-8;n+TwCx8Zcw(kvBVtbh-sxV#(luW1bCw
zXXS{y3AMERuaa}{QNptH8gg&+hWUtgW(0vj0oSUQudi=rm*n>Cu<J}$trz?OBL}hc
z1OBat)owF@QSdLt*w5qs*iQG?p$s(u1d<U_`N+zV3Ix2v;%FKLRxbq6c#a^_5<NJM
zBoELN^Y^z`?7*myCQpO_P3)?eOYKg+YtOyRe^x7R#y&_!29mkJ=%)fk1NH?MTBaZp
zB}Qcsj&8&CR$k?{^Xh)mm=R7F7_xv6CVeO2q$Y=@cyR+U2>okoYY%Kr6@aw^u{X(S
zfieTP67YS+-9-D1MI{Y3-k6(Od9W4{&oA29puZ3;_G-eIA%fmS`Dmb0$gRdU()FWA
z1-A$)U;=2uZzmlc(cbU_oGu4ppMxs{lpEA1B4~?k4kE55ZWK)Np8#!~1{8&`o?uj?
zfR7b?0v-vwN1t7OW@n|4st2~u5R7{?V}tozg$#Z04RD(?q9yxDKLJ=}{@34tIF*Pu
zhTsff2Sxi(c0s}KKZ-IlG6>BJ8hc5wbwoh4flF+*RdXZ3O8`Pa&WylWob~8lbJpLz
z{{7~toE(7I0!|z?Hz)SO$uSUFBH^K9DC>~~jK!khbj&%Jhw5PN;H|Bwvu8zrLb|?p
zXMABfope-4(jzOJh?JDU|5_r#`o)U<AO>8-ELU{FVKg&Cvp>vCpxeg&J<Z725B|k*
zEP?P()!`P^;-*MqCGrMPz@V$B@n_D_2(TH9Y-}V#S?po|w?|dP9;$A8Z~@TL)dJhx
ze;pW6kDa>Z2EzzAm!T{vvnLD<g?4d9vsA~Qfz!jR9|kfu(66&e*FyNhVhLzd+s>|U
zdpBowe9e;$Xp@NG3)lAHo{T0M*dDP42t;J-R&6+f8Ssx-o_3-0XPZ#b_19m{xVUKH
z)QMPt0DCr7uFOO$!fd4tScUefQx7dbe@3M|f)FBkYl}X+kOq%*eM)Qa(gDdijYCN^
zhA;z~LSo^lrQWD2gfE8^WdD=r6*I@Vvfjt}KF`TvgAWW|n0<sB+dB_qqy#+T;^?d?
zK@4E|o&)u?`GJXE#+E|s2BpCVwEy)f+aBYiAKnS_i3F6<kF{pb9nHS}>*h$7L)gu~
z(K@S+!339If+l;lSmbKq4<lcQaM-kt9ZNtLO-Sjdv2$DnhAA4@ufi%GX|$zO`!{Do
z=s=$EB19n-L5!fD*e(KizhJ|RMo1Hdsb~hWGaB-kUkIBp#ZJ$GygeJBkXPp4E@PvM
zu8ZdC=yr!Wy&r58b}aVH_DAW2QHI!#?M%wObX1JoA+*dSh{rl$vwpc1U;1~!6P_;m
zg0642e~K*si&6TqtZ*UsUIIe_3(RROm{|94=_r*Q1W)_`Vic`!7shh?Ve52fDces+
zMiES7N}k|OM57FZ28KI;UfFW|!su3vw>rQPq}*RmA3(p1%Bv_rn!A22%+`4R2gS6=
zB==HX5`U60h%o#BhHXVJb3Ds_qP4fjV#RzKUji!hI$-Ix6C4<6z1HWUSPPGKEqSxE
zWyo}RHpzEH{{H=&Hh0<i8pjP2V(;yTKZX-nuzoCi9I$p?l}>7cSMYsV6Lx?)EF5*%
z;_-Nlu5>-aH^n$%$I+Y%Y!5b2bi<IM%IR2M1YuDN>4`e>IC0N-o(eZMH*>;Nn5^jW
zBbWdnYB4j#VSE17h3`?UjdMpW?z5EN@6NvQY&jZ0>aP7a9<f5yVF*c(TZ6^=b$KxG
z4<yd^1WfPh&&IoZ?%4>3VmW@X09bJ?%we^gj#>-DK`e>yAJ*f}g<EZEx`mjy2|y+i
z#2vZE<2VsSs!Wt<BHoGueCYs+DY}G&#;l2vH_r{2v;1R^)ybeWb}V8y@U7TzVy`4V
zi%={x05PN8DW+(7*MowBg7V2dG-$W_?c!`xKZd~6-`{8rA%W6kIrAT!L29uP@0h4q
zc0YUlY5D>NT^(#DTBlDxD#4tn!{GS!ubF?te=yzuCI)=9(X{1X47^V&^s-lhV1m0j
zgZPoc(9BF5*a6_`2(@`-b$JqdFEC0LdA67R6+XN^Kl{4*vk6)B|0_g+Q{bZv<OhLc
zz~kBuUa8vhZ&@tk3m9C-$}h-caV7{Zcn*P+X)!tUsRU?af;U1tBo`R-7?6-5SRBG_
zu3Y<fJu;Gxz)%PQ81OTaKsry%vxg6>Jj|a5h=OFB@C05z7ZnkK1)7?ewShm|iux*6
zcvw-cv4oH#-p9jI$zDtc&?FOmze4t0ss)w+G!d*=udoQ<>rkSYhK-AKS?&V?q1e6W
zp8Oe3@OYHAcJKf18i===0!4U~qYJ@SAsaEeQNtigz&)hy{GT<JW*<z$7_8!M2u7#j
ztDHhmOOfIr00;txEWyeWDih!w@cPUmD#xb{H@V)wckkSL85WrFNPllY(U1|#?;eMY
z!4v%lWllD*5YUvC77fNYsf>d;xEp_oudnUu`g?nqAG#^QY2Mpqg0dstIUenV(KV1z
z;IZn$zK1?v`_MjB!m`j7>m-K2qj%=>E6+X58Fb9eIf)M%R5PE+UjATgxP*|4u}rK>
zu+zXAz6t3?upEARwdI)YL{O#wu81WB0ZcS(sR^)lvjgRV4ymLccJvDFH38(qb%hql
zIO1U0_6Aid_2W8(_2lS)FjKAk!k=n{1$aA($xn5_mlXhxPXTi`(zocFDz8J)gFTky
zkpaX-08ZJ8gLcL0(is>%F_{IB%IeK7MM)RIHY5Q-?<te82HGMz!Qeyq!L$#78G%7i
zaPK}z2PY`932?^a?!e1LvRde6VDQclk^vGk5SJ?w$&`dwh%J>gI)W#ig!EYN=!CH%
zF>^{EH8rgT#EiK4$rVuG2kfvK4Gs;#EB+3r!$vGv82#AJ6`bzB#g_Y?;J^Xv1vm;F
zPlI7J#-c*=23#xBFI^$2i8qMte?&rkrM+v||L;lLQDI@qaBN8l9=>3O34=`#gMdWR
zm;=UA6Qn&2BHNiyJ1*VvFmJ9lHoJ|42C?=dASDfVNvsTK?;r`G&~ZBxc5*;>R@N%Z
z1Fm1qxv)&J%KxVCS6cW{L&k?68mp;HpMKwc>$QZD(cv9ZX)g?4ZM$<$gGEbALvn}i
zy<J**vz(?DBD>b{O07HB#H6v0`xFZwNBFBZ_s{pLTiid|@w~V9M{wBk?C_iZgaRem
zmaYKxfU~l)3rkB3oK{=dU{L1x^5sj3_wWXwcx2Go`&>YP=$0)sxUG;EUQS5ZOks;A
z1tE0`yLa!`)unfJa}&OO$a8MQ=tHs!7l>AGu^cZ-O=65uQ>ZQcpsuo~mN4i!^pNe&
zojU|Lhi`rjezrdFJb-<!!f$Pzopfk8Y++@^XgN(9bv;i4ZKf7|+509Y2sa82LOX;x
zKNL6A*Vl6^dZeXU_4D%Z^y55nZ}%yNSb@C^4~^-iq@+mXfs*}xAaIcWodVhnH5t-o
zHl&U-N~<JjO|3qbKN`(yDT0fO_YlJyV9QAG3;+^E!cOoXf0!MZpZtChV^KK%-4Ae>
zaN2y+-p&p*5P1b?pBwTmn5e0#s|-GF5otKNn>19Jo}OljuA7>kChcCY;62e=mhZa2
z-jMis_9J?F>;B!Z+;u!>{99CHP>6x9F2M}FE-%M@Bb{Ig+yv?I8bny=h<3K7v5^s9
zGo<o5R?CIO#j2VbYTSHHSP8LoF~;fCZtTX>@@weg1k$Vyc<Fphey}t}WMo)@e<F(p
z(#y7|PlrS6(20A?S@5zaiViSzWQjcd{ojJoP5u>80*DUuq}%|?_P3s%b?Vz92NbbT
z*@JoP;pwRX^3Lz^aj@=~&`_ES-=`X#Zzzr)J*qKK<f3PaSXl!frk|`Fh662FI6}zK
zD<~-7bRU86PG~x%yu%<xH432_vhP0-T;4eJ5J=@WKzpF~$aVO++11q4Ry3n;@bK{D
z0ToH`J%rqWj78t*XcWksLh|yP&~3N3w|87X<l>itj!Qry=sYwgH~6m9iF!nFtnBPp
zu;}&SP7sT49uR_DekuRdb-Ke9zAB!dd7jojio{*Cev0Zl!LkIv=JA7<p7|6aJA(5_
zje#wLyr=o`<N9dP@&JVc5tr;OU!mS$eR)f-2*>S*w{CH0*cUCqLOg<#na}cHHxZE?
zyLLSQs&=N*etBu>BQ}01C)eADv`H%I6cSlK6VbZ+5zZqYU0$xxxZ!zASeF>KM!gp|
zNNEL9O%L*m2koPe)6*|Q2$PK~`Q#$6ZWORI=gdGy$FfV(s4;LBaGiDx%u!_Iw6uJN
zK5WLpmV<Tgi5FvieHz&B*mfzr^}LmobeQmRPo(>=EO@?0_mbFwg%@ToE%gZfqp>i|
zWjSJ_;72S~BgmC%X>Eyr2V7xnVq#)S*#z8}Utf7=3CUO?@AJO7`4Ns*vvq>_0kLe;
zLrA?!9eDiYiKxH7e@=F`ut?vt@@U)mkrojVa3h3}1I7*ndMl`Ln~2B@Q@*CPmF2*J
z17!IiIT#c^xUjU*P-uuXqb$a<c{6?3{>$i2!l&dFDwrrNVqsyC20S=~VS6!Qr9Jy=
z{ocCz{#)|Jc|POWvkgdhA+Xwyp|?P<%gD&s@Y9QL?*$I@3}5LiVZoOG0O$&mw4k(I
z|9+>sz8*dCxbchcZQrlYPcW;lBqb%;vftiun*m!LSpYCKSU5RpMM7d?n3<TF?d<KT
zZXXK6x+N+qO30QRTil<ZoQjsx8MrWH7psBMwmo&KPHIMO&z^(l&nrPQL$Z4s2GXj2
zzqzqW;%G__&Id|@1ly5USQvWwGUbsYM+~j4ui*|82DrMeL1fY0#-G?%_JLZ1w*aq~
z#=*g%g6oXCdu(Z`GD4XFjNp(Bja2(AEqTBsyn?N*3Jt|}Coqx4S4-q_db)6}y$paL
zH*VaZaCCHBU8@pnSl3$rVfOk%<7}Ph9r4E54;E*KXMbLol$0z)peEvS?1A{*#!+n7
zfRE{8r@G*7XJj?{=fb7&TEv}B2m~><1{+v;kY7N|%4&2t<`v!rIiow+aI;HFN_z5r
zl|l670bYfm2ShSJ7xnV-8B9I&(@QD)vy1Y@1!*@RgHPP+Jf1s#6dry70RgQ~#@^~c
zVI0!e9|{5nEgXn6XMBx01gTTe@4jaKp_j$s9@vD)$j8|6aKzY#_9>B;?=Np+Scgs{
zF(gjolC)-dP;uhP=a~-DX}$E^v<NHf;=%&a6D{<bW8BHh$r+;M8~7tr<vo81=FO82
zW7!h%VUS=G0$1db4rr~FDH^96f=$JNgor_2-MQSo3qXOiA`p#?Bx?gfThEuCIddkv
zq9Wx*K>={gz9D}~F)4AYjGufS?CY!gPTB&f<sL6jjI>l^NBzJQxFMvUe9&}#_=4sK
z@2|HQOF!^M0@;SM8lG<`ppJ@4N_MD88fKeH@7WV!nHzpYOcU43!LSbLa#jI>4H(oK
z^@u!`0D!S`c8)?PhHtOWv*UfaQB=mHfcy<MdkR58L4Zo$w6(D!EyOwaZ4RSD7`-3?
z_8{9`bE^7v>}&Wq4YBz2iV6XDcXtBG#WN)Jd_`qt1~#j#c3~{1IJxN@ITCvN_9kTT
z!T@xeUY19cC9I^hdAKDt!gNO+a04_^q^l_)<b*0zR=BjZWDnd9k`Y6x(gXed9=|(f
zFc-0PGAdLW8yW^<^XkVLrW&2K;_loDM}*Va)s>F+FL-C<V3_fIS=iV@kedaZGOwpY
zUkltyM8dm}9$?0fg84pE9=o)30Xr(dH~V*Sf`>#4BI<tNlWCm&v24Klu94YBI%N?S
z0*;IXfPzzd$bjN&^ou8%A(V-19UQKp#7zHj0%Sp;K1mK2$f!{4IXH1pAuk984IT&V
z`dSG5d3br_?%j*TlS1*XYWQ>My}QdgRxt;tUs=Z@i-a6^#XfIuSyYk8Ls-9lJ%Z6U
zZ{AS2yyWro^MiJobMDa&Ny+fcOi_&RxVv}P2LuElZ%fG*MF4;i1qHHeVWnhiyRT?;
zW20>AZkbgiDK&Q7?nktYP4+S(K=e(D!W9rI(4Rklq@|_bAlr)16S8m(+fNxgK&<lw
z^C~SbuZ<H0GPJ>vN51Lk;KJ63E5WLL5NbQyd_lE~bkINT+_@9$SM|q_Y*_S0P#z;a
zgF$!A$j@ih3dP_=jB&VCDh`P%WIi?_pGlBB<V85<kwp~8p}ZF_LX5MuaNZ(mYm5^5
zJB5vnjimB`#Xk-2gu*?pzY3W~7FJeD<SieY?m!Gj+F#&@*`D?shOpMPQ+gW*10r;L
zaFkZ=`as5QZDX?-(1VZ)Lp0=N88F5W4@jf4I<lW=>Cun$avcN|GLQqi_P$Q|PQr6e
z9QIbE1==aUcKR;91#1zTCt2_A-sMVG^bA=`I3h+6)JEsdl?Z8y1wHsPIW#zUBO+qG
z`mH)ozT~|DcSv$z<iV(5otoDm9hQ(OO<6#wv3kF2Z;!{(No9-QnX;X?4<!QT;oTjK
zWN;gy?IJcCyFq^#d#Z3E;--%fs4k}Vlz3hNmrNI%IfttIlP`YXfhxt5nR>2%ekK$w
zh=z`iyD6&YKc?qXQ&BnLXtgqic|aIM2>n8S$NfEM>94U3?TSE<*y|sIWlqSzzyN&-
zj3a8@W1e@6z3c8yise(Pu3Vk!)BK85YFKFKRL!doYu$}|L;pNQ4bl$hO2p5Utb~j1
z^@?v|Vq*L_$KRuhu{GeF6NxnmOffhRX^<?5w9BW*8)AAseIgiOtTohV&=6sqU5o8P
z1}I+*ao4pYAOZt98-t+;X(%Q>W}yQPw+L<dc~r1G;Amrk6Aoz>u{0`9o;6Ny{Ht@v
zOiZpI^(-tW$AKCe+2FwOK<wT-ar;xyLtGz;6amY`oDGLRii3;G0s9;Or4S)javnpy
zRvqEzVn85*d=D?LU#*?{8r{j>PdX<A-${u(B1VcgG&>_MMdj`8HRpTVhB^b1*HJjx
zDwkVN;!o&mW(XaKQN-TGwS>7jA1vaTh;XNVUQK?OpC3m6N!Sxt27Mikws9a<)xt2%
z#P*{fqxb1Wz(gdg=r%dITdAKWnNVQ@l*1XB?m}d}H;mb>C6IJI=>@(7blfP;H_1Fn
z1Bz&zxmih3u@w;5<9GBnsvJu4(zt6KwTi;!o6b&dLZsy5=T~`Y{ruId1oft7jQF9U
zp=8do^mIy8yjIlIlFJo;LF~Z|2stFWDuFvAap5JUrPUoB#hT603zep(rfIl+z_XKL
z!nJG4lMcQJKjq})rXG$YD}BL8nMPPja?DeI|4z(#*I>0|h|>v0JWhaJw|`bp*m%($
z&NbHrc16K?!5`IOGEFTkd@i`g4q$IlmUrHJINP-1#u%;TO~*0LUCFuG+4NeUgv&Tr
zHensq)U5HF<o(R4Rm(;HnoerQyLa8o?aO}eXsV@|YE~qq{5MP0NHev+vY@i#4!8Rn
zx|L<<A!%4+ZvHppJ8g~6cAFOr(dS>CI(hO3ydAABFHs1_fb#J33&#!=bk8&ZW9`18
zClxq!^j3WzUVp+nmGU9FxyCI+MMPC_iAkfp#KMYA6XN9u#@n$iBhU9^Xeb<<(odg1
zSM~JpL(+3@>IdZ|bnL=Pj)Dg{L^Zs7SA{lYwr9_7WM^kbAten96CP?<TpTMN<9eJr
zG32PRR^OGh2|19dfdeg;r|E@hQnw)CAn6b6G4V|a(sd-(?cd+p9<oIteAM;#bkwL9
zDASB8Dk_)=q)%uaKyg#}NuX5(K~)Dq@eMiBjkJjZYbS_>y~OJ$!+)JF6cp?cccIX<
z{^ah>Bu)kwuOA5Zyn+Hv6BG6iO-*eOYBO!#O+%6{Sl#X1+?cT;A@E{h1=KMbbzu$6
zO%$%Mh`hlIfNN)_0MoV#xehxgr${6aRFUmiSIkqh68Lm|wa+o^hWNFwv1=pl%gUk<
zJ~A9(D(<G*%9JMgECP;0z7iB8I$2p+jIkdmQj`3KVyq^}!oW>aVnxS+lAQmsiQag3
zvH^fy1Yr}s<%}XBNDYwYT}bl_Bb3JCPPQxbutGhPoDQ}gzGJ&E)mP#<K<nQ7I%;iY
zSY4{5j(cd*jz08!HVZ4x$&`h=C@>$y0kPK-VNLZwZXq9ZTZ(-+?2yAe5h|eRo+jQD
zSRPA#2U%&dIiuCeRUCD>H*Y4P7P1650N@aUiXyqicqfb+CMKqRTKFfImx)1wQue|^
zHHBP<g^t|){Ouc4>;OgK+=0ZTB+5cp`rgvJAnUJdeQ~uCdt7sKGil~Y;LsM9mepUs
z@?hCKc=|NDaPPvQU>$A<=8_P|w)o^4@cf{AVSH&BlVy-P6ZKI3Z^1OqC=lhO8E5CA
z^Z?~T4T6WifB#B7isOF7-+9RAQDe^I$H7ri^jK)Ggoo3@hf;(64oCa2++2xBx>Ki4
zp%T0&&2)zmN>MmE^5ZOnD!h}sJ3*WY$LV<d`DF%}9gZ11sFWW-g6tU<0fZs<5)!~r
z5`?UiWJhfs9U~AkrlVAtmzTE>oga<|jLSo8N)WP0_P&KgbJ0!?A<nAesqpg}o#tXp
zvuE?srw6<6X1HB0ww(>Lqqwt=d+597L*uI2+R$84yVdRgZ{oW>YQT!%BY+o#+uc*Z
kQ^kVd%>Mtw9~TAf&Y15$$3XFq0{`i1>uWvRZyWM|02RcRpa1{>

diff --git a/master/_images/sphx_glr_example_parallel_manual_spawning_python_thumb.png b/master/_images/sphx_glr_example_parallel_manual_spawning_python_thumb.png
deleted file mode 100644
index 8a5fed589d17fe6ee62d6a9788519dba471fbd65..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 26794
zcmd3N^;cV6&~<SuCAeFl(4s|(J3)&T2rk8;P~2Tha45yKc!RqHEp7#hLvXj?ZZFTf
z*8BYp-w#<U>z<r@Gv~~mJ+tTR2sITsTx<&LSFc{-D#(M>U%f(#fBBGMqQ6{mHpumQ
z^~%~w0VJvEnR)nlooKF`yb|H}xA|C<^cUE&K&!B{#ar*%RrUI{fOnuSCL$03!--#|
zXbLuXBIIR=G00EgrNBb5!{Wg0L?-(}Cu?Haf+0F7%3dlH!I7q$mL$Yecj9<`jSzfT
z^D5Zgt0)_Z#Ecth&8aB^q#^tt+jp*g;+l3Kuta%Af{32Sylxr^V1hg&jmnR}mkXfv
zAbI2;{|#XPvE=`E0P<S?^9AVTE1Z82S>eA)C<&;HFH;EcKm_ss4Tt~~BwnThPbC<U
z{y+Qy-yssCdl=>g5(+AS2SjXlQXBFG6FYU#Fiz_d=HGDi{5xv>!;RVHaotlES;<Vs
z54)3=h(~4As6dmO)fSx8D!+i$VC>Y{+4GrXjSnu(wiwX$)Rm{msq2XKT(PDE-xEgi
z_(VivIE9l($F=w;5%X502cyd9S;o)?@*@IUt%-#RKi)is3GxWf#e9B1a0p}-K^GkH
zE#RcLH-hEomo}wtS-`xST`geg^-vL1i@Xk;hL(*ihF<qw7%c*?Q+E}k)1pi-WN4&J
zrc9QIvTOJ=TWorN19X#w9YvE^0xfTy!~GL#58!=LRoJfOBeNadJQGEW^>VL8OfGro
z7;=ZwLXX%~LDV&-I)%f^Zc@iBUPr)DOdL&SY8Fe4jRr0M^?OY#0}B$nzJ&L1In9Al
zQOK9T4G@xxBY?1&=&}M<$uJ+xfTKBW>PRAN(VR;B>we1G_F=pv`)g>`Z^uYRx_jEF
z<Ml1t4a&-D$31j+QCSBStgUGDG#OX))Ly@#cs>LgunT}hWZL=J)o}-tSc-H43?kvu
z2VGc#wT9ESOGB&rE@vy9Zz3|&#<T%q7(yR6b>)3Ae*}ua?Sht~cbom|GLpPlTSm&X
z8g90~9$!qLFeLh!xc`P0$(R$Ja&h&6EezoO__ACMkpyr<B+{;|AT;h98iu@OkGbHl
zGYQkX4NM%Rey;N@d@DM~2kNR!mU_AYWtf#)e}ui_>*-p^zc6Sxuga|EOJ~>i$53|i
zk3uPR?4WZH-H$~ya%J$=ipQkS-owRa4tw(P{%YPDc|taynH>AT=I|^o0}49wU~C{J
zauK2=H85m@3Fzliq*hcd5LX*{uUE#$CpBIVU20AE#`cp-cs73fa2V=^W|F4HJO04v
zh?+`+`1w}s-C)?|2Q&Bi)%tp}=My{T_L}AL?X!5%*(mp1+HX=ICa0VpGHL53#n?|w
zrsVN!1h2aM0663#mGa+wr{K=E&1QL;2u+|ml4!Vcv|H~i8eectf>jNDUkq^angTdW
z)#m!O%sOuC%7f^!Qht!qCw#tfe81zkyyL#c-mQq;u_=hU8Y&X0P3ZX%SuC=j%cBp_
zr)IOhDL^juo$ys}!om$%Q`tjo6J*I-hLpYq;KY(EwP$7&><+R6Cs-uf#l&-#?&}Uc
z2}fsB6#K32O!O58@ksbMFteP}G2(2lnU(Xx0EQ{|6Fyl`z1ErflfT{pGkLUjTO->;
z)k=c&e_nqG{-=f;xxac&YcIobydWM*>-WCPDRj}<`M?J$>U-1W=e@wPjOqn`WB6h3
zK~jXSxO+anw%MV<(bYq5#QL@?&$9}E9mOt!yYPhfLyOPk<($)-DALZ(Z<${Ys2e!I
zEf@cqbht`&4<v}sP#}Lx4n!!n0Yy~ga^ziw<gSkSH*`h*QC-LW26MvjQwQ^Vr$sI=
zO3>b(^c=S@?tyU^-%7nsg>47yyUusdaMN|<O`K4>9?WoX_ry|GMNc@rE*&1ZWvhRi
z_S3m-n`bk(@O+$m8Kp&CbTqx#!3Ro^Y+#WD0wD>iyApzndvKsR&ejCNHW0D$L?Y{&
z-+!(ex(HR2t+z8ezJ13p%e#_pEzeK4BMasjYdnwJ?s{s>RYi;(GFI&e<aS()?XjWO
zoYY<N7r&duY31K@BufxfDguxb7oZWLurjz{vg84UbI9BdjF6-Y{<Q7%utyW)+c$+0
z<MkF<9?8fp7m*z+oS$&oEZ>z1i>3C&Ca}f(#daZAeX;DyOc))i?10TL3cyU9coSu%
z)362xbDc4*fSbB%;E}Qhj(t3Y;h0y44|>PJ{Wnh8tcJHxpgfb*cWZ`3JyR%I1eX@D
zhJN6ES7k-g^nBRA6+Lz{hr;f=v9U%{mDsRr?D`_~Bl#>!q2hGpQrdVLN8GXyT@?Oz
z+7-##Pu6db%j-<<)!r^-I*w+`UXXo=8fYr1;~kdxDdbuG7#=k3gn<y*eN*f1KF$Bh
zuM!f73PNVVWu(3KcN;%3OXZZG_KI`Pw|e|IE$cJPXsj;kh%dz$%0ABb0gso)Q9gp2
zNtf=PDrDsl^fanC@8EI&q(M1hW$#)lTUm1o_)})e6B!%S2>wVoy*Pse*8?CP{EM4h
zidh=Lh5n)*Bd=X5LAxl_{94L7yp5$zMl^#3{EGwW`CdAsJ<I9#=;R3}l<zuf@=^R;
zn<3z5TIkm7d*S{Q&!Lv#!b=h$$&Th_>ZjClo-A{sUr5=r94uSOqPi<ojUL#vQoR0D
z9@tomD=TNeq@pL#T4Ehm8}_x9rr7iysgJ_9?J!1H=qLr{xNdprpE-Y{97Xs^)JR-0
z+mCbAPM4RgJda+_w1ZhDF<f#6KMK?RRs9%DGo70zPi6-^cvsZh<Pp*;9q$hp1HB?1
z1Thj8DEHa%3MWR`2XbC2FZ(&PJ%6R;$UL|(@;7M#``OoA{*nKc(^Xao#r-E)ODh-A
zOQX3n*Lvs2@mM-*O*5azC5U{B1MKB77dxh5mKt#I>$Bx2XqwJ0Xj>oqW`wvg3e07J
z)B$YP_^e2>9eu-`H$Ilmv=_G3H|}OW)AT=Y3o&4{OAN>c@maT9IRH=16i5c13XIbq
z?_;?AkC)0lxBghxk!|c14fTiYk`u>n>GBPA8g4GyLuofED1?CZ_ogTPNTn=*7Rz>`
ztYLiEcuec^&_!PYFq(!2r=O|z#Ft@!biey30#@bucj&Ux#$iwD(4dukXbV1&1KhNt
zqjE7f0?G2qOI$YC4rS^$^olhr+rM>-n6e_o=%}lSws$wIa?!a=?|&NpC0eU;zsWc{
zKWScWoG(_FP6kwvxFXl0Buq*6*?NA&t1N%CNhYQxAHE>%%=x@2fFKMkfr^ZzsGxgw
zHOYm3bg4i(w(q;Sv}7U`6Pvh6Lm7Ptpx?d8)x>^1RQ1|vuG<vm`}&L2^y60pzYXJ>
zD<jpjZ7a;%a|D&NxnAb?nXhfne1x>|M?Wx8nh#9<4K<gGSUSEv7K#rQi7l9~Uj!&y
zKe5Cei&k^WbCf;y%2w7n*r#WSHtKp+%roO8#6_LwL1~<?Pvf`t?S*N^=01T?r|(|z
z$d^E$%(J*;d8ErD{ZEFps9f+WKOzsubk@s$VbOjhU+t{w@7U3KK%cp%uNDtZ>3^p^
z+tzr;24w)=wRU5Hn4~bhvixzHRwBbhqo*XVUm->z)SC1la7x*{$-z(_ak+M#+ZP-g
ze<IxXQ4aEqNb_=<ma`Qe!{XjvZ9eq5Ib;szIj%cDn`nu6F$klktHN)8IeDy8S5qM2
zC%E^-H3a_+ex7p9xz-wfTD14=;J)uV*XXUj)pBRk@Pnh8CqmCZ&iJ1QUo^Z)B@vIg
z(FlGVdndb1<*-47yzX?oP5~QapP+;@NEQSQ)AQC1_Ty>JQP$k`X`41*d?oW^%4JZH
z`2Ld2K6XjjT<v??>~MILbr{_xL>OfMd9;zSl*--#DwbWDM4Mp%J~N_%)nv|~hmH9@
z)4zT59opehPLJ^KJ6=VVO6Ho)jzT3gR98(NX*61GGVm56{>rj2M18)d^HhPG&iNrA
zw3~jo&)!0v{GJE>iXI;ylLdla?z?dM+T(hzjWMg#Y7nngVIY8XR3%u9|L|1IzJP2%
ztaEj&!l5_dWLvR#((N&`TQ%Z_aD4=0dIY1oPwr0rTMp@~krDj&qrOo>N7s>&+3e2#
zj!pO%Q}P>}{K5VvO(@IX)|C%9o-~d!1M>GPa<ndYmNv;aFk~!<is==b3_pJ<30VRo
z#OB^{wUn=&24gx1tf_VnB;eG=J!0z~bgcJJX~-Ya3?h@$yfOIL)G*(evb?a%VX!~f
zqp6{~#WEH-Q1?9^ryjP0Z$$eQAyAnI&7&uR5P|1uwXt}Tyv%;tYAc#KvBk|6x%}OM
z9w~hzs3Y5-^N5AaGZ~c=G;%D?B(WRA1c1N1778yvhoaD%3rgBv|7EatdJt$>$u9}~
znp>y9i{nXjS1MjmdxDO!=)mKBA}-}mBp?edtY>D#G4~`p;ncc6ZCuu^9162{V`}LD
z1V6Z{i6Rj#_U_ZbKJm)mqBY>vgDo5E5h<Emio&5$E@AF?G3Le0&bO!{+=wz|)7zrU
zih#Jkm+X}@+r2~Gu{6>b5`g#UiRW{W{;Yv3BoGE+jQunVS36Np2enN*G1C~E`p6Rz
z{j*O=Hr16UEyfQtZjl0gr^mQuLw~P`!;^`G!)B$eL<AKUJuf^<NAv;d9AiDUw_1$1
zJPo3x5S=Jc8|9~ObacN3AY*5Q!3T7T+fx>?HKwu<G$dMGGeKe#-WS@H^V?x)LO8fs
zqmwi>JHg*Av%bEfp+&rI#S6h3Du-ZF@PZN*czynUs7CxrIzi<4CKW$DVUD<li_Y|i
zG17HQvoz)>Zp4t@iHd(8;)$gN|7tcwXZ*}bVfvjelr52^Z_wbOI56$NP(NgU%iVHE
zF4%~Q^N)t*gCUrPDQc6o=jm9zod2D$s=kAl9(4KKJVtQbESUEoxkjYn&#4)gw9Hf$
zxkomp*C}$oszKqQn1bcxZ?n$jcvSPa-%+!QHfnAo#hP0@4Eu7#;_sn^YAFm}a5$kj
zR>ve*17@Gze8vcL$w+E@`m5(=Qy*Q~>OI;1*l|C1v1PMWOw3eL9{)(8=9b%IM}6!U
zMcHwGU&b-gd%E#WWQ1W&us{G=3v*Gczbd>nU~XA+RYmPvAK8xKIjItpke-|6DHLM?
zV^m(Trt~H3n8Djvl-rxNw5%%KO97+CBR|c6-1xk!_MSCj{y-=Tl)@CO+~o2vmxWhz
zY)tX$O0oEU#QKS|IP#<h`b6RAJ-6|QMRwxTv+S{FC=#_NXFVnwajvlc^v;Fe8pX9A
zhf<6M-uxR-uogNQ8hEK)IoW(Fw~$RK7de(kPizF2`4M=?^(HIkpV8g2>9wssiTL8{
zJD#*5d^it313T$)w|h}CIq*lseFB}(#~)6PjENhpg?^8Xm3F>+oFgAQ0$j`746Z8!
z4T=vRkY=AQ@V*?NvGxTdbw)Lbpeb#Hwn$jEkRl=kfxy_mFdnIZZ|v?Bk47e(MA6iu
z=n4|<iHvkUa;Ob(KQEpZ&J@0)KduZPmaKWtq0k?3_hJD2Gm*qyzI97E{Lc|!PSUoE
zK1%~C&UNiRN)zQZhN-r>Yw5OvgD)#Z`yIX8TL)kRf#j_X&X!Xre9^FbrGeT&ua~Ez
z;L~rWd6P4b4BJHLCtjp5IQbK@Yn5nt(<W&i`^gZ*x8>u(A%E*th>PWT9zmg#$i405
z1u)h5>|%j;2S6YL!d+3OTS>AcGo{_X_PRbtC^cZ~OZKy980j48DF6mKe0<jN+^Q!d
z6_+wYh)0B~28opBG018{`&UQC-Z&pbLmX%ur<=#gp5NWH^gbnTA4PSVe6dX3@YPg6
zDr(C&J6Z51?||W#t8Zxn-kdrXW+H?^p#W}niWn)GH&MG$2$2AjQmXqq)5=*Mm#Z+(
zS55!)cU_L+xQ?>`ShOy)%)$Gpx&xYS;yMa=?0Luop+4>{HHg<+3oxDAZ>#%N{Z}PF
z*y2V1MAwfF%&zQdI+R~N;zar?o(NK19$nQ&-1jXxMS(~iDLANz^Z7LflHO0Vw(S<{
z-&<r5eM|PgPITIz&u6(UmdqwH`L~qt)9ZR|&VG<JT1tSay74fpHXwRH=;&6EH^5|0
z+LuTHxkiqxO8upJPHvd7O7}8jG<V61p27Jt)N=R14)#js<10eRO&x|CW~Yj>6fR1v
zwYJ=7+&AQJkPZmGr9o>Y;3Dk51T1!Wny3eOhpSrVHP)r8j)n->0clR9O!IoX`MucX
z^Zwiy9^o(E6WZ-Mwy{?YbRia8Hzm{^kdB^~ULyM{?&zV&+;rnvab{}4=iyZAU08O5
z@MV3_6vpe|{fU%r_^adQ-fn4Eqe6TqJ&VNr&47I)qv~~M_?mkg8<^CFz(S?YiNc&J
zF8sSlt6VZ89ZVI=m8J2(?|e_Ih#V2_^euv=$5qy@pDZ&CU`?d!{!Xjfsm*EywtUo{
z7M`bz>Lm*<Ce}MtHYm{7r(J5)5q}n1Vl~^$H^%&tC52`G-pRCoZdtm7PEwjEjW?3{
z11<BF;^@L~F7Cc<j;XTn3-~YX91YzpO}BG$c$6|oZbk!Z-y5|EV+OKNAcMg%n<sVj
zl-e-={H)#pfrJC{pT?h4SGInUS*~;n1&|2O@-f=eNKvRZ4R^(dzj~9cl6p)6{Kv$q
zqIam>;&#qzeUCS&pM&I|_VXXi>@ekIJ%6OP-srY6;NEm+rQ*Vm2SP=dpg`9v=8of0
zhOkW8VtL8ieYY31ECO7=oqO;@Z;1Kzm!dv@X`5}nJ?l8#I_`)#C(=D2%+*Mp=89va
z5suRsE0+)2#jpPQZ^r2U4e8Tn0Artnx(_PB{~E_^f3atL)FN~J89#@h6>YEJ)SkOU
zkVI-6t3OWeJfD)pik~F|nR+pV9<?<+S1*Gk>ikPTcq<UXdw_)K!lcpOk1iX@!U|Th
zoEhV5Sl$g0E!rygh}-<LV39X5LUDVTxY1S(aK&!Kr*rX$uZujy4v|hnhJ2+@m<J{n
zzX?5@^gCzR)(UN#6V?@^t({?Bx@QIn1n)vheiuiVE^S;DDalwPZ+w=+Bbf|;LoX&9
z8F2XSC1R0>m+4>0E|mxI>`U=jt6gulFW~I2NBxSomP5IND~h{YMxG>1blL@4eF^{6
zk#UK;tX9dPQL#jH$V)FC!y~DdwEKH`o<?o9|BR|2<^vN!e^m`uTJGR!jYdHl+#*lk
zs5uVJf+{6hxY=%{b1<y_plLd2Uui24dyK=?>^W&`SzLFoKY%L61$Od-y6mNtoA!97
zb$UD;2B^}_laB+luIKy^Uxf-;n)u=5i3a`beLsWhyI-)zg6gb&zCf*CjB)ffyJji%
zQ#_YU?1e#Jdx!n7Hh(Ev_|i!PMx4H_`%&raZn?l8`iT!XKN5a0UHxpS!xGFhTNf+u
zxpL3g?fDyTpaY9;=4WvgLgnhu2&s^lz|K2HyQC6^rpuT~EyYofcy)zpZBfmg;sx&O
zRGn#3W%*3V<n*|z)Qkf>l+k#uA^Ks&Nc5APO_6`soi1jM35BHy;tl(LIhwN&a^a`%
zJuR6vsJ;uQlY9qsuOj!Xa9BQ;(SH{l%NmW8PkWZ6_$5iF;!BonGJlTSxyBnademO^
z=BG6}B=bz`T7Vp#E}gK#fA=$pv=-a?p8t<XnSlE>oJvm3oLf8KYg@E~;oi^ZJM)2>
z@1e%vtsSw=PX_(a0OjVZhHr*B{g*Z;W~=k%vO2VPhKypEAzx81pO<(tom_IjrLqcK
z@zif`H$2E)oZ<bN=C9xVdlaU6(E3^#$Kr3w>V2<j@7jJovp;i8k@<L1`K44d!+7h^
z;j!IeS*PpL6VpURLOzQr98v;{UR7Gn4&D`3s(EoL{Hb3^v2>FE-3#8Buc{2P?839y
z#~{t0<hQk=<*T18>*VWy@2eH+)q@uti{*C6u-6YLDb<8RHk){z#wL@!tX{(Ia1#tU
zO(0i6^t)Ueq*UJ0(u~F1x&}Xc`0a@7^&Z~y#d7ER`@Xo6>+bhoC<@-^DTsWJz$8O@
zE6q6A|0WUD9M91pzApN8$^s8ZUZ_$??^mF<Akg}=Aj6dC0d1XBWe6+yIu{JYbdiY4
zRV#EJzDK4PtaJ~0363$VH|~2P9v-U1O5Y`@`zw}ofRT_oR8hMLibAhfS|zpBh(>i=
z)P@l=?Y7Lm!mt}6Km%zR(K8vzVwM*Cph$R+Xh-GcA2iO3omIER78s5{b=pAhf9(?R
z=jTLCaD?TmqKr8O<U~<+4)>I!4VC|q!DCP=dfL9nazg#oyW^g{NeK5d*yo|ECkDoA
zntQ_k_T^y$F%++7ZZFRcq6{D9EFXT)`{k$vq&&@tPha9Z|9HA1$islU8DzR5%*4N_
z8DoCrp?%1ld;gLWA)zb^F16gFV+?4T=W$cQrr1dXp6_NT3tcXsDig~m+Mf7k+r5{4
zaX`hIACv4>RGTPbo0)OUixoJ^mShV!I=(iQq{0#r8*6Kg8efTj75)X$kJM+F@I}UK
zUT6J1kRe_tI*P!@&+7A20@t?a1+@I_mNWPu$Azzu(L~LJHNe5}vc>A0`ZhB2cvGWA
zAP_fmED`+(u^vqY)Gl~k>@lkPE7s{N6Gk9+#3TqD@!RmIUBVFrL?mpY%{jeTz>Ul#
z^|pRz2~Co5>*5Vwzu~?e8!^P_RU&lWU@N*MZCSemvk~1wkWh+dwBQK1rWMVqOV=%v
zyWG(np@yt5?UDMjtFamAdo2)+Vqc~!LSR~pYD@3!kZ<Na$LxL#pj0rAjQqD|-&1$j
zy#(nmP~x*E^M5Xg`E%7amTX+r3i*l@v57pn?=H|ni%Eevpn^1ymfX-wy$foPXGw0Z
z1mw|CuClWHN>umq_zz#BBQJsnet2!E?A{p`ONN@O<Arp+Ey-6;LrrroraY~_4`3M{
zSw}L+i>d?Dfmf(G#Pe1^hkYDJ<6WdPOs0PO-=ay0Yv^)oC(`N1#}Xo#K?6jVUH7&Y
zt>ZCkQuHIG`z{jqel&L+J&lkI(N??MwcMz<!|5hMT0kTAH8^+~^b&)pnX*vkErQ2~
zPu@;et;%C0=sU%49+J^q3X1pBn*(_u(Z-0MEB<C}&nqd24lK5gXP&vny^1Gezm2Q_
z+4}k|da748EB;qKA4<mdW-dcMU?lj5m76+sff54`Km3OuVccWpy=)w?d#3TqxE`h}
z6l9ih_x!f;a;+y-2LrFARpjx26^wNDnukYzyUV0&I13UvG2=?cXJ$kgj4+>EWwuRE
z#tsWN%OvjLGsAyYLoMu0+RDt)d<{d{D_#C6oX)J(P(ujhKG$ds$ObrB+81&6HEUZK
z6)Ny?G2BiOc@i>AX>0zY@kpJT=631_c&}0lHb_-=K+_#y@r3`Ov!QI~A~_f2I%EKS
zJ7emxmhe7MD5eWcXU=kDb2?}pJ=Z2ixgDeUC)Cr0Dgk+Xip@@Z?fW>sIrSfsd3iKl
zA^E2N1so#Rx-^RoLChMgcI*of<7Fc`lfSl9eUto$^I#>M(!@{8#<(NM6gWSTnNcQu
z#F(n@U$X%u^XVuNW%;9#eG{+J3*S0eDbzOK*JU^!PFZiL;WG-ghu)JglFs-`?gx{0
zg^83973w0~@wfomO4tTCaGoAw5ySDvK+M&*B-%3F6upN6405V@Led1c)Ncz<5n^(<
zPiF~QP0i%xt-8+Ans2bE6p^{j(LYwh4VMN%1BUMdM~<t<Q@y^cy+VT`F&9XO-UmwV
zNQSFdQ9@(h9NntM?`(8xNGQ6_Nxr;hVsb>=lxfRktQVj7Tbz4GRfJ5l**g7#AP{Ll
zg26O4>0syPBWE01LOh{Vto8R|+hzwxjMgaa+4w4znI|@APC<L_p3o7U4Yvm^QVm_5
zOVTvNz*$GCz`?SqIMSzE7St(9XFgH5M$3pE98RZ&THMK=&w-ai#*K*W-HmlvcPGZx
zE|2{<GyASOzW{5A0ac5FB?W6uG!rdIrEI7rRrbvHHSzD#WAq@ZMp<5G1U+3Ot38^0
zkWUd%2mn+s+69nZC|8j!jec@6cB^Hb=$2uG2_CMALSGUF22lMX_H9x|#-j&=|NN3R
zm3MsbUbp3kXQaeGa7qqy#C07rvSnB^KY)88=v<VG3e-?(hSrV*cU%;r4bEmZXYHnj
z=YJxI7X?am{GN_kF@k$GTAc#_;Ns8cBGR(=(_bTbRFcwwQMtQurq!C}McDx^cmimi
zLAKxbb3<jnqHw?7Ob(Pc29cn_Of$5{HnwD@ef{P(d^BGojR$WAu=r_gjLP$CQG%I$
z%4-rTB?K(yt0^=2p#hk#Ngr`76lRn^!7j&$f_FU~%2Iu9H{5n#WSYWDG%R%8>}4r-
z_j-3kt^W!$v8L)O)(h**BQ`G7xTp5~*cIxL$Cbd!4BDqq1iG#K=~<?S695vr6nLB_
ze5lVfgCMMs$o<^jHKS+&C+0>9AS9TqsXP_zV8f-mKeyL`)N8UjxE6ah@%CHBRjiYY
zIAN$UE$g3x=1lf{!xt1ZRgVcz_({~_;ON4pN2KG(x>7`@%V6tAaX}iC*iEyNuh{Z<
zP4>B7WT23ha(Y<Qv)`n5lm-6SK+9jP-Q=@WZuFraml@kW?^>zwOadZ+-><%ECjXix
zzBw)Eg|m*luIPw7+s=KKGbMdlsp~TSU^^P?aC>~6uI9b%gV=r|dD=H2?R4lgxOouS
z4y$=lE(0JSD=3f=cism*huq^ph>fl#eruAl&Qgel9&OAcjR<Alj3|8+Aa?D7u|RnI
z4cuE_e%8kAT`@G7_haq-o0?hACI10@peGt_iAj_kTckd!D#<{VT33EZe@1Y-4A*a^
z_LEZcNew!=aM4z|te477Rk)Kg#_|Dfz~k%pOW!YGvYaJ4-%?DS6ZBjA&d8C~ri<ER
zgh~TqKZmiqZvv5nGo+gIxf6*Rwk_>k)uVWD_8t)s7{ATQEaE!nr^~>g=%(4pS8+Bu
z5=G_(yS2XmS6Y<<>th8PL}|H=vh_+`rF++VgzfHwF?)POOi$iLaTu-X)KFZf9*u?T
z?)}Bqv}%JriAp(<V<mQC+Jv+oFd~ryN0m6QSZ>#fdiklAyZx}<fs<VM%#R&iP`U8S
zOCWg^;(=)Eo+T0tz=;EEkv+9t1OqTPasza(9LhQ|<YBtAV9Rl<t4SIz+SeaeFy7sb
zJmT+NaXYej@T-2JltU&JP>iL7=i?Yhzm53*70oyBuUiOR%j>5$SfU;Qg4%7qnxL^X
zxI<aYkVGyg07x>~FR3wjB?`nEIEh!feQ?EB5fxl_Xbb9SubMyRuRRP*Q$EOTP&drs
zY?gCRJmNK?Yq1V9fe#~p=K6_U2~;O55mu0P&seC38CLy+tv~Wp-x`q#x&>ZM7-Q3m
zW4eD3)V})uA3+zGj}#kgL19Tb|B00v_HN*#OGe|6oD8pH=^tCuMly2~^7SO4!t)?g
zy6@oxDL?L3C~3=371z7zURp$Z`m{P~37?4aXWtd&fX+CBi)^dH8Pk?Ah@fkZ()!`)
z1X0mZu{ka}gkauYDfc*`#Qvobr##rtiB{U4PY809iCvoe)JrGm!V8g_?{9}sjEN0=
zJ7<CK)qfT-JtwkE7tN=pto<cUp>$GABFPts7amvF?Wxg3FuLN{L~fQO2Ji_C{H?Zr
zfx1Mn>%xiT^c%1j?7^kj&;Ai(5oLtxTKUAKq*cME;>S!-zWFU`$Q5Qp+Y6LFt-d-*
zI|Yi&1D5sozyVt6Io*3Y(QbMUH6Zt&!WxBbM#)2%rL4sK&J$q7R-gKw`9Ij&^fvO1
zmLCl&t)><H@>E;}-txS})rj3hexI>&3B9Lw&HJr;em;dp7)6hFS<O(ccpsv>x)BpE
zDm*>V;wiw)K?<mVEhhW;@iA7j3ILNZMLoPF&x{hn&$8>Qc**$GE$ELU6tzKjmkL)s
zu`UM#;VDh!tL`#Wf4#+=lYGrdZfsUx$JbO|r4oTwXfXkrg4`=3V%Izv8Kj(cTDGhf
z58eQ=^xq<-^r*F|u>!{;_Em2^J)jAy{&s94JOl#v*YHux@HFrJNuq_KMj{cd4B!GI
zyQy%L+U#GslYhy*L&uRMJLVUA4D-9ylSREQjKgNu|8-2>jqtiW+McHya%Ket_zkuj
zm3$1|$Ks(pmrP4EkMi}jCyvG+`NH5bpZ;|q0Y&){coHggK%;uf=ygA9Gjg@-z<06?
z?Ix4=xt4ntSo@+9oy6cjU5elzD4ub<xZ(T}<I6I7?f4SEItGZCqH^<VV-82;is}N|
zCJ#Iem)VB}DuIbW^(ifRZW(vT?%x(jWuuHGjsQB<=Ca$*?Z803l=!O9dzPD^C0fKr
z2m~N1WJ$ToKe1jEO_ejA(9L6h#ABVRAa=9%1Jyi$K$hq6Ap+kBGrPmDUB7Lrbp7}D
zT~(ys+@NI%xgYT&n2Q#8OCR|4nQ!zGoaEPRz8-8C*(dO+(WTLQqg~S|jv4N|PBNb0
zjg))lyh7kNehT(miL}8oD{hw%^MCcSz(ND4@xi<I#8^W?mgCIKl)sq&d`dbMmE%a8
zZ%Hd$rWX!{XmXV0v02hdx+S*&R$ekJ@^R2dY1h%ElrES8zk8r$p`pt`@N(yoa$K8(
zjhK6o-347bbur0FPyBz2^m{=@pz-rv&bObZwz1EzCUD*?!EiX3xSDA}=rb9C8xwGC
zNrz1eL{|I#PEL6#J$&j5zf+699C`UqL5~2qV!2i!hst1Z=Yiy{?8%n+gz3Y*e1piL
z6O(<SfzEwKP5Di6F}9!EG|9*jG3lry?fcTF&yqX}`=^~vl)#HXi_dK&J=(Y|Z8<*}
z30@Mwy@}{(<;t;;0%r<52#kSC7Bu51jNyy-t#x>WSy9ggSr6AII%4s~kygw*$EfUf
zCY$#?+H2MPU7Dk3nKJ$X=vT#P&+<xyH8g+ku0O7RnV>45%*v>a)aJH8>pmXiNa08~
z#<?i9?rpIxfwN-`_yM8u7UP^;zQna2AKg3DlMak0P;y?5eH=PSdr}C999F+r^0L#3
zC^_JpLv#E;X%i<g;;`yFp%_PWU!s&&mSJg0X@WSs0r!qKjk}nBa(sb>uSsyGl@a`R
z?7#3vi8E>bKryMKl1od9W<sxngnJ;(4%kmLe@X$yzeNlN(!>2}*q@jjedoHB;K~H?
z;h{>mbHAOQ?yR})5R|r$vm|$Ig$ZnnL#t*8KK5}E+@@Kv8^(9yG?EfT5PU1;lin{4
zLSUjS%c8>N;~$k2tlzG=zm*BM6jW_WNrM*T)#Gz|3{UBi_BFO`gWIeuu7`n7qJ7!+
zoB49j8dB$kFYl%CkKkQAyfMq60G`QsWATY0(n^Qq0MqRiwat%l{@qJH_O!Wp|2ASV
zp~0AbnKEINW-+Zod_NkA)8QUU3OqKkv6{WUv7umiWmN6w4y`-SB|;BV;-wz#^2nX)
z8^1f-0CkHspK8Awr2)AoVa;dY`-(cwmj-#a^qWnY@txS1{hg<?>b=G>2m7{0c%!)H
zbs=y!>P<LW&QMLJHU2(FX@TL%w+0?R_@3OuR}@{Lg&|edk7qocJn&~JH@eW7osFA8
z${K&8_u3o`BS;wP(94`+cY0*KMQ^Pn!zqW=({HA?)YFG$Dggy=&6wNiOg1+<AFBE@
zZw48Wylw=jke|06tUZ<MZt;wTJkCAv8q;y;Hw%GG<FW<SZJt=!e<_wD0%!MN`hJPm
zeHrurlz8sVi+qX5ww`QtsWJQ+HeWESuU#cztrb5u?LN_a$E%6jk1*~Pa%#{$x4pRQ
zkb}<7AKLw!Ny9$X$MW3rj_>=MRpDu#0WZhGe0hmuS$N;@)mS|*4E~<ZFs$_6CKXO<
z1gc-%wS@=}LjMsBXkO0zzU&v~<KDHjs^+$;w(=)gvt-zC>72mu{YNq`RZI2l38VG;
z&9cBZbIPwX$TSBPKN?k&W?cZIRHxaR)9Ch}tUXnw4yeb9w9Y0HH=?rzg6qU0_kXX&
zikwQyBcGjclOqS5tHEx2(=0Eyz4ua<7wa614t|PXbnX@1^gG5j4}b>F?#wG?KKQUL
z>D3%@JC#ak1x48sS1KpB=kC_yJ!7{-np@4coMInIt48ux_LJd7=zo%8N)@-)@Mqn?
zCB{TDMLie03HEs~sahW1-369Q7k;$b8PLlKc|e60hl^hIV7!27_os@JOqQ{&^m@It
zOU+BY&Kd1I<4B8XNEOGQ1fxpfLKZXxc`gJZVp(4YLUrlS#O3!<O<<Li8^|ViP~L7q
z$w48`7Ze*B>F;mrX!*VfOSO6<2k<HrvugZWKTT3;LK9K#TC2QDfk??*7a5I^N#_yO
zlk5BFkcI9a+tRGJPA{oW`Bx4JhUDGs&@j*6IQ?(=Va@q^QQIdWfsS)iGG_#=im~(I
z#@c8z_?^~$u5IHC!|pN^Wg8orei*{_?;yp&F8Qjk*|OneM3YNAYj1+?PD$;rZwV=n
zzRaTq6k(wsa2;IUu}FF`nav--XUmtN>839IE#ij3B6`Z33p3V1pDGW)z||*)hzf4|
z#5U$te~Ik~F3J|qKe`<wNN1Otc$I~u5`XtwL{mgdIA4xb^cB@R&G$qUX|DV&YaYo%
z&L(UQ+fbU3rKLh<U@cfzC3O-2z?eQ_-(cwL{I(P80|tK|D1sOdPghb|TlSh~KPvqa
z>0sb25<>Dt`sQv!b)&q;?!Q=CJ{$~>URp+ihe!Y><Q%ep+4I|OL&zo!K@3$9_XK}G
z;>>%EQdotYO8c58TuoTLmkppGqebiz(tbA&68SL)Ym2PBh`wyUU1P7^o-V&xxQ%&I
z1`AGU3bH%$YNuf^IXQ&0!Tdb$I7(3)vaau0-GLq!gUc*E{8a*|5H8aRMZy`Da}U;Q
zk0R&2Y^ra(MQ4s*7$;THElm5vNno7%&0ev8lfW&rjt8_-k;O9IROA2H67|DhXL*}G
z{w0f8xhpks8E$F1j1p#xIWiXZjBfJ5^IaMxafGqW`53v4sVk*MQwPMFnyS<{K}hw@
zd!wvxfN`sRTTVb-x;RGfyoA$6%2mq}{0gGHfQRm0M9P^IUVT{HFSh1`bO^0`(macn
znI;iM@Bp<><Rc~X?fKpyA~tHnvI!PdWl@Ff{c(Mz{#_*v)u754f48v>s_2kCwF=JY
z&PSrL)N>3YjsOL_gI3lipNJ2XrOul1Wkw+f?&ac$$3qPNi@V_?<6r^H*bU;p$q=>>
z1HvZjJge#_N_vj=kpT^tx;Xz>wFlov3AYhs1rZwWA82QuRk~@988N#nN1&YOzQHa3
zdf&p06k#ycU~rClDyg|%s8er0F|}lCP@)Z{v-VzuK5T>5<=`+i$=M^z<d&;ru`drv
zD}9p7>g~GidW{KUn8}&u^k7nksXyJs5{jRl`tHwI4aOs5*?ya1z}BK`rSbW4Osuwk
zzP!7-44zPQ)BpsgLeCxhiNT!Za_>?f*CXJ;7qQ8#<^K=Qzj13VkbE1||LKI?N#Dgz
zIi7fmv}!*%-MEm8-?F^f%-T%3{*>l+mq_<~HYo0Jb>M|QdT}F~I?YSYm@dv}bQ}pW
zFz@r0I@tf#7ANC?Q{`j!@F@?F3h&0m;LY#5vfPc>xE`I4U`@X&vyP-}ERVw+47+d*
z8XbB2sGh5Hd+{W*%UR)T5f#vG^F8`sWtR@~8)B#C?QQLfjP@T}EoZ+c4rjM1ZjGme
zhKp}sp}+!o1JpBW<=)Fa2TtF_-3=QXVIC2VH8rTfxH&^ZfK)EC-4LZ7vc)gt$j;&B
zE7W-upbGnrXAHR~IRB-K{$r25*hy70%I93yIb>rw9p~$V7vkmchY;zmv47Gi0=ZcH
zG-@I9F(-ru^j;GAl76&@L6#uC9|fE$kv7K>zz5BI7}0z>OgtK##ulGp-w47^V_Q0#
zHuCjpJ4->8o$~7oo7i0@Y29LCCw36}NAQbYir!Nzr3tMo>~3U^4m{^j8Rl_h?YAiE
zyhGp#8%ummkQ_Q>=-qTpeH?NTNV6{LaTjc#b_*38J~BuhbgN0&^W0t-1FuH~#hhC1
zMn{OGvG&h*`xt>KB~YjZM>i)`Nsmwxb)F$$r<YPHwV!0V&ufqKc|0#JGL-x$z7jU6
zO_Psw{GVRRH@I>XEa__@Kz+APMRe*11R#?(&0!_hWIB<((OfR`TY$wQ>BK1O&x$6u
zr8b-zG`lss-&MNJ<t*)>83H{XmrzrxLCXP3zx616s_fK!t~=CSlew@PbyMR61D!a|
zsLmv7%{S${Gp_IQYMXWX0v;SV?H^Pci3At^G0=Y~46@<M7=4?8k}(Y00h#Mf_oLZt
zEaKFb;z6mASb~Eguu=$Q19<su2WT2XS;fbTS~>dK00jSvyO&zMzs`y%Yr2nYu6#9Y
z*<ebMtaZPY1pYJH@gdD`b5my{U|!8nF|1wUcOG>YTghKFZCC2=D}K1OtgE7My;a;3
zs<??3)@Os1_A7_?q*<m<poqUE`vs{gNE#3&pb~Jtpqr7P??e(>O4r0%Bicd7q4#W5
zm+Xvgm|hj6LwVVX7t5zeMTO<OPj`cE&#GepIHe@n2rhx=mL&(@5j6PcH#(beQ9WN4
zyIpJsyi|#V-!H)R;a@QlA^Sv^gHLsmemx4LH>VcpsaPSw)$>PxMkvdb`kyB?G;aGW
z#kZjiTUtq)U|p}b)@9@V<bL(e^H4Y^m}z1uIIpXRTP|1tiM}ZzqJ;t#i5fFPFcqfn
z2z=owWH=(tAaZh0Ynt=ie?x=L2_;a!FftKYNz>#PZn`{gD1Td0g|}}0wF!xU=e0hy
zHyG}uiLnRu=sn7Bh;pjoG`}Fdv!-#0vnYegTi{ni^fepVyOqj1R+r<N-dlesR3mD9
zm6yuO68TADe<cb6Yg!{id<L<`O<<}=d~#(&Cbq{P4PwJd2Q!&O0rFb3C?c_?WF{aG
zZ?hU~C0biaE*z>>Tla9W>4-K)UAhR`x5(nWp0izSdW$kHF1DHoWrOr2;ljn~@&VXv
z!;to}?n^_JyK<p97NU<|6OUgi6*^+*3?Ag-1|ge9mf(w0fQT!Dt#yUCTC~%ww`fXn
zYe%j1#Dvpwgl2}sXR*)Bi@5Pam3~zMMZ4~y|1|3I$4Tu}z&RUh&RMNk?j%v)?=jcW
z>~TjZwrA8#KSX8dc!BRrSBlPOK7p;OZB1NX&jQUu>3+zhyr%DF>>dQ*Vj_VXb;zX(
zIo_JSRUrlqS!=a;rg_c(E}I>bRiUcdPKf);b;qbq#k^^jYshOCbOAWZUqIUnQ9<E+
zRok82*>fUKBEK$&)hXv1&qdgv{1eugEQ%@G{a2OG^*wXuG3b~$Zv(2H3pToh28)cr
zYev{e<VQ3a$$qKuIJ2AsLNvmpKCza{4-AWhV(06z`y6=Ho!wS{5X#=ko8yJb{f;7i
zLZulxI-bl`%@^msa=c$EaYC^O>l;D$^5k5A9V}#~b=H}vTS^YGz0)O@*21Z`O0#~2
zm-{BE`O_B3P}xZhi&+@oFrY>(h+qjmQno+gN&jLJZ+iW2Lp5>J!M6k+iEQybA<igG
z$gTS1+Cf9P^nkLzr`p<6KqN`O)ppZ(S;pYC?O=*vBqr!*TnwJi-ek!Z-fC;Q`)J=X
z!!_TEG3s|BpO9!}p&gL_)2w|}bh)RXA>pKr0+#up4UogAeoM;JMHEB#{VVG0Br(~t
z;o<{@RWdi)*lYX)(Ellg;`^QWQ~uKqQPcgOc8HPgMHiIdN%#N6MK-MwOPxW=U$AsA
z6hD+|C?>Sb{SdlDXTqSEA{YPm#r(u&eVOc1!~;E`A@Ii?TO5jMV6eX7U#3^=(KbJw
z)X$50!WpsTP5Ivi;dG`1(Q1Mn-iO$?#VW{dUm|NOB$ja@QE>~Ph#-mNJ6c1`0#bL8
zb^;qmJmr$E@2Yme^?T-@OSlKhpe$oGY@`wSL5|#*;yAw8FZzoULOkIY8Pjn%67i)}
z{l}6W>Jt9?%jS@lTQ|d>jTNm#PJF*w6sHB{j<*k-*5Fdj_2a$#24~eH=jC%yHg2!H
z2%3x{t~SIdKK>vjhfqvt_1%^QjO=(dj%>Ar?0bGs(A!D$zPa)z8Xsv49exN|7Z+O6
zQRUPUKl+X}fXYb+dctaPfkRo?R%^7-e7>~5?UUu~$KcRYt<o{R+o1Mwy?Wcy$&MqM
zX6)=qi6@i8KsE9mLTjqme|l?11n=)oV8n8bC54`TgdpCv2(x><>l?;Xf7hVW&f)4q
zKa7KyF0KN+?^GbTpU)om1euxNw5TZ^rg5oVeTlQIZ`^tJb~Uv$0s|b;q%9c&;>51e
z`c$)xs>`*eJ`x%qvrd4^9+drc4!I@NZrA_@HNnBsw{F1nVkF8z@ec`yiSEx14LDqy
z(CcM#uu}3QSi0JRtqHAV&yzQlJ+KM6c}f*i3M;DrmHQUjI^~nLX$wx^x;4UaN$FVt
z&Pmm5(XnW;jlm;3k}hAJAQFW@KVom}CA#H78GhaYQ}k-~GuPz3;-Eyv;kR5brwy=B
zfR<wKis`>_?_yPB`{N!wjgu_j5w7@JOz2OZGu;OL@xz{Q;*pimlR!y=pCB{#+t5AL
z32+inpvA+yi-KRXMGWN<G-5CAPp+`+`51>P^XPZAp}vroYP!sQG4E;+EVla%vHfn|
z#|EaX2j^1(GZJzEND26^KBVP~l=+=c^IXE0_g9ivk#n7S?_xastD0^LpRc#-hdV_0
z{hMS2L%wspFLfHM`AppRfmS04pWmTbkzSqy*eo0l65!CmA7^`)_Abq90Nw2}UpR@b
zjs1LC*2a%m*M;*QVbo&1a^QSx7#|CBpliq@NG4%{O(l;%KbSXgE&km>;pUK}O#LY7
z=hKdHY$ZPGW{US;_!1p?0+d#mw!Ifo*>HlFQT4NYKK3Xx=%8w}&z^)%lHyL5sH{iV
zQlF5F9w7Upp0rr!ed-P?%I&>cfx#jb4Y6>zdKeJ+1*oh}oS%=)jNbSgBl(?h{7G(h
z3mkpT=d=<%WB8h)flTzOV9O;(hB8mL9kMwruOktKJSwJobi9HGFIkOasl_Se(vrzI
z%21W_{aWwfkgZ?Y-lndle*~}utqX+WJJEhlYJk+i+X|(+ms~(05=ByP)FJUvJ{*@t
zA0UuB_vEcydLk(s6&v~4(`<o4;R*T{&oB-8e<%$goyxpQ?J7FYYc+<99Q)lgt?@xA
zX-rc?;u#A1GRWL$gJsFtWG893Ku3h=i@qaH*m7rH?^fI3dBIlxoT$9-J?M?bJblm3
zNSf4ot>h))Cc`&Un_zxPiYeT<0))7e6|JO%<o+Dy>vqVKb!nHdzRPQar^O`$jL}9z
zcyBLNT;7G_UWPng87D7{$!@~X$jtVVDcw-^!O--QX8KHDsde0RJpzB=Dl8CGTWVX;
z@z}4;-=5e^MKO4ZMW5ZX8QNt)v*|Dr=8Jk=B(Ie^;aW$=do+aqxqxHb|3JavGWra5
zNAiX(4t5fh7us?>RPI#)^Zbo>R|As>Q{=BSGd7fK<Qw7V=POlq(v^hD5W`{}-=-TG
z?Rppmv{;`!1mS<Pydzg9Su&G;Nr$7{sv4_rXpAg+zV<=?5*r72EPpGHSIds~KTiFc
zid5uOgrDV_@}54LVWCG9hyESCUBCM=!p6trMfDSKx}yhwN`2b)sPH_?H#Gy4Z&@}?
zi-J1yh5g_uJ$K5DMwa$r8nKfr6gUS~qXm$K^ji+|s8C<D(KjLf!J^i-ww6}Zeuc+O
zkGmNBW(F-)pGO_sA2+=IcFk4BTCtor(T-?tqd2Se$e|)bBICmVjHoi24i@iUlXxRC
z7h&F0_niYUl;%?K$<?)UzeUsIwS96wnf^rnhCSqZ)Lc~7SjHyJYjSj<h})i~cqo;B
z<!FNZ;9|poC*VEd-lpWNvmJMdm~s|4<!`2F@Jj=GBLwAG-OWF0xtFuLL4ct4!Ryp{
z8M<8OnMs@=SZPLy79X_Zri#Zc*~nA5pZxm|HO>Nt-!1Ld6HSL-vHePI$O@TM*k}5G
zq^A`yH&o-#vlckF*uR?APTJx>E?Zt)SPk|+!z72N4GVeicTyS;X^jnzEuc?(yP3P>
zlN08Op$ef~naLYyYId-keedgm&K9%#VT>(izue;kL4<Y4r#AKvbVBM_M%d~z9I`8}
zO<4Xscx?5Wq<N?A*?YIla({jg0-q0{zTX#?(mZ?QNe1#qNwf@mHiT$g$_#1|qw2FO
zRE3Qn>3eqt%)>!v%o88*#7LwDY|8_^$tgu9%gq<`i_mHPkDUVII;Qtw_RB+^7)5mC
zCO)~}!GYsE*TkmRB3s_G>oXw8BYs`AY5sg;!geCmo1pHIex)aefr2!DvA#Du%ja7g
zDT(m(jrA81Ja0NuFR4?npk#SW=Tgxj>f>|w4$m(>3*rxA$Nt%CR|a@Y`UTlmzk=`{
zlJa_LK&$aMpOzv$X4@i{)S2$*P^*VjVyi`d><LTua(@O35dB1<y&Gk`LK{T(DyY8R
zt$d1V@;WlCe!`me<h|mj>`~!zR2>ac!W2?GYJ3BA%brp-`1T-|#o!YMKf7{>+eEb0
z!+5L{^jYlE?$bB4;fJkXuVVT8R&evJ-b|1^3j?y}_D<b~_Hzv8%bqq0jXutN%#N{S
zp56+LREhdWa*YoBWmMqM5huJdnuW}*xjc4<-!<yQkD?eJD{Q;XM(<yX?`p=F2*o!?
zn`kWJt`8|cu`SQ-s0u%(tB!h6I41m=3A7kNZA{j^E^;t3FxiVyKC@h@$qNeCkas6a
z!XmbjC)uAKdbkR<=q8Q0-r#r^_XywwUeXdv7D!0{LJ)HGp#a@e=Xhq8Jq07}zaP#Y
z<Ld?88FYt#3t2=VVcl3z6mP#O-&*edewH>NB@Qmhb~hIpqim9aAH5+>c)Y#7rpKts
zZ(aG`MU5}{u>`^kuxaA4s219<A|G{NOKDv^1`iL59{>GVYtPiXZkO2mN)&g2(3*85
zekL#c#|vwqsPkYpbh%Qbr(NbNcExY?J_YDh8Z^w7#{*G;43yA?l`OC2K^dy!{v6SH
zps#MZGnK;JmJV?nmva<nJXxc&4RYOlv9SV(r;Q)tIq$Q{pIz%CD_fh5NcdZ}s#nkU
zL<X#+F*O7vLH-~Sny2P=>Y@bw2t4~(UEcWEs7l%MB<!}dh)QKkNlS`fL!tZ&X$I*W
ze{f==JFtWVr>_4?CHKX#Q!hWaXl6K6-TZ5uZ=I^?gOjBz|5}r&69B4}z_Tzld}wGy
z?b%1(T+-Xdvq{HW@~S}iyt|<G&wq{s>ILpxD5FIltHoy%AD(m<*-9OrjE#=`&No~h
z&a5kP)IFE8-cJKe!KPQ3n0;~FRcZ(VX?#sEJ~^5!nlJ&J+T5a$c0~u#e_WY-)+gKH
z{YZ+(zrN=FbSpBh_D&uHpO)o?oyPXOrEcIeidy<KdwN!Dq%<0}?*eotcp(%c^=+zA
zKA8yv?l3uQ#A&X(-ptDd5i*?>H1w<d%ovWkm7jj9q7B2GFIt>R_`PpE?ueekOtlSq
zD?*Um=U0_Q!WP0HmHr;0IDH($;EznIm)1y;*Uw?!R{{+X&=}A!?<5WRid`iRdZFm&
zs+(j3E{U~_ZlOKp%p!IEGLh?yA26aZ^Jr+5KW(&=>FBfk$ha?qWhR^MGpKL!pMKr$
z7H_L>>D%j8#_jS#U#seThtMgb!S$HNFps~D>K6`Rg~$(tc6ZusJL_?`wc1}j9<yCO
z9scI(xa-mJZ;=@q_fCBw#;Z&(Mp`&azZuk&vx=6taMInqJ}D>}iJbfaut)2sEk--+
z=k|Oi_(_-E>o54eBAm;1UF`l>4h<9QdG{*Q0M!SAvrL_28w3V~MH0;eJIaZKp0v29
z&8L=>zYp8bg30W+;VutS>>Hof2%kjM&0lXT{-?6DY-;oS+IEp5#jUsnEAH+Nh2lYq
z6{p2DxVr_~QrtbbyA>}E#e;i+;BI$*|C#3lJTLPmbIoM!wXeO`ah$Pjy|bz4_Md;~
z1pC@Ph0t7v#4(dWv8uN^bN0A5tHRbFqD*Fu_d=(_{>f6i@-qE|xqU<YN8oj_yB@#2
z2jz*7*GY`t=-?u|()ju^Uq1RQDLlhs;{PqgoZc2bT?(S7!J(Y5g@B@6x8G$f^tX*F
z%9EGK)onwTK7{&cZ+-XdI#tJ*rt9yTF%A4-|HMO<fT-?D9c;wwDAKXSuPee}g_}?|
z`V-|jdWZ=>7m}YZ4OB1(@)vk7l|OabbzUsXrP!g8cW9xZUgV;}&7FRAIGETMG4Qrc
z@Rcqr_DE;dhCVJEZw@J!9ZD-9!<^q9?2uEH5}_nyZc=qL*#3$=-TZa@m4Q!GdR6EU
zY4dEtdO*_8qNu$V?|#l`$KVS~E7m$SZ(prQt+X8QNkep)GiGJi&2G26{L^daEZycT
z0URxTOx4*qHtXF&(amaJTQ(5U6R32nkJcNKyGF-aYJi~Z%rhUGfjt8)G*Wh_n4ye|
zZ5=c1H%zCvqvVahp$pm$Mk{QSf!7M&wdCw*V-Q<*DX?w&RYHw_PTRtKRD@@Cq(zii
zt?u$wxSHhJF;K4=a%UZOB&L8KibDYJ`ufhIDyMtYH6BaRStSfkIWkwNG4l&t(SxYJ
zQLQF`xN_vv{wyCaZ3~tpf^SRmTuiH&-0V_x6!%xw6rX|$3ILYPtUHWr;FSzYX&~<H
zx%C-1bj$?Gwtn3H8oNz*VQ+ZFS^gczqb6EX5kJu%>@n6+zWdPH@-*=_8Us~~T#}hK
z@N*_gzQ<t_Jq<NQ<OvytK1Fhc%ZHBvS{3PC+x9Q0{t(?bPj~v+kMhM7WcVRoKnBg%
zAgDwd4&-yx61pBK92Zv+Y|mm1x4pD&7t-tL`2FFqX0wYHeK=s~xP965x!8l_CVVZ{
z_K0x8tkMprNd6zfnKV`j*+}r$kg@GK<UJqBgw+E+_F;;cO$h*q;#NIBAn<O(*>2b^
zx|WPHq$4pL3fGtHr0?`l)0fV5ivu;yzAajO`U^J&U;4RYQ!5P{U%oY7%`9|Wu8HJR
z(-Gzn&XqZwD{+HyJ~^i-OJi@j#631hjddE()`yHmt57~FEo>vS_;W7Y*7Lb0Gi9*k
zn6ht=38plebV1|SSqhBbd_+mA=y(3l&F!asQT$W6(CKy$(Z6pXFhQ1N`5w<2w)15j
z9odK{jC;(c*JjE&C%`l3cN!Le0fv>-!1OqU$tH*A(E5nr5y;}!?O8NyN;W%q`bp56
zU+>u<+@g=_-?kcDzZ^l_9i(>K7B!WJfC<RD0TbF$YULk$o|4$ZTK3Tp7Bn-Juf5i#
zmk(N6<AO6V26B^qw`9MpSlP(*ur(oGT-m-STm930KOTO)9_R>p&eEp4?|?8Okn+GS
zT>@*2p4(bjyL&ZTzw2()&PmDGE;^QvR2&Le553=b5P>jB87lho$Vtn?l+lg~B_KxR
z%i|BP(Tzs6D+sv*OVrpQ)s;aY6pdCX&P7H}O~-L`b=Z(tWH<g$;qF0h&_~1n^QHl8
z_BwTu6izC#KH)!)2mW5RbF#3bP8sYsmHDkGCsde^HRK$tEa~}XD5;G*ImVKpg>AO;
z0SjPieNi@v_gnd&TD;`@&{MWwPoa7#qokn6N#O3+-VD&~dsROYw)rA29lkd|5CR?R
z(!8zwUPV>1syn!rUoQL%UwMJqj8@iI;X5^}PCrMF*a9p@%>$|C_N5b0vDn?GY5<j=
zMB_yoaMQ1J1@AT`y;2t2BfLGpLzPcoLNBY?9E_@=a(V1uWt&(_Az^HzglWJX(l{ng
z$AM$i^3LRJ17&=0UXtUTfB3l6H6k6;KD2&cDj{lWFj(iizhctLD{<n+R(xrEne$F;
zWTMc3`1>L>x`NI|S=4kN8UhC-(u8?-#qWPk>GvY!hTsQ3>$elCyZ1wiF<g4WLy_WL
zLuv5dgHX?Lhd<wLJ-5e=HBN`ChC~3Nb*tamSY<%&m6m>oM=6^zt53mj_xLU<jsFs$
z+7!bTtfWnIX*qrRZsl2C2PpNKBcSH+B|k&mL3diwt&@f;dm{2!9RBko{UfgEzOVK;
z(u!RB<EG$+pVCLuT%|5@R9oMot)jSED)vTrioPIz0Oj^W@lj%)0jowMC7G1x&|!bS
z8jo#;Vag_P&tS-zAD@%yT2%EL(=3kPHW|{d7tW{;cS)YAeMP7hX1nE0-Y?`Xo2WM8
zo{ZDZD03qs41DNAjs+};*>qaDdBI*2;&ndxKcbao6C<K_tbZ{)FZ!d*E6D;Agfm<I
zGB3Qk*WQj_vvgTHBP=3o(PfIm$7y|iLDe!(AEBRH7_-Rhzc=T&CRvmwFn?dQF<b0M
zqWP<U7P0kRz5>nTY5RGuXUdiv!~|-l`MzV}<lMlqzS2Ole~~-c_V|0Z(#wT@^wsI4
zr@qG&5gW_55RCLcq4&nPY3=THxl!`IDoRpT;zpQWv;xxXSwG^@XoKd}L*FO%<)zDK
zqovItLa9eyc^}s^NBSOwpo5eECJF{8SYO;@WB_kXji669J*RJCWvLGTJAQ8B)wqTg
zfj)0_8h|8EoDnVQE9Lv7`ZI%mq#5+An@W{tNTARN6$1|QPnbPSYg9La+9fs9C|ese
z<Jqn)R>lS+#5;z<cYc#G0-dtWAGWtLrFQu4*f_jDF0ESpgBr@q^&$cOh*M~p<b1(t
zd3&C#1)A@3@(@!!eh+7LW7_s(zzB-u`IC}@y}RQk*MGf!3C>J-IVr*U%&TC*u|K0^
zf}G~(aVqCOX;9?~8S=Gm%SmpysOc;p#n2FFL+_-qi<KKvX}`DTGb7dw{e4R;uXAZD
zz`RwG@dtyo=+V7FQdcM+GWiBILj12{8vS42NyalL3#a>6T)j0v^vhY97Ye|3{<X+X
zA^wT)?;DEPpj^N&TI5R6kd8pieoY)&ny`o373Sh>p8cC3X>-o}cd&TX9WRcHtwxh2
ze~ZRP9M;=crn5a~#rp#|M{`;hwOYO^_%dziUVV)v8hoIco9tODo1YCDXl76J4dkHs
zJH|uIgFy6&y2vc=8okc|G~xbf@a2A>Qf0xX(+!2Zk80(AxF-L5UPvX<_+d%5TpcgA
zqP!(O(--EWZ=7YKE7&jl`82e-2$#Mk=OIx_xb!V}a)k5)RX6E|4p?EoHZNl`{meSf
zElFi(<2+}sSK$)C+7E*{(ZZzzZS(WOzBm*IT!+MBAvn<76h#v<gae?u?iK`gXeH4w
z+w#++lyySVTbEkiMGm?8*PW8(ujv2EWNZHjS3(u~Qne`+c9FSIXXJ?32E-%c{sy8P
z>fH7~=|ALqjKjTES@TEU0!YgbU~Rgb@O-&BENqDu=QN24{s@~={eReNU6E4=aQBoJ
zzI1Eo9rrzlHmvAQ_pT8W)21an2KAa{)?y|A9Zk|x!u5+hSi>x1juF3}FZ0^t#hMmC
zTk?1Rl`0FN@ZeBs%}*q{pq``Yw4`h#`F;oqqW#>abcnmkW{@bmgRNqTdHaKWnv?Tp
zi<)ShZ8BM%r9)_kR^mH7aE)~=<p2}_W(~ZNto7dgXzZZ3@BFjj4l}$Bm-7pX8Zs;1
zVW<vNMM@6fb*Ofzkv8Ct6&aPW;4k}yzGqr(cM0_C!BnDq=CuT+8q3!Y%mVLPn31l@
z5D*ZL+{5qvoB$1cx$<~~>mmKIomZ5(N1(F&ceUC9UU;u$?qz?Ai4JImr~!Cdmrn(%
zu2>#`Z36A`a6A`pZ(gfeo5zzY_B*8_>1=Lqr=v;|31BZsE;nqdPa^0o;xzjW-b08E
z47SBlhPGEj5|*c<lUoLJBZ%KRwC2qy9n<FWPk}ESn>#eE35?#=v_aeN<$j4l4(6aT
zV&CYNpDouTGUyjCi1~8B)6iSIvL3A4GX(q={PrkFOw>mrUyHOm+EBK(61DJlxa}Y=
zvpU3;A<G}b&EnI(YrI^Jjk7L51TBb8@jBSlTD^pj6zBWX-?)_H=C|6h#PAzI{WZvc
zM8v3O7;-6BI)Y;1_KG^HJ^EIRfvGe0h6tmQR)T*>D4pDx3XhMir#gllFyHPVBc%u<
zIh0{{Q4)<XW<r7YkbD6v<Tr`O`W4~hD(!zRffL^7dSQ2s`ExQjHWD0VyIM_yK0H-=
zbo%E82StvEbIzRlX^k6~3FTU1%ikrx)V^effszye!*-1NzCUcV?sxLX`kq=r*3*K1
z--hVr-4w-*5H-<v{8JH6toHLrqbHDQ{%2EJce3OpokkDSO`2=cMEx5L_SWVOMGqgj
zQ*U-5=F~=YXAts;T&lq3;1O%5UclqG{dM6E)qg2n1#ni+!>H6;%620zIBt{BW8Mru
zG0#P>RK_T=)G^~KMfa>G_|GE;A7Ur_*1uo1drGF38p_<^QW3E@9}!zoGx1zVLFb-K
zHy13$*N*>Fr3O2qj}Vp>9|2}$wy{d$r|=3x*aCzS9Hh&~kS6BdE*riDGG1ni$r0zd
ze<-U=Y|AifYm)6-Eo)bqks@)j6Y`?jS8~tbvd~zfuO)y#p99oYEMYrOrVxuTKXkq%
zdQ(gA)Yawm!!m77y7@&vM%W&6u2*-YH5aIZ)MSx_SwZiSXxb0S_J@S_V~BL@<_2(%
zkVMalhm5@ruCOY%ErBRG&ASCDhgov5@S83%31k&=CA#qzFb$Y8gbK&gCJP?ur+n6j
zScb?LO@*dr>ImjWXMO>#pm{~#CnB(8MdF)1=^*bI3CS#>z+%>Byr~@v=zrpf04!XI
zpK=_lh{0<a@q{Vtodsz!L(DvKdJZx+V!W@|g;cR$L)M5^Rh=0?@U7$<iDdlR8w^B_
z))@5<WAaIwGXqPSGv-Z4{V3cM9}`LO9OfH2v|3}IgWr5^TfyfL<ZJkhGKn<Q5{ZG4
zijwVMEA&0qp3HMKORcTAZ-2sWLndMvc`%rT2aYLFG8n)GK3{BsgxKJ9JI#;B!Y0CO
z9|!qmJY|3-k{d`EQ-CObw`(gm7UEO-^|=?BZ#i>lECh@VP*o-AD@r)(1tWldjgD#k
zK}+*BMMMV%5!m_}!(R44G`0JHo9DMNDUT~4yOG|iC?^WL^W<<c=8!Nx%F~?e4B{K}
zr&n+)6g$d-ckU_+S~R%75W#gkO%LlC>wRcpCpbZ&tw1EYJ#W=mWT|tW?TCwVq%jHJ
zZ5KS~_XlicZ_m(Ff<)Eoc0DP3&ouUna3KCA;!{(JC%^V#j@Ed2+jk$ox}cAWLT!99
zV{S6`Sf5E!pbkhSHnkJeY92(9pB&KAeQ|F$V7FrHKUySb>|r7t@3da`;o2QjYJ2-E
zl0eP3xq|9^GmNV)WVzfw{1vw<x<UN8+a4;={(xMLiS|whm7MPX8G9BjL99`5gKU$x
zM#@>%PvvYEccJ>Gs@sPY-40sv7BDqlLtb5#lERvjv2ldvj|V@y1W4oI5q?6foBgkH
zDYz9~77WRA3H2zcWY}o{$s6S{99wPto*QrJpn(T6L<$P;6w-mrj^)^~3HEc<$@Ar#
zu1BGaCe<E?A;(<Ir41`B7_b<oOAFJQlVN9{7URam>cg=ii9d_l>Il^PFOmd`>1Qb~
z3pRU>UF}qc89<MvaT{YR_+a8OG#G;wZj!;{F;X<~G?Jngx)%R(o8Pq{&G$ks2@6;o
z_5%^nIT7igk8d`?7dk<&){Hm{^Y6MEmpLb6NAb?RV#L(-e3a}1$K-nHSQ56BE?SS-
zzKRfliD8b@0NCiaPLNSckiejZIHIcS>%|)49<R_GW>*uDH*dNz!cq)(u{KS#VmiE!
z#z$qe#Kt&K+kHG1Um>#0Piq}wKO=`<t%W;Yi=-RhpG{6`<y}m~$1DI|ve8&mlVPxt
z<i^|K)%Eu6gpIP3g$X##eakK9BP71_oI%LonY2<1uF7D~HA|i*C{WE74%)kA$@Eg!
znPYM-mJl@_i?{w9eb8vzh22s(Ln_;cIm9g8_BT?FKvXFv_*g_Mq-vpyfK$8>$y!#L
z5HC`{Dt5pvJN{Q(&OJYw_BM@;#1qD%k!V==&6QG%_nQBu7E_mv2C|uqtcIY9)u)?u
z5sMY0Mfw`Z@GsxhOV(Q6|9WU)zdO;8+(8lH8ZB*5YiQ-W8*FdZMDP=0dd^~WiJm*=
zt~Pt=;N_Vj$9W{M!4vHQ`Or;(Pp*_=$=ODDBkM23%0^kMV@2tJ)=W7|KG%?Vh5QM~
zBj=eVy1zJvbxCgG*l|2zr?j|uzMwrE+R?2008;tk7UB*)p8qn#u<j3QIHQbisSbu>
z=9A~A=hKKPEt8%I<crIv&?w|Bx@rMVvw05G9j&YqoDfGX2N3UVk*4Re0)z!uK7{Q~
z%e*iDlLGaxgh05VyjW)Kcw=ha_FGJwCasT<KAGY=HIc#g|7{h8^phzC-ap6Ma!1hH
zQ%Y&DVJ<|?tXgs931W}lWc75o*s!b7=x^NVm(!D~{OQC3<%dNk9-VxnJV~JtTESTO
zW)RiOr?*#OQCIlepLkTYLlW*!QnDx6k6QpY*W{yGaQ4Kn71sTcL4D(nF^h&f+YT{l
z-7@pp$mycb-`-UkUD+^UJRdqXH0!Fg)&M_~hN#9Uv0QeI+P7CjDaYO%`l8wO+81e@
zBsMvTeoX&kJa{IN%;DYWI$O)%Vhj@%uJcS1eUI;UN_&pK=Kq$xDt~~U3*~q)gagvR
zW0jxubv<$6<7IOYFm~Em-N37TJlCi@Lq1ZC$Y41xK-Mg@!D2-d-s*W+^1uh5P1xDZ
zx=_}nJIER;$ZyW544R0)mYXA+dc>Z3FM0FhuPIo5<s3|8CT<7%HQIFS*5I<=jE`v-
zt*y%c(_sab#6Bua`t5F+J29Xc!iz}rJBD4eW>z8pGu~ccK>W+co^5~{Mp86<2`P0(
zkvNV<88MPGlZyLr@fBNZEM3R{2<F#O$f4xKHvI=2HyQXOm?Pz9B9|^DLR@F~7txrk
ztMG2{+*9c3D!3Mmx+#>{76PN$I^<Y09(%7jq2*Yy_^^a+C}@a18J3tsGkV5DnXsyi
zEcNyR^XQtj`Ja{I`p=t91rkwir<v60+##uzb}>S*wI1vKvJB^%mra3@5vK!Vs5)Qz
zKxPI(x%&x-3=2WCh;6hn0N5$<-9|5mWU`)a+#XO$g{YP!XE_8PCpi{~AdTh5f5M_O
z<*Dk791XVh6f68NJVB?GLtFtlxQp?<{#8(I>vpkiIj+uXRt@t;heRS#P;ovb%XpWa
zdJ30Pu{=JH(2$iHQWjNI824mVzT<d%o;=Ow$5ARFddw{z1HYikruCQc99Uf^(w8cP
znh4FD*JVlT(2hF04f&DoQ*w%$mIOYl?3^rcGe>e)$?T-J1fKX*KY)`TS+`t;+iqol
z;mIsxo+5kX)byt1*akU;m|zZjd(20OcPS>rCAaW{cUnf6CRwPR2=+ewrIo2y7W-VF
zFS7;ROoKQKJNuVuOz-*04Uk0oNVFL3Gs&2L8Gh=7oE5ir@4{k%K!#b1l0~(2u0FP#
zRfF1t6rL-;Vpogd0$)qGX)Z3?W*d+_err<Xe<3W>bl^3cWsSyyFJ2TWMw9<}NIV~*
ziQ$ZD{ys+0Bst9nhuif+JFy#Wo^0?g<R5{X!!vmxcbu`8f1~fA57SGp634IDCaAp;
zza;?If@S_G7n7&jxf$PbT%M!)+WYxSU-f=|?El<Ec)H)a<ftaW={W{aH{+IGk6!PI
zZrOQ%q!d?k_J9}CdGlP8$vd+hds~W^51*e-gH5G9J|H5Bm7Lv{T%3_J4PLlb)}87m
z>5Y3&j?IsouNsM;Z`kKNO1HtFxceUu1Kd~|^xTRhcc#Vb;5k~SsfFAZZxddhLo0G}
z3XQ~ZBoc-(SrF)~O4ph4UCZZ*ti_wXrY-J2jmOlhiis*74U^XGDF#o|;_$zr)y@`6
zw8?qBzn>RP5rBm6&UEAVOJ<k({k5PV(j+1pV|zP)`lHP(b>scQxOX$9`!*+qW8l9X
zl5jG0%>Q@h5l6c*)llUKpMT03e3@C%=U3Jh3r`HSr_QCkk)ozTX`djmh^}FCtR8ZG
zR+$m09T<-8{%9MBmt0E=zEEQ%PkR|m(%sfdJkBxhXuIFF-ULoL!_6H$65z-;kZK!_
z1a9-*N96h_x>)h%1)TQ=r(d)K=^~jl86(6e6q79lPbq~DD9ExIoj{%4*L6b|9@16X
zgV|}qEQj`+s8Tw1M}1<0ERG>reUIxh_04_#vknTAnY0br@_k|l!!<*`sA<3Lsk}U!
z@UkCvjE;sa=?oLjDp?*_9iegDH>;F$iE8D6Z{!H^dd(}?mLaV<>&K5c>C$U<rDF(+
zBwcSGhT&W4<auN1z|zabZ5Q3#{B2P4xZm_@d~R9Ch&V3ZFk%$`cpJb|x#HEPw)vvE
zGgX3h+a`dSulU7{PgMB0whKp|ZaYOSA)@=XU&+F~d<!&h{^TFpJZ8{4F7u0)dEhqN
zo|=NMK0%0h%=<cW^!b!8Oz$#x;lD-nKT|YZfbqiS<%O17baB8Nt9%Vbm%X^F<TNts
zizBYZSMI9Kn*{T95B4?m%I#GyPd}~tn^@Z$K9Ab|=a>%^qoD!jRaumQ)@)(XA$Dkz
zT(^4T&z~#)WDnRDwi%qp4jcf1$$LhSF%#wxlY8}q2Bw6@OQdw8t9G^Y4RJ+9nT-Y6
z%O1u2;bw>3cGDkqMzqvv&h@`@i&zkP*uSDfu@mtLPZfv!NzXCNJ2|zh{CQRV-Lr3M
z;Wdz@&=+h5z76>Bb5e|ZO-N`7ZT+cNYk$-(xkE(rpgMI)(I^#%)(-tkDB*fTcPXhp
zWvU_bZ*%x?;7BvzwN;Zn+nzX3P)`);;cv~*ed&MKE?<KJ5O)c_tKazVES4`8si5tz
z+7p^lO;7(^@D+St&<@1AwW_~+536oHW3f=Io{~7{%e%j{O{KgMIwcP0jnB2&BU@Vj
zp?$AsV#utAEP2q<t_=jfmeOEJ^BDj@8mN=3AX7vR<P;3va5gGRbWT2`KxH5p+cye~
zRF=A>5$&W8?2<cu;5QUsDT0bA%-=1sIb}*UN|5Zo=)*+hQO{{8Gr2fMt3>sBO-h#U
z`lPO986`baZ#KaN@%x(30G!*1gOlu~1wG&M#uUBg056Bpl^@=B7zSsPAADrwOX|h?
z%==Mq`?o=+f0Y>lA=aeN0Z!%Q&y!LnBGVHP0*irT+o@UU`+OPh<7o+Ou`%iv#h|kW
z>rLj=$8^D8fw7SK=;7J-+X&I~J{6!$SfdwNi{UOzQ2vT2(#g=~yxi>SjQ?|_q{{q5
z`(FbAoJULXX$^7Sl_^O-myl|I3$tfb>NGU(2xK?Z=KSN7f(@{Lx>MY=MO;G7F>K!z
z-Dcl#iD@YGCE^3Y!`V6p4%P_b5{`a{<?iyr!J2-7uSaIUey-h3p^~S5z3M+%s~4pq
zIY<6Qd;}M^w(_a?_bFVylVt%PfVXye{5K|Z0p@&m)Z(9p1M%+WXmR{YgmIqK?Vj(J
zM@J18CP9C6S;N!?LIboiXQk8h%yi^JOSc^DLFzs_t*b%hjtIu`hNp_6<pARSbhFJx
zC75=&1U$fPwvXy)TLIptk;wKMxzCBGj}g1B8IG#62jnTUY9M5DhKHSC7{CSIiVuV`
z7}*eQvhMuL9yqT#{g28k!A~b_<f1q0GB>7g`dB{IWl$iz2o|JuJUm%mzTSZ+2dV>L
zQEWu8EL9JzW@1$jlXLDdjKH{?Wk)<WPD;mL)282uUGMLdZbn*lZ@hyCYld!x;PGS6
zVL;Zal-I?kDwC3Cu}!P}#4u;*NJ2y|QchLSk}M}fGkZd>I-;(;o8Q~5Pi!XcEv5Ii
zLotVmMB92J%aVoMyogWLL7dnwcfP$mi?8L{hk_~P7fPHD^@5_@hHP|J4WZt}On=i9
zQ%CstnpM(7#dutJlU61^2=elpvTXWQxmp2+<3lTr_E$E4ROWJGg9^c%@OMmNBHc<{
zU%|z;JF8#oEw%<P$|>{2c-Eoc7O|@&k5f)`zriTO=`BbF2evcx?_>0iIKAimc}4GL
zC#NJXniI{}<U}FT!QNgwquM$>M4=wMb|iedDvu{l?4W@C2L*l}13n4ks6cvT6Er}i
ztbj9P%Qkq}1%O40*57v0tMK*wI{r_$dkv)k{N=2RD><Av^gv(ZM!PY1C<Y#8Wol_F
zfFUAtw1H`LOE&ih2)RJ*nI?Hw3TkTK&-W?cSnI3NA8sn*?l}sd=wTfa?E%!34h(<)
zYUtu*8K0#2S#f`OnJMUg==iT43Wc5&YO)WL8+M?5zOc7AI4#z)Q^|&IS=1_SRQDGz
zKs8D!fgpfy|JJNUv8leI3B&Sp+}>8_LDw5s?hpI>D|x;Z$>2{`J`HFhlE-+GF!<ga
zXs+*pceI)d+AeoFk92AQs>3m0x2SB_RP1;euJ+`AzN7xNlP#=50gBf~U_7@K{1mI2
z%l-*CsG7lZe?0Vc%nxxy7ubgq_LWae`ohJ+c_|(vIpgQvEO`-UXFKqDq-cE2M|)>m
zzL*-%68p9`aaS7UeRc*GKG#jPo!trmsL6+%57dvN1a;4YFHWv`UX~7PJMQ~!&v`9c
zGe<(GfF#s6E!3dzZ|fvA`9y;)nQSTiLg$wuS5BF4t$Uh#)CWugKj!wJ13Z+|pW#7e
z^IqR_I?SjjS)ji8b#ONdmf}3o-ZU98-y1XE%b4fkvuLpn{y-rujITk4K-N1!D&I^F
z2X}VEH>}(xtMXATDh5|O(Fs|QK{`!DR^1N#C|^e$_Z0Zdfx`AJE6KSJY7yNoHOZ#p
z+*+C0Y`^QJwY_hiW#DU7{6@s;)ZvIdz*H~r&gZx~PI>-;B+u*Jh?*!*bw2wDO8D&<
zis|h{r$<V&<)b0dlm@sjj>JM8TD5)7v*=M{=h%6<w(iGEE=*>>m8Skx;#@my@P{oN
zedyIcBCmG;^A9>O{?BgRP`{gm>^`q)nAc!Rc55tlI8Sw_CQ9AfAR%hYen|OCA`+(a
z<Nm7L4oA68QB{RKw>gl=SidR`qiAW#<cd1C<8n;j0;VtexTBPv#qkqd+kKQNt_!b)
z^@7Ro&DnF#&atldqj}vo9C@~6-X|m=POO|_HbrSycrLE`kqMMR3(2mJj%KeV<!|S(
z9@-|^dk(t|M0n=K*qXKFy@$di#ll|izQ*%zC)p+4Z?8PRE?Y|`jAW<I#`I3z-o{!Z
z7M%r((@i-yZqSh~kfL|Acmb>GpGdr~@r(DyH_FVyW_>VS+bq!R?fz%R7suiHW@xz{
z8CIvW51-`xL?0(B#qw+|BTTtYii*ml&&Ra*-%ZV$i@9Kl?{#I1i$fnj1<|;AOjut6
zg}a$irou3fbij@gEUR6u`zQr;Mlk=ATDe`mlpPbqRV{iLog(BP2|V{2AFmSLeT!x%
z;<J<be%5Bnk8=E1;ua$<RO&Nt=?lx*4)(;tw%qzyKr`|7tqOW2%W~d~Uz>Z{TE*7G
zKtgA&5zq=-5h;Mal<JS1b<fI{rIsLd*<*!0_410^+b%l(JGFqB{8|TN=tCRj>$0cS
zR=;iPz&p#?@PR*1kXBO`<B0^zI**Pro8pY6UZY(Z)`O3{v&4h}q{ASmcLX@_`l6!B
zj(Eb9hjA76dtr~f(m=f)4T0s}S~549VNA{f1Hxz$MCFSSsfW41f5x!W+f~7*ux@hq
z03_71w;$3;VMG8TSkd44!M2SBXW>ZKU*9etEeQ<aalAmX?Mbs4IjgCDy?7tXGSBJI
z9|aF~>~p5yf^s1rwyc+V@2!>n4x#3^on?vQ1R;{;!(WHM2(v8>E>0~5u=l*uyjU@q
z0%VeEVryW-eGNHWQ07XoAFmroP^W$4C<;n=9t*ihNb9PdN<O6?znqWe_OD7loRnQX
zFtCedt^__N!ZEQ9HkoVA!-|)wMUfyqrbBklp-S@D-baf>Y7cj^j(A(`FcgoePUj9>
zdPOl%N)P2ev>9A5M;(aUkGf6<o6aLFJlzt{%toU$5{Uw_4JO2Jn$H35OJ_1A3-6-<
zORcGmDkF;BrRlGKxR_43cd|a!2R3UB+F9)?s$8~ah_`pO04wE4Ccmfw6;}O`lLTN6
zcn%?g{hheUP+C+?(1J8&m0{CC(8(z!UV+-r538MWMPO}!dvH}dXw^zwM9=tmS+#!H
z#!Z;-dw9G8<NhFj!)M`cY(LLQ8y1&!I63827DQCT;T2(^(0H$||2+6QJi|&fFsq(I
zws#p9&B9aG8fRvPS?{8wAU%jn?XiH!M5ENX>8?n!gQC(&)6yz+2s6er7&7QHkT=k@
zAv#aC?b@<x-D+}YSQqn6)((6o{jC1f9QIUC<1cH(AD5`bd7GZy3n5RIrCJsE^>|MG
z^jL1#?NMpCTavL<2H;9ASBLzgA1s8Zeo+NTZ~!b4z#CdL$as|leV<&2rR+s9rnq%f
zpiRyOAiutW1s4O*G_ZdG@}d0w_6%fHH1??TDDZjgPM!K2(Q9K{mezFk-yxGhT1sp`
z5+&=+J{6R)wQnqC2r(nus{Sv0+nn~AgGglM%nf9-B#N2}e><&ezl8`obK*`0VG>(y
zY~Yt=VPZb%07yc<uCB8v7Ut!Da$`o%`yE=47M4q!P6%MJ>|V|lcEi7*>)|V$PmwLm
zbN=#(XnC+0j572(e}1m50%Ul<`{n9eN1T9T826Ck_}2{I*~|#(O)p0^#{8It5^pFF
zb&?q-ai=CnK|F1gD3hD_k3kJHU7j{SUn^007#_v0QO<9CFrq5_?qTZMRT<7-YHpJ)
zT8+_pVKgI(^@+=(wal$=_oA4hN}w+#fJt(Z3=3aOpeXlQEDx_v`w1?9h+wX`m+ivb
z+)ZTY>t}c9n-fNE`Jm+73>4iCCV59~xd%SAO!1LyGDEF>;QIW|_rDAl-l`WA1Vg57
z^xneX7r@l_5@Qrcs`J8^7UDb$-WeK4m+XtJ3&KZ7S6+^$NZ>!&X`PU!KTz#P_GAY}
zUxpQ}?a#t%EUg?CV!M87FVE7`zVtM28Ws)S<jPCEvBkrE^BEaA=sg0$|NlpWAAg<J
xk>A`7AtOq~2ECaAz`qv_?F~C3qSUCUM{7shv@F07{vSkdl;zdrs${-|{2xkbS>yl!

diff --git a/master/_images/sphx_glr_example_regression_001.png b/master/_images/sphx_glr_example_regression_001.png
index 47c08c27b3830ca8c17c22318433b34af5537f70..8c4e82cad6051caf8e711e911e11e1fd947e2075 100644
GIT binary patch
literal 54882
zcmc$GbyQVr*YBc18U^VRK>-QrZb1Q2L;)#j1f(S-rAtDjTTnr|LAsGfkx-;dMClU5
zJD2Bt_rBx1@$WrjoN@Nxz}{=E=XvJ*)!bpXZYmR>p*@2j2=R3lMNI_3^g|GgS$tgh
ziO|T@BK$|pMM=lyj=hD8o3WEQqHgTsU~BJU`_P2h)!fPXp}m~|j}Q+(H?x(Ci-WT`
zFYlxO`3pSuPL{kA#@mhXCIk*Dy3PnfYK;Diktg%!A%c7wx~?dv<({_w)lKJ)<uTso
z20@mrAD^PEqhCR|XqkV}Co0P#!XnR8`cLnwigLybw2SWPX;<AXcxe?eMq(2mexFk0
zaduH?3E?@6cE<N&%`$vX10&-W&5(nYceT^(Vqz=TL+$M{R_ZoV=R~8hSOUr9@$&^0
z#~6{nFUz>kAOz?uPMZHOFH%Vm?P7g`U}HY-UoK1DM@L6Vf)6Pg8X9Kj=ly+t|C*bu
zu;d)_Jlte?vb&05X=$ldV#vVcd%|yaH91IG10OEkB-K_JX>j>QQEm4%++rXle@e!*
zi|Ay#k5|-vllf-4c+eY-OjcG_`J-BW9_468sarZaq0TFV#^c2X6LroURNOibz9q{3
zeXBU?s0C3QQh5VcSK<3r)<Np2!p4*3=2jywahiUuajCr0bocbUaP=z1@8bhJWU9^?
z?-`v)*U`=jIW28?j&k&T<H2<M$2S^t3kz7tM7cSk#Zbnx(NQg}e9g`j-e>UaSvuC%
zi=+^BcJ-7%MyV%Rci!C8skTktKIPIb&dSTfPZxizIeqZirX<_-$LC&T^W<m;pM>Nd
zt}ilNWuvW@CQ4po_f@_x;fj!9MOGGqgM-uig}aYcN{YeI(C`6u7g5P`^i!~BMM-pZ
zb@OiL;3~5AaSwjhM{0fl_~<u2p-kj4;M~!<apTNet$a*3H@Em>Z!fQy=g(hCzINGs
z<F3DzB+PZMuw}k8LY-a;-5Q~eaG}_~zP=IqeXlWF;1L>su6!u!b0SL5z;NZsk2>cS
zljZ*8-Y=T@=c;VRyWly@E-zb*l%W@rYa;Pv_Y(Wfw9v>%qT1S8W(f(p;^Jbe`X+T8
zKUn(Kot2@Hw|BXgyq)UkVWGN2X1oGSyQ8v8OUZbRnzWK}y!O{=yu7_Jkfx@lTKBD&
zy&jahZQ-VZ`S+_h-`*>_zw9F-BJ$7XWesg>lkZO!!23Q{goOkK20l36{{f2^q*RkO
zsYT`B?k;LKRr~bt(0xfmMFl^YgyHRpJsuujM@Pq{Oy`5229jQfWX8tEh+k<b-&Cz5
z)=;L5Hly>&p<``bUH0`jb}C<W^Z>CSF-b|GBGc{)G$#W|{8*Ml8A`IUex2PWcLe<4
z?yNpn=J$G=lu^T)^d5Id(gY+6Shjuqc-GF&F6pZInYOkzGFqrOs~=twrU*XtpKA{>
z-CO&la{YQMtT+F|4{Xl_y#9XBvg>RMMkJjGNpf=X=I(CCV7kOeg{214DHL^O;5DYK
z^jMJ|UaU$W9F=U?4hHqEov+nXUqpYIoSYjs^6l&hJ4;SQg<GZ*8GAw5<o)xD_vnTG
z9`~Gv2Fd<3v9o&BwirmmRu|nwjXjg~r?=*x94E~fWc6Z_lL^8UG4r+FJ|<=EE5F3e
zUAft#;;}M2i%3XF*pLRmzDWKG2LKC?q+{><cq<>#@?d&IXT<sTX~V~-&oug3DxE~I
zthmk)62`uGft_GpBy~=cq^`cceSAF8T8%<p1!u3FROZ2aM_A9mfD+t(ndK0p%g?FF
z&+VEokzsuLZWb{y8dq1>yt@Sno#DdIx1$$C7nheg<6fMcoD@p3!aB(J4i0uLzK<)j
z8RtP-BsLlX8a)rz3KdO=&mpvO>~Sn2BIkvyNBt@)Dyl*hdTBT~IFJ_Dakm?vJibrD
z9vE|G{@31jYEcK~pHp>M@Eqa#P2r{R(Vrdk9e5J4kUdKFYj;lOZfR+m#r^3<>x4|y
z!sd^f851}Kmo8nB9Y_%(r=ba}nSRpuS>a<irtI0_(<>(2zt+j*RXUmDZ}9t^xYLU|
zAjsy=`VAAPI(hesQsR|g(>*f0fG#qB+frX5dIr#6+TY*TsjwjSIa)fON_vW%LBhh;
z#)=|Jntt_uZs&`LVvsQTwfk+15f@gGuHBIyElZU)<0J<gGs<7zJ>EK)@iiSQ)Lmld
zuk?z1o)l?Joh}nWb>-?>1DyM|!>xrN25z)y=6v}0apa8#Q$K89B_*ZM(9jk*yTjFw
z^lV56m@Dyy!$t2!HPl-&$Fa8?&f~qg_M<hlKaMOWEv@imBQNg~w@xYIwlRHodS!lI
z^D0kNfIbiJojlF6_V)J1csJl7P2^Qr4w-Duwz}`Gl0&KSfBb#S7#4fU$w*`Xxqp3Z
z$%<{@Ovl3AIhDz)MA6zo7arbYF6U8BU3vlsy_X3#U$F6unhjVNZEbCOsVC=(^lGks
zdV3eW#S8rW<gn7g(b1&(`ufAK)#=hi9p#rA)>K=F+R;P5`lW_9Ej@iCOAd=(%t>)&
zLio|=KwM(lmXQ&1X=&-zndav0GMUpuEc>P-_NlL5=a!a&4yHXx!@|OPyO|{2kziv7
zIC+)!GmJNH-n9Don)vkhfsy_@7XlU*76(U1xr3iyV^dPzdQC%#9fs$~ZP-9PJw5H<
z?A#tmaMs1u_41`lEy_&Zfhj^Z^4wH8kG@ogK6`do`n37aZ&E4G{lJhAybcq7e}DhY
z<BjI;$Dd#$lsR(unRJFzt`28oLCHn0ceX8v@Nl!8w9kAK?-pW3shgM+doqCZTxBen
z;_Y+yU$n<fx`IDyIhZD<Q|EM%f`VeWP*=dr(sJRe+l-Q$8d0%9eX!lvciB*)+z&Ui
z2Y1j5#lj_Ofit^>?nn4A6%CDX_jTBW{EyePxOL01VPQL=$!0$n&WwvgYuEhtQeS9T
z*i#iY`I#o~K9>i&wGJ%HLz&S9+9hq%(~JtiM6Dc|UjFc=3s8S+-8LAZt!8bfB_&a_
zv9U1=3sb?OD7d%?Qt=uV?mbq|kfd1|%IrS<eeC%A*W-pir^l;fMRcdf8!|{s?)A7^
zw{H1C8(rvnMyFF{9pBr}$r2c|y#iohb-Gbvc_0<9PCqJJA!NQco(rw-RUsC%QC4j6
zH^h)MEzT=DI);XfZpZtNczAh-W_QY+Xc9#mh27k8Rr;bC5=C=Jx_iy#8;G8{u1~aH
zXQNp95yzq7gw(c&kbd{JyLF4~U~k-r%YHonPKaKOUHY~#Vt0CS)C)Crk{<cFI+8;>
zbnsC_#u2KyzJ&#QN!gp%ui?+>MC=^syB)7zzfO<X!R_{%?*I7GjWKItV)EjkNT<x-
zW;8E$d%q|4f|!Pe1`h{^vd0YEirD${=Wp^PI>Vnbh`Y=}`R;P0@BY2v`)6T&eZ7`L
zBO@=8LsBv=Rl<#@+mZe}J$?U``<0>N-}#mV6ai0Mh4&XVGBP4TRu&gKy#M?@pzka9
zH&WK_H|7z<vNHB?8+ws^bf;4;;S?WU>)5KDfj!Jw12u7LM42huaiPn>%`GG%f)IAk
z?jRy;Gp2Sl1I-I=uXoIHP2BBgr1;}CW<EYLctG9Ljh?ZosRdpq&3{hYXIlfvsj2ay
z0go2xokw~jvv-96O&~~XYb)nLess3S&N64ZKJ>eo^z>(d9O7wK05LHq_KlPpD*zn0
z<O=}pmHzV#@yDDASM-9Rr(zMF=RrHl<0cKLj<#`X?Ct(Igv0rFIC67ynJ-<!PvkY~
zf<@^2@w`O*401({U46_q2BVwcIyY6p)ahzYw8QE!3%5}dJ)8<TFE2^|fPlw8<FF8D
z^zmPa+F$6F-*tHWxbsticA5Q*^x@HwajQR$)pUcHVUrgDvN~D80kyI0(HCL;S_h2t
zJo<>=Rntq}>YXw-_l_%{J&j&px1E6Y3R^EUI$HY__V<I|zxEKnGSlvP*wEC{UVl+W
zObnUV@4fMBMsd(8m!v$lvFGOIbl$mK7L973C^L23npYBh_@U#o!r5~;epD8`TW5Dg
zoy9%(xvxE=4Qlc}W)&5shDzwZIs0^WysV~1`OVGr(kFY?`prHJq>Pd+M!vs;Dk}x2
z8a>ZLM^grDLn~qz2`H<z_;^U>4>K>XhKB?X4+#$s4|LFb$)?ci3E<ekxyu2#z$x>J
z&TIVbUBa{MYO-)J0XYXjks=T_guc4G2Sq7IDS}GW`<Eb8STg8u@IUhlsqUWB!tH8H
zqpb>Q43ROT(Lfw%qaHli_(lpH;)+z|Ra!wSztvT%-Sx?{MD4Q`L(&2dsnp}(S}CDJ
zb?ezT;*gS(p1?tWpiA|PR>&X9AfSD;owf}Pk&-e=cch6qW7oSH8?$r)ZUSh%&>DcJ
z1LcNYBcre^)UrL8nBQgfRwOk)rjgIl#rxH^VF0iJ!DZ=ISirWxd##Zf_48|ej&XZ%
zPft(F`#6q~T1V5?E>$=ylHuXueD>3lg}N2EvRh2QZ{NOcHJtS{UfDt#Jt~nsB-0Ax
zm+kUp&LEqN<zu#c2yvO~kNfWK?pXk>(5>;#^+z~>PJ$=I0G$<;l^4?8rsbecm6`WZ
zt$(Q@m6MYr3sanE_LV8tuk(k)Yy!7&I3K|a_k=DHJRSfFgn*h9OAH%bs5++0O!2No
zQa^>O&%T{=ZS~_DQXYeMfuE~vu#iP3F_b4j70iJ`WHnLB0&R?K!W|eJ{suf2tC^<8
ztB0w8C<ilTjIsk{kO!hs9|ei;elxeQkY|mR<&19|7*Jm-J2~-UyLgd>nHh<Sin^9A
z^W8~M^I-7Sb^Nk}LYxg2n>ktv;|~*IJ~`K(we=<PIv#Es7w8*>4v_&=_y~7t3NXR_
zX!{~`%B|lAGdcD%O;k)F#Yye@Rs<6_^c7fVKfJB!6o2{`@$2~bNK2m{ntJUFh#-E$
z!?)IFnwcK2jh1cqaED;Za+xT)I@nE(d~Rok^Ue7TYCkCoA_0CNNT$~|0n~weul<#j
zlrfm|U%399dia9{ZH}cqJt|(Ct$00ygV(C8KS5KJ$?`n-ImVc9_Z9`p-sCwer>68H
z86>C_LP(T=!HK)9`W+q~-k>|dIEPT6WCXMR@pmeKZ2=oi$0UFh=Guawt;qp`E0Pi{
zczN*w{8FHnD1G-F?Emf;m+!<RCgMSDUH@8texqS03i{01O9&f$mP*j``1l}#v+T^k
z29e=cs^{0I8|k2?)NJ?I;$6Wwi+)^Nt?Lh3lt;p)L&-2+>Wvo#HiYAAZG{_yKC30h
zZpVpQ#|!A;1L`sg$MN`JL%H7dN1>Eps-l2j?gvxqE^7ukxU3|e3m34U(6qy4p%m7z
z@d=BxbcVkS!zHAUjXHSbD5n6AC)4>3q69mgc7A?-xZGToUfc!vBZ*G4j}!@$G!Zl&
zz;~}lC`u{IxI=P}ANKTf`=#pL#m0Uj$7a&+RJ6Egw&bg<q=eSQTF-;xZE0vv75a&A
zwfuMq=37gNMkdn@9x?CD@QH{NwY1L3%gYx^RXRPjK#CT-_RI=(C<2P;7E0|8glxZ3
zhfu1TI4nFVKjoln9Us>|x)7_<!WY@=jou_2nz_~0&~Zag;#A>Bn$k7E_@EC5XJ#@1
zue|-{CPSf4876Y8#?And1=P{X7r9+6Zm@#*!A?PoBeW>k2W=rtKA}*Omlh=4eg@Xp
zOZfcW(-yQI#UC6T1ehn|<m`;H1GGOeGBPGXpILOO+x7r7+WFy?D!=>YgV0Ws(L&v`
z(7s2H45gmz;sZSLgEx|tk{YRZvzK(=ls$5{wdFSS{rwnzx3<21;hD&E%b(K&nWpA5
zu!`(A)YUsZHyT4Uq>r$#CKI8%8Fs?uLjeTG?;hO$Ro9JL%zQzkC7*mjrg}Ghem%0h
z0cfOU4_Yz3PLEbpR8(33VQo!XW}>PD;zvv;JPSBM)L|C+Io%imP?r;{Q8keV0|ChL
z6kVf7kKYYqAgr<Ybi%eRfQ(Dw9!~(a94^LbXbnP#j)3=g@aOdR;X<?w868~&bd-5O
z&fOc$f1<wDyO{z-KHTbNx|?y(dEO`s`ZT&%fyM_s{P=nfcmi#=u9j9Hz{j>vpD3Y|
zwgUK?15pMAldw{7o8G^6<w8gp900(j$@nrZbVuX0Cs{5hK)ZVjvPL%AK47=QVs4>q
z36?NMg>8)Sex1B(3sqM+eU#$N2ao<(g&MNa(Q(_p)-@^~{ZQWK6AE&2a>UQi51DH6
zrh~fI2~T2cb523fY8XM`I(4uds3b3j_V)L25m*f%l-l2gT+*=+DE-AJ8}A-}Hv!Qm
ziO<--SIzA9ZE|7z>BuL0YZ%C{gN<mo_XW5W6vM}{t8eaqv0nfJ2^27E+rz$L3l|;%
z7J`=A!3?Pg*d^H@6`6O@P*UO~n3s7UyPDssG3_RSkG%k(7JyXDjaTj)jR!9ltfp$q
zPDPU8?yp1Lmi}|%j6!YhzWJ(g!}h0Gcx3eSwm-7tRnbOs{orG7|4DvaQ@`?u-+~d5
zz)SIfdQb~k5J9(nAp`o3f{6($@?bRYHh>IesLOnS%}<V2GI5QRXT)x%i9X%jbU?Q)
z$PFms`0(LF48StUuE|MV8W#*dU`D_j3(9>*y1VgE8AHrXM@J|53jqNEZTAu2gIU;r
z0E6x}ozKX~=ouPPhR=f?&u=|?$>qmqN}wnmK#lM2YeV5AH*|8kisHZ^!t?&-x{<U(
zX8?q~o*lwIhuqh(uy~r)oqVHg$nHz^qrZ@7s=?!|qoX6@2c44T@?`?pxk`Y@afzt?
z03gx1Fc_==tihP}+>jEq86%SUy?YkA<;$rz+B!OUk|6;B0li{0!6WWWedWQ81h0V3
z<iJXBZYTkoLwOhx%~fpKt;ktG2P4G>5{gPn+0gBK55nQ$m!6*NXb9R*(*tl0)Zy+T
zl{xWakZ_gdrqc7YTB5yzp_c^pPA*y?9v+ePXx>w(MW#T4#Jd)PVU<i~Tb`n212$3c
z?yigblP7Jht#TK)+J>F=>ztI52QE@j{!w=VZrEy=^TdJb{P{PJC#wNpg@%XAgNQM=
zy!?7E?TTKtBCH|5+fV(#pdc1#5!u<<{?A}@aMr(k$%>L^(EG3v^zQLU=w)Gj3nk0L
zpJK`q65{XrE6(RR)9B3q4E2-cjd=EOYrYe$TWk~r=vDdqW9m0~(ZD}RV$SSi#Rh~j
zGBTC6lN3-4@JJa0OG<d5O=Fzfn2ZPuYo)Vqc>4BkL0fx!PCuLx6!~4^<ZJ_N=vju8
z=luF)1;~^U4@x%Rl}_UmArKb=;3O{qvOk1VE66ATV9RT(^Ze@9dXWTfT};pcPk@`C
zTu%CE@$5ND%DFH0%}M}fZslAfM$srR5GFu0EGSONEl0ImTvEnJI7S!<sso}Y960qo
zM%_+1C<N>j{Hv7YxjRXQ-T~YR2qpsJUndYJK+`1Hcz8L`%TbjU6B83bV6UUs2rUPt
z{7{$?AR+Zm)j#`qn>)4^x^EZUy8=%_a{;|mC}}$Nu2*0S2B9!)W4e*^qKTT2%@{ZO
z_+W)P%FX*q0o}oNzq?Qd?E(e6*5k#`LPPNd?WU+ryCUTP%1lj7&91LULU+kBYW6`Z
zLsv9YG>Cz70i@Dowhe(~n#l}ePS`-oT0jZ~=`%P0kJRMjYxQ#{1d$-RSL`#}-Ugls
z>zAwfF!@!>e?wz`pl#BJHNCoe_L8HUVo8%6zTL9rhl%T9Xd#33;MDD^mi>#xwAeuJ
zO_K5y9xE{-MNm~@e`6-PrA3x2y8pzblYSX0>phC8eo{Gg?<->j08{YG?xJ-C1THvW
zm<YOK?iK1h<-Ge=$HR>BJhUe^3jRFp=9H=s6l3TNz!BhcUDq))H*W>K5`ah>G&aj0
z#&_M%A<xegxe$s#!>hi)%S$R?(cc2<BgWsF%zEJh`tZ4;3!SpqP-GLRqa`no`MLU0
z1T-gqV`{Ho-`<FNY_s9`Qkgkp%mVKn0ign9)z*OlVvt;+Q{-)LfxHF38=ncdDlcsT
zWKk5*gN}phbzUc1-2gihd@iH+b-Mx@9w@xC0FLtv8^z%-gU$OATBkFpE?#-RJpBmv
zX`y}{1u?B40vM~x=KPa`pU-T@i&-w^)q3^TIxdu&y{A}mhBuG@a|QTljxy=tuV3?^
z>8sql*%eBDVRm-b4>S=%@q0J2*t@R2(gW1}LK!pw)S>V>oWnz80nqQSPv%raOY$?N
z5+a`a>-kQ{J3}(vaC_uX8HkYd^z<s92WUw`n;FN=G_)|^ue5slN;T00{#K{bDwdf>
z2@u)?YTW$%{7vX-?YEYB=0X-k&rwma3JFoN#)e*<!e@zZllgNJ0LwEE<OV2)PPrKY
z+N=Jmu|*%s>yT1GV0W<fSfQD~6YV_)jjG_J0F{0Y*DHozRQLh)Q+9CR^YZd~(d`CB
z{w&AM4p^Q_Do)|R^mK-k-@msGw@hzbWxnR<>U!<R=c@S5N;t!;^P(hZAKv};@iquE
zplG4Z93>xBA@{ybnj!8BUBqaelaXNpwkhheN(QPnCPIDH41aAjKY&)qdKP*Kx^Yo7
z1N0D)y>TP_#wMPuhs4E;O5cSD&zu3tC#xYnIXT}s65az?83u5*$S2Qox63Ok0va2o
z2I#d}x%+MpXlQ-|9|qOLq1s_1yQ?Eua5fdIthG-2sV2;jiM-nvI=Z_RIOFHLnS46{
zJUFh6UIwa-4RuYM5%d!xD43wXp&}wwY=K`X|N0%oPom%{s2iP?clj`}r#v5ONItq3
zqNR^2r!7yhP#N3P)3fk?)fv#H-rOanGl^>sgQJuMMb+x-J0bLB!OC;#wddJ<slI}s
zEfzt5K1h1*t1VG?J~E(FUl~j%1X2COV|O){j5ml(Gwf?^M&wtC;yi3$ggJiM^dO3>
z)=_Hm#)S(VlAsho4cP3WvuB1bib7UEF8t2RYG+Pu)#?r(m2wx9OX}T+u86b@l^pVZ
z`#;zgmlI=!z$$rLz8g6oLM9*C`5(ZmrCdwhPmAc`!-r^fw-`(dgXXmReVi0PX#4k)
zX1gER<ixgi1nwY-qrw|n919D%6|Y@m1@O}y#ejkoIQpoN*w)uaI9_4|W{1G$n4Wz$
z@L9wUm~0kQN)({nD=~cb;>EfCOc^G?UGgC07M-lYwpRM^yYi}l?^-VJPOOW}v-|ST
zoi+fYc0mI|H=|b3eM&gBsFCroq=LAj4%vjRPt9+F{X8y?B~}HMJLx4{iP9z9P~*vA
zzJtJP?~``V`}cB1_p63WjAY;wjgGwFn4J9jUXmp0$O>u{GgSn?)1pfJyQd(9+a)Kb
zXkUqxDrTM45!?Qm@l-8^$TN%yG!DEAFTF>XX!^=Qgrt-{UKi+DauT$iAVXz?zpCT4
z-BP%q`i-WKzjWZ^9AKk0uH@Z$L*n!2w;M|c5HVaw=Y^X2mV?@Do_|QUANNApwzMay
zuREy$-gpc?S5zeRcUV^R8AUP4kRa6JE|&nY5`pT?6&<;Gn`a-mtBS{2vQ?b0j@v8z
z6mH7E)^x2D$KoD~(Y~J{x<zn`$R=!I!%JCE)w!NUQggAh<D&ftdg2`PE_j}d^==!l
zXG`d_5g?xZvTP>9T&yZ9{EfeG?;bw!qvujyo2EeoFV>%%Nu(7!OV3raS{v$Uc9jNX
zTt6_KE?m5b17>I^lr6p|yH=k!v<Gh?D)=lGXt!xsuHg|tn^(u?ND-G3Lksu@D7>9#
zR`m^p*WAMFCTRf2m?b6Y0Se-DJ5r0-(evonW>vJ|mKg}w;K4O<KA*Ml!M9%1dnoa=
zC-=mfVOuInr`x{ycQ80ttXHlOL-Ribw9aoe%xps%k>rOd3zSqKHZ;HTmErou6AQ5;
z#y6zqw1fq~PBMLdT?Q8ffc&SolrLYtyw&XE4G%L3%o*rS77L?!QtMJDEPd{n1IGx>
zgy*mJ{M(*!B}&O%ZBGE0g76*)DkOruzL_2lOzw{Q*E9a;GT`&mT{FYm$o_t=W|RpB
zaVOQa+2&?|5B>fo(?=eB7Kgbu9F!mqXDgtT8^@OtAIA^+Re!G#8>UYjY0GrZ>)zk{
z8K;}6UM-80V!eV?HvwaSwiOA;7-|+PH#gDN))vS>(N!VWu&=4M0KSkb<+iMN&tA+z
z?(p!_7w+R7qW6NXAfYUNzpDlE2`?@QeJGfkNDF|tGM8_fqN1WYZVx)RQK=&4%<RAT
z;)AaatD|B`D0Av%k@nE_ehR<augkve|NI((${66wf`4&17s51Lq$h$tVRBT(P!V(c
zhrhn0IKlAKN{NP21#cM6aHu2xv6l7y`}eI${HA=i6MQI22kZs};LU6aMHafxdifKU
z9N2y(44OvKzr#uRR4G$za?6Z1QT^&|&<u4|W7HPH7Xd@KgHDR-N&&{y-1jT-p-swy
zHV6`k$zS0{MJ&NZrl`L^Q~X<>7u#2jis4Z1A<g72O4vygP2Liyp!221z8wTaetv#b
zD_mb+m3sTn?K^P`RLBl5^l~{{(rTXr8x=iY02)FX8YEr5j}gPec0c}61Nx7N@e6F&
z9ZN~6bj|N<y`s5>`ja2(lcSzF%IJZFKd`h60Z6cr6D{qpqUbCaPmV6GW)u#nC#;U6
zbwu0%;AFEBag@!amU{(xhM<6>es;9<FZ`gOpge<mVL-Ma;Oy&qT6Kh9TIUFiaZ>w_
zvpA_HC2OJ!&>r>UM)U972Av0*$O5Q==T68`lRLM8&~7<oE=Ny+)lsX?2si?~{-VYJ
ze+G=pYO?$y$UKqIlpcUkigse)bm+ZdSrH_`A@lp^kl9qE%Z4tMaz_9ho&_r-NR3^g
z@yQ+;C1nsS2N0!>;J?S4GEk)>%kZ}~_>E;RL1sl&SKpH@JmkS}mfXa<$6Nr83nj0b
z*!&g6Oemt_?<^M)i#}dE7feLs4|aYV_-!Em3V{fU6`Za6K$Gh4afnQS(v|ON31v&i
zBNLS5hLGZ%&z@AQiG5|ZlUGp>io5S6I1?zVa&RbxqhL#Y5tSGDZ=lAZuYl~uLo7tM
z>nyxPz!WPIEbm?T_VLZ?RGk3TmHS{f^Fx^ezb?#Q<|V*fID$l7Wlp(2(=lZ4M9<A1
zYhJMQH(jD)&^C8kpWv^w8o~URF9Nzkx1z#dzL-n@cOQqgLR+{ju=6H7fcrN&ye^wa
z;*5Y^*`F?P9$Z%iYirIF5&J0MbPpy<jmH>iFDqM-G6M*f^Uz3(xgLij6}7no5(W{i
zAU=RFJj4qQ+95<#Kmp5JNe;|=i5?g3LRPk$6RL-liOZh!BO5E3zNpU;8ykxO28-9h
zS2xtQtO`-;y?h^&J&=s>?WM4@H=liOpKfRJtERX>v@T`S3Uy4Cq(Gh%+(&?GsKaY!
zW|jq|8<lr(d^^}sB{VchE4PE-zg055;)JS4Be2Vdfl?P5H8Uc7z?DwmS%A`4xeew&
z2(oZXDoUzp4#Z)8GmBWI2YLBU&5dtlPS}|)iS$iYb8>Q?fj-y<HFNIUw=fXsacss4
z<>EQ-!eQ3L7-od>TNV?ZBnpBvM&iB3zQ^(rMeqCi+TmJJkq8r+_S#WLb@bEI(|e3G
z`6}qf!VfnWgJZ8?VSzC<^_G=+n2?pc>Ln(p-8}Y%@&M<Bs4>QCK`}8gei&G|D6;@4
zjWz&~lkhz~f4yUGd#Q1bMUhXWm}5a@_lnA^quygS3gz8!e0c`}WB-(GeQBjw(f?=$
zcjg1hZNE7iuj{}`4FUmytgfzZb8D*=7&Gca5k@9yT|=KEw&OpwQFHlUVXqY$pgIX=
zpXbpzrFq3|5v*Hjmu`#W!^1Y6{IJrNRE9dUAY6CAF2g{;51Od3B!zIxm8%w-Vz)uX
z?l9x-n}dr?@1_@XA>=N2I2Mr^t|=Wv!4YWQy8=75M?F-@VoG|Ey(1;f`%%9RsE16I
zc3)|sy$7qXwE?Adfc<>tG<0%=h#B4vlqRhNZ=u+Kl%~-Fqk-9-@4T69%tB7=`uF#E
z&R1Bz5M?2{BYj?eh?Z8IDr%wh@%MXbnbHhEHS=^U*wKQu|Ks6NrVj1^+_G9ivM%$g
z;~MAvqjOnoj%`<5_|oXyDsOCZ>=GrtqB*YP>$WXtG4%km0F`^f&T_njTnebFvVc-y
z|Gh+&Z<DJcHE;bH-xRYZXp{--eHO9gHkCfVoT9!gk>FdBq10jdCnPtQ4cgcPkj$aK
z;8F))T?#Y*=aw*@8pf!&h~y;JO2A&|%iGZ&-M&1gd3*m3?M@xvxb4-MGPRUOeDGoM
z@$p+`eE&QH2U~MfxzHcFf9QKfR@_Ebe;3A&1@<u`dr1C9p#!I;g!n%fFeXA3g_W`F
z5bDXP0OG0R-rC)U(_i0l4FbqmZUf*US9B1NA!7Rau3s&y<Lk&G1>dOKJ<+ayzsWU2
zxqsh6h$;a~)ZuVGMXG7CmH^}(5R>cWRyY)`Dgo$l<%co|_@^GR<mH#>8YbGPVDlZx
zPI}t_Bw#GBo?6$d5LU2kvGp#=%FZNordk*iPI#eDBEfWQe5BEgiW<0?iDb^;^9VHY
zY2zgPS4`&&ln5Kr5G}rh)uh|ECpEi6wl9gWP(iNZyVRhsE779Zx03LC1-|O%@OG=U
zYKCKmf=oyk3oH|!C?<`^pX?GOM16l2LZBU!LMXT<iJdBJ9~2Q0F|xu&nJQ#LlB$OZ
zqQ3wzKV0}6VC^gbQ^3x>E1i*<iE1IOfc}BDt&W#4!6A9=6cs|s^b&r-0eB>EtSwNQ
z71-l+f4w4m8T-;S_{mpErS3?&@yl^{&+A>v)XdoWDzY+>Wl;B-%c^IkEv>Uw0~F4L
zqWh*h8`=)r$pY-fQ%?<K*0X1$C+Oi~_74ua!GA^L9S-y%V9jMiWC5*kkf-27(|-UI
z(exikE-17DK_^YhlL|q7=l>he*E{+Y+Z$!y9hvpIX~{<{X@b4!$LjVUiEhu5EGRL{
zbfwAo29bK~amFCW9iHqAwK~$L2tNu3XpjYs3Nt#5%=ksb#RuT9{FTc<Lu_blM1w&9
z?644EPyv8C7SKE&F>xnc#umt*bI`J&uwWpzljS&&m$?Rw1hO)1-%~2F)%sbsrU}7B
zd2>)p${s}fi9dj>N$}^7ox)dBB8fCLq?h%Azez=J1WGv&GT07ge{SfGW-{%GVG(tH
zE$V1-a=3N4R!}mFngy}3+ikS`Ap-+Q0()Yg<L0c~;^HDREFlE&<Uk+?dp}ol8FXp|
zl=EYtx+*z6eId~)k_Frg<xHUZ2Z1klf8iAwX3GhKK&Z#v+}|O;EL!sRS!J4k@6%n2
zq7MT%y7t|>d5{A;v*ZE_rMTsH%iklLaNGEVgbDz2LEA>Xd6bI-?8?)8bt8G#gNca=
zjC};r)YR1Sm+4#x@2-UGjI5m8`$Rhni`TLx!7&ueKa=G8mGf+M^v<wS9xn^>TUOgD
zp(YrdhdB`cuo1KdqOKAcWJQu!<qyl=Bc))c0vZj`;_Qr;ISmCGF%OW+qH#H#lB?8x
z%cR0EfJ~{IXb(H)gW5NE{~R_OEg>>ZmqBEx14A`Ii(hDXv0l5hVbCvy73g;WE^!;g
z!!Qs~_S!*Gy;a5ed+u5<X=)!PuHXNz0+_;;Z0_wTfsqZlf_$eFSl-$1-w6@;;gvYb
z!rCIyYjL0#%5%nZ`a;@=Z9Re&dwC;CE|5cOg-Es9sP7&9vs~S$?hTyuu+no|Zyg{n
zC3pxPO~REDpOY<RRP#O0YltR^NEpNeAtDy;FQfSrB#^u)bKP_2&SimO;NajOtELvY
zT?P$e9t1U52}ni;1<k>#xSyWXZ6|>$(heG&dWzu4lV@;WoT@{Ye8B8;cI{_wJNweD
z@Zy&e0tTyh!HUeYeaD-b3|KP+p0hF%xEMARTn3wOF3IhKo&Ud4m(bh)FX13Sk@YWc
z04)QR*x~c<yjGLb(V>=CP{2S8-npRW44>DbL!n_K?d8jtlVp6Q!3tOO%_dJZ9*m-i
zkFVPwBN&kwi(&ro;yLFpEOp%cSb;fkp<DY}m&!&XZvFKMsQHjTCp)Cb12i86f*C#R
zYS<l-mLNewXoBCQ1D}n9LxD9G4G)OAu2X}_L4@ek*fA)wl7nOw0%AMrhWs@sX9fpJ
znooamp&~X2Eno#k(u+~3CUBRHsj@0w04D12%FqSJcv;PdZ)IAXrQA3OpY^CzD@!*S
zco`m=v$%Pa1d75IYF&e%HpWQB!j38$sDN-(o0D@v*lsEm@-fy@Ks?0{0eS^7(y}Px
z8OTv16*yM%7p<%=C<+(8y75+rbGVSm$;pY&y!YJtL>U%>8d;`88B!2O3yxR57RuM#
z31BPq(Z|M(b#FG`ohIGqOx$HPH)Dy9*Ajl^AiI|B+mD(ptglnP{mK!UDq{Z<Fe1cK
z@ZiQFqZJ9e;?AZr=Q)&fu+N<heCWZBe8d!Fp%6Ue6Qa8p%KV&E6(@y~5*S$ZBRZ6U
z1qNb6mWBnC2SO@tzjT8oE;j{EVJOz@fiHM}1@IZDeTlOeyr{7!6~R;VBz06`3@CAu
zm#o6Ekgxf%(4E^qq!%EzFb}GF3zSR@1b{U>s1pPtg}~j(c2Fp8Lcy{hN8?H2e8nS$
zD__y*ScpZ3b-2)~*sw7i&>u5MdXN~^Y4Q@2{<V4`RIyU!8Yj=+dmuNFNiJLAQ?9@2
z%n`5iGB^Ap=0LdLf+(8NIRujwrCLC<Orh)dzAF#ZGTU7tX6a!2b&&qe<a35#_Q9BL
z`GbnO%b4ml!>y2Ci1n8d`u6P`O4KSJene}LwD&LOzlOZ#LzF!66J#>0jhnvydHrm;
zLGr8TH$oq$gH4JW4{#1|zq&zbJ5h=;GBUEcxmmhDj9*5LqWYvjb1_LbU&4nQzwsw0
z?Pt>+&#;yvH}=ypXx9+hQVw+^@`u?8lKD}Fon7VJpm791351+#4#WXavIMp+&^>wX
zC=~0A*y(F@#jfhb+}OE&;>+=IZdnG);Qr#vhcPub$SiFAx<ptMLK`1zDVNM^x+IIk
z{{NrEZqJ(|{x4U{nfgbcd{HMO5YirPnLayOO75?U<KBgBwS2SZmSt2fg^71rSsAMK
zf+mN0`X$Y$wBVj#wBOozr0Id;CVuKZCs7J-)fIkalCSc&?ihtCc6q;F)985+Lo0^G
z%0aNGJB0JzrK(f>33ngMHy>I<0IYU5F*@eo#{2{GRu8IKxf%^*7uN!+@HC9KKA58k
zB1qqZBtBgE@Qw{>99cmQx<yJM)=-k*UG=ZKielb8n@!ao_ec&Sm)p}9w2GviKaG5>
zNxYrDMp4fh)<F(<kZdY4A%{}%(fQSwc9yA3g`%5(Lcl>sZ9~VvzzGs`58?LpbCd0g
zsz+Y0B(r{4>!Nx6cGuaS3QgmUHDZY}4$*sdqeNNRov*5{l)<RL(M03Inc!$Ubvfor
z`}3QQ7Mputh^*Dt<yx6oUfrUhmJe+OHgo1l8hh8p`}o=g(d3hI8F5jS0Xv%}Gu*je
z5}5xh6iesAqvY^XIwv1r=hINPWthQSPqy9zT?45^E@^`=IEq9qewRh|=m^UGo>$y|
zJ;WXAPH7=i8lwmAYVXZNC`81hlG*GgHf77Dj*U>?)=wFtsP{L?Z<L(Z5pz0g5=&Dv
zlsiND$)2sl0mzMsg}nFsNs8D|IZ%w!yLZ><%ru|-|HS<G%lxE#YL!NUMZ%x@Y#Q}D
z7rS1~`YJPnsB0ezWhD(Mqh4HTk32QNXbDO(@?cPgtGX-)#L<FnF)^)(;2-fKF`3{+
zt;CsXUUj`hQiT^8WeaqXgkm{nYK{*T5(m|Ft;aj2d@J#V>hFW!0D%PK!E#;?293Y(
zb)7jdDeRTagsuWWacrSI5Re7UET~!rir&sHT_3IYny5QsUixx7rrGPvfPl+dxN4W=
z7cq={&o|9)x3tytD1u`c<RUNg6zQ0ZHxu>C&Hps}pjrQJ-i&hZAfLS<#OGqqO1zOr
zW-@4PdY0t_z=-)AW@S+sh!wYos;(tyV8qGr?TkvOcnStv2T`EDeC1<S@_;9M1=;2W
zDX5vIYyI(ZxJrExNpS6Zi9G3buFeaRLoH7SZr&lVMt%s~ly}7*X-<rN+Gu$tem2)E
zEFDYq-p)r+l3BmOR^8r%snO*6d71mu<J;1WygmI@WEBM{v?;h(<N$oc-NR$9hqw7<
zYlwuQlyyeN$3QJXbiHsZFMn+a)S=~hEk?~t^4TT5-gfx)o=UnWO%)!8dg?vp^R@v=
zs^jj4{0EqGVq9~NEyQ8RZo6Vd#V<?_5{O58R6F~>_pMC2TAu>(5s;2eA-&zx+ba)Q
zR&pjLn4EcH^;-=UOG_~OtIx6uLcRKX`)n!khO~o?zt4a0J(}tfh{s+CzxQ}j+ZR_a
z@6pkxO`40Q;i@ZNr2;<FG`BJ-$C}Jw*F|z{D^iG-Ylhaj9QN8NU{h@qG5_vy%4IMl
zx>HTJI4y=Ftmh`opPC*#CGm#YZkwcJyeO=!qM*iprVFeX$DL($KB52(ncwF?rwwfM
zem|b{%N2v=T|1fPZKyY)H28IaDwTJ*i6w-rqx+lPzOBv`WU^9^SG(PjTbw7zysbx+
zar-Kp=p1Z)yY6#Fq}<j$c59MuU0cNWyZU@J3=HMDOU7dp$!s(cE30O6pvB-fs-*}L
z{3`{3Do4)9SiE)8wVTE>Fl=`v@P+o@RS5QZPyTWB=UiWOV~57j^nmublY_r%!SuE0
zJ4OyK^G!pgZKIaW=S|ElMTM9;H>4Ik(}aU}^#*%;|0bCrcYz>)#{=O-OLup7F29wj
zRSlJVufS(4xGn1XjR@+Y7v4`@i>Do|M>D#o)nzlbCTu-Ls^XPLg2+9$#c$5IVy%pe
zZ;O-8EV#^781YI=GeS-URT>~qya^GK1?ck@#HWFm)D=`b<OKZSJn<hlg+3fj)FsSh
z+x$$oN-K;V*)#aoQ{<>$+IoJ(gYyF^?+4GcrXRS2&m`p8_Ayg)yqU6kI{R{$K7FXX
z5*9#iyt53GFs8U?&YYRu*ogK%-k%!~nK`%kE(QN?N^s0}AoQdDP~lg4hTqJ`rUG>g
zF|t(q43Wc)_uMz?EO!qp=LC**QhGWGrE{uVUfs1f%lM4$P0*+^NAzM&29EDXZ_!jK
z4G$IndVUBIMSd`;Pj)|QAbx0UAB}z`COSMO5#ndFG0H=d$ZT>T%-3J&LM*Y;#c=ts
zf?4`^plYJ?+!tTbADbqviP*%+cXwACkv(Cyb|E(T;UBo2sug&~E5+{wZC2(;o8RLN
zh}(EWMi)idXg~#^BN`TfX@ZXRRwh~9<6NhgG|*$87n#dC71Gza-DKfqp`sTRZ5|uR
zV*VTxi1lfqYRLaGej}Te$~E_~WQM#k&!+9_m@mZKQZ@sQ^dPi`fbRpe7|ms(meR-9
z&vc955Sagn>kXX~9jts&e1=m%;4DQLHcVDQ8heb9gvG+rh9(9C&n?@Pz^hf?6is*!
zBZUizBD%(xbyaN~cIP)W4UgAOEW7cm*?P*HfHEvL5*2>?Tus&TeP_Ot5Y6tOd=1I+
z{d0<n=WG;VV=GAuLJ18A?-EC_#@IgZ9$)X}wdtT2#8Sa8t83~EPcTa<6?@njy`}$r
zQAcdOBO`DB)TyUj(|Bc#7%w_S94mygY0u-?ot}$w3KmRN246;L-4k4v`)P<6C8;$s
zB<0YljZfM}a{%p-b}J*F^J~%P)a7zn1Udt?AVNf;`Ly)Eg%Q8~7_fOyq#|7*(z%da
ze4sycFgPs1Zh#R`yrOG!eHf5qkT(^Jx?iE8-bK)iotAE(nHtcHQO*Z22zKLRX^lJ?
zO&4$WlS?+4TCC{D$e~!~!blBsd_pB(oC^I27b|}2KqVewLa*PVgy$W}42~y5Z}Pg|
zJW_j&O*F>6_+!$B(Xh)d;N(^HLCdyMN6>v&br{mxoNIRkqY-I=7iAAWve(wut~NH2
z`R1-3s~xIK`}y&1Bo>&v$VV#awghS+AUvYBE@vwf@-Hl~(;sirdKgBeq7c4L>%{P=
zdySMjJWf>fUh&oueC$Bz`YrDX?UG{C->mbn*s+3KWGEOg@8d%oEU-dcOSw!@N6dj#
zk#3OK(0mFQpz{!g9EPxK@bB1{d-xfd1jvgj4Ne?@E|*ty650U>ibsn%ty1aSb8bqk
zOOzlc?jov-sx}S6r`~;B0PT*hj-ae3@=FMu&!W~`{zI4E;b7Ax|7WV9KdB+`hsMmo
z3x}CRXGrEOKw^<v%!zHo5r0cLoGg<-o|6JPLT;MsyUcqQ%Npss=F-o9zR3prz?@Ow
z!0Ao4YJcZKe!d&c#Gt})Qeeub(R&mZkJpCGuet^eCJDc$C(JZ=ZR)F&=dPFbIiZFm
zLl=m5#K_@n0BNb+S8-7>u`M4a#Q0QY?l`#z+Rc?e&u>6vO1`(whTjrn_0RL(e(quS
z=Ip9%B6BOgdy*L?yB6*D#$$U}77JfLu_(GI@S6Dj+L8bd(okWg{*?~=;AWG>o>-WF
z2`MPxs($n(7{KKAlPhb1hZll2EMK~;P}?NPv35*>c>+ew6Moupd&?d1F@PFl-Icsz
zV}aV4L9=E!*LXQ-&qiJr*9_I~b$v=!=l<=-jqK(eyqpghfxLGPs8R^Kd$+uXTvlnv
zE7H#r-5(TkrK$p(vch6O9ynWUMn)7=1JvcYhUe?Y=Xv$W)#LI&%qu(jUOK=SxoLpV
ziZ{LldX2a=c87P8;Ny>e)l~I<U&gn@m|ONm>&A0xfeR+<>(5LiyIe5&#|vK<?MJH1
zl<O;x@gWZ8XRH%_)IK^D^FKX?xmDCaMF*J>bmSI|$lNWk{<9Fkd#65q^;FI3f-Mc3
zye$D(CNJ*5xCpa^9`35fOT|7i-^M-U>GO--g?%F`G7Z%7+|DT!k^HO@p{Wg6j&W}8
z1_gpFUve%_m<&z8uQaUTuH6%M5E69w1KJ<IML)Hfg$1P0uq0g9Wg&~DPM^qm=ga8B
zW3^R_^sEdrXMpZ8PB+-qgU=?mexJyCE!{9=6V*1njNf%=sNK_4T|FjHz36r^^21ZE
znU=<^QhWi+o2?-BV<4bewL`k~R=w+^)~>rL`Co62%ACc~u)DC3xb!5`@;$t}Dsaaa
zHeD`Fd6(8N`fSF34NmfYO5b(%fuQZTC7s&3iEjp{R|fR^(&X_>y$B{52U2ie3XS}!
zzZ&^u_ARf2?;363)9ma+aAXj~dAYy!`9)QT2Qg11^=I!|vWl=IXd^m?5+!6WiE;dL
zr><7-1XblF&g6ZbRfmw;ht=6nV9RK2<vkr<{}D<)ge!Va2!Bz>&v>}rmoQ+^X3%_a
zd$i!*S!YW%y55+Xr_cp$UB!ur^{a4xdhTB{neP~bJj2tP8WGox>1Pnh#Uv|s4!t;W
z|CUDRE1;Ua7nN2BABPW~2P3{ZpZQj!wBB=tOY{2d%y^8wrbNVsl&zubsTlY}x<R5V
zWj)3>)=t9e1YLal7+SAbuZgPaR#*2wv?;6tcs?`p)G|{kZ;i>!W%j<f!w}fJf1c4q
zKyd+U#^?AE8ie=p=`@}DvORR~p9|ol@7}r|^w#C7fB@t%VBQfTkQP7YVl<)TgkR(r
zk6==rV&Ffha7fi5xmPljut*k8y_GhPmt4LeAn1PAYmUl7QYqZ~>>@uSgmPsoTF<h0
zeBI>LkBV>RbSm_`#`Zwn<khKWWsok<?`E!sUj|?CE|h<yPxBa?XQ-WqAze^l73JkI
zQQ-mt#3&|ob?q$C)4L+C#QdrxB1|ax*EE^Q8R%%j=aFK_lmq9kbzUJ!OH2D}bwQp7
z+-6P2k7cIHY2uG>EO{__ZDLg0PQGmT4W1o3bB%g~Fd&7BoRnPmTBb8*ayfAPs=5AC
zptWB2GQNJ%Xt>jelUp(Lxvc`p!YlqmtVlw`Ad+i0ntCwQ1z4+3l<dyv;ja;yuWs}v
zjrTaxSKH4-<==V33ehPTRFyR|V}q%MEGNUgpO@a+{vi$7K0M~24ERTamj+)Iod<y-
z5LwWnL6JuXi0x*Yh|6>eC1JAJ5j3u_?P{3p+HPKQ5<^GkA$^o4;YI>JJ(`(8L5x@3
z=|EFQ^~3L0#_NvDxlPw)I-3Zb6yF7C8a(<^N&dt>{mK>RHP*%R)x)ZZ!BGPf<0f#i
z!$UV9!YF+>i-k^eAj6HGZtI_`2q7}S**$6fOi|&*fz;bT>b7Of78t8kSlWls2%6f1
z&@7I9<6gqH4-D@^z|LB-JJa_x4D9*^i1dK=EeENIQphayXGqb(OL_1L`+w1fdVE{6
zGVcH9`l7dnopL$#@C}RNtNIdSFj9wqd?jK?%`fTHZ!SPRSIF$DsHp`)NGyTZC<>%J
zIX2EzN4p3i3IdkTlU`h+`C2g0v%9NR=L@9=#&2m+<sQUWRuK{HqltgyZX%}4=btUX
zOHhIv1-<*#mN-~oJ^lS!M=aHs!E~eu%NaCf`x%8BFU@~`PyLtI_k9}L1{C}K?V6S@
zeZuPJ1X%5|13VtjmY!2;>8dG8=8>;)6EL%-=jx7VSB25dWA<h$7GG`pZU`(Bqp~<8
zsL+g@faTyUB=yIrPcU9cDyn#xe#C&W>m+G!@oD!3V!=nBwLJdBuqc$mjC7i$2L+^Q
zJHbQbhup=|0_dYCwpjWB`Ko8*_sdLh0Mu>2f#j=&yOJV4a8pBaIq~RkpG11tDP%{F
zqdC|1pPE)lcMs1BK`80D1<pLE%el5R!QBOIfHg|n_IWAA9|)Lz)9fju{4<RiQ(P4h
z^7skin>TOJ(VryYN3;;m8?Lc8(yw<VhG?^Rre*PIUm)2|={LmTL+<rjkDb>}8)4|Y
z@d$iv0iXpmVvUANQ2n2cot@UZ=s&b?v$f4I8XS?GcCb(T`mt|JF+6&U9r@&}C%I1G
zvK>~76)`3;fnn8<u++6}kqzx0=36~}#AL4Do^MQih~oC$^ZfX!xn|M5siad2@^uQ3
zyxxKoX3mXx&O_+&TM%$)ZCCg+hO;;H?sE<j$os~mIN4sF6D#!#Rvw?nwi%iXL&Lui
zfPBmdA$C2a1z7*B5<@985(?jhz}XEEJ|d*@Cmp2!-ZB&Sg)L!=>?nYamAZlb5{^*T
zB@*D&ePb9dXoC0TJF=GXh;!d|1v}Nsgx9P*c-V6wQ2*Vc=Rjh&i0t^rz{C!1gt;zi
z6deMIVM*+}4GASQ#|}wy-1P3}?-iDcrkPo)uMH!m02IK>%+oGm+;HMrgGdPs0HKeu
zqpORRnOSzjd46F5LsmCb5DXnSTVfe9zENwR3a}6~P6l=zI%SyWNd5(qzV~?d+CqM@
zd;5kli6>7G1ZvRn2}=-vQ%-0v7_xlbgq=h1Tj|QKp#Y17+-4(tYTV#=QH3F!NQf+$
zc7&3n0y**kL;*DUp>JRi-%U-?qQ~l?(vzFsJ9YE9mYx47M*qM-CtM&(mys5TXGcIh
zSuOc04kS&`(LVRg?;z?bLY?zDJ+h4d><jD$(%^7>(VtCw9~hwe2AUQHG70MuUxfjc
z3{A{Jiu3_kiRhOk0PIHj-?aIIJQC)z$H|VzF);5jy`X5KF?uZis{efHMeB^@9KKSq
zGj|E^9>feyD4LLBVao1|4N5=;8NT-+9BeVblXHLoyZ-#1`SbMh-JA~;0*`{dsDcVb
z^77MjQ)_FzUuHA7;PVGT^_AO4prD;c%>)xu)45fha*hI+aRI7vbyXkK^I$DbCCEe0
zt*pE`xt;a~cAlu?JoY=6Z~o|f6N>YieSU9(Q*~*`7(ztoJS6DUaQ@p%jobOahCri;
zFWiFrh4VL9Dir5Khu_nldHqw!RF@>|z;O2SCk~Z{15EC|*8hr;DeS4vOn;ooVvZl0
zkQCQgdd!34T*(sGw*-k|a==-!X=#zrD`sKy!c>V`@~lRlt<58%-`r=a@U%v5YH|+T
z0YO(U>T7-RyvZ($g3%e4rEq4|7VvGap;KK73U^P3E?m0QGuuJrefBvFT5*2FQH%p+
zViu-|L{>iszTklDi2~dHqEM>V6Vl{ip^Zv@OXCNo2PqFzwr=k~ptAa=IWqwngW%m!
z*YqD>ke0!ArcVQ<nLohvL_=8cATf|X2flxW*1U6{@eZ*!i6EDgQrSJ4H1v|pQ&#sc
zX}Q%vLK%rat3xlfNAc_==b7~z6%`L=6(kAf@270E@87%`bwd`cc@hYNc&_A&_lCZ>
z)lYSQUe%y=D(ETkSDV*KTems>iPrY6JF<w&t<t3)3_mWxWWr7HIdAAq)mkww*=l)u
zdbmJJ#3W76YuMli-5wBtKg^h0OO8#YOGUzyE5j4~=JUBg2D8#jhM#qjZ@RY0UTT`C
z%!h)KWtq8$!U4Wh#OL^j$kHRF;Tjkf0-KG%kw88a*RXL}Zt-P-MV9psp={N>Rg<u`
z7Ox1V#QsAox5eQrrG*YfBafe{A3t}#0`w8TL^_jz)&(Xjj3FV8=57re+@C_KX%;N+
z`vEd7Cm0Q7nMy2c#M?hA<I)Yx$|xiQbyAoG{6H5`hS0YoOuqM+T}3lIs9pwz*8R#j
zg?uE0{$>TNBL)>CK6v)1{nivQt7wXuta{CvyfreI#Mu9ORwCe`b@?DV*@h;S&=H>S
z?*+xW71zw0tJ#-HW}p4$=G@n{&X8NuwvVxY!@$84aDP1ZjA{%`3*_9+f~kuR29l6;
zp*<dI5NdZrGP;T^kvY)n_SMwf`@DDVU<O<^S_`%ttGQ<U;L+jWp-wrTKxWI}s98G9
zm!m_+p%D?NKL|)J5azwE*>B<{HCKdO&Q!V+=ZcBhILkg}7|B3S$e((~X!4P@fifqz
zTyKk(q84uO3lT5OGg0vu9Pdd<4mv3_E|75WU^SJ0A(Puu%Bu?uE?d%vueGSW(6_aH
zh#PN4={a=Y|2*#4#s9={_swu)G#*mkxv>fsD%z8~U|5D2K{JYF@D)0SkMDl!zO3j$
zuO4`5#k6;Q_1PiEMB{hP7RSyori&anEzFXZl(e<d!g>UZ@;553^|SO)Y{HE_mNo=}
z!os6~KO$GQqL!E{lDSa2Gv31~RB*w<h*{8$4l|qwxg^15xo+XDE*GV;P7pnKXW;p4
zma1&ldB`iJ_q7D?KM<zM?AM0m)U%sob?_|<Xz~-jg8>sk?Jjhb5KQ&(R#PEMOH!5u
zrz5ZBDgwLPXMb-Pb&TB6T;#hb*PNuPVD-}t|5A9sK@8TplfR268>^Bl@I{-4*Y`Kg
zL);iG!O!InC<Pb|l73i8xR@pL?tvfx!b9=F$!Ue@3!|>(OUNV6qLM$9s=G1EdL0v%
z{ci5tO6MC_xPE6~q2?X>l_%&338X?~PByX7vM$&4sQBd^t);>hQ3dWs>Sy1Ng;^x@
zUZ;_l%<F!9DKHxN;tNd+qz&??cwjyWMm$lq4ZS0IeSb0&%yiw-l7wcST8&nTqF&r5
zuzgk}OjoB~CSEVvbQu|)=BiC*0<*=Snk4I~RFZnD$d1s#P41GN%hNkB6s(piObn0{
z4?(}s0|Iru(rd4A9bWfWelralb9|>_7~PV}GFC6mGRCTcOf-1xTs;U149tQ_d~|f1
zj9kuFiQGi}wOK3eJKao;m>r{g>WVz#!To8^nQ!aHOm@hs>6NXFw_E3=)KFsuS0XJe
z67{!ATjK2fIz^H!D%lP5$`cRgAWB8e_W%KGi34AF<$LM{nw9Ki^jC_aDK+y!q(V|Z
zlgx&U4H<n6Dpsw_x94KKa`<ztFSvCS`s!8*ou1RnlV8ZdT0F-q#g%v7OgOmMWT^Yk
zp|*z}x$T+QO=A`G(_F2m$oG9WKI~+SD~=v1s->LpY;SKv@*n~fWoF3+Gpcb-10#sJ
zMl+i~0M2&TCs*?_noa=}oFf;23y(YIM#O&x&1k12&d(YhsC{${PH-NfTr4GH8T3d?
zaHbTfW){da89EdGV5jTxN9zWf&s>T24=F)>UiBWjaOeB{pRP!B?hl<Dgm`lN?2vwA
z5*H!M=~r@-KI~+`N^ZPy*lmvV@V!GppiB@#vE;m%ov@)c;4j2RHwvVAqHa9!WGU|I
z%JnjPfBM*;x+128S9f%m@1b=Q%pW0Zf{CPcD}}2@RgEi!s*ek6iB25m1VQNjFHXT4
zyEe14Y`1hTHa^7MSzhg;3gW)Kcs>3LjLQLg$unwha$-8e(!nRsGZUsr^h62!TfhI0
zPa-Oo%pGnOKKq|CTi*-fnRrBUlx)%7iJ44VN&4`+eFEu=d2oM8$~QT3=B7bMQfF#R
zq7gpK9HHz8#z=#^3R3Ii^^}zQmUGL=*u=jNX9tkIVF_$7vHc-s5ZL4YsYch(^C;28
zM46_fY}=q*cu&l4>#K|!K`GIDFLOoLOS?8t%QC$KAjn9748vDGp)S9n;pgvE_g|iO
zRx*w9bqk!}%)x58pd7U+U9DYW*ku?TbK|{!pIhs|2~5=1sD+(l=`C43-R8d6ei}P@
z27PQ}9TP1hqez7P@;4ae<O3xL9il_$!(prrrsQ5%>~lWQYfS1Ryf4fVA2x#VAXD1g
zouQjgt&SU8Jya;!`RFE_3HE!RyS(0Qj)fV#8h!)rfyL=hvAV06a_lePr<3lRT06u|
zP3dj;AB3G{SXJNm?GFtS0@5M~g3{fkl!%l_H<C(-fGE->U81yvv>+fLN{4i}NC-%m
zgh)x<x%K;t`@i=-_j!Ed8_wBh@4ePsbB^&D3E~ZAdK55^Zi2G!^2-ilDym=r2Z3j6
z%dL($MTqU<Kc8{A%jQ9b3NCMAA@}qH+nrVg**;B>*bQ)CC6;qK+csVMIvh)ET7+5y
z^$nGv4H>e<q8h-U9|cGXX~q{+kLjF0Iu!)lolIY1kRi<?H<XtvoCy|;{Xj4zS#1$<
zbNBb&;Y#Lxlbp8(L84;5HMe>ipF5vvjIFf^f1Mv1B=Siw>{d;aieQipoB>=!9)QKb
zU>5}wur;HfBQ38BC)r31PM<2U2|*9R56eeXR1}do(@A+Cys_9e5NohCGjo8sz|5Yx
z5hwk@lMb4b$GiKbBQ0(wsjZmF9!GoF?-VgnFEKd4q6HlU1U&E;fbkg_UxSCmzgJe-
zJnU@aR*O!NJ6iqGMm0jk0=)?6w?S}*FCs3U2_EJJJ`_{)xSivn>e<z2_TAM>%F_h#
zyhn6|D!4q%te3&10mn*N;QpXYp<qYQ0-(7Hm%7Fq$$46RrI{8J4`tn*;0qc(GdLxI
z9~;?g{rxcl2`>i1oE0fo*qCKIpX;Je_iQqjT>JX^@q)dlNoo#<dQ!pt;%riMI8;Q!
z+4Vat7a;Cyg@=X*oHQ?c6_w3x&o)3eB#qtxFT35hDsZmhLXnFgvP?}cBmN)6$-(G%
zYR1^C8z_)3yi_OVbGG=j{`|f_$=LSqp8^6t?3XwwvL|K7zqj=3d>BpGZ<)kHgU&O8
zgaOr7gCi}#TpYoCTeCNAdo!gtqhzm6UKZAqyL5P@C@~QcmI-1lAOr$OL=N}@(2v?$
z9=O85B^}(u`}BQw#7FMdi=!{QtiIfa-v8$h-Q$A?aZ-*T<`|7He`7=`fEAV$OZLOl
zbecke=o-~=T|!xR7r2UeEZT_?AvDV8ywm{-9%=uVFJCq;<p<}HUu0Ok23rP@7LcAQ
zxD9WBCH_75`fFUb(Lr-{6VzhQD=-;^T1Xq?u8!AHnS3}qcRkgS#kuCMnZop7`HNq;
zQv0}5;1eSriKB!jZ#_YhvNiogg@YRI#r({DI5i+XaHLC}oO}`KQiIZE+1rd+y6`J*
z3Bij~*<S`AcsV`}JyZJy=iVR?gdrLw(Cd4EvjN#Zlm8`D&gp&4m{;Kl@=G&vlkn9t
zljqbOB5t?O8Tk%?5x=4Iv!;g=XuJD@aJITLuaI6EpL^@6@t(L5Icg9n%EoXUJf>D}
zlL-)6auiF;-1Kz(Zz1@bVsaGV-VlS@uhkKJP{$yyIFRPpjLby+_m1LcUTE;a<cuEc
zhRwWkLDb9W&>K)e4#uuPDHDmbkAjCt&7A*xWZ{?ed)1Ig=&|?9BJNhQqt<qkmEK3O
ze?`sGVm^Q7I`!4kp(_mqOj^Rc{!=Fgm*2~)r1Ww2L^I)=mqux<C*%IC!+k;g=<L)J
zDQ;nqfsR6S&d@-T8_91ie{;c)ldJ45yO0crLNw|T5Co7z3iyr#L@qH1v9EX~O%YxA
z?RU<+&eg?}cafK_rq#HLxU__<EX}evrRm;&ku4P$>9(<r5L3tJ1my{dKoe_Hsm(A}
zJTx#WHp9&I1&DQlezQJVIb?G(Kz^j1)^--*cRj8_XHV0Tx_!Ls-Ee3GaTkdN-T}DZ
zU}utoi9itOvcAL0C-XXvIJrqU*zJ&y{y#@Tnab`wdupBSmSh)E{kg-(p5H|sscTy2
z7p=zaE=-T9)hYFaGU2y)Rh3eGb^pR9)jOkX&}z0BCIqBSge%5t+JJ(_d<L|=Kq~o3
ze+zN@Xmf-#eu;1tdHjX0zM4z+mBmG<Jd`e3ezVlJX}KA^fJ=%xh3Blo>qHDRL&0#e
zM=B-6staB~Ku@qYGvk)vao=@zy|lmfed22sFVWExlP36+x+P55!n0{qwe^all&Qt*
zLe1;=K0{M!27kTo#^$4~(PL^-WQr8U2!tn{sYAy5eO~9(5Itubw-4CjcSIPbC{>Vh
z1>OTQnDU`~kTxE;5uswFIM(;LKk+H4$(8Y@M6NP%o~~87fb1y?2u5y3qli%;4u0=(
z=rmdHO%DxSInbB!!UzOfJm*~SWZtbJHmS2Y6XUkU`Mi&^uWs*@xQ_}R$Nwmv(;Kfx
z?s;efqrg+~^h_N)Flk?n&-*S8{FJ-=(@bH5jg0u>66Fs1Y-ri8+^Nt)-%5Fp1I04&
z`>P}|qgVcz?RKBLnhgH{N(@Znk>4&K9(FKH-8=lb@aPdsa<)C5nDs+y5)_eRuX$s>
z=Otb9#F6Oh=WrQvNG#5$i<RgS044%RqF4a-T!bZsqQ&}9qkfiqTx>9mSo+8=C5PMC
zlUWfxbhf{~mY}V;IzKMwkD>hYtDNuEnl{TZY|~8mU^!&^B^I1%UHZa5#xu&yBHi7c
zQeb`PQ+K6*^5H#LjVzV+a1ZKOKq74VbG`#H-=k)La)h`aUXhc&li%#-ydpz17eRJ(
zHIJeCZOw%KKMyW32TJ+&Z5im}z;oaTIL*h){b*s7+&4i}_q+;|LAwdQ@O5m_j8c;i
zs~<eKSC7o>s0Jw+JiB#Xi~ZAJ-EaKLtR<yL0W;J3e3!hWUojZxk6p1J<)^*lO-HOx
zAkVhq9QOq12m!4K@zae-fYN==kPkz!PG86LymLY~!|fFLUTk9(J(qz6lJ`1nNB|*1
zmS^DnG{IwvKwpO6b=1>hGJI4GCeX(+8hgvA<Iv8RdRdr3dyONZi2LS?V3Mriu-1|g
zD)5O6CSbiYjbzqp`1MnK(ai)}aPai;*pCa$FD(JOfd<&7a<J@U*2k3`^tMx6>-pNt
z$6juBrMU+7+oq-_#88w8DtVdLk;KVYEOPF5P}+}K5DT%Bvrb2%?tByocl&W;bQDMT
zSZO>)DL)%e-AmzqB3RX~dr5$M5rqP1NgK?4W`6vzf3dcc{d6(ToMCZrhP_&)>s^2*
zqaUUZ9Jni?z-+eK);pb-h1V(0pYpPO+NsyaY}-<__lX&`ak_W4tRpeWG!Xm&VrYYk
z`<{bC6nWS6wg5W&hO4PJHt;9(i-yIWyMp^qR49ju#2#wQfh3W#$?-MM$s4wW?r!B<
zOr+qx1_ywiyGS$keRe-PT=`AC7CU2*!qt2g#{c<Fle5qCvePzZa!O<=e_#t*{2R`W
z&dg?a?)NU{uczju63A@6P~QMS&AldU#M2Kr3K`(JYz_Uy+p735cGb_>NMNs6Xs4T#
zA%GqVE;5u2=*I@+hT`-5X1mBxo804nkE0Ji2Zsu;<GfQ8uvyIcRa%;oe7?ak+FJ5c
zrOG}>*S(e)s#-i2=j-^w6Ur8DZelm1=pvw4MxlTO{NB-q-eQ9);wR~Ab}?UF8-Uh)
zj)FE{*(+a$tPHl=<HMhpQclk2E>q}eS=F=EDIapZNHVtVIhe{z(jSbqQYzE^qI?6*
z2f+w~6YgJ#OIU-<W&_S|&1Aau8-GofhTW~=!>T>M?+4MB0dVBPhp8%6a9=#jG`Id6
zn?h=uTWtD+4~UG7Q-(S{?~UkPL3?hghbW$5(jEk`Vvvv{#SS7tff<9_o+AzY>kBf$
zm3S1WR8ePakPyqm_yEy@BP1B4+PHD!kAvzBm+Y<Q8OvLUsmZ7@K`Ib4hL(MFAVziJ
z*4J^4IOYp?kR~z6Q0IUIgNZ_TDnN(xg&{cVRQi*km<@nbr2Ibk9j)Z1A>3i4RSJ6%
zlybezFK=gqzP1utI$4uonEtCAJP+TG#{GzG>wzb;)Arz&ZU^tUTc*uLg{g1+Zv4?o
zDJ^E6_YUhELdQnp3T-k#6L1yB8<d{VcrmR0jR_l?kEJNWFG4tc2rdEnM36-{A&X8U
zeZ%$TkJ<`^GPncsP$&TB;T}}4$UqDD7sWMa2#yJyPXVwfTICvB_a!jlsvNEU`jI^}
z`W?gP3&kU!#f;jj*2WG3?pU)8Jr46JVTSw4tM|@r?Qq+^h>7t-I;9{h!#&3~p`m-W
zUbric`~pIL7HHrbpM!IIp1~(B=y-1<U_AgKBD7Tys7I$c*x++^WpbEKSlXoLtQ>;q
zO-K~g0m}Vr(*!c1p619<bCwTYRKkIRnHwfI^NL2bXfLaHliStMq~(I`@g4gRwOr~6
zddEc^X#=Z#&=cGeuL&Nty$eEn^N*f(<NH_mOc`Vbj?|+MYq<`bia(Z0(YB$RM!)pN
zv;ETRTUN#&Rf3V^=WNgg>fx_BPRUs7aMlMf+8gLIu<f7y_3sHR8v^72;rzkgi2?fv
z!i)t$p$sfvui9W4L!hw;wGQ-K$PKp}|9}Ua)J}8ckOK;nSzsi=bhZ&_3MFPO=e|y8
zOesCHT2+3LvlP~BZ0bDNNo$bsBAneIF5O0$B>$7EhV@osiId^i>I>G)(~Jt5F~<!5
z?s6@EQ_3ygjs2ZwL)*o`y)84QY}`sz#ix;-`>Cn7_Ub%$nvJNcrKVL%%V0S@Z^s+X
z+1G6hO92=@wkXsewXU+-LH~$;<UKi*$nY<B{K!aEvA#@)y*R(?GQowU7UyRG?}04O
z2*432gwSvPTK$w8X_3J&cOcyvmYZ`3guu{{7CFv<k@NA&AQl|!I?b)%bcEOzk&+GO
zvRl7aH00!79n%;7zu4iIo4c5`GZO68r|WatwR@^)?T2>+vAh+D-UP>ZgiRa8fA80G
z|2GP)YB9T?-TQTO+N8)lk1^Z+5>F}!0;FSP4Knu!Sz9^159JoT4!f?oK6|V(cH;4R
zPco>*{WqQLpyOo0M3vQ0_cW#UgjX|D(4FSsrB~0cXAgR;PdUo3ZTfPb6fRI%+KPGt
zjgNS1!SkRm(iglz@EG~~2!?=rSjH~t^3Xj{FQ9dm;+{XR@bDqml-n970$8B;{6&(Z
zoM_fxiL}6DClH<nj9MHQI>~Exzkqx^`v3?J2n6!!)H58^dqDFbfE?%#BVMNe?>sla
zb*$O%v31dEpY}r9p+2cco(3j;=tH`5;B{P{u0&D!sN?5Rku;P9JJW?hF@wFC&2KRo
zm^W)c??q!(bh~UP4^J_u1cWx=eiT%?cVA@_2cKm2wFy?<>gE#rKZ^3qc>m`QIdkDL
z0BUFibi?7!!{kpMmi~uRH6axhB3U^(W-Ei)SJMpGV$o4>utr1R^Z$nd%JKC08LHtQ
z4^%1BB)kq#BF}!|?CFyF&qP?(doh2*POC1l*u(#P94B5oXq_0?)v|V7c^Kz|Ye41N
zJQIGWiSNl_S^NPqwi&_pM0@d4Zh`m3*9%<5=TR6$D?yYD|B-aLS?fhm2US5;)zY;*
zKBacU@=0&>JKN{5u{Ya#Jbb7HVrtk!A}*t)BSZ53EX50mH4)xQ-a`)n2Lma{6~J_e
z79PQ?{2T9X{y*{VS79aakz-Up1>9iQ{Pu<K3Hr_vn>&PrQmE51Ijo{x!YL(dn>|i^
zlGJfw`3zZiC~VJBJn1au?aQ>fV!31b{@LA;rTMHYN3&?L0T+DE?p6yWp$p1iwO`3z
zmT2`KJ>=uZj|PLcVj23jtCmRd4M9*hfWHT%Br8GP?=<iC_|EADk7;TuUAJHE0z`q6
zbTsqKr)jT?;gQ_kV6g#Drz%Kog&n4vK>=pJ-tUUQZ-Gzp79qqaN~aU@mM_vwFI^<{
zVA)Dx&L4P?(5RG`GaRktgv+U#aYyO)%4rVf2$#a#!hj%l>cGM0{;HU#;4#-=b)IXs
z+!D+6{3*{w&yg(ouF${3`|KnL;p&W+&jyeRckufZVn4FmzFYg#lENA-XwfU^!%5V4
z3bRzkufy^2#AYf`Z2?>3PNfm>@u1Uzz|Nqq5LkT;^ukuKK7;}trfGWG<XJtSxZpgz
z{c}MnR<8jtJO~5vSvMT!w0if~$M_e_laRMht6<P6s*n%4qR5K?3mB9(`P=coq=-d3
z`Pw&q!_zgzz*6GolXC+fqj(kOpVRm8EOQ)DxDe10nku{V+glow`6huMo9&3i)AAFt
z&?;orqCPf%BT>C})qkP#ng9^U5K#-Dbkqy=1&`t6w|r+5&-#m(J}nb}wx0zT;hV|{
zk~?HLdScy!ohlk9(r@@)1MuiP6wlxydr;#&kRcD`O`&B&kS=t-VnxsbFu)@rCQkq9
z3LTN}KuZGHGXjr6J^CLceWZ&IUNccR@DQPp+yI~kAdj!C30(<VXe4zYh1}wq2N5GB
z_kmUYIVmc&5xqBeL1=-C%h91q{<i|x)lhXjox(7<C|I58KABYVo_(sV9W`DR_WA<h
z4bIm-xbGUx(GR#)c2uVDbYJG(?R&3KB}_y+nwmXPPS`-_nzhy$oqb)Vb@#%-e6l~X
zoS{&lOQC?1ih7xqibd-W*&FJ#tGBorTk1mef3(Kz%P%v29bY|fqr}G;K<rHX_<VX5
z()WVCqwumQ>>i(?G=U`%4>&Cdm>61b&z?Pl4NA_B*Sysf&`b#K3aDFt;Pt!$qj``?
z{BJ-U#JP=-YoOL^1LXDnxMEu8|L&%Mt-jc=x+=|^7&OY6d+T${Q^xNE4={WHl-7yq
z3|(FxkH@yRm?GKsL13kFDl;mzq5o*}bf7@;G4owXu^Roo5xPfgF?YjT+!eq1%wLkf
zT4=Cr8c;K#xO6nu&mTLwyQFd{>>m;;;6DE$p=ufX<05MK4C4Xn;plH17s&9(I#qN|
zq_K9Ht{@c9GHrP&?^6ljI_}czcK(sz0bmcbH8222`m;W>)Czwi3Eq<VX}>QkYeHTM
zC?XKH2|zpHWoZThowA!`^KaO!Iv>~ZWJ!e7Er|<n1dm?CP$d-P7^oD_&@uU;^!RTt
zdz9m->T$%x(a8ernDfIJd#9QEuRZp~^Dk7oHI+z<8@DCJ)MT!mbgRrzzA^%dBUtK7
zo^Bdjj<H}#b*;gJB^_qER(xGI#zM2$OE9I4L?)OOj%G@f_pn3b9if@$2nD~#-k3FZ
zN|>6pn1X_W-DCv`LIibFM$RU1(=R>-Mk%sV_U)`uq_*8yQI_b9VTs2+VR)|gT^}`W
zO(V~E^RR{U*Hv%(IqmxCA5-|oOLNuJ>X(vZLviSZXGT0*hUZfTH?1H1_7jeqi;lss
z5MB$oQoaA`?K#vdV`FcVwTiFn0v0AGdhauOWk%_l0*K9_t9hHr#-&+~5kZ=Q*H9Yz
zR!E&op&|C*4fgH5#1iW8*7dt8?<zY?yzicy0db=672kbPy<pA#4ojvjH*RH>sezlL
zQ@7&Srwk6>e<!9js={Y{vA5s>Q^fL*=$heu3Gb}!53IuBrH8XcA}y+O!6}?hk_H&w
zTU$I26+?&(?%A}FH2h9J-o2KMs>KORxBBjtl5&IWL+neuo^M3%bSDuiic<sf2J};u
zSbKTDrx?Sjgw!?>n9=~rF<PsI<l5(JB06e+-XtDTyk>Axg2%HF=G9nKL=0an;m=np
z2d14jd|qT{q-6JV>E)}8<$DG@bJrP2-EAbIKBm^RNJd)E)xS$}JGz*hRFfH2b1Hr?
zf6tcS%)h03h(pLaaJIadK9ewN<~&brL&Y1i#!9ZFm~dPMx%++`Xu|g6WKgqnYm`2Z
z{@MD<=Q7BiX7?Y;T+NZy!7j=cI^0pbeNXO*Dg1LD?2<6CHmJEP898Ca)TGy^i~NA|
zz~QG7xKB(6(EvC&IPX%+XrQq>r1#1=<*hqe#2egRLKWod7~qC+8GIx_Dyldxqfn6b
z0!gGmVq<afYO^ISuedn|dP`lSU63#uoe?o-d54Jssr-X+KVwS+341CMCn^Vt4(9;z
zsFDP@#s)C>6&ZygLMbo_z_9~N0=Rg1yvq`_>^N;D#37P`*bT)u?T6{)J-ZWmRdE?S
z639?o08atI$!9n{;iHyE3gSCWrfJ)J?@_fc{7x>rV1^&As1N71)XhX9+3^JG>W(z?
z1sdyt58wKNKBmVzBqqoxUYVK`9)ZRU(%H8RJOi!<0_cuYlp9jeu&pXpp2*BQpEmOv
zutqIes7DB%lII?}YN`b5uq$o}n|TbyU=&Bt@xlcZ%L8+(8`ntcKI{&;;9Uw>uQ;JO
zTK}o{WQ5|^Pa>;tDFj}et%&dn0Yv_b7UQ`~4uD06{uu7`TEP@CcQqho>9OnLQ2{!H
znV>CqE0qEU3!r)Hd}2$k2wJ^ATt!ySdpdl8u^NA|;QGwNzjH9X*1C`nP1=#?jlM!r
z)_Lw+3QP)Xl8Y~~x`pqvst-;@_HcDLX<-JgN~?4nlFxBWVNfxT<y=Wvxa|1!^JL%W
z9w4=Yvj7e|9EKIde=)AIpo{?qg4I?fs0cuXFJ^rTlA{C93VN``Bv~#-5z7qgU0LK2
z#sAu0Ux!y>`HyxV3O!c;>rnTB)`Y=SCFAyt=3Ty<TCm}KcGi~Ze!Ic$d?5^uxow><
zl8M8*D?Fjj#cH#ezWSzGAmWQPnh&7zP^ID5Zk<}8AR?n+vY1sv2b01fm?%~SrEb=p
zT9O2EAAqePz;%qSQbd*y|Fh8C!rAKhi48*>OD6&VTWemu<_r9?I}XGNsYUkwf(({;
z-=zJy?aS=!RIvw%xNnl5U5-WF(cytgAP5Ta&e;wC$lw38%(A}zqnvBOX6;?$wfo=I
z+&N`c5~~OS0^irOq}`z_341y+K?0o^JniQ2X297zKym5A<1tDEL3tI&2c9M&R}3MB
zaoqy{Kwhl$vf@hv&s0ftVTZv)9S}<95u6HOC59%7s-y@8oWJ>jQrHSG;daAsm{9d;
zzR>ie=+TYsut4BWfZ0QfoI{3YAej3O)B^s)$uBHNW_VwIXNC~p^@0_Qh(;h!W&@2W
z&iXbEp+>1GTGP23u)W}dG-PMe*m%i`_8UShA27PvO&)&#97o~UXCf5%+X4WpJ_lxP
zln)@WOhH)w$cgmoK(k@t%Day7dVZN(hZci$H^$Et0zUtP(F^}bL=)=i{=d63EcYrO
zG?h5bPo0L1mgvQM%RH3faN2y0o_yjreYnH_;^}@kd^DMKSpP7Fl02$tDo2^bXS(9y
zq%za=Nu)yLvEM@@9EKmyFVP6vh2`a8fFhx+v5~&Z2@!#f7D-9Z)>5*iQOn2e7Z}*|
z2{?>VI`fQfh7;f2P6WEJ3_33Z8=OW~)}zkd-$=jL9b@z0c(pN>Mz;IvEps7X$HxmU
zHBJ)P9CP{fT_6b}6~pC@D&k#E8jnp`n7!zO17r334_@?nwIOLrDJRTMr5*#BYlaJs
zXWyPQN33;V_n2)b5)OTlt*D9h<o)x(q6q>R?WG)?XQa|wKf=|2i<~IsBg36Se$~*K
zfdou^uIpXtb3d59Zf;(m(-ke9+q?K!@`pg56E&&NPWF!z&Xf?y3JIR7ZN1Sf_`D#@
zM%_Zzx`ZH+P~(^M-oKFpIm&UL3i;cRD}9z=h#g4G@b|;a<2BLDl?tf*5l5wnppA(Q
ztB$)Kc}q!lVhonu(rTg81dI^iEGZv>*Px_f@+7v)!p(o@%s$OWsL<d|<e4}KLs$ns
z{wntJ1ROPs{!ry}BCdGBGv$*M5*apOX*aYBX88c)g=$O8T3}fGe!|^B*{_3aRk5k|
z{2#h-P3<TiZSy&N8#!Y<t9N#LX&#Q(KrX;pT@^z8p<_=0^`W+nHh(^sA?wM7inmyA
zR|Pmv3izMkOFy&fgo}{@CeByPe>rXQC@%?RX9Y2@){Mc#hJPaG?TpiQNs{#<%DmJ^
z$xrcxC%JK%q03rTlgc}~SNVg(pH(MHZLNsHsJ3aDL<3f7AAH*4k}!@s)qB?|axv#1
zQ<2hqf{0XxO$w|eLRbFkWO}NX-?K7Hc5CUk2->kUHyYjBRM%LuOV07@s_Oi*?46QN
zpkj50$p*f?_sHkSk*~j1BvGTa>zyc5k|hZWmn=#K^n&fYqD_~!A`_RsNwg~K5?B-2
zsPWwT-fT+5nb~wOVztn#B;CHonozBQ&yaNth^w!!06$g}M#O&GjjAf^+!j9F4+$GT
zG#tHF4bHY#%+J=&UbV!MFQ^H44zTZ~Z8t8txb5oU;!OXh!8m7xrm(`XijuhL+IzcM
zG)k68qf|Mc6E#}B-WeR-<a!0l_R`BwRvahXe`gv8MMl345xl=rFCf2V6TeMnmf{&^
zZ!Xl`&&Bhb>bJrLh5#1uP1r@=YTla4wX{<_FD0$DtFt>B8a1Hc@=<-6+YQDfkLWj=
zcIdAb4AT$qi@8KJWovEnh0x@4u1kEMHH&(ZM)B#rM%z<S!T9Yv)wxU}pUOT7yHON8
z-eN;1PO2_+#eO*_?Em0FP8`QzYxMObs6>3f=Ov`wguXt^-Jz!fSwUrck2o?UKE337
zqNif+Gq0z_oHuB~{{BGn?)vwkJ^5m<*9>!C1hB<*95m^-2@NNHjs6~g9MM$PCzC~Q
zoV6er3+<gu$B%sf(ytwET6yO4VdR~jqU*2juOzg9aR}tyafd1JWKi^(hW_d^)sOkO
z9mjHt!3e&J%C4Kx-|wa2#liB?*Zhcj&-dx&?1cc<$d0-vn0iZB82WB&Nok9v^UjA!
zfm5r-3Xc$vMJY;WS_A~O??qRtrf5(;e+e{(i(YXHTZhLc(XseFT}2?zix2c2+$Pp?
z33LzQee_^V$u`=yo&GX3iUDb2IA_iwS8T;2dB?rqL%>UXa@D(inZ^MsA$thTvWiIX
zkMjvplyvR7VDC$pcH$ap4wR{Y{dDdR*x?>F7btSBOCO)V811x&`&#7b!R%W;MHx@g
z{)E#;C;?ixi+3aa)wXc%t0j<xP(Ah@qQ*JxW`_r*1VsgUa&9kstTETaC2SEv4~;c~
zcm}1js?rL1(^&q<x>?Nva9}-P4g(YyE#xZ9vnh$QGVIT<oTJ+H^kt7{IeUv;(T$D@
z9;3wV`4;GtA(DsV5=d5!&=_aFJo%%TAfQT;FYK?Hm}nQF^W|Sh&GVBkoS6MNVVl(E
zkghi7mngHMSr9JR*91g|My;Z>onqglkfp_0$PDaarT+29(9ZMZC+3NU=j!*=#H9B|
zCtfkvpLq5-_4jK1u)q5^z2S4h4pME;+RkL_@|X^sEwb+?A9;evSI3${R#fP|PEJ<{
zTaS^}yOyt9Jb*amB7+Yhuk|T<d*Vl!{Ua>(Ij&>xT&P3dCkb<CqQd)xWw4r-Wfqcn
zUVL;F`{PG;A`uDq9{viRazC@R8``O>XA$%KMF&d9Azv8#bGV$Xj3Qh8qb}yHH)*zT
zyOuM_e>p5LSdBt|U$~8|H0Mz4m7!Crov*waTl@Fq)6Ax`MEJFo>J5<ISg7qHE6X0`
z&qPu5CMuqtT$)D}|2MXb3${03BCY$A6r46kOZNjCy!%S^T;|(}3-cFh^!<5_OyL$8
z-al{3Y^{8L>Sr@e@}}U()oykTe6_OKx!ZeF+ni50`#b=~o?T{^QbBM~ZS~oDIi3wI
zh~fF$G)ecskPC|$Uf*yFx8K1fg<rDJx88U>B$;fWJii)IRdDvay)IOJlt);_$dvZt
z2ia5arDz0@(0dtvPTYT8pMLV@Nd0n{Qw<s&XW#3Jc$8*5#vxgCuS)QUt5k%`HdExX
z7Bg<+&6*<UYT*OQ{CRa2!|LdWw?|?_{L3qrL#7MenGC=B7SZjr*#I_WJ(R6^8)qVP
z!S2nuK!^}A0SwfuA6df}yWoi5<%vgSFy>*>Hd@quRN!aMV5RYWlWnAeHqRkZduVMq
z?r{dort_TGnVwprav&By1rja^a(Uk<;d(JGRl~s6qlmGmdIPuL>hg((K&c8iHMBd9
z6N-t0++WF(I!iw6F8f~RC8F_H50sALD!)@aAz5fXP_A5XDyGLX{!-O=Kyyk#qp&%*
zgB5{&fMkjgMJ4`(LtA-0D5CoAhy5u}EzEJ9Ck9QiT1ul^S}$(DrP7#wK`av>hjv+=
zFN@gW8O#K}hDd&&vSj1y`|+GZ+^CYix1|NG2V0HI<E@IpQY0_}u1qpqqAcjUsr^u~
zpINh@bWo|~eA%QHQ0&xS?W-U63St$?s~YPbJVjHT?nZ1`Nbo0Uw=VNPOxv5PUX@QT
z=Xso?%G<>X(=T(#dpRe&CJxZUeqafb$GP2j32&Il{`pYrLSxaB>ML%dSMfh~7dlv7
z0~5!tQ!e@i?#t}ex3&_sTc(<Zr%%yO2TiIJ)Ta~-`fcKm(7W`}jE&I_?oaRvKpw@7
zZ4QIKWjw_uuf=UY6?^&g5o$3F>%uWZ=e+w}V-Q;gnKySI=5edBm0lApx6n9$pX})S
zFy6O~$i8l(<Za^9pA(NBjsVN`!Dq1j7CI`;aDPLix??z{pwrK0M)e3K^?+5uwD!ir
zN$6OGPdI+wmD;=*%XRkZLgz(D_7aoA<6V%J*Y5?)jv6r9outjf_*;o)Zj<7XjlBfY
zP64KXSQt+|qh__ZLE;#1R*l-!Hw`g=ZEPQjxspyNi<L1E5|NR++-&Z&&(q!xY2MDR
ziB)}j>(5sYDfJ2I8PY$F*XUu?vQW#pq*vthZi=y+xaeET9e(luF~B=~a=?<qvY8o{
zQ4>`nU9j=F73aI~Vft7o9XqMnqzq0|rDK~D`7zosHM~QfU@y0$Nt><Dj>K-5S=c6I
z+yBY=nM1BxoWbhJKinF&vZtgWS3r{a=E2ZvjddN_f@kMzCC(Kl=42C53Ki)B0POvg
z(y&%-<+3j0(D`1vi^mWLl**C^9-(LXvki+`1(TVD6e(g1<KGgK@*Z6|#e1?@g)Prh
zaHmFRo3G0BqK)-&PK}AlfT`z#C$s{2xBc-%1|$5V_%V7M12)pNOrFtnnvL}KFW4D*
z)4p1AO9>$E0<wVFQ200YYozWnBsUHLU(sFXs+B6o&oJ?gdSXWlY{CQ8r|2RhGio7o
zkK%q544TBN4VcP5OZ#W^NgP52IhJOi=tLAod}Yp$SgLvD2r|D{eV~2t&?COg0kAzA
zmvlPU*93Cj3o`^DwMtTnUNl`wZryKgY{vJn+FwE~YCY15>;JLw%VpzEJha<K>x}$p
zUGGm9v<F9DVcRG+$T--g8q}ME9t08<P<ON;4$KT#7fgG&&#1^hXg`L{5>6(3b9)TJ
z3}0q8m}Tzi5h-LitVk>ZUIu%UoH;<#7*-&=q*F;3%Minw8L-3WAK);!Y!ZMjCJMjT
zTmV~JhWy@_fX$z|W@+Yh7aG=TEcXR|veR@!H09Z;JL3F-wkR;qT7a0b-7CI{%A*=9
zml=HZ>|vo)Ezl_+TRq0G#CM0WHf&O)$vN(7DGvd3><St@-cI<~+R^Hek$qLxJyEf-
z>K!pDop~aP48HQc$DHxjmX`NE2jG3J=O$$s@!jg};d#(&Hhh0VdZD7T#=7`%N=L>k
zcNnKqt)^XUi4o_JVDA)`Rt%QADsyYW)LE$kIu19%ehp%4Lik}oqhBB)1@R|51ergF
zfPoxq4mog0))=TA8ozzJgycaydGZm<1dAl*qom9D`>_Y@)$Td2W9yGT7q|;@7o7gy
zu}|r;lHo*K6^mkgG*_lWg5na1lK(59(%KmPXz2B0T=0X<$ho)P3bVytjO!a$KeI$7
zw~vjVhhd4!2K~EbvikU?7yK_o%CYH+&na}q-Gd9j9Y6Zz<oa<<_I;^LSj0;QzZl{W
z<$&nUMsQoceoc#{o&4=pLN@T9gwC-~4r!fpu-~)*-v80@f#bg%Zb)Zk<IM&ByX$Xv
z)(=^W=PU;;{Ns8AkdZYsG&<8~zzLX%e${hH9rGdMd!)vhtqrx2P9L9Y0lzfwtZ1>{
z4eY(t@=l8tC)yyeXJA8L8<(Xfs8~F7hFx2dGmR-GH2>G;!a~|wqgDOPi6N+56!~oo
z);%@I_P)rsBfowN@YhVisDiM5fBl%yS>pxU+w<b$;-76znHFkDas)Dnw41KG1QOqt
zWQV#C1Stg&oHm%PaBh<ZC-N|giwHjp8ZFr)z3w01bGvV_R-}fkW4gv92Q%Ug+RKlO
zU$Xl$zb@hObe>k3sg}P|NjwP)$3+E!1KN~<?PYfIix<AY?MG%A(d6W2dN|fR;W^Ll
zG`+mVhF;jxLU_6FcDe4B-K~2LGPvKH79naE#vO>G6=}pmaDtq^eg{D>ObNelY(zrV
z--0JJiv2)~1j6cfyoQYFgN-p(%w_iiIg;P@4Q3<d`u$V1mm7><n3vGf2gw;>x53Ds
z>lb=jz2RHmd5+8N`fbbW`Y@74BJJSTLHiX=JYG4l`qrrOeMQocx0h~pVl;(@ty#aF
z(Nq6w;Cc1V>;{W2w=F!)>KidPkZc$LU1DNlA`~=`#>pa9Aed!B@)!e5CxFST^DGiP
zvM~4r%|pg+4Gc3u(UHj8+1)KLVSBliKdYZ_+N)w@Q0>aIA=K7++yl#RtK4QrPIPb@
zdQVT+u<Mx|ox<sVAs+3TcBicXU;QW=FgIpFpxO+ciT`N41zw#5y>d_fmAbT8s7V_Y
z`gTh`kdT|J2*1A1@6IpsJ%AV2bsC{WK!3jA4eiuSu<RoKMr5EwNB8j8GaX1$fH4_9
zubi2chrxZ;xE@H*%Gd=4lkm;)q9;67Lbps_tY(B?tLbwF!D8g`1~zLB23O0zxHx)$
z?Q@bF+XPmC(nS2)Am2qtAt+iv$y0#YuoCb0g>=M8;r!)oHx{{kAp0P}%aWEBPES>P
zvgfsU49#Q<PIwg`z_zGapiKzBAqEO@fx?|;YXetRR&MT}z^P{tcl_@p2fa2G<)%;l
zrFE8NZdzyjZbG+oUaSHI!L`JJqIl;xd84qTNtP>ZM(r*+c%+smoH;C_=-;R{<p=9a
z;vX^YLslQEzuvC*Gm#_Lg~u??kPm!r94GX#q9VIb5#ks%H>^-2&}E~7sD`Z__S5S#
z03fXKI3z{p*D!rULwTR<QtHOZgL$<)H5dF8dHS)pS-vI>mX8hNU)U+s%e5Wj;OF`h
ztiG}lQE-c=|9V0Djd^waHZB_803AMl=SSb)Nsd2{z(7tm(Z|(<=cV#<IdtDSj7hZA
zDLUeiUK$P~DC4kxVR_5>SDoO83X!>GY%Y_DIO1<<SV)Txto2sb*6h|hC?u)^E)wk4
z0lB%?5SAdkW-$4rgdiFuoeW{|KL(~Esvlw*5du31wVQw~3IWRjpuRT;Jyjk+49krv
zc>SVNs6KEeQ}a1D$ppY)yxm-4hF*%0R?c#9m_UW6eeDCq;Dz70znjf)=I6;+9mK`P
z8MxK3A3b5i`yBd{r`;G!@2y`;Ra{TOM+3~)BiFxB;*HLn%wnCY?6;PL@f;aT%WoRE
z<ZvnNaB@A|)We^W{es?5V2)<wt_q}eiF+cdO3s?>ArTQRz%(^lJO%waqB%t}8bOU0
z2pL!zi0~Cm_|WS=2O!gh5)eHnNIc_1@|%O@{$y!DKpWyF+;jvH!GAItqepKZ3`vD)
zA7`^`dh0`9@_}~x)L1$q?11J*WFAxSFxfAlJkM^kow%!0D+nJ%o)GQJ*_S}DUTxcX
znsIc<fdK5;OmE8Cfw&NPZ!evxz%JpUZg8al!7DExwp2mTo~l!)-%nDn^}dc_Al$lg
zsi|5oEyKr=m57J>#!yVD+npR+Y#qwDh4*HP;ZvBYm1)l}O77&XoF{;WLq#B=FA$p8
z%tsE|+6z!6rF)831902^Vq_tOj-<+L5Y($Hl0k5np>%pqOHp$9WI=-}>t=5N@UE`*
z)odja-iCoZk=rnfRmSi77eP7Ff1p)Q`FzY2UpTy_39^pAQ+?>J``#37Z{v{M|95e}
z7tcT1($>5eUK-4@T0^CkQ+6vlXqGNxmuO$vX&K7Rz4<poFkF$pB5v8!_SUBP>PuzZ
zxA!|eNpau060oK*Vj2~~k3qa-SW{o5%xi^0{G%@;f(pctoau8K8?ll%!-m@v^9)9f
zo(ADWK2K12-qh%uFp^2FU1W%-Q*y0%XNkAv==U$IY%tJM6p~F`8AdlP5}6_55Q$jY
zD<<#rv2mj}R>!pTIA^+M_IB!1sL$8GH0)vOerhL3+5QUgv>`p;#lx;ztPsj;JC-+^
z<WIs`Sng^6PCk`0FfS0v>B$x&PvmjVrsYc1ovrEVY%z&E91{){HVov^R3mXaj4G-t
z&87~FYkD`EEn$`DdtoWsY~p#YWj#Msr})Tg<WqFl7QO_($>KJPk0gfZC$kTGx;ayy
zpvE>9`UKTmR5qU!`N35yTq8H|Su|$ukkuTUr#Pvln^CBw&CZ4TU;IIxY1xt0flKrw
z7%k<}_C!e=n>#^^A&rdlrz@~cnflO}VaF5M%L^*ZOr05D4%+q@Ayw@uU^W2%WrU>P
zNG!)E8+<2faKyG>P;{-$VH>QLuzEy4*FkCYYCn&gaEEhNrPcwKRuL%iQo7XxW$8~W
z#ctnrph~t#kNBN>qlF`Big!msHH9(QemU_R<%)=9TnTJ5@b`lo@@Y-JRUU}GsZ68T
zRGQZvx%__k-5XqvVgjwYuDr;1WM<k5Ga<9-?{YNQZtKi)q(~P2aAA!LB-iVz33b)j
zG=P~k4VI4_yL0UK{ve0yPwNHO@wKn!d>CZb@Zf*AM&x$8=Ogt^!QfcMG-*r<K?`9Q
zDJ7D5p-0aMEQ)1b@956+&{z>##?Ksm_cGqZn6}j9jgRb~v`@n5({bA+lpznQ?wm*}
z&23r7hvpJ{>X;^5@?<8<tkL3m;-?C~Rt-wVhhW~EJ!VKvWuQLR)sT=$5cH<ZrRuf=
ziYHGwoVV+1==M}iY$$!7*aI;X%75#1tW9IM6tg+zokc+#_sH7Z`qHd0#*PfvCxwoQ
zV}kXH>W}P)(njxtLE<M?{S|(LBL4PEESJuc==Lx>-o~;F9TYpW9F+-o-7EOc&$cwQ
zTX)zOsotUa@fQUIT$Bi~VaVDA_FIaCv*KkG{=!B9Bn)$!kGGY}oV`@a#DPFWki(rQ
zR?-do{a@i#`_aBb&9SrRm`;PE&F)Dj1ud7PlHQd}i!c2N<B#F(0Jnwdeo=v2{42D}
z@Vxl0d(J-{!%UoBv-INwf?ILp+I#oSg4*a+JORYW4J(}drInV_u11><$>Af=YX!ie
zZQxcrAMJRkSi$Depq$3HDiHZjS(=fC(cqPa8&#!D?i|#sZ?$FUl-WDg_GxV`i=Jn!
zW{XZ85B${>_JJ<fcyerN$i)24H4FQy847^+oXXQmGLO_`y`z}drmw+{dE!}~y7_#`
z0IVap;j!iCl3(fV<<wQdYvKh@Wkdj5)4j-y-m%fhEwLh=E7S_7uT~^-9)CCUB;!{{
z^R0F43TZl1UB5Z<PA)=`Fh4Zj;gYSe!p~OenTe}JsC>dxQaGL6_jEYelU<zK4X(@U
z;%NCX{bfb@D++}}D0o3>?g%HqCeGz>wSfIGIw=GxkitWG<xS@!s=LF9YL{VNWRW^i
zurGpgjallkr9L{X)f|qVUfh*8E`7a@nmI%1G>nAEyEb5Gii>L;Cfz2(i33TQ?{K>w
z4-;fzxQ9edrOUM6Xs+Iu7?!FfEG4k-V9A$G`e?gzf|0@GUInp8f)+{`SfK&L2GEZg
zxRu98D?{nhTkVc*w})TbHdg6&MPU%P;ksLt%A^FEU29+G#CTnMxuv}ILdc_oCzpTk
ziv`<Q=M*p(l5tOuhe$qSa(k8SAUgc&iER>@wW;^6O$S|noNwVRxq&A?e6&4xkI@>X
zyX)>)so5GVev}WidqDhqUlx}^=8fuQId*w!qFrrE*u<vXmjbv}yM^yOCcKC){FE@U
znnttdwq|W$;1kP#!Vs23y_(2f>TU&`ty3MSe#JN&<Hgl|i3?Tfm8v)%w=fY~hfjM6
zI@N;qAIe+H_J8SE>8#oA#yRiac}sLHiD!ABoWe%=KzhLwJ#=~vl<1FstX^Q6az~3R
zPm#g;{YilA)4`l>Wyg;#XeqviVa&3o*dYUd8LEhjXp6H7ONgBUs>b*-0@l@ToK2s>
z0-<KgIN9#<8$^@kWL(-Q&)Gcde^1HP=?$ExtC5o=R2tjAoAv0spp#FZQ{KjntOLR2
z34^z`5AN#t;Z1v{sEjhv7`FA%JZ~dxe5xe7k)q+Z>B$V*0RUJ2y32GOL8-1TQoGtT
zI8jVz^m^1|?&7?HMjIlpM-i_Q=Vw3S#d&mx)gj}yi$g|<-T=fE7Zlfho)k@+(d8zR
z>LvKKki4ZwG(sPIV`=KPl+w#_Qe}d%QR>NenVs~+RjUVj1zQyit`1r}mo<#?nCM`E
z#DuILV`aVG*|Mzmc*b-O7o{C(L&XPi(BKYOdKZwOq$4WVoxQ_0^X}lY@Kts($YTOR
zN^5r_zxo}PCX<Yh#Ju5wj+E0+Hu(y>o$`d_snuK~6owq#yyN4&<o?}cpr9?#Uv2I8
zObNYyS^r~x2ryG05S$)xl+cSl?pp0?^sr7BP3@2hbRdK_>%D%0%L7Wmce!M~ykj@Q
z)MF05w+@8(=kT3|Ul<QhL}|bL{K~y$iHC>%_kSmbfkOsL;wHcgz3^O!4&1mJc<lyI
zdLzUl`{+K-tH?qVb2m}?&AtozXG%I@$~NpDAG~^BQHhHP+MSpB_U7#(14FYu^a|bZ
zv^wS6lU!EOvXy@;SFRU?<VaQj?eH(N^~89&lH~u!0cEG3j<!I<y_V2eBxUcPXnO?Z
z0we>PHha6vcAKmA!;T(tkt6;Wch|N5d@u!_aPIM0?+Vc=%IBqjNg3s_B%&S}AmtfQ
zT81lD$9x4GAqFM)`&&DbfcJ1Fx75z9d$+tQ#fX&5P=xv$a<xIq6e3)*+{%402?_)n
zL45s#W2xe~;rMS6N0;7OeTw5~vL&)ZHLV@uX}L~L9@KgN?6wyB$o8r@DF|i{Ay%jm
z<3-({7w$r+PK2aOIeV(-_PXGCRCG=zXY%j$95?L`jBmfschvW1uknD&t)Zg>4>}Fa
zP@s2yaJhiq*2xg@EoWaug<rY#EV8rrXQ*LKgH`TZS}*|w^yP74k5*jB->Rsj{^`(I
z>8KE4K;Eg*BMS`xd02qDXqCKVSjUE}s9M&;{2e@v+pJB*?~)$W+_piXpo0y+ACluc
zBSHigIC(^RA@IO#E8(CMR;-`#zhB;9a){vMZ)#=_)5l2^(6eQ*K$(^ep+$0TjM_Dj
zezN(#J?oi$n--skKn%xv+h)vKHSpAc*lCj;($hm=ykuO(?@(C%y2Vz_fE79#ntV}*
zy~`G<Q5-J`kLEir09W4})WY!b$N>dT_1!ebXvZ)u!uj;KZ&u1h=abd+RcD`uPI;_S
zd(^<2=h+bPKLq405Wd3X?I)X?;ioMLF-lJi-U6`#mc>Eh)Y@|!3R9o-x90~p(+q-2
zST1Ct_wPAK2Qzy^(k=~ztQdc=oYHp}4Rhd`r=28E-SoO^wb$~;4DYtqlW$peS1azi
zH<+Li{5%{85nVorXTw0daPdTp;NEhK2i#7HX?`QiFL&nG@q(#Fm<?0L3S0OMOsQVJ
zyd}Vzs^vlNf*O_yj|;b7BUAgL$*(>?i}eVVtnu?y?`RsGA%Myc{l@2HPA=<h!a#}X
z6FGBiKA@}EekC?`$Ay*=L{rX!wz>txFenrR{>Z@HSsh*a^oj#hHjaVZH&#sOQw3KB
zf|3IfH?(Mg7W$%8k)+v%>njXpteS&e;?LbRX&rr4QYaFY9d^yuV8f~`Z2Ee!RM^bc
z>PgvNJ)A24tV;e=Rj;!lYfM}7>+l?TWAxsF@`<F2k^bJ`Gs5ST&j=6=yXBTJq$308
ztTzL^BHoI0sMH?`={ttpY954Ot7?Ep4@OK4Fg6WtrekY+1z$prG(=1`ew#DazBkt@
z<JYgJo%>N>AGq~QhTo^z?%=~k{sq??3EIxg{5Io$#MsXxQ*nPBq48xUK6?5TbdKWv
zkgQa=Bp@uM^b;y>%{Ugpdu2la*v1!9(|?kv@dw`VeatrX=g9@ekbpkVDqXe-!@S?|
zP;P|5TrnNc%OlzP_Da^{d;ZjJ9uIOh)HDeh(RMG#U<4&9XuM#+0t64j8-d6Qg(XqW
z?b9`bn7-NzU278DY7eD*7i3w=x)WdeerO6)U(jN(yJ6B$n3gls6YZlqZC4a+SANP2
zkGU^;!@2f^L;(isaj{+s{hkN-j3#8`<^yBswBEP~aSh`5o`e6p*IDo9$>@mD^Y4iZ
z$?iKM1kW}J?OSrGB0k=gKDH*O4zo(0a63_gTY?2%Fqi6riM;>e1*>^bJfKZWaIlh~
zc;Pmz_t3l?b$<=}wTSSB?xHINV}xFM=v<E?uk+K@PxK`=bNeDY^%AabT1|j48-Q+W
zZJ+S`iM3vlvh)LnwaJCaQ+ze;rxf#sNmK5-da9m^OjT0u%$v!-ox`9J9nQvMk8oa)
zL$gIP_!~Z)ML{;eA84(vFD~!PtC+wMzfGTpjCz(F+ex%Ermp?<pLp>7qP#Ao_0U)<
zr)Vq!Nk&BV1NaXC8v=l@|Hx2w>I`ACKnPQ8FXi9y14Bh35C84nVwe7}g_#05`S=xn
zcC|}`zr;rbj@pCQ!}UW)g$h_m-9OjvXu8IZ2i<3tS?7iw-9R4A6w8b~`3s?un8vXK
zKpD-NojYmc{}@zpGES07^jj3oFW>7<72jQA>X6_y?KeeoVhNfMKmd|?vA%xb*|_6*
zR($M8iW3=c|Lp2tjuDkqRk^8jx_d6`t<izi-pnHfF?yYuBXL1X)+vfEf-##vH;Ww0
zhdj8@PG9(gQJhBR9`Unb8Cvm;ktUn2bfgY{Z9b(?=z#rZasI@T^cKN4=iC+12bDK^
z>?o+cyb#9>LVWZ$7EarbeU7Ve*89BZobo?M@0W4@g0*#np%@7+2sbt{_apxJqS|Pk
zoU_!Zp&c&|T)c+*rG|6O+CqgF$HgK^#u$o+s%9-sc4vy`v)W0XfnEa-SWj$g7;Jly
zZ*jyz*S`-IHjxQprz!M8aMou0GgX&sL?%zkzRzy-5yCd<wP_0nW{ZWDQ|oJ<iBVZc
zRb8K!Gk!>_Z>^T$x@cdUtvoUaj#86W%`I+_!8EYlu~y1Uxd?8)H`MSo`V)nn`5vz*
zFYb%^*t+#4w^hi-d0_zGv;o%DDff-i;?sA|HwjR31Q<Tj-yOmlYb8I=a6gmU^!8=1
z&lpHd#A5q&$mO^;6{IT>Xq(G5Cv&WLD?8Lyq@jFK4Y%uCAj~aUqp~bjWpLkC!25l}
z&VHzhLCJcrgA#3-3-~;$hS;dfYd_W6y{Y<a`;0)Qh(J`2!3tTg7SnfRk~zdP6L9KW
zS*fLU`iVhl^7kIYRnZUf1i}=qiN$;qNwN6HA&$#ur(`vYyXdf>YT9pI1u<%fyG~jH
zvr?r5&(j<sI8%VbkUYH5UVo`97Z@&&#CS2xo`9;fZ&uRV^1NSyl0SH^Q8#GHLiVYQ
z>84S#_!Jkb)$UFoaX%cviBaNZ8>?UaBfS5GZtCi5XEbNUAoIE`&NHqUeK^hRC+}l9
z93FXsNF+BfuC%Zka%$HT$6-n1)Wo?H_ER6{PUCpW4T3j$7`ic7-EXmdY4LreL9c^J
z*0L>@`%&M|BT?V^&ghZ%-`KQ_-O5monHaIh;|9ZTN8b!-`ELVl;CSPA<v_8brw2Lp
z^fU>wh^?YX8^4+fWebJs75ptoIf^jaU*%)8XET5ydlO_2oYE&t#8zopMf=zP0RcUE
z((#R?_IF%27q!M0g;A+yaHcFRhwZD`T#<__<pi+WpGj9vMm-Lq!T@%MOLmzxM(FGv
z1-F6aF>%jwOqV`J{fB85e`5wlD(kxwN~Nmy7})mIR&&Mk)Gb&?3rtj&92c}8fgnZc
z*NqnM^SBJ;^l+>r4rT899Bv`ev2*iH7RlsfimQFn(loxdaEkk+@d#=7U!M~V#%-m|
zmt-MSDQE#DmYyGe$i7?oMQgVAe9tou`%d58qFV=bA48AnAh!_|6(JpDdiiITNA>z{
zJvdwEQC`W5N`3mwq5(kTd8gj>iFHp!Gd1_SO#8i1Tf8sP-!!LwOj*(x>hG0Sdd8H#
z)AT995UGs2@_}2&wzJ`hT|be7$0WVC)0{G$!3xJhXyB%8LHi~eL)I*icoXDaWGBt{
zYtIc9XeR08)F$R&;{9VBPvGCPyo%;5XB_O9=)bkDh07vV?84!(U%?t#w*QW~Jv>d6
z^dX2;5plzEGRa7}Vc`RMvxzsEmnW*uU2CEDH~$c>{(>?mo8>9;u`GXl-&Z!3?PYLT
zqoC@(sX=zhs&Sc<`{#~tL0)2P2+mpdWx-Q|e5-40`+H;V2TitHVzl)e*q2S3C_c>A
z+JRf%m~*<bOX8~2wx5Pnz2rkdcQ4#C?66n9NvvqEdF_{f4VZLOqs(#Y&tD;|3xveA
zafv#t<B!<hA=jU|6y0?=C<wPTf>us`f5-WMV3}n;?7!V_;J!Wh^NlRt=6n?%sr_~C
zGDM*0;ONeW(V&zu8p?+UAGTr{9zDU!D(ZUt+E+TV{o-3XhsJAjSwD>t-7?WJ`qBQ6
zHUs(VO?s#3m5lc8lOh!%^0;oo=uh@q=sh*iMPzM%ZD(IUqC{N_N(pns!%eMSd?URs
zblIA`S5!?->WaFQ_G`zJOY4R;kEHqJt>c{JTZJ`}AzGCk0>{+=TEks;fR0Q7oSgWO
z6etLxVWD3VAwq~LU{c=#?@`Tm>qStI^3w?1PnWnVDR~L8VsY0WqyIH&+yj>}l#nH#
z+4C#N#&SL4jd!R=#&2fxJL_qFcOi;@BR<2es@eM`TWqAUS~hSNkZ#L{7ZZkEgnPx*
zxKld@?<l?RDO)(0Zlh;&C%<^HVI?(N!(P^U+37v+Sp7K<F&&uIW!09`QK~Aw;AX=+
zrPX5{U-XKCHyL!lK2P<Z>#TB7dGM7Z;ieE{&I9q~2-h6JGBPk`tjBKxNUj9T;s|dR
z4Gj%BWg!+-WTT9Xj70eWqmNNklp4~{6d=nCv3Ucz=L#R+;2?LdjD|(pPcH6^EByO~
zuXk%yzb|fJF8#iQhJrA4{+m$`Bf`T8V7z<pIDYz#j-1Kx@f=g+X-e3;nwX{7B4(B@
zQ-~fw1vjVi2>I~m+@v>YwL1Ix!w{jVcuCBmz+>@#0c25K3m05J#=p6D%LK(TYN;CV
zHTf;~zi$%QvLm6!vGoXtupx-1Mz(dxq6e-y;817g=8(ZNGW580>sAB3_pvXOWDHD9
zn2-}TYBKlj=`>Pg-+>WzSZ!QrQH6(+%7N9SP+d)XLHDFUHA>U6>`p_)vWgT;^8UpC
zdU-nbf6Y<3h=8bkeSJ|r(ND$SSW;5=dQK9KjqwqdV}|)l30+fZYj}(U1hm1G%z6&h
zv2_#s6!xq^k60$!5ymzebSG4sLj_~jd7UEbs_qium#_qp4u$&a>E69d3PEj13g&_*
zOw*9@?BPz!*c*_32SLmsANZ3Xx=Naf6bcx83VLVx*5LKaD?CJhX35w-&&r4O&-JPL
zQ>L_~-<a2@J$~>^6<`__-q%SGj5lyU8+jPJUzcEBtQW<NbiwF~lPI34umAqik#sk-
zmqn=#yojN4L3^$?(_|b`QBwFM8z=j;|B^&62|tM|>K4>Rn3;Kovg#DM35!`tgJbm`
z;yNsGL5CEgT*YMW{3F))$(4>kn_nov2mR1XLc$n%O5m78LEwhMf)_H!F$aG^o^fqe
z=iJ`j9zqVL%T$z>Mmu-z_PJ^+E30#Y>*vBCsEI`I^M#;2aCmR2_Ocm1qAM6|?RoK%
zG0jnUMJzO8|I-E4Ie$m<fp*c0nNwaDPZ?gS-_gXv+5g5ORW$5my^<(wuHVF(<~&H<
zP|mj`Y!#=P(#veX+45SDyg-O4uV7g|vZwOTk00Yij!GK(C%De`eC8wzp7Cq6_$U+^
zj+VW1z5kW?w{~;dMo(~Xm<`Ig&!2g_Kc%CzEI&g#-jN_6;Nd4I;K(o}-Y*;`W7BB5
z%P#o9vFpcp(L(BYZLFKagRny-hX)Vdx_rx9yQ@uM79FR{HSbpGmvHfNDFM$r{=UN#
z-fo@={vCqqEzf%6nxbwvdNj#;Y>F-&NrsQ<UA#}F_^Rw~Gah~B(iY{`{`q)@wT5e7
zh5cloZ+^RVw_DbdoCCYPJ6BwMB)n$C(pF?hP`jrzRh4|gNu@K+sW-<i&7<Lh(J?#8
z@h@vso3zPm_NvI9*z}@uO3{bi<%t2rgGrOx8V$frLHPj9FKpD=*;%DcMn>il<iLc@
ziKP(1C_Be_T@O4P0yGgTYp`MvPX7MVL{+03CH6qnoAz#v-)+?gNp7*y*Q6u~f{XSs
z+A?GIrM+V5AA0g(Nwr;X9DhSQ&N@juOe=8U)+gF7K8-(B>_1ody&-z_L;!ObTa#j<
zBV@RVUo|%$yZOrFC+HSlKR6@1j+;nUFZ+KzgwUf;*9}`D2sIo}zTv#6cElD*q%}6e
z$&@(Z3DtO-zAnR;V44@SeEW_yhtqod55xHi-cc(EcDHphh(O)Ed6N;|dBFTu%teZ$
zsB&zgE{O+H8oQ#Qh%a75l7?SH+5-SPa8R^jk8%AjBl*i2`*B?h(OnLG>QOoKJgy{T
z4#K5k<c{N7fvj4+<>8&?Q)+YJ$Lf>bOE(`?X)aCFHu8LXcIZUX8C2}MmtK7kB8bPb
zSdNuFb$#uwl$bou__zPn+*?Lf**@W;n?@QzL_$hL1*HV(4y8+_K~hq>OC*#=Dd`UB
z?vPeWx+DapyPGrn{hf8z`S4%o^Kt0{7Tdj_=eg&exn}N}>$2;ceURkp8Z-16he@?6
zAU?43@j9Nxcrv0GyX3hx`+Y8S93Qa<P7!`>soc}CwOp_~x&JNCRAmjKq@scpl&iD1
z%gf3lVq!3X49@|M36V_I!q=yY4qLw`)iCd~u=J`imjFEhaz-WDb?FXL1Z{M5G)~Z+
z`@US9$ekAamQ^>-$V(@Nd{yGKQ~VCkC6SL_MA1K+=cnIYF|7*rxOiV=YNP7b<DgQZ
zJD3CoGr#}!o~>xkUV?p9)sMM6b=W9_jg^S!uZPU2$JFR-jKvZVI($j5YD&K>Fjm#i
z#S@V;Hf^h@Y&;933S>8|Pw=f9gXN$bLoSEbuLaZ=5Wh179p~(A@M?qzB;7TV3--0H
zD~4tv*?`bF3$S0?Y&RY+0B1xtg%6^{YML9FO~XKZw%9Kqv9#J77@~Ng^UkE0=P4md
zQ_eUdXWZv_9g)>#9`&k@pOsoQ*6sb1f;Yp~DK{?v-u^?vN&Nm!wU77_iTIkGcWI@U
z3QuvdIth8q62r|?CuxmaEHvy>@+<eeNklf)94#K(S@xzDtxw5`VVf~)Y9nA6SBVWP
zOB{%4kS+pdmIm#CJ2`K~o*|Orga=pGuhoM5ZxiN=pB+aeq6MQNET}e`_P;#v6Ers;
z3hi~GKTsF)eJadwuJ@y516|^Xa`}YfV{v5p69JZOkK*AymawT?>HOP6#+}MzR2qb;
zH^WM=-b|i~bE}RlNrvHDDdxQ%{2L?oPYjcb&adL-<0puFk8hJ4U>6CXMaTMlwj*%*
za_=4%YXGetL1EfSxx}qRR)e41Qz=^oZ<+$+9U8Uz#%$P)6I#Rx_xETQEdJsdOZ4Wb
zhFY9kXH&o5iuI<;W$UOoVjq7;{xpvHw0yFvM`0+0L;yWF&pE^Q-E4uN^qr4Rs@E7P
z?-dbmlO(@n74^YXc~y{MSitmD$-;s`?x^|OrxYhACp^um(VT=g&1Li1>d!Z)m<tX1
z(glLc2ZcEaqc8B^)Rog$UhdftHe-FbuUL((-JjUs+2r@-!5BxdeW+I3S6sbB^zAM2
z&7x^e-QgFm)c@MXY4huF6{wpD3afr@Bu<_Ae%(6$!}JgJM@pY^J;jIZT>qW5r`=m#
z-JSEZ%G{iEeh~qKRLFLFddL!ue`|afcus`dl8MfW>awBo&MyWx;_d~kkHSq|4CY$1
zI>oR8te-sIQTygDI1~@evjwLIX3dCeNH6Qq%u))=zEV_@?Oik@?FW&1YV}OPvkdiL
zV)1Ic@C??JhxNJ>#Rq*}<__m?&o)y`O%5_<($rkVVlF9vH&f{iKgny3p>oL*_8USq
zM7yVJ+QNK6_ZC+ojN;~f`njIWNKUXWi9TZcU!-id&n-<cq}TfrBUQ{h$O7LA;U~Si
z{<L>ZEqWBVh+>8Eh?mjpIN7AUr%Q0UioV~n8jG-5sa`c>EaabPT&na{m@D@1s6uei
zt>`LC6RXyrB=O6Pg|8b-ZPfar_IRSuw=2Ge{fkB+lG&$=86gtzL_<tyGx@vhU?~k6
z!BsQ@3Mb&5(h#+5yu<U%XSHM7j7eNQNSKYvvwu0`3qPT#*1Sz@IL+knj^GXnrhw7J
zv#HTN+Od@Bbdid7zvzTA>&o?87Gs2(<wdnjRZfKme}!^h;4h88)D;eIeK%|OTj5jg
zTMn>YAvg*Q%)Hutjpu^*4P`NP`0v6PLD%|hV(hchjQcBR5E^<ZlGk}?^Y8fc+hpQh
zN^$H6EZeyftq<tLD)-p+6DEBlWKgKbX+C8Yc@Z7B8kZ68ezb-Yj}<WLKc2?=<AdBS
z^k@FUaj95c^%~l*o#<-6ql!F=m&2wjB*gG(r_1xcU>@?8nh54s+SBqKwikC;PErUB
z8YF#5yFQ@uHW9;!Dty5%&S^v6_pUXL_leL6N09dnX4*l_s+#*h(E?#-9jV8m@2!m-
zAJN-&WgKk%v!P@w?B_b>nZ<XrLfLEf7&YrGdOiMqKKeab%CMu_)2~Egn#_=1imRaA
z@O*Li=aYNlIX@g-?&`<!IlcPXLEKk}`U6G7);4Q~gUR|EZLR^6sOvI$!XB+#5`o5e
zzU@mcoaBG6iNQZi<z^+R@CcxvT%1veZLQ#?tT1KH^@KDUQMJ@Cbbs;D=zP*upEfh%
zO}CtJ!_kcu4f9{>ZWKZc?mfc7+x104+phIat$SIfeYtS_yu;aeca3m~FRzRkk*!<K
zUmrOz+_LM#%sxIT&mWITFKfu$tlmrd>`!kAo-C(RPn=mX68uQ+0%I?gZ6v;xNJ!dR
z7|)`Il{m6nf7&6d?|X9ge0O?Eu``bs2hlw(G`BD-gqGq1Tl7hDY;F4X>x{oATW&KS
zh{T@vh@;Tj%D`>DSk^F&;>%|v)kq53jW^v<`Z32=a3C1M;)+i-OzDJK3ZpG@;ZAke
z^b#ps_h=23&VT-KRC4=ufV`+R(>oQt;|p?+m&3(_UK73iiaLGxA%>J??hUm>+C9R5
ztAUcmw-+DW>1v56)rza>r^mK@c4g()KZWt^Wkk=mnGv=6O)>1+OE?!y<S_X)HFMgA
zq97J3Y<S<+G~9mvdkfDqXg?1_Tl{UKI)*QhFC{lAC7w|5p(TD>`ge!ugNC*F8H=98
z6-Pb~@0yVVp_uI+H2s>tyrkh>t|Em;Wjq@XNz+U_gap*~o`sAHtXB{i=#-gf>DpU#
zp&C8@SVmr{I&;XAE>uI%JXl<J(X`|#flZfr2wFNk#=WV9r$x%MZO89p8H?9BlgG=Y
z5qJM`PsuTyKj4x3%8t;v*uOWUnDe94TlI6Af69mXFM%<*KCRREvU=fd_T>g_TX(6Y
z7O&Y}aH5s|OQDl<ex*X)US~8~XBTQKpXy4~y=$#>@X`L%Q~o$p1!h==zA4?rj+Ee{
zy&IRLua!CRN8#qs&QAn-i0iFCTa7*+`UjpBUFSC(P;TkjW3aC_&#Rc+{mCng?ywo0
zyF$Ao5Lkvg<a_@sA8EYsKEj1p@U?wDk-$Kjssk;bq%aPd@RXW+@uOw$2NU(JyY8pX
zxvH-U_xS>r@VKx--6<)@dFq@9^PN;5!W6REOK?Y@^Dx`RZMzsA$edx1KI8l|TSVg3
z^GHH%O?cf=FYPH!7SnOsa;~k?b+S?EbrDAXh9>N-M^CPBj0fJV3=bbUJnTquZ6>Js
zGlqr{or6&0x60@HWfwJ6XOg0%5s<z;M^YPKp3Aj@)8&Y_Svu3vcIZw%C6@(5FPVpF
z59eP6-EfQ2c`jwUMgU)&YCh_%)iQQDFYcxH*Hjv!Jbgj)XDke3N9GI*hnXCGP`#B%
z#Af!2K+}S}aDi4|u7Uk?dEFa%C#&(FG>>(io^q|ati_ARxZ^pw6n8MKd$$ebnEMJ3
zqLZTFkfu&Yq!9=O(yh|}?MRxBj^V;FzM>{Ff0do}L=A0$n2#TYqxtA_ao&>#-Ej8_
zvTkVZ9OAR-)%a#YIZ8#z`Ha-vaeLzyuG2s5QVIuD!+D;f`)Vc4ZBi{6?`XZZIdV1$
z{ufkjmmIi{7!xH*$F4H+h^*<dEg9(ZRR^!4B_6o_-ri|C5?Z{~EWN{!93Zd%_QH`(
z4_BSUL&z(I0By00_mV0=_JDi3mP?JO16>e@Wc<aW+Qn=fv3fI_$&24#Aq0)TOKbT)
z;Qtzv*5xd~4WrPHNG#DJEbfC8LhFuUs)-IqyRL!Hh&}aa>LM=z6%UwJ_eCp9SZy9E
zF`os~Z}D`-${C*c@p{DvOChltdZ*mtA|{8|$`y&ipGjpS{@rB|GBA|_>x|52rSq`8
zDVBWwtd)$qo&8X@B;i-9EjKH34SlSyaH^B}1d|DULp`C@FXE?R&L8H`SYQ0?9gF>4
zc3Mk$E639ALTF1@>4;J*B^x%sS=s}HrKI*lvW=g$D(1g#WSuI_JR3GxQ`P$%*szJK
z6#a!R?{euA=GkX~%pPg-CgO)ZpCMoQ#`Wwm!3|V#lg`J9e6G}A?^LIbuPQmwO;$)I
zna_Xejy@*SGkxoVhG-EKe8j#<tax!b^Wq7HxBPcIw`t;Y$@HXaITX)7T#v4i&Y6_)
zeFeQrFb3syuB_nk-sBvk-~<12OHO@9|8eCK=Q}RO2*3~@84cB$D70A#ePf)7GHnGa
zw26vh-CL`dvN&_fcMrGh$-XC{`hR>lqyMdsX&<+%&t1qTO1jzbn*3mc$+J;Ml^ye+
z-0M$1(sLqzzI*ddejDJXlQ7q~{l9A<!SE^kct~S`B4B4EsMLS@J@^*Ag!5ei^w_K1
z>l#!3*|`#o1S-~YdmoL9@XHNV%tKF3{OfNwte?7KAR3Zp8zq>1Znz~1x|8Tx3caz{
z?iqQm!ot4UpT^t}_7klK5xw7QOhZ#UY#$%rco>pcU2*nlN!U5hSSR-a7jS_BDWZyg
z$-AKZw;|+%coiO-Q3dR-^uPUu*O<qx3%<TDoHIueO(5K^PnLLfh94DUPloak@poG%
zptao*Vfp}9m2n*-ch<peE~xl6-98ikkz4VHOD<1=7W7Ov;ZW?pck^_#a6dSt6ga}c
zJG(cDv|TafYKl9IT7noCN;p29(<fa1&Wh?z+hc|O<u*vs@~ZtrTChiu>_rgcFFO>n
zu1c`PDA@>&$;s6tYg9Drw|5l&?YpcNldR8Fu^f|ooHY}lbXpO6sXS@_Ux*xx=zim&
zXeHz~<bhyAHw!yDGsd$M{XjJ1s?EGB`QqybBkZui7(y|Cpa`G9yX;w%5(?b+l72@Y
zb6Yp?|2EO8Tv~AH%jQg!M=>{Lv_nkqu@*&<K^wzRnKEJ`$fki3^?s4)x8q?h)(P2U
zMANJ@3SPrognsU7SIMplt%{eBudSk>R<V<DZP!Pv%uY*PW&!l_Mz2jQ!RSQ7D+ij7
z9Ji|@(0sbr)C9Z~c}L6@gY8LM9}hdQ>kGTr@jibsRifAslcFv!^3}UWzOVK3Vtr&K
zEzbCCn0S9cj?Ly~4H*d)4AI$A%cE3tR9|d=h*5O3!NihR#wdI8FXK!1ku|R|9+vIx
zF-9*!$X3}zw;k$!?HKcGGWke<JLcrdnY{75Fx#|GQtcX(msylCr))olss){D*GqYC
ztG)B_-w?M$yjGtRx|*oSz}~|Faox{m?2oPVy|pT{?>o+9AC7r&T-%__yXjTLq?n5c
zL|u0B6=*!*KD5LOQyxGa2Nx877wlRNz?JMW1e8TfWB!K21w(ztXUgi2+x#TA|In~D
z;uYQ3w{P%|`$M(el%aIgRn5Kqg<<_oDZ!wWJoh_TomK}^tD?fc)THfyUDWK%mn#v4
zRT1UdbN5?>E%^OfJ;SW^zerAxWZpOQYpNlk+FnvYF<;-@=;dX;eXj-IY<f)mQ~mtn
zx_@^rVy9sw0(d>1+ovH)q9fyL*A^K9X|Y`(*WFpmFBU5v&w!c2iAv|&XUE&`?3-ka
zDt_^_{>|R0rbNVyzi}HIwXl*R>l@uzAXcsVVPM4icIf<pVC}%tK97psg)rh*YQ5WY
zMlQ3Av|t7PcsFiF3Ob^RBP;}<k@mQg<dvB&ZhfoX8D(rbznTElmRI*V9wE#tbK1nZ
z%=Hu>M7aqZ{1Qi&O%Vu^`*JN^G~OFFxD?XE{&K&j<%ChIhKmd~l9&!Pr3EzDw;pP}
z(;V)~_aze8WI&Yg7n+5aCMqlqn9fJ<akjpsKvyJ7d1ZyCBK4-@XroYZo)MNsd)!Pu
zBd_Ilf};M<kSj9LROV?Cq!adl-OnK~@J7x<v7T!;gwGN4?696>a&i(XgY7VP!3K=i
z!%@WyovUJoO3HxSALj13&l@KQdWTLnX!g%V`5B~{xK$->HP_Is8!MOpTCKc^_vn0t
z^rAM;m#B6L2f(=oZaOXF3CZJ}{oV5#%C|%F9g`0V8mmU0?UX$`+_2Fx4J|%S6d%x~
zn|BQxBZ5i=t^|rc^9aPF?B11k<;($7<0Q9UF0;A*^uYG@fr2vw9UV%e+o9d+m4T?_
zeEV?=Gb-)defeU_a@Q}UvUtO#OYgP&4lLFF4V<W4md?ccJMX(ke*9V$TGAAYjge0s
zsW!up-_VO@eETz_{CQBQ-JP=7+ih0U_uIK_4L(L?;%5Y4H*ClY^6QM>h`Sk@o_>*N
z$vbgzgBu(G88`r$!!44a<VLXw$A>f>9nrbp9(8=qVx{FhWkR%s+~GtjdOje<NwDb1
z#>063=l8L@t<^T43%j1*e1MrwQqc~dolrdXoHDM`!KFLdad3fxWTI2_n|q=8_*AgG
z>CA+GLUOE#3|NR4|DmM*bc%YBHJ>(o<3X`sTV%&nJ7Ma|`6#5`*mBDP%2F~%d#tS)
zA8U6wqXYwa8}-EPqAK3}^(Zzqdmr1FXPRfh863>4GnHIYd(#pS*5h<U$*o0Kf~0h`
zT~MpihZ&*no}Le4X)VBVvNNKpIN2y=>99REe=WXXcIzpHvq&vMWb;ANo1jr5wWzz(
z{=YQpBdR^$cL?(37q!)^KDl|l9EPKdkI0?(wC%Sq_=8?AN+#|aeL28=^vP@9g#cpG
z6TY&yNhc@>*Xd=3lhuL5qok3If3ja)0W7X=#%qgX>W--nsEnJOoph(wOS%(~h&`7@
zTEHtEolvC46pXiUBT_+~4bBlnfFx=onU&kvrL+R=vsVm?Hg9jR6s7d4IW7~WFU~V#
zTKd;|W#p2*xhsC}gcQps9eiyGvM;F$UR;>Wxw3dZCQvoOYnj#QQdW88F86&8xtS}D
z1~z6=nQMZAgoo#sC^0<;@dwYxx!h~BjOt|Yb2F1og${ogu~AAlEh-h<(i6`+I+QR{
zLp=IsHt+_v&?mq#Sx916q@1F3_+1FfAgzLwoq%zq&Ex@eLA&a{E_@BeU<o`#C4aEG
z^gDY0M8|l`U^5v}{gbo9ZF9^Ul-W{=wxM6&d^;!B{@LAjnCbl?X~H}wt~~2PC%DCR
z>fsYBskH~MMaG-5y{j|jUgx1iGry1BHC@BkFW{JdlRg+{EBJk%YLgez?Z>3xd4GX!
zD4Ss?9`rMb6~0<Twlo=r3oh-%lVEwZR=0AzeHrtKT!-1)rZ%FSJ(mbQpg{KYM9(ex
zO-1f|UttDAc8la?qM0n})`Lr%%?bJ!<7GiN1G}<}So8nB#Lg4Kx{scE@VAL9dUa~z
zeQbtA2K%D<I$|ux10U)@o#L>1DjFJ}kab0<8Db!igw{yvO(Ubswsj~yNr@To>YEnN
zeo6jvud({gqYCa^9(%s&`#f*oj!70b))7?w5G!JSIN5`SJ2u~0pBHgj8L0H~ece6n
zr(;pJs_437pQMC@?CSa7k7WJtBcUG!mx)WHySuvur1@+?2L;jMi-Ef{ruFVS@Kpb~
zgq^rPvy8Ce=4|Mcz<b!BPK*#&vOU_64;=pRnSAcG(a5iw&R!`r0*)YKR;^<8ry*s$
z2Eu>ROLIaJ=yy2oQ@k(>9@*o9$VLI@;E&@*5N|+{&jpy(lvPxGeS8o|T?j(qi;~Ra
z?z*ULm#BVnpHHJR`46iNCa4)ug1i=Y4ukH!*w%3;#3b;Q6#JOSEXf#^9!owR_mx>u
z7%#2jD#LpT@&aUOn78aMnFRxNAv>vIvdy4cN2Hd@JmdfFG+Pwzl#x-yW}W>%t;Ehy
zN)!ZAQ5d*<VZ`josb|RgN+PEMgC>)&jmyb*f|aKnXjIz)y&+mf^5+~V^E~Zy8G%EJ
zD1C)uxoDTath%2k3a2bPgnq3Jc*NNHBU(Q?i_OdAuCJM=p(pvc6t`NaP~L+b06zwy
zs4ZmAWqfQ5*JWKeCJA!}lF_<sPp0Rt0-nvfpmBV)5LSsE(iw6+J{(*OXO_wN9(Uj~
z)Y9{%YmsD2Vw-<znu<iruv3(6W9bc7`lwQtM62#Bewwx+=b5H+r3C}zGH(|69)*Dk
zXKH|(MSbybp(pxp3+FZ_Xk+YddtXz+fui%Nr%c#3W>sIHnEH}Uvrc6()7|*Hfq{s6
z+^fFvhi-2_Xq&eQit|t7UxgHm$MMFKy*{E`zqdbnM?7_D#`6KsY@flJTF#G-LTcPK
z2gQ~W?lP#E#rn$p?}JN~kRN>6CTywC+B@w+1ha}yngdTs?L+FHS6SfiA3BZUUKOHo
zJljz^@x*-WLC46jI8S(+m}2R}tfyGg{RW-{GQy-JpVSxmk(Ym5V$WJzxH-o>)}n5>
zky5KKmK5L=A7C-<`!(fqKWvY)L20p`rcPVvMw%6LxIB&+5x(%&t0a@T{Xtk%|3CC)
zbseBRMnYG&pFUX>Ha3lNgvN0f)j;MMIjYRZdtvIQP`Gm_;0d~*UI1Z&><;^C0ajj%
z$@ubK^@qkUeRJNGB-o%JAO=fJSs%|>2QPGY@`Z=VwD1h6ILX6it@>zK5a!LURr~)g
zCO=(u=M)Vuz)()lkjL|>_r{WFid`L|k#ny+?LFQf6%D}MQ-1JAJoToX&A`Vi>`&f(
zFDN$^O-*XM;Hso0Gy9<M;_SEDuij{meP7I!sYU)zxW&7!7j?1Z|4Vc7f}~b1zYyK}
zN0@f6D{H=Oi?qTUEAYGFq`%SaDc$q0T6UfS?gO7SH$SoOE?ftG*Or~)@&{0<Wsjsy
z4|>~=Aj+lIF$~d&Zq3*7v019<9@F8H40V6K(aCs96zn5U_q#Qh&Kxg<#?=06RH_sr
z{7W)P#L_<x87K8s_Z2nIiGJd?x9Tm6@92-e<SA@3-0u;@%nMq^v!J%9xmgm&NmFa5
zT8YL(aJHa5C{o7bh-Jquz_>cw@B1T0<Ur#ayRnpapywm*XzzOTW`G61;u)4~goMXi
zS%4yss!PKIs73o68q(8IkqU{xT^=RFc-MV{9j4>C4TIO0e2KztEu%)Lnw(0jEYWQ3
z(R5E43^<L*4JPk~1djD9(CSgg{E>^dN;7=8PKy6eRYBG~!**K-MZmh-jc(YJ4ZesX
zuynq^T#?<oH`A32Y0$lgF(SCa!Q))<I3A^^=>3UKk0)DexFf#&h^${NuDR+*>z>>p
zEo!6=gmrzu+e9h%D+A8%;0RA7C-Ue5LqR*V`<O$^VP|ni<J<0GvON*{`E~QqshSW3
ze4)71VpCb5P35uEDnbDsF(!h$0&mY|8xDTJd@Pbf4>>FAr&-tuzayrh2?wU>EZ+Dv
zj!tr*-~Ip-ui`nwbX?c0&(#~r2h-O0iobl;-k4l+ysM9T<QW~1?AgjP=KMk@XT?EL
zhwXu%7WQCLHNTZ*Vmg5`#^}Sg*NAYnZBuih@XLWyxU`i?=Ju>Zi!c(E<dT`sVfJWo
z>Pm*>{|?;ev!ry;#)&Mp5`|<DPbF`GD}Lu3Nc8Z|mKtgx?6EnkJBXh~93a&+(NHw_
zb_ZDC4bG1?Yp$=(Q4pYZDFsupk@4|CadC0k*571-Xxtw-wijVY5(6OsliL#&W+Wr#
zj*VOE=2k+uRN)4X%TGTPG5@f?IYQ1eqZ+$Nq*`^uyN<Bn1yyP0zl+FBbzdKPR)<^i
ztwga5>wg@m(=&WJ?mo;==RPOq$G-A8+wNJH6ECd&a>nv0iHWN%NoF2$l*PKuY#$V)
z!X;8i7dnvdD<+44nhOG<kjV3XZOuG5-UT#1#Gzf+_FzRh@VKC$Kmr<FI$o>td7Q9q
z)hrSq^>HFVGqc%C@aD~%Yt$DgKH=dFv&S`Go!Sq!kqPDgphV91lt}g-_3t<=y01$-
zLRK9plA^{tq*r<F&il}KVUb>T(*1WF9OH;KbJ6;p`QnWt@3QH!qeT(=kACS=s(F?V
z4Rvy>^*?zu&1OstPZQyN8Rulz%I`TPk3<?idJi)(q3czYC{#c&i9Be8*;0CbeoA1+
z>;fGah`t)2#nIB%hSt>7L^cj$x(rNAU4J~G>*v$Ip4j@4n1DEy;q^oHO7mN~-j|L5
z|AG9E7T~|<`H|zQP?&@i1|(PxR>9oNb$UYR!8G^?0;y)I3Q<eNnMKr~tt+gG+f<=V
zwv`zU^zPF!$-e_PyanIbXW3e%al$<uQY0C;p(|+Pf90!rA(Qg-pvqn)nGqrieFo~r
z-TgJbi63fLo61Pbv39P8v2Nf{7;Zj4Sp5KUa&+|cs5frh*xB3rw(kkBGzjLX6Pj3`
zJdpyy42gkvA0Ji%g8|S)bW|kU0?s~J$p}wXd$fM?a039i;ez{^k+tULvep+2JhpR-
zgE$06k95TSJO5S#SZ3<Nye8*VW5uuhNf2i|jk7#5rtjWqjrzFgAH{ah><Rb{O|6bY
zIpcvkN<|_~JB}(pj`~L4_DYl-HC{^Gr^C8HHT=e<{ymMdfd-ozu7`)Dl<s_x{dQaN
zVOaLV#4>P{ib_dgLZgAw%a{Ie3Lk85XK%k*bmu8kY|o#?MZ;ajY`W|b*yJApNG`x)
z?HnC-!YuR7>2k_5mmO1NdX@~m4LINJtm1wDWjJ?OBedn&-e%v2;k#+{yJCGYZ}A(s
zgGF^e)2+&xW;lF9*<GpNrrgwf;di~#{QccQ{z}OTVj-vN`(V+%FjTUHF&sBGQC7Zx
zERw1U@3=eT>mi;Z-E8h?GFfrHr(jZy4+@?zle_AvJUcZog0|0tVO6}+)C_`#Z6q59
zG!a0f4*%!RpLZq<!%+~xae?G{2Qgvp?(<P+8Tp*yH!yYKvR+u-LM~pc+ZYiMflMc#
zJwK;fT}Z*oTDmiBrjGqK=Q^cNffgqwJg!XsV}Sx!h0Yh;v<i`pBS*4ye!~BbeRr>g
z=^dlO7H3jdY=ub$x1KeQQMQmKS3lO?Fv=mZIFj<z89o?J2(4Tya!ye?*tn&#oFt+~
zuX_4)517v77prg)w?WYz$)5~X^vwEq3+1~86i_rq<`CwU6rh>i4&7*fwzmnPXQ$y{
zNcLlBs8tEx{Fz)H2pZu7OGA4kLx5k<a<V&t#aKb!_(&zQ(V9}EnL3umffDDV?pkD(
zJap*ET~w)=^-w>FT}NZXPA6$eH<_dl7=|r&4;ctgSG`-H*7)rUY?RG=I-m_Zu~z;7
z%{)6*tMkCbu&8#?Rl#nlLhvYi4;=r~(CH_Lr<lqhuY`?*)7p4E8w@gB=olC%2=MS|
z$;-R&(9_DwZZb%Pv`Eu<j_7aejX$h}MKPSJFoPa0ADOIH0ITd?4g1s_6raoWMOb*y
zX?YjN;}i6F)E!;>y3<?`L*9Scy4&7OQD?cQI^Q7G+U~4$i(wl~{*C4}EpRgWdpV%}
zoagQ#*Zi$3>?9Crp9z38M2LU7gLl@Vru0`lKU8&(e!a4xZ?7_rtSh!SS-9uoOcV6&
z+ang8I#TF~7eb_ZUtb}mf6_BDC<`~kM#79xK6AKQ1c}Ph?Qi?E+}IY&#wq)fX0a`3
zReg06-Gsf6Tz!YrPWan()tkdfuinuV-d}8Y7Vz_)xRXN-&Wxhswhb=*Q5zgv0?JBM
z$dhb}{Z!E7Z2$x=C!kS*f<QKke0!Ns8#~_oB!x2!l_<|*5`AYuwbkT8su@QzoF`CJ
zK?gNrs)@|P)Jl&Zr)bSRC1H9wW2K8wV*4E_9D^sY)K@>z)KroIvrpwkZJD;lsi}RD
zq;fv6B^bwRe?KTV7zCOFp}EiyxOe1-q)ZvjLvVL+Nya}Ft=j|90Q<xC=d(?L!<E{O
z8V7SqjB;!CDM)9H;Vsg`sw-!)1$8O%nImKU_zxKzWbX&8+7<=~7Ctd0ivS7#R@1q4
zW>yxL*$z{d1b$b&Xf{EWd-(qiIeIiqW+<wSd+e&^a-k)I9E}JZ(9V|7$o&&T-@8y^
zQBJHwR$7#GTy%y!lL~tZ*6aGFD!u!ET<|OH8o+v0<5Qp%Ft{^lz8)kvR!sB*nKU#+
ziP4Xn1?r`!TU%T5L-$|Hj!fYGM&AjmHa0|6X-tvwTY9#mkN;|{eKt2M;nDM<%6`;e
zbunVL#y!tuGGW5bjz5jK#f^<-@N8f=f5m%T1mt*Vkai1n_$7A#e%{BuLX2ngy`L)3
z->3fp=~jT5-unE&$HlaNA0immZ=}9Kr72aGh0|E3*LI(!W;Z+)9v(;X*{z(<Q~mD)
zw4jcdT%q|GDXRf;8e;jQ2GCgZTvoR2BQZmK!{Z_aM~ZPy3Uc2JyMDzX!ACEXD2&E2
zmos7vGGp0Fk9o6xORquAyPy4YKLenDO&lD=`X87&tXVIrFEQ>R+Ytzd5Xb<8CvV=o
z=^IL}lXKK8Ha#%#|GYK0>VJsOvBu~aAN^XBjLOMzx}?rh_|<r%I6TeFtPCz6#}@K7
z&@bHC>tW3R3<^Xx6HT7{pbkD&u=#Dhh;V8B&h#yl#cpS+pvr}ecTj?qkd)O*Sq>Rd
z?s57u8`xXy*(eiN6aPt0G79#7#Vb-oU2aVc`GDXtmDASm8weyDI*^l@PkqYd9c9vI
zFETyAWP9~Bl$AS$$TNj~(M@I+GOiyPF<zULnSP)0vy|XO?fvA9$0nwP$cv;?qI*kx
z6vUAX&0o-yBn@Q8MjUsL6HlNVix8MAGZ~^Z7!QQ=l}+m&ZWC?AgRNnk!!7NH1y;qD
za;mquv*~~Smosgm5BbJjv#CA01!PcxFw2%+S4RP)?d>3HPE1KT{kxBJln5pGu*x!0
z83mtv&s4#~op0A_D)Cl59OsEy--ZGxZ6}S?t5;+|dlC$|0tEqHDi9=^Nxi&=%8Wlf
zEeg>bGu6}k@7cnlW3XSbfeg7V@Ey6hxFChw<G3uycuFZ;Gj%gVG`UBvTHuYJ%Zj?A
zREvxk>4<|gjnn3oKX8R`IooS;J-k7IUn$H+2U$amGU7oi4;!CJFS!R4e|L6wMWNA(
zvpR@$u3N$^dS7(im-6ec${e30<jK8h2y48ZN=&woq&9>AYCeazTP8%vs|xMz?Uhhd
zQ~T0z8@g}MMO#8CpQWng#as62!S3aASn&b%LGDQkpK1KVNc$1#eq|b7Y0rE3hhLCC
zm_4OHcC-2W`vaGi)#u~Phx)_<0}KKJ4H~o4U0s1*FA`IQ?845vdDj@0bHg*!ypTWC
zW*0F3E_;OZfRIE=!IV4Pkl1rdDE;8l$M!&D_B)X$vCo`UHCg`qZ;2Vm_jpdR#su#H
z$CE<o_mJ1_7r`AMg#Z&H5}P8&`9FM<00NXfEw+gruYmzdu8aB6BQ}dM7N7zW$x~Y>
z?Yb*&ZEek2RkM3}u|GgA==QX=wH359kG5tkY!lf7K;N1gj2|#s>&~Ax+ua4pV0+L3
zVsqUyf|jWk074LHC1BE!#4D(CxgaH5F~xe{GL(vUXd!Xzw<199b{*##jA$8uSiwJo
z(lnsCQrE$&;?<0lKM17pA_t+}x8p_Bpka=lq%a;ulQsB!H3Ry#rv#)yQK_MWx3D%G
zp%0DdK&=v<XRTvt$%14L8{VzDqC)9E|Cr1m3t~m6UAA%j&PA)0OI~46PQ(O3V+kPS
zT3rWKqHIH$iERTB@(<A2eiOQPK>`SKVR12`v@%pNe>5Emqe#8O&AZ;qNzO=VX=L9a
zWCqBhF1R8kW#u<~rXY96<Ge+6lY|60d3OO13t72JSYvp4!}Jmz0m20pE>*yN^2WA-
z>sS>!hE4k7{UN^JyMG_Kp+FpzwnDRTmCCeP&#mz;l)2cxefzevUfgT|1Ie&NxB$-&
zmrkDJE6C8Gv~Ugx817Ov?)UR&8x}T(BWMK#sNgPJK`8-A+4UQANs#P{utl+ak{McD
zX;2QR?p$7-Z=akjh7ZF0VeojuC9gr|i$<`c0=Pk^@LRWTu{%3EAN`Bfi`T8U`Gbnq
z5_VT`O5(AlE@A@oD&*#HeOmy|P5?Q)g<D~6?#Cbh?5aS{r<2?edL2y>6A=8Z`IbTN
zz?7=1bbY+|nZ|U&XGdZ7PRQwoCqTqfve?Yx(gFHzF@3T!Gub*bFTFgEXPiTpU1Pp{
zG2}CaMeK$B@bc9w?oF(>sAHb(<c8t_cu3(P;>PU4!^p^#6q;#B5Cx?P$j?8ryM=|w
z$jGQel~BXLAYC%0!f*Mb6e8g}B*VmHWC5TUESD^B2XY-01hNMfT1Cl)JZe^dL(-WB
z{4+dm2XF1Usx^#_jp-mON>>~)=fcFNFk1Tp#80hYx@eeqY)PCRCpHi&;PqUt2??q&
zD}hwjggk4(|F*7(msiuOy4E^Vs1zDv=lB?lcJp{0WCB+J^g>BZemR7k9krvQgN%W)
z(BF~f4)na^LuWlu>)%FF{Fjw+43}1oNyCam@z(EEz9d+$ptO_vaU`*zbW4Fo`3@MK
zmH?|r&&f9NCdTh}xIVwfT2-k~^aV+k&b@y<NJT<8n<fJZs-2ykl1XfkvL%Nf`iAFC
zOfG&eH6uE=L%Zzk?0yyL@CPq~BJS1ImDRU7B*7I(&`3c0FefKTanso&;3M)!-g#eN
zq5R<&T5Q-bPqu28?;-_R5%(^={{8(s;W-x4rAgxM34+8Rf+Zn=4(`C_?7)mgw}Bk^
zizXV~pZQb`OPls5io&ftdh`eqvQpE5y9X<SXy647s78s_8NehFNPq~m|DQmlXJ$!V
z7H$v2N8S>&YO}qMSBJ4aVB#a+9Wv|rfB!~99y3(e1Fd@86{s1$JAmH<VVW2$#lpfu
ze5tYgI{?T$?neb3XPO2t3c`EG{lGs_f?ooOVsVIke2!})BYl?Kfb1nTH7TA_;5VJE
zF(Z}TVYm0q6p%?4guB5dE~R9EBRA+z<b5tD2NYDM_RqSW#G_i9K=zEJYU1GFAi8<e
z7r-v*ty>z2UlJ06A;noT)92*$vR?0furCDp>URK%kHgj-_((}fD<-|re83LTF)@MX
z2u);qHD|Ki)DT21PoOYD1cYU7IzP8xKY#vl#<b6+pb{RAJ1j9ef^0xpYlQ(6UnQia
zGXPJk*V-|j9c>1KpMa;wY2az3sO27pELK2GC0(!CE8>R{`~}L^?_i#EAel`?^?PFg
z9%^3!G*<(0G{|j9!sj9>=Q1<z4euTtIB3_}br|SKYTX2NE})?c7JB&PNoT2Hmz?L#
z|EF`j)uA7pEiJb1sxY6`AgB~Zq@;v{CMKe=IfdL1*?N7nG5G;F=jdQ+W}zl(u6SZX
z9aOtJ>6H}V8HR_4o0+dWEfs`sypWYeLj*79avevKixMemXuKGRMC%kP%PAC;hM>%2
zzl1k5G$f}7yRHQ!<K(ZckCt<1&F-}`;*zoG|M>BPg_$|{5&)Ow$<7}ZHnxbF2G=j(
zTtH5C0JLPgVT7q2B-#3b2~NY((vq{f&NezKsss+5OW_|oVgphI(;S5)1MjO-Cc}_Y
z;0<Kg)6=VZHz(u-7R>Fmp-y(^&X1oxJ@UuM4)c2Ou9NUICV;vN>0>5=6(V7sfeT0a
zQcXprmF$UjH#o?jWo2a(Fg~x*>~;B{Cz6!(U%gvco1*LT2P6Acz6Mxn4~QRmMn^|~
z?YBKT&l+A;htzoD1EEki-0uY7WpyPbBCR?HGkcFr%;kb`u(Ap0#h!psBn?1K)79^2
zf`ft#z*DBcJ2!@t8(RMO%*evX=ntN&6j%+*8D(SG!WJjOg|E8FGnPNHcX>(-M0$>W
z9qhZLf;xXYP&W?1O!!w&cea>o5~Kue(4OvYHfmN@R#O-Pqu<}(H-tC~!Zo%m{QUKI
zw0@s>ovkuqwk%mdY8k>V{xA;C2}3ckusCbA`(e|mBcrt7-u^!Fp?p=zGXlXs%g&HH
z1wiz0M@W9rc0>`bZR0x@jghFRXz$ESlB9%0iD|NHXSxT%u&_R5;}9QF<+^V&j12Wp
z^qP9X`E)ciG^m5zw-#`E<rfuMFDMT4Bpwb%GryjIFwq5F>0h;<m<|pOT7saIT*jdK
z?;NgUV#AQj!(wk^Vq)Unwzdy2%uxz*ZK^w6<bu*6yju)zVH@QHh=g<A1#-iJtHqCi
zgRtrTI*D5RAE40x2TJ<?{FPNJxTIaEKE*fD04poN-08oY&lMCfyKK1=TApP>HSWVA
z&+t|Le@8qbV$aLJ3VHJJgu&qsxNM}dB2t$E(9Oy=Fr!#sEYe&@I9RKqTLFTNGU}T_
zh&`X?20aza*@in{Mo0xJBu1;LuSc>>a4)}l`4W!cGDK&`=jNJ4x1c6i%|L;2&+jQS
zNXEm)mXMH`hhWtL2yGW8!_ab*syz4mo*`o-#4%I^$ZQ8OrFh%~F9$KK3nDd=8e~}T
z&`od&rWjz8034i;ey#iduxfcswFRz0y}<>EdlB#fD=Tk3&Q^13JWe^_h#M2Jv%e2F
z76*5T5FtV@(5}5185s!xQEaZsb9eN$o<_rf^PFcO)EQfPdT>E=zWi;DAW{?yjujv4
z!oo{|4G3Ld*dt&>0t-1S5ic~y5H8>gRj5IC>&Y?<WG%CJ6Q=Ir9H0TYf05E*$;rtd
zgMwNiA|@C;jUgI3fKE%liU+#7y1C!Jad@UgMfrit1Gr@XUNE<)s1u+YV0fF8oi`WY
zRFob?!=}$%tpl!0Lr@4sjtbiR?Is2%t`9`71ar=7BYE&-u#mDkFd;d=7%m*3l84mp
z-ATE=z(PbMCx>lpSbFSyhn6R>*#+=RUGP9`wtn;WCOjil6g(t_@izoH>L`R=q(~CP
zLl<~(rI3gRZX%ignogG}qhey>b<d$sMd{V60Nce*6a=KeunOUj#4!LpqQHP5>{251
z?;W#6j&SXuF?9>@(RZkg?m`BRymS{>+jS2`X!GcV+~q5TxzJE*{uuVoI9hBLcJ@dl
zZnOC#HoQAOum4Iz!>n#8n6*hCVxzGyZJcg?L4grGv<<-C`XTIO6cj&aU6-kcccCaw
zhKyINo#d3j;GMu}N&rq29x0{|<Nf<nR?CpkwIfPE_hZ-suw?J_bi(D?dNJ8DMAI+w
zi9C_us)w><HNx)-vV<W^VN1G8-}3V{V6o2uEz1`pUlU3rJiB!o!X6VF+l45B)Pb#&
zyyxdn1H=SW0LTqPtY#a&=8c*MTR_lh>%B&A@8~G4o^|V4NCY?L>mLvh)Wk~n_ec`C
zyZg*xp<!8JtpfneJT<!s)m0O|SwIlMj^5rEyIA08!2(#eRF#$AgK@!_hmol%a-K5V
z5MUC-p^~qd+%mGV9}*HM1Mu$9*UrF@(gSAZHZ(#3{av@0FK=x%>?<HzASq^<LP3<4
zmJ)0;fBCTkOqB+}rWwa!j01}Y$Hsw4{Lb<HwJy8#$RayZxeanHj?fF8T00Erv-x~+
zr}q*T?*y<ni`2z|n;-xuSbxlS@3#47Vu^-0lo8Ld&0fJ?L5?~8`Sa&JaCmkb#awHs
zJRl{Ti;am%N7me`tP=VfL8*Lg1QR@aY%#mSpckA{yq#-m>|n3e*3EGumj^%$kQ#|m
zBIDw);bp)CGvwd_;OFDVk1ddxgIB;5CHj^Bnl(7j-d<o}8oe2SGk1z9XpJLv-fL%v
zhs$+S0*Qw9l>WOa9Dh{s$)5+d$BSu@X*m4*-qqE$#S8e0$7zEcyp<5bM<J&Tl9OM>
zdcyU~a0Eo@ON*4fgX89EUTdoNBkPCR<AA_Wz@I=)_%V#iX?u1i>X$)OeEix0=PdaD
zm0xLS$hSw*ZYT=uqW_+mK|w&R^tqy97qaRJFc{GSp_*!z^}z#dMn=Y8=9Sq#Q*~1M
z5Sd|fZVe6i64{+2v*K<mV@TJq@$lM_t`0Vl5o7=`1`O#R2RKAbL>w6Rt3pkG@Wu<x
z*O#oi9=no(To+e+vCaAr5Kq?iu7lur@mpUbA`l38aELBjqyp~V-ky@CCJtn?U=jwi
z4XzV)j?5gKoXBp80<CHSP+f*ouIr2a8ZZUOXptTX$y5vi%8>`WO!D!5_>|m83SN+d
z<Zj%8f+Ms&c<lXPLUvsM3}f@Y@(4VJVQiv?#>QRfssrdgY|W1}Dx^@NdXon)q<n$k
zp7a5*oSe^HKgnzKi<p+PXB7iysI&Bfwuow=zuyD^kul^Ua!+E*0i7Q8at^SCO0|)5
z-cx}BS>Wl(2^%<lZum5AAjcWJN7PBq3{U~mWJ6d~WSo_hmIl43^Iiv!j-XpC_wM=G
zQguN+tP|d%=wj~rl4gCXlG)zb*%ZRRW&1~BKr0;%GmvAqK~)$T#-X9YB*pXSYui1d
z43IJ8M6l4)x6=RdXYYg*S_3Qu`6AD=gEs*D@DS7Gri6yQ+&BE##fPLJUhePh8H5Y$
zVV}cRi#ts50ZsA}$V(G|FPt6f`P?@1B1b=%Qg|$xz#Fu)-!FR<|6ehrLgpz3bU#2`
h?L->y|DV3O5SGs#%vB90YeB%D7m^ARMPmBC{}=VQOkDr~

literal 54864
zcmc$`Wn5Ni)Gxddq+6sL6$$B<ZUqBG1?dK93F%HL5fKpS5Cx<Jq&uWhq(o9eKuV?K
ztnJJb=Y5}ZzMc>B8-~&Q-g{rywbsAZj<|hOg$SP(A3+eJYpRNx2y(_BK`>@;vEetu
zLzCa(U*axG+Aen-EL<KNKQc#d7`r&yI=I+cn=reXKXSHquovVJ;t}L#wsLWCbe7=d
zwfoOMz~k`9lD9WaaR5F9&rwy!89_*l(SI=VWnNn&$d~$SigH@+uhzajcE4kOh_ktI
zZTGswWxLEMN>Xc%aw`JPJ+3?E=X2J!1hC|E4$eHW)yWFS56lRc&CK@?XU*1B%vkHD
z$bFvpMc%ST<x&4e?6ddR1aCfz|1y=ny5c(i{rf9+Yre;6OVt}GvtlurEWxDmxNig%
zM;MX6KbCUiBY5afE7bo_en=rk?!VK+3pM65?I2vR%gxPAl5pibIy##9@gqVm2zz#G
zzLPVy_VEUzfzO}2A8Ko}eUA^^cfUPmiT}tg=_!v3*Z$S49Ajqx*HWqV=S%OcKPxZn
zBnw*3Ev5T)>`wc|o@2i`GnB0o9vyvmiZ`)mYo(w#tFDgz^{w=dO1rO{KdYuqngh?i
z@Wn@;42wtvJf-p-=Hc^{lvxU4#A7AKm@lsAhPwTnnEiZ9CPa%fq_I(EZhoHMebWeO
zn)W-5zE|{Mwl`TY{K=D+Oogz`37g_s_&!`*Tm-S5{6OZs*ehpYap_fRYSI2Mdb#q2
z{X-L&gJvtkZ`}WUxj$6z^_W|?Qr+XQJ(5a8LnC1QTP>G^*Uq@*&<7_|bHshImqt-h
z@p6z1vizlhDo;H*aCNje+vjkHwpi!hz0k5UfrSn39#(dC9A95wBkB&qfoJHQV9kgT
zpFMllYA^#sk+p}re?kxW;I(V7SL1@WG>~5GWT>o+4Hq^0^Jmc`FIibcKtSMmrxwWx
z#V*(SXN&eH=e6tHOCz$`(Z|Bye|)%0Mo(`WXEUyaxSyW*sHaOasKhdc6ciLF6-zlx
zN%p=JR@<-D;#8XNh<1WM5Z-PQ854qDNX1upEPUqO6yFyYTRwk27ZMV(xxHOB_Q#GB
zsdL|AO_THpBxaDf>=56xus0tov)N8%II}t1dYPA3#lRq~n|o%yGnR~+8W+JOBy6j8
zUef9>=pw&IpB)hyNuZ~v_i&PjGLi$i02dn0+}+c||9D-mP`{QEA*G>-xT0H`p<nAp
z`R$R#9i1kAe*O=he>gQYHJ{7G#l=zcn*?t+roMXhG$|?HySGTU%5h^#N>N1x6UoZX
zcKWk&SJGo!(W4iZlZ7&pjFQq~Ye^AZ!ZXO5H*c0d<wgqZrHDIUx|u2w^7JWbWNdq$
z1yWyEH(29plkiniQgUW#>2;mY(cV^)unncRj}Lsx)`+eHi-ZI%tTf`ECgsJ-%}toC
z9Nn_=r2wud0JgCyYz$h<y}w(f*KQz9DoM<Vb^F(^+11rZhsI+{SoyQJb>_!D*>WP^
zO@B^QkVYs{3t5qNcXtmK+~>6&E0ObP3<|;|W|R!)xRb{WU&hGDNG)hV_<~!<q+R12
ztVs?JCkID!p-y>;)4Vbwo5-VIVl&Dk^l-TA?R?os3I9rF85t(>2rSJvTC0q*FU&Oz
z4D4s)$h?-mf5*bX!LcC;f`^@ae7rjy9vRu1z;(Zgk8rUsB@DK3>yU=)N%XB=7NOb{
zjQ8gk78X?1)G&L>Mi}u+6{e=9U=ed@<AX^x)qLO>CkdG2KZ%UgX!P}28!N?>^88)0
zQ}X;WB6OzHN%hvPFt`s|;fH3WS<!7LPV`}Ob0+4`G;iO2nxFy;p{NRbDM`eR77j8t
zve$TeD6=+MeaG@5xlIN#h%1mx*xufb9)QVj--@K@O)MuTClTaUnxx$E!5?e!BvSc@
z5nT7*G4t>c`<)((DzcJM@)|^W{8}9Ue!9eg2nFi#;2<kp>QQ-nbGJ1#jN?yDnS)7N
zm^z*Ipo;MM9JoABNXRs94GoWt?NsBfH~X1st%Q+*XLU-+Pg(yM9<$ri>$yon*5o^D
z<Cuv1Lc;wLyRRa{1@{NK**-^$6qCVH1XItpgb*wbW@4g$NJT{jRs~z;^r*ZK#}7H{
z-==sYMWp?8y6?lE#z{(5_TSJ#`Oou@o)sfb_dVc{c)T8U_3?U!_2<03Ujs7yUOTq6
z*9a?OHOS-n3A2A$4rM7Rr|ur@|AH%dRuXg(4vMYT&6~v7gw*~}DYLS&&Xksxe(>HG
zNb@`Kup#MVevdQwMC@M7GhKF;_`FsPa;&Nsyn$rN>@3VuQu_PTczAf)B}N$b-)eH}
z`aXXyf2c=~Q>|x^eLMHk$;n9z@pCx#r5dkqrnZzAw>m;G5W9L6-KenGPxRMU^Jrwb
zI6jn6S(htp8Qmom5Q%9Ey34Pc!tf-Jz^m_3eN&U{LaZVyMk`<U^y$f=gS&^v>|mzC
z@>nSw9QREq?5x7V=b<90K-rR%k{Wt*hn@baoBBfCidqw4t5MI3E7W3+%&_Qi=LGa(
zj<QQ*BDQ1xlvf@!efe@80|SFv(!FH6VcPeQh_`O@OkI8b?9x)$sDT#|9P94Gjj4Kq
zP~#HI0miz$xkyK6X9c*G%>kLy8F+@_QBlFKU$Y<l{t?vMtG+Pb5yKFPfr)LjHP^nm
zyBns>-SPcnysMiV2NzfC)s4EKP$IgP)>cgT+mg?oJ$t@;y5V=KFx}uYJ3o&}LP8Q8
z7B)X>==Wv!HIx;TdF>Zac)12EZ0=7s`Y~O3@V;rh+;VF%#bLsH6X!N!b^a{X4(o9w
z%hLtv+tvCn4oQC)67uVk<_}C(JC{CMxIS6uan5eyy_~Tzi-E_&`3sjW&22CAEPtyN
zJ=pmf>9xE5LvmoQJ(Az7i&U@r(b?j<ZG7b@>VQ|T=%YIgJ~(1PeG!2w5vW^bKU+2J
zd*@mTFDGXU>`2Gimb36_@)yXBXPN?pB}%yUK9K%e8tBYbOPpC=4xeoen{PZl<{c^0
z4U1>LHMjHgD{PT~Zd3Ae=bU~m-nek_qVecE{ju7|yl}<}|9HTzCbH~*)ePOvX`%ak
z-O0f!dYLE3yM{>9@S8gt(As29_Au`k8??W@TcBle4`9kzdS@(C_N{bj_l^2p=ENR;
zlQvwvI(IVITje{0>aAqaIhq_e{_)pyh1a<_IA-l9KQt#ie@^XfaF1tR(^6>BKAOyW
z-)TnEQe;-UT0Q)A<`7)4c!Xls(ed8aJUwE6dUEW^M90Y3?{l)h)IFUdWZhDvUwd!=
zDq??fwBJ2VKtLdlOhQ8wO55mU^84Oq!k?I)K47z!oSaOLtV0#2?{w08@PHP_|7)3f
z<aQD?A1CBPTbts79oN0WnFp1&Z;pI=llWg17xUiN(#om(*~8oT9d@yJwO+pyeW&Zs
ziG&nsABj#U`dxUSvCfW;GmO&S`2uN4N!0dJbsp&FD=K1DSq-ro475JqC54UCqpGg{
z{R}3iIPxkbg>JV`blQ=-2j7=LpFga6kT4@8VNycJbn(`C9G!lj=aUlFrTiR<%F1t#
z479bwA5VGU6g{X2tf~@)t3cPPUeF)*kj6<>SQt*5Nom=i<QhgP&x~vFmvD)Q+VeHC
zOPrT(QczKGZJYwY;<p{+3l0sH=cdet0)*T)CHzuMLrRJu;o;%8t8e%a+I_9O`{O$~
z14DF6i#%-W#tyM!gL?dGy%@_h5A3wGwC;4kXqOXn;5^y^5+c;EacPE<h;ED2;&X6&
zUoXUeEHP?6<FYZuKtn@=OG45aLO{i5KPgtLzkOE)BO~Nv&W{5Uff+Ft4|n(cqm%xt
z8%A)(v>Uv|Z>C8`Jbz9B2x@tf9S6C2^CoAbAmJU;kN28;dI+HLwY<}>C8eMUK1)EE
z0hij<(}Rip{`r+kTwHvt&O@+2O)9e7a^Q??G{073-W=2G1G|ZiCKWhf^z`(OTXRY|
zIXRd8U|}L$S4S|AOX^AfP>YR*^EFVU!OF(AwZD{3=W_!p{fphrS%tQ?HlrV(asiy3
zyQJO(XoeJE#m`K<Cud{{oS$RidA<@)m2ky3?Tl$}3cw^IBMX3XI9cy?e{)q#LLv)}
zuy(zt@H@TgKmf&a@cdQW+(e)fhX8V5yL1T~mQvo{{+-e2@oI77ZB5OrBPnrl8hFZ9
za7odoU#lEQ;K{S!O3Q#g$0_rL&Rg{1FfJU~j;VStv~}1{)#;RMkpt)}HuO6_4@aw&
z$?pi`)~#D@ot>t?7W+60lq$|91eEeYQ)n4tr!JceL&S(w_#A%|h3fWDU6FcE`{2D*
zK)u&)mcw*|5j8fG_h!Jga;um0lJ2CB*C(5yQ(`eOF&!PO78~hM0zwOpjU@rdzxn09
zA<N~<xV!7$TA`kut#$M85bua(iiKO7hdQDC!I35JMoM03_>$m&WDpsfveED5p*_Hi
zLk&J&W)>E#{QP9FbDIl_8=lYzT4bPwVlmnBX-Y~YK*;R71@{7@69Kq*XlZH9)bCEU
zz7)1Ww0wWuQdO0Wq`XonDex<kMwY--k$97-^xjs=K|O+;&ZZ|7F=ntx?|(nHx3`z!
zywtz^wTjpC&q@e@R$Naa2M33>Z?#mXyMCu(Kn08e|BhAIP(jgWA&)3Q5jPYlfR-J=
z5w?a@nDE?w!byH_HFSoz{x|vD+}vQXp-k%8QfTc@0BD*53a}b4yNEVtsIS>h^Bo8R
zEB5Bj>nykdv___REYhGT@9!#+O?|6<yfqie%Nf*IHRVYFHTBJxW+sG>zSH_;=-Xca
z99iO2nG>!9_Po2!b>)f@6eoFk`I(iKhr{~>JZnme{mGnoEHMqyGcrf#+?6_B3b|!X
zkCmFnB_@VJlhs`~kgp2^oRVY#jgeZ$S8{uMd*<iQXorS<Y)l;D_j_V4?evTCYEAd%
z&eISK5FcL|!9z&{OTu+EwXOBq4Y;@l1W{B{GKQ74`jiuqpkg6Tfjl5%d*ts`U_Zi$
z4QE`_^5}4*5k->=J&72|>G6hNsqJ_f?RoBpSMiZ+slS1!J@CSX59a2UmS%LQ+rq{t
zAkZi_kn#o?x%~NcY(<#)`R7<DqI&wE=)(tU^n8VrvE7Cm2S=Rs;>Fy4Y!p$U*Y>9U
zond1P;PzQKCD^Te=+=TxHB@4(0Ql#v6&_$iU@pXf^q~1}0%p(5$-xQ;2pFq&W<S|D
zJsFr#D958f;Fz%F+~c@>xe2;H850u;vOHeSUg<C$bMR+X*Tbwb=!HK@uQhG-8oX&3
zBwS>@y(OV=lt8bhmv9LWlo1xfR7IhFEfoBbV#8SI_t`*CqiKa(V9O6xJ6pm9T^b@}
ze)EeuwDf@rdR=)veSPg2azTHu3;r;mS%vG@@z456+(xuj_yF^%+$YhKUdltBz-|}5
zz|M}+#jz>}rdO|CmDqoiSetHSYG`N}VSITb`=Sw-i4L*#<FdGPL+sMAvn-juq!HBR
zdR+e#)cedmio(a<0COGhqpi-rZ8-iTDJ|`j;9x9hDG$7?+yX3+qVFpSE`{39!f&j`
zm)abf+dfE-I<H3V^yhlb2@V|#AK&eLwm8)$zNl_HsA9?m_ljUUKCw*qp`;VBy)9h{
z_w+gMhQjUJq<Z@LLpj&Z6>68BK?dJj-$9z7DmL>)Z9jIc>Bm6Qy|?jLSy^XhX9JX@
zX=kDFDH<DRZl9n$3Al!s`zAAVT)fAR9}_Z4QUe<ptF)tgDPo5Y^{4K5=j+p+h3-ya
zy9o-lM4om4tLO&-Vl#m!Dq&b~xVvH6^^|3EXD1Z8^}+52DnxK@+`4|f#d$yq2jH8#
zmsd9Y)jhzeox8W>a9P?E0SLJNtQg&dUr_-AadL9{;PQj~tRFT|68v`t*E)BKs8|wS
zyq)Apcwurl5@|tjreYj+6(zT}KZ(o{si?zL8(_ArW*jD?ua$O)|A6!%Ar3Kp6Ocv6
z)sZVG!=YJDmGNUlN?cd(<-C5)<#+$}Dm}tX9<e!YnLZ0U54}s@)8ofS03q&*t2jCF
zOCPQYu1(d`H6HKK0S@JJoRLNA@7|wL!&09^H`|H#crr(uLEZDx2g_{l?Y>eb@)~wP
z5#|SyLoem2H7~}+$%)cS6pX`O7js>qK%sZ-V~6L7i3*qEjhp;2mM1E>(9R@i*>ChI
z=b91quii;GAsVi8ZFui(Mv23S83GFmxEc;txYi~r2*@Ma;1qGa$Feo%?wM_xT9t?u
zFWHc2(ZzXbC$9YsAi&mECli~92vB))@<2LKdn4EtD3OJe5e8^}CWOiWY2`yp%62I*
z4+Mc?jD?N;us-yiGtv}ABl!IE8*CaBR6vKjmmE=;pWl>d=o?|t_YwoaCSi>F`L${u
zB&N;ZzZJE$sSSMh&FN))>0!~jJ6(T#It$hQEF$@<R}j75k$3u$0EC9AE#>9U08$5C
z=(*fC0Fa=eMCasOfc-DenJ`x|YH%k8Z8isnW^h<9#euA`*-BFCzGYIv9f@voubnkH
zMaAHXFpESh?7@~#>t>@#`*+y;xRfSRZ6l*z=jY=B5kX5WJmLfy;4=yjA95l_z>rn1
zU6X_V-rD)<Fy9r&hHa=aEgr*gH56KkknI?$fq?<F@I#8S@^TG5y+X;3Z{PH3TrgyT
zC&8wFTjs~l%#8A14e`e=E-um24pVh+oL>1I|Dm#<ZXn|2<z28rRfbQWG;{LuT1H2y
z0EQ?5SR1TzFf7!qx@qor5lYo;?(5gtApW356CMM!3MXI(_$TVJfW#J-5&^@!p*UO<
zahQq%<h3wRVKe&l)hj{-<+z_deJXMJq51szb0aueC=Zhx9deC9u?##ng%lCHO^|@*
zfTMK#c!GA&?JOoP{^nJ@F?DoRDYwm7<n0>(y(HY$^0yxV@|sEaJ81^32t%tMH%;6b
zJN4>gWmp`vLoV1TfkL3I5<@Qs$H?hu&&l20eWrG!p%q=R*;(VCnd8C_&Q$W1jwub!
ztDh$&^~WZno#{`#ipoMUTj69>RAC=Jh@pi8u#7SgW>9cuzkknp=5D+!&SOxQ-@8a2
z!PywUHeRk$vXsyG>LNS4jeg}%xJ`0$^4lP3fCyH!WINp;HBsdd3wR3^ezvx@Kq@4+
z?S-x|dRI%!2-HXI>PHum44_0~<(4E+u&fv2z7Ru))K<-Bx<GzEwWo>P;eK%NfsU<u
zBo%MRSBFNT2*p4EY}ti{gj76wvi9~ou%$z8N$+FH$;ny4H-ebaaC+her=jKRS322+
zw0htmF0iF}cz8Cqw%#}q=`UGL)roU*a&7`b%?7B3?r{<O$;)HEuEJ0EiVdQH*v>2u
zW$&FFZfpTfv66Ia3n$xK&W;^y^z-A^DZ^@RZboZg7(Adkl>VWg1}HAmu+bOQT~J&K
z$_f?i7E0SX^o9YfkhF#pwf^}08UwkLr``-CicnNklq<GyC*z_K!MO<P-;qS49_oz&
zvy^HGhCkl9ODLfqt8wLmLXR?I@57%}xOjLN2$Uk@KPz8=kt+g0cidgqM@w4wculO#
z2{r6bCD7Fd0U1V*4!XvGL^KzFOS&1jBr4CLRWw&I>W#%fS~LL_PX_cy?Mho36y^P1
z`Jy&oy1u@Sijjk$!M1gDpwJ6FkAP?ZN6oIU$26Xv&;?-=wV+58n2|B6=FUfpV$t!k
z1P_}-y_T(Y`{{V_+Zy(O_v!I2DkPuvJ7X#qGhZl$tNT1Xog_l>%p0u-rW_^j-&0sy
zTf;pCfSjta)aZGU0cXeHelp&3_u8!KCP&5hcmmdIIoF~3@q_x4z-_9_+~ySy@>M*j
zzkn-oR!5335to%=JhYu!S>*%5@~tr;cFu(E9$|GZcylCxa`i7Z8hblDtku@uJ_|<-
z@z*K0Fb2iJ(ao)LyXzq@K2oH5!DO(|L+Y=>gj@Pj%4^4{B^dAW<;(ui0^uQsf=tpk
z(fan!W#mG|gFlq^00(SUu3f{k=zkRfI~W5oaGyO3R3Td<TlpTN&Nta>pqYUll%FE%
zFb6xW6CgaOg;Rc~K2U031<DYVRU@0QyA<K@GgC&QN-J<h(B3&VHb#o*s6#dOlO;XM
zy2a6|f8WsOJS<JVVWaev&#r!#nSh$id8o3%`Qy;@fNr6x?QpRnQy2+TN11ug%(80a
zvlnL0#cqy{eD~hzML^r|pBJ<JT1f=hwiB=>8KZ%YwE0|Hgm#%(#5O;Cc=zdxo&{8z
z2hlnF^Xpq|Qc?`CgG1xv&%NL^<XP6{2zW`|^V}YgkV!0nP=41HEfj<z{!oRXMk`n{
zd6-CxP$3|K++Wy5o4TRz{>5kU@d+3iwY9Y)qP=l!Dvt9VM4(y`f*fTv`i|iRkNy*J
z=S3rstOCs_fP4lRQv;=GnVx1MR7kE3W{R1=ZD0@snp3zUYquj?@Gu@)w}C&Ppa6b{
z9!F?bAwfYw6=C-*$Mq2HR6U+f)B?%MVg*r~%VE0O`SrFgsy?CU;<6>iz$@{f@Nir#
zEUe418~<f4*j&-F#RZucq~7`FAe`rBvQcxe<vXJ3xFW63$*bD6Inj&QO$4JH3knbs
zok$xX>`lP=fd5?6ZeaU^YNO=hA_T`}7JM8yTr41SBYfU_jzZR-u!v}dnm{w=v+Spz
ztZ^j(PJ0$HGc(J_@<X0CTAG<z)m!N3(14!(ym4u1X_Y=31UHmYK6!$loyPX2763G8
zPO5^PuDCsqw0^}8i)+5201crHZk7Z=S8fZ~LY9FPr=JCZtpEZ%;v#7NK!UbfNo0&b
zeGzk-!)j@1;j{j9VZjMVHZDHCyt+D}=AAo3g*t-S#nmS5_}m6{V;P-dVq$~8g;^BY
zqxJnRyj!MH)Av*T>3VYH0cVKi$`zuRm>BJfhvzMNljI5*McOW%lK1!bqlFcERQP$W
zd#I++J;JwwN~=f{zzqxpXoUOGUTHV~lWA8wp;S=^#vIKz+e;_Mhw_r=<y9Hn{#=)+
zbg0iKKf>_Wdhq0@F7F(`X|Cuf(qAAO*o?kY0@>o}ix;7AlIB1vL(PS=ej*(_C$!u>
zZ@CBVB|dw02EoODdvu8bPEx`B;<=CU?A@j@j8gce+J$<`BXI3XI-bf~3b-sC0RExa
zjC%A2_>4A}Y2EN2CLSLjU!HE1K|2aOaza`mToerg<hA^nNk(LAkLM0L6i_Y_T2cF`
zz`!$b+Sxcb@ZiWP15N>RY^$5AxM^U31gyPgl=gm7<IRNH@P9Ep9G#p>x65Hyp9dyl
z(|hzLeldad=S`7{;0taWAfC1XINzLU#{BT%!wg!+pL5)?DlR7~htdJ}MlIzj^y%|w
zAi9HjH|Xf4z2O(Xo8W*ut&Q<}AN+WInUhl<x&fcnAaifJjN!gJ>~2(!fDR-F`xLAi
zK8s%Jgqi?2()RoNFYjN8l6uE8t1W)rJ~deVC9%pi(#Z!6xXOx$MRsI8W$21a@6BL#
zEj&WOp(^MHQ0)t)gtEf=&iaFFQZaG}wnZ@P$z>^(_)9l7fo9^1bAzZ7uHEa<9<br$
zs?;1(-!&<Iu51#g`zyUq;H-?vl{!|_A~AuibwH#jj6)X;0W^ZuaI~!k-A>2DESRKr
z3?+;^mo5yGkVh2sn_ifDhl}^8d@$VB5I#}$)Fk}+CJq9k4E%_T&$CwXsVlq&dOr%l
zQOB;f_G-iaA|0TT)Igc5@)*p(>;qKWG7N1$^qU?K5x(~HzvWa&pM2Hw%;0m1J07|g
zN8FBulUehuZ_NZ~P?iE1n3;_Y6V)_8nc)W+ZeqjI68b?fsUX^HH0C8nlF;5|GW;P%
z3#a5I_z38d&N25?$HLE7L@s5mfM6UR5z+i9SFPEJ{?g4?;UKcD6qd-@CieZrxDfnw
z;zHI9QlJ8w?4(j=n!dth`;02Vf1pjE$G5A$KQu9s3MCSR3=+c=t*n<akf*>))i0Rj
z4@WY_ko%TrddFs~n-~K-nFpJy6P0=plz;;~d@1T+zYs@uM+1Q;*|vhNtV+63*CQtR
zU(&MJF%8X*Hv3sq&Kj3W`PQfEBSGz*1B{3GgCb4`rz!7cO$8Quk)$TBQC%1(Y$Ke$
zcN(b>R6nUNcYl73I|>vtRFrN7Hjk16D1gS_YFsIfzOeq)3`+3~;!1H}R^r4+6Wz6c
z{F_L5@V%6$<;LasF&ESV2TKdc3E>;9C(FxL0CSCiT#Yc&l2Kp5I42*TYO3>Y<U%q?
zH7(@|e%U)4G|#TBO}$MqGJSUiRBI-*GF`lQ7D4MPw|)%;Xr3bs)cgdf!QU36U;iuy
z)2q1PH^U26y&+zaN&B8&K31IUGeX@LM5X}#@_THXQBYE%ooCu-m%(#)-R`F`nPApM
zBk1CH?ic5n0&C?2iAgm#cdX$L>pO!tx&bkR$!)H38X9D%UI*PRtE7Y!?)V>B!u7A}
zViKnPqBWp?)RocMXxg=;Fz?v0BAzw=g9UlS8BCRF<vM|gRn*#^uU<-P6`l6R3^gtR
zksIxi&@f7N2E)(ZMi<QO^3R)Y(aAEdpXxMozJ0753@^NuuRb>s18sf0>@29KEzxwM
zsH6+K9<?cmiCqsX-l+ZEpjV%#8(vy=st9?L+*YCQ>nn}g9}(x+vq0WKQ9j5hC<(M7
ziPomR7y#-C19{~B7t?t!Jv)=@mz?a}etf!!q9<tWj*gB^C{YKoS@uF!bw*A28CiHj
z3Lv07Rn>Fl!Bnf$CnI06Y}ObjHWiGpVT+mnYi*}Ko{|G$6Rn+K4rD^lhNUfK#Xp4A
zA&yWaJZKMHmCOU!S6hUm8XO1W<=B`uDpsI|IQSDz``hODC%CAPWm3#Kfn^M*48HKL
zed0eS&P^BRDJ~m`zR+wsp{dNStwo_4!r{)(pyJ}<iZEqc<nKOWi2Mf`R^*+-<!6AZ
zR&GIr!ZH9=LN+6W;PQdxmW}@Tf4*!<krlh|<!Jd1UpP@$Z*LnggDs#U7)U5F!xJd)
za@>?4d9{uFU8S}Q^jxo_bm<&{`x7!r)5kJOq1<2k^vF4iNrsV7p}aZg(%;Alb<s?V
zf!X>iJe_sS1-lsOW)NR1`$s3sDM0o#?|EVT@!3VtiBVx2&MuGusIw-mmr;!|TmuT6
z(gBr%R-g<5o|q#1c-zsqI6?W|V@c27xM&y1Qi^OU;|A;!26F11O_kqSXr%~m27j|R
zRS?5!MOg2@YTR2jVkYS*H0%L((B|f56Rh(Vs7g((t*sLZY;OMO60=})RlAmT^rmwX
zm%Tc?T<!YF#Plu5Bb<DYgZRDvjSeIrtFM(;fFg#&Q2<QT7Bk3*?hTfmXScyoCGe!5
zT>|z-E#Y#x|CKm6!@v4OQLUr(%a^+YY<L<<XjLA=^}3$#_<WN~pl??!Yx<VVaWII%
zETW=RCZ*KSV}zkZ*uu%Ys-)HgTLE+JN@h#ZLZstwJnW33HeB>bp>^zH*j=uKgaqS`
zC>m-$Bjk~@GwPRu9GM&_V?zG7CUmdyX9kjqH~F<aXm?oIJ%b#D`I*G`jP87g+h3ln
z76itOgCt3LiJTn$x&>y>16by%zt!`eMKRv_|Dl+PRN2}&01F1+76*--fZDhR;J?(O
zj~2{eT`#K~$_y!V2b=G_x;^`kZv2zgsoelo2H;XrgB?LocNP{SWM;;QD>iEQ5?VY7
zJId%mF=3NwBs3`{ke5GZCfBItFdg@jKl>kgvT~6l*jR)#0$5-<Vz%<R1OF;@oRt@o
z_|z#(Mp>T?OIZC6$qLhY@V4#-tCLdW7|^Uzb$=7yT+Kbr3tIoJ8$7>ljHT;YyiC`F
z28KWn&H{BAU^@sOI5m$q>=R0#O~j%b!{OiJC~Tj2U9Hq-RVj*gbZAX3Ln`6=tXkad
zxRZ8LtxLDHNX~F%wrE%WHW(77TpUkB7a!H&5LrN=Zfj$x;S(7XH+%&Zvu8uh#p%bq
z^q2T?cJy(29=>cQ53NtQ&X09b3%4g}7r(Px!&?w<|6}cKZ^3INg*N)89>3d3)mMke
z!RZN8z0d+2$Z>De82==E97JT6;KcV=$2of%UaJ1O9_PkR2j}_R(ZPXQE?-wv`dJX~
zf4b#Gk#hzc_j<MKm4_U3v^X`5&A1#*m0-B7PefB0=9v4M(4gz!{&9f!+;1@j8cojj
zv~%RPu=`%@3~re(dnxfNyk^YduplsQAuMbt4X4fszzHgsv6a0|an`Tx5n6{gSt)CA
zJzr(6Rck$|FUys33FZ4<Z?c|j-e9^VTiU#Isg2@2&QsDBwZ8($H=?DBe^HKL&!3<5
z@Ha}Aj$ee<-S5>8PMNA|yNB+?StTH@ydru_{Kwd#DMgITf6^-Gy=-sXPi5IS?>xX=
zTj5Q{dJWU@fphk<k6K~S-}NH13*>vK*GVfF0QyRKqJQR8`3uwHq8}859>qysq3gpx
z1ovjl@WdSRlbLUA=$c5O7sXXNYEKJ)lXOaAkkfScNz~cApKq#p`$VZvT4hRq_*m}*
zZ`j_SUBDDcDA8n?-hER2#Ic+^5(p0(Gy}!i(aGAsnn5RHGh(Ea)Z`jrr*(SLyPp4U
z)IeB8vu8|Rl*r2>Hu=p?&YlH1W8sroSKGn&8+H=@v%d`9bf3W%IzKD&p4b#}T%di2
zfK-QpD96x8@HQTiGp^hJZn#LR!?svbeQ^f;4ftXK+T5Yq+{%D=!MVz*YXsTh&g+{Y
z>FG=$Gx7s8mh?T+UZ6+qH*nDKK{Q8+^hU!WkC{MnZaO{-Ek})Y{^;->m(L$OL&~)O
zXgt48ra}}19?owTPQGKW<+o$o2D1LsUP@i8lbtaP3aA7E2pnR2rl3&lffod^zH9Ix
z?*NGaW<?F@8%1#5nw;oSX#twmJakml-$hvifV~;;gu)bAn<3UmN=Fw7my3aXe<Y4(
z0YTZC`~BlHgoHYw5rHlj3}jw+e>|=x#z8b11P`}Tdi95|-;THxuS!*l3>#pn^R`?8
zBN=p9@4eYDpi(G;1yz0U?OiUx*jJ?UW9{T6pmLxRII8-AqKpQ6W}1UA5Kys1A5=UH
z0co3=lM^45>=v8i`cMGWEFh^OXp9QgVW9GZma^!2<H4Cri%&hT5~Pf<uS%^)ztOPx
zktY|<lGZjLKNaWd8B6f04s$Nwn36iq&U14HBb0<GCZ3%u-sIAam%&i!DeUd-?=jMp
zQoYA2ROU%Eg*8Db3@jbmRDhI$6|V2mTRtEi37u^oz8djIFO8K|p5_Qc79`uJ>FKen
zqpkyWKxjtR^W@Ym-}#nSc9r|c7<7JFpml3J+9sH&utCEbEntMhO0@u;yu8OIGO+l<
zA7ciR4B;^`%GTB#usKQ{rlo^JLQqr*>=yixd*ae?+u+3`oTAhB{tXT0ffitHewV3c
ze}Dgs6rKEA?b;M<{%F!v{xyz(c0HQ13KaKDd2L9c-Vx#ty0r=F2u{^iC#Wdn%81Y)
zmA2#L92^{wnt~N#LOIzhF{hToQN2==vydCH9kC(JTF%Q%Lo`#xYpqG=@rHOuJ{5gY
zClV>xIYTUGF0a~0I9{XO$^*<C6r^_W*AWCHoKBGaX6ED#C&Q0RK3+7kw6qj*n4)L6
z>ehzFZ7S__LER<>Uj84M!>w2*8D%3QW`voKuSEG?$=u_1{+GQeq9hQA+joXUlW{r)
zONp;2kqJGEgsFB-#RHUkL4paVz~pJE(*-!KauaK}gNoVALT&9p{g*)%MzP21^xody
zCGa#+Qc7_|R%;rMks$&3I!L>~<7oN#5icSlqWcgPPC*ty^<sqdKRNtclN6M}!5P-p
zP8zWRE__*S;Ic0+KaP#tXXXz63c=7^-2yxYTTi79*6;IIans_%n?7D8m1HMlvC&-B
z+I)RkZU-b%1<nLk_+)s{fC&5n;Du}a-$;EZ96*3W{5B)pdj|&)^q4QInnL#?B*zd0
z6hhQGK}*n=g8OVh4AEE+$Pz#3XiI5E*7A%u3QNQKhkZ*!v7Y9=xHH6LEAVfxAu=GY
z+|}gmIDkkOA4HOlz>PxuVa-ec@;N?wRJPrP%bx~}8Z3)uIOK6@X#kJ|L3%>-q3{bF
z#0Xf{|3Cluf8h5Zk+nUz6+s%U$lCEvt)7}Z9Pj*l9Z!<fZQ4sak;20MLcte<`f7}z
zbR|i82t0Z6WM*R{778FSP=m^y5?ub|f9ZYyp}xL;sqe98p<$!JD*ug_Le0gEC!D_r
zaanHB{QX$GxL{-MJLSeVh?t3tKN^*b06=Wfb*K-4qH_j8IR=_{M2QXDBw48b8t}<6
zVuhVVyb7wmPhQKz9A(DoCwacS$hYy51Kk3C$KQ4rRK?G$VW(7l0Y_yJK%DKjn)9r2
zxTqxn$pMZ%whDZd6-*K0wFMJSsNZNI4dfUv$-{mcT4GkB^VN|)tS?108;a-$k6)Jp
z0|O!U8V<3Srshohgc_EL^gfgp*d}HMB;EEJ;w~-xXqleHacol0Y8qhe^^hm*5WCfH
zk3H~;;hmq-^WT@2K<}Pv0#z0Rf#=ilc-jZY&(9Bl8tE^TjklzTpW&D8eCeh0`Kiy7
zJhRr9<g&D$3~E>@74483Ll+!%nVOq1dJ=g%AuER&7f$(6_2*ykE9#78;vFMwsV1Pt
zo~Zry)sx)78(tkw^eBI*J6x|t4=N-rN+DliuCF8iQ1HN00e80V>9M)|x{}~1ql686
zVp0n07{x#RiQ?#Pe0+Qm5Ggdc489Tyg+UU8dgqG$pU;eVqIc+7FXc0}@P$+`lm0*a
z3C?3kY%IhKl*8Kx$@y5?m|yNBmXfl3eoaSTs~!@xOR#Aq9<CTQ<2Ce_*XG`L>EJr%
zb1?=dj+>ijJG+U}X0u*;xDxzJv!Doqc=W&1tnhXe$nb0bkzJekqfUR#NyV28=hYcU
zP8#wEE?|XhSvQ`-hf~L4epeFjIysSwk5s$US68DjWWECD0r3Z+!UQ~m;oa0>o%dMk
zxcvBRk7OSzsU10X7YXbYD05zE2lE+q;kUIjce3#A>d|JvH7RLolB2DlHH-uZ$a&D|
zl_8~Y)-UH~HjONBtp85igR6*xRdcR1eY#OEl5P{(@A*uG+N@(B&@$|#E>K<fU2^At
zqVu#g9z=|S@`VDoo1AKX`cSd8K5JqRggOLJshYb7^;G`)T~+o(vL6_~1dp+yY()85
zZ9LJN_gYK*|L}X|^2I-d(v5i&-mC+jMg-+9l281ctj{`+_6lC=`tT2mA3Zp@XK2Fy
z68Ii2n{t#a+z_ZSr~4$R?G3gcLi$ccm}LhtR&YDpzk0Bw-bvV1?7QIj-6gy+F<5zv
z40+;kiAGAnO#LZ5p1<nruev7}cR<S0B+YLIv5NoU@MTnwhb$);(QV+pChY3g?69#^
zqO?C;qs@d$F80o#eo8Ep=`)dG0-+QM^wXLxb?PPVB#zmnK-2c#Y{o&QLj)Ct(OduP
zQ}Q3!$k0|s`;&Spp;k#Y_<VA#PG3Bt+P<C#Ede|AzMq}6KDC(pem3ha3woXUP2<Cg
z=S&IIf9c#SzFFF4K<8LU-!2;eG`+4Tzwz7Y$*zpc&xyM<NA>%QDXKSb-oraV-X5!l
zpXo!$O_VD<;;aAF`xd%=F<Xh!QU!_@8F4)1E;0HE)A~Dk)dVN82wdpOmtM8%eN@dK
zTPmE_UwQKT^DSyodwL*{ptEj5sWu(VkfV*worn$A%7Ts3n*%&Pr;oTD!n|E$V&o+E
z+L`U#^5x90+v+Q(FasAV?NHBX`}$@1*xEbMfuZi`4<lQTkN<C8gU<LpojS_|^lL@B
z_*|+d$2ylyR@7f*swdlBfgmm5Jbt4l1O^D^P;m%=UIj#XzW82Hgs=%7e^AZ)kCOk$
zg`8?Hk{;wKYNXtPaO_LVS!%HZ?Cc*aezy-ctbx+aT|44V#bfnQ8DZJzk}xR0F5gQ^
zLuWSJvyaJmS)Pm?gZg-SAX>*{RB2Uxax->zM(5UMvh5uUU9d-^!3;*JEW!sVIc~!S
zdVoR_#UFQ%l`5F7*{KNn2jljR9UJa|28lXo?>7cXfn5o5a0EDz7{_G`PU^Y_W#3Xb
z8E=-f_<O5)O)h{M-0O^Sv+zalNynJF>eX-in01lYHY7{^6m%ka+92J}EjZrMUkNfK
z9f+4rXxbK{%0W0JCVel3J%ePLN_<Cxhb@4RvGPrr*-I36X=B~^im8$w%2{>w0GmzK
zZfCfp{=r=gF)prAYSQY;<w8pj9lyt<3PDjThP3jW&&ck)Cuh`)8TsYKOqJd7eyP6g
zox|0@%;eDy7yTi3ModfH$jC^3FiHsNL<nJ^#C@xsD(=~{Kyc!|WUSI|pRndsqS_#q
za!PMVb5&T$tXBmm=H^vF+Bd!Q5(ikVf)6j_$3Ipuc=~3c^vE)tkb|%2B$p+JmF7my
z($iV$Q1T9(cY2IS#`CREy2MD?aUvl%aeP@}|DoUkH!K_yLrlTq5ROEXF56X9)&o->
z91t+a>L419CJ^Wuq7SLh)u87)6i(vb%g8oJxpnT?t48RKp~#B`Y1yevVF@NvtH-1b
za-n&=n5-pM%(Z!cCcPVEH@}%KiS8-XFMay$u2B0ff4+*Tm}Td}bn~!v@qf^K2<MoX
zm_*g&b&prnH9mCSk!HdFj|Z+(bGVPM@_ivTui!MGOXE~p@Xyu{f~QgY0!>T&-c@W?
z>084C&y#JX>ysXvhZ>#ubzmZ(835LEV`gS<TDDnHxH6}zFXx5ekx|<TR7u*w>1(4j
z%lbY2*+TsB-A}ko2fNIin?(Ke^+$g$6fL_KMhK2ioVXQQ`fxX%_us#Ivk5vVlkXl2
zC|9;SGXu}h%mj~}gG$Pe2^+!#)n@khhy#}{wA3#BeqDwCm9=PDIZLOkiY5+oh3Iqr
zkzPp9_uQ?^VpS>*snSzl$-HS-_C941oyuyZulUtAiMq|06jTk(uG-qF`38d$15xf-
z2u}v7nl;20)+E*6S`i7I7yDJ1I97dh<gI#vTs2n{Om7>KurZDaH8ZJ~FShCHj&3|y
z$Lg)3*9NtM<lc|z(T)#KniCd((UNp_$BA6k+HebgD*VPsk(ky{@Jcp<G&lBsUsx~(
z>v<1!F$|=*euuW<ct>|{AMgI^ZMueHzs*;Q*}IbDS<%8K#1ojR;WS(0r%JsFcg}=D
zTl#QlPY`rY`K{o6eusvhiZ4MjKJ&BXo$n6Zux=D57c!sIsht{1`DptV)K?Z5|9HGH
z6$17y<TpdW)buwFPHq1ZTDE&lUjdZ3wo4YA!qFK1*XBP__@|`tv8v-yxOQyhewJcX
z=pI5|8VbUxE3cOki)n>6$`7hLD5$}~!OScy7@$Uze?Rip*pYr?1Hwm6OnFG5$_-Oe
zk39^+CQ1te37(%A8f@;;OkpkGze&E3y}=uh)j#$~#Jj`vVc%8fJ~K;mb7Me?dkcw%
z2m<H615!uKajNc823@>7&*9<IPGIJr1vkPqppvzDabn(lj9O?`hR3Hlpz=`sUd)<L
zWZbXvlL_~L_{XyJ!cm<c_TPA3K};>!Ll|SlQzbpf(KrUQ{@J;?K$Ll+$;yq3hAfRA
zE>97JB$6;gp(Ik#UZ2K%lVU{z3X<rHg{Pn#?eCkDM(UG-z<ry<pg}1xL~l`xq()EI
zMI`?$&9u`N9`XCWhy1KKAHt5AijmvP3$k|Q@!s+G*Ni<hfzd8>C!jMV+AowK-4XZn
zsXqt++tN7c>zBMSg`*>QqcEH9njgM29bjV_QHDiuTLupllrRQO>o?mX5pCy>O<Fy6
z!-R7zOG$hDTFe(#7!ygG1B4SXqty)fr`?0~Ec&{-njo!?Di8?z16YXI?w+2sLX4RW
zAFNV+7PS%o#{)k$ZKq>lsf!d{5xi2t|H<DV6G_nsFI2HR>lU&u%c6Bf)Q0NUz22Ko
zvO}Be?}~=grxppM!RQL?V+yqCna-<9YM&2!Kc1<4UQya@je4TeE)wLK=5OB^Ame_e
z;c)$GP#~0UUY?pwybA=pYq>MA!w;aOn5g`BtW{EYa`CR3xzx%@hHzS*6W)1xq~?i-
zXz;Xu$c&k}Cjmb}#4RVMn1pPQhqmJ%CrgTC)|kN71Ba6dSXmoTds4`cL%ap`%#@?m
z@_BjW8}8$lssK0)X8@d~4?_W|Ld6u|xsuVYWIO-p*X((9AY(<8>!T+>@bE2f=L^fp
zP$6r<vRz(HM$?Z`*81ah@)wm*xZ#)HG+*S<G;BVCL&9hT@fLLU2m?X44+P~1Ak#B7
zufaY{Mb)qo5J-@q^v3$q(_xXnHwg&Lo#C<g7Ddt@ht(T_d^xHmrNLKT6?o|_CVkQ3
z?xfTki+lxBE{byj96}e1aT6$-8_$KG9B0BewI%VJK8=sZgs>VU)oCE9=h>MNd)rWN
zuK_0~f{yvdq|w$V)!-yL;4&sEfXWOy^{X}fw<lFNtE?Z?l(z}6Pm?@|vy`fsK4#w5
zvct$LPAp>;8=M*#`EXKEIm!OQbiuvM6kDAkrShVozR=lSU<;`9jSkSjoQ%ub7#X<X
zlo$3egTJP~Ijzs?9?W%+w`D-Fl?uqw5FriD%EC|WOn6ck2ki~J%;t;zmpT{Ab@Q7I
zf+9EWp9X#MyOlDG*QQtg@<o4=h+==h!qX+GVW;4eQkvRZcceZvIxi8Uo(d3YOazhz
z0iYeWw1-T?vn2d!pB1sb`{dFhN3>#i1z0WN3}g}1?_&u&(%C=#X4Gxjv-)t?na`*n
zzD+K<Ye>h`-^5DxqM~C)|GmRDc(MiL`KwtrwxJuJ3#)q%&^!dpCw2`Cgh2?u5;Eu@
z!qGY2Y)zW@rsza`W8?hKLaOC+tU}OGN(E<&Fx}7k+v~j>IgH}UG&x!e6Au*d85I;3
z%kQT=Y`b?{zc(Z-(t76pm*imx5yDI;0h)Y4(9}E{iU-fjwn;piO);`&Wg{VwKt}%M
z2e}PXZieM^CXja{uQhu}Md(FIzJe3ccGpXUud^knM>%6Hbi_I7fqZ)siGF5~fKTw6
zp-rTZ2~l#z=c<jNr%e@}gT4@fL0viE;AfED)T>l#FSrAso^Sjb4_w(eFO88Q<)X);
z$b^qXd?U*u2(7SMeQTsD%SFz=Ohp$10gHzXxUI}@mu<qE9@cy|xu_*Xe4{C&_ZfEr
zza6f^3GqQko=nqOMaBkPPiIlUahPn6P-loAhFU4kpKk`I6x4PVP~BW&6uREAE*eD!
zNV6=0w_eKigok>CLZ44Ldey32E~8SyPKVd`N%s9ZLVPQ>f4n<kOk;=pzr!#4?*^Dr
z)8>!O3B0~i`dgq<+WVBa0h96v16?{IYBYJtJ-9OilA%EGTEU1WK-~8i5=&tc2kYz)
z#@O4g_<bKKF@5BFLuSQpj%^=RD1y0pfgBG8qF{!nB23|S^;)gTwXA;mF3`Gr51Bg_
zUj$Wru~u__qMX(F>UH`R;m$(dj4vY2`CF|cRdM~YA*KtUXuW;B!)A8lVd@F;K0$~3
z*pFB8*z4P@l(f%A3RvN+&^sZp91KPS8rOxKU~6-8S6c0@>0dS154D5%PeGjufexKQ
zfZQ+CMu$S?HX2VuVM+;|S_B4|2@YSkX^rb@D1;^QCD&0&0VK_Y#%V|jq5(n3LZBHU
z2)-GiNi~=6qCc)-Wc<8VpBe(`@;{8b9mAJ}Yz%H~IK9yPBUe;o7#V_#_pI--$q}OL
z-1P9etc`jG{0}0lm&YX)BoCcS6PRx?IUm&vNj>IcMGa$g<_&^MLm6_IXlxD8WnKU4
zAIFA1g_V2lEe2DVIc_D&y~fUJP3$*cp`&T2e+Ff?Zg<KHP5Z-;83shb7o5-p(zA<d
z&l@4*#dWI7lQ<7iNF`7aZ>CE}|NiX~#VAD!UOY@CRV$^FkP%PwP<zLgQ`n{a5u|Uu
zec=;BKIcRY&ACC^^6CDO7ve-L!#I`ViJU|<Z0V+bTf3Tm)Bl_1;&br38JZ?YR0FHt
zm3j&qR9@eTuj3j*@azQQ!JLhL5R^oNa;RjSXyAFx1Fq{TOo<7nq0!+FPV?Aka@gM=
zL8BpP0REM@GwQg5L6`DJ9KuFl&j!I$&9W<RSd{P)){)4laH)Jp8jL>?{P2y%3N?}V
z$Sc2MP@m9U0D)yk7Z)_Sjt&upD;}k&jz1()l3$TXaX8DKa2*2y3y`yFW3u`jWHuGR
z4&Vp3^zRheRp&)QuysjENi*t-AUOubsPoV2sIb9n5R^en=9Foxyxg6>%(~^K1VTc;
zDff-97C!o}2X6(1G8VMOffBRGcq9(lO;VIIfvDbHcuP08C?q+vsJ3@}y=dl$MD=ve
z)xcC}aUb`1cMgVTeRgPIz&q~c%O{|>&p@0VeWL&o{t0rOhqdkAxC_!%#*Nlp9{rjO
z80+f(3ehGMvd@4Fngn?*bif0gUCxbxuCCiat}pw+4}<PSkoS^hjl1lJjx2{-DynB*
z_RCU(B&$dwl2x>1m9d5`ImYe(S?gWyvrG)T|4A|)eDJkVt1F}LY3_A}@9Bq(Kj=bj
zq%8mNPsk6<LiZu07b68ZTSZCfEH*KHI0$EKoSfbHzmf}N1X<dIL>Lg5X*mzB=bd#a
z4Rr7rol}B@UlcX}88plRMJo*QpV`I2@bCerFAP9P>IQ=}Kc>?Dxj)IW5mg0!FZ_o_
z>I|?bHQIGE3P?Dn{h=dvO*B5=<-TLbVm4?IliQ7gol@h=O+WC*-ga2MV}4!?7lqE4
z?EQG1K7-~BWKQW}dXvrDd(ZCvpyJoV#)FMT?jqK<C^43bI7s+tp&1u+LLHs)gcPwv
zEIM$BfmAt6E0<gJkwWqkdd3_2Uwd1o$Y71bASBi{Ll@d4A@IOb=bEAKdS<9nr!iy{
zbFYrMKnG{=>-l5sZ7y5=YD3$O&ySd2YF9<ni<fhR5KaUeycK$XH-#l2ckpll^6Dx~
zKEV(aWO`Ih4{|89>@yNB%gkN-MG1OlBo#P#rId1*P21a#a&mJsAs<`UQ<Oh1M(j}k
zhZ22yIXM>q+AsU4LsA5}Oc1B6qB0Ln6a`yo;u~rDua^Z3e+ajO-kieurzNbi?;w*E
zN0FHCT}A57!NZlpcx&r*5+jx%k<40I>Q8T@H>jh<@1?_B9y-$|=Do*<pqOgxz4f`e
zx;hu>{=ufHPal=!FZ~`~7>Uo>A-{ld^YC<_f--1H7zmpC0b_m+RptTEqwheNT6zhu
z(m>OA@Zt&>c2ujjyu}*V)YpejIitoFS8?4L85tRL5E#6pHW<qb25B8ZJa^V2N=o>m
zXoc~i#4CfpxHym=3j=7MxS2{h?Q$<h;aRzEP<>Ya+8ZZs@uN3cqk!3L@RAD^u7bsn
zw;KWFW?W@3Z)XZwadd<qYCWpj9xch7Vr1v?<OY@oNKJp@%xhr!vyuB<hV5O^AG6{m
ze@kz{pR8(Wq&Kj}F~wkaqGo{V_3MEJHeaR+L1boQVQJ2A_JO7c?$9y>gF(Q4F7xCU
zRI7P#;?YW+=Ch;In<n))mkm3xP!fH80BkZe%L)Nl)9+nz5Da{O{hO@(T56`Cy8mJ_
zoAZSjZO-K8v14=Y9+p3Dlstx>;lp2bjK;h9J-|h=nqU+=5YiH#8WJ6w$?6X6{lw8>
z_2&2yY##6}(ab314&luZp|D$x0Z7#a%CMBZb4{Hd-p#)4#c{rf{Yx#$wcLwi&N*0?
z4JTP6><{85)SljC2FpMWwjNk0GYbnLjVA{LnVFd%d=5omGa4mcyqkXxx6~DUeQ=X<
zPB1a7Rx6F(QR;)pw`p`YYHA+%Jbu{)JJnSlJ3)yrUc3bSY~8Q#(oy4J)zgp^-wcZO
zl@9g%y-bG2n_$Z#O@1f8n7nt!EjdGWD)=AYC&`?O^**)3x;cg6G|U&h`@>G^W9jBh
z85w&j*3jpzk%XJ#itcV-@h%7o-l-maR|65^`sqt}EFJtVOi2g5aZh{9R=D+#5;sl+
z+M{A5-n^&yT4kVR@r+>Tto-h?N4ZTLS8cHTLM53LAh_i4cP6E*41K2yXgDUYG3uBc
z`rk3iH!y{a(Y<tI*R`B>;$x-YKk0HLRKOe;4(d<|zd{PmyMO>WIF*sxkCg_i%zG$d
z6$za1<ilx?|9s&T-?K7k5VHJ8W<zoM<vw9Dnv^K0Car6MBx^!Xv;1H~&Lpnf=Z3Jo
z>d^}6dtGz==g*%{60{^nhxZE$xzV;q#cLpRc@MLfZj4~xmm+J2d74=AE`v|6ZdMp+
zCY}H$ycOXQsE;_J_LKhVms$a*pt<c=(mu2R21w_~gGq(GEs~C~xdR;<cZIM?>STJI
znX>jQp2-Pn$XgEM%WXsvtFP7<Mwga*$X{GMg6aQO$i&3KJWPhbkUfZaoQD<r0+vR+
zW{w<+@+9!}D-1sQxeTRLfmr@%{smkLbXpJwm}j7C!wgMO*+n{*Hp%y+_dhopMs0-s
z2JsYh@wbeff?w}+MLl5IOdaA^*)aD}>t1P2`$y~GulDW-V2@@sK~|AtjhjhvXppll
z-0~Xkr5Ph&(0lp7>-LbR<}_my2yof(^egb5VQJE7kmHuJcl1(zG#7Z~TP2E5u;4iV
zi1czT2HE?n+=3Pd1!<6Pe2m1mV|=kIFROUe-u_4!di}pcv&fO8;pdx~G^dK;Fsp!C
zH+x6F1~8BTsa<-vjj6tZcbw^4_po>T^O~l{6mgHQ33gmeN)m{FYd2tJeyEJiuf#>B
zndyEj`30h|B4X1<503>*<If&`VE^2x4@Nzchk`efOBih_vz~~`9Ls*5N_+Qgo#g5b
zx#+G|Dru>yk(vv3_MYVu8sh=xQ#kyTT-!a3Ppbk?nCGFOqZ6l~sv^in-7jM4qfG^<
zCj^34NxW*~dd;~ULZ`YP6}7e~6J`b;%(GPleZ%%=jZ?L(EY`F4tEYmu&QzRF*NI+y
z_vR^;*@{$Ro&e7x=M^@rqJd)$wVn78LrtZtb_>pMqtP%Vl9`c#z&m=-9K<Fhq;dHs
zUiN;O8+dimG3svpxp311@ed3scp_DA4eKg8Ro2qKy$GoG-Y@IkfPIU;o&%juL8A-u
zdV!>@NNRFP6fP3>3u9xpTH2#)Qu4vZ{;+5r-GlB{@++0Y#r!7fIP{Z#t;<tA^89&l
z!w0JddU{bXd1DHa3*t{@=uHO0NS=rJDdcT<8k{QU)T*N96)Nrw$U6EQj1TYO@`o~G
zpfiXzla%l(9dtS>J62l2`ZGU^qS%3;G0k@>CwDDcgq?Be&h)0eDo5<_lNX)@0(5F*
zzx5k;s1tc56*V-fyM+&kaf8BMmm3AV<_~zzF`o6sa2Flsg8{G)zQ+>Oq0R7KjZ&EJ
zuRk8^`}TUAViBJw#AJ1zEto{^jI15ZLqz}+PWc6g4Rf>4@k-+R;1sg)H1D97Hlu=D
zt#4Xrs|u*hL_+hVbXP{lN`I0A*$UrbpnWLWLE3?>*<l`euYR4ra~86S?dY2&G&KVu
zVujkT{QL)3OSH3Y0d}gy)5MjKkuh9Qfe}8SC>?N3=<7pRc;r-5+e;r&OsmWLZu7vu
z1m5`ZOix&tIEZF@e%i)+s2%Qjp&bHR9h7DxBei$CY_u$|8|pQL)D>veuZJkLqs8y!
zSNds2P7W3rRI&gZif0BieO|8Ovz+>nn_LU*MLfN{Mpe(i`&|B=`wjVk#p3#Gu*b#I
zCMENcjNZbt{8`uG-=h(j0zNgYjjE4l%+=*Aq*JBU>y=&eDtYGbbk$0<ucoa~KVf0G
zItQ<%f;kY(e`iW$FU;-wOuhO!xk9UJpv<nwBzL9>`g6e%4|xP?aKnpxR=<VG;|}%G
z_qec<1^JfFN(dSs5)s~C4-<4zkWSCYQ(f#>J(UclG+X@qqjlIui(Qd9xBTgDiLm2l
z;cM@2`sp4RbfP{XI_070L>l0^v3f7j8kZj?gCDMZdtF_bW$t3KffdnT<MxxY%^FPd
zWQp1wwSQe9i!-4w)U_2v>$n16cIvVRUFm-3)M)e3r%CP9V~^)P7xlSb8qesJVEmjt
ztZ@$y4{8Qc3KU!-8Q{vGgJJOQ07Awp;RB&jk;S9kYvH-n3<#r`qyKIx%<&5D2m}vb
zTQtb~H|=V{)jT(zDI@T>AJNtN)gU|W>RzIAnS5XF>xpQ)z<U-#qQ$6cWmkD@&6>b9
z2&J1q$%PkZ@t~vAo_`*Kf2?o$^RB*a>8<kmOmi1aI`}eTYCiu_C&Jb-a~3Q4!jI%V
zE^jOP2wxZczP5n#U7>IEY-IjipX%?V0^!rl0|6mJ?;TNEFQOX*UnT=6z!0u=yRz7y
zN|3A`IyiXy!i5XW0s^1%e{rao2l>4GNpvMUexhj^d1}7@oA6>Vb?ZptkKcB_qr1&i
znsHG{SqWlGGB{^1-u+0BM`5;D-Z+$Gv+Vh95kIG_tc=fXP5WrOhZpgOcg!$=u?tZS
z{yn{yR?}YjQq5snrGNa=gEv`JxjsyUKWHYidHe8Oo|oH4+UTYI_nyw1n_%S~s67*M
z-=$`(Rm80!EGm6m5O!`t1)HtT)n9IO=9eA@?nKQR%-c1A0lfvUw}4EF5#+wnHw_J`
zHPj>(r0UvP_MA#b_>_G$XhLplX`w?AGZ20S2mgNc2F&SD&XH4s=U-a+b_R1$sMahx
zt>)lx^q%jgu>LX0dc)V2hfGi9-zUu<>M4xmobmr(WSw_Bm;KxSKggDq?97zCLuHRt
z$R-rBM+n(uhHPbJCbCC%G9t3~%-)ogz1QzJude(4-uM0c<9b}zAJ@ZYo#**Jj^p)u
zJ&$(Z7>G52ZRaC6Iw4RC^Kc`8WLXSmg-fk4#ytsz7dtJTKy5%PiV&An;F$;23piJp
zI5_bC&5r$<tY&V0$OVG15xyX|gl@^E=Vtd+eF2X~M^Z+fj*>R)%xbd|0=J?p+Z|}_
z8sOxOPDsE(Awd&J=@S7B<4uo3SLeu?7wNHSc!Eq6KQGkKvJkB*0V^1YAP~B50yht{
zd_*(U4}b6Hbf<%^?x^l{=uxqL3kY0`5-?}fSFqWUaNl7?oO@9HH^7V-GVl#B<y5?N
zy@WvT7_Mi=3%DqCC<B0R2SG3J%^|5PC>IeD2zX`vzdt;>CL3H5U?IxW{F>G9VA-Xf
zKI#blrc&dRxesYE$J-Bs{}tM7G2K|tjaaE4OxUOqTl@(>gM>QFhY=SPc<`VJN2o`l
z$7^MGQi_F2gTz9lPmZrPQ^V^6B`wA#cy+Wm5_}NpFt~}jd-ap|c|CDFL$evqWksr{
z1R`0bdnEr@6{u5ftQCJ=@`879i#=xNZ8(S<?2YOEE;7$b)ljqp$?v1q#7X4d=e-t!
zWBc>kWFX}wyv@Y2u1;u2`5+~B0$yR1kMOiB28?4~CM6}Q8W)D=bW<}fv)n~}hfHTO
z<lh00Bnho9a&nk}N&^0Ni>c2Q$Q!zbL#0U7kwDa(uRU-hPA>XCZrpeF_>W~Cc*|ol
ztoJ2S@3(EUTN7m6l+zr!q@mt?FpndQHvQ#_{36GcTew_G_3F|o6f2asojDyG93Y!m
z;<jxDwqn_Hj9<i7Wp{!PH-FI;wZer)LZ)%ZxCNHAudS`vNT>*ie@Hq7PLJSWHb1sq
z>823F)z!<}m%TJS8S>0C*~N*0y-~PH8MpMzaNQ|oMGTVf54F7MC_5eQLjBYkPdEQ+
zUjfI}k%-odI*)eY2hG!GMnPv0c(+<Eki4F{u66capXntD+U&+e+PB)&G@*Y?*w?xo
zxp;V50oa64MHUwq-+}G`{DIexFL={($ip?hczs<O{`IWjFywgqP%Vl|qO-u5A(NZf
z1nZ-_=@GyK7GSz~X}&z*J3)P;hU+iMl+MN$&N=D{7s1~TDyNGm|JpJ}#z<th3$iIs
zTaxbx#Ct1^r<vYImkTsb=qR<j7>0Am(sz8aLt>P14v}o$=RZ`C9}I=C%L*P<lS1dZ
z&s0gK!1GNl>4r3?Ocd(bSg9$JbOMWifXEd_A^Bj#j_*Wj#uCKa+nvuN;(P6H#Ya^=
zE_x&#v8uK0EBbbTEj?g!&WUzrREk1f=m`9o`w6^H&=F4F!h%1{^hS(M`<14Ot-?P%
zM1Cq$oYD&?5CqDA7&c;QM}T0^128^N2=1PxkzHJhWskqFZfDLGPyb<MZ+^ao<&!3*
zpL|4Nqxfldwl%JH<98z2l7yMH*OdzakEBQLL|dd+uzf6Qb*Qe5SUA3gQwVliK%!g=
zB%tyG9rXF1Nh&bdk>j5s%!d$`SKrIUscI?>TC#TAAC_E=-q*sk=K8OXPt|}lfTP6*
zt&qP#Jo2E|AS^tDU&0H<2Qbj$+fzvwT;nJU9<|5bpOI8m)x1s%D%kIF>^W&Ql91Xs
zHoJYpGGfhID#pw0eU$<EhHKbH|Hj7!;{V3SEo+r?SWF)8T-UWA;>Iu)o@-os7rg0P
ziBcQg7RW{t%^;_Ia#+Gc8A8Ah_>X*W=y5uDF~0ULs35cMS+kbDds)g1g~lK%N{g7b
z;J1AS!FoXeXb6~fNg5iyQ-QmW9r?ulGiTse-kl}NIcX4!jtOMy(c2Yip?F2-7`;sP
za+N+a@kbhq@$8R@oEGf&|JqSD_h=}KobIUvvzUlKXVG=tcf*eDt#JtBlyzPFPM!OJ
zhp~x|pw3{&uV_r>ac}P<WDM8Vb{837z`U>(TskGtu7zTHOLtG|TC-_wewh7<<QaIm
zj}wXC-;bh_bOI700Nwn{$^`K6@VcOt<n-$e%6*Z%%&8e6S9*LwFgn#Ar(lFgO}!%;
zV%J}D?5|i~vqxgsl2ae76(36r1r*8RSDE!PJFY06=utk(m`+>F<M>=ZiDHYVqdGYW
z(1IflKv+n%iHSOyIX|TbPk(wrlp!1Pt{mA$IQU22aq|ZdH{p__vOauZ1ey($;h&s-
zkwbM90t7{42s%nX_9G5m*nX0i7DvT;CxEGio+{!A;#%1DABrJr4i=KavAbgNjNd0$
zp)GJA&H^G7p*0b`%=QzeIXt-cKYS3eKOwvm@e@U)Y@ScHg5~pnIEP%i0WiKMA|#AU
z`#@^7JTk6OGGrD`ybW`uFaViCuCE`GlwmZ&gfLB#fq!--qKIj(%VOn<vxF75usnKa
zHkOb@e;Nu}7R`w-D=8n=`k*=|qxHB#mf<TFE#$Fw-ktr$vVNpjVydB*$cVFkvJ+y%
z0EK`KBm-_1@IE3m9Qe>y{q`Eyhq?-@@t)B5jTev@i9?eItWJcz19+@!%|CHUsN&if
zoE98zyog%mq}iS%28`J&KSz(QTLMNMq@^mH(P2hmaN8*+3PNrP7v&@Fb%eAoz~rj}
zL7c>4IWEo!H@A*5Vj-+}P9=0{WkrT5T9Q2$?5uhoBF*<fc!itvmW1LrQVL!}0t5Et
zhpxPC34C3bUax3K+^rZkQ15CE7Ci;gp2O~jOQeSNo<zOpPex~(F9rLbo@9(?cwu)I
zh&}>{9&!_bf=7GhN{P{7P{N7T^^b>CLM0lxT=)+g`^bG-$at>@Y?y~9G;xCAYlz%4
zPU^T-Yp-U+Nl}TmA(;5~HLTapc~bakN7i6mXgCp53)hh*4`F$ux`PlOVR(&1-+^KE
z4tmN1`$Ln1WP?cge6B$@f1-4ATT8aMsPP~(RK+_gNwt&~XA3L(tnmoT(dmzz!}Ce^
zTtBk61gmP<b`a^j>F{vQI_iq>*Y@&x!qDseRJ}~p)9eXkEzD*x62LDBd80Yt?Vx<X
zI9cHBf!eBy@LztbOF7B&&|)OHn&qEQ0Z|Grur{n7cqUlgZiynAKUsVX{;z=__E^35
z7H{ZfSVlf^8cF?@_uvPMl1vJWP7r4{9P~&q4Ix&jU;E-)vo1KO$~m^-lT558+z;+b
zp|D>5oJwN?9qAhtMbv3uYsdE~wQfhgjgP(fq|k&c&a?!IaYX6<JH~BAeEd)sKz~+x
zKbt@rhs4hiZ5E=z2oxA>rt?AcFC^hIMuIjlzeUI~#H-L4xPu+~XBLmoP+hCa(%m1{
z>gM-aMg#;$V4TYfjz8L)4j2B3ts^8~fUBSiM%e&;8o&T0Gm`PFbKmuY`6jnPeOPH}
z$G9cr(BZjA0dt}6T>mZ1r-ME(qE)o*I#u}bxp?UvF)5|dk(mWe)16xrL%1*3tuqtu
z+03LfCf+D+<v1|bA<V5Rdv-V=lR4cpFoAv%g^+v@BoClWeBs6F&yK7S$^aPsM}Efu
zb0~|{1PeJozkjl4z7t6o4vY;ByY+z&_)r>OYi7YI!Bg)v_))NwZLFiIcUqCR#dCze
zZB<igo-I||T5Q;(!)a{5EG~F$8OeicZI_lneGNJm!f64^Fie6W{E9s{cf*5SuSOkM
zT?x`rSBW7VGWp3K17d$KA&Cq0MjmAR496D^KE8})Zf;>YXXI<ir%5+gU+1Q&>nFVI
zIvB&kz7`S?d7aJs{aTeAPR!#VoTnAc2j;<6?x`>1Uq~<aP$^?X=Gou-{J=-u&~fSe
z;3(0{%3FSc^GOelSRd~a>IAE7x}%#&^-@(#T3gJwfolaJKte<S9mV_P>jlKb4+Xwx
z3y}q0TVV>OB2EdBbJ7J;5EDS3Meu~|Rp(Gy0><#hk4J!GY=Ayw6SROP7>dJL)`Gx6
zN=+4(_>d4TX#U9D2Bua(1i^=*956aQTM1D9KIw>I6M1nWg2ZMZqJY4uA|hh;jn=#P
zj5A+6AZIcW2}}>b^-anUMmH5^K+hNElXhxpC)v=&A`-$^{gy<c7JahAdgk$~93Cs@
z<JUSt)IJT(6YE18|AoVkVn!F4Qs`G!HmTI%eC$Con4BtKdMrzJ#SvXa=YsBNSIVt_
ztHxH^?=DwBlxHO66M6oF82Ry`rU8!&VgsfW_xv>dO9gI*;5yQ))QKYYOj*il5m2LG
z5i}u4Y!T4irrlPqAhuMjlwZxSE+OpgTTZJNpc6$-2gF1I^IC8;prI0gbNu3Pi2`An
zK<SUjH%7+B_Nzl|sMt=C;`$?YP!{chcLG2vV~{YwzW-kAmGgs%uxK`_{M8jj$B;Y%
zaam2BnmnoG<yobPFm;lJP}#`pU^d-c(hJmY81uYZ!Zc4c&7ZYu))<V2y>mORvziD^
zfNBB$um6tJ>9ZIdR7M`}r(HidN{U2)dT1!bybSO4om?n+ZCa4F38(1oo$Lox099jZ
zBjs(i2gUWK7O2la4@UU_G#dHP5R4wmbTl_Jdk;Wy1P7t-{*wh1v4i7dW;l_-X@!MC
z>P3s;yvvZ$^v}s*1z;Nv;zNV{duH-wxcZrwyWS!cSQr`ps^jKDo1lw`Wd_UZJEzg1
z-ds{oUd3GuG+uV8N4ut~b>Bj6lG8mBn$LWr5$BI_eceaIDnZrwMy&iIL(HToAxuvf
zP6_qTHbUqKd-qPeNc$}ET-u}qF0`AG@x90^_P*g6BED@yzrX+Gcw?Jq28DvSPGF??
z5l&H25tb-EpaoJuXO}AVp$cq-)Kny>g^y~SFSfV0LsnPt<^cw~z$>%{U^<}?Y9(@d
zz?Ix#jA$l;wv&{=4ManM0E`7R40eUD@L-b$13{6&Xf3A(gj3*Xjiooz+c7VuPA$<r
zP3Dxn>zww`Ey#VNF&Om>{u{OTQzB{g6^XlfG9^ifEHcPfuO5pcP;Dx8+2#1Wk->j(
z$?f24)j7K0$-1iRelj6=aN6+sqg_m{{F{~`d-R>-dj(o|uUHdbfy1(<!UncR&*X<G
zRhyUDsd8&eKbx%y*t}ExfmHh9WK-S~G%twINUst|2{+%oeVdt+Q}Ntaw1hJh;gW!!
z4c%brj1|aN|Aje+JNSQN&Q1QsoEsWn|0*CqRk7O2?oVU`??0h~p~!bF@jmKJnnBAi
zcb2foc@wPl!omI8ygja?hVV|!KYEKWsFLbYR+yVI7ut&O#A1y9p092xcAd}%`rNi{
zI<{i%-J>5`r@JHH`<u(?L}|6sgEzgXYxBeXja5|=!m%!jL409}`Zyk!F|+d0Rg@=a
zTa7;sTEIo|g~n+b1h;v(vJqQV_!Vk9Aa6E7U-u6Z1!3!$ec_(#rF9z3qwf?xV143g
zJv4DC^s%ehNO<SV5|3Zc#duDS$q#3v_9t|I+^6IFz*?RqskHbrTU59~%*$&>y>?o6
zUzXgAvBT1F$w9m|56hPpAB&**b*qTm7Lmx!%_J_t%ezY~0pCesTnHiD+!K9((*w{8
zxvGp?!uVE`5i1JtVZfm2%^)K3(9Hw94gg>WaAY<={sAbrkkMn0yy@~Mx#N&d;zdEN
z@+Ariicg<5j)||2Jc;;hJ>1(ad;p4q^#!}XsrlVMXaq4+ytiaDCbmc81k-OG$`XmS
zGq+R<4U*@Sywp_ad#bwhYoXCB{q7>Rjoin-vCzSZ{;S|82C0Ap!#GRVu*;M=&lO)k
zc{lx=pM=aW5>n~lYFttjFlw|Gej~CSFao7kzzFcaMctgqnG$1p;Egc?N7Nrnlf1kR
zigJWH1QZ68Bm|xieE^W~7ogj~MgbO&%&6pHAB&ZdP6#1YtZ%@XbX#DmfQoY&|3P2r
zQ=EvG*l&XN-jEOBOha-W{E8K$DaNutw*9VC2hrcSl`cKFy;5ek>Dl06G4qFWi|~cT
z#*G_%<*em@Bnut(lo#g%-w8)b>Tg^qUOq7HRsDU;1~b_p=#>t#&cvdj&9h(<%eJ?i
zRO1>iHw~hBJ*<esD1GTFu51!`6XrqS;rqb{VV`f{Tp$KCB{=2)vV)L2SXfw4C<s!4
zwZ?d)KnFR>5RV9wE{9my|0UBy)chr!w8p?cM$U)st+AW`$7N3Q__?<I1M@Tn*~@lV
z4Wivgg%KE(Bu1^xl3U5k*8jFqjL|vQNblA@=CgVvFDiWC^r>pqOiifpXnQnGu#w?Q
zuuPz%bw+V+Pv3~y$!ixnACjA3S0#upem;75i`|c<9xWM%le(=aUFX|9!6@)IgbwYB
z64=aSUz~9d@Eji`kq(aP(!G2LQAjNFVMHcI`9O*;cd{NJiD6f6lS?jzWq?fyu$dZo
zQXDWun}uK}f}aN`2;ks+5rYxL&aR2k8_{&Bx+vNAH0WKQa}>KIuQnNNps}0I>AcW)
zj451ho8gA<D-E-2$ifKz(l1`^mH2i4T7$7uRlp`%vG=R6AC|F}bGk`XZ)%%XoCNDz
zi~g)Inxxy;@8S&pG0D(zyn=f-fS8FFZmaJPhtJs@XoJ8U1dT)EtaP7e2kGa}pAr7#
z!!IxDMvM~$E&bpJ7=CAc={UVmCXjf=v*<E0Rp;DM`l;*lr?(&BsgWdjV7wRO9EeyY
zx**Gx)dI+8TvSL2mWU}@`w_F`mpB{7E!>eBlhr9}tgRQvvYmCws&eV!O4I#2x3seD
zlr5Id6(}p(Rj2eY1Goiygl{LW+r#@K1MLOC*jqJsuS$bnRASOjy0bQ>9;`rYq$(G`
z*SK`dh(p3a%xjMBGrf38gF7WqXkR|s;r!h)k4{*Ac1vC4aA#+Xm>F@80~TiS^4)2)
z=j*tp)g4Efgmf<qL-VY#R_`~=m1+0)aaz=m&3Q&L2N0(xOybIdZwgczphx>e5`E#P
z-t)QTJM>uEnYKR;^^QRn-G_0M72xraRS;_ldET#~-;yv}VP$r)FQ(Rb?Mw;9D)NQG
z(m*M|S}t`<s9N6ZLZqHupO)|6&gahhZk~%zzr0-#<;Jh4z=54wE4>s9QVHUDsI3)6
z3zP}oeLG;E>Aod7x4S@o57ig%sV1hfz1Bz1Wd7pV;qsDc2&F{q#N{!tMr`!HzahUx
zh(es|YaJ;T=T3xoPFsxRpWF`4UTZ2#lkm|tDR9VrM0-Qg4n5)~UN}c-las2+4Pdyi
z;Fewh_LVgtHnZ=vlS{8fm-za)Z#6Ktjzqct<gR1VKu66izUi3VOP}8U$;q=l1fn*{
z^sOkD4W{Q?J;0UWgP|I5p((&oMk(ZS14_H1GS)ioWmZ8#4a^@mD$)KKTI2yNsDuzM
zHL%&u0eJ)g?)1Y`4nk2mlgbJ?&8d`^a4hcA@n#jiW!qs<LY?^Q)!A$sKH6ZH-6HHP
zJy?9Bj9@JgM|)r(6FAyoO9fU+w!7XfU?L%I?@oQW2L7=*Ksn&NL2qDDh%*F&F*0~Z
zq2MZ3KiuoKr=<+^ePsLSBA0YQV_znwO*=+l<&8mEvAe2ie1k4uZq+vye(=|>M+6Hf
zr%eOTqZnlnjNtHH*+F@|T#DYD^tRhOd;#o&`d89WDS!a}mWfm`-;k7M1ib|JkSk8f
zEt%U9783$FPpz{G%W(3>6D{8t)1@%(u9Ep0$^DUm_!LA?Fm1*fP_r#dmUYFq6wHdx
z;OL?hdP)fu-34uJZ6uV7C>M>OT->i)ymCZ2?PZq1>sGb;%gW?f8~`4g*D!rTF3iJ8
zg2zuO{q!Z&n!Gm3;V6YgO``c9bAO|Ljls0#-1#95OnI)3rfK>S!;}}S6i|KIkwuxk
zYuN&tN1a0!cw{&7X@gwS#)w(2Cy>?R-2+QJQjGjXY9hIA0O~^Oy=B&MlWkYcRR~Q<
z;)vqIc`A89p9ru)sNZxnfEECWeI%wX9npSGg^ChgqdWPszPOK{z$VDi?CXv@xPj*z
zk*cXqL}R$?R$@UdYV6ZRKZbE#TZGH*_%vn{D4Dhr+URlQbqtZ|AKf$m{biNbi-yg-
z8sI3QU#_&KX0WJS2wdgtZxRv{yMkyLpo|it5Cm)e!8|@<rgsu|jBy7js5-C)cyHn^
zzBNznuektWOXF43yH{AAV_%ek5I|d~LzK&+z)-S{htJNgy?ZZoKw;Mg9HSynE-%$T
zd8R0VU1dDj@TWR5EW1_1;PaAt;2n<*ZMQF+E8<}9`<d`N@9Nf;{OJ?=rvB{o`@|i=
zw8V$^T|Ql~W;L=&UmFHIpAoowT(=f+f#ro5{lRQGp6QsFF@z7_7V{g71{kc-)S@L3
zi+7Dl`SVOE1fINV_`2WD_{oie;q;i%(1~-(f&U39N67hufts~Wa+9z9*^H0$oWEVU
z=v#TCRNEBbSKN;fq#N)9;g`XJF_SA8YLvvQs_)l*$Fuf+!W8hG2wn+dm`GD09rlu#
z#b%0&OfLA4TY@@~CI7+sQ(EtV!P}~?XV^ciZ;<oO(MT0(N52RiwI3}qpSO^HKO#Bt
zdMtl`eWn#Y!!xA0Lcr3S?bp-5x1|)dos`d&!u9bbk75Av@e?(a#nVPj>E};$BDAqD
zddUS}Y=5n(LPU8;rnDT&(dDPOPZG5>i7oO#Im)kNzRSIGoqXn2x{IRa1J%2tOIVT>
zns3v6ot-}W**+G~fQ=&pEwIZ1+6D;kAih{IQ{O7K-by*-7i^l)$Ju@PidfAB=QXBc
za1`)Yeyye8{os)<;0YzC9*?xdqm)aHE1tN<<GLkyx1Z!><DjaQ+~g<c;n%Nd?Wu2^
zQk(j0=|B7Zsp;vbHqX)jIZ8U=Qt%DRTQn3rP!NEcg#~M|HTA9S4_xqS3Uz;9uPT8I
zRm#aRB!Z(lulMNTguUl|uYcpF9dGGu(3tfSzIo50t9B*65<O`kH%)rZUHoHC`O-tH
z9%oZ%!f`(1i_;AJmeKC>av!{YcPANRH_QmkN}Dqu;P^K|H9)z5?9D(rIBy5u7t0Di
zDxZ&BZKk*;eRqoaE{BM8Ky%fnkCh7tCjA`_e=-(F?m^vh?cw_%TLM~7YCkk=oj=T6
zqj}@rS`NQ5+X0H@TDgv0Wv<^RGO;w^M)K_b`OR|7@!|N#b)qRgDP`~6vZAG6%zQ<A
zw(s<Ox*<_{8Ca?HN7x#_kFvTvar@(X$P<o`Z(109El(f(JHRia{v1L~qq{28CQH=K
z?jL!?%pNatxF#V`M{rM*w>_9vu*ccn#lcs--Mu%Ht#71!ZMBhGI%d!;KqDtr-N4{U
zAW7q36VcogRV-17#6vaSUHnUonYIIu=F^BdNdHByhE)<$`^<e14=6aiwzcEmrhbFh
z2_cgWiUs`#@KZsj?)CH}=L_MDH&kS0k3q?NW;c8>tWAo_;@5k&J}u6gJZuBGFdkB|
zWv1?XH=+cz4W1+(=9#R`L=&r30fDCE-e2g7e8%5Tl#^&z?==%N)9?~f0(Mksr{<n%
zGIi<njMk&K@3062%M!Skvo#Q1zryk6zJ_kS=AT#%qH-th8h$1p<LEVUGzbwgNH5Vn
z5$4uj5g|M1Qqk<?b>^NQ^=;EKt5@W@ml-~f|0b(03h#40FZ7CloO;#GUx`za{y}^P
zTkizYNoiZj!~1c7yaRKcmFLE<ao8Jnlhy3<4NUSQ^a!xJ&UqQ)7$v6`Kh<p2;Ckc!
zQRm0taLt`}g)U>UB+$~y2O3jCVTy)=S<i)^90jRQ)=(s(p7l~BwW^NfG3RbvjR~)1
z19_MJ`Of)sDW2Kfsq^`r)<QI8b!s~tSx8Xnlh{c9Mf@{B8mzm1Bpfvs|J)=LZ;X1L
zf?8=<)-`{<^H{rXh8LfNAAI6p51{>&d$)Ei*^6e#^+7z3SaZ?u_}z%(rLQ+EwCqc_
zCTX9Ivf;f-byU^4Rc`h%tN1}mCBcGVw#8NVE;sl4O}*B*<1Iy$c2R+^Wn11qU<mkr
z9c)!(C@a+ojE2u!Qcv1SlZt#wuI@rhQrPcs`OZSnr#+&x`_p3C+Gamjb?^$#XxaH~
zW!#O*?)zHp`n*<dn46so1@c9?zB;4pgZCYZrmMd>pXwG%*e}NV?Z*oERDYn|6WfqI
z7WQub?M>-#o-^dH>16q!|L&fr%!kbI!Y<FBG&M$S27~d~aNkLxy}Ov(jmA(BHt;~Z
z;o-A0Y|%qo>P5`A9SarRtgGj<#a~_SL@67tfQ_TrCu4e+D0*|)qR#kVrb@?*PFz#a
zw)Y(=K0%%p2TsvXPn+*uPtTMlB+Aa?xr_kDXe!Bd>UpSkzCm>A^|WE^OPnp$AeBFZ
z9a;5#8&tPd1p2N83~`MF-?ShyyduMuk!OLwSMEjZ;Y3Z)8-d~0&1%KLrS6!!8`$&b
z>Tqjb@u%H4DFUj|Zt>a|WpyeTv?z8jP(=qzrBO_ltlY(ID9MvzXhk;!aSoH|<1<H%
ziw!2#M4PTn#1Pf>H^bs|+iD*m_m#tlR1CiTe8o#&fw%lB31h%RpgB0<RD|Ujtxb#8
zfAhcL^)4Hv>T7zz8gID3>&Bcp|1=!lz$Tk5SFe{(L4eC{#6=WjE#1uT=i5<O%#_h=
zMYPP5(c~wJeEp{-PK--K!^XNXBb`|D@rSX64n0^WjIRd_*^;3~7&SzyUh{eRUHow7
zoBHq2Ekb>$CzMV1RaCNxsSBx-bT9GEL&3#ZQ_tVym%=A7-D^b%?Xl-md(7N9%mo_T
z8+kw1i{>~#nQN+$Tges+ujtmzi`y+>35>hS6hN$yF5TyZbGc`!ECvsr;pnX$N}nU<
z{pk%GqytR$U&XrQV9O(hHj{l^L&gE>AGpFY%?<rkB!Fkksbbv)Ty;2R+J^zPlV=0C
zho_{(v#)*4E?@7LbiT;+G3uuK1~ul4PbEaqC%vj(_RJEgaQ!k?AY1ZG{Rg5kapg8_
z<H|sNButM}4cx^QlG`OZBWT^|Fg36dFb>0!d*3ncHjZAmiuM6F1N|NKGoUN}{|J@2
z{wMEJf1}gxo$L4#>Xq`X3X|ub=SIGkqUnFWtgp{Jcd3uv%*yGsZs>N-cc%I&1HenZ
zw!r;F$XC&7Yh&21bO=w<hH|mH6V?8so2L5R72KGs+ysSh-M`-Zs4b3u$MX^MY+rft
zX3uX<u?bAdHW8OK&$`RDyy6M<a}3XZvC`R56`O0E#EClf4r!6P9U{?o28ZV(j!M7s
zr%yis9{_HXXFFR#{y{t<S`Q$TP8|C0hmM<xSqG8OY@Q<OVCiOEtI!W|HFJh{IIP2O
zv<3TPo-sv#dY|Sf!sHs3v@V3w7B(QrJ-Ed3g<EDiJu>|^@6wmWT<<J$LwtT!h|<@v
zrF{*iW#H+M+S^}$TgVIm<_*BmIfcD9jd9<0eN|<h!itvR%~oL*nXDo$?Zvp)cf-0>
zYORN{wii{mqGnKPbMPlRTOm}lS3dfgT60d%BK!T*M|JPLo!`K#?`M0x*nXjh?^X6X
zqW+>AkK0Jg7Tb7S`Q>(bOo{`pG-#lREwDZk@>e2Xt3{fjI$KBgb9RCkiI@dix7U9<
z9k)6yo*w1%j2%TDXdHSz<Q>2CAay`%HQW7hGpbl}Y63Hr+><b73e|v?e3tpvXLF{3
z`H{D($-BK+<vI1%{=O{{XlU^8P`_BCgL8*J=s(|t4iM5<OzS_)4MQDbJeE3Y6Emdt
zqc?u=N=E(C;1Ry$hYh1_ooY_KLg!RsuFgE3i!&S>S~`2fvg*0?N=_tc`)-%zc}Zwn
zefw26-txFLUYc$lsoY`<yh=aT`C5pvxgL)ERkB^l>)mX4?R0h^%@>N!QZn33vvAN8
zRT2>YJMq~U<ZGfxO^x_*a>ghkl6cmu5^WN7*ty=Y4ht$YXBM&xxs5U=4K}<Z#zd9{
z(Zce-Y&pv{u0-LTiuvXXm{j%*2vjU<(ZrYIGVX<g|B0f!!UoPW?G*QS8HFNc#5SAd
z-3P%9?|s*Fmw2ir$G&*u2`w}Q)2b9}dd#|<`o1-aOe^|u4@+_|L?HOhW0_oqt9YRg
zKiS{JiJ4MKIy7dFuc+V#MuuKC*UxYL!VitlP4%=U;*ggENS@I)yxevebqF|uYWdlR
zF|MS1dz;4pfh)Db%#(m?H^a_St(iCK9dY{9XGV|n`RuBZ<nZl3V%wJFAD55AQwsC-
z#a}8%+2vbJAYa*;Dmib7<+#9gdhqj5PgBLN`aS>;%AfhJoN*3w0!A~>hjB7+7(J@r
zAqG{gq4dc5obCSU54E}>q8Zc>+sEo2qn6zTt+$2WOPGGj!NBQH;pef(uxkw#tpxvu
zZ9qn42V@u{8>|)`IT}RI?^Dw2n+sHOrREBVLsHm!t#RVl%~`4$#T0=jA!-sg=!zNV
znxiC#Ar_ocW?x}i+4Cqscfd@i=z-bBCFuaf=BDErf@ZBHkfA2lXYZl=<pI6%#5`3}
zb3=*4qm$nvd$e4>M~cxJx_)BvJ_+x6h}EJC%6xcd)Fh61#3VC}u+D5bWvR_&G%+-l
z9?J}+37eK!nOj8_{0{Zay{8Zm^*Jcc2MQb~??;U<=@B#_El05`>XwQ#2Hn$<!PnOk
z-;R^__R)Pr_Q*Lo3!FF{_{QDq?U3Xj^Fn_ZyhBc`6?0_ySBuOhJ-JLlVGI#`N<Ig&
zxYSsGGz{IuasuP{^5Va1Osv<|1v49j83S1Rps>@)GLyZ3Nti*{o*7K4UrRa#U7gR#
z9?Y~#T_9)r7|cHX=q+ZKC3_F2qMm5IfOtq<L?msK>f(ot2zq8|l7&cu){og^&4Coe
z^<WA{Tr}Vj_IW}n8+nTIEtX~Q*ne8nTX_RphQGV}zoCd(SGl05qeF_ifiIB{giR$L
z4PiXcoL(N@ke8*fyR$H90N>yvo#Zh`{t%Ul>ng$XUBf^1$P4R7qt~aK^A-*$h?hkO
zXFRtrPP_l&0{A2VzldRhSM&Ml%sCpWIR#5stgqz8IDa44RE;ngpnAG&B^`b3z%2{3
z38S*->luSedf{LEjIiW+ab*H)!y(5)!ig)rPtG&@XG%2RCa?JKYZ*lgKMao9J=o!j
zrywIU+*)jh3{Frm+O#JZ>2{Wqk6jcFEeAHULyqI|{!aIrB}Y`@L-(LpW8HzXNns{W
z2S^z+|CCgV`FNjfVk22(uw%ow1Dkvdmd;9yTm7C;#-@RsnNN#=j$k*IBqw<jY<>Wv
zFDxrd$4tW{3EA^{9zHv@l9GLVq_LU1>2e@CdpM~A4Gk8yynLa@5_J7ACTmlk{xGR*
z!RiGoEowK%l5kMwQ`Z@!N~xf4L_Oz;H~7ycBljDZ=jOd@`KH!MLtN@!A@DO`Is6j;
zI(}!f%9T_CKi5vU;V%xRwB^^kFD^WY-z;a}pYPxrGgVT#EKaV+Q(LT$5Nm}L(>Ele
z(%fQo<_rq99l)(L4{Jh0SI>8l5DSb*k%2on;SmT8g2_8La4m~SNeC+PdVW^SBi~;5
zhV|11Gi1O~8u^>q7->uVjB)1&$r<h&S&nv0t+^M<|Mn&Cua{wR3C6quF9pzsTS450
z?WPe=uB<+JB1=fAQ{Q1flF(m@rQ{vDn)ii}kgi^6xMs8ueI|hTePnkx=v;%?Fb{14
zUql7n0zeb~SyvgUJY1iPY&Zb(TodT-8{ivU0Fyxjpv@xS<1B#Q8t01FOi*t6fZcXK
zgf>%jWQZE^<c*wtK;nthFSBm+>o@D}(ymf<gw(wPB+Dx9u(<#g37Jk~^0>s(<m4lS
z#EwW7pt1CSF)-kCYpl4Fx;|)5&`>I%8<+LlE<JD6DQD6pFYSZ-KDF*_60!77Ni`}e
z5)wcN$`MwrH3K+3D&_HTA+oX<Kz?^jOft7BLF16QN;(H$#Rw&LiO;H34BL$}1g$&1
zM!dLYkh!FK;>RKrM|{KS6;AHmJIT+PQhCmF_um-Gw(*#0+=~0(8>TiO1ZcF<@z(Hz
zrh(7H;ElGD8)_-~%`n<DY-BYrx1wpn3(@Xb)4WT+eX9z*CvRHJaO8aLAo9KB3<1P@
zUir9;PHzB4_G=KX1K7U7boJ^5$gCmTruwp#=@FHZkdUygtxfXuDDPa|`T6qzz&64H
z<MPxhaK-|IP(T`Tqq>nSmD%oJ??SHxu}d<(dg{&^RAPK*xy@1Wz4NWKGo}&Em1j0%
zmz8S`r1z{PB%N3B4^~oD;_^11uk2sp0>Uh3=R7O{Y4KRE!~UP{Y*Gkj#k8`QW#n_c
z@%X3%&$cj0{B8@K*L;}Y#@^GOVx!&C!%f7P`Fqh!VfujMMf1!B);5lpTlxc6pAm5R
zU~syV=JKxsd4}<-J8#;GBd_lxwRcN-%3#GwtBd6Z!U0pd>-o8=lca$l*x&rUd5uvy
z;eJpXBf%?!)IkrrIIeF9hZ6oiV3^^V2|_#`?Cq5t7MHOZ>kl~%cyZG<xWK;HOTg@H
zT>sWLsT9$)8O8J{GR%6wO_MoSUwHHwCNDsH3<8S;1`0&{&BgXhvBto92hS_yFmrZA
zL3;GpVAni@<&8513)hwrvsqN+`s_CydZP{e55yjMBPD~CkIcF_ql=clgoI7F)2BfP
zLI0q7`1|)XgXXy)b=$jz7O(o2&u+w>4BfolvGJ3pm-%>r@l#DTveOCD0#*E=P{+5)
ziHDb)%L_31<aR}v=`>k$C*uFNA(m;^IZoABpx~XzKWyWLx8I*wl{mmumK-(>YwW{z
z026a_X4af7@8<sgepuw#1T#Nm*o|sH#+GrWn{ZQFi$#WpDuh2ZGJ4$NY^ZWIP;?Y8
zIBdz4E1pU~VaK&o<Nzv0V}+`<)6eE_bVL}&97}tQ$x^NWe+}WD>qc_K-+X_FZ_Qik
zDv<a&GhHgk-Eat9^v<Hs>p#}2&%5rmUZ$!W>*wrT-zD=?@v^(f#rg-h@Sh|mG<%IT
zJg!3NS_k*!$B!RXZYE))VDAg8bn=DE7nTblE25Bl96*$Jm6X1DpC20_m`c#%(NX1A
z!*~c~4_q3E!yNQ`L%5a@f;6?Dg<)Z~9Klr$Y^?fD(<3v7|Jc1BaU%wlPA`oN&-+Vi
zTLeGw1^H?9a-Y-p%aVk#37zL-%M7!{Ju?o`^6OU{kR#ur_36A-jH_WPpN~D8Ysv*C
zKBYGp{NiL@cEfB()6}nb8miuX%CKXO9dQ|_ml1o7RCRMAJ<cN@DsShGA!;!C3E6IX
z0RM#$yO6DSkk~+W0U-(t1g)4D7$n`E!ZIbqz5v^sr~qJy?9qX}SXkhypI$&v6tYad
zuxR}uC;4JE$&2#q#u_g+u->NYU-`=lS_p$l`m3mrGT1nIr!7$f==CPQ4f2@%Ki$8x
z;w0i)mOH%=7;X`|7afj5@l%BoQCyLM98%?VG$L*q`+<AU_H1%!-$?&6qev^*PvjFz
z%s_LwWZP=y*IhV?hxk)vmA_3lCJMY5#^6`?SSjTUg6(y?BehBUQejeMk%{#hASX0&
zxQmER><^A-RJTx&R7ge>ODjSYplKOhQV&KLP8yg-<9iEy>}c^(dFaZLBp33oDrp#K
zEj>gL7=>stfndEG=tXdS#xS!ua>1!^;QQn6_LGm-?Y%bH#=~7NL8tg>ci-5v?q}_E
zYpvRbILw%k>B{{${<Ma?TXInone()ySCdqm4=lUgH7}TeDEC&>u~S&Z%h)QWv7NZ(
zLUtBbIdW=&FQi5pov|X6aY;QOYufpNxk9y-T!Kz;RE4$BfH1*|S7D!-`|H~x<aAH3
zoj9IL0!BS*Z$kTEy=L1jMNbu{M!GPe<d4zUN2b})R%D$WY(@SY<`PaZqS4E-^Poi5
z;u-uh?y0XD#|yzfrGz`u31A*cUxlff9$ld%(`klOg(_W*y31@+u*%gf7??d=TBrTO
zO^qy_>|bU1-S~ZQ>|4$}7qJu}A?+XU?Q3dkJG~)jD&$||C`?8^DeLh)dDch0sPFQ#
z^VjFd2%8|jI>%uT77w@8`x?2@UUu*(a}|oeApJD$rhwF8+%T{C;l*!Zk+qAA2UnFg
zY{atP-KA5xU@Y;wI^g7eUPC)jMIh4`prL`~N#G;tLO7M!J~#I>`2ANri}9fB?--Pj
za6~~HJ?iQu$2WJoO2B@uQ|?oM_H(RiQ-Uu`IPrcOSN5&AQwuF4Q`)jgY9}S+rAqep
z$#&~tC(s~3SqPXkED3FP;@5H}F@B7{7XS_8t&;)y4Ar&z!WC45T?12W-7^@3_T})>
zv&|G}wDXet7H)p8=v<yl|72t8T;T3&9jnvS(aIiIdIK`#f>AR*nHyNMhauDUH6IUI
zjYTV*wMc48aJuT!_eGbgf8PC)^77<_UnDEde{f%h(QmbYA?X;^d$~yZ{R*uG!E>p9
zz9bF!S*8OxyVJw}yuK1ke~{YSI`^lBOOcx-L;3Hh`S-m`Rsyz1!L>ABt7bf`vZMUD
zT*Cli<4(W6I*LP52)t}XkF}}2n)_5#yE-z!n0L1xg6+Gt{-YbEY4+_-pGJDv>kOU+
zStxhEtN1LK&AN7|8Fs?-O|9~h=v4!;?yREv+HQpYCRT!J&SIDQVr;`%?u+l&_!QXN
z`59{JukgSZ`MCcXp1LnXce}fObmkRI4=7cf+~E~Xi~MIAA$~jOu3(?W`GcA5fy(5<
zyNP`d19Y0glEkN{uH}g;D|MzN=8|8Duoxu{?-YA>n185je$g?nn|iBQb%hmI8h|o?
z>f4q5C!Ie?G|0awE~O<@k9Et(Y16GLiY(6<H@6*3e$QdP6C!v$k+n28|1o8_BN3C;
zR{~MzSMTeq5bYjlI6MGN+KnUXAC*I&HJBaa;3rSLa+H4|q=@U*0(__M#QJ@tV0DVC
z^N(KZn+d%Rm6c3gs{*#3&$b*OgW;EP6{pls(c~$m)<X3&+U${UGLDkh-MGoG$n{ZP
z-7ZvJ8QGE(ITn=Em?MX|ZGTMxEhuJ$Q|`<X^S~l|TlrRxoDghR@7##PaAqG~tx=Y|
z^s}0}QBW_+*KF>_FF#T+gk1!(DFx$wM+=UP0-L=D(P8O8iGSttDq%HpusLe!soX$-
zQPxL?^>Lb7^D{CBZ+-n;`~{SfZ-G0$@ME-;EQYbDx~PyP=P?QyV)gRUmbKI@hEbNB
z>JgN|M3+sLvNoBNe4RMB)3f-KbiLS;-G77A#tkli*SRF(^xvO4$Snn)bgZSQuo#8u
z^k|jGfVb6ho?>&cLM$rf2>q=3IsxGYv<-GDb-VAa8$UPWdVC&;-4%*qvbf}R8&fIs
zl685Fy0`5SI=_;XrK+lc;fjHK>-$6{n~6H-ToFOWe|8YK_9qdeO?|jYz_wAq_0FCc
z+oE==Z{dL3d`2`i_+z|*){Bq%+*G&G6VU886nCC#hnYoqGkcsHEjg0Fk05ES!f1j*
z7S~2P7c<(J84I%Z#f7N}FboP|?Y`6Syd63!l1Hlcv}Tlh#)g3GDyj5SD&$*D!6i}t
z$==P_wsvKv=%staLQS%l-9G=PE3`!CB-Zw$H+2^rdwJ$Jn%IQHsD;s#vR`g1uFlF&
zJf)dIH4`(*n-9^KrQhMGy?+CPU#8{Vc_#)okx+i#i*&SOO0MvAAuPv}a9Ce>cL+CN
zBdahppm{>m4wsS<PqY(1b>mjhk$sDEA{eI_JH3K~3<$oR1>oO`L?T3@w+J+xNpOKK
z4ywwEQ%WLbq3I(sw7cET)<gtWRavjxIYcq#XACcB0=&wLAW^UWr>zt=ir|)$l~2RH
z88~2=q(?HxO`<ZrX=F^R-kdnj;17^S!fKsYW~f(E$Q?o)^u_*q1tm)u(TQ%QoQ0!z
z-G`e8<gOzXE1d}pYR0*jmAzeGeUX-W6}K+IzI!SHX4s!=_tA39Bst<FLV?w?vO?q*
zULCTxq&$U&gL>w^ue!z7n88uf{Jk!DyfpXnYi_OCEgSEhP@7jcG&<Kby$eMmmMX&M
z+Fb@2G_r1i0T~~Df~wWV7!H&+iYd*IZ1dsBy!x!pOe&9z|6Q9-C)TcJN7MYMA2+q#
zY*;Stp5CXlOX*dT^0L(7n+}=8);-dB$4~P~S30<8N%{1cHQkY(WybX0<5w(MYfBm?
z>euN`DdfG6vs;*7a<ba8hjXopdb)|laSc_hM0@}91}*_zOrPNK-eABwX@*b;!`44J
zlN>yz-R`)T8dsUKp0U(UYOe3kmRggsEKhLBh+*8iE=;8NXBfNH-q3nY!34#N<1V0I
zZBk%&SO#O;Vtb2Q5B6u)okU-yy@V!w_Tki6yY$+UsLOI{Pb(eRd^tH3nZo(X>$IM#
zMi`FH9^XP-Z7_G~R|qEM#>*Nh8XRWHJnEuK8!L?pF$q{!mUjDM^Qp?bdSu{whb<M8
zgvBqV{8Y44Xh@BKEaD2(c{S-PCH!{?Mz=%@7nZ;sVL4>)xmUDwMJK41`Pa@h_Jq&z
z>N$S%%kRD4KBmAT)ZcYIP06WeWNaP>zrQ=K>emmq7_CMv@tH0kSK@J>wEcSYsPU4Z
z%?}*mUco|z&VPnJVPM`xsZ(v_shdc`76b?m&iNCG`+&t9*_^n9iyLfl>50QVVtF2K
z{nqx<v+S+gX@(CEABu79Do%B>$x7Kf@-J4XX&$ELNF3pVZXje?%mLa6I;M{-PdXki
z{ylZ#_AgaM>~3cZR>zla;6s0^*w22|qu!TE<1(k#&ew3*a?=V_Uf6#w0zpfZA>b#F
z`gCq-2@`c~{ZP%|*61}D>mtii_<aP{*57;I(Vlcdci7_@*m!vAaZ`zl)z;S2fFnAn
zxa0N5jxcU!{{QhF81H{dNE!(sqN8PzLbT8&7!vCl!$dlBo*;2C^`++!W6W{6+Rv9s
zK-&gm9_%Q@0J=ZeyTm|m*UP9#V#_m|*}?uaj3Lr=&Yfufd>8Y(XI#Z0IT;aaO~}b2
zK`4dF^06*vT{R8ptP5fAAi{&;v=W@<bpfn%Mh~*qCi3*{dSRG3y&z{ecYL5d6#0`7
z71eI?u-pO{h7nj13&_q(Bw5VK*)?18Lp}E1>E~(VeC%gO=Pz`7RBeP54Sas@ex-B`
z_}0+aL_y0jO8xOCH><wepOtO#ri{oJs?%2O=pCLjzhFN4li<G6)UW$EjLUa)8f1B7
zO6>|8W4!c+Tf1HMuAeK4!Wap7NldVe9!$8Ht@a7vJnu;rvyPipZiHhJbj0mjT7c&T
zy;bFzEz#A-*G21M&%-k3cV{mdQk+y@q%sI^fvFKf9~0nsq$gz$u6e$f9Q`8u6tNRr
zI+8l)9x46@|1wKQ%8iozz)L;t0|hu|_?$@nX~k;X|0a>PUb84*UYTfH?bQ7jsUj8G
zFK8PPyU?=}SX(MW|NN9#X8N)J+S*g_w9Lcyi?86nAfT(^gTigDF^{sY^_A$VFIQgP
zz3UG9n6d=lzdjX<EJ$9$(|Ln;XR+bfeVCp|($13pXLa4DlVQ>r-f3Ndj}_tVlY%2?
zI|W@^MQ2B<9>Snn6p%OqJAAHNE572OXL6WkkB^oGO}Di>n0EHE7%gy#e~OJ>S6E@Z
zO@nO9MHcAPR-Y{yr*AwSD5N_<L+L$wTRwZ+kmURPBd*@E<*=f;cCn*N_uwmEdo0i|
z+2XQ1Qi>TfJoD=^RaJDIH{Z@K>RJG?hm=A?GeaP^b?*M$u(rkbgjP0mrr(`Ov2{*f
z?zJCWCR2YOZ&DkW@ogY`g(0s@e(?<TLSgIkBHbwg-EVz=m)@VmZh0f)gBmdtpB7W|
ze3UL3$Nw&jw*0d@NqkrNsc$`9gXhcyT$cD2SWyWVXGV;(#2*9;IpQ7h+vr3zy&)x8
z_BzDCBB0IO_gsF@H5jAJXjH_CO{Be_vq$OeDn}G#e*A*k{7$$M6Iwa;(-tg;kv2h!
z&xF^&y@Q^JR<IP7;wunaQn(Q2m7Hs{OqORhN}1D}RYHr6Mc~(3I7)q@v$eYy5+>Ky
z>Y}0B?&fsGZo_R6Q>X3r78TIasr^l+1R_3L;(sR~TBM8(vr#Dhb}IqiE<60~Z~yhT
z6>V-_AEJ}{wx)YjS>@wLu=M*Ip;1763DtD3HFuLy_^4hTgY#ziOt=XCmr<?3u#AZ4
zpPUy>z8AcA66pS&j=WjQ5HplDo3Z&ofMMHfs+raj;<;#J5!=gx-@+GK8%q=YvC{9`
z(SFkOdoXh!kq_~CJZAelIr?5GRQAmF_KhZ)-h|lC*Chp_X!qL+za04klKK^*KJTxg
z-7u&1!sc{$kDUA9OYl}x@Hx4*h39^P8gf~QrBgNNwhg@%RHv#j$@*=NIDDMbf|>dU
z177-nh)PzJ{qnto{HZkTsQL=I#Bd}7f|{ipyGXx2iEuqEyu$d3_lG%tB8B&Td4gBX
z!;346XHT^u@a7uh-mf?_^T4a6pG9QyM+AlLKHt#=nE@yCWQ?Q<*{td)4@Ky|v|#l-
zzK+M>=Mn|8M7sF@liC3Iw<?R_{KIOv5QihZqx1qlxo9nnhbI|QVj=l+fNjjpzB~4x
z>Ki=<&TDuIZ@JiY8EEV;giW>O>jhs6<Tg(Hyk?vK;pGr=Bi*%;K{Mns`K|P;r82_N
z`jAN$@szTg7sKddO?>mGj3CdKEB~F1!CG%rmzXALpm;(nBywoDHBP??&;M=r`oP-h
z|JppBf0i8f!Ja&d1TFcGbY}~vab7Q;?$KtP5l<sKh?`RXzT7pgx0<1D;KRs^^QwsY
zyQ6Y-<utuSyD&?-4#v3v^K6^`lGAdCK1Ic86CcS?{ODA9hfPL*{oln)tf}f{%b(L~
zt|hwfl@cR~R+g~d!3Y{*M(N!(6Wx}@PJKU3kQai`;V1o0CBjHmW`dZDItny#7n*+Q
z9=fp@y8db8sNK1gx|V?EHZ2prqNXTFT`E5iHF{fuMW*SLI#vCJ&eD&I4+lodDN0(-
z?xE<~<TU2)1ZZ5vcxS2zw%1TrRKd%qrh}i;70gx@a&2k9=!mS;esTNV*eU*z_9peX
zo8|4)b-H^p`gX7$d;Jh4$?44JdQZ=b=Y8er8@Uas_bZ|Vf7JX&=FW7_uK_s=R&@f^
zm4Up5;p$Ltrm3Z#T2dO#zu%iFN7Bee9e#C#`xj%Te6W*VGPP7WoG0D_e|-GKY%fRG
z6bU$fc{AU<Wc0XtSU>$g7wD<-tPo||%~oX=TGo2*DWAFPgtdJyG8DWZzTBZ!iCJ-2
zO(KnBK+tNRMM0nc^8f5m%AiWTY^_Ag(VFt}qNeU!&2A(6f1McW<!aS2I*+@(?1UNo
zDlg!415b;(b|$p<S0RKnpA04JZG0cmX)Q2YdhGxcrdUUc8O5(~B%ypD5x$u}8sy$O
zk+CoEqyHCGQ;WHDTxrGd^2!~Y5(C6Dg7q=3EL-UptcKL@Z-RVF%2p**WX)uS1(U5U
z8WXK8h!-QYWmb8xUf4y5rvICPrKo<>-L6Cun*M7z87)o)H}RK#hX9(tzw0MC9i?hZ
zE3(&ZV=JhZ;m!V8ySjHKW0pMr(pL6|(wM?@AqipV^~f-NnzO)smQ9L6kCigK^z){^
zfV=K~9iKSD9dT+=NvYnyd#4;%8WR2C$9-FG-ngd7c)j}Uj_P(fH?g;WdB3Hkx-6zJ
z_dkE`YGRz|&ET+U!}WoAOwm0zs+TYMV<nX@kY6+zeJ-vwS}(k8d9fDk8b}Vo>E<Vg
z-c&aa77kc}asOQ-Ld+3;RT?A=q-ym;Dml~@n5j?*(VH@zD&U5M8SkDiA5(g${&SU3
zA{-7uC7rx|PyKIH`>$A8i|`r_Y(Hw-_Ouc-lx+$9P0hJ=kN+&BX$w|qK45O*qrGUN
zZ9VxL&cc(UI`deu*NOTHzsq@4wPQ2geEyk5*aS?MY4$1#T~!#H9zuQ|#=wCn-Fv|=
zLj0OmP30q=T~bxabT4@SZLFHU9$_N@)_9dHX?r;SM+l_>blq}^>SPT+EZbWgXY}4`
z=X9*Yyoy-xx5p68hOcU)!BW>_gu86%_2RRIJ6FZv;}&_<*I20W*fSJ-(7w2pZzaSq
zqQTu&pU!Mk8<Cs;owxO$j|}Zpqu^!Q5k2PsWBUygs!~Fp!1!#8aRLuW9J9ykINA1?
z%c4Y*iCl?#zi5SBrwL+v@Nhw;{&+f3X)G9Ps|pqgHZ|spH}FQSNZf&U?_ZXAv2Zy1
z!Ju_6ej2|x%nw=VKjMrXQ_Q`b-;MQPy7^#teM>O(^FF>Ud*}9laeho8Q&QD=_0{;v
zB$!oH1TpRsWa&Sn%&9!d`>;$)NpfFq`tDY8cp4u@+Mll0>bhgp##nMMlGkvqizgAZ
zLGTnKB7QQOQ<A?=Z8^+crIU`EK>p^9;C8KDl5qX>8Jk}yqWNz*7Xog&-MR~?^4>`{
ztKH3;ky)c^muBkMng6z(TX>va$934)>wz7dRzwdfcs^psho^RtKVLiL|6e*jtX?BV
z?h`(}Yn#p$$NX#|1Z(35$Oj-662>@yJQ6~x`U>NJSQ!+7Y+r>1svXBSBbtCxi$FZU
zv7ZjbDbfZYkf$X+Sh(%_tFUSV!d9S!J%P;@536jiK)|zbBg~3k3G?+K_ZN1MkPhaq
zVpR17MQw8S#-|wsnpsWBm3Gp8+-oA@y}QNZC;AW~FfVQ`d>bjCY2*DtqKb+R?*A@8
z(f8P*`qq?yOWyC5=iRJZTtdGmKgN!1@h8^kio82(Q$|(;252N}cE0bC0ZJ(FS;DPH
zQ!ThcMD3&$3)kY(IWR!XTg3y6ZfHmdvakxlDM~$k+D$YITbCBVFx~_#SYlFAKO*r1
z6etGB4%FFI-NV8ce=&@$AV5O0#}`6APX;pYf%aUrGngAIXqvY5_=To>Vr-4X9PN>q
z0b0Z3bs4w`E?4ZmIQn@VAplfYSAXZ;lRn8C`!Vo`%Js2QQ@{0mq7z1FeIOu<$ppv4
zA=}H#;yNhd*jk*DoHaTSweZ}zLgm+k_igW;cWf@4&(-`h5pl>d>doTn4SuUT`$+7c
z4@3<g=on~xkk+@Tja(4<$AK^5Ee=)|A{4?&11PKyM8%B}2oo$VdU5<)7$J#(XBg#!
zKt_vex9}js1PTK6!?o%mEzP%CMCmy1cPecL14z=}X#`k*`R$*tz!MYFta9~cPJaE(
zFQ5=QE_XG4c8c9azwNXd0O~BWpdcBX-G<Vv^O=$5Ll;dp7Hcfor|wt{(ARFJR1dnq
zhf<5(3p22J%%y=Y!gZQ(Gx(Vyz?4hItW-_Z_aDQwt)xuM2h<!yl>|OrY}LK1q@w31
z+epY3HTzd=)q<rZeRUo#VBAj@^B9>oE(*^xPh>|IB4Lj=K)bU-C#rPjK;x$toymXJ
z|GFgad$RZOo(+&hbf`{%6o~JAVoyrT(E`Px{lbBL%!;_$$i=5x_N2b0Gzd2_Z!pw#
zkuoC0?`im`N;b~)(0*G9r}<V_ir;*N3FBAh*5GX8rx(eYesmlxO$_^)-dqaH6|s7_
zG%X+PFC`ak^NRSG8yD*BS%P=m$Z9tqKm*54V4$PN?3<dHyaP)+`~K})x4w>!YP*G9
zGc`43<;d437B<LvGI}4LHik2k{v0(ztf4}}xgyymHZ<O`oyEiVY9IF(dvgdnetyYp
zt!?yeMK8*sigMS}>*H~T%N&bs#Es4f#xFVS&vv4oYK;&M4UtFK*+|LXi`ruuK`Wen
zve=dK8BF8fXEx+x$6q0|py0i^LlR@b9;>(G!(Zt_neZ#tUxQgE1=DA_{NJ_h<4qrF
zz@D?vP;3+~m$Z<1iKEPq${^V(%}1>zbf$D;(95@|fULOrLW4ws*y_@WZ-HILCc`>o
zBKr3oQkA}MRQjm|13k3OM0Ad5AKAxRL&kkcK6uy<SD0;Nev-(txGc@yV?m8W>*XJ<
zJAyt@WvOE{<xJ&oa6&eXtuJrd*O4nqH#_>E$cjq)#s(MZa1)85tdFC-tc7cGE?Tk}
zxg5FG3=4!NZtI~F5)fQK{j+#+4~7&R>-Yo&2uoJuPkg&cTYI}IT}FGY+cvDOkClK}
z0EO#LMNs|w?sG?XLO6x{-p>;~XPZX($GuF83$>T$hEmj5o+fbqg>PiXKu)|r!P9&7
zd!PN4t+DkTue>{bfpyoJo``%EO?Ro}g^a}E`Q9H<(Y&oAjD!6LX8dnC-@Gg!4#s_g
zn~#12*Yt+QJG{FEfl=scL+pK}?C-hIx%^-gp0XVA*adX<8uS)PCI+d>kHRdB<h}t>
zVsD=cR0B+|UD2#Yz@=1?)Iq$rNnOC@tp^W<h*z&iPbkK*?ZdjEOb$An1k35TJFAz%
z7-%$`eV$Nj=U_7oX4@;>b9_<hZhIviRm3GEXPf614;y~RzvCE!J`J)oATz@N-Am>p
zr0KETMOAyWnve2<U?w1>7odHh6nVygngtN-tQ+!l@LP%NT#D+{>$Q-#HZ8L(Yj>~t
zEdG2i>m^6R^%o~b7kk)8GrqQqdU2R93B`MrP3yk(zds*@J>5jC^8C?xZc)EO>L_na
z%tK|8crtCq8FR|7S~FHEa#mjOZrl81?%7{1x-uP9udFoWt?O({WMZJ?m$c%ZFDE>L
z{;BdiS#|C*@=UONiDn2Z5BUWIb|qlvATxw$#Kpyb0Gi?dHTRZLS#4kZ??ZPf-JwW_
zG*Tj<(kR`b(j_6CB8VWM0t!e<NF$AOs)%$qD4mj$ckc5W|NG{S`|7^AXN)tzq0h7T
zUVE*%=A3KJ?+5hWUO~A_8VK1YHxj7=aqgf2osE?hX@IKVA5cMdpCLPuC%vhNaHJNN
z%hgheBh{uECw~9QI8BeFMv`f1TiH+C3l0IjflZSy3L?j!q^&E_`06p)>_}5TO|B<v
z4D-$S)gAJ)iM!tNi9D-^(o)#h%-8(9MI%%OW9t)-UD<S=9r{PyMW>d+?pRxHxv$AR
zwO8sSlTw6IA0;1mc%79<(5X?;;8~s}l4F?!ORP+n3YB{688I{h4P8nFUihJ!p%JDO
z_S1S0fYAJ`jY!gKf!-R+ofn8$=?P;HsuTn_5KUgFb~sV$12r#?;|8gy_FvLMj)X-8
zKZu5myVSg+cD_+8&UfO+UI#9D)w{;WQeRTS#eV<o$@Bav*Vuw<z8gm}m$-xW&a_Q|
zVYW}=sUF|*-0zo{FDLX`BKJFpUhK_1SJIn4_LCB+E6;=ZBceE>qiRp*CFVB+>nv86
z8uc0!uvKk#eNgmW7XSYKK}6W+;9bZJ?bm*YAo6-nduM&31GjhtjhJCL(dD7ckA+=S
z#U7>vf6dNyT>@2c_0j@D6t<WQQ^57lN@;5oFEM{k5ou^-5BIANk{66EwRgH*7GfHD
zw78@%NeYM6UrfCqysJm#%Xp&8hrM{$@GL-ObX+%y)%U${^$1zH(oMK6pCUfgwW8!=
zFF65$o4AIUy(|JxuUkJxiLbY$1@Ekp8+*_=vtZ?mI4Dp*h?%`2C6m9fSGUxsB!&vQ
zb9=!rD$NIfvdWlT>S7hIVV})<ASEs1*n;sm^JIY~W2wDr`C`>YX*7>uOm$UBkBEZN
zIC$W#(Hz~S%qPZx6D#}fy<PLJ5YcIc@jL0YBGJ%EAK`^C81kU*=h62ke!ZsX%{;H?
z42TP(J``&hEs#Ga;Cs7Sm%x{lMwVXZl!~s%(n);KytwF_(akpV#BJ)G6McYNV0OGr
z(1a(e$se)Z@C-Y887*RpwTBaY9Bo<B6yg=zbuHeHT^18rm33#$G;K^Q%53*#ba0}I
zwWwC~({S`O9=hYyrm24UCv7<@>!e02`JZj4gk-PqG#8;sGbtOl07<<*S|*oF^|(0=
z>e`mfHy64H#-k=25<JgZr#>fvk9*d5i}N8jkL@ooIE9~ZY6hxd|7~1<IVgp5c{kj=
ze>S!GGJtca*L6KM@p=70RY}>|0-Be96J^{Nj<eq<`JR<mP&zwu6GYf9@dET61>5{J
z`CBylPU<|O(KCu!_6HBlrp)~rL2&=-V_4$OkPFvcU4BI|tvK`gUmq5|yj!nB^x5iA
z0>#|oox7#SV)XC7BpxL2<%r|u&%G>>k{v8r$xsP?(w(^zzbv`=PaRD>KF6e%txAlM
zE`4)UFN0EV+VEu1^HDde>rcnHzOeTuiMDj~Dm^bZRkD()#Bh5;?dk0@q1k)$ezF~=
z8hydW6yqa)+_sO;tYW_ga2@~k%i4UgzUet+Ou+ytLyWdB9C(r()0&*SRsKO(b!62J
z2IMHmoy3hpZ0!`+?XPd%a5+r6GwQ8P6t4{Ajht*tcpu+>Xlj!^UeR2Zg30x=is*Gl
z<a$^0u_5X}`k3eD{VW2J_(N*)SAE0PbJAOC!M--s|GOa!h(cLS>Z(_j-bdXRC!XO8
za&rtb?<iIC;12jy+lWi_skOzBVLG3-PS|exSI!MAl$V0W2%q(hU5dm6QHu%e+?_dm
zc$PckwbH0H-<=P|o%5@kk;`e?%yuWL$DJiFwwZLBzxUi_lA<;F+>YneV7bvKYM=KK
zmQw}EJ9PFZ?maK)ZZ`e=-bPaSm}EneUDJ)h$**x(hZl2LB(;%+$zCYh_<?a)L3pFQ
zmsns8$@?H??oEZ1DL+l`vk?DL%jO4~_J$5DL!lZT7W|A3fe*E_i=`D=Mrg#eC?^MA
zkjmeiR6t`Ev>J2JmvTh)&)4eITpEO!+@(Fje$>h4lp{;P@fNpa&tYY&MPn{gCW9nT
z9xK7s%8z$y>api^_@PfxWs4wxRqQ?SEJnY&IK4r0^}GkwwvqBt%QOieC#@K%tIKo+
zVdpE%A{P*}x|9-dM9p6$FS`Vu59Ks5lSv&LYn`qtExDuV{0lpGiKa$-tSx2CrnGXm
zw%D}iPUIf<h{ltN$A;Q;DJPZ$<-EFgO)Jx+#{_427&ZhsC~ucPyR$Z3noUf!OoN|&
z;nAy-*a`{afvSF7c&KeDfzKKOQ*4Wxe#ht7{!An<4=f2x(jVJiD|Kq;%{&n(u9LEI
zds%xDCow;A1B|fwWR4rkjk5UaXM#h*@j>ja2eFR6zjf%OC^t;7M@==TvmZ-icgSb%
zv`+_0+wOh!XXgpF@13W-D_w+>KEwsX$w&z{N2qE!-qq#rPMGtV-M9K%E!mUiXMYXF
z`j?ld{6LrT_^&*wuyIU&C#X)Yn@8GMc7Z)CM<&{#u03h)TZ;Gk+mpu=<{}3hk*o@z
zMVrIoMe>MVXEPgZe6yc#q2Q3JY31QHGD^6y=Cz!@MQPL!K_vHChY%)=KQdw3wZETe
zoMNViUY(;AMAs1u{9^yhJlxf}HE^x#WbcFSPyD#@r&(#!nveS2X;zk8sRF5Uf^&0Q
zy<~OGE7CaInpHM#I<|Go%3XfjBC-i>yS+AOVCG=)Z3f<LMd8?D*x^;Wpa%s|6=Qp9
zgBh!TZOB1i@4CG{VXWKU7n9{*8pgPvp9$1=^2vD{vXVbum}r*D@S&pDpN=pR43Ul`
z6Mn{K*Zp_y6CsK^)Y*X1=U_*^c*B+ak+J-ri!jl<ON1D#W!)7zuBI;Q6rqs?vguXF
z#d%gdA|9fT0u@VK4RwEhE3(Kg`P_*ei}+K-P%F(GbGyuiNuVW9c?3pUd-@D7+x8aA
z0b%RBJ;}~4owMaM>FKW`&bn6%Z3Se)++-KU3@(<wG(Wh^JuXaRmy)!YyQpmMa!Y)s
zKN3?FD=>_X2xR)l*wIIhhxL&D+RR7S-p^D&_@=@haE5>VTJv~<yb$-nw))QHV$(X8
z3!2gG<m|ce6+2q(e*2V!$^a?wn@>abj%exR5SQr|2g+eFsm?4pdvA5gH~!;fsF~^d
z6LOn>`P7f>UdUGx_&u_VaK7_E%%{?gMSixoRp{;CHGF=rMdnhciy1Z%Sr)yCZU^BK
zFzV!iO(Pq=B8t!APVY-0BfL95EI<+={s>HO`8w-6|C92IhuIH^_4H`Cxa`aYiHxbD
zuzlZ+$lAYR%l<fjYt<~)=Xm_Ucd+mAjP~f;gW0R%Za4lu<E!vG3uPccsap!-Rx$<|
zGBsR7R**~7hr=(-C{05{d?;jtQgXS^+Ex31^oNGdCkj831a(}vOIZPRJT+@MYd_Cv
zCf&b8dnfZzjZ#!|e7D3{^*a3zt~Ct)*TW&M2Wa7RX(eWtRItA_K2aGO9(=9;ucIjj
z=N2XP^r+)@?yj`y^Z1*yb(vT9%!QQ>_^>a%8~cTvisg(wq-hj!X3N&~VNy*o1u}Nr
z9n~MpZ!lBo?q1dRBk_|oV)*ZNU3kgdG*KMsl{i82M4=sL7_+f^_hGda;A6g~B(7-3
z_4}j3k+pzkQ{dwKSf)Aq>ay!nWX{(95=Dbx{xh<zH=K6EadV2_HTS9WN&2oa1|RPf
zOw?JDbQ_yIcCi%UC@8#i61zYP;2sbE_S}}3;5oI~s|_p*&*FUxZvVpbb9{eU6@h!%
z@p6!~aW!OJN+)-rLq@7yn3<CJqT<vBKQQd=SAKcDS^1XWXPPM$l=b<2hds?Td2+1t
zq?B9XCbu4;a)ZZl835WDBRop8a(PVgj@{qJQ-<-%Z)M-r$#Sohb8OW9ZB^YTGRn$@
zM{5Onnv2<bS<S&2%!^lGPFk2S1sk<9hGWYUeWbFn|61h9m(u>yuz^(m+np%%X3@OQ
zr<EcOt72|RZ|L8djm5P3cQf;8mKkfCms6-rpEO|wH-!^+qM29m;M*4S(;QH!N~gu-
zU3Jb#K4mz*?icjn{-?ZCPX3Y>D-21N`qOQN&DkOMEEm;>(Zs#=(Qk%`JlKNOLMLir
zPdQaYGpBpB$>h}5Uy`5fy)qbjo3(6au=q1N+RLx}e?=Q{eV0&%TwJ7+k=dQ)IOH=e
zCvzf3lscoh6YnGU`U6Za_=GQyq(1ffV>KU2HYsP<M9wXl+cKuU)w^*lRZjjZcs{Sz
z7b6TuPk8NC-bd((uP__Jts}Z}5`N2wGR!2EZs%hcKK(PZ2>zI_mVf>}GOl%yZnn>t
zOJ5r07QK7L`<{!&1dhR%jBg$>*k*ct-N%O7Q}=G$PN<7hO_Lt@Zz*F<WwEdJa@VJ%
zV506MtrPCV%I&uKdHK}oOIPP663DknM%2B|VXt6xcHEUF-!j5h=L)wkxk0;2)Y_8=
zS%(S|Ve~O<gSAn1;S-$)55gRie_8$54_=w<eVT4vaGCtQ)h=Y{-?(nIk#aBNY--H=
zz3q&}K-m!-DfQUqo>PLnxyJ{|&-_v&^D-aeVkz+4$W^0TR%YufTj5f=Bp&<t$fBu^
z_3gKXN=*|E53k{Zv40<9>W|EBB`d9uIW5a$0eScDP8M0S=r{Mzq6JxI9>g0B*s0a@
z9Mq}3^&X`)b0nxwGAtpJ?r8auLX0{OU1h{$$%2#rw4C%_n<LX{R6qTr>V-w36HWNt
z=~XQ6t1AsDDNapWG9&_zPvpF6XlKiu#%uT#XK8rLmpCk!pXpAYeeZngPha0@r)Mm9
z?EtT89G%_wHuY9LUFeiYU_0W>H8B62?+u3Ov6qP|js6P`SNix@mV+&A*^}#}XJ~Q|
zg?@3y)BBs>ret1V$NW^M6D{s%t`NF4`v}$*`Q_0Z@-Wd?D!q-kGjkq<I=L)LmH5R!
zm9Lkzl<^hnd&y$p=a_h^uhBf0@joLo+od{d+~AGk31n*G_a$(>&OdrUJtPy3b89Jd
zw=(nBH2H=>%;*k-z$b#gXNxn6D{Dr$rthMf^%@nxh&^vidn|HX{?13##O*!eBi-ek
z6IHF&o(y%4LF;m$QpcsxL!cRN+0|MBt6gTu*Y@IiiTT+#n|lpwgMlgS-<R;o^4Gg~
zQqBB)s;DIGiq~X`b#hw_vRf@;0wX0U>k<+Y(C6zYq|il15B6rcgw%uhPbe663!1-g
z7UVeL9~O7n+t$joy(=2?<WuW6j98}m;!KfVQ9-MmO`l=G;6TX?2Sv6t<ftmwly<yE
zGPte6qe_gQ1%}?KU5H+q*sR~x%F4=>iAruzy9x`&z)51@u2A?ee^Monmfo;xa7QO4
zG0&=`yUX^n&2lG~$efj1R34qN@$Chc`ogt<iXuFtlCgo<1Fz&vYcne&0+ZU5Pz>K<
z8j(@mI#;j1{rLN<n|JqcLF_9&1GU3KN;`rEh-_WGDuE)q9w=4^(HqzO6)jZrr~7Yr
z#;wx~Hsze`EAEp~z8tVhOJc}X`8r`~D_hp3t0MJWYy>NeuaFSMe6TZkQSz<7>eR?O
z_&&wABbt$oEF&V5nCh}Hy2DSYYRFB%QcCtWH8g!yqEVN=V&hMuWT^a_RNfE=AH{@z
z1uuNJtbzA}txzby(%HZH4WNBNHk0kf4IDl`zT=gw7(}$$MR$o5y(*N@Ap93)48f2w
zY)>Q<%U1PO#TV^_Pj#_%_4-)RoV&`TeizaYUzt={UQwd<xE-ZZ-O{HXO2H?L7GzY}
zP-f-Jsr}|eOIJgKDk^4^SKnZo+EU4gNTegF?=UCT!z*>ZIn_Sl*+|L6Usn0Zi$Ws1
zP2<d1Wd5hK#)d1*2Hbvs(?8uH@RkNYl|RAzi3vj<qbfqkEI^{{HZ)g2k;yBdTJ%5J
zxsfBOo!BUR9W5p=k_aVEpwmD&C$6pXPJClmW8?2skb}XyMzgtTr$EK$zbpP&ts7SD
z7t3(0(a=x^q2z{p(#*M!ujZSMRF0MnTa{%%ek77>&L+P#`egRjdYDXWK@~0od$4k*
zV-tpWV;*?15s!t@k4aZ2`EeUgF%4590=H+Rk#d{s*RfCn{wMB!`~7!8cUb6F%<4Te
zM{k*@GO3t@61F0W65>rx62Gmk?F72xf{mp9#2<Y`q?0?3H|#6EDa{mb7kW8sF<`A!
zZ5N&rZ+rLPZ)F;{F{+a~-)`;5V=)p84<pN((6`u_DoN!$UR13GdHa~MZ=_?%;8zXv
z^6b!}o=y87A{lf^K9pbWUU}{h?<$Ib8_1IdGmtILWrfzs*wq(io$801SR{;6L7*=>
zQe+e^d$Ais8B_gE9dj9Tkhv}I^ByNEcVKNGv)`*-{^W1f(V+PH2`d*jA^P=Bp6W0Q
z!r_$Nl*S)+sMxt@R&{Noy+p9=8pC84E^bUzIB0iZJF+=+(%}SqWIflSNV|hR!x2#y
ze^_DA5Os1WnbSK(mgJ*m<NL5T==jfH!sVv}wC&(g{pCRULG!KM+4**0*VpYv0@TWE
zlPsdi3H?e&Q)Y?k!gN}eot(EO3tlT^^{yG6XtiHbp3I+q;!1D*jmoGfk9XhiwXH5Y
z-WK#=b1I9y=gH>Wt@=E{_7pp(wh@y;k149Q3|&FT@3p1}tzW8vC3!oC4`CSWnM0ID
z#eXS`8XRoNf07m03g@`oi71k1dMHW|OE|H;()WMg-#xgu%R3-r42(ab|3or%OH7HZ
zttGzP-0OMJJ~H_7VB`L)I;AL7uMoXS>eQp7$C3+65QzDyUEeB6i-ZfyD+eeQS5R?D
zICJQeA?8i-ddK}s9Piv2m#0U=IARxg_OW}eDIMEajDj5`@khuNX)J{+T-qpoCw}Pj
z%H5>#uN-Pc0@Lnvc&QbEmjS@2U{F+4+}Yo62bALOMtQH{#&Ax-lTEJf(m3^@E6J2S
zKX`%^(5KFv8xL56u>}}h-_jI5XF+owQk@_z;Swe|{4~?z;(O+u7ck>k)zpLk2&Ie_
z6ST%{ekR~@<wlE^wuzR~L){k9PuAvceH6?e8jW6Ll~|52Lnyf*PL7gM$S)|^1!-NY
z&524>5S+)+zkK-;WLGKRK2#<z=Bwj(uIw^c4rZD;$;A9nu4K^TPhE{j-juudK4*nb
zm@-5*z0qfUVybmk%8TmgB0h&<KDnNj#FK}#2HQldQ|cahY94R5y}MmaMi^aBODT$z
z|NNS8?ga^C=T7?d>=85jT%56sx_sO0r)MUpO{b=&a&U5%P0-g0COu8nfAWNVVPU~#
zz7-4A9D+r#^Xt}j^Y}Q8RB!21?=Qn~;@NcJvQyrjc{pj;k3RW>a}<eflEpES;1*g{
zq;^i<>@yKpy`1#tO%%TATFovAp;9$!52c!tbC1YKRpQj$sXwTj%#9h3<+sIr#QQcl
zoJUVO70cpWetDgpeF0d`N(KfrK>7-Tmb|&oxv2xry+3|-Su;(VayijoL<eP(<6_@T
zx$4zf+U*uPT#;v@$`$;_4=-OL`;e9=#cXu2@6DP2rQCdDc61Xvf!96EWlGW4hcltH
zA82!eUO|{woPdks#F|1+vP;zmSRj&aCZZEXp%APr!X6hNtDz|$Gj$x8{e?}OEywYZ
z`gqOrp7fc(;RJ@CfBs9Kh*7LS3>WGKD@#(^39?b=-;1a2dhhbrN?HcHu@-L_nt8WL
z2-I`(;U=J)<9T6nZ-w%gOEs0wZ>*2r_{2Gafg}|IG*K(&YQyzMKovkZOK?)z;!2M+
z+SUUVELOQEP^IJ0?xD|DHkT#!-`mDEl_XP(WI>r<y>GuPC?KSuAWA+Ob^Ob^f38Ng
z`Og<4)bc}BhKK9*6qTEO_jgghz7Pt`nC;XGS2f%Vw@%2scHoxf`|R&d=-vHCHL{|N
zNts{2naEaZ&05fi+(TxjJ~>KK-zxd&W@jm|#1J|yw3FiE;x;Fm1-t}FO}tbshZ50T
zM}D^1uWV=RC8BommYg4LY=_BGt;cp&X~;BMrZUdO=6DiU?;c;fLVVF78&aV1c6oTy
znpe%#jL`CH()jx-!x8JV<Q<yJ9CNP91*J%n<%+M&VYJZrh3iUcNHEqyJ6Ml&DI&^@
zP-%yp$-<@QL6=Q86lT3%vLbuKgfPUcd%M?$^6@jGBE#+5aa+j5rt{fih*q8#&o{2n
zZrN8+mg#fFPuMt`m}rGd+fS|KTZWC6SGLy}>Lw~zx^G*Kb`zz1!2NvNp0+ISHl)W$
z?`EL-T4*Qg#o!bI>9~T2#sd*Z5VV?G>Cb{;AoY+jDG?n+^@PPQ!2W|hTsWas@IU{r
z^sayU_ce4CjHqF~V?$T-ZIK~6;hLsZieTwNv$U>s{Tdlj^?eZ){}4qg^C6t~LEfbL
zO!v5$0*NyUoLRiIZlat0yTU-lSY{&7CV`}P`bVb+yi!@Xg_x-LrxYPV^J=^}O4zN;
ztBDI4sQ3s!ERTE@a9KY6)50HX*>Ifcrbggvh!mZ+o*tXHI5pCw4LWff7h8?zA<cV)
zzRyhL6iq~Hlo!H;iiMhWc5roahxNWDcxU}Rc2$THcCNKlOOv_qx<cm?*+I+uo!vTn
zjwLL~t=z{C{Aj+we`HhISh|N-)ZZ>|8a|JFg;N+1-rh#?W97TcrYv#omzi%odUrm2
zC^4hxpe<YfLreJ>A|OGMh#sfUl#n&DyXuB25+M)#v{Z3N_cjdo0N6pv_tHJSuU@^{
zU3kTh2y&vKq|WyT$LhU>VD0y%Z#O=bz?rNz4k{i~xZClMd3=*-zi1-V=<oc&`4+KG
z%bvo`&Mw1}MiK1~uIMI_jfw)q$(3rg*XcsYkFZ30T`&D^wkr4~+9sup4y7M|!$J2Q
zoxg@_@-ce-lCv|oq&VZ#Z|ie*6H=>nQ;+pOx)88+1U_gE4rR5(H1;K?^7v@Eoh$NX
zM)Qo`^2aE%wxfjO?7M<=Qt*>gL=(|X>)biTUkjZKms@L`ALi$CAq$2<(me<$vWRLh
zbb~fwm5imp+sEvgaTL|~DZua#>L(V<x)|HQo5p~SSrPk@IcSt7BEI;abLS!ZA?oRH
z_KlL7cKHsPG_&dl5l&92ymib<)z^&l4)ENEjKeUcj14mK+-kHh?EWz_w@BMEJ^KD)
zA+>@8k_<|duina{y^nThTio6I#b4$)=v|&k?YmWZ)*3K%d8WZ3c}$E?QrtX!6cBFY
zCduO5<yT%|?N&6tZqvCP``8vHM)G;(z>X~zI6+Bwlj+empshdPcpm4x_eGHgkg&o5
z9YSg5e?WnVdjrW9+K(5&Gv{^v9O9mJFNY#`?wWr26Cz4pL(FKUP=wb61|4JsorjJd
z&vKPEvX?afpp(QoG`+Je6RgDfK&2<V2usj%bcBef&MDxWQg>1OK!>+;bZO=8#*nvx
z?lxK<cHI5ZLVj!ORZ}9o1J^=c-=KwUS%uvrCfaOihK483ZX#h-|A>Q;14?6Ceu8zu
zhXr)-j?^X|p*Z4|$*8Gq4!!T+#}jp$ia;2(FK?uxP!1Cnc%bBbPetX9w)VR${$Z}c
zr3UDU_L>i)+5zYiI_hF4Ty_p#S9or&byky#gOCl&^$wkHjXFWa^Ajjlo9kDrzb#|+
zT6j24OIp3Ck^eI3)o5JBqwE{CDl!~BwyKL0_wQMyH&g9^DTkg>iMD8i-kp(C@sUCD
zX+dRQNxPnz9Ak^lcVo}KP5AUXvh=9IdByEy6*P=ypspfaxzEGGSU^t_h_Bcvq^85J
zaw3mc8Bk9B*(y{JEKaOR6ciL_85-K{cgLxxhzBF5zq0|SyuY7H$eiy#+?Z+jdOt)~
z&QHk3-W@H-3G-*YVmHG(dIPHi`iwQ<_B`E|o@)#nowChu^<x+VI+TiZz3#;>oxJh9
z5v=tC!?z@OnFdZq(NhkX^B=JCh4GPZjBPF2s3sngzddy3nV+SaACWjDJtgMUkDIF+
zNnK)p9($PC6O95QswK2UgVDr+7U@!l@tcGUlH>p`R&;XWL$s#<{CVuI3R=@o-Q0>&
zpZ)m}uV3ZxVc0*3ty0&>2x)<^0HIkpr-sUtE^uZ0n*uMDK8|Yp3#}-Rjov6peZWa2
zjsE1oaL5*y^WS*uL>-|7Qy@-B8G)vsmk5d&MqJ6(Ooj07Mt$ecJG=8oSVsqR!uT~7
zr?g=q45vz^QG==SwwF27|LpTAo)ngiYtj=s`N^P*EODOYsJp>N1nZj$1+u`pm$RVJ
zdFRoiYrx$p^*wUpAdkdOdxj`$$_6%GECa(GsXn=MbNe&`4}<W|&;2?%ALh)x)sPAP
z`qdX>TmHR5tHqfR*;v(|5bu-9VjAIQ%;hC-pIT<&LXaGPKvz2OB5eILpw_!G&VKTp
zzj2TMGqSwCx}vlJX#^~kqe!X6ZQNzd62U#viz!R+ksUnvlDABGZNK3rSoDn<+aHR0
zhS3PR`!n+Dpsg$DI6QCX!F8Mv2HBC=PW!i<=68V0urkwd3r^W#Ekq<G(LzZ8-KN{+
zh5suzY@+Xfp28~0GL!6yGdu-CQ(seH;Mq<aiHnCv2dwup;5$FTyKt}j;mn#dC4Ul6
zk?uUqh)M|y@R$hv*`={T^zyk|Gd-VUI_tsC9msmD++20ee0`I?Y%j)_YM$=LJY5rQ
zs-O_^1NjrvQSpI}*4oO}@Jjw#5?#TotyVaTI*iJcWZhiN(-o2@>X2?QhNKWP;l<wB
z6Se<d2d$$MW}h}x3Lp}wFvx^}c?L2UXtyH@lV~WY#zD}qpj1rf9?&_;+JyJeo;Dh3
zujsh7U60X8jSx+dIwn@u%-f^ZBx?6x$M18KTuxltaXS*xigQn2d%tsr+TWzY>>QPU
zI53tVxiIo=_}%ZIx8f|8j|inS#$+vT#7VYU?;F25T@Rryf6{5nM=jf5Pm-S9Z{|v$
zc<BD9fJS9qZd9z_sqw$D&ly}oZw@1%vIHS~Ts*uMXii7E2N38P;B1iCBZ@{y0Ez^T
zkB=b}zQ7;&j{tc9V>Dy$y<B#cqUl;U9Qc<YH+~H@*Bf{tBy&6$s+3*4nQ_`>#&K>(
zPk+)pZ4%j{Leh4h%TbQ>kwo8~_tG}URJHM0<=VqhDvOC`!>)mdk0Mbeih2gvPo)2$
zo^hXRE{<a;^M`fB%RM)299tTz$jvh9Ss4kg`cykGg;;%cs5$evt$q788gK&z9(^MP
z&~r;~F9`}T2uy(Qo`c+rqKAhFLQ?=d$)}dtT+T{Usm$44CA&>vw<72Rd;$VQGa6iK
zW<I{+meFC-s!&W5mpsQqx_6~UOlL0%LMm>@jE-S(Xs8Gqa1{xGHyL~ygvX(FF7KZH
z04heNs(SCJdY2k7$z_n;vRB10pg}248>_!;7gyxWnNwVt$ANP_WsPl=S*o*3Q>RLs
zh?uYLd+;md#XI%ex+Rz_fl&dYtnjNhT=QhtEBO?MU%5(rP3z&qP=q1WL+j}K4w6``
zpk)rXs^<Sa(<Zcbxp0fkbvr1CYNW+Vljnaa`Anq?_uHvk;u*r`{&INw_}cD!9jX`_
z=aL;MB#<SKX{g6cjd?{<`W(YxaeG+HM5Xh;v%LD2S8p;_1S@ntN%K;2)0{OTeuKZe
z;NAQ8UuqG_yV<$11-;H$5v8UeQY```AOVv51KnaY6rhqjfV9D!DAsr*O`0%N@Wx}*
zwPKk$?lpqjYZ%;{Db4C%uE5DGO3rWg*6i`{&c(FZ{_U7b;dF2{%nzLif|X0PP_~jM
zPg3fZG-Jh?e%xMIGM$)Y&xV8Wso8z&<Ylb2W$vmMu=7_X!I1skNk{d)c#$)Lpi>aE
z7BX~)hn*lnfM7$yO{!lP)%qqR=({cM?GD8)QtzK?Wfmx9r=odc(Nv+w=Q!D0_WrZe
z3G?hZwRh}^Hc}CQ?x#-&c6*NI^ew46^!fSu0&YvHQb$v69K5_8AS@1br!$r<Y1xM&
zl6BWd;Fuk&C~~i5>yxr^ZOUNbaCS=j{icA@oNGAoG&h>P9Jd?c8mvA|nO{AH>z0@g
zR#khVk)*hUFgnMZ_d|vNF+jrA)q(4%W;*Bk;6L4oci7)bocTCX2<V$w78TR%2RDrF
zbQ-ZG71a$Hx^3nVOLc7Ja{8L<eWl_mgyi-BWj}5@gZpQZ{~Z@opSLwRPKqLiCXvr1
zBrT2UHN8YgJhxuey&aq}b?>!A{j2(BRDFD9hu!=gJ1WQNTUQ7D=}b-R3Tiza;jk&G
zd1-^re^5tmQZlXTd%5EkF(x}Fs7d(L1G(xrZ7vFgH3A3&`RoT4pv(&(@GjUejrQvt
zvZc5sc2x?;-=b5sdEUm}F<+KH?pz<)cl4Y4oZM}@?sxO8FZmT=dn&9znC%lh%4lM(
zaATxYRN<fqg)C6;Ul8&+){Wc*4ZI0uZ!Z0~yuFnKefuA456ZA;9+1dlj^|NUg^#US
z3357*rnYUKILX7UiDaNk(WtEu0P*f<(*=c&jt(w9ek*K*0>GqBXcU$<QcJ@S5u1C#
zn#U}4#ip*^@S#rP#KRg0!kh=H*sIVD?DCLjlujov{iv*>BIy11Q9S@Q^YZgsq4dmk
zbs$@p;tZ|oVd|9~>RUt8YZ4nKmTEk{1Y(6f4Gy?Q%l2#UVBaDgnCOF-@heR>rlv+5
z*oC)&Ig5yMLneYbuHu*Ygzk^>)AW~`&zI*QBg(<~dBEBE+ZFpmn|u%Zq9V}&{UOZ{
za7xt{l@M~(yoTphcsP#C$qE~faSf?mC9O%F`)x~0Rxq|kCpQ(#C^&Yd?i?1Oze~`)
z^Z6Z{aN(Wus56=?U+GgS7sCv{_2DwWPFnTpdO}DRN;7CnjP2GfI%J7noS&kh0M?1*
zkmfgeiz5?o?N{n54~th-K<=;N38d0l>t>_o=FCCF9Ou`sUkHv8v~93{DQ`<zGWibH
zfL%Te5#ZKopV)U@IW>4!4C)92>=g>Adu}L41d-L+)>dXe%11~oke<!IoMGks*!5uv
z8I17rio4_}{63;6Q1@m8(30Rg<?ArsM#uU0RvUQUjgB!^@YN?`solUILzYQE-EnK&
zeFJ4Q0_0mIpKVPIBmx-ed8{YG{IsyJApXaPkHXrKLg1AxZEbB9ZWr)%{k3k^XON{L
zPR8gP=@a@Aof_x3ALWo`eT`W9#^egkxGtcA`Aq84L7xho4B~(4a0P(h;1Ux%#ck_y
z#qWo=WBVvo_5Yubwr4CqqLXHuq#xI-@$eLYoC8d|9uPHvA8<1-(3^D;d3%_@FO2z|
zS5<5);UY@C-$eiWk}Kh<Z;|WUzqD3Rge40f+9Zrd!J#-jq7alx{6Bt?9D}q$i)%`M
z3E;2Rs`YeqET9+~=^UJzo<{P13-A(oy*hffAqCf7<GR=fem5|kpdy~e(K-roLtz1g
z$Y(U?X=S<WEvaSs!_F2L8;i<*fCsz?#8&?K^M|<p)spqPM8i%i!ISL~K$AutdcRK_
z{1MN+n>>4l4NN`2to_%|8x=*E<qzFlb>OX}`mAx*)YY}l&N3m?9AVmMHs|c|2yZ1e
zr6>X47*p>*6j}a^OV*gEw15*%ND<|38&NSZL<4JLy}J772)Fb+YDM~HmEABG+;Xkk
z@|vvu>QMxGGi4vhLIK-yCGew^_k$)!U=@mc?Jy&}O@#CXB4`oO(T*(zjsaZp_)t9F
z1rP7wdI};=;g$BImarEgMcGqRQvyEwHrhHmt4H_m-D`%wihcL)IVdU3BZJk=-?Wiu
z<brF|i-!Wns@1;J=2Y#V&1u_N%7cdw1?@+8z->TeoM6F$mvBv+D?Wh9D6rAa-hOxJ
zp+-hd4z}$1{?#gn@eqi2tYD;hT~*Z7e+r=oC7SwtUy=3!HLLwe#zT01;3ex<+EF0(
z02ban1aQsI@vgdTPKcoP8PTfJ$mF);5Xt{t@Jv81IzTC{`q;NU;!Xo((Fy}EFDR=)
zs88sFY9}%T=hncgH%E;DWsku<jF$j4`Sn@Jg>~w(0nk}a!R%#U7S`0f2e-J&ZCSmO
z+6u6f)~GS~`h!z?8Nc_g^j%vs4eO=$&#nzFw^LhzJd@Q`c2!cToCfJ!KqO!XW#{DF
z>})^(cLMeK?-f`rpg-WmABo6vw7CpCJSwf6o6%I$$gHA>m}F9VJ_z02d4G5q<zC;=
zFb>V<>f3P?b=K|F4R<A>gdC7vPCT5P-&a0Vyf}(bLK*$PM~AdoO3?$`laP&YUmd{8
z)kt&t-0i3}K0f}fr$@wx9TL0|QBgy(ZwGTV+Pb?5AjLJXEd&fVfboK2Sm*rc)SCwn
z9w5MLko%tlyny2E+ra@wzP{@J@lcqUm{1EBejnkK5|)ReoLo>q8uUy6sc^{V(V~UL
zNt8SACW+0=%phxr^NNaJNl`Hj`M46C8s1Y?eXg7>4~UqCyOi%`*cg0bf(*)^%tLDT
zrjJ?B$nFtY5!Qw|6ec6dAD{h|V9;4d4zCat9;8L%l9LIbK)5_C9m?%vlanWJQcbnG
zA2}Z5$Q_kC89`imc77gRr2>4a<Kv2gWl<;-MugFzZ^6L8z-zBZOb}4d$O$XpAU4+|
zcBNuq!7Nlao%2FyfPu=q@cXm0Wau~QKFP?QkV=M9O_id7=zC?Xj17U;0Xd5W`qF(5
zH_-Y5Qu~qZ7FHtUe1JS^wLR0YGF(74FgUm>ozJ!(h4P6(F@sx4#b?|M)264Y@<me&
z2g<z^WC+(TzrqBC7ol+P9xfc!5QWM2hv+Fy)Ik;V>g87ecCU<<Fd?8zAPKd%w^z)?
zOUwY)$mL*7C+lBj`};m2$a_yzJDXu*VX;69IXoU5$rfbsmv;bF3}bb571H=EfU}MU
zoEwXXNM$yEfex9E612S79c}l6XnmxF+Y&P@OgO&e^_QU3!mkMf391aR`~bpIfJI<u
zX9rMdhl*aX4G=Mo-J}{r#@dkP@v$mWXr7p8U15^dpAH&b(=uWx0NRBj@^V5#XccM8
z!jKa}Kwwa;f7ivO^ukRrun|7Q+h||sNJ5Zd{pdbK(<qqToAWdXlnO|x7&yf3U@IZa
z5A-{p58r2`A@xy8sHi?#fx2}>rYr~z0}A|E;+5l)k|Mzl$#cbnOjdfwEE_(`NdJFf
zAt^ceb8zS7+268ANEkFp!`p()iD;5ue<}jz5D@5F`S30wVF5Z#POPuM%LCXMWK~R?
zdf}}!@6})Me>_|PJRb)Khn%J+X;Ap3gkYxtJPahjsG$7(j*U%Tzb8BlxDBp<eo$n|
zNx&E?XFo`ru<Le^W>bOJkJb!k>wQ`rY3j0;iHX{O3;EG~U*8x@OG_8{44~iE0|D!a
z3uuIsOQ+qVy#B}6+W815dCUQY%X8~i@<&OJNI=|2{yRDO0=5LK57NN14h8YIJU5V(
zF8!<5ub+T=8?WDRPeEZ}lu5l;9P&S!x&26?1r@*Pqs!CTOKPjABE!l^c&%md4$A?#
zSdMp<y%e%>x&{WHWxHUD8ncR#?gVgKn|i9$cgP%fE-Wq@ttBLdRW1Y1tP@ZU-{8ou
z=Sz9{IEn08Gd3Z0GxD#%liy$Vg3S#y0DeFU&_G!kzx(_5{>%KlyuNL#_1;9KEDQgW
zeGA0qt7vFg1C-GsARxd2awdp@|1~1?<LA#ua7Wh5zdl-lb#l18IOpNviGo;BHSl%#
zIgMcF7`X5)u!|UYdWwGa`*+iKF<Oke%XI_F5Uiv6ecRt3SSxJDJ<f^m=RIfCf-@8#
zit=!9M8JG1SN|wk22NwbhYz1;O$`jbKlzpN9W3jIf&%Vj&{=?cwQb>c3V`Wpjrth?
zc*bbK97n9Zqo&J`SeTdswZ^~TW$7y^z37(!0<J!sT)I7bX04q429o0+<KyEio2P}D
z?0Y|FW`;=r`~CK3Z*R=^M>i?2TxmrBd0B^OQPNYr-v`RknnfKaKF!|!33X}#|H7gf
zZY6M%<1ekRcT@7{HxJoWhHmG0kNGIUBz*$hO)Dt1SLI7A-`R0FoUkt`d-LW>)}GYQ
z-Uz|Yi~U^Lve;PCrHzfB5RvPvsz%8iTSd*Vbr#C9`NZ%D5uuDuq!*~I3U?PH{_~1H
z6{Smnvh{!d5h%g@fBYrZ6LO-a%^*jTb?FVwa_?_FD{d=cgQI);t`EHEv+oe>AxJct
z1;&HI8)!7++6wa5@1Tpn#^`c&@$^*k@)A$rHzST#8r+y|0Dk(}BGV;lWeHn|C>;4*
zvsE!ev3&BU17-@0*z>hDJHMmpXYWPqDM4b9lHb&!%2`c~2-1ehzkU7v{E~i4{V!N!
zX8EKDn=~NQ5cx|4u&5MG8wR!nm8UUT49H2y^(w$thC%^5nmaY{!rRN+y9My$uvs_r
zH=Sk5T01)OF7>7%;4e6BKtp&Qs)^EKYG{DEv4Xb86~JIaE){XYK~WIG7=CJrr`J%;
zpc+fxg^URb1s*f(1;{poABK*&Qeb!;_oZk3j~MEnZ4-yd2ZN#stx4{6XjxJ+Hcsic
z26w2jv2m=<gByjkfKcCj6mfBp8gOC@VGdrZ8{|!Q{{Fpr;xOCjKPUccr5wK03Ys%r
zTwQPL>r;bR$x>{;zI?>uf8(q}j*kV_dSCX&jT>xW@$TQhk8m!bF$R(|L?k362>LTu
z{bL9mNr01<CF^vsHjM1B2%RQH%KJ)KSlAf2Kf~K#c(^>bre1<|U4VGz7<}Wot*zH!
z5EuVDmqEyAQ1ZeQaLj@-hm0-q07xMOG^|*Fe|ZeAG-7B-8#N&(A}UI`*cpAt$4AoC
zZ&$H?x06n{gbfneI3$b;-rltr<p86}%*;fG&<PWjr_F^xYQYxVdHVF`wJUh+>M6LX
z%bR`^2S`dA0%S`tQZ=A>EOWXc<S<iD16Wd<9SJ;^B5Jnhv6MF<(<ur$3b2$?z2%_e
z%AVLg4o$2FNMD{lI9nI*K{>Y!JQ{)2<%2XGJw2f?`Igs=(m$xJ5R=gW9M6+YyUHXi
z9Gvy0SAv}{NSJ;Bwz3WWnvE?K%^Zf5vSjagcaadq&BZkksH>yXVQgYDt|5E=sA3yZ
zc!-IxiYkMgQn_h^4~?T)g-zUeEg{rpAd9`qX*&OsM@A;Kqoc!MDowDHQOc_sRSL9e
zfllh~U%#H9#^94QR#A=<6)(Mk7sOEhY}@q7v4)(gDxvT4$7fN8<5iAy1o-$@&P<_h
zOwkR6$HKb?X*QQ<&%zChuzU9q$WsEJN&f5%cm-MBfsn6|hvR$T{6zhFsG}1G#8X7K
z8QvbaS0|2u;04$3rV7S7Y)42o65<Ie|LG4OKIG-)HE#zT-OB2<x3k-U4FV12G*x5a
z&VT}E$MK*ZJ_#+QQV3NXQrvQoxB$})p54$kOj|V`gD;o>$-Es9&=~N0`*3q|WiU5p
z#a95F(d#La9zSf_ORB0UA*BG92`VVC5B;TrYx@zJL!MXU!6I20)Q}AW?6{=o2JYvf
z<C5!QVq#fo-QC@SE^~6Q8m2dlVH+W;!w!?TI1;QQhZr&76&QOO$G6>oi0X(x0uPFi
zNroQwynD*ZNXQ18-2P-&G+j8@PKU8lG_VJly?e-AX{}AgDzFL?1Pt3%j-MlLW?5O;
zldi8dHNtBFxGY5^U^HVX!6AGB=D7cqgjxXg)pIkXsY&*oa(0J*NJxkabl4Hni4vmH
z`)yn<&b%Q)Z(g_nzZk1`50NH?aPeo=x)QU!z5PKNtT}?Gwe(==Dd{DmA|m>zDpnvk
z8n5pHIAj(?G-H+aRK!d&QOl{m#E{lTX=-W)L8OS>H2A+*C|IcjYyZIZYwGG^$17)(
z<f-{$62o0V-rMus6p}!ET<8!3bAn&7b8ygs1V;<)c!=67Sn0s?bu-uHUks>baPvzc
zTHKhdCb|6A8;GRj3Z%~%y?>LbCW{gUo(!gyI?u|SOFRR=6a`KST9+$gPGHIg{r&qC
zkZ9)ftzqxQoEed*4W<<W^JenjCxbttqY$JS5+c{b{sm3G&5$)hI^VR8Bp?ZdkcVL<
z0mrB>aBFijnwUY-98Oli;cN%@3(0~PEJsE~J%?R-#&1szY@tE@YlM}yK2ijSLo@&s
zdjZKU#H$WGot#xL;b-|2ZM%9u0*;)zOL5y%H=#k@VQuL1&(YCD*cB)#DLaw<7!nk%
z)Ps}{!0QX=j8|Uz^QQwWes7NYM=pKLR%$EQ(7~p7{~zZ8*?l{x%vP%T%^Fj%u(5Bx
zUL1oCK5DCp*7!hhtP1bOFsib@6_k~g6@}P1At@<eb_=qZJW^6Aa8*%2&LN-I=SA<Y
z1(Su*497-L0Ry&#lk=sbf&$`RKp@`*aTYbI3Kn|jdv7vatBK4QxN?pr#~3Vvo{<u>
zUe!xH@u$CW>7={_9&(VYe%Ag1F?ULlS<}p##})Yr_Sb^J2L1nS?7uS<IL(p72a-Pd
izw;>nzxZY860P9<**goq{T3AbcVAImp;+EL<o^PO>sC+z

diff --git a/master/_images/sphx_glr_example_regression_thumb.png b/master/_images/sphx_glr_example_regression_thumb.png
index a8b97b20c7d75128d2692c14d69ec0af21ea6c86..c01ee72450194f4b2855b15de149cbea0ffe4c5c 100644
GIT binary patch
literal 53963
zcmZs@WmHw)7d4E6gmg)F|3EsVL%I=Akd%<_?v}1gNT+mzq=Ym`cXxMpz5D*3_rvqy
zxnnRKT=3j;_Fj9fIp>;-P-R6a)VG9hVPIfTWu(PF!N9zV0blrt@ZcSL?JPGKm~n`V
zxTu<I+F`njE6&tS@JjMr>mG@;^lmoZUdw-<D_>tTXX%SRX(}J;9C{w%w)^8sg!$s^
zwYVTs|EgE0I~ja>s`^HORSf4p_#hwz8A|6jq3us=yh*#@g<$x{t^PtA9|NzVSZO+k
z7|-IeQRsKPJfkEbQQ$Q!lg*Do0$%BHVlL!-&=+0Mv4|t|rTBKqZx{MX!)bzr{;x4!
z21kUPT<~)EG)a#DiHqwjRLRw7apx*7K0oUv<1pw#Y<;@c3dBI9ND^>kr;?0VjZ+e$
zz(7nhi*o{39M&|CfF0wEWNE$iueF%@q^&)e>GRxQZ9IU_W&kJrbcxOQrQGQ5d`mu^
zF9DNEveNa;AYUodJAY^z`spTYX=bo!r=94+F^P#izOP_mG@4!60ua#i)k{8K9?V1=
z4<u%r5qdpb=Yco0o#%|1beogpm>a1%rM`cst@%DArmh}{Aw4dmzH<PcYtocjro)Jx
zVt`#JGg`>&9$eLV{-_WW&FJC!$n)uHCW3(D2LZ=-3|w3ZJ3BV#KO`{W-C~-W1Y(F_
z6<#4CBX@4utEcpua{g$2IsS8EKg2bIK}I$rv+uPBz5E9w%^rQ$(~W51^Es2uZ&eye
zoMz*V2O<%YR7dl*hQn#RmP<{zGLXVA6{M*=Hu+yFbPs2%0=9=zbIk~~>aE}CZ}dd|
z%Fq9tLdV7y+a7=b%WAVMP~Ncqc71(a@n#FUH)@O{!Y=}C&-aYrcd3|}(G(_zLrAzf
z{C?%-oo&V`y<WX?-<v2v-`m?0xZlmST&Ro8$)P-MxzP>8kk1rKM<8|eeSGxHaNAC%
z8YV!5J9Td<IKYB#e{nhqzq9Sll1q)*1nsiVi&uHm{(r6a`y~zwe#HNF$8xLYEXBpe
zS-l@!uMg)w=1NA2X{p*YpTRHG+t9s#9~l@Jh@JAMxtX|arJ`3-@zuNVH{sp(`9q^%
zBa_&T-{wk1(+UcvYu0>^&B#FWEiN`0%qQFtfj}T)FN)IA9|mQ_rKBdR412LN>VN-^
zA0HoA)|S}W1mQD2{>i}55D6KXe`0%g_p!jPQCr|@Qq}Xc9U+347ebfe`Z{&UIe+N2
zc@}IDvo<qcRaI3JEU!d-LW0WY&-ozW)6I>*az)^?`L}r7u-mPRGq+rhb69<%F_$Ol
z+Y?C~DH<sIjPk=_d+DRpP6VslxzYXQUck`z4Y1gycEhm4YzX#z(7=~VWI}rKnJQM7
zLfyv0#XY;*$=tYkP|8g}U|?a{Y23-kZD_cSNYj&&DquG@HePgD@mln{S%Q8&K3<G)
zTF#!zr>fK3TWkuCHY+wiOjBu5vCde(rNg(?t2RPvY;5#=dAdO%<P4de)i9qb4jMD1
zW@3sEAs2CT<5f{n(O`N<NlDpoIWEVBkK!so%Bj$~=ZJ-e=b};B*4&&{UCj=DT)A-c
zAzks`zki76=)KnURDpjFuGViJ&K{n+y1JCVetj!qWJFzD41LSJ#WY3I(UH?n#P{l8
zCO9pvbM5cO83^iVY8mhoZ*L*^0FV?+y>JV1RU=DvTT=K>n&Uu3(JEyOzQNPB%~#<2
z53DuZN$b_7w>`qLmxbu(kX>)8vyXbB++;68FCmyz%IXP63-uRW*h;4hHf?qzf;Xd;
z-#(T%ZWAUlX?6^!^Pjm22t40z(1&$2*lpadygc`kxGaiVS~5vUNc^^(g9MYh({rBW
zE2QEJdkZ-KN&h~SV)t-V&jB{F&SpgsEX~b|k8n{@k+#>>6sOH{nxMz^Xsee8Sc2JR
zHx97%o#wUnCoMPQ;B(nII2w=Y7C=BFNakY3^>>|b4cwg%u*4cC|NhN++<c+9?0)bb
z{8ng0gwgHkTJY2|ICp7;t+(sQq^_&4nw$=L;u%zj!`ir>oT^`E!}q;hf78}Vq<l@v
zSrG>zc*(Uq+r-Wo*f`57{!h2!Cs-^jEG*aOyB*$>=8N5#3Nk!BPd4wiOppo*WFTVd
z*J!~3ey|AV&krXxrel=n7Z)EL%60y{Q<%UPbm!z*a>NeH8Ey0NVQ$>S>gebINd*f-
z{$0b<^#TExIgZcc83MY{Ejjd}ss)M?5NcvH1SNqpIPlaP11xP_2UnB>?BG6lRy^at
z60+<6g#l}g{*Fa=FbtOwpVJg0THqX!fq|jm&>Lh)IEMfJGw2Gz6gd40EAV*QX%cw>
zv*Z`3JaB$?v=V3pM@>*<P}0+4*x#1&WrB0Q&MLfJUw>yfy&1%$<zfRSg?OK6Uv&hA
ze_<g#A*ZSD(08!*pLBIeQ%pv)L@#&80-!fizw92KmPVZ6b*=fi>}ycI84fP)+Qpr;
zygW4xO%Mn;aQs2^(s6UY`#BJV;SUxa3mg06)BV-Q1>>+B5X7<|Ho+o-<Zrj+G&5Fh
zEY}l7GO*(FGV~E$unXkWm#G#|`-%oZwb@;$XK~mbTzfca+j)6Dd0|bKH<>TpkaQu(
z?W>cHKXpZ%+C&V5?+J9L<I0>lkU$hA$3T43cXPZHKC)wYy4nsNXyFL0`#l=5Ll*sR
zrr!;?_a?`^jv=0v!Hyv0xkd-NH*el(+jZk6$<qpuqKkzn6K{>mdbX`7kz@STXRUO*
z`2LPb<FAtNlen51PHb#!gEq5)fr06CDFHau+TM3tzs;xKN~nON!hhTVE1{xMWza3N
zgbWIG=N+hkji(*RW|M`O4m-nnHY=?r^S`hqqy5y#Q!Tr+?>{t>Eh+w0f%7eM`$E~Y
z;@IIpfCBG+x$D<&lw-HildbPMQ?3K@E-E+yq@XZNmV7}6g$+DjEOjV|G%QTaS0d4*
zg8xmGejrApRY(b3IoWA_IL-y}q;20%(+on5(|j@n{6wEU8G3NSa-cQxC$kQWw4!3q
z&c)izLy=ya(8bP3K~@&|W!*%r#TQF%B4`-CJgmI9t_9&qD=E=&aV509Jf|-GahYha
zGX#Y#%*P|u$#K{X#Jt;XcJS>{{R%$6GZQEkk}B~WR?{*R>AyZ%MgFWO0g(<3Q>XpO
zSLFQssUWyYKOd+>Uwkj`tu+=3d;ivB09m=?h2J2;Vy4{q@%GH~tQ+seow<HNK}H6t
z)_jWZP2%2e8qwp^&B~(Fj1D4&D@e3^W4XO;85b8L!XQsNTups>Ni|cHU6IlAdxUkc
zmmBPVKk}PkN=#PvEeP0>Wkw099Zssi)s~xO!^H+WZO1W*2r}V}+uK`kuWcZlb}-o3
z*e>^{4&+m_a~%|80b&BbJ8(Ex<90n~LcnXknasTx|9>D@y7_uV+p_`4Fd(UWKOFJ!
zxg2|fbegxTsR6}h1^8_%BNi1cJqcfQWx!q~bC?YK47>TZzUJKm={*nwKe9*pzI^CA
ze=%H0zL_+{aef^OM0qwiP@Gotm%r1Lq;|9oR$9Fx2)V+7knzs{F;>_vSk<SS#TBD2
z`KkF0Z1tMPvk@vW)hZ&k%dxbGD=MNL%v2PZP2^9O>!h+7_F!ORJ9KR<yX}yGqx_fV
zr}nvxf-p0HXv1zh=>^cjsybMzS%c4OPYuF&FpW3P`Dh+`ecUBKV*iOf!}}Ah*~VOq
zT$OCO8#O+sa*stuSQ&^bSPVP9^hM8$p=NMbaIatWM3X7m+1W|EZ~oNt{s!;`s8qHa
zkvt$Tw~@<WE!0_U_#u$O_`i3fQ%viF(8x(kOY09Nu|e^TP3s*zSgOizox9_C$Yh?U
zufSP6TxrV$_wub~E=kaX`-_efE-V7t=GITWo{h7Z<Ybu&-Ij*u`vbOQd2nJMy@fNq
z?-@Yk&;TH#k}FwdGD0eJze}}f&GYdGO18_%@|0_@ljAcHD7m^`SL(~27YaYnNdAGq
zQBqTzOy(|`A0z3bkWXz&eske$$obEd(CEYk1M^$9CMbI;QV?n!3DHDjU7shX6mF}3
ztL*`9AUs3E!VJzg`=MEunvM<|<hOe3r9yCyI)9)d55*=Vbb~tMdbjlxRH80}D&+!2
z66-%l+;vt9Jw!Imq6LcS8g1T!VBw)xKp|icd^r9Sz1A6gF>g^ZR&S&Cj#*n89BbFh
zF)6prSgGClT1MJ9q};K-%e{%Y%5Sip!RR|+eK$)gS}Cch0yZ~|z%zx-Sqe?nYe?+~
z_Z~0(X#|(ltTK3G)AqzYqvNs!!uNMJ!Wbhx%T2YR)+@=}h@3wBNPwsRZAUiJ3C#HM
zAxHOi{!3ugY_z9|qf??463UpbwQ$%<_yWbmXJ7-?&s<qkhh~qiWd(Z<CW|_Q@O1kC
zsOSVxDR}0<cq~Vv`D$8gFhejkA_B!zh>SKa^c5_8OiGHZyE`Ai5p;rrB)MZD3rDm7
zB!dmeyuUh#%g7i8iA)*ijSNJi(cZXlA{qRgmi7l@d4qpCzsryNF;LowlR<e%Z}ock
z^!4k%G+u{chMiaUZ>a-w`sm@4WrOa-^x4xF-}7Y&CIGqt?o_eO`-z^32`oJn<pciV
zdRWza3qG<+gDD{p8=Y9#a+bBc`CRt6?Ztb4reXx3kP*OEBKnOC=BkZBSo_9!zk`E4
zFD}k0HkbBtx!RwatkBB@=!Dg%4+BJ9O1?_HwN4;L1PTAUf(e3@jcz=>&H5GZ4NzD%
z!&zEKJ$wP+h5b3u4L)@w5DB{hEREa!UcpHJ^>7nx(O+|}@ha7~sy=gpD;1QSA)z%B
z<zDI78FCP6fH^>!gEAVIWw+pywwDaBzvXSu?yb-F7S|^$(L>*J<$f|lX~X?Ng(_GP
zkQve8{y-u@MuR}G_9u&=;Ybf)Z6wb!IjG8$lWInqU{PtQM*!jsZr2w{GzOm%V@&X8
zX*fqB{Pv)t&2+h$cPN!-CtCRVuMs;NcuHOOJsQafd{B$lqxg@ZNCBh?Y#9hXyHUuo
zfTQDcJBVi}9Zco6TF;FZS}!gs=bNBSXLEdC`f-%Ay%<;1atRXLZIJbG2WNjeQmE*{
zBITX?ZE<nf!cmVUcksAJi!RIO5-#c-XOYmE?=?Z&r+1#Wo<fJt9`rFM&VK+<2AOiy
z4eb}q%f(7Hlf%Ujj{~4N6U}ZHql+){L_*|y)>Rs#oT&8+B=`5yyI(c<b|{6PZ>esU
zJy=i)IadSFg|cMgXies+b&yN)$p^5G_-Q9v4bK^p4pvkD<j^N1)=TEHZNi`_tN*<h
zshiAxoOng*o>Ba~RHRA=sM~+M`99_Up{YL!R+>Y3ZOFvl=3YVOHn{F8ppoAJ2~2p-
zA65rsdG3!33okD(;8`&l8BqgU1_O!jU0ZZPEd{IEtIw(oDD?Hwf{dIT6$Qnbe5%d{
z0h>WrGPkJIPU&|j#Q}Us=Z`b=oUxmZW&w&5nJxJtMuhzDU@}Ml@bGs~4jfOG(?Klg
zx~{!p{&UC<Af@T}FQo47Zh_0uk09^HWM+<Vj0k*eaqpcul|gKmv7Xse^UF88f111-
z;Q-u-)bmvIYn^33SU`}io|CzYBvq)0L)!IM>lUm(La0GrWyzOM7BeuQ1Yo_ZOB{+t
zx((I<_=o1#f(-XC5aX$O-giMDy;S_;eVNPZ2Kh}{*>}o(ip=u`0-KIbMMX98n}LsS
z%D;0>AFPlks2Kvcs{sP%eN^r`YPge#2(Q~^hfOlveryf=<oG^_pTc7^JevIx5HS+~
zGn+4l`8T(>#|u@cV7~q~ZJ#VsWgLo4O&wV84hOI&A|gUok(!T>FuWT})@$m{MO;E+
z&qrA*dvMM8AW^_mC_yrEBSh)L%_o78&(o!voqe%XVEwkiGi>FAGlK}7(g6Ea^|egk
z>2jj<Yi$_1@MFyL(?jroHK1G!jpxY$z6{C*%)fyFbYz3OC-1QEa6f(ciEJ!t2=$&Q
zOZtSEXlq=Bke+ByhJgs>fB!lviRQ<spAK!vFZCo5+X0|&2j@W+oS3`o1sfR|8QzKS
zQ>NyDuz<BeKQKNX2e1?*xYwWvqEf)6<X3&`gomah^O^D?Et?jJ-YAk}KBxU}?WJJ5
ztsA%Ak<YfO>sjrlA-#KP2^Rbl&XyYBqf0|Wy8YOg4v%WchJV?Vk|GVHkCz*tm)>78
zajxs3bod<KQQ$vDx(ZC2nh%oP0}MS^Z$kzsw!p<8$L7usl!c+d!-5^<x0?U`Y84{K
z{H-Gp$?<qG8L%2syVj2-HBOF-cw&fQVT*m|?Nb#w(($~_x1Re1F63fY8`>BA8G<FO
zG+dOWOXNltT?Dmz@-_BbvOmzt6v3?szdUk0J=_R`l)we{+<-gP4e)+Ouf)Ve$zU`R
zG$MY>*-F`*06so0BpHabLZrdfKc+xh5eVWfm-$rGUjyDwA8RqHmjx;M3~thFc*fq2
zKe{sG>3Za;BQ5h}3KO@xxzW()gp!1f-sl2=8UQ({dFv%umg^9<H_4SStaD(%$$*c7
zrI13)db)loTLR(m<FTO1tDz(BCr>q|juW4QklETK`$!(~VS^=cI4G+>2WjJc)-amQ
zY@B+@Ee(##;7v~Mg<gj_7yITs_vQWsOq*Vw^N}=4564USJNh9BxMo}OMsbJ;Hk@pA
zln2_BX5eWfn|Nk}q_)Dheh3W;nGjG`Mt+-3007<#>Tv1S;LGQO-A4ov&0lM&3R4Mn
zsL~L8Y4UJ?G`J(S!6<yZ<%&PGWkQWwD$#=T#8L3_YuVt7IdjCTUTMn^k-0Xjyr0BG
zgmVQ<x^Xur+;pjl3($+H^xigcJ;3=VmEq`$Y+c~uL*TG%o_rr=&@t%;MUcX2w$JHU
z>FPghR~Zkqf_3c;(34i?yRo3|mL5OKB-I|9)Cc8dB#F)N7ARp*=nt}cmz@jY6fZke
zS1kGlpi!mcneM-V7EV)qsbEEh0w#}v<!0e&I=uCfl}L&vw6Le!$^3Mp@BZ+Rq7Jm&
zA07*@C(p&=Pa9MXw4B&{o;^WuM0Fc{+a64IJe)m~Prb|(h02*+A^x6pdY(=Ghfiom
zng(0v@?qgR+Kw_4s?KP0mcr%gz1j5F8iCGzVbZX3SU-#esyAPfdrodHrnPL&202Ec
zWm(tz&KOX$;oq7s4CX!CrBP96FbC7$<7!s?<}Cl3s`)E|Du@=m6ETlFxA9%Rb^?PG
zb*O|Qc~JbS4PovR2hCTIpnE|gas!X<3RJ~A5MYiwDyX}R051~;j^M41_61YR6c+zj
zsYvI|g>hJ#rjEWt?OpFTC8fKb#?!ZP+Z0>o=>PwLHGsu~>HI&Z%d|4h;+|{%&sD_Z
z@{u&GBp<zT)u&GHUQb$J$jJ{T&)9KHnJF%+kHp$O5pj*Xfhw5cam)*kPC5wCOOlXR
z)3HD?ffoow)c|R^@U;{Rts4Df=0#HBBC|1VSDA?#im|;2NNj97ROhX;Bml5MXhzSc
zCBxW>5UL<6qKZMXtp4!*wp2&1;PG9u=DQY^`WR9}Idr|k>a?#W8A&K%ZvG#Xg#f+@
zSgHfO7qXmG0lWcg-^)L@5Z|AWMzU(34yR2oF2a-L+Je!h6u;h$Zy>)z@hgx#FF<}d
z7~bH_tpK9*a{$isBTA5PD~#{x-$rLbC(^VZ3`J)Ib%&Fc7BkF}QN;Z}P>F62YeoQ?
z=mTdW?SGM*h874vzh}mVK|j=9wTc`Q5ad%W(UaO3SG^^imU!`nNBHp+R?z(tkbG8?
zVPbGPMWH&k>QMaqyC0=I2ys{2`ozd>h<tJ8;<ti`Q+hH85)cvBfs(7{sYp<}vx7<3
z0cP6yL1e8DU=pYG<BEn#dfBcP;B8YOhW$qN%WX(&7qF|^9H`mF;@3^J^TTEl3f(Eo
z*uDM?`jh_Z8{CX_iFltEcuP{>Yt^v=9o-q!43I(xjDPAl=PQhhLQkL?kr=A-^^JJo
zNVE|oR-T<QYfa-4x8tO&6<ZDqX{JR+tu2*pO0jpm0XDSWC$P+}exKijiyb5!_pALN
za6AByYPu*C1tF7OgH=B;(&|NnkILTfl*>w(0H6UbVqpH~#L0xhkf0EA!5VPxWL&&|
zeAV@}M3Z7qDY^2biyRdxz89=3EsX%WEEK@o0n&T{;@O~TQv0aP0TXl#(wx9~X5A~9
z0{@W9jQ)0`{DN+!;CjnO3K3h?xF;FrYNJRk6?bgIwAeVF`d;u0AL+yvces`VBUm~%
z(5YencPPex-wGrd>OF^S;#pE(SIfH=P|RxSo>pvM>>T|vIu5?Q;Hc!>l~y}hV7&US
zS8q)Mm<t^XOEiOOzAezzUKZp=i{QZ)I(*JH^Cy_7tLDjttI-ri)TpOTjc-yxX;|HT
zH3i{)p}GlI!59@0nG4G-bXbYjAo`-GnjMl+Pp$^CNt`HR2fwa_S{vq+bL)PV<zBs!
zk1a<EKmD#o-6p?z<YP*ubAYdYP0=}5+s2SaWYK5rj(MFQmPo?uo_^!n*KIs-H`58y
z>>r+=-|0>N(6_-{1)$vokW|N``2>I;V-piUAI%gx*zcl=@04~NoW*78MYUrr;z53`
z`aE#RVTuu}w2ECMs`8x}D%UD2A=eiAjtA*|E)^>-_^JiL&AE@HuAZ?X?`qR^GZn%+
zwnA2syQt#a4Phu6W2pXZF8)`lUyqDRfYiw0wAGB6F7HB!w)BX`SNxd=T}9x@Dv<C|
zQ{0hAY@H+%Y>4{s-aISz$aO+r;;JVke@**Y@J`3ZaM2^~?)y14srxPkRO}qwCTzOO
zhlSoTQ>PI(28k7wMJ8+><sn$;z^=+%3$h|V{nh(X3U<Vbi^=-ulGpvGCuD)eC4q}d
zCs~&LIHMw=VRUEH=!iBxcMO%Mh^B-cCqhh7RG!utt3&f5{b@I`MlmS%wyCyf_2QRJ
zs2<_3)Rpd{9Svj>Ri1>($?Q=*9x)$5CAkD_^uAnXCqCQc+g_SFeVF`<U!;sQ3f1Zt
zq+4~ug5HTnt$CH`I4hOeuW8Yz0f=x9e!0@K3)=3}rV<GikD=8Psxdf7{nWXMo2s`P
z*oc|PFtw$t@iz=V<=-8au}GTY-jo?Uj7)E1#o)f~e!zUS((XYolO>w<;UL6eKIK|8
zH0tA9Q+#7q8A7@1duhwLbYYXB@$_EJz}DBGQ}*bcK(c_`H{t+EBbI4_XDd*6;(Y%;
z<<H4M&#_Jre&pLLqAr74!e}?`#jOn(+p^41BUY@jgU=x5be6U=*gZ8<|NAV5dou@P
zZpPf&te@aE*TI63@}tbnXFa&pAM2fm>@bl-yBA_x>iL$iA2sH(8{(K(fE}@exkW^k
z)7j3k1Uq8jTUl4?MELZm51B=u`JAK_zW45+*BCFEhefK0>+nQHSNtK0+nW`ZN>BM1
zX-i&Q%nsmuB8#4)xHw$>qCGN@(4jcHRJ#$f+~VN?`dcukfQqt$iaTF#Gorw=6xgmz
z#k`i*aq{KP?4Nrg(|mY7QiNpkD_A0j=fAWCLqGbYk*(&Zd9w|Tv~WO<^4n*UWKW72
zx;C#acV{iAMLi}kCG5kZgz5R!(=a{q%l6+y_ER84Gm|p4Q@Y`KNVU2NBL8{4-1SEi
zG8vpqF%CJ8_%tO5p#IE9R=*rC`n9tz-Dm3Uc`2LrKh4zs9wu)o<#*05Eo{beALcZ3
z?aNgS>nefd&<nta;>Gq50caUa<SQ70E+e%619YCy1jaAbo_F^7is}DAcLLf|oG4Hd
zp7@>_cQ~ihHQw!^f<q7|BWlWOGpu-SKY~_-(oO0$&5Q0^Sr-tp@50{)A4m)ED!KFN
zheiIM{xCAJ0WSpYwZ_cFe-7g_^y(zll68!2++I0lKBmT3T@*b`exhvBb(wTq_PWtw
zZqkQ|OyxFQAA#06rhLifR>m+fAZ<Wz66o8%t(Q{r<bDG7M*g+dTmncC&_bM&4;6I4
z8{MEl^Z*LexCxOHyFjGC5F;O87gMvzo^F&^$wpSd&L*#;mb>*Nq@l6K!m{`teW)5~
za`H9JqJV+{9-6Nh_WDX#rZgjrIH^*fb-swsS<<cLBZ*(P2KeT}$hA1*Q^wJ2vGdMe
zZ~PTO#gT)9tm$22@nSA8eM^edB6gAo8lhoN#5X{Rns3*`p{P`|ob9BNk4_W>6vkrD
z`NI=XCxN!HwY4>S)Lt+C|7xKc0;(#X#QKT?uH;Kaj}f1jkH`*IKLcoQ!Fvj!Y9=bf
z5{^}U#)^wddw9|7X!M4k!!D;%{uBZiKj{8*KDV{^Q=4I)U2#oP2U+jhA}i6H5oPrZ
ziTG6u^5#ZK%9nw+wpQurV)}9?p;)BRzwpudm5fgK?aEVSvh;=hvyRm_eLZ3nxC+cY
zju#uDk5Dr84`{Jasx|UkhOfYva7`o-1GJ-{A`p<-Tvz=NfPQTP!U>4G;h;}C+U(}Q
zsbF|8T_!TB`M;JeR5fh|%(t(45C|#JD<{I(KQA&1oYN7RTT2OihG%~)V3YIp4Ce);
zViIUNkZMc?DL>&Tkwt&~gkKzfYJ8GsH2sHq>F$oAn%Li;EF<go<|ZZrPF0sfGFzM$
zV|lIEFr&BLhXbrccZ~-s%=YBFNkZPnfAyXgxy;nUxBSL(Z4$_?NERLpb@^V3LvLQ>
zzzqT+T;OKWLECLB0jjii^<<>1whkv3NPfVmUl?MnX!Zx)8=wh*?3xPI(m=bQ%4&f-
zEG(=6s2`YOtX}ty&^iDW=(m$i{u|1re+#shOB*ku689MD&||@au^e@*ZVBdf><&a^
z{+2g-DFwuUyRg+UW0cAvemH0C$tPCP+RKEKb+COD#PoS4WkTh{8Z+x-<H{BP8~*jP
zGJ`ukLwjRjJ|zRNI&5_}a1V>Kf1Z?r=_pv!z93p7Dc>v%q7%zCowsOkDM#XE7%4}^
zd3t8AJ$}sjcfe6l*Pb@N-BAAE7@M5j54e$LgDnNvOuqUh=Y?c$nH?NepiNAfZvn(s
zP+1uRdU-KDFF>jq&Jgr4oh(!_J=g~Vmp-rn04*#wCdQU_NN6WCtQref-Ko~DmI5vh
zS-9Z#W=2^jPh&qlxXg@&IJ}is84}J=#CF!H+y1HINu@w#AiOVw6=!GOjogpAE!bp4
z7OTQzqYzD+JNTQ`B<jtyy0onFima)=!b#{aVtLU79K%Z2(@_`vSCF~ks^WLGzUV&3
zb+X^3-`+aZ=iaqQm7bQ%K=pLsiU1;fC!l?%E6=;a=;<|aAULUIa=vx`0Nv!u@f{>2
zq+s4q&@$t5E^S6AbvRyKZVYT5k;-usAphv5@@|XK8&xQd2659QcB3ghzIP{=m~!|=
zDUIm+`d4A}IqmGA9J3GGK4Ok5Ye`YD{)8O2NNQgghHKlXWUyk7D#+dc7RnkLS$v3X
zv8J;*fko7-ydB0!l@_ljigJsrZjWGSq#3?Kjztkgr&eR7sTVIVDsI9B%tP1Xhzt4!
z5PA<XOS6cP5omtM6{G;ag}Bs?tDQ|o`1Uz@a&)C^3>PSVr$<^YYI!<k&E&Vy%5R*M
zZ)BX_k)7f1{LqYlpNt|M=+c}rJCG?3F^V|5Dkdfq)W?x^6&U97A=k1LzidFV2-rX*
zmBV7g6OLf9pgJx?hN-OGl{}=0CcU{{Ox6ee6R0!>Y$9hsOK||K43H_uVge|BRj2=E
zhf5@a`}(sOIVWbAAlv~(MBnvp?y#ZNI5Mxm>$1UDZxMwx2<!i*jaUCW+p_ScdT9Ap
zxV0sX_2HeDLfM~I>gEY9DWOnm?6Iw{lug<m9W!PTa=F7vX0Fc6u$&;?DN-%0!!7}v
z8U`{_k!k@N=;i!!I+zBMV)}&0kXJgO#r`2t0c_lR{=eQ*x4~EYv(=$dW@Wo^1ro+9
z3xA^dtsIc-wjG~Hy>ySc?Ow(zKGVLOPbO3ErCp2CnE#3I1pD{7y^({2iD>jsfOJ2?
z6<5?`S=Jd^Oe1L3cymp=H7r?Yc%I8xH>~|nBl<p=l+7RQfC$K-!?f~OAQCkjT+jC!
zgg%8UUg`XDItpUA68QJ!*`LLQF7ciuWLIFbvMl4gcEWb>0e|SAeR*3f=N^p?Ja}O3
zp9m4N<FT6hz;`%I0+$O|h;0Qqe^wj&|1V(|sx9}U?xpen5Brj@ez~(g|H@rDAjyW0
zVJ}SVXM1cphnq`~gRtz=WvG+79rLtYx$2yN0ya2{7diCM2|-qasi$F*iWJ$`plf;i
zQ(B1$yt=2ch;Pf4g!D^wQ}gf1fZmN>YYt53(x@;lm?6!GeW&w-W?ZNRgR}5w577Hf
z=P!3d5%^vd{znxmZS)ux>bDKq3I1DO_GWj+t@OI^{#sw;6;JiXSdKO~ffo&aIVj6?
ze;|JN0cVa-VlXcnYs`37WrS>MM=`KI0<cL+PixEf2pArFhEjr(4hiep!T};?Xw#B9
zjCt?)iAlX5+_3*Rzsoi!gb$Fvco&M&+`#ZoZs?%~=M`DEPqXA+vr;T4na7tcyMcuI
z(>{j^!<r|4x?+dxPZ!>*PL9PQ6L7&ZL2o&)X1jC5kky_=AQ-REpZ&eEj;U^H_}Qkc
z{b_2KFn3~5XbqU%aA=CrXlm$Ims(Ol`JldE4_3BR36A_SfZXr>0{`#J)2r7wTKGsj
zEiy(U3mC7TZ>^B!-ko$CsiAGYW_7t%Ed7j)r|o#J78(8Xd`lIW8qQYnu(VT9?5xJp
zdwHWs#b<aaJb4#RcTRYZb`*+;WXxMkJzHV?e1FW8K;m%}10?NwKYXGreM*pi$I;=E
zMFD{+5^J2M&TSm%f7(=h>+K#UqOy%?VsDeK^!zjT)KV+VVnX1px2vB&gJB7=R_nP5
zh&MmW_?RK~l=`#|JzWt8ybXJV*_IN-TouXB)JB?ODxi<{{h`kD@p%IK3xgJS&>^}t
z^-bJ@omH~SWEF3CK6$9VddsiALM)t$RNT^kGrT5aU&#4-C9<f4n=6jnYwuX7WhhHG
zrn=DG8uv@{%t4ca^|Fb3FrzB{&=P^Iiu?Z3LR!45Y6JdT84K;+uoC^#fgz_2?{bTU
z=vA=)HiU9xt_AC+E}@}*kgj+n-L<A(KSGYCbV)xxEPk~*K{%JR376L5!;HKH$t9eu
zkwKor{e?69L?q+BJT(F57ff>*w4Mf2`R}mnWVm~T?4SSOcH4Rygu|2%IoR1GM`K}1
z&ANTj4=Hxi?oD{U!sN}2^7VcD`rqeA@uM+Sw24vs55xlZw;COIPhF><SNSVfRRi8O
zztr{q8J>#-KE9|(44h|9wB8P~7{*E7u!0wKLv2Ek6g}q$L&9q~Hk-IzTL7uUzMV>-
z2AN6<Fm1$&a1bxQZlmjAQ2Sj+L#4z1wzx5uD>#UvGOKm)KZ=xL>H(Khfc;^x(8U{R
zh_>-{VxH)ZzE^1Rn{+9d!O%$M_3@a<(OL_3Vb4@0kuVW4@|eo7*TW@vQ8=foi%l@K
zoak}^Xh{&lFK(LMKCM#EU2LYhHkCt%9TkOw_XsFsQZ~Qy?zH0{uP_V$gA|GVDftyL
z{wC8<?u5kN9oo(ZJ#wR+t&25Pn`}#0gxD%r3O`G3?5P)0Be@}YmT6u=XjVxo%wQMb
z{|TpVY|ML$lEAg7wL<|f<ohdaSuYpco-~-eAD-F*4)~IKKDEJuqWH^mD;gJlFzm$$
z3F%c>AT=C#ueWhcse@meV6u+S(T<!U(1|b%3)5$|@vsaYi(^+s_9e}C`^KPXA}R1-
zS)3c*sq}02abEMITlFBxsjPEpPQ2R9YpiK*auW~Es5Ux;2Y2zFIM;)h|KRY^a}x1X
z1dh}LSF|V^=xwcZ!wk=i44B<dcX`B6$sd>h4J*H1bNZ3i3zhw;q6Nf&wtagHWd9;9
zJe%)r`eOdQp)FT$X0^}+)-wHueUL%rBF3IG6-5hwlQ3!%p)49IFUB{#{EiE$U!OMT
zz$vh{F{uR{*TZGsUx?H}VN+eN+^~@_3b)?GSKJ5X6tPhB<2s$bimbd~rg1)vlll-G
zdRwALUfnxXXnY5tKDkyviZTd$hKy%;v*)Q_<|UYYeQ;OR|9B4J6(rc<NK*Fu3RZjR
zs_L7w#ur*yJdjgr=uI6?BS8G5R_!OvuxL?L);Gh9L^&zj=^BL`*>ln4p*^3h1w*Z(
z(L(q1P}h>Ap)&P&+9L6V`Ur`WHJ~MiXEV3e+=k@~VIj;j8Ff~Q=@Ze=xCU$)Q_vP=
zS<nJ8{vFk;{Q0YpeV+69%S9p8zB+pIr7M~@!V#T6YZ65Iy#r^QPKUG+jY{4)q;KPl
zxFQc7y|ua{_a8row7YiQ`K?}0P-rP<q!0ADKfHbYMl<_}1H+WEaKB`hYV02}dgzGn
z*ObV%56BK>0dtx^8q|y&@QZ7heh1J_>PKI0P&kij1diI4rAAqDVy&Dob1P@A!QQKW
z{nP=Pp3*PRPtg|Tut1;TC0i1k8`tx%lUrSG^8K|SELgUL>7+>J{w{IY4s&|j0790o
z@L0Jc)3Bxvd9i^{{Cxq{oCzJwwe|&085s<B46PQalJeC35D^E9%GhY?Mc@Py(Bs=P
z+Yo9wZ{UV$>635Ro)<r+Bal-YSx^IfMtl9w1d}k!%vS{2mx7y1gTV)}XO~a>vYDpA
z_teak6je3dUEGn>T|2RF8WHP9i{yNR#H5>}klxX6`!SlIpIF{Ux?OCiS6;x_+Gt7N
zh3Um8v=4IhBN~U(t<ANp2kwmX@{dCEX(UbU`^3q*DQ{ArW3(9R2qaX+xwPd{g~GhX
zq>1^gSv0(_oZ{42%Vvn>S2Y-ZBDPvmA{xOKpdbaNyYK?-PXtZ`5`#mq@@$2p=oQ;V
zv?3@(6^Llf|Flu7P)ImUD=5?S#KGByu(CV_sxM0R8&;ZWD(^l<^=s8=5&^W2mj4|O
z`D6$70tLlczII$`w-u*JttWz@6J51+C)p%pZQ0{Qa8GC}rKUk_<2>fQh<X-ZxT3jG
zdI~g}%u&qUDBSc`f*p>?U;=Af;%_Z_gU3yk%!jKz?c0chES}^6HNnyhZz##2%^7eC
zj*L9C!}c4uOk?yGuMWfpo0oIHg8islPh?kq7g?PrOhJGy`<IL%vR;;&z$5&ELF@jb
z#CzFp0d`mC_nmr`2p|h1@hFl5k_tzBr!hVcGJMmxE_+hQz5|_#*YT=`LUFOd*?Kns
zsaU{30g$5NNQU6Ym5FBNw334Lq4^NrJeqG-ETZhB*|7=vIeg(&oQGj-LT`MQTlK5`
z-u8UGike%L+c}f~ZIFV<#WZ&In*=OeD+;dn@BwS1JUXIdHY_T9E|m7&&TjpOBPux-
z*gjJiy=hubn``q;Fw{s|?kw?@v-<aElRI`yx7;5J;GWacH@)q6wkI!GXM}F+${S_r
zZ}-d=wYI6$$S_GNXx&#~5tOp<ffDlSC?<|ndPC4p+&|=)XMLUzz4uW@nwEi6DF{sv
z24e8{Q(LFq2!TDR7<phAHQCb$A5h-Eb-+RGfADC;eSnSW7TKkqdh|aSz~x=Eikr6E
z%B!{U%QWmQ$?D*}N;@b}<-C%3P2-lN`W4e_4L1J$6Pi!EZp4{cm(G*GkG@~n4e!!x
zkf3P(-DjqS-VtHTzk8R!=$+12z<OJpm?YmGoWFguy8P^NO7I^UBDsZY-Fw73J3(1G
zp_5i8BPIMFE+@1wOjiq^4vuYDd-B%iS_{w9Bu9z!zy_9WioEsq49h99d83-(=Wj#S
zi4pgSAs=J<*<SYfK1sw*)RstA#Pudsxg%N~)d!>B6y%7w2JWXIVIN@#F^YF^ZQFk3
zhu%<SfN_YCg`|*gk@!#8DLs}y8@oFJK?Ukw1@aHn{b2;Gm(nlNfwm`KW%vA9Vo`9>
zKAD0e5E7zWuVx;yH87`m{lJ4o|4-`E-+C8hb4})cbOk8~et%8(0^V+|oy(jTN2g+v
zf`O3;GcCJ6Pe6DS`wQvzFk5C1-F_cnQvfj#M~X&oJtO^~3`<jeJw_&$A`4;zoNuH#
zEJv*4qY{Sa4JRxcH!>Q5HL?=;SL_LOL24#Q<&hk*5u!zCFl)%5uaOX<_)$zt0WgCi
zuMwX3!8@2Cov2zf7L?e~O)p+T8otnw@)^U3GLONBZ^cuA^EaMS{f9S*AKq5w-B8~@
zjwPJTH3{36r}|l<(&bzz%R8k|sR{e7KjG(CP%t3Wr2LzCl)Mf+7+ow6v#K>i%naT|
zJ}%?jdAz#-eh0X)99F{~6h4>oCVqn~YJS%dG|oMU5u$1P2zq<18?(8cw*<-CEK^r!
zrk+}3(z?s#ri%{<qdpycQxW<DVG8l_A7UBn-2LD8ah;$4GgS<NjYSpLvn9(QkvqfN
zrX={Alq6;pof`esBIi!<*1?7J6FTN1(z&q>Fos?R8QPv`h-Eck6p8(2t?rGHJ+h|h
z#$dt{Q2VLkYg9+}?rSC2L2#L-a3EpU?#-9JyHjWJAYzbA$%-6SD)4in^U;XA<{OBn
zdzLU>t$brBr}6QUU*2)ab@PG&VVzj`DwkgGo}{&T3ExmPq~)C8@IAaQM*RExrcYs2
zrxqXHDNtA69QA$Y2j*9(i*_f|Cj;m>5a63jo}Hbgn*UxjLrTN`uJ$*E3{rUvN|j8o
z@lt<c<-Z$w2xf}xac@1P78T-&H&qp$u)kX8L&HIufIDG@V(mK1;=ueK2NI%c9t1N$
zl!C3w-Y>b#j0{4Stk8j4fjNEri8?N!BQulcW<Vvl<%v{;Ta%53N2(C6B>)2vchE_T
zLIw*+PttCVXm&FK5=Jllg@XHxpm&To=g3OVq}35}=)~V=uWUP3lHJvXJ$FPKg_HJ~
z#4Uc%6InTSh`m~(qsED5tHfI<`{6=qXKiFfVyt3>>UHaC^LGNhVdwK!;z)}DvDe+2
z4fV_vHyc`rIL5Eki%<ioIgM(Py^@nrNciFbe2SgX^@s$u__lBu_HX>CZ<p3`uQnaY
z+`CBLJl;?qAKTgef6E3iiL@l60$(eY!VvSzi;Umg+B>V<KB)pm?<98fapzDBbl}t&
zkC#ep-{DefF^=2As830Qb_3VZ>EVA4_6|mIK@d$FaFYlIMCFm;%5v4<B-95Zl+q;6
z%8_-}`yhjG5_#<bfTG?lIRfePcKm0My-22p3%#jgI}JeLAel3w$$5y>Aadbfcb!s?
zKy*d33&WCrUE6&6;Y1$S^uN0}vaZZO1)tJGsK3BzqkZ6b^Y})y!G)Q&xl=vRr6C10
z0;nEEr6^(TmMV;|n<XL|F3>2cC~qCEiRbHhpFh_C@vH+p>T`z|{>DC~79|={ZeXwK
z!?6v0@&o4c;roWVK?w(5e|sMLN}=vXZ`xQv8~zexMW?8UW0B{Nn_CGrW?pD1w5Ia{
zv0rP^Uvh<UO_%MBrUELvR_)(Z29q@Yo3*T0rK^lu%Hc^df`Rk)@x8qKaQIE##tNX!
zH$KCyG5ck|OhGCPWicVj|Ia<02%8A<c#0zMC()1EuxJGnbmgnUA=H$7=z_;$0;&d7
zgGD=Q+WdI^g?{kWQxRomE7<%e6u0(uYxA51<Icoz2jVek($4CKoicD7L%yd`U$WiO
zKfLpd5{+f$JQ}%`$ZHmO{$vftQgmJS)G4h%9}})IY?($%tWMHEF-l+`-_X-*y}}K1
z=z#hnuN6cRvhmM1j0c8}J7;9T!TW;AevNanWx~iX=;9A|@jTys#Z#NBve6mztoSbb
zo+0@P%@|pTZekR|6<hylI}voC3|G4q3e^v6gN?0c1JJC5I$QP)>r(H@sbQ^BdnVdm
zRKnm&!rY|CRVIv$<v!GXe^T78R(#v9PkEM)w5dcGU~<A|r1gOfFRFdMaG+o5KgM7|
z%OE|foSDchcw45VKnzjL+tczdq=O+4BTzHMj7GF(TfC5Oo>#D$?mS*$FMIqXDku7}
zG2+sfbi*sAT2<Q`($KDyX7BYM5ci8;1YOyW#T`6i#3fKlW~>S>1^Bqafrj|n4<?RQ
z0ctRVnth3^8~pnEq@hC&z-a)?eOW#)&nWn;zF_zU>V%b*m2KF|PX&VMSTgrZ1XZB^
zTJsXlHa}Z(0Q%fG7W&1@yu=f{9sR!|H&{oe6!G>hws`3=`RIn9)s#Hdzp8{w;>WRg
zI#2f8UHL!fCh*lOu2$#ksl?af8QK$>I~@lZ@t?n?D7~V=nQ_Nnc^1qjw-%3XPr@2m
z9GU-FLsH9x4zPT7y{qbx<G#Qx|8qwR{dJPhdn;>MOL2`q?eT-p#4=+T|0S6hV7hEM
z(0^moAiDXR+DHmcYR4b#aHC@FSstr-vIs)K6uw#uHkV;bU>A!rY&S9Txf3x(7_9N2
zf*6oL@8dlul&NEm9l@i-%EVVZ!Ga^&d=X=4O|qZ11nfua7k8|Lk=md$_6P>xn$LQO
zpp!_0nZlW*g6^X=CQ4vX2~+HJzoY^>t^jPyZu><Ax0e%2Y^mrgZ&l)JlxVJv#yCaY
z8u2uRcIk~YJ2a0!s@JMIumt{aI8viVweK9xT~x!P(iM?RG`=?B?fb`+E!x>>L20DL
zx%MRK(0tO|(*-}-w8Jm^z2##TH3CroLG4;$Sa^&1iY1C>Wo@nz$58t}%%yT+11eg~
zQx(uW2VE1qp_A_T*^n+RTD5=ce4Gu$^mEZ0ClzOSM*3#6$svkg5ICBBMb5m5!)e!9
zX2Wa0nRM{$sPcv|%XI1SlYX?+SzN#~43VWY|68c+C9T%Brar%W=K4Q(4`{FfizywL
zZQxv6TO;QLQqx;P1PSsn+p-7Pqj9Mc37VN{1DLvZu!?uhd|Xid8u%H3#~GHKkujXt
zVH*Ym53d)vU7!P2&?^C(tRe8OH3M^CS9kXrK!??!CjlIXNFvtO%;9(}k`QWkcJ^Tc
zx*U3Tvn#{dW;S#M<ZX5M*Mw$Q2CRfiQ-U!u^3H)Uxu?nuKB~Uda!K;)FMLQ-(H<9i
zt~K1>6s>=7wcLEe9eU~d+6FuQr?r%Q=t_?Q?rud$?CdvBKu*x>e>3uV!!5!=25*Dc
z=UL-A^!yV}At)IUSDRj5WDL<G!t$B2?-})9JU-f%>ETCc_MNOP{Ykl{s90TQkACb&
zUtg{w^u@Xp6!4gdVZEG^jZ&$S!glGN==E9x^Kv`Dmh)|?$;tD)pAHybHe*g?rD+$|
z_WC7}iF3c|0C~Jn#jIJY&hp#%FPTNVUJ@9nHxqQs{GfwE<8mx(wjY#nXsEBh*UZts
ze~(IXa`V8z%$#3Xh?qIzq7iiDlbq3VFi~bg===xi&ysinv(p|=)>u)vo-9_ZptWRF
z7;S%J67|d{&~0ne>m)-Qt#-nyJs>8v#r^90XhTYSv4!0C^eDux*984mGGyc6!RvAO
zBe4f!yNJ_Xb95|@*k1iHDl&j@x?8R{!+98p&P?KvSX9xfUXkVF&K|xQh2EBL6xB3^
zlHsqT?lNO}&UD%J{veJVPi0SP<JbbNr*)Bp&T9#3w1tlKsH;Dx@SKnh2ds{i?!2du
z9Yd`6acZW|iZS;PC2KY)^|8<!G+WoT`6ETlhV4jl9gdZcDkx}ZXf(<`M@Vj{s0+Cq
z&38lRC7`BBcXwq2N@nH;65zwc69V@5QH|hEQ80QF6chx!_Q66*ii+EbTGp@pk3PWq
z`E2&S1Xq2H)E4~B<%Pg0X6oN?i#;gDhi0kppnR~LN&oU7SDqDL#>Rld3)kO|>3E#-
z)ws0#j|nbK&GIWvII=XcT-P40SuLHAmaV&&dD`qZwvSI-<yRcBR<YvO&0B?hpHfQj
zv+vOvPCmHbmoa2}vwA91c+>x0O-#dbYG{r7s%vYj=)YH=Qj#>zi!BnOoWQR}ty(0+
z`K2F8uLgLIpqzHeWyKgYtRg)2%}U>KJo^iV_*Q&E*4Daso@~7d*01^&G$`Zxz!=T`
zAj+vk;D#-=p>c~VrpngCL%w$%UBz{zvvmtG;1SA{_4I#Ab9dwaS;UQig$Q>G=0<Np
z%EITdCJxJ4wf%rE;ClMaP=^`sV=x*GnArsVXfXWy5j0WmfU6Nh3_8x#<Z_}248C|e
zj+kJej~GlLKu3g{oJta-l&3a4I-e>qi98w(Z;HASmyrRv#2PgXAs&3o_ph@<FUAkC
z5P>WMw;G?OW~FnwnG908I<OSwu*G8fiVQg5GGdH8$4)9#PLFciTk#3wD>7uylZuiK
z>Wb`}ixiQ;SdN{I4yJsYx+ofQZ{l*z_y-@6?i2OUUB@>-#QR%y=pvFe+Y!=oSoL51
zAIrU#*v^j8v@KDtuc#<&GPqG&7=pMIqs{tdGjCPXzxR5TVQA+?gYFRM>#{;;L)#wh
zPkH+v>`|aCRYX6o7P?TLRw9T6SiyiYv&{c?_~%<#^v<aA9E8!$4f?TK57Re&ZhR)a
zEOl*(4*#~CmkcJ=CEyOQkEumA6=7<7yBQk9Md?w=?;QS`C+-6S2(ztT{9p`@1B{6N
zDk$J^IbQsaBxKBV6O8(_pWJ|n($L7rZcFY7(DMft-@UmSIxs*CyV`V6o(qgb@jrh~
zG&=0W9qVLW*c26xedtNe2VJ=8_obiOtbXZt*CM)}ZCe_?Xnk2-X(M`tqvR3!YfTSG
zPV0tWfed0h_`>-ik72fCDKj+`sF3T?t0G2n_@VXQ;?JAGr$;LhSfp~G*%3$_h2amL
z;?85u+LJWlK60JF_Gw4l@derE%)1oz53ya4p`z8t^)fPL&MTV9bIL)6_aUvjqUS<n
zel{;?)E_0Pxb<z8q5}_lGL>8Yf#>Z6Z6GC36qET~m{AGgRI|gsw<(>Go%pdeTr)g8
z(4l;*qFz2<Q^7_T@c3N)a#V6=LS}y=<td?Y0A{X?O5fhMz%F5;`ineUKXyN`(M+Nb
z)!4hy)|gmH#Tv)xKizw7j_WLPHnaT0{I%=!Or(|y8%vEOp#mD!zyaSANfa7V*i<=W
z$~peW<)l-{1`KJ?i+Ix?EgU?(8Ga`(6Nz|Y|DJ|JNW4xBC#k6YHRqgB?&iqNsqv9y
z9UO`8*RS$tmoruM__l)|d=YOkiYp-VA=KE8pQYL%^FQKWQVHbSP@|=|$+T;&R=)m?
zx0zYLFzaJCI%cH67{zx#Qtx$Y!PF^-b_*t}AN8FRsWE+)h&Fwm$o<ykjcOuy{cbQF
z#yWf7bMJyP2<G6(d>+}*1uroGnfXAgdb*6L(A{E8+K2Gpv?#lIgA0?W6Th;h6g?b{
zKV%O3^ysCx<;76odb{1jrg#a3!q1T$;RB4U<B{z7p%|j`5ejz%gt3k_@;k?a^9hB`
z6tVeL%X34y%KxHL8rU|WgBz(ft?dU_pW3{=z;b=jDVQkAmaY~%%!^)MsDE`cEdi6X
z&q)mwP?$g}q5$Ga;4vwDYib;1>Wv=O9*B5LuXQbWd=$>slZP>O^&>_jn}Qt)Oy)5p
zKZ)9w`~EOuH?Z+KoI-{KvN2y(Mgzls=Yjf1_KPHHXUn1PwNo%L#cS!Q-uIoT`fhdc
zn-z0Az8?C@^Lq^<C)7L?{u!m-MrZiyZbk8*-@ttcTb4%>@!0RAyFmXm0hpC)05<0T
z^nNfIhRBN*2L9R|^H<-mKI^-Ft03`ys7U*&-1oCCM$^q2NkvbiHQ&~NH$0rIBbMD4
z0oDFBS%|o61KX@0kE8pW8e2*{JhGXmabAz2Y1}?w50C~P2?j<+DlRVE)H3&*9Vc_z
zo(9hSh&#~Qze8-pvxqy+e!4sV>NTf-X22<8ug}F@iy_xJOrL0r;qT874dkoZ0EYS!
z@4R6z;oLw3#>1mBmM&mveO{LzEP5Z>x+-*7#{9udOM&(c1=@E<Bno950tw#6vQI+?
zzfODY;>#X^yS=KuF{z<M5BTbMHof^?4;_5OTGhDZayot>a99d(SJ(4AdjnO~!32o$
z`KC>wR(-KI1y<~n<m&tfgq7!XM3K-4%KHvBDfRXQ9N1~v7J4lqR8CYR4DU+LL#*Z-
zIAdvZ)Xb1SCt{P;ubEK!4U2PzYzu~+=6<{L%)NQtZ`wgj|Ercey-yN7b5d}E*Q1ad
z;ghoRUodeS$TT1g5f>GG1s%1U(W_uF8<D{A;QV$3uYPa&Q1_bF%KW35kwc8pdUI++
z=^WuYvkw+ElHM_z?2Z@Q+rFNxtSpZS$~21?a=?gjlzamGPclT{w8{IEEVwADd#{M6
z9UtL#XBmRH7;E|l7!mv_0@2b{D;Jc?-{l!?FW@}gyYI3LJuGOEGrHD2MV{V?wQzbq
z+$cAlI$)K11+1@Qk$N1e(1PK?+-Sj`&S-(KHTrB?Idd%tS%%_WvEJyTE@<)?$XK|@
zF94(EU%krUbK@1NK=RU$HU<dMB{)2>L(YfqqmN^-A(RbQJ^4(T>_O*;OTUupW4+fT
zl2XwHu2IK%OQ01M7q@%1$`E88;MWI7w-BM(NvyEACr!9sRa^PpA~RMD9gn_AF0v`6
zIs#%{GVUuq<8Cdu;*um^s?3}fUGWS4SR4Y)!g!{a<o9|?#%+^<tc;HPpML)O@3Qf^
zKVwOvR3kBTg)GZ14(=Y^a&Y*lJ$~^X)zEBZZ8rV2`WGa>9Qg=4GYT2(_p6<aXCwNa
zV!XewJfBcFqie?XGN}XDR*k0ixfY#vNWfo6I2)*F8?RTgA-)i5Z?xl<b!EbAvF`jN
z;qS{<MKOjK#hafVOsUl&(`@#d4NovanX%v{gZ>Pa(0(jQ1<t+>=m``Ov}6zD=`|VV
z4&?9w6!A2#N9(M#z__zAG<}GI(=nv=1Iu}nQ!!I1Z*Z+ljh_-JA?J~ulOl!&N%iPE
zXg00QH($oYYw-j52r1qu+il=uII4_4?lh4`?IfH59vE#gb|9@^nF^h4pfce*0xmrh
zGM0khvjNnd#kdbu!O0;ZtPX25dz@B@u7d|oOc~;9(Mq|ATZSPr6yCD~plc=$!Nt*1
z+lmVD3xj*@fzrzB@wf1s@?U)pcDFsDo+tw86<tBsgEAhdIU`6|NTxmBB;ZCp0C;`l
z9#=U*i=l*l>lbTJL3&CkiZe>Tm6%?!@f~o97+#ysfGkm$Y5DCN1ysjP$p<nR`1>Mk
zhKh%%GCNEk5sBq!RRlKPbABP1{*fN~=bjg3%n6Oiyb8Zihu^=2fCiUQ)el3NhwE*l
zK|e*R74`D{|3}hUhE>^YVVLgj&P@u^pmaCVD2;R@jndtn(hbtxAzgw<H_}Lll++o%
z^UFV8${yZ5@61}yeJ|>}M2g;wMr=Bk^fU#Ep3Etn#ty7S3We_(;Vf23a_6>uR}%5j
z!*G7N!t?*G-u^vtHr-iI9^+zSy$+IO4EDR=WQ$S56!B`#R=ttIA^yB=)ZqS1aM48+
zuA>!E=r^ybcv9l7>b`=5rrH>A9zK4Flmv&0pL~V{A{|3hHK>-*Gph=YNdjMM%A%^>
zS+>q)C`<rSMHDSPGo>d_=tZE{Llp2;t*c@+iS<5vK}<RGZ|t|W^BvoQ-}kj1H<f)@
z*vdcXVvR&%8f>%bW5II)u;`b#AQE+)+1H>TI5dPurS0x0n`Ln)kuKQrV+7(3V%<rv
zotdut_Pzsbh}>pCWV*`1S&h3gM*;*ejhy=JMVcfQ@2sRQ`linbkEM{6k>o!ifg_1S
zK0ivJE2DTWi5-%h82`4bVFJW!t)LwH*2|i18p7m}$)b>i*FG}JB2)%@exws|z=$4>
zFHi5^{fRBRvm80HEV%QtZvthy^x_3mE^_Y@=HL?JLTTCDS@?s{qm1<GGZwUhozLx#
zU7PDMNxJT(C$_&t`R=PFXi8e8J4KKHaS~}I-}?tIi*R<l9O3>PJ>%`wgG{J&FymFW
z6aKW~{1cG8nj;?p4hp0qeLVAe>f|Le-=JIOlx3V-C8BfVxIs^4i|xmEqExqq+l4m*
z?r5iq*H3C_2*HIq8=8dl@;HMF8bRv{xs15Jjs#NO-DuFhHRzRClB2<e{Rj(buh@r4
z(wS=(qqyai-;|9jWiZr%Q;|=12#rFXPKeQUGQbIEzbb-l!BiJAc>AR%eH7&zF1Noa
zyK$Ic>4^;}!znz*B*zenp8*{4;ucrg>0FBNe3x`fPO+jx-#kjGCw&cdsN?iIo~;hu
zA`7_Zc+A-FCxWZ==3*z=q9S{Z2~=YX(){k*HrP8%HC7B$pn3p-qyKz3K@=#64V5lW
z8M5a0{^p;>;AA6e6OIN%IEYNu4k*@^u5*%DbHq2;eipn%73B>eaiQqJ9sA9+mv1jk
zb1^l%@HKMQy3isAL9=OGC`rrC+WAgXIE8gFa_mn*>2P$J^>CeJ>WNh-vgAKh&fy$x
z1aKK0CJ+FXOzQkY6@YOdkG3&lt-8NSDfrtnJT~^iksE<dVaM4@J&1mG9L%;t;y<mD
z4i~ep2oO-~_Js*Q8`vs+p%88g0eS`5I-H6Y5tkj)C-+V@7i;?2O;HSu*wp0Y+D{we
zASMZf4Fs$@DkT#e_Z$4r=pZ~0Y_eA3!Q~zju#7EaU!~FN#R=9a{&M8lGG;1w5D#Bl
z39%!YInEXv!s9SD*%j@`Px`So{Y~~1x^{rQ(A55n5V}l!WR%dr_D+gw`#T8ub!r{d
z222j;{Dq3?_Lby>Cc#tuoKOg2XCQ0y@@>mpN~z%`B(L?{H1vHOINRt0<e_2)Z!`vZ
z49f!&Xf-NLk>|jf2=CSu%(>0Lfz1s0kUJ_tL_|7)I|yM;+OD>FO8v`oq=v3aK0rNW
zD7}*adfnUCHv*Om#3nJVbpAZtfIeUG^*$wI)(tYW4D|J`cliG2vF_~oPCH>8t3N%J
zwXj5rV6PvbkN^OZkt|wq-qJCBP&=N(hH#YQH%;DmIMz@@>xLUkXAE^h38e<j*okfw
z95UDie+ex%)@q4HG&a8DVNp<KXDP_F*b8ou$FjKRTaf@HZ|_@4_9nQW?l#UF(DGh?
z5*wsdP4+N}mP~p1i^bdK9!+cmM24xuYwk~nsYqGra*c1}05gb7PCgE}r%CQ;V6I>X
zaN+@f)&$ew9%^nGO!?J*bk0+FtGx}5y@$ZxMnJt7$BM~v>hkTy*S4pX61`&8rT-`}
zu_;K$QYG%HUmmAHC~HnCM@vurN;)b$nw^f^)?K&huK)-z;mpQ|0(izz!T?e+P6q`#
zhC_{37}-&ppr1nl1C!@(bEUC)@)Vynq2Y8*s9ItSZ~R6wK%X}h-b@hVxQI@)R)%~Q
zprbd=K+sWV)C2RP7qIWVnj_K1-5FmAa~{`-Gd;3n>(RU4HYTW}%S=N(3B7Fe)dWoI
z<;#JCfZ)a910)xlzR&lRfVmxvR_;=f+xwgQEufq5#Qbi`yR=OD%QPGUnyr&Z^`!|w
zFJyaeOn}iPis|)N*G5Xy31<KGLs)1VU5L};_;Jl*%(3Lr_&u)U+FeprMkq_iu&88g
z2GX;PnSdXgJ7Q~2tf(s<{IvC0#K_~(yNaKc1D)O+i>K>;I=iB3qG}NKkzu42skdJD
z=X$(%I(c^i$Nud%WkbJT=>#>^f^(tj;|aa*RG;M<6cT$$G47h#0=8WahY11*>+yki
z1$cUolPbR}yOv1L@3xwD^P%~Qy_Pk~JSS*8y=ziCT*H-ir+EHcHtr8}F^<3!0f2NF
zpaIg>*6vw@=jTbEEw8)13q{W>%=h+G0Dz<df~F7YY{%>Wj7z=j^Hy3+O7EQoKzELQ
zMfZI^Q7)rEYA|stl8IOF9G>k`+E|Lx=LQ?!qX@HcypdgL<$7Tg3<O`MECGgTqXlx!
zn5!cl0vC$Ghz1>PEXE1YWfTULAhB2OLE91-RJ6l9jYjn)#1VMkRq{d`T~ras1&ZP_
zaqvk<y?eZga1XYY*h$|ChJY67M|s<R5rFW%gtK1^I(nYRjW2%!8kwB;MDpDBtHhto
z=Z#m+Mm9^4`<F70hah-~sv@|h^bp(%$eqsfz0(qpUEL2PZesl6f<Qa9L?)}Mie24#
zO#*HKI{~P#0Rbyar=G<%mNMScb%6Ng8!Gg2(~i2{k$Air4Grt<jRVzvfx!8RK6;(F
z<osb6<@FcY-oAQw^tI0N_;H*B$Qk3g<=|Q1u_y#e%<9`-i|s$>L0<D8%%Z%Rp(>CI
zx3A&Lr_?-GXX{K78poqnjp9_qin0WrHHW#1Qk7vrqT;0_;(qEVup@pDcxKxq6;1!X
zjY=;Vki%Qe?Q`kGpabrS7v}~*GUyf4KFR-GcROJ;?K|jvllmZSeL{oN?*r1F`6;^^
zOPqqlK?gFdW7Q2tai%Kdtw^l70$5qJYIZh1fDz_Jkfc>@Fb?Wu<IX>YhJXa9{Llr&
z`N4VU4kX}{EYK0Y_QDWSb~Fby53&6|te&sGysY(TVv&Xuv@~F#s~7IC-7lZys2f2@
zfB`ACuqn>)vs3EafMR4AX!QC;S)I-_lcjB!e;?}kQIXN%MG(A_2=MCNG_{hqD$(1q
zemx=t8Rmk_E(&=446Id9rhM7Uv-~|25djU4t}f1;GDvRu<t}>R7y<lqdB^}68fHXw
zO^BUT-a>i+x#Yh`i-|za2C})1@|i<Rv-fxVk90q>np1ZU;D3a9f3>x7#sk@h=WJ2O
z-R^WlUY#1=gAf&HC0JIEb(UyPDpG5=xzvfTMN3*#(gpD^9SI=80?wyFa2)}@m6sFP
ztgC#M@5suc!0>`PnT|Kv_dU%KcY%06GX+k^+r#N~GOxf?2$p4wJ1axkDH0WtEau@S
zd^1#fTCB97jHho3+GLAgA67Vj5CjW9lEi79E?iNHheb5ChWaq)S>=zOC;U>}9Dk;*
zxC#buy_XnQ4R0zHlXNeePXN4t$)dg%s+5P7?z+rPhJ6o_KI(|kZp1c=x#F_Cc57-s
z<10eLxxb~F`=OgxU;e{q0Z83)x0~_89>gq&9!ny@aX%T<AjS%){3_^IZq4i`m=;aw
zAQiI=lgMz)#KYl86%^0qM%-HxfeN#9XW}3fzO6o<zjIKx!zcuPw3mi?rEarYyPF+&
z)I{ecOi5g<2@D+k0J>FP0hbC9&kvP`Toe+#5sF`65~F(W(x<LM{+SUQSvDdbM7Vl%
zx@iHd&M`Pw<a4TPNva<k(BahEs}vMWzMf(CA?>UAcw*Y+o?8$SKVHVTY^0XVJJjTH
zw_X9vt(<uO;6u2`71%f>9C1;DY(--qqVcW3rUt9`p?`~!lagA(V`@M0Qm*F=0e+a6
z!wCY)|2v?Ty?am|TkW}Vj)i3DA^0SJ#fU><FdlRe7^X#tAVB$~kW}*l#NY6rR>*rs
zpPB#p0AWZXIB}PiB>*W(W%HVRrA}jvx8DS@l&UEULElG*>mGclJjAqOtF#IN78;bw
z8n_Y0I5f^3g{~e@kWhU*rLLQ=1Qdq!B|ox=1N+BgOnPO=EdRORDLf%hk7>eUS0z;9
zi0J&uH*Q-O<1z$_*c@~lu6S<4QupScMRoSG!s$HGsFj~sv}olR;@Tdjl|?uXx)4iq
z(tD#QyNf@EQZSg#*VcSM<F+Ll0=HS}y{e4@gZ!MJ?1J#Nza#2jgiVb5)cJEJd^5o3
z0IsYTdS#ZU6TG^O;{GhrMt6EfI<~_MCj`;Mf%ZnCTwY;+G?o#!ShYg?RBq2{Qgx2f
zc|$6=CCFT+(5~#@Jw9-oMRpIiJJ8ZVC8*N>>qV3$`gg%G%`f$xu;TaJj672rL8Lz_
zo0YtXAylB)H$8HnOrTR*IFLt*duRL_s$_b^`KRz0BBgh~-J`D4doKJ#HuuCv_UM*Z
zIX!w(KVuS$p1%)zctbbYkI1CZEj5)(f!ic|XwC;OX_-BZg0ZcGAj{Qa-i?puy!JnO
zv`_l%`F=G>(R_5{vY9^`I=-HU2no;7(yrjhfh3Mvtt9f`IIM8TJueNDTFr#`G-KA|
zYtXgn&0NVJy)4@a2IUdln<}$m1YDi(p+)wj`D`8CYVh?VE*ITZ=JozfdvuAd&2Zg@
zQb;w6(G525_j1I0hD<@g=tX>GdNj7ZHJ?*F*5uB(-WBcSM}7#JfN>YM9Y$<%>Oj}F
zeUMk__(qU1!fOiT6<yx(%pQnlzGDFFQ-R%b6A>E15A#bVWR?RB$}3V3rQO)5BbIAS
zC%gV?S_i)~%Og&rDUOSe8Xfqyo#r>4O1f?%EOtOgcmdO+4bBvGt>=8%SJ2H@xz9dT
z*6z{!xpu{#fzN9<JYbacXQB0V^z@dpWz1>Ueax={ndy8fRP&!-C9MBgyZu}WF-5yR
z{7!XdR@GPU`SyAj_FecYVp!Ou`i84D!R)Rma5b@YoWZ=(s`w4MYt%F}0pQ04tuswR
zzdxBiK?jJlv8McWPq^X(?XDDSQJO?Jq$YN}P$Q=B0m+M?&_R7#jyTz203lX%CCN)6
z%umMmr=ivG|Mt$L;%W)8<g(S#K0R9GKN=#=F8<B@Lx}BWZrf`=u|7~s1VLGhL~#)c
z5KehB|5Ft1VL|urK-wgdRQ#VMq38*H^;jEH#c8;9H@t*IeQCvtV79I=kst&c2R`QI
zFm0?+?FUA~#;5PmFom%9#>2(5?qV~)?y6PnFicO^Isu6W+)KghrwmL?;XqJI=I909
z@xXEgjtTK`pK3z;Ku-Z2b7y8|CM^~h7lC#J`18t6C>>*{8=%&rcmRA=TiEoi4wlNV
z+lX(?o0Ma)WMj2HVd^d?U|r%vhQDOyGgZV@4z#`4#0=lU#i1m3b*M*eMo{)VmhWX~
zPckCbtOKNYN$LL?b>$E3VsH2+=ugK~El)VnYMJ^C&OvW<7D=hd_Ln56=*6grHKik;
zdT0W&q=wSu>>~^JRnTcDpR>^7DMn95Eiv+B0!g01)7zA7^Fm1I`@%j7a%#Bg61$HR
z734sSfJGwU_WADNlC9;aPC;U;;*U@*&uO^xsux7t&+n`%uy#h^H0XHl*V|=Kfduuo
zVBm-WO+*L;QUm0rB)k>;eZY%((TmCk)nkCnx8^|$PS+NrNlXu*6)&66544>>K~;Ac
zG@W1U0idPd;IN(z4k4QJbeZ@_IguSBm}>ktCPBMHM2Yg7m8X>SImaI!c<ykAuo%ar
zc(t*7*x=xHoZv2RX8W3*Ut8JfOn!5AlFeZjF=0#GBO0y~3B;?4+h3Qh4z9-f^%-MM
z3sE--K*T+B4a)n6qQ8VyV8u#uKQsN?(e}CzlqwXT4LHXRgo9?=DfjImdQa&v$X^KC
z9ZmCL*xX?e(<X3&!9xO$j%&TA-hvAO6>rQn?XWYegAygKFc>&A#wHJ!u<-4K4!B$2
z?)=DUN0rOZe=VJ-8Hs8+-=tgQ7{MB$)7M@6LekeYe*`@njU0ivzc7MX&VJVAlFEvT
z+b=2uP~v{Lnl}XfQ~(KuFc0al5WFZAUs|fH%`3g2z7E7nUcj6r`22AB5`B2F_5u;&
zOHmah)_Y$hUMsE6z><S8P8%dz&RB+%e;{7}OtNEBtjc-ln|l{;ifrhE0J`i^yG)qT
z$Vo>G&l#b}NxL*7IK8J*32?&dZ=CZ7>u4U_=8og<c0yer&Q|#~csx#b*_f&N)}geu
z;;>A=F4YdYz&}0c%QOsins9((UG!OdoGCK6gz-7k_OMd;7Y@rNh>A(Xi(a&clXItp
zRY?TGJBm~3(sImMhf&jiP3RV;QAu}x;FC-^BRoMLR1f+=GR9-7yF@+KLo@(R{trBS
zmfdHb0T=X|ajAP$Sd+|^pM(rVKHr5lC-U?|aNT#fX1Y9aYEy-su4Su%gGHgt4jwd8
z!@|QiYKJKL4zIwKETnIvv<@*KpAmdXse_Dw{nzDiF1oi@9Oy$p%O0s1-T`=f{D8#=
zC|h52cA&kQ$Dar~8k`^+8-xX9t0akkx^pTwSKCQ6a>d>BwBY=EJ8x0B!1V>gkIh1$
zjFv48KNwC;Mf6j59^E~9CC4W6byFOAg<*z-*1DRS8qEmn8I3{@qKF?0OwvL+brD4?
z)l9nsdGDb}VP&Nr)0mueACpwPi6yj-Uz?*3o+sQ>Kxdw#xTR%`p8TGdqTFv2+wtz)
z>ry77{IV(roHELv6GgW`aA!4%tq}OGMQi+fWV|sQytamIpj^+FluR-PsY~GkuL(Gf
zSX`Se`+of^vXEgTH_&Gr^VsiUuuS*C-T25sr2&URUm3Q4(fA!Zf+#kj*z0$&^t#jR
z9h)jRrUCaa$TU(w<@_8<=dpRgP+xSdF&;mF^Ugn?QP21BYI}P-*GeZiDp(TOHH~Y7
z3R$5Ze0=HR2f<nQ0u(H0$2ea~u|D11ZLg~bjmlP8S~NANDHt_WXmB|sDK%Po2InPe
zr3&~j#Nwg+&jQ(8^ggZlvw5#UB|xwVEi9IIA0`E%t3#Le8i%sow%sa=b^aKk6`#_K
zQ8nZ~$D5gQqwH|+E>ISJpG2b8uwLf*=X>+I={WK`o);9V@@IC4U0gh#zs(lSXo@oR
zm+u?P$-Sp^4i`l7&kvLU>I;&>c0Gp;x*-d3p4Fg{q*&DXQYw&FOjJwAQgC}1iu6W^
z%HRE`Ib>X>*DmLLq$KCXBlHW@=wH8EA1c$|X+e`qlj1ne%rQyZ6L&cwmWRBdn5!5y
zo*iYu>LF;`b1f|pz{z@s1207oM|#oE`2Ajfky{TB50^Ssl$CV=tFx)O`32befQ=7G
zSJ4_D(SR#>@A4Pmn)CDX)80DMqm%L$78V9cM`Oray^08!wMKuu*IO7w0f6y_P>9;~
z4PP#D((oIA>JtpO)eQe2(oJ1>l}+P&L2FsX2}um-oF3rjEUe+|lM0^?p+Sa8r{bYw
zflw{9JHUc$x`u3<gtZR$<7@js+c!={O^qK3^Vx`$Mt-<s`{2K2AzQ?W(P^Qi?pdlS
zAK~KdVGO`Lt)Rgj6TI5DNkC@eC;#^;83NJHP1$+x%D=Kqpu%}51ko1f9I-sXk+<lU
zS!GDUa{oh%_Tx&GcTCRBoJl;O;a_d0#Y#ZaIBtShDn9y9+gCyFOCGPsL)$^2JIKwd
zKt!*yJyq*1-)MzZkWK5t<6UW0v86EIs3U*e?5p}TK}Q}R-%t1rRwl0jjbD4&ok|H@
zZquO048+Pn?9J0)yijL``Z~3IwAQ{kLn@L08^oGkECitA-MSSc@Slef$d*(!lS3(e
zh=4Tj#Z-HHz6H#l7?^@*P#YW1RfN5ZF}BfEf0~rTk3N$ES`0+2*)>(`vBCQnM264J
zD+bcyLK;85w;n;6uUa2(2%lzRCF~zcCjvZx3Sn<hKl@2rJ8HW-KJ^%mTw`x|_*?l=
zHgF^F$MlW{Y00ght1m?_Q%~+SeT6fVK>;>0tLl1VIy}9?H%UvkKt)43GlxPQhfNh&
z!I=9`?AuX?*P?EKt_S`lwL+;`FkA#@4euW*9nL7rBeVI5-zk=l2pk^fPhtj8+Y#$N
zQYi~<456;~&1c}$|MJK}MWKJeX*v_=eI&)gv4lwqlQca%pXK~kW;7e2_#ljLBDI^k
zUJdHi1jkDZ?+@Hc3!Hh?DusSAD5hbdSPu;hID!|>`?tJ6f(?qBS2aH~xUG6UPdnfL
zsRujfn#WeOy2F`vlx>}{!`gG&)(@si_7buc9~@}PU*!hIMr)}aLCEdz2ilA&AtD3*
z-mD-DwMBpaBfBwck*nE+mvn!vo?=`FLUAq#FomU3(m82Ft`rQUr`e^>Z1Q$8B0N~6
z$sj7>ACcdFMA9SSLjpu$B9idr(Hohbc5t~<16!!sD7`M;>m{0^jpk4(AZZ$f$$$5d
zO)RPT#8tU8gW+B2(t>c<8?MqoivG#-`@)k`SNEasj6A=G{zg`)D)%w|!)MQc9$zWI
zZQ}f#pc`8JR&CtgIOUeCQOo&sS{>v1bgyAU{So*|f#XP)F(yyF{sqq2$f2YeF-inw
z3&z`H9$WGdxOOzHv&*E$ht5xI5i|u>ptCfv^0to4RXvt<=t=qCwT-BKQy`GOL0QhQ
z2Sg&Shg<2A1y8RJ=G1r6@TnMAe#R=VnBPo1N}EyR|8XA@*;X}>`K4yCcdF(UN2;*=
zIpx+buh$^o4<01Cg%1Zvplz3{|M=Qa<V{!*(yLVmue9vIlOfYey@!&N9U8$M#GL24
zARVm`o9Ux1C#ndV={i?CUWd`4(|KmSx?HnE2pU4CT;jx{a^qVpVBrQX>ixy~lBz*n
ziF3lEZwpWHfv0_SqL-~mcqapTCB*6FVfu;LCR=U1MB|QRVQNXKpBG!GZwkp|6)U0z
zq+LEbNOQ=lG~1i1Y%$;2la2ZIrZ%EWg}@Y1ysG5;oysy?dYdGhk`eOBpenHq(MCQV
z57zO$u#!iK%^Ao*zO#?v=vq^?`vBwteWyNT_1{=eqr4OUR_T}jut-&!^R@j7j|!h$
zv-KY4WReM9bM`|g=K@r@*`bchakksa(Bb}!Y1x235dDUNvRke7Ec1(RZ*vRe(vm@B
z&k2+$wT75n-@zy>eIEX}fu#KT(>((fd;eT-CeBO+^S@&{GUq46^HW?s0y0}Y7ZxV0
z#tNwUl+RRh9}e9iDhbgJ&p<VnVf<~~KO~yD#5vM078cVInK4EEOpX^X;HwKKvq3K-
zN1@Y)TkQmnl7{}wsf$uxART-uCBGQW5YS2b0Vbekrt{X%`ItMx#}|HFQ!QSJezF6`
zX3H%oi|@8{?=7Bo%KnlND<1d!76tz^9r#2QhTYlVj0XhOgkHjijO6p?0$-rC_Wl}!
zR-DFjJJXYiAQk(pwdhJ5A%`Tx*2)<Jov)8<P>+Q7`g%p9UUZzPMt9CW5ozUwg1}}+
zS4H$u&L`G2p+iv^6E75NZRqr|XxqjYk_NzDGf?0wfdd$NNMQKcYd#QKz#K2Z$vYP3
zNToq@gUIDt!6a(acg@u<|JITM10H}T+8i9KUfyOhv!<734%GL7q^gmq@rq(HOka=t
z(ZBc&XE#%*GVfaM$4KmT%Qn7wdC~w0&O9Il;?JNkGS!Zg4HM*-fJ6Av&6%-%FK7>B
zSZ=MzihRA1bf6X~WFA}r`Ebn{bWmb>a9O%jG)y0kksba)ib-3Ii^jTS@+Ra}*pCc_
zwI6?m=RScHd!GxHInB%R;{Tj*1qE6OE|nIxv>2O&mvnA$;;X>`=W)thD@m93>|qK%
zYErZoEidK59?6_D_T9?F^R~`3MTCFomXc~}3&@6t;75H?qk>rNiw_qVE!*BD+f0+q
zB=iOj#EHXl_bwUwgCtGq(7}GDf<`3lTb0|rn|sw+qo;NtAYnkcFUzR-p;r9V5#Q0_
zbmYeM_|rl1uqhdKP+T3`2ea;OIHG0(os03`PEmwg1E;kWg;aN^;)mri5CRKuKh=Rq
zuoI3l3TEy-#b`w&&royys`I70d=!}U5{3dnVb#5^9Pq43Uf!SW9G61I4dl$;C&Wb|
z8nOhj`I-EBALLFF03=6iS6?nRCoFumn5^V}6}l%fGP}g55xvVwD#blm-EeyQc`N!n
zit-r&n3i_{0R`c_SWrqcb^fpe`4&;gBv<wH-TiAHh^c>udtgQ<9l*}{w27fEv*)79
ziqU`U#sz%#K-xcXkoJ^0aEN~PxZQZPrXTdHR7G!l<KAH|A&yA8a;ax22;llEoSrc0
zL*bR$l*VkI;TE;-^1ii>G3xTdHyAbY`I#b(#sT~zn#}>Rtl*7?B}1GFHjqBba>Dy|
zgrD8Vy;_^})^2CizlBHcGmzj-2Oqt+bLVucoiW@9#eM+SbC**0D~~gxPj>{3L^Fkz
zp5s;gtM%Z6p?V8^ozK*}4Q_b|8lE6<a5amW#(Ruu@a5d~qy?0yCR9~^L)c297CZ#N
z1!6%*->*njU}ag08orNG@`@~=)opsKy*LpfQk@BOp~Q;pqBQ)Yqm6uxMXFo4huZlC
z&TXXxh*K5XTp$M(E2Od)G(7Q|mZ1bKal}s}?@+_w#ALoo{6&TdXsEZ8D^Uk1#o2Am
zEK(k$axLdslVn|w86!`>r90Z+Q+>tICmseD;7m*hIBn#&u@VSWFg_3P#7H@ZabKcy
z;gSw2u;`K?Q5;N~!}XMsrAs{?9dIAXtHe3ehR0Yf)>g8Rl=@7Quwg)%HZ-NuM)MKt
zDmngsfRW5sS<;fZ_%;Nx_Jpl)S)uso$OWh4yp#J5uETMAFZul)Z?~x`Lbvj2)bh@d
zDevU-gY@4&>v8{~V=I{Fdln$59$#wU4!3-)MH~6NCIPgmKX!lTm^s#Fy!AcF^$Gld
zCl#BwKO+dXEiz`deCLB2ooL1g7ZtiD<BS1&O!?5%>VL%J+7vp(R5nS~U*(T}*D_*<
z=UM^DAj?VREAjL2KI~k0q|Z>)Sm}OHX=<cK4zAQy=v?PVODMZ2S+C%aSYIPyVBUVR
z%U1uU@Udv|kz}_a1sJKq{9?~0B*Koo1)%6KfcaRgM@$Yz2_FxiT!WqQwNw2^Z(ak<
zJgpH!qCIbD`I_d|;q{*NpN135@bgm*f9QxR#Mm0moH6Rn(}>)>m*pK>L6c&-*nV*5
zZyqD3u!G-MU)v(NFM_TKq*x-`J4O#%M4c<j@c#)_lk1l80)q->0)LhmHy&U}mh9%;
zNQy+t$|4n*?6v*Ks81B7YH3VZ<IWyb3UJs<c!;Gefs#v*z!V+d>rd_c=Hvx~fkKI3
z@$Eq<5Ic~SK)upsGqf|pIF{z*fDlwUswz?Aqorv)*Tq1x81t>cKN&PJs+XQ7M-?xy
zkEs3nBv+<UL|(pa!q?(>YddUdSr)QrP4)II3X=V~*V70JIO|i5a1xBjv)L$<-0v08
z<c$OnjVPB(kDpUhYZ84_F?DhtJ-?foTow{F@Fz_b!1~gjYKC0RC+gNjTZrh!44O=j
zvZL`=eN_A0hsDyh=Ce-`@~+{2h6Qd}vMdCTKQ&6s^PkPLrDSpRytKIvATmJzm$vXh
z030Yy>ftHheI&0gc6r)^i$E3kya8{R{2)YfaA5ox<>2Qzv=o;CBx^YFVOlBgHv}+{
zKi%`#DzzuN+?Nc`F+Ko(-ppY@qa3+W`W=l*X0R<rIi5*8nlzZQ5afBU7F5k*8Pp@`
z5h@1c2DoeI%s*V0b5rntk1d8!(-WK6>g$m{_n-H$y0{h9^t&OxeoNxn5)2KSq0;`2
zy{q1K!~iT#P^Moj<D7Oi)p$1=tbv9IgvUxg)Lj40=EOh|U)Kd%la_Z*>lYDonCG-D
z%rw0*4&g@Xk*uW{qJ>6StO$?44M03N_q>7Ni#CH581%Ag<_O6hy;AM1ws({2ee8I}
zkq-D|jebihx!@yakwl(|zQZk|vsW~ikS#qk1tE+jcf~yKp`^Ypi^}!*C3jzmkrm8I
zJq%k`1gurx?H=el@gHWf8{;c}Y_Aq{2bKy^fkL5Z|0b2mkRw@z&Ts&QrfxYJNF;vn
zRew)yL<WoFxh~fmM+g)(Vyv&Ieh&{|IjTtfN8BMCJKALi8b?Cx7s@5C+F1Yc8a8b*
zTrdd9=9-E`uqLD_P!@UJ;d#M-<z_ca&}a(A;qjjlE|5LMKB{1ZdqcJ+uPew&V$gdo
zr-fFANXHzWCZ_zWz@F>+gw^DxiUtC)o)6DG^$W=}5qilFujZx6qgav}Ovo=^-PRbz
z!Tvo^&q+!u^}XWa7yZ@np6S-yRw^H=h4)9^^`nT1%~Jd5xmJ06-h5-8`#J&~40EHk
z%)u2cKMx6*mcRvsYCexvxWw&|?CE%IE(%35)&GVnwuQBA$L~MdtW}TN;f||^&>ah^
zW#5wC?6&3-0oGNrc1WfCWWJ?&qQrJU833tb31+?R>i~u?ANgWObwXYqr{uRDxD`9k
zF>4(lAcA|k8H>%h@UuM=<UFCMhk?08k8}@wfqM4_GFpXACBQ;`He8he9<u~!7z|AK
zbs-o_S$br2<P63^0>(kD7EYH;h*)Qvn-+_ZhY&NuCaiAln0@bk488g9N20Ci8klCx
z(;rn?@TJgVk0G@|p_H~9x?Q#?{+lqS^^!KHm^0Uvz1WK>^v&IpI`s%8+ZF|_@=Usd
zOlkD`8*npqYNIKNgL4gpGNYVv5T+zmFcxU~7r87CO_0&S&x{YXwv_Y*0b1v-wiVF?
zm1XZKk5u*mTATo4jlIBI0B9^@HcA<fq4HPJsjM3dx`t7mR|BP0IFLm@Tjai+Zc@W*
za|kWf$Z&*F?1o_R<8J>B1_RS4VLPGqjr(Ie0A;1AFROH!)7Nn<+^LjOkU5J$%5Jj2
zt(58Z57W%%V;&2e91Qju6|o)CY-l0edr}zoS-bZmgO~g>=F^!XiyBE$6~l#h&&7#*
zs6p$2HhpND2qM@dHuc0-zb=q&K5%FSSp_Im!acuEpyP9$?u~yJZ$c<8LAD>{0fihy
z@P_UZ9QEP;UODop?VcsnmJlVrsA<p;zKV*PEH?#An*$~*<LDW^@B9W&uw6@qa}Txd
z*_}MS`5|nGv5C=CfgK|^XaK>}tGtJ`K(7efxsSp)b}7osSz&IW+Qe!>Ul!#*TGGLS
z{k4zLx&aEp=-B+zy9q?T--~SxG@wT_$z0Z9239fjZIK!K7V6it6N&KsFWz+Ra}}@X
z`?8`SW3r_X_$I8X9U8o|)Ab0?y#$l>*v-dVRJZr?5EWP>D<LGNqhXvY2C&Q1gDkov
z6_EmtF#4i3ZO0Z58PAZm%pMD%Uo~jy6V+-j-t2!1OgAd2e&6nS=;fuoEMe82Ky!Aa
z7ho~(1n?NLOOHhou^(n8AO~U}G%Y6mL08X7v8I5EDlgtj5EF4DNcU(V<4PhtN{W>U
zt@s?_duGDB+)tGL2`~`#n#M}eUMUE{qJrYj!!QaD5#X@stG9Vu#{;9#c^QX>*qp^{
zyU>HVZbyR^7jHJnUN_2_#P81oxogfHQ7byv%xfh#?e-O_LU(?pB?<$f<w#wXILfVM
z@Q#g@Jj!K=Lnp{Qw4hn01+55Isq3#X6TYzC@M_T?Dx}-4WhLUxyoB*{_(YHSu|d{j
z)7Xn){NsQ_4}wgB)59qSwtjWV$+UuO#|}D;fL^%DD~CR$Zc+(JviQ!^UX6C9@MiCz
zRxw===B#NvQrfhmM10lX_?x*{^xF{H{Rba(-{3#z)#@wfa*%;A9D>NsS<#B<HZ3QF
z9Peu;zO^?nIm$tsxHu#}Bhs~*Br6Eha06UxlFlh)=lm<Wgy8PeoREZ)K$Snnj&~of
zFinXHSkCK+64|HjuR)EZihm4qYHk({wtaXX!dSn689MZT@nw0YJ1vBajqPlg6GF$$
z^N&BfzANvF@7w|ZGymptE6Z&g)%Uf3UMyEjls9DQ^Wi{$;n6(KFhKsU&<h)b6-DkU
zkJ|E=U<f^-V*e65V-m5?+D=eoTb!uX>%p{bBeV7%**`g_n=GSy)xg!j!!1KC8c*1D
zXD{F$8v|ZbAu6ySjwYprmXwY|EO25)ul$sL@6b+8jk;o8(<MPw@S9d`OdJY0l=L{F
zFRrG;_XYok&13&~Ezs`|_fTp!+g`N4cFhO&x5%?EVhE?%N=UOZQ+7F@#5WCq+Z%1w
zKNBIHN4Ze@QoXs`rvr({&v})Sr~qft!e-$063w#lpM^tuNWoRh9#a};O8)wqWI_+K
znC~*bTcJqsdbarEm-xT?+L)Ij5v!1bq6DYPw?+6MPZZ*+?pw%MPazEe$dtU6$*~im
zB1uwmQ(rFuU_oLhLStd8tMjqJ6M6cJ+L=h_^ZC&(Mc3gw!y?D;g5W(199UiEfDJJi
zI&GN_44Z-~B3$)z3kyFOad>APE|x0+vS%eB<V>4lnvS)=DUJkC56Lw8=N#;o8Z)sN
zb|O0A1XJ|(rDu361Y!noDA&eS%P2O96nIY;EOa7*5+Sw3&a^q-zx7iiCo8S*eafQq
zHy@9EJVQw`RquKbfJmIN8bURW&i|@dv(87*p9q)b)=vLhoYE;Yb=y73?KnKv9JFYg
zzkOo?*bIR>;s+ng$|R1@f0GNZj{^^ei?Y1&<=|g2^725`#vyCXLM`5Oh|M*Sa$=i4
zdl|uQ&@L%#r&j|IDICg;v7)u?3cfO8B3v2vh%F{_KH06PpD8;v@jF8nMVS)*N%CRs
zC=d^OY4X*ZXfmeL!41>Hg6~VN*A8ZBT|dG7PD@gY%I{_E;)|OAp_jO7lWb10maGk#
zBtDhK75iR%&U*?RR>3J={bL<Bou;aR2($Y*2(nV(pc1a7+E4Kfhw^|$Rm5!xJamaQ
z`k?hloOszyR!XB=bF@a(l)QfWkB7H!z;Er`%nk$Lp6yPg0-c}Y3T#wuCY~M$)}O^n
z#k<?UbB8vRFL&$A6_`pdA^B5R^@w4f3~0)Wa|Pn+)23P}f3J0wC9vniQNbilcvE6o
zMj#soYYagHTz3ltR3UbVVCqBCOaZZxZ@SB1gf!~Sh#e0+G8I!T!Iy5FkFj<hE6^-%
zsNPHR4UG2$c?d|SF6h@<gfO{`IPVRnBlt(+U!Wjr>dcc)l3JH5n(No=;tYpkc?Q@}
zZ~5QOq^`_ez_3KO1G_-riw$3h<d~E1X(bD|S%G4t{(UPXi+2y&2sHAtJbepYo-8b@
z)NEOQGUPA2=AE7{>-N^sd7gVbfc4GQzu&fg#XbA$ZSp?GHj5Mzy6;nVS>qFg*p!%#
zg`$W(qwUG;h_4C5TUV2ZgeRuF5Z#q~GqfYXIC8Mj70lI206q%f-q!di>!)AuyK}P+
z=kpPTpCL6?+c(+BS&{B_oWf#)!#AGISG~+|gq2#f-v4;e#zi)KpSR+N2AzK9zWyt_
z5Fp^pAny0Ji24^Sr6_+Pb+Ro+U*^i@?|EsPqH1MCYi;2=q1_-7MePIk(y&IjU>wZ2
zWbM&C5b>}19t`Ol#n%P{Bfw?wEX?`O{tTWAqEiZu9A`>M)+d(qfWOKkj(OC&G~Zl@
zN=%hKbMjYR1C0IsSmgd5>F2CHl91iBCe;wVTxxb*Ddu)2o4u)*yFE#BuBh?ew#?Eg
zz;rVagceaZdbN72+5?{pk>~Olv*xgc+1`ck^9_=-`NLo&TW5ck`-;C_N9u(umuNyw
zqadY;?W(V9&O3ZC!D6{PxHl&o2<D+BOld%^G@^6%J8tH;i_w=+6ey^fY0Pjmu+Y(F
z3O7^VY^-G=>flLJGa(xW)xtYK#Qj<C7J0!aiHV6jK(`AD^b0wxXBa{L_N6fZ)WjuV
z&Z(aJs_XDYtL`Yy>>-BNS5ijZ2Xyf(@&p`aYA^dA=S;SiyxGxTGdNW9Fli(}(vu}6
z5R)Mf{_I2-Q@k&8>L6@tt8Vcfh(-ZnQxx(Qhve*Fqc_J!RVC4}kA`AiB`0xobczB8
z3g(D^au2;mI>7-pr~4X=u+KjVlH>}L+ns33L_NPA-2gn@)L~(FjV*lvkf|7|_XNbH
z?_j$-E2try@0>^_^fWuXTB|^Zh;Sk*tj-N(A0dGtevyWO97+|Irl!ECD%i;bNzVVy
za-&BM1T5N`_XNO0nVOo`fRS7F-H&cBg32Q|0T6!L0Hf2kUZ#nFzwt6W(_-$7E6O$!
z^5<GPyz}6#<nK7qW;i%OyyLZ=rtw8aHI+oUgztOEH9iurq0;M#PS<!$yPbLNvyTV6
zEIi|Y(;sF)v!s`wz-MHQL7?um2>Qso4i|Xe<4n}TRh=yl{}D*t1~4_!Fsdh*S&rpg
zhmP>_Qg>X5LJEjAH9zlM@e#c}GPXxC93nq9wQg-ryqh)IR_sU}jh37s1BMOL=<cK+
zg!gLPSNNGW=2(m?k+uo(Xq*Y!(yXtHpGgCOe%NPf&uU2!I#I6Y6q6r-*99L7Zsj*)
z!HOgUDBe`N7z}|089=Uv{Fhm?$8MLq84ec!e+MyWK)n}(=^QT!LC}y-0h%sy#|h`+
z&;SVT+zDsWnQV|DESUHKAClI#aL_Cf4;M<WiVTTT^UF8PrZ>^^yzZRI0TnRX2Yboy
zN_Q7lK)Z$g6#seW=ep9`{LKd23St?euPXg#!|1`x1;72*sZ&#2$(Lg}1I74xay`M7
z0hp=g49$<ulKOw4wC``Yz~n0f?;SG=l9g=wTAT@}(u62mn47Ykk<||WB8B=7ZEpe-
zjBoyQBqCyy1a}K6+>H{8<;et+B?;dRS-)2^1znHfduwj%&l_@C%$-&+FWX|<H&qCI
z@N~Jd_v@~PFP@;Wp2=+egMdJk>XQoSXiOKsxXEuHeE`YDk<M)uFn>rVC`g9HPZT?(
z{^u4n&M$fpL2Z=;3>EG-<>$LNQj&|7Yq5fPi3sSV#8T4--{!HIC@bbklz`6=ael8E
zgzzi;7x%@FV#_7<Qv5vJKXKW(3HckVoiO8j(!9&!IqglXlB{eP(aC7J$@ppst)X;s
zTMGogNE5duk}H$Bx>PvktOi%QpUi*kV}HoU`O4$@I{I?2?zFE}b~m>mm2Ox|J?(zy
z9m?1>Rgkor8qxf^lR+u%2Ed9P_1Fl~=Y#o7%yyDC1M5Fol1Xs-b+f}7UK3PA$LIw*
zL4f2}J+-aD#t^aFP^gmMil6^Eg=FVtnlOUIB&2*cVwR*L#eSuQ!}I2(=v@{+i%tWQ
zADAtMMfm=$T|Ef1M3M4gfMI)J^xN6}W62H7J1$Ug_n#sIy0)-=CsqJjV>6ES6QvTU
zO&_n(Mzgnvk7qb>e2s26RY5$YJXOQ_ZDi&Gk3vFXB&R6Bs|!|x_h-@BJp=QYJ(9mI
zC%Z{}hUVfOW5_O*KMf3$lEf@r1dRzdv+ZI{sM2BTRHHrD3F%xw2s+OBz|{~|YM6)!
z-+mMvAcv#GW6=wKAu?{8rgA-3qg8bZqvTm_?+I{Zpi4@V`{E1vjqR9>tP`7>CFA5^
zN$*vClO#987_scZ%8HuQB6q=04^a{097R(7l}N#qHb+JfLrs4dWc!any1;2g+Yx-9
zf=RYSLHu{9KOA>tcsLh)T|iKr158s^q{jtDNidGi;3Xjj21uOEs4LgH9%@ne+-}G_
zgJFQiCV`p*1d%&6FfGq$;wl&Op#Uz#NpHt95r@U8@`VLcEEt<g<t6CkUJlrno@>v|
z?rR^_G_)yw95h`sGr4|XA;gg-+lngs(d@R^Oi=C4&Put@@>Eon+kCiEC)DSlVHC?}
zOi0{*)3;6brMSdo_VH=|YX~+P9AZl4&F8-#+sYlX+E(>TPKEX!CWN7foIGt0NB7M+
z;>0Tni<&432MzpJ@u~QiLS_Or-g!d#4jAAac3Dg1>3p0V-zF@|!$vv|9-XXyn2CC0
zY;vTbrJ=6BQ`9C;4cv(V)L!kk5@euGFYL%~)$A>LT+yA!r;%%(BVJ<*k^d&rQCs$b
zW#(wDH%7VJ!f`|nFU8~6%p}{js>M!7`&ICi^%lkvvNB`&pDj#Zqt*2MPaB)rt2_@+
zA^swZ<Z7iPj(cmdY`)}A5#n&U19Zlx9-P1Wd}X$l*gmm-jo^kvtV%z8#x$0*^qs6#
zRD5R|LyDM?nfU`iQGp;RBP~5uYa$cSL!~BP)qcVcCPN1szQ_e$eq&&GWg*xAMc%F<
z9|}S_tT8i}yz4#tjE&*};u#NFx|TP??u;T%REV^@Go+R)>={O8*u7=BG>eu3sb(=5
z$cz|KMLxVok0}Z=aA68zkPqr%mREA|j9R4eu&+hGP5SZctf)?^>}Tr*i;NDvukS8!
zo>=sgB!PC(aFfbZx{1aOUq45mI4N*+6Cm$f4l|A|e6-LVW?RFL;bDxRYWr)mi$72^
z9Un~}K=pAglf1BSV5&ST;{ubi_x%~H4#14}ghWl(IV!6fOG~rINyln*8fu1CS_=|{
z+ROi}ZIu-1WwuI;{OdDb$=IXSz?a#}`!L-G6%b~yhsE=o21fI4K>+Rhug}BPUVC8W
zSis6nRqFD!x!RBmzKkHleDfNF1M4UF%PW8Izdk?uN71*1?jXiQ0PDN>&WldBH=;!(
zw6l+j^Ykv^;H&bl+edD}Ia%JNAzVCO?mw{T7YZ<5H@hOlM2#fGXH7BWNWn&dsAki?
z`?Ht!1gzGQ`82`e4UDw*BLy9`O>w2K-SbNc&{iqG5aUQ#?th58=V`EBw0C|pqt$LI
zPO{(ou5t^)jh|$WiG?KVX2(inb>hu!{~;=0EccT@*HmSp-VGZI@^MYEyF@KN^6eF@
z2_=qKm*EMTrQC_A?S&GeVATRm*pcrOTi06uw&Y%mBh~Y(nYk|?+S|hZePJW#r=Ki$
z2YD?m%i8f}RnD1Z36flX+M0edY}-;eiD#+g_qy$C5W^?3E^>uWnytu0yg_wrx2zhd
zN^{&WJZ7>|Z%dCna;pM<ZB0+k`-=LsM(CeUt=`~=cL`FoZrXO}f8lbqC-(6qiI_n~
zLA6p^Pt3&eKbFZ4_4k{o(x0tN=tj#^E!g2E6l3;VduWLm&yq1ZWoF?$V~mHZct=CR
z%;>vgTxdKnS4C+BjLYWj6NSZ6>NU*-O6rrYa>i<K<opj!FcI|h&!}k_h3Jm6o}X`y
zXSa{5z~Hwjy3bL<C8!oz(>n}?Cbug{k)6r5(vq>iBr)EkhG^C!*)dr4N43N(-LXyA
zY|_RdV0#b7xr`(-#t>0y)D(Y%F%yB8WED<K(p{6uIETn-x*lB*ny8AQplkk<o8~JB
zOIdc~QsaHa%e>$mO0B>*S^Z?!);6Y!4~sAN-S4&^jq}~Xu$7_EF#pfbw=*6%UKzwB
z^_fwm>ypv-tPa(TptDh_9upj(-!^<TZu0HXydwnNYQNz$?Z@{XCXYEA+nVT)^(JIu
zWDxjfu9iRO`LOuhcG~g-4TvKxhw87}%P^h0m)=Q9(K?=ENM|L3snL1xn@Wm`82v6(
zcEiQdW?&fx=U8;}4ik11Zk#TDU-u6#>5`Z8&8l);TBSF=mUfiH5jb$m{}+pxj*8@v
zEx+ye2afsD(T&vK;wN*Num_{wAAN!Q982t4N&T}ok$cC+%X26!1j*_l4F$^GKh6Sr
z7mv9!SEZd)4LGK|nez*CNe0yA-Ji+o9$)?P?YL0k!IQz8`%7^f$*X=fLP`Bw3ZDuK
z4mNk7AN%tK{sKSd)^_A;h{c?Ohc{MYyzIs7^Y=N&Gic|6G8BfK%#o90v)dx{JjR76
zY5seHsb9@ew~U|AKWCOExqSI6OU*mi8{rds@%ZmrmJY{Xtn*3(>yy@2B2pwkQfa?P
zpc50z>A6o+twalFV)NFACWeow_JWlG6HgsnFV*1S0{IObpe&C?%Y>g`8|w|FN~zG#
zP7M6rPFcuW`=C{5H^Yd}?<adg-pyLX2H-H;qY9b-Uak!@o1W+LRpp4lk!KnHS>A)a
zk!v&m*!&4^%BikpCx&J8LSf<lso;Y<t0XqRTI%pFD=#tor)whr&9FSH>3NT4He~qX
zjKDRu)Hj}4V!bUgpWFoS{q}z)hLgFGx0RDASicsT3s?d-uBmwZcUlt<B-F0_2;H==
zYEq1#+OnonZ;>m>m8C|B0-Q*-#Q!Xj_%P2@aWl$E{u@;_QBxsP-Pod~iS==9M*DY`
zdMqCOPBPN|zW5_O{SeNye*-ljuZ){@&+c3_G{KI-zdIP8wxE-%Qq9)a`J@m;w~Ym7
z#LYcbWBz5q|0tHh0ewxV9NXGPsfeC<ASElhv`o~_XM)ntHS&FYhYhT_YWO}MhK==W
zsAzOy0w3Lpn2>WK4vRQa-ND>{@AHfqrl54&s}`1Timne+#Skz`f26cQ{w0k8Hy6CG
znXpnumvkvfy!X&YxjW<{?YybIqiNZdzk(;TmnxM(5#+~OWSxV*+1A9j+K8#}^6z8&
zh(Aim2c~owGKrzLsB|`#I)+2CpOJcvMQ&STVAdWQH9FiGo)RjNbm7kTR+Ls2-Dv}T
zNan};!JDHg%3ks>xw{`yQTRx>-yhMXxa+r$U{;%Qmv#+ebyZx7>Bl2@{k|)-k3l#I
z?5QF?x4#`yyG>E2_Ap26ps29RzAw5$4k#4LKL!m*Plx=`V7+5d3ov_nRk(hKeVyt5
zBcn45fBkg8d^(OMq_NSl5Y0D{`dj;?YhYT}o@)E=99cuCkIROS??d=umW~&EAS??l
z_Vh*7`hh295>T#}PjXB8lqL@A&A`JSp2RO+o8WX@u3?>czFj?F9ZU7J;u#+zwlQ6T
zTgrk6U3K@IJ4WY_n4GBi(ST0e&s>v8utJD@%AkMkO2br*ugRhEJHR1?1!98w3*2$}
z9t@b*+M|mrU#YNAk=lHFI1{V*D)gy3N9s7UI}`;IUNW{Ledh2fxF+qN7Ci$&F|3tD
z5EdGLD|zRAq5AnaBJf)b+)n0pt$XB$>XTw5mH!%%#Ar)O|MQc|tTa54AV+p4HxiS-
zil-4(T)k)V#*TrBd2mp9xNKFa)`_D{B@XFGT8JE0L8ALa^(=LJOXrq(^PuZ!RX4Pl
z2?-hoRv%O4LxyC()I7Bx4{7t+KzojJEA8K&ct7r_$?<mWwUv}rHrrIRf8Vk$1*Ij(
z<?H*-@lT6=Hf$DrpTFF;gv{prI$QN5`r=pmM2O{l5M8+b2ru>?K7tiyfI#d#au2tZ
z8I$klMG8K4^AoF&@9J}{xhwXGG+UN4?w>U-bZ8H+vkigrg_b-#COQrdx7V%tKUEKv
zC`%WDyG>t3z^mW|;i#Qf8vaE%p){wSF%KtVa`e7&FqYPH|33Pof4iG}+Mf&-hP3<p
z&9Cu&rM1cGKRkxifj-8{-E>}xy|+Uu1vpkq4NwEbN=JX1jeZh;nmj)uN5FAEF}1en
z3`0W@pO7TsJ~oh1Hlro*#D`Dux13TSTqolUAf3zr<#Cj!GB56rzoQyVi0<cJg;0k_
z*HZfdM0H)U!;{k@mMp+q6fpo4=v^l_6T(HFF(uJ6{RrlDCi>21^5#>U6OL@(=Aiw<
zN{=3F<YM%`VlEtlg-Q2Gi?+By!PAAz%}hw5caeS5nLHfR9Z^Gue`qW!?Ee-eeE2<J
z>Z<&5v8D>HoM$`f@UX*6$mX7!XX@YEnPJJGS!AUuZ(d|^1I_sxD@$bP5Xgu3PW^k_
z9B<dy$z^p-_FGckNjWUNM`C<LKvyWi;d;;sqDwL;`;gV2<4Nggq4aO$iHsFVllFYG
zB#h(}b=M~A4Cq=uP3a?Pib<8Ex+AQ5)vKC9DVU=9`ymb01s6{?c)wlaq2zL!`HY7@
zGx_@hI-%kk+2xdYEX#cj+*uz&Jx^OhM{ZUk^?5+JFTlVsymO1H;+sej5>*_c+-EgB
z)@`}M{zhBosjbD5F*>>{)OD1Re4@#gCk0M=rJ+S{yzH!3H|R7yzg!`Vh=i0ZY&Mox
z&1Fn2dcq_mUj0J?nE%oxdXPNCUv)jkAy|eVt8Q^3CG7;iRz>8Yr|=BTIi0mPzE}?f
z+Y4DZw~q!qtEl`V6te4Y!YrBFH&O#@=Xfy5cW~tj*c+RY{;%O8utzl_GS0ZG=p42d
zSIJxmsk|x5_MQT(VikYS353+xDZo!430zeP={X;JSlb+?@!F9TNJpdOu;mZ!8#as>
zHYFO(u07g92YsLGViUImoa1m0l7CGQ1+Lnr0bgh_YowzKd3cQoCN$+#oMaYfc1u0O
z^L8s&-KGkQ76bt%+@7(YK?;}Ld3%EPN9;Nqc|(hfn$Hb$GUbXwO%o9he{estdd-u@
zreFAoFUOFSNZd0MUUaEHVPG*8taNE;XwWb;uDC9ZjNgV`?w3JBcCVAU<gF0svgCzd
z?$|;P?jO|n1G`Rl8MytcEpf9S&<iajo&SB!T4-{nc(3nFO-%e1T(?V;FK-*cjF$mR
zTa=ww5~Z_MeeWcH#^EV}A3@h#IcPj&M#7mo`jP(}85+gIovmc^G5q%pk6|j+hxhV3
zi0<l7*Gk-WhECsGJ$L!{i92y!L=aGFrJ)%Pr&S(ZL%qmFq)Eh(JZjCaFlIFo(9x>>
z2~uSxf7XZXgSafkeRhRR%B0VzQ~tZI?XaI##!iC9RiZJ_suER_2pugcPozxfH21L#
z_J6Zv1gq9QM6anV9>{c}k-TZok)GQ5X>2o(m{+>s^pCdd8FOQeK-`GCE#C9JDtYSI
z<IU>W%OC>Er!&M8U>&RJf6B|3hHzUX1W(YSt5jf@)_&9Qc|g=#?}+9OK#Ih-v6h+H
zH-(vnhkJD)3jHV5Qtl6t80Axb<pv319ts%Cu(jAk2H~#w7Znwr6VlWGg$6rsfywD{
zdhar~w56Uei*!O`?8<sI=A=82j)Cv{hj(<n6Lx*=%1<&{{(Hg<1uK8@LOs3l*by|~
z^iiId!KGlpg=z>Yue3kY4Hw)o5vShdYKb{iAU~GP1-@xVlZlpF{2xnS0TosEeoaev
zcQ?{4H8e=4grt-R(hbtxC`hMtcS{Srbc3`=gLK3H%<ubLi?vudbLY;v=XuV4YVXt5
zgTK3%@YyrwGqpcF=+t%M)MF+wqeIeW90#*kC47x6PcKToKO@L#8GE{!*0WXUw?;+Y
zbSeLw`Bj|bKLWIR;`c=8?z6+U3$<*P>&Qc7O0WU&d?VUbY;~`d*DH@`iOs&+p5VUy
zqVclYuI+a&Z#fy=>EejX^d|143S+<Yi0FZ^4ENcvy8Op;yT@l&3RvXFqLt4aFW9o)
z3Ukdgu3X4h<Y&jm9*YZfvD5^oYJ;8ruu+Xm9AyGq<CzMVH6P<lf4qJBEXqysOQTn>
z4WpnI$xv4f-)EMy8wPth5Y`k#cS<ImaHmysFpE6xN!}~30B#vDX$qbWcpvC-LyBNJ
zcA#`n`O>r2rRLdDIGQtL(}ogoW))&ZWA*1s=9*&RfPB8ei1t-XvBUFAo&h?!K&Y`>
z(qXoLj@0h^L1oT2YDy#i`&F7-Y)es_PN7w&dV~YEQ!It+XWUh%#keJ}d#64SCq4~n
zT`zy7@iElJj|$`r!1!>w7Ugqw`4^qz)s*yWru>B^gu~IA#?v@KM;DjWpYz7}IC-yw
zIJT0)x>*{==zp4;d_vbW9~0WwTk_`$YGk-You@I-^nl&J+8i_aD1x2*qWJcebI7kR
zsUz_(l22QLG{irXoN!Bl!k}333upK3&#98E2e@pC@;*1bwa%3ikmH<o@XA>h=G<DZ
zMLJ@QRid2-e||aOTy%U8dJ`npPW3LHebA%<ae##4#B7RG>ijDwZeq6bR2|7%<C>W2
zk?rq~e{x#4=M4=MbI8O6zf!G*AGY88oUHwD#kdm|U^aTB$W<~kxk1M`c$OU1Yv9Xx
zW<mM;=1U%3fDuh8E5gTU8~tY*I24G~{p#$0o=Lx+w`=LGlHDHyz@2&oH4YGq3ubey
zhrL#YE2FNPNc((H!BHT3ohj8K_*%+FF8Y)jI0hN4t}U*$R4ZNzZW^bNgSx+SY9;F(
zgbx=ZLyU|}tUus|A*6b0#qRTHdB?=CYsLU1O<3}r=6FmJ3{KEo!35S(R31`jBkP{@
z{A)wjwJ+{X9~dcFDbeDo6)qEMwWN!jyDLFi+FW3sN48(`gBZhOxl`tl*(X-Si3LaI
zbVgg9PDGzR1>;C#;S1aAKT{q2?LsxEe>2iVpdD}t!*AU+)*&ovWAiU5jd<7#(3tx@
zw3u`$$THNe-?iwCA@8){CP_}^Z%?iAKO8XO2cA^XsMo$yHWz66FT3#l3!^5M>1}$+
zcX7lUEEk+~xyy*d2$7w0KNl;v1m~yjlKWd6-%yEMxj{U!M}z!VCTHb(;jj2tNu(cz
zZ#2)=jQ1$jn7;}&OK8Y%qyRI?0A_*%IBC!?{R~-q%{Mrg<bW$y)*PpaWza^OuXxW_
zW%4C~C&j!xn4yxaj1GzCRW1D~IctM>34%=n18?{1?;(Oz1q)*VQjueYW^=`}9Jfqi
z3$Yi4GYZ5P)KR?j`&I2DrP)_+){d1Y`9-V!+rD-M4+1Kg0ET_}r=7lq=8x}wHT^~P
z7_(BZvk>HEouAsk%w-KqFJt`eK-H{99yW7#VLOy<JZh_dhtf-%Af^6EmDVRKvU7bv
z`(@p2Pn=2<N5R-UQf(mpo6k@3(sAOXs#~(N-33~{L#gK3iqckq+ovX)+QrDtRXbso
z4PIw+L~GD)N0M?;a=hM7HL)>whBFsQpzBYXtz{TOM&_RYAWowI(_i}83*pRKZBXwP
zKAQtA0ZFJGlh*8aUvA0m{i@Pax{6I(bL+voA!`-$n{N=2sylQJYpxHDkC>~g5!f3`
zO>p}+*dA+s0K>WQf^c))$OBI6US!yR4~bl2^9^KSA+I&Sd+$CZM3XK<t~#h_$=0BP
z^ecgbQq;j-N(RF8Pm?zI0m?wa90E!<buEScay@M3Vd+H#h!Pyj(^A5P^@ztSC;C-c
zhn;RGq>)<s!X(du`d8Y`r}3m`qimG?tosY9&((uV9}qxqYutScNS%C2!~BCBDZZmg
zy5JY+(+7drW^wn&``p$vD>dA{<zXEA*>24+jl8wTg*<?R@x`DWEzbv8v0G4-{b$}C
z0K$}mW&6b0a`@%Ab+TXiDl?-3Zai9=D^Wg4^XvJd!KmS#xbc2=75RKv7Mf<Zu(8IK
z@q<GdHj-aFFvU!-7qx)F3PUr$1a+L5spsI$$XXZ1>B-lz73FRwr^g&@LQ(KwkW>EB
z*rUQlnz->)TjAfioa1P|I(8kYzjy9IX2WQ5XRnN=v$GW!^De?6a61Y`#*nW2^brx!
z=cA%ROs@QdHhG+h-X1vEX`xS4CRUN=X*}M^F*ABiJ(kX6FHosIJi^?ClDWfrPhYna
z;$T6akzXZ=ipYZa?s?8~IF|3&*$jJG2~UAO$XVIiPV8O(2?zDk{Q0f8B{SBQ^{1(=
z`d?{$6gPvydn#8l?m*IOTG775WZ*B=h{1>gF<+sWEeqzzvlmrt`EKqaWP8E)Os`^p
zHG;&QqvS!{0X50Fkxh0!P#*=i%od}rm@%w$w7_X@ZqtETN=b2dPS)t+0aix-#eMpy
zJFK<7nHSU^FaOg1_ug;11%y$H$3O76@76)qM;?}+Hhb?Ws3lx&H1?7j4px)aFE(@Q
zE|c#AXVHOTKk0z$Wk9fTTf&&1v_a;O$b)P}qluKH&}3oi+S&y1o*z|U93Q<LOhbqz
z?JL#Wox^A)dhcI4o825g3G}Yaj?>Lq*+_@&iIj8vzbkkUWIr|{vpq~3xxr5~4fLb8
z6I$Q!)_k}Dy9F-_H87YjRWnpdRuLnLwigo7-lTWdcdr>1E7XuSJylpXJqFTq$TT$u
z1cf27hssOB%*olEGx6Vz|6ihl{FQ<u{MD$0`Nzai+#<GG#k0%99M#{?`2l5emM>>~
z_6F0xW%f6a>F4Zx=2tieunGAd74m9(oM?{y))Ti@Vo6gKwBvcDQM<!i4Wc3%xP2|d
zb>sK_l@;1QzI6SSEBT27aial=dxhf#PIGOdNtQBrwkCXyYO6c}4}uVNpOtwchR<_8
zSfk?wveLovrjy9ULTET`Z}mD))0RJ&JF^21$t`yP3*zdpp`#i4J1wXzBiu2#0uIcZ
z9(p_GbH=yszQH>DpIybiUWvU7j|9<A+AFa9s>j^?FN~>%%mG~zZ?dom4;vF7Vp|6X
zPeG;<$%L2Xg<DJXEIcdy$C{L!M8&N|aGXKkB#_QSp<L-YC*NwG7h&YQkHmJGtH$Gi
zqO8Sn@Y_-rG;l}1$3czpxnSv9%$+g)6dn{OldDWX6P!1flA_O!Dolarzg6)G3<~0-
zM5;zb8|m{Qt|kbPlODVH%O2E44x@}XZA%)xCYqa@1GC@Q6cnhW?W+z^(b0-(YD_#l
zxL}A>9&l1BDk}C54pK@-$74P}&ZWH08EyKtt1<fx<Z;ZT#7?`K=;^mLOk~v7-{K~}
zjLDrLN7>~nP*F2a-)p9S#%x_f)XxXm)qTagd#W1No!d`}%n<Wj(;n$mwo|$y4=lUg
zPRN~TGYx*|UY;JF*jC$TUN)G{yr}7gszRsZ1(*^X4SJIYT<!<qAS%3){|fx44oz?K
zJ|N<Rq$Ms`s4Owvtl@e(mL~4IEV;E@FmV@~kyCPz6l=%|`f>A@?yCQd?8#ROi~TQ8
zd7&Uj#f@2roRHCIIhs>^^{rbYS@_II39o2kB8ku2V;*kdZ{$RnGD^eNIevrCfou}%
z%pR(jTluoGGH4-?3mx8g0m`QGStJWeDbB~wzj@{To#z9pu<($B$)k;F%g2ApqHQ(&
z#Uo+rI#N=x@F~oemaobv#P+mq#DxxVlq@eaaQjXf4(f`BPptmkWe)s)Dz+w>6U|@?
zAE^0U5Hw_M`c>ufV>;@~lG6UPc+0l{eB5xj^d?SOeTqtRa*T3`0<b8|^HzDXYViEL
z&k)NOH1(|7SKV<xye%i!MSSU`kRXO{`~wocj-vfq(Neb04Aw$wrehJtO;xaRIhwmA
zWq^wmQ<v?AEp_d9v@bK>aJ7+Xl=^3c^mIGwkuS%XJ3!u1*C8E_aFWKCa=<F6G38E6
zi&?2Y3xbC2mIN0f$md~()XxjX9-CKT*KNE1eV0~JLUjb15fmb>rhotY&TiO*4web1
zgp!E+3Q_s~>rsm<WgZ$RdZN9?gjY~)NH|&#i9>`ZfX9fG)IltpoGQf1!IjQ_nzdrW
zQ`*GX3%U(az@4%1+$*|iIXej_@i-BB7I#A`HWto4ly4Pozt&)EIm!QMN^>&(=J%*L
zqy)}P1Dg<$!DhqY3?~-j!5lqsL@e*0`>206YFm4{sox#!^+ShL#~O8$JJY7kI#C}L
zYNm2`AnxYRC4OzrvmGMC+slI33S;B`M`2OV)0g?l#XXCqW-ogdb*V;bt`c?8`VnW8
zGP<OclwSf)lwF;NEWQWms5!*rYZ92b?(;6M_@eE4%rz8Mq6M<TBT<HiQ4xObyP7u|
z^hu3?iMFYB4@Vg9%e@}J3Jk_wGhL?@k!kj3im@C!uXM|~%gW1_8f@_($V$q}8gq1x
z#`lI=+p*LVo3uQ33)r)?HtN$g3WBjpsIX>O)k^q5+BU1|T-ei+iT-t4h_=S$9seeG
zjuy3lhFVb4FqKQ>_||;qoj8^=>NC)UKYT~#Cl>ncxddVFAgfRDS@?^Lsru30Kv=$p
zr-^}_vonHq{GqJtu2zYLbetr!DZ<Zh(}*86-4{<xCrzh<jP3RH;NZM(#{Q%zXUpqT
z4d^wY4Nyw;E{4~55X<Q3;2qAV<d!KZ@Q%XChPv77n<%||M+X;7XltxPoRT`bd4W{B
z67UdTjDB0<dFr0t<Qs1eyQHtY)9|Xc@i4f4DDOs+oBPgW3sk7E4I&$aurtQOCWw}w
zkrCi*(!=KLin_YQ9~a{(LPmGY0g1N)OKD}L*ohv^N|P?v@zd3;U}hMp8|(2S8}aI;
zkdrq`IYnyiyuUE96uv9IY)J5yW{G}`Hxi=whpPtu7@fn#&({(q1pD{|^9eb{TE<ix
z2ieE(DBLIxk5lt;B=glN5`$Mu7@GHe>^RQYqdHQN#1J*sBo&>++^)nOTCePJCjt_d
z?_T<mZ)<m5Y3@ddu$Q}XAEq5I()cYtwyc;s4Z8y>PGXy8Jdb%}A)gpIZ)4(E9fTlE
zbeVEWJ(^N4MNMa!a`B^{Tq#HpZ)Wd<AzsO1Y_%t^Qgm+>+;s6wgGJ=N<lDFYf!l+D
zfC7@0m9?a_G_R{G8&H%0wSyck%U{^g$~>UpP}9{V47eLLf%4AtTK$)6((PY4_-lB0
z6R=Ps0Ycy}pz)(di-HINPQ=C9vAAkJcz|@8d7>VJj<@C8>;~n%k9fZ-p3b|}4b7xd
zW|zE($yi;}4Xy1o;{co|+$Ux`ystb2I`@#3_hoWkP8gv~Sj9YIXpL<w72nACv`0jZ
z<k$D`ueFP2)KfzkA%TuM8q6$Z#VxZc#nCymcDJWfWMvr)2XtrN!4-yDZ*5DXj}2MX
z`5E3+y=J$cxR{!{LY$}itI4}Pv6qchf%0zACtJBrBIC^z1qQ|HmlA!^thPeade);g
z9hJ2zJM5r_82xb}wi~87zjy&9OSmWw+|F9@!fj?le0LQ;HoWBj?I9EYI|fkZ9yq)L
zjBwRlUgwrTdi!oB=UE2ObC8I6^Oco3uX*jB-EPMxXJm{L=sh-|mQ_*tVm)}~4O;+J
z9`(kjdmWfYJD(s(?q*T-E}hI$Grm0fCR>%vP@74v@m76*y<B@-vad;gjY-bcUuoLp
zq*}RIYPRw`;ZY@2#3^;i<#(b)D+!9bd!-uY%~Y0j!sjbTm9bFm1b;y(3i0e~_>X#o
zL8j~O+P5ii(X1LAgqA;}H*bPqO!ieD83L?J_$kV_Ny}#A7U2mpkJuTf=HdAnzmhqS
z4Q6K%>}eMHMNu>E|08h0BH&;Q1$jNKyUk$YE*OP~J>kp}*zJ}7knoaUbU}UFdA}$X
z`qSin1f&`^zZn`dQ+0x2&sOlWGTpF_VEE*o|JA_mbJ!tOLK1=>R#eI!ZG&0S>{r|!
z0|nPhX1iJYe*ZX#)DRQx6gn2kbK_AXCT48W!=9feI!K|HlBY?TnTbiUQ{&UdqB<+B
z7d*bO{>tnK8DE`VtnEaVok?iR={kNaWU!l^yJPv?1aEOA+OOxQ>~*9ra6}PmrD1S|
zHWiE%re4bu>w4xFD+goDyh%t8K`f-#<uprquzf>ISU&Z|ty*sCVssL>PLccpX0`pg
z(Bbb4pRYJo>VPUptK;(j%b)LD4WQc^ZzTh{opg70*=S|DvKP<Uy+7-Z9U#y-CKtr~
z6M$r^&if#I@Iyk%q=4e}?rO^~$=sO%!FbhY2I9?(AN}h1-(-I)G^WS4y|BW=1VX-O
zSEePuxme0~EKoX*kaf&;^b(horInU*m3ASjb@+?UKCSFvN@IJ}tB00vL!$TI1s?7B
zO5lU2djj(fZtdtsupq%X>oMQFVBQ}5{w5op?WqU0Xqj6*i}ZA>A?uVkisDDA;?4lS
z`cR}^TwGlge#pXPEH%Q<yxBlj<oid2>k8=;qi~N2XZv@ZPSqKBZFVSYS#0Wb+fLc8
zgO1PNzWiL;ew!&aZXfLOJuOoR6vAWA2%aV9isJ*<mDAfooiJ{+irjihE10-j;8h6n
z=8$RLC5bR(nJY`+W2SH>BHQ<n)>U~&?R^dS;eYIWZlRi|csQ~vE37?no2?bF{yp#s
zH{#_{=U<x!{$vsI6?6PsM$OXVfMFBl7PQ4pp|6hp^pEY5T5p)kg0u@3Z8~G}P6BWX
zh_uR^<hgzK+N6Agl1HodRATDMu3QAS40a_KGr#oLd(mhr;rrmD?7qPAi;yf#2jEq(
z(WrNow{lcU+T<>L+p1!b)xTyzBvVo0tGMfkB?q)H!<(ie2?B$94Ji9pwCt$3pGORq
zA9Qi10@&>yzt4_6SeGAUCTuS<t#(Pk1v4_lF-S1SAj|xVy?AXm_nH}jJJadERoq;y
z#OR7^CLG>N(gVUyS@j#b;G~H6#W+hYe9yfY2p>@=31T^#E|8fq*>C%H@rV&ZD7tY6
z6mFC&qB?_rq#2o;vuFA_5Sd$i`VzZX8$kcJ!bhs{3*TqQY_;u!37n`XF%RPL-)1sc
z?SvE84ESQaBHGqdb?LT*?;ocID4gZx@$izHK-TqB3_}rKuN^EMGK5-#%M~7;K;A&G
zFeT#eN~sLZGBe%Ei&(}d3yqMd1aF}%B1xQ}tqz`RW^O(M9(7R9DEYF$xtO6R`(L&#
zfNOO20`@GO5tQBxQ^D1e`dQliHYx8OHZBQFbDPS-uyB;9;--{;q1w|HJxkFJcF8VB
zpb1M}OdcS=i0T_FfZOIyw<|SyB&;oPWV(y$-rReGFW)?4V<3oS_7Hg+mV(@lVZAUD
z*Bg_OmiDL%-D0&Lu775)5>1SMMXtQl>yn*paKg+UyTCaJ6r9s~^CBOjp%Oa%osmD0
zv{KsTjDi|J<%ctE`nP<UOj+zHb})1~5Y_eQ#@hTCo4fYgq|=TwPPLdi3+2Dt8;Gv`
zXFS)pG;2xWeI?RJPkTY?r*T%{a|5EO?mhen1hCm6x7Y7d%yxsE|FND)4p5efdm$Ip
z3KT|vKq$?6z{>0#r>Wb0S7>y>vlB>WyWB?4-m!>EL%lK7_~#R?clL)WnYVA_wBk;c
z+VQXl2o){YFA_yeiSwxI5lAl0Zb|RXbic}&CdZ<m`~LJADYrzJ&x@YNh;cQe87V0&
zHj3E1*}zhOgUve0vM<yblS&QQLW_dbG~wRdoYuJ=I}lcv3>AMQE=egj{wL#ZzL-lq
zJx!~Cev&W$)J>x0fkl`8UMhJ_-ChnhKn2HJ`bzD6LANuOoE8Q1gQEsXO1D3OP7WCD
z)qKUnI6-gr*r)KsRJc#@*)c7&+Vsh6q7hhuV{|K=Fj9ZE*Q=g;9#TEOAQ}empYi|Z
zamy?}GXCZiX!uGni;*P13Xyr|7&F$Tdun7u_I9@+`+eI7CQTz2#go~{WD$c>@Aq3@
zAATq0e?j&hUeW!cZ4ataq0|m^QF+qS!c6>lr>Yc*dT~CZ4y1w)iQm*WjFEbhe(ziR
z#2RY3XcQzN&+CThy<<l8_9K|fc_!fyh+X7$p!5DUH#_bHyZ>qH%M%xbHshR$uk<Zg
zZ?V6qf0%J9IC@Mv5KEg?8~NcVePuHy$;x`;S>kL!-hAlWDpBI{R|5vZ)|||6KJ<_|
zlGo24dzw8s{3eLyLZs<z8ia_XluXgazo0GYtK{&8JFOAWkq9v-M}$~r`VG<qQ^O+x
zi0n9FuCa>jj*r6kVO<zQ+Ps-mW^iabq6Tka_pSNR%l8yR3EJ~NXz9mEi*;>Do@dDt
z?Uj~_Hu76V6S;Cspv<I8Y*!-Q-^oIc=csZB*$R;;km;x53Y%pZf8AHpe0Xtz=V)&3
z{2!MdwS2@Z{Oo_e>1~MBwD9lkj~_tMepG*^eRJtfjNm0(CfA%OiZ1=O77N$so}(oo
zO?r4V)x5=+np*Bljj$xvmpluVlP|zqV63ZPw0q`3`UdYBx^p8&s}IJ0WBt|q8!4~j
zidq?bJLpAcjZ8%BygrYUz_-b*EsvfZYchjf!`1B%_u+2ehA$Kg4S_-TO{Pc(1&&jt
z<6n?O-)|0x5Ol-{8wv%9{fR1Z18BM_PBoat>kik~i|o2+&o1c47+T5GJ;T;`M{rC1
z>0r|GcTnw!+jIf6V+TmzVqO;-y||bE>;OCNEOcKo4*RB!v-JLw8v$4Li*+A_RU+-}
z0Z^@Z9_R495XiU56`a>Rd$!D^K$@;O^1DPLisEq9(!q!%%s|z|E&25dVfqwyst#ha
zgfuN#KT}}Mo`DC4PxR%7f-h~-*tHXVIl&(ViMFbQ91~^Z%G6n*;Ivs)As25A?U}@L
zwTl|oz)4|b4pceJsND78PTPh3jn2+ct3J@mM>7%2&!2kha+3GhDViv*w<i)J=}lUC
zKjF7n|FHdwA_~+V<t*rW+Z?6XW2mDFj_E<0)LTqw)t*LWrCSPt0k%Dvv{8((2)5D_
zO(o+mZpm*vO)mnPuX_A@W>f6}fR&lW?<mMLIzSzvr>qkOSqEg_GfdsY(^un8ZfMcr
zOM@$o9UB($joO>kj}+yYIYK><AUG-anl@!_+&*WNT_#E4G(BM%Jtu;D3`82)-eeGA
zDud}SMKX<x#J<<+no0g`oyVKl6>LTB<7l_2f_ZXGxE1z`H-%-Uo=?!yXrd?|{^u8s
z>zd7v{oo*`i~q!C2+on5L6G)~;D7y<l_SfM?zBL6Olt3aoBSUqe<OjY0oBPv%*M|Q
zuJOK~4vI5@?n8~sPZi!6+JrEAiN8juV&C=QbzHAG868gT&Z$mfPn-%2PGLtS6UL(Y
z%EA~N2|s3gfB)c&8%3**h!AP^Y4&v3sgj}S2#s7r+BS*1BpsrzuYhGQ&2Fsn`BujU
z<v4C6^VgU7x@N~hryV_Hf{?*L6u2*hL<VOvmm=ZItpW1h9l0T|ho`Qz*&}IuF2j@x
zDP^KnL|L!Dr=^ioKnqhsuip1A7b&LjTjioqG$_osK1>#MQMB&<mOb88D^TEkpYtbN
zl6FD-nH;{nE8l5p2>Ij1d<q{#QE!Y5m0%d@LFn6EmF&WiLbPeKn~*iXxRkvLH|peS
znpEU6+LEt@ZUP_n#A^@2X_~$>jKDxO8LjFc$tSBLY2%{$)+B0U;~4!u%{-GAlx|et
zHjE{*F59giEkzfkdkkyheyF%;*f%(nYYIwY-Npavvc{cy_&rJPP9k-1^^)&?vd-NR
zD<pzNbCxeb_e5FN<ry^i;YCU*_*b@dlJEWucA{s$KsxxM!q95dn<n-HdkPIQqlmu|
z%_ldSEzREoiNr>>j@|zAzCiRw7mIbF^o~;%yPs;{{{AD@g_{&2<;p+{r|ioJYEj6r
zlsE%cE5zsnQDIp0+8-&FK8Ssl8hn5eLMM9QG{XnuB@bHc4~@9SDJ4QNsQJMRdvNzL
zdup84cwS^w1M@Y4vrdy$pd1nPgFgq?@BUa`oJFoab+P(~>;V#J%>dP6$!HZO=bxb^
zI8J#?q0}G^EQ8TO+_yly)>QU%e>&^?j9-#x6m$+afw7MiewVEL8i{#Rv!B9eQ0)qg
z%U|n%PxA;B2Q$=8I;`G-e+*(7(u)`1AJi>=^LpXsneuhO5jU=dw3M0kgv+nATNVs9
z2|1+*kvrPUdsHQpR@6Kx$yz-8CcnAAA&@oLO-Yz2gq#RTqHplBgg08FUzrzU4OFi?
zVv~{b<yV#Z{>sjBdSVz(25UpSjZl%q%d?EptdoomX+QN3lqhCl`l#9;-*175wHvH;
z>IU`W2>t^2GYcw6v?9t9;lC}m<q*QuxaG`2Hg*ARWf%6rNhinjh2K612Pv+E18XXG
ze~tRTIqpw#XOC7LC>v!4OMakqi(aEvPXx`tFy?CQI0HB?i`P(eBm({B`jxB&avtW(
zugE^P!wZqSv(L+A`+DQHyF30hsqXr32gnTnJj`5`-~9Ef$ZT%+_PS~)DJ_=2llOT#
zOD4)rGstksR;=;ejgl$<Z~ek>{_AM<q8v*LrZYJjekDRwxfYJ2XO(PMv%J((UJ;*e
zx|8W{(^I~b{Niz#(r{K(z&K)FNi>=%xK*b1Q=~a5Z)JV!Qj{rAu84d-lUaa?MX*C<
zET(&d?G~1!R_ttdmB^j)Wb6VIJ^c{MuR-n)TjDvsAHDO(6*WXEHiDs$p|obrpPS*E
zxPBoKz5Tku@YcXx3|W@UFk9-@QWI*PWZ$&yZInQgci#$f(Wm9D^t8unP;4oXoLJkZ
zFP5nn?9LW;u{vKqRGU67RQ)eQFy3Lc4K}xZ#CG|a1ENv$Hxr^*)H80&xv_Z1t=MBe
zyJurhfrL}RCMzJ|9QBOK!;;=v#{@a(N4!|FBny|1g{QCe=7uy<ryJHV_OF~h1h*K$
z;;TsIn{&S@L=>Qb4Y;FJ=d)B_R})zMNoRgL?<LAV_9W+PKz@O?z1L#PM1e$=4G&wx
z=s|rI*38v`!F=tv-_lkdbl&X9sNTjzzq6*UywmPs>bQ<&Xcg}Vk1Tam+PtwO6#wG@
z)sA-NCQ>0Pp@5z3W4`~Gy|1WV-eN{`6?*U>0oA624^NYOC)SETaL*=C_Z+z>6dTa-
zh|1Y{i3^ys$>PPXnE(?oOivI${>z0D9O4Q$zW&0A+5CMt-oKnchE45zpY>vozM;KF
z*}}EU*K5Y>c*Q05>lq6EPT2Tg?Ud!|S<^3JH#9YIrX{JTFccx0^uggxyXBk$T84)`
z=6usb58N#Gs_3#C2KSEns;%e)Rwz27(FuPYW0xsY)5A!0&-nj+ti{(s5rF}9J+KK9
z7VEs}W~Co~9i6^`=?Im;t{6e%&%}kx`?5Cq#ro!a9uwiSD~a4-XT7b_6N;i}5(^8)
zw6k|#0zHGgc@I!+xA97ZDU93npi0VOWMmAF184zG)ct4!Bpl2DAtxXf@-YG`r<=Ql
zl%_kWE=#)H5fOXhGKU{3J{?8pAxvp=)zpmpvmAG<Cr@juXg8b5pm!u(9ggVN`%*w`
zr-E=R&1oy?R|!BOd(>+;r@#L}&0->G-%}!6fscR&R)*#$@?^_-f%>AKM5kQ1MtD>X
zOk6?~Cv^z500%)qUeI*2W4`V70(H;Q6S}7Vi?`6z_2)@*P3&|Gl7C@GNx1lv=2+rC
zINGV2j{Gx?HPx<s4guo~?4&r;HypZ^?UdrYK?b~vQppnP;Egw>5>4zbS7zqth87|T
zCXwj(J(E}o^U&rjpah(+SN$F;WILWE+U)JW-;C&6+xg1-KgkfT8d8VCn$QP_5K;l*
zl7;PG!>SgBG`?a!ECd7?$Rzf25Z#f#9p!&t8v8{H(Le3mlm+!wO<kot|M8ppvT4zs
z+a`(RI~EX5$&1}jBI|nNJM2BgfDlSiDD}z@B`%&ae-d@=Q6;bIuVmi{=TuTQ?M_*$
zBr@nFqS=KL^2v;3sD0ubF959(+}6M4j*q%&{E*-xNO=Mys^|i)-E8-{e>)GEULRFS
z!BB`W00lWjkwo`V9{_KoOpQsJ{yU+6e`$?x*w49a4$!Y9s2Qhnp&xYnku7+f(zs>)
z^52f2a^b{?rKEh?rt$)W^Jzj(v{(|4T<$)H&&2Q55>C50WdtR)Ap0#+W@wQNQInj0
zeK6&&iF55(=9k8AA3pQC>AdEPc9wi`ipK5FN8wYEVkN!0lCpF;jZY4@Ov)r(eto|1
z%>p5~lBezq{L44$U?lh*pTa%EQf-$=0i|I~cKB2>&W3{p8f|!!HM_#RZMJ2ZVTsqh
zCLLrjAH6Yc0U|Ew%gxDA%S&+Gp8?HHLmAzKAMLDZAF|?a2?QU1DUfw~Q|Z4Yp1NSC
zZHtF^EIhJiy#{7n5WFwx54&)V#o}{lWpiS<xqc-E%FC@!eU&MMME`PL9>W2t*lBL{
z)hmX;FM)M0ti1{S@_}l4PJ;VFnY!cR?Gj9}9WfXgcBSR-f_CJuFhDn+tO2o$F6Mrz
z?vcjS-ZvXv-|p0OQmiG4<h=!1(DfX{QOOooObG`oq~KEK7pye<3n5c%0`xT@QKai8
z(CLP_x5JbQG=E<;(OrX4c76p@R;&hME76MG9q^dluP+uIpEii<b3*KB>Ru3KuDds^
z<3;K5=Ugcn89&8<?}I5_Y&OS)&DRSFNT*%8zbh{s$QtTi&r%D|%na_HH;jHMh@{uH
zQnwWb`7EnEqu^S+`=o3Fpbr8+Z0jWB<E|-jNa8DtD|;N;Z!>2wpOBQPigHOdMTbx4
zP$ngml_^+7rinFOv%$cc{Yci<jAE$Exiusf(rB#q7)oZya)D~wa9Itxr#VRBRcyaA
z_AOdLNjc8-zQ^xqUkI&l;6g9N#zyWO|F%TpQ%r*|?k2KkJlZ=BpWaR}k5B6+0pTWa
zReAeN#6|IIf)=}M*=~2}H1+a$JsR!hR=g=9n5bMwN3hmyJ{2LgqOf}l;&jiV>tj~`
zGom(le@zr#7Kff*Zez9WOK&QtbafSNZ+;TR6PMR~-ev_ejE>G?LuuI8oUvTLXEx7~
z^V%POlfahQt3n-F>!vC_L`F;(2#}uCi9!#<_n^@}3`AHZjHY`{Nr&CPyosOm`iTu0
z(a>Aaet$K=`%B$Im8_#31i7ZLfo?wv*quWm<<B)Q=AD;&3Y8Vz+{Nnw%>*9$>oi^%
zSy9i76*}1k4Akp+4dY`*I=Zli%e`RAUCtHIgr+o`rcK@|t?^da6EAr^W_v!6I84gV
zPmD7Dm8^z>k`hEm=iu-V1u2-Gz9CSlJRe(P%cEt%4^bQ|43;^;{jZ9|{u_Y`&dnKH
ziOEIcVB8nDtjB1OnfRe=xsf6=e0}fN<0vLMSJ91AQVOYM)1MNq+(%5hqNr##0Z6}3
zm7bYEVB))C{=-Lf*uWqPVmCy6(Rdr<6*I&DMFrfVqObOq>BCs_7q&46s_n^bJm!|7
zxOJkGrarSjqe1tKlVe)!ly*3OJyTp9Q^9~W+y>`rShuILf|*d}2!oVHB&n+aiUp+z
z*SIr2G7?7DqdD3-$FTGQ#t|l(#h{zT-oN?s?y#>^TF(K^c?)`ha2nFD8Fp?Qb44&u
z>;dMCgRY4EdRJEaNn5o4(|w*TDU`lAL(uUrz*}G<XvsQfY);SQDP!gAl%UA>O6-lN
zL%_0i(e@E<+#XK1#9r-t$As4UrkTY(=^1TF>(nc4_jhXWG+aSQS`uvAVY8&mABKYS
zEIbc-&sA^{+wF5xZ+P9`$!iw;&7J<;E)7Z1R4(d-7p<gAWd5UFQQ%_m!R1(5xdK8v
z`-FyQ@ou9d-dWiLKj2G;POP@q$GcTGB*M|aHjVMBVp{yzbB~jsjYraD5&<!3c?<b6
zOk2PSq}+=$(J=5(E4CddZjHA+Bz%;Cftw3087EH3X&v4@E!}MYUmu0O_M>=K_3qI|
z;Yn%#K)5b(+mP&Bae$16PG}QZQQ>C0t;{uv^kqUh<`UGQnfz^WqxSEn$`~6=<F*jm
zFHDAVZUbE?i>>}h*Q3QmpkE3~hi{T+XM@X!qUatOjEs%P;IZEPyxgB@-F;@-bHBLi
z2R$hj`X+vz;e$w2QFW(jHM&4@MC5hzc$k}P!)P4SUYVpvy8%C;fO^eVaq5jrm&=-*
zWTd3hui5L31dRWpJTs9v5CbjC{RN0V$R(_m2i-SDV@Dwnge2rLGfS4v2cuT95$>bt
zeEt}ewKBrYTEbAg9at{Zw<3Ah-n>2f0&51lADaG-GX#z)Xi`c3%Vqc$Esoy=wbdVt
z@S``61ns51FO(g3AJy$~`m^{G7iSzqa$`9qWo6;p987CNvbFF7hwyUS+E|VxgBoS`
zs?I-;I4Q?QR(oF1EPXIHKR*<9piqE7;OOMz{;W85?|{9t1KfQ7>CH^Fi3B?#T4ZG8
z(Q141SNNq4FW&0zXOS?o8mlqViurlJ_NStiv()3C>EELH8%Nc$eeTY8e!I@7?6r*p
zTJCY6E*9{0%*jdBd^?kmFRZDQK2K-hSvCMtKLsNEKTGelRoby$)>JA*&S3YD+wQv=
zmzndl4`aswv9BxlC|-bkuqb!oD@0FpvX7&hdwv(;LJR(?HC@p8p~m{0zIs2E5<FVa
z0$Q1wlEPDwb!7qf)hkcw>>DNJmVD(K^hm-iXjg3Q4>yLom%9gTw*ZFe$x_c#RZ_h%
zIel1o)@Lysa4DSC389VKhD%~Tl>shDBWIffHrX*Xi<ikkHA9}5fv;#sl?c9Un88CC
z0LrHc_Pn5mdqv)iNzcoh3}olAzUZ0weFgLj`DLtOIQO!S#7Je!_X~Er<5^xuwL{pD
ze?SKt7nx7c5TYr1bE+L6fY@1zIY*-08yY}+QqgXs^9=(k>GTFDk7xQ_FF}=eBwBlg
ziFM0IT6h2aoAMcTLL}G!hgX`aRj>mC$AWJ^oBZyKJDX$4^QIy!*s24SKIp#<MxLMy
zb$Wy{{@)sFx)PHF^x1=qmfAer#jE80zq4m@?nA3L*W16nb`o3Ew=HOBz>(FNatl_?
z+6KK&KL4WC`D#Rd&V2s2TO<R9DXAXR^<{6=QM7lqM-4-3_-`aNr0(H71tug);B9j#
zhDxTg-gSH?3j89a-4M&Y-G+l#&uLn{NUP9xL^8787AmPC4r#>pxxbRR6!wgQW^)r=
zev2fer;~zG7G`)bhhY<!%_KGG#(_$OfEuQ@p59EY#i3*kpbmevaO^Hx*Lsmv?TMz7
z^CsxB4GbpPlocGz$b%WPPyp^=2nYz?X21Ckgsso}@zg%Ad}^lH$o0oC7LnuXx%W^i
zdn3b)Aw#TVUzej+J5Pp2se10O*k+{mR5mds<`(fZZC^B%WPye}nPEW^2BU(N^nugV
zF_*KAvh6j+SJf;$jr1~VS{m|<njfx=`{%of{#d={+ap_Ygq*K)>q=r$=tT>dGK0ph
z3!2}P&T{z1rgP>Dd|n8-)F_l8<*~;KnGT@HLMjl!`ZQN75k-e;<`jaeYOp1Nn^F?Y
zB*z?OdMRwYH^&saoX)5@QxF;NLHXGTbn4sO6c>43Eab@+n|@s1t=ZxMK}L7n_VtR6
zzCJN#n#bN`U^Hh}np7~-8DO~BzVdFpnm3G$e%Q~@kG_EUN8~v8ujNKZryQ=JG|_Zl
zz@B6ks3AGFfEP{i#R4S|F3*1(eNBc!+lmTmeQk$$ArysQw9?ml%0X`&6`P32<iFKd
zc@PO2$Q7Z~<W@51N{r&g1j!KHimyD{Lqc1>r4}<p>Kt^$nNcb7NA}PAETg(sgK^j6
zLeQQg>PNedn+TV_{^$B?ci_KV=WV_thrqf$-#=bh)D;g7y=xtF+8q)k`!}I~6_xPi
z*@qXp-sXC9RynQ>{(JcR7n5ec&N`lXMJZZ|@o$V_>HQuKy2&rwWzlPTyNv+}v1+*K
z>4E?rKfyQd_9BUXm?q_;{r!Iypqpx~)8}H4%#peX{Z&!B>wYhw{&)Ii@eX9~D+DYj
zUquD+p;zD}mMBtqZJu&o+~z$0f+~6d%`bWLt<6T|tFI1yHWfYJ=AuKCvl-O&BCT(D
zxPtcJ?5BeRA+71s*kosaEPBY2codJ{&Tg<pk_v$aL`(+SILRf|k+lxkfGNh1A2_^Q
zBDi7ids_)#(W#9wZj|M2?zI0qT9Q$RMQy@kUYBZ@??NkHZBiFK&C&L~<$}P*hx{-m
zOGy73Qi~NHZ>#^rGQBMw5)BVpP)$dVq%JquBE*gf3)f=_7rc-97@j*`)%^H7-QdP;
z@Rnyvy6REKzyR!yA*`WF_&{~WxkM{1Sp9l$9KmZTJF7Rbmk*x@`==LoQLEZ+CfM(+
zP$tA5*ad%o?lX(!fPc>_@eF?zy&hr;^KnqIaP5}xJC?7?lA+*+!MP)T^iR%A-_w@F
z)p;BRZQNi{lBr8vr)LOg_f{l-(Jy!UbnKmc6)p?XAo)TTwMF6gQNMDr+PhITx+v8)
z8Ea4x;P{YJP6z5Qe88lMWsYAD%n61`*fa8IGI7ho1nqOhFmgHkMAgnLr#Bw^gK;W?
zY<=j=$GHe0p?=f53D6c-5{MT;DzsS=Yin*k-9)~C`NxJP1hRz~kGtx@US}tGv2biQ
z&FhFuK`{#a?!eDLcr2fz!MT~ztSdO?yibG^RgKzjvlQl~qicV-!cpL>)Q!0_8rts)
z+_0tM8{XGV`|}rcs%;(l{$uAHpo}b-={jy0yfO*Jg*7Td9wSxq!f>1mO*bdcg_K83
z>drkY>fwgWQ63PW@l`w`cUD<-&%L77F<0#-Q8vFen}`LjEpkkah(FT?8L`MO;Jv=l
zfBab8zxIpc&7U8=<o~?+GAZzNEmMUaKZHTMmZF4>=HQQV?T<#LjbP`gyw=F%soyQb
zYL)Y_E5ZTPR{f{`XKw<d-i)kwV4ha#Ae2QzD3FR|4v?lDO+6XLPl`A9rk6=jWa+H>
zub9Ub&~tE-)SFi##jfN+NR?c^$<@q#EOanc9HI8V?59g710V=5<0X;+`6n<EsH}i~
ztvMp6_{*b4xp_0ao4u;0onjUoKM&3vPAqP!)N=oX&EpA25-QKqdqU*m;{#X;r>wRF
zr-7>nK-Z5nQFSV<l6+lFf)i9Cg$J?<%1TUN`tL56U^Ve+Fj4$Wi&WP)f)5=qi^*8#
zaBuDw+{X_!-hOgzCd&&+=aIdpiYLwTwM@1Ep*6$>BAW1x@Ap!@wF~gxiGbH(!6ZN>
zN8B%NSvX&<HD%DvWMi!cJE24a+NLe^dt@$q70bt6Lbcpg<8ck?{CYOzJ(0z3%v7sQ
zH;<Nn>{}a-ceEIi4E{${HP)wW-)6{Iu~<J+PPy>mh{;pA+FLhkDR$nylo<Xa1wH!p
z4<z(4UK{E^dVm)RNWhy9EGdf`+TM(0%H9C9(tm(}s1;BE>hC8vLO83Lz)ObQXHjjh
z2@!SXk;<Q_A00mcmkrV6TUh6d__V_M?=HyoAQJ>9vr=+zgyLTJMIK3nh3S~U;nDz{
z_7&uvV8K`)%G^v2wZ3r#DC6{_#9II#uNU_^g4B~c@re@*VOa@EKtbyO%AsU>dOB2|
zo|%PZ6AH%wuMwbBr(m>`c{5dP^c+1TE4M_<#dTOSo(GOx5LME8`+w<nC2P9&ixA&&
z4k^Q_U@3fTWIIbGpyB&R<iMognOYIvE#)-$eBgpC`=pZv3u;4gRB>`Y4mO>8MG_7U
z>>m9k>aE7|ttYZQ-(*Wu_L<inF@-ym=gOH%TnX5F!{ojDhXlWIHPhlm6FVSF4EHY_
z^XlY2FoI(pXX@gu&9?wy${Kxkgy0_C(Ci^P2(q>g12OvZdCw5B%FLwOQ%jB(@{%2~
zoPN@t=iioxXDS38owGY%p{!S|ybXv<74ZAhMQFO^UqR{+t$RH{-{NSkGvny!s4t1{
z?i0#z6FhU7lFsJqgXp?py9G??b*A4XPkyXXE`R+?yM_p95ZXOX8F_i}GREA0ul)<b
z<`WRm)X|YuRULQotJlI01xz;ynVD~8Wcr;%i?Q352=MW-p`-Igzn#RY+p3atass{>
zO`nAgn43YR&-@>bhyFZnaY=k0IxW7GCm#>jMh;9#B6Da7Bzo@|xVN{reH~rOjC%2x
z({weAs^ulS-7wWhAwBqb=v7qy=TVO4e?a9ZAu-X(Xk!0huKey#vYO{T@Ck=xS^3OR
z<g)(qd;|EIE?Szesso<bfF9NMe;^cb_(TBNaBgh@u+r;dOt9}JcpFDn!xIRBM8?Ft
znf`R&xBN=fOplu5Xgy<F-Das*9e<}pr~gH_BuX+d@9WUBE^5!jHK#D)<HCQXW%;&w
z)kB;9VaupGI(_AqGxG(2IAwE!m3&30kJD^WGES?i^)Kns_4<#Yo2f!ToDyESIX4eg
z(1UhBf8_N~bU$8p>OmwT2fq_QX~%4BZHJa)Oa}I*3SQX6%Yl0AM{o)KjI(lL5m%er
zAyq?{UPm)f=mI4I+VI90VgH4MPcdf1Kx|B1Qxo#;9f`LKR~u;XJ2*H1l*bW2elUV|
zP{R*lo}KSUMn?^gr-0fe_$A&JDRJSjsI84OXW~^zP!j0bPDx2o*U*?M<JR}`^15wl
z8a#H_($$sI($Xp)`e<iYQd#-wYO(f_z2QHT4<B}5;~W389kp)ypxXKVl)Jw1$B(eO
zx;kAHb~iUSa5D`eO?i2NpwGEUfIDVrJFlbT&107mODX6+Y-qUra;h#XI{=%L+%(9A
zPD)DJyY7)X>h|-ob@fCyCwU#$1+Dx$6v5g~)8F5m#t)cpLUp3SGlZ%Oz{F69ejDX{
z*)fP;M+Le>P)RJHlH3p2g<2oaM_GVPLuH?KFA}q|#w_B+!bW$trsSyxcARmjs3w4L
zRz%$M9VMweSs&Le2pl|tq=_)l(E9$yi>F+-aw}1uGB4oe8Or+xK3-dA=WK|oxUdkR
zeRUivAETtCl(F>A#AE_oS&oYs%6ad!B?HwGhX!OrQ~@z5N#K!%xr_^khoTXM!Iy0h
z#%@n5bAI9#g|!+>ANcP(FYK?8k*&ON$}ONgv$5uL`r+pnFqt9m0(zj$>X$sXBAJ+(
znSVMo^L`jf5m;YaZuj7XhJ~@Qu?j5E?r@Ncz-%^e*$6^w`IDrq-|5ZI*>y_`0}U_%
zPP5k%h5ezHGd^9UeCGKkhL}5OvoG8QXkLKqwFywSYI(X|Q3NVu->tLFKyYA_J_z)}
zod2t<fA4ptBYwX}2i~mu{yk;D!^%30(ABI~>t-0)Vwaz&n)n@knAv2pTF$?lBSC2;
zFQJ#L&lLuZN^umTy1=5F>)P53lvs8M^o(d0D&EnJj*jkx61`_;V`Yu|?$=f|3Zj#{
zh3;pauIuHGno2A=-UlpqU}Y4m7uXyD(cIl}^ZE$j+*4Ei&LeyO<nILpm`4E7SYvBt
znAxx8WdmWi18t~)k`haiQG-ol$G={*9q=tb7#nA$XpX$Aw;En?Xji%S6$fH<6QkWv
z$Bv)uiq}t585kJA$4NZjT6h5hBL(nrK+uNz;{5zi)2Ed`MXB#u+1S{iEL1Fl4!@bv
z(9m{(4$ucbKfjD+gLm&($Vf>oxrkH12yX4~F28o8F5gyi3I7hLM(l@ApDMvegD`0X
z<T~DYcfP*^GIzf~zkid;af+r=&IHqwe>ZpqNG2)!|Lb{mzWlNJ>)}E*^q!o1Ls05E
zI*g#YIp^T*iNFY*A8*g<N=iz`KDo88f|~|akN@3lDyPZop%@Z6qSr>m&t26<t(x!M
z=5*>xOBK67&KOFyX73_elyd%X+MNp0Roxv&b)%Mwijf`rKQjJo4MaoND3i-9h@F)+
z2`q8-BT$>N)Y3|Z#=<+G132LLU-%B%e;gH7K(zqcXRrK?eKSt<#1teP_S#f|?4hEn
zDx<h~c2H1|g;h>IbN|4==4YYacSJtdN1G{Hsy$ZqQ{{S`U_l47#e9glOkoJ4LqQ}}
z3v?d^nstz+?Csfsdg$g54=*nXpA9w|A^S$tibFwue$fB#Xs^&^E6#LAZFfJm(-P$%
zRK8uf3&SAWTd1Z2Yqe-EP*-2S|GSk}AR`uBGOvskj0aGI1i{{(4QR?;nVs+Nm!YMj
z11jm#8u&nNO+iu7Y~yDj`}>YGkafwbtK+#oT9Tf~76W?h=wt#hS)v|IrCHc^9G!q~
z45|+U{3P`H@2gV*mP`KjP0q-O98drbfILm(M?Ih(1IAb!FK(znC;oYSz+<rJNlP&A
zxKJ?I{XDLF>be#BJ%B)9_++&m+(UtK>T4l#Oz;Kdq@<)g+iNPBIXFJdSG@ZT;-dck
z{>jBf1!H3hcqDXcM#eB8@ZMy<1Oa~l=V<@t=Y))mhm3%iG4ME+8$S;D-K@U)2@W)m
zuE%p_Z5@nLe$p_f-}v}VLV(1vkIK?`q0cs);|$SwA>OvEFonndW3g(M=SE<H<Th}3
z`ZcDo0vVNaT~D=jS75V4l!iZ~8bhCw!9a3hicR|b0e2r)oZ`H)nHf#I#Iq0}w<H3x
z-{3C)O-f_I{@ay!0eU-xri})+_Vx(W#ORdu6ZQ*L#;v>AKAyW-9{O#rwzuu{A;=T!
z&n{($_uQlOZD9ND&6N`Y3#Lppb|=9SzomwO%3z!LT@irBd$fS!pdM2pj7|>d$-UU(
zL<e<nKrPgCxxu!iZ26#ljVFNPVwupzu1Ye!w5sX{5TXMfpviGv6_`Y1T%6VO<DJ<=
zc5<@I+fSEwzI=Rq;8+K3{;sEMS>Rgf=jYD4l?Ji^@9wR2ilPy{E-WjH0y1Dz)h690
zl(&LEPj9xJTW##^`wy=)-@Q|ml||Ii(Sb?^-QIe7-<*t{?~ImKR9K$;ZoLEmL*)}X
zQZUq5{_O5vfK{O>qGw}6>*IntU!aB}t)vKU3&;x*u<6$LKY0Uu`D=XKLCf&s;=&7<
zed|^f8}S>VSxcbz#!3#eR&R|}TU!e*9IaEXtD&LM13q{kK=q8}C3C$`pE$mM|86z}
zI7o+s5YeEk^Ex#e2uNe&;q?Jc_93vvSU5Ns1O=PNK8=lz((&*lRaCHp{ma74oC_A}
z{@xqdp&U?>1KSBIHM^-Jnflsz`Eh6*I5y~GgIX1M#ORc7(tvOm7_#XD+X0~AJWzHl
zEG`DfXdS?bUT_rvM1~ygQd{9{86_lg7T)_{$ZG7(mg3<tzRmSGU4#1ITVL^$mhHHL
zsy{x=3M|k;ojWiyfDOUM!RY~9a(zI2Y)tItcVCTJ-`+|yCop$Fp~~?a>}n{1CNQDx
zw8aN-%3kaj=Kuw~+aZbP9uReY7ja{a%Af2G2vC)H0^=8E;FN?SyQ3rf*X>7j0*$p*
zRdGPIH3G0LI)RPG<N41N3b&vNm_S@92s=nInGuJ8VBpVivKMGv2b<%!*KXFQDltCm
zaq_6BD5$O;w8I|&04k90>Uh3CEGbihYJ2h9;RC4L6Z{Go>&;}9v1k>@9||<$LB}8h
zo(@eESQM3&w*eJHK}}5p0IW1`0>Em%P3MaQYa0g+;~<bDNh~acU%`Hyf14wo1!S%P
zZK&Y7FK_yfvqg)%>uF(O;kRjAfxmt!491Z30peLTXJ=>6(CKeL3NY#WcSUM!Qy|oc
zPR#WM_?L4al*;S!_ia`8vo|g-E;Qrhbx>NYUuTJuYUDiZ_*}410lyUg*>q2+GGR3!
z!}=xQv|DGkw}<kh^0h=jbR|bl<Mxzp9~U-+y!q4*jl<BNV}*I>B9n4rz3A!g?0R2q
z$-iRvQe6n$5bLn_?JH261^)<fL8DPxS_8!oc<}YOxrN!G|5HcBafZ#p@V@#c;zNhw
zpeVV)5f``cdv1z|VU*B`T??oP)XVb&&`5en5Y>2X+?HaU?S&LS)Ca(kuX=3N`T4nt
zac-mcmEH4Yu>?gD1_xBa6coE(YTkb|pC(Mh?Ee#z53CismZn_4fb5dz<JnMYbv1xt
zA%IvEm}}(1;fJe(xk}cY^0G4hdMgZYM9j|5o7XQv0ZXkxCCt%cZ7|ffK*0eD%%FY{
zyev=>2HXfxvNX`~TbQ4(12+NY6Ae+2l~`EV+0oI==K+q^(Ft8SwgPC=ezAt;(Fm9$
zWqf#BTU(9kd*C8GJAios*D?T3iGUSFe(pPz{{SNbUK|w4A2%-WfRUP$t_KIJiR>}3
z4SP#83ZVN5Agw;TMonPqz@mhK*Ms6sKr#%iG=V3q2@oZ;Falt}Fci%8hN2A*51Z{x
z=4tEbXc!ob03W2NqOt=d#6f)y;1yxOGAMl9Fq6h-iwjUSi00aL{DM=Gpo<p$x-9}=
zz6%iWd<NSjDJ|_+zHD?!RTZXQ8W4{hv$6ntE+sY9W!>+F8(=EXMw0s8n*}=k1Ef!L
z!JdYuK@;iZDf`|D-kJ0p1*T=HGe<!b6;Yr-Q%&p&A7IUX^V&&pFhcoX0fg6YvL^wP
zo#4!aO58)0XTj)DX>qnWco1eN&d~#q4<8u7T+s$7p8;6Q5<|)#1%{A#%=?iYy6&vp
zPuIr4J%U5K2jJdM%Y48mNz2J0L9LBM)ZM;ZC@h~j)Bk=S%B$MEq5_U8rW{`$3Q_k$
ztzuQcSj^>dVnj+t*5`&Cl^^m_S;h*I;NZ}z@q2K`#lr*DOrVe3c+9l&CxFrg(cHP$
ztkD42umxD)%fp5E35l0X;6kbKyFrd2A9_V=>tql<Brz!W8iV)a^@*rJkU?r42)Sow
zqK7#A?J?ydC$6rlVx*x71kZ1}%g=|0heyBF`GaQ%P(+2lECFXfSY}?Rr19QjEuE8-
z)9-+nClL5egYi93_5pBy6H@bphSd3<6M_u9HxywpF@=MDl=%JoouOKJ<S=X+8`xT?
z#tTj&Sw^3G5Yfq@M?MMy+@|R*DE>^jQN#}gmz9C!J}=1jf8#J5gwk05>%($tUqz;|
zv9qJ);D`nO@eG`@bv9G9fL8U@^vujss|ypEfITwXdfTc4pGrYf(>LJk>YaC#y6z9y
z#6eh8(_)5(f`S4SV*bAYlj9V_<YNqq?!X>m%k8(z7A;~@WU1Rf-$B7)d){5ph}Y-m
z=hJ}^2JB<D@yquef6N(v;r|aA;0l4p$j-bXr;8UZ9QgC|v*r6epARv90CwyGfc?}y
zLEtr=;8`6FF5um9%YZeHPRx!8Q?<h{tqfKNhB-Ks*jAe*AMZQ*{4*$>W@Tkr0L=h5
zEcXAmn>uys9bnUW%Cu=lwZBRj&e?qS0rkVS->*9jtR>xnHC{(Y$AdMS&t;`P2SyQa
zijobuvc~DM39un3eC;W4Bo)+?yt+Dk^1OL|d3k!kyC5Wi*LbI-re0j=%>L`!+ta{;
zVupFX9IzJ->dpa+vMn!TX3q(j5d<6>27>lhP2dC{5L_w)7Fr<iWP^b$a1InWW|^31
z1+1=t<4uQw#|1)FgSO%U0U!6}Mo_srLk4(88dAXuG^i-zkG*t6Xm@VOp*;*h;OXk;
Jvd$@?2>?r%<W~Ry

literal 54239
zcmaHTbx>7b)HX=>rMnwJKoA7!MmiNlQX1**?gjyo?rx;JMY<cLq`RcveShD~H}n4U
zo*BpSay{qly;nTzS<fO&Sy37Tl>`+A1_nb`MnV+^1~wM_Awot3pV;eUxxv6p70616
zs=KBgrn|V}X%L32eP77<xtAU|E_Tw!>`z0WuHNhanb|*Q-Bh&tMz_jEqB=L!QKG~#
zFC_H^%HGhi$?)NCPaiAf-ar<=tP7zSEY7B{CJB5FCw#ju@Cs9JQ4~uOqQRxuCh_*W
z;-6yOAn3nFec)Dn|3a@cyk^)5=*7qE2$z!!E(g`Wh!WD^+RYos2>tKdve~2%a&V<5
zi1AYKLoY_jw%1f{Zf=ppT)|EUv)N3jl-{@0w9?UE7VFH-QiBW3J+EdyM3V8v#|ZsH
z?pX6FNG=2SzR1j9C4y*Z+#S}A%~u(~(bLoa{=pmb?d6Bp_FFRE4`<S|Wm-KY8l|_}
z$$CV5_H@I^?55x&Q}+evQ=hfaP2FLU?w;;XWEB*QJAGkcSoPY|hF`Tr@_b#iU3J^7
zw%>lQu92_T;az7w6)eZzmbg8f>~Pjc_#rATA)#lv$w37VL179KnDA?K4y?L(4%y##
z=!+Phq^A(&ZidIJ;^N{$FY>Ey-^2Op0QQcD`1|9Qn1xzXqlP7BW@c(8CS;h1KL4H#
zkty!(o*r0aOw59uoSenCQn}UpyEbiC1RwOg-hdmY<*g?Fx9#G1lIVzV!Vjl#nV#n;
z1Q9{SYK2cO7@-qpdEz0MtRJg<kCqx}kzZht@_ad%Ex$ZkN=|?0GFf9Rr><dqe|-dF
zwc7eUF)`6%v$wzBe!(~yw#saRrlRo=DtNBq%{KH9r<s+6AGw8}AGi`26lhpjP&G8h
z2Ko1cV8JeJPe@ZDbRJyUA2;pBXJm{9U<t`cNg?+2^(}kdteB<tmub~D-mH2sGcm1)
zF=o8|lj7Y$Svlv?{7MwqK;h^@lLKR}bd1Sx5?d?S3mv!ZZ=PU3pTEM{zcm_&?lY;y
zCL#Itva+>%wavqPGEa79V!E`l{Z};kI~fH9dIpA_z<y``^ue04dnrX&f(V3&KKp{?
zY+wo-Ehp_pV4L#_3%Oj5btb2#V3qN<Bg>0diDMfZ8%rvU3=J)a!cI?Zn;rM;yEl`*
zeN)lY)ZG3s!RH3fxU4L4a8SlSAYkm$-qG=CMqOR~blxcDY}Mn0(_)&5!fgxNif8G&
zE`c%@hRkOb6&2AhU-GO)6tOWw;6i$!7M;%j<zi<9YNTaltOfGlyXlj4z`LnOxhEjm
z^F#Wcr?+RB5qo@gR>Y-uBTr{O=-GiyG^|}n;kD6i<DxjF<f1AJ0Ont9*cV>nGgtB1
zL*3Nd&W<zF`yX=+Yi7R&MfaZL`LvpnimIx;)DAGX;^>}@^$d?=gUi1YkraZd;G1Fn
zQKXrB#xvnhBzL^(m1s8CQWBO-JApg4{+KuTFEwq~J7L=&uRkGTQOJM!LU(p{MieCl
zmVI{r_%>bf(D@y(sqt~;S4~?uz|bgSkTf*xnsg*2C1-4}-=!;(VIpN`XTu{PY}hy1
zz9kC`420m}1s0gQw&{M+l1ehBp`l@V`4Wzbl@(o5CDxd@axS8zBrqgo^X7h`(az{w
z+)H-$fz31E|4~%PMJ4Y=MPU&G{J^1j5f<jEg_M12h<<n681UzhVf(-Rm#-Sv{khum
za&t&ydrnPkwmPL7Hb+)rawN<uHzf2fUgX@}j@?N|lZE9lK^;t*!Yz=P%fxo>v&#f{
zSVj9k3(c~RT@{~OPF)2A9!`5?;(POC6I^%GT^b*+m%<_=^BNmdX3IZu8FUAh*fm%!
zbOXy1Q&xtsY1gg)T(UO$mB<Wk3|BZc)M&RUBP+X6Qd*H?Je+j6(vqy0&c9*%cD~u^
zC2*5&i+cN;6_+XSSbKZ>#;e&+dn+xxQn}a(L%q-d^l&kTEg3gGt$y6NiK*-J=mfrn
zjD}{^8;bqL-p0|<Q4u54>zWvN4!fYo^%k(@cQ2o2FXx8dbN4BA)!i4*Uo)nz1?)Si
z@ZXb@4R`QwR0~jGh3Bkaxii0d#lXN2T$t&d3cSP~ge|)XNkZ!3*TRpNxc*e?#tX^G
z$y~PUBCB3EpOe@=Nr{Sz&ZLm?Sczz8^hc8qNAa$P(#c+)tYt>9x2LSH>lY}dqr&?m
zShpPU^hXl+*mOMh-=1wY{mqGh#!(2o-}$*c)E^iWQhRqZeWK?MB=IC$b=fQ1u1Q9I
z@S1`^`f$Bu1C6CKr60IlPxUEW*J0xm5|kDmXD235ly!7|5600pfyFjGKi!X!Y{E9f
z7K_>8tuJ*TbRi?th;6GxdF+P!a95O+)Y+_21ax(RGjs*sGT-XPm0F-djI4q;!Hv<f
zqL?ccRkdLCv0O7v(RAi*01{S#%gJi%^`d3#)4zkQ?d`Lz1f?41BW>W}jg3T^W71Yu
ztV2UXz#<b;Q{T(TAjieU0awp5C&3`$rlO(oudAyo*?cV<53D#hD+@*|X=rqGr}>~F
z7dWPT+S|B%x#ZUd-S9uHSN_EugY2j1q*<XO+u{BAu}U9y&HJA1`F`!0<GZ}6yVgLI
zwFnFLR|6U93FU7Ix!J|~k>;+dn{I8o+zLG>Hy`*0Yinvw?>u#!7mP!5LYIyz&OebI
zI&d)mJ&C?)Z53#CIT=;pUFnUdmz%G*Bn%1)(y{BqPmyN;Cl5B^P6czUH`$oDV=bmj
zUmpZ}R^8SVN}mU8uq_`y65-?HOGrqJ=SZRfr%u-M=KH!>CmK(OC%FO6MmnBOvr^Yf
z^+pT?WrmZEXCXpeS89b+Zi9{Ppv!}~6M6Q%zzHv^Z=npN*vSIF{jjWi^mNgE%Z;(b
zJXT!0jfl|^-EWSghQ8#OPZem4o>B<9S9w3WuX^8q1(DZ$zKR;e5#2U-YUCG!?w3^D
z+<3kqpnn+FVrFMI-JdQVtobc?vn&pRT+{8J2)E5BUbD$Ol#$doJzri=w7A&N$#z>e
zOtpEqz{JPLgFPCok=NEH@p(8y0^6Wr^A6m!)1>xnkT&LY^xetnZ?3V}j10y1?@{HF
zUm1^O2y%a2OneE_TIQ+=dHS;e0s?}33WpyE4jguWB!JaN(8+=mVhpm4e_wy4yXyQq
zg;U_P6ON97;YI9^9}z3o0~YlZ^_je;wD?O;A~g#r7GAKbgB{Y5c0Iuut>?d9dR{M>
zKD)9uEGd|qGlM8vsBa7q1xC`V&%e@LR>KO+q3P}Nc;%il=5DVr69*5kSL4IyE2UsQ
zW-mrBq;uZivgMqN^+Hp_MYzB<nzoZZU1~G)^A{2;Ou614HGJnb#|c2h`19?9E;T>@
zKY8{8%3RXeX9zYnHc09>^6Vgq2utefQn;*n@m}nX?G7<!ewi*(F?E+MME`$iY#O`c
z`RN_4RAi@NHHT3@db87kb3wA>`-$8t5O5q7Icr!iCgs>UzYId$ulD==5zwBT`r$6g
z2bl-9?X%5E00?36oHMB<*LWeJp#d-Y_N#Xo0vZ8?l2QpII;CMxuweR{+FG!y=yFM{
z8$pD+>$9J11gGRmxE$eB(eUwKOXY@6xEf|Vr52nmJB-C8Cl7%1WWL(Uhk}B#4Bi}A
zC|UK|51V#b;K+@@5}x-aV<NO_O(v?1q@k-wGu#<XA1+eKty((z{hccWlbk<okA8-^
z8I}({d3c~quc6w~6k94+;zpO*@o09v(FM_;>J&KrAN!`py@DUE;H^&YAE74(P>;H7
zjC`h0`rXAYE;e@8=|<0YAuoP^s@S-=uC|+1H2C-Dir-%cfh-BZ#q9xjMjAjHJ)g%n
zHLP4#^Bhor^f+mgHZY*Rzd0!ccx&E@XJ;&P44lzWryrc#)7|cwtH9RQ)_8-BUPZ@~
zyF3FfLjsCc(%tQ5%mM(Y01iSSZkxx=c^4Apx^oLLza!Ibnu8+wsvUW{4-!R82%(;5
z+?=jQlx0N|G^CTXtaukBYwZ3-+~P2Ks`=IzajnlCUc|C>%+AO?Ef$ef&Bz^1g)9>v
z$Q0iCa6Zg!v-%%^HO=%QXjvHr@UidlW=%;|6$ffKhYJaHb=<)iq?&K~+r92@k6TV$
z!Aql{q4gM6$EKtVmN%@v2Wzfu|HtyV#o2hXKWhHS`MBfRI~bGv7jW2~P;6S)z5L|A
zbCs0f$g@GP6}VjwfVvKVAaDsoffHt50K&FftP8`Yl{&rK&AjL%)QkQ8T^78;oQ?yA
zN$s8bnju&$xySV(6t|S=wx#l}x&|Cnv>bz|Q<wd2;-wvb+u?7fUNgKnI_zAP-d(l%
z=DSXm7$=Ug?#;6*lMxaqt_N$AbadnfmV{3HI#5aIE^tivnJi`?V8M!kiK$3C>a4d{
z;#KRZ*sAA+Li^ngDM(QOc5YtH>6tz~+<h@I`EzjxA~iY@hi_sc9tc}|)5W-ZdwT#7
zEnD+eS68R`Jl*Bx=f?q@d%V&TM~5N``B-bBppnl7++IQg{)3JkLW}cJNr`kND0IGf
zKVE(KcXc4GrA3TJz<R4hj5Y2;r6Cg^`UO&)lu~J)BG2d&foY+XUr=Cwx1H<&GPR1j
zdSIU+aPU-EIK<O&aV$=oRRLB4xi=fV0goXVr2Vnf;zYdGWaj4Po**bJ`#e8__zyBp
z*xCtF(d3xFueh!X9-d^e>e}Z9YjW^4C^E?w51|=;T~accB?`x_VDzUqtWl?;gfqkI
zTFc~jN|$A2TR*Umf;pIpi91q|Cj=fZr{psPzJIcA3K{?zL|AK8i<j!#$8xD=M}?i_
zy-%^m2ns==MO6Hb?*%rh;YFz!7!U!l0Uj~{e0Qi}&07w7ym~%@dfs>Rz&Qp?Y9~PO
zR#j7T11n{I`O;{)(M}|bWdVw>K)?(O3j@Z83vd+x_wViPUxR8WG)D%6gKq+I>gsri
z0kA4~;7g)-lK%ewAAedALqi$cr=NKMd_lnl*afIbI*bVp04TxvU4Gg7{^)SAK7!9-
zhYI-x@XeX1km%91!NEcOt-&~}g_<BNp<7?@_;X-nk=d38<iRBtWuMz(j+V}=KcYJx
zC&>pOZZN*-C@Imv(vwuNZCObf*(o{I1qgP=q8vx%T~V)P?LtXENJh7wLg1}R$nLBg
zNCUcjC(O!HclD7^*E$F!<9=8-hfU;4Pt_O?msCGf@OUaut{7Hlm9SVhwCq2}1Xcv)
zU+8UxP<pQe=POXjm1+erh}&}ZP@Y|OuDgUiG)JWHm8|1_yd!Eq|K-jrS=IWZa`1%L
zFi6EC23{Q6`?gQkFj$#|KP!Rhy$9IkP0!`c=e~`S4kQ;gP}PRmHt7gF-EKnf3#|#l
za2Yl4fBdUdwLzEB)Fc2^c3^NYDBTn#B@vc>e(gE90wsvqLJd8@;qmb}&WH1&78Wd=
zoSdLazRHNBmCo3s2XS}?S~!B&t@gOF7J7bg0QG|>C~j<u2oZZvZn#WG$?P2*09Gpa
zP^n8tPk#n7vMmS{IAmn<wI(A>%*+5{qk!{OR#xuBRT2>G2_E=xDm%rkc;2X3NT9AU
zb5c0kkpTiv_}S*bRI`&M0h_jjw)Wz8UDdW6T_Om;fFL-9Qz1Wqa@`X^ZBW00q$&aM
z90;fVAj5rK<c8YYVfBD3fD<VovfnH_D3u#MwTVRZ8D>{iK}u&zgm0gykzP^;2L+|?
z<%3ceWqW%Y3hSWZ13dVm6AtUIn@YVhJ32D_Yrag^KPZ~z+I`Se076I8PO7CXe|jYO
zyCmK<&jdIm%<()~To~<oi#|}>tA33HpHz9?*#igR;;K!>M!~`wFsxP&YTtk0I&^M%
zLUO_*+t&SQGd!@XW;V$k{x{`WH*|O57YG`}Z!GYdnwsQ)@I;9QAf|&9_h~;0wpzO!
zdX#0KTas?pd@`RN?>`uuoGjFafZ`nY$4KzNCT!=P<E*XWP>sB-ED9u4pI^sc0s_CX
zm+zp>#A<uiS6Di<{OJYq{1HL$%DA7&o~aLPwhw^NXA7p8XX^o2ljWay3lx4x+1sCg
z*VSw*06SedH@sMHDXXZ+z{rTyv8Dv_L25yE3Cj<juZDxM)O2)or~eKrr}E_py@79&
zjF#o)(OI|OnNTUWYpi|!n}(9azB0TjNS`v|CtNwHiiNgwl%=vMgN`UzSs<jUs@nbh
zbR!IEj9;`d0^ML|L%NjNDPpct-5o*zdV25d%mXmFz*#RY9v&Wt@h?13hBmA?lO8SB
z!>#{`U^hA68ge|Cjq-y-qOfc2sjYQ#R3uC-P|-P*!oxc9rkCwhIqut4+b@SzSuuV)
zCqLk#&zKaPj%@J`S+s;VCWPM@pv*`jPJdwR-v9O(|J|GbI>~WXncZ9bw+YwgKuqOz
z8z|KH0f~@%RAtap5{9||cbX9#Dt+H*^PBr|bqsjejxwA~NO+$r`_r;TmV4Tkw=Hfo
zzOWx+4jsgM2tKFM8&gp&DwWHxq{Q>I3rXO7kQSiom=C&cL7-&jYAdS^|98iqm+L?<
zRWFUY;?CV^7SBNhpZX@oU|n)=#+m*qESwF^w&E6Lf{7|s67uH#Zd(+)3*aBL(^Y^6
zu{A7rgPP3kZ+1uuzf*~|fTI7uyZ6xh8P$^zyMKN6FJlQ=OB2*LJ^E!GCl*hWkFa(*
z0WBD!A{QQ~$>g1h%C&z}_U&>S!#SmR-nO9$N|t@B_6I4OYWOv9<~)$t0g^OjySi!6
zmfmHRg24a56*G^;(XEU^^3Av7?Omth?m#UGXRsv7o#J3@s905=V!GeNeNA~FHoGY}
z=w=65ef%``J4fk7$J!XFO>0+{Xn@($Pc}5dSJP1g!V+tKfB!xVRCAlj{XXSo{<p`h
z{sE=GNIj?0gP8ekjrO~n(d{iCe$<%*SnRC_FSDA6JQN2=<kdSyv*)KrD0GBo0fCF(
zT!3&nZ+scGPeu&Ii_^5uCoF~|C1elYK?LQ0%$ANF(}vLkuiC$_Uk98e9}f^sK_=AI
zl&9L9s7sat^f3>M;6?iqnlRV=&;cU1tgB>yI2)h@_VFU3uRD#;;qPSr@ORzG4#fZN
zj_Z8Hx>Xyasiu1@3DG|)d2FnV+nW{KuKUf7AWelJO?I|xX#2&O8VLh3&JDpJkg$AA
ztcx@L4pL(}V}`3g$js+}3c$s-+d7!yv)jMfC<E+i=*-JnX7(Gy*Rn)WINp*5+vlvF
z?f>E+q{{{zNhawkf+Qs$5HJoOkBgE=FAoo`Dw=jkGTe5Eb-nM5g>Kj3R$QlH|6xi<
z7gA)IuU|0FOum8_6-mNZulEl1Avrvwi+s#DH&#0(Jn301Bcx70(0kL~<goi1&{}}>
z``Bnp{dMUlK<)qLa-cMpyDyapz?EfsEtDO4k`R%zQf^C%fIdTkYnVb5qc5-Hk_B7~
zl2u*`WyXM_(GMVrByhfQ;BU<jkrhTz-d9b@Y>he$HTieRw*{u~_wE-&|1#GuK@C#|
z0%HQFQGXR6R-tqR6eJ7&+UL&t@4@~|D3?NvQ5AwDSxh9F2h8(sW{Vx#(eUe_d^G?S
zT3%jWm44R??K(5Ng5=9gQJ0*9ztddu3_Zd;6sc=+d)+q+!gu`V9YhhlmOaecpC7w6
z&!F`l6e$=df9|jBcq$={nTl+O1Hgw+^>PXq@9`8JCKa^;gG-i77*}n%V^hZJ=6EHN
z<W0E351unnQ*k){9S0QDGaYn;chR%%+MIZPcCf~r_htGa`^!qyg}OR8Z+yJ+N*{6K
zhf>K<-p9Km6BJ5+`!~`5+tH~|ofO%pM2|tYJ3~xQZ^CoG^g>Teh3hQl*Xn2}h$9CA
z+Yaf9I<E+E8RXYNY~2kewT7-a+kDDrAbQ_QQn-6^f={7{g3;|{E%cyL5*ep)cBH4p
z6eA4RTfAmLX+fZW5U*^{i%IW+oJd1iv`(wKzWp?D#IAYa=T}lFj}A&=Xl?-%HaG@Z
zBrl|#T!%&56Ky|_0~(14Uww2kwkDUd&h+Z7X+=8ZUY#3~>kI4d#MtP3b5lc`dzepC
z8HLUnQ@zv0YI{o!?4Y;@MbtrmE=Bf<h~p?~nj$6N0{ur;srdM#=T`2r0)N}owajT+
znRuo`-I88NhRiZ~X4YFY>=A0G+@wzfH{%<&(326m+rkAE;Q*kKL;DPiYwsKt$$+Cx
zd%V6$I=xChU>bA*48jpVoo$~+xf09tj9yLVWYEdRQU|$1!0h~38!j_l38v;}Tsdvt
z-x%-yjD8HzA0L0vm$eM4H?QRXp@u4SHMhe)g0bA=wFJUc_v#5|UTjJNC5Tqf4kQ&<
z9W<%s^B!xxv*8d>fF?XVc(&gkzZ(VlRm}yK|G!Fl+Va8s+xHGDdY2Bk1X;1E7dZhK
zq*CL#F8e4^`OU*aUN`PKqy#jf%I%IM87rkSM2~O$tB=>L@~PgPxp*s1u#f#&>cB9v
zGNNI3TdyOoQuEuli@jOJcsn;H4aw5^5o|h_Nw<8(=0D_bdc&eax=WQ>4)4;6ZsfSg
zIl%f|`1`|Hzkd!54E=D&hM%#Cm|F8|r+sw*bq*ey9ThR5+hwWqic(VY72TdbM!1Om
zRlAolqd=0!y&($^yF?wbudi>o<%**~%Ljw6&h9Kz-*G-cg#Vza$G}bC*jpgZs0(<^
z0_qK4P0WE6<K67SMEpA==8C8b?2(X_o-3}?FXhs91aa@~wDF|*{1NlfR%s0mU8KwL
zbKKJs+5f?)<9FeGto_TfIWatzyArUbU%Mx*c}2VuR-sl|y&yk-q=_e4e*HK<zm7{G
zch$^tphXsGgD+D0@MONDPK>3X#{94+Ta$69aJ)r2<X*!YBrNDxP&{z#iBqI&I^x0P
ztLp3<zoskZlth~<2_F&4`!b*==!DWpL>R}SKGsMffPq=I`RpwkXfG+Y^_$`G`RH@e
z(3B0<_^(w?mMD;s^E0NOew<|x<mfNRa?{}M2;ZOKVEEY?Kih<<>Fs`MY?HT@Y$|$&
zxVnzdPTwIAVzZXS|9xyR52?Q~xF}>rT+Q&MJj#g5FSHUt6?o2dcD__jTr20F!xK!p
z;}0yUW&WbQllzAld{6Azlm>fgcr|5J4${;Jcdz2(yUpgAaz>|>eiTSM%r)wMIcDxR
zd=o#*gj0Lh>*IuhE!(@o7nGBk(W#dh#96_q@<Bj1_xffbkewZFW>@kp(Km+hB3<@q
zd_5zI^DSmVRqlD*P|7m|Nns&Bos=-npV0pi%v^!86HI<~xEe`SV)l(*zHD!oifrR;
zc~w>H+0Apz_C>?l9vfl8)=~fq5%Q4-Y1ilAv`MFfX$;VpVhZnuas+^)M#jL{-uYp}
zKu;e81;cMFXKhFMc5Ol54ek__%om_A94}U5v~E3<LJxMgLwRQYqO~z_!c$|YMNm>=
zz#WcGG@p2Z#8X<!M@wTRAu<1S_V%YqTh)p%X`fao9*!5Prn3nr{ZEOSpe%QC4R0(_
z9H#G&BwggM`r@7U4}3oTxTM>A(qf9<zv0;bRjB+qzmY46DinuWu!6R1Y@|NA`pXc`
zaqER5yy$uW^VHwf7Q*mS_%3}Z?&pg!t=&syo9HTj$G@FDSi<rEtAobUhZH^shEmOP
zLr?%d00suMrh+v}HP43GJ2*gpFo|7n6nr##S@WE5xT5wubGSB^juS$h-<^ZdGAL}D
zBx8)+-O=p|lY;xhi1IC_0&cORqAybx|7Ufp$2LsU3OB4M>>;|<3tS9}&r$&cAFv%(
zN<(+DNCem157=Hc|D^#idf?&iVsGjlq@NWb=%9a-H5UME*?1WxCDZL+->jA!<9>c!
zItL{BYrv9&UT2TnZu-ElZ!fi)?CBr>UFCu#q~pGq52*Ck#kWutZZR{*?<8)gf&j|Y
z^l0Q9-?2hf?R}wIw0BWvHAn*`Crl4dD!-C_oYbXT|FDQ&X-rSNhd=bP5k*B6SHpZ7
zwal7Pz#ObjrqvI(3Fc(U?~ge@KW{v4K5+dL#^?$-%{u_GL!+VuA1$Z(jHiEGQ2y7(
zNgfVj9R~d&fBzRYAowGM^!OE+0}>q$0_r<S$t&SsfcWQfJO2XchLvWg6#0TFZ}MjZ
zkPSYvB6<qptKca=ux!L>8g448e6;YAzDf|j(tW?%9#8<kjGpaP!FW82RkEGHRDl!j
zof2p{<tdR6R@L5MbmGDoI%%i0J+kP_YYEq%!p$1hOe6Cqfe|B)IWXM=sc$M%`l>X`
zuUwMXe^Jlz{i_@GD%%Y)sYoKf8H;VOuLYkPtO1baFZ6Xc5aCu3kEeUp@nGNVeDn(t
zesb>a{NU&LHf!yN%S{QOH<DArveaOM^rD}JKL6h*ML=^x`3umwvD@sY9zOrxnC2sh
z5|*vwXpXisGTM1`hD(9nOP^d%#d_F*?YuXnqqNX{gH88h)KVeWa}Tfk->QH4UDLO6
zZqyE+z<9kiDUKH;u3Q6DL4A<PyI-5Bg^tHR@Z>#^>$}7B@v5<i6fWPeQxjRV6G>vY
z;W*fnE)x%PNHJh9Ubv%Yn?4=sYG~-XZR2OS?=uIZ6L+Jrw~Bn(S|F{<QcDNR1Edvu
z>*)(n#{-NAe3uNQF;CE>k@!Cvv_(A8bh01IXG<!tjLMotA@J^3<}#%`c8Pc&Amq66
zwzlMEY>yZtyZw3@F`TcPq;k?a-XeA3c*<*hnk#5(<oPYZy@0&OgXduX16MbsgPRZ!
zAuqtx^+gfMQ~aaGS3FY`{2#kiM8)B;lb(KgEwQ8GMM<~HX!c#Y;ohEPBM0b#yA~|;
z4y!bco7*A8#p1~EapT8xIE#;qxa}FuG>nWP)kXs~!1JM4*ShU$maB~}T}{4?gG#gX
z+d*YVMotb5=$w~wKLO^+`1*K74m9IRqAPU0t~CBXmGLa%BfU#$)bOc#sl3qm71T>o
zpSa(<!4@<YNMLzT$n}igMmP*+x>FvL4@05|x<^SxZ|{sO1^1ct&iULZSHkwssLYhx
zFVS$0pJ3&B6dCIX<SdZ#LM0x8?jdv-3b4%FIoiDmRK(cmJ1iYMFeJ!3RMEx)FkAjB
zO!)LvntZ99g#ZN9Yr?6nrEr;I0z^g!dQIQCOec${!d8sZYXjUjBnaXC02Z36x16&;
zFa#GUJ><Tf!60K{6@sot^Vu!d#!5p_Ym9YIi-+J-yqCI_j?AibLw55pbBeR@c0@;)
zURnUV4I_fu$v~5}R1~MW9zy=bs>FqR$?JSCw}nKue}8WY?B=a!0UkRXIWe#vN1Y&u
zX09gY<W=a$o69SSGQ*%QJNZf#ZYG33Qfe3)Nnf{%G4Ljxxq?~MOg`U82TjH+1tW#D
zPO$wdgBoZY#sn_Vp{y<-KEABBx`B5jRAYArq7Lbv^fM$2En@#W{hG9Jc9(Y*9o{{)
z9`ybw@O$H&Dt!fffsW0si$tl1qE~M%zr250@yFNhtyp(DLksh#OBDPiLZhky9Ou!a
z6x>fZ%f~cUdtR|<T$~bgv<ux9ml%{Bh|F{(bN|*LY&>EfYoYZMXM8arol0PN`T2te
zv&$B3Y5+2&BW04*=a|yx@A?4$LP0&)+w*=>8{5aLxy(KKy=hJ$S1G=c=~D4i%M8GB
z$B+ns6_<GJKY_}kOgx`gJe{ibYN5ND=ODmzj0eJ!|3(7Y!s|(;wjQnBD<z(|QZwrF
z*&Hj*z;iHXGj$F<yn|{fExQinIZlrTY~05m*WEvAQH&f}c6v1n1!^DE&w#~3%Pt&T
zTyXZ7UzcW2(&I;qvvtV~_WqJ#e2RFQU%V8*G7k+$^$*LU;5GO<8E|?bMMD%>q>De3
zYp7@~W)`Ep>WSc2M6HD5yc(7ACsjXOf!2O7H}HzD#^}SVU%l0b5h8Wpn3YAxe>9Qr
zF77$G*3WQC@zOOu(!)wFxtj@qFacCFOuWmsaN@z}2G@rRP}73q_<cdu*z`aZH9OYl
z_>t1rx210p0!rCO2v72t_-JJ!t$Zyvb$Y4ZWv^8B;pw&Hh2xQkKagjtzTS08EA{D3
zZVRjF%fwl{Kn&Ej=i<&pQcx+mZw-1csZIX+^=qj7=F?BB4G?xWK@**SN^X*G1W1E+
z%S2G4^64HiO&+Up3irk5wFi^f5L}mEKaO>zWH}R3)^~KUVs+^k7#@hB777OLqtIpZ
zYd(fDcvRhxs^rFS{=`mV5VU=|d+D!&fRGof@4uKvl0QF*!<$0~sz?I^0*UIUBaY}^
zplRt%;WRRvF2Y88{wVLB&_5{ook(nRzreEs$eDl=$rQu#c{wVVTlqUAAat#=#(G4^
z{aR7b(CyRAXQ{3RtPO2mxGpo%tHj(p7D!}B+Hth88uEdcT2x7(zZunae?y$rhO2XB
zKZHc`HtQ_K@W(8V27c<Ffi^Gf)kB38oXP4p!VTKzdfpiUnLdm1))jEFcq;1O29Zev
zUVT-x=n3_4P2MK}oH*b|y8uHx2%0S^w}ZJv7d)|#)8)74E(TQt*lu@p{!4m5SW@CZ
z0wC|GRsaA6N(q6!;aw9rQW2&CN<+Vg{H#uY3CV<n7?SgRWFw+ggM!BJTIBqk#CWYr
zHSR}q^bHgV^i<z^^Raz$?r+^E6a`yz$;!s~qA%|6a>JUAHQ<N97M!o}k}kBrp)@iM
z(-rm25hEO}^jvmwQL+*>74er=i_z+{1yBo0rmeKNTmat00Vs;HHSlV0(n(K^Phj9~
z(QfS4dp>arHj}AY<xbVPrgH@yOU&_E<4MD=rYRKF>_+}xYyK+05COYD$i%vRQ(9+*
zA2OnO5BK3pL%G`6yA$PAn4hwqii8>q8euVuzi6jkP0Ry(|N0{LO{jz1PmXh8N`;}{
zyl02nL7QWHTkTxvP5Mgp&KYGLOy93qvJlS|=$O)<F+pJl!^Bjk|NLYxZ-SzN7fS~k
z3>?H5cv~%lLwskdZO`yC2RUD^4-PXu*%0)i29lFiEL&{a?}7p8YE-YSRcD5s{Mnfp
z2ERp<diO|#1I5q73g;{tZ+)TZl3O=u^t%{fw8%&m@*SVv;XVI*QL^O6`IDKmNAKNn
zTCkdA?Ilq~Qw6!tt6|X1GMg0$cKAnr$MzvMGW2JaNQw=8s$O4WVl*cUzRiwy;`xg$
z|F*=@JKytdHIySaTI%%5&e<Hw5RWmz?+11^kK_~R$Ta@2TW|}FJYzlHqemJ?50c;4
zld%%`uthcF&~Eg<PQul0N#k&HD-`|Qn}AWlt6H!S4ll)MVnW+ZrBr$OtGfO$hPj%L
zYC1JR^=0b&t`9drSkflqq=&EbV>h1ri5LAE9_H8Z@b|ZNG?_k6e1J;F!V<hxeiNt4
z{PqPM9Mh9H!T|y#VAL0<YSdgBc+Ws@n*+j~-f9HHJw(FR^g;HikXF|FHbzZz>E0Cr
z4+x>SpGSYr?(2Qox6F>{X|JDLHdd=&^liX&LMS8;6;2GlP9~)7-m9s^$LK?_e`mRU
zPW+uAzu=D*N^(tx4R<1WaSuUfkdQ<6?+V<xdHNc??<2Oe#S==c!n7Tv@tV=l7C9*~
zI*aAjCtYO95X-WB?hhf$y#-0uWRu(rl(09HPpNAW9s~9deG4gS@AMU|!q|NX@AW#!
z#*U->4p6<Y#nn-ZNPhAV;yB&q4m?<Sy1lbyeP*$3Wei95HJ#H_x4MlQZ*QnE=s^Vd
zQWUh_)e04j$FoG=^m7!}(bf%IC9*x#l{kKtDHNTbyeD(A1|40HU-%e~BlFED^fVY6
z%}~Gi-`h82Rz*)2i{ME{e<&7ie9Whd$_z6aQSxJPORwyWpCyJ8L+$(88vdbXFBIaK
zpYF_CwMb8sw+s~y0`)P<J;#xLVx*QCTEs?|&)m)IaDm{y6FkS9c1`tk?ul-IE%Ev6
zZ9yQk-dDZ8mv)?lr>V0#M`w%WrVMAn!TnRP=1XWPVQAitis{K2T(c&ODdC5(6xOuI
z=85F|XQ3W@0_Rp<y#dZ<hXF>wgNIfI5zyJ!MK@AcO%v1m2EXFrQ^$sO^}I)(`(tt)
zc1NMS5+Vl4G=DG^Nf_}H0SL!uCKWw+Ymrr}ytLZ>F`H8pd`lA`RB?j*jo4mvd27KE
zYO<uIHZ!(ZM>t#1R^p|^p^a?eK#l(?DL_v$HjqWSnJTQP!tfWN7p6#Ge-jlV_YwqN
zS>3Aq9$I7(_263mCR}{fwqpB7O*tP1l5g-Y9{W9S^^RaO--32Y?rSzwN98Uv<en_~
z63!&c3xQt*o{{<x5gg@}d_3<Pj-5nj<Pe)qi8Bt$#`k@C9of3)kesW*7Kj7jK%$!e
zXg9!g6&b$v6=sTjZsSTXzL-5<4|q0uxCLrCX<<#=5z+avrt<ygw(kc_dS`D^OCZY|
zs#YVHD}x>QfRG_e$O2|Gz2TW=xeG&*Ga+(&=Q5}5bZkZQr~7{VL6@wwU3zpSxWEkQ
zn<_-2n#KNQr$GxAf;`He^f!Orz&-=yh|+68x%bkM6Bwt2#@V4ra^t9GCE<zV1k8B_
zB1JZ9v%XlnSv?ZyfefUyBvdjlQgaZ!>1+Ylibf=pm;3C7CdFmf&?-IQ5|g$4m7bO{
zb9aZZcEHgW=xU=uM5C<t`KoSpAF#|@*{zorgP6$Xn%wwKm@#h(A23IXkJ2&SlY9)S
z%5TqT3vq=1i04$eC-$BR<GStj0NKnlQP&q*i$OK+l*YipmziCV@1aCwVVpGB>Q^V_
zSYF0>wq+?&U3Jgd>WLILp8bPg%0}F??@D8siO_F{?T0{bAZSe}a{pE3zLBCi;k;K7
zOIG$B+@d~qvXDe?J{DNLC~C9>$HI$}4k3JW#O~21@(D+)VlN|kBSUGgJ|%+>4o5aD
z)afBr`m3!8uPMw}jL}v_UyVw0M1-1P&+O$mu6}ysDb4?kc}X5RYDbMAjWgR2jrO7#
zwY?P`1d!m+JRgBHohl@SE2Bt@&<+k<cvvHO?I>qeTROmQhd!n)sp#~Xg=c!PWA?gG
zy|a~9Z@Pm>h~SArl+z{g9&cZi%t;=eHbmQqriSlSmcK6uXSYFOdRH{xiZ^pWv-58b
zrs|Dw@~P;J>gyXe?u7_ynz+B&V@Ajkw1C+FxW#YZiJH)K^TC0n<dD9xXZ=@>#y~+)
zQK-jBOAMv=L)ZYNN2JFwMFdmEmPdX;j{C3D!UXl>6z&ePqL`eEN~mo5>cyJh=Pe%|
zlCfhAWJI`ZFB{tZ8fm5X1mah3nQ!-n03w4C6+w>{b0Y97z?nAS;Ox;AD=W9pW-<u>
zitaD-P*jXJ=85ei`zmSrI`k(pnVfK+o->tg8eKTD6#m>lJ!M2;e>`WhZ6k!ZtfWT(
z2?<6jHNRTG%Su4{I^BGW{^&e&#RX7Gp@vCuq=ZuZ697cbfL?q6UlT_Epg87HK=3RG
z0a;0mKP^ADd-4H&572I$9<U2f6)Gv{1$8(IN6!f7@S|22-1N;pq|stAkyPb5_0@Y5
zn31HC%yTZJ-iX)1r;jf>j$(qIf?6RSLF3?WR%*I=c?)fcqwIzfv97sIyNd5oW|+78
z5kBsS4r2f82=!HeT@=r^0}JhH^FqmYSgkxaxE)3<(Lv6?5b7q1liAvi`EYebXsmL2
z)C16(R80`J&|e|3W*texDcvH9fo$843fZpzB9M1}b#X)BnX6v%DT;YuNWWT%F&Z*8
z6U$OBPwQwX$&l5TkF%h#SLu$Q#8h1*cRCE99vJ8bt#H7CWO$y-B4Uya0;WmF(vQc-
zj^}>05$ERTH@{`fh@g~>N(14WQSBuH&l}{7$b2dk<Q#1Aj{>4!Us3&u58=yTiR3q>
zgjOb8JdfXodG%~SOH&d8zvv{hUShd8J6?XCOxfAug5e)l;t8*y0+#!?5j~Xmp52}i
ztLI03{&|Y52nRWVhSSGSqUzwMDWShY$_rS?ZXaHez4S&QB134sx9%tU?w_xcY|(^A
z{kbydFh=P#8Bf!sffVJ~`v^rc7exw9C0SVN)|b<)#6PX$hZnnEU`O5CDrbB#k7bC=
z_&dcdf5F7x?)WE9<tdv}C>^$B{eM%XonW9cep23OP~V&bSw0Zr4~jTyu_adCl7RJx
zCiy1oUbr!Rzr&&a9HLn&SZU6JBt#itbayFY$BrHQyLb7PpzA9;|M9^%-#wJkO<%wG
z_x+8-JH-&fT+UOTHUP@C>%t4w?q#+n;fQGCMG_%w*?zS2^er_UiNOWfw|CXb_~E7k
zA*N$VC266lWCWu5{?x2LS;S<8s?c2eao`=O&Qalt6bUO7YOA^TRlnka@*nW;Z;U&9
zd^_l{Isi<*gaLIkL*6|{Ue>11UrJVYW8^4mgUh+ZSDutJM^ufEk{>@^7n=Q3yB)^V
zw_HtLk4~Or;l-j#-l}>|x3sg-@}u&zToDk^I*@6$VaRz?r@bWnbQ=J~6g09pwWec$
zmCAjIaw~W9VR8LE9IfB}D`}^1np{-&Pu9!C5m+nYBIQM$ZBE#tY@?;rs>K&4q97g@
z<Ng{Jd`rG+fvKhPy2`rC9?F?t;|dOW{}ev{{1Th3D-bKy*9=AMs>Ccjn7$mNE4@Ns
zN1)_TRH&-dT@hY=@3T25>nk;8=z04X3@AAmY~|sR8D9E-DN9l=GN($lR0^!Ee;ad)
ztrpw88Fyn-Xw?<6<>g2Snog1XPU*K%FRaQDV>xy{;hg*smASDTN{2#aL9rnlo9gGQ
zY`ihob}33IV7Wr5$vg})9YB?c2-U~KW_xk`RnG3@qvnielCiG)9z9fh4!UewjkX({
zQAm)%H4b>OAQcU6$`rN5S!7#?d=klxGDmpgs}%ViqtDEGLRWdm5g;i6gguR7;QTV%
zfPf5>9cfKpQiY0ZpvSL9aK4^q%ioLa_Afy^rLO$PF;b2`?i|BmF`Z94*-7N<orSs3
zfE*gZK*m|C#jV9jU!P~V^d4S(9i?E1B)cuxz<d&J<e)oo{>9o;xvz+Ah#W^>;fE_;
zDF$3n&xZTB@##CNa};vZHf|bAxAKf-%)I^JjmCt`fGs*@TX9G1(myE@wwxdli;kz@
zgELlfrP0DXQmNX!P7ab6hWKA<;UJ@{7(CC%4&Sj}+gpz;7_%%y!=T#Q+9oh2atkDS
zZa^#pWaFDLd1w|cfYujM+-r3i|AIBUa}?HeT~Bh()_+PT=7CvwyGuBs{)XxsJA9r&
ztH;xy&H)!gMKixUA|bN$S}bTc2nEO?BQ&eO7$?z?|9q!R_p)XpIt!0F^hNvQ(y7VU
z&e;dDG6nanA>H>T?SX$-5HMJH*rXCxfS_LN2p5G0WWYe1XC8rfAADhc&hA2-M$l$r
zI-^xHX+ZMg=+9HB{>q8y!Fmno*+^^G-s|vOWOW40U(xE}$R!myxQ*g(?8^Q{Pd1%U
zrD9&lhGq>Ni&zgDAm$lfkSV{&cP<@QS7(QR^HoW?)cg03j#w4i6FOGbNYId;1l@65
z?HDTJFtP_^121dIa;83_xw_!9w=&bQj2!(u{nh86mX`=LhkDQ$l+zViezH$u40b?P
zd>`Znf3Z><dB8Fx_-U4JhKAP7?0JUgLWT%X<59FUV~y*B<E=_H?O3&|&|;)T<KZ7j
zdT$$WE4H$*UH_sY!Wk;NYGj@ZcZk^~Qn+)VvWz-Sg*9&Z(f%Wftd5dx_TkA`c_|91
z&T|swru3OF(y!AOw!yphrj02*Zlo{A4Nl?SdTbG$6oD20LYCCO65}ySVn!0Jv6g-3
zITPhCyzasIIFcnR8s}$v*>zkJZdQZ6V#@Rr_U80b&gOi1N{Q3+);7cYpZTG)NBvWH
zF4cMnpA-fJM%?v9AR}3Q6!dvFm{ucW)CUuaOcY4|AB#tasWW3nJJ7(|(dMhXg{Rte
zlq7>93(U(tCJjY3IMeW4Pruy$h0wzMU_z7eyPl}qaDM6%cGhI+%&nE%0%xdGbPryG
zK;pgpPlxSs9@ex5w1y=m9Ydzi;k>k8FLTrhR8bFHaFg;hH8#<5vYR`pk$SXap0aS+
zsOPoQZpL06zl}tK|8>y6@`Mt!CJ?AoMJO&o69>9coGe$F3eOFeb9ngTj;|-?T9JJm
zazd_2qK#&5Ztb1@VB!}fCQm;wn|TuR#!w7}=SO@OaC^<WtSlHl@}n&dP!V0pEo{no
zmZv@Nk5l)~M1Z6X%y1y3?`RKafl~cW?EZ`B0>%!1rzycq5|~?&Pv`#z6bx~(v7H#a
z%OXJg4jrZNdAuS7R5$>;T>70bzkdA!I#nztHnwfR|A8S822=MtQ#>urTdywbWqf;$
z=)U%}*P0cBf8x8?l=gH|*9QEQq}Fw~2m`R4@#a&097ti~!Hd!|wxDawG+LN_qB<2g
zp6B~M`*qW<=}UJ{x!E?mT5}n0JKQ^9CNy~m;}%#0u@!UcAFuJB&L@3G#;Nj_sw`B5
zO;MZPBKd6S0>i3)b0B^$-m%HP5Yk2AqjN_nEeS#1^JW{J5zcPb(b!<aCXDX!Oz{&8
z<*TNby?GM5J1;h)V9X|eWwJ^^Y2<!G#mwm!ghmLayROFcoRSl6x)d6@W@(5zliSGk
zj(JD~YA%bUof+R>&Wsr<n#Y*#W)kJ9As^WZs#fVzZmr1x7F{S4D|ccL%p#3I$2_2u
zRz@*`{XhdMda~NKH&uuQv{+!O3rvOpaV9*NE!(XgpxE4u5$?CD_+HK`D_x0yX4AE+
zX8Yk5k}qyEmP4hiyCXSON85j!Ij!TibatqNEpv^FU|%MvBA-oECmwyCP8{?$UDrPV
z%Zq^_0XBj8w3H#RA6v+F8GE-X|AyN8QvDZ!SLl8*RndsZ;iFVqjpbW1A1#V)xk4(!
z!@kKPwCv9z6x0`=a~-=p02YS_T|n`VtP^trN-}fV()fvqXg8|uj_;|+c-yc*3KW6R
zb4&JD?|UpSCF@7Z(B%y{5z&%DmYpGIx$SwLXN5&IG3%JgWEw*KiQ{R2SN(G}ByjN%
z(faVfwQM&)4i(@922L+~xxb$KGol@1R`ENI)3<|z@a9@ru^1K`-K5)nRb<|d!EZmU
zZ5j;@s40nftcYO1u$@ttKLQLGo#F&iGB9N{K<=^(bdMO&893-{9~dHg0AipoU%re3
zy(0`J+1t+Z?O|+qzu&)qKL=u0FGSMuRwcAl3h1-4s8LLkNT;c8!cwJLeM6uX5$VZt
z8-F5rVv(hZSHHJQ_`{NIxM}qe{i004!$8(t&$8ve3Uo(|TM{!cWEM8OHf4EeZhf(;
zNOgDrtANKG#ABuOP4sXp-w&o+3xSfQASyqdSlK5+7bElKw$w72dZGIU(DNq|c*pAt
zlkdQY>$JToftevQ5fh_!IW=X5n&oFS5<|DjMIIoSv5pylV(r76S^yti7u^+|{;mnA
zxo<6d5QYBzr3LC6_uij5%uKNx6~Dd>IQd~Iy%ig*i5{&k`rt+dB&bGU9%R|uBTM3M
zAPQ~@2nL{=um$pBZfh%<Y<!75niUVJv2lon%<z9IM=)!6>)EdM1XCV^?BAIsmpWD8
zCBX9|ZXLq)h{&C-w!5V!l+4K=6?;6F%bazW6=%tltt}*^C7y&258Lc0R}p89>DRFK
zScze>M%wfC^z36zra;Jfz8=}*qT-;d*?_e@QH6p-(R?HF1grYYEB5D2TF<5>%L=a(
zZS~}rkXJ85{rLB{kxosXH4f(6TgkWDo!+djkV4pCHMEtP?;pLE0rT?nZ;vl`C|_*D
z#==k3-R4ZJ7Tb-+$N?<jhfcT0f7!Q>P#vrE8w!Z>HO8?cLMnQylH`@y#~(MHvwxP2
zMm07`92z!sZsbJZRd<Eu(oaO*%gF&bC{a{@Zi(mdYTJ)wUH32~O3&EKOOZ5hb@lb-
zI**&<RxnSc@pX83xO(Xb4<A1+(UD1=i<vnLsI-ZF!A$7-&HWuagu;3|X{kk7dA<8f
zVpRy`r~dmq2+%CLQK;#tEx${Oqaaq&*(9)<x9tABk*{?@#P9gv;YS}8zfh@m`Dnz1
zZyZ_>fPeW5FjLI_@6apaIK60GU(c!h{7b8D$~l{~i#>aQ2){S~ggm45(MiG22Tq8h
zPeV1goU_@0O1!-$i=ZUftWH0&?D4k5YNN%~*`_Rmm>vH8)W3+@alAEyCwEwQHmYma
zv__p*89;%%3fk$=K{ODxtF%jfcI=-jIv0&FXCevdXsdbK0_gJvGujR~`ra@cE-xuD
z1yCsDewiJ{d%DFdX+dXZN-KJq*}Lj<i9NgDqjF6B`Ze+<=X=OOn%DGqX!h{DoREf&
zpvtFmqkyq5cJ|$x;a8;-KXY>Yp)(klmyS;l7fPT@{&%L7@TBdABDJ8}=|B@q%t!+5
zF?2+yxG)1ctOQ<<Hfd2EFTdhdxh@Uz3r<caA3W+*tH4a*;8R!^KPm_cH~)K$PiyUL
zc;39&Bby}|rwo37SjFtNJ`h80qVv_QbbZax0L0uTqK#>fI6q$NewwT7Z17?CN8Xm}
z5Mv%?I6ohVb|arP0c53^3i&pbA45RC`ODu!<)w@2jh;j!I6qBRsxw^}LHQ!C#x0Q`
zaT|1Fs8~Xrmu1dk-?d62YLtq;gLnj{nW%U2cp9Iy1NGHK5hhH!79Vb%VWGYVMu?!j
zaAvgvThRIbaNqQYh2eSFu5tDmRYbWZBDCpvhJ}_IAqdy*?78He8#AYQ*tX(31+ul-
zd|1}c0E>~k(;ch4(|hTOR%>`!5vOm;w@09<gV~9<c6s`8XKY*R_m5iKm`@Bell^WO
zE^#D#)J2>4GT8&!x@Br7#awXl>FFb2juAS%0|rdNG;gi#Mz=O}!g&h}7>Ohqf6-!i
z{TiosK^_d<4Gs-uD`g64$SQ-GPjcw^*q$ShG&dwve$KiWdpo)*YBOpwjrxJ=T@xyX
z0-}*8e{g6gn#jT=`{1(~UHCgHB;cq<I44i-i=>xpb`N(L?T;R%#$S`!o)W7rq%?*X
zD=-}FE&o{G`Q~J1O4;EVyq95gpGON#f}~-xcRV&n3ATiDYSFo0S<8f2>d`j#7~k4A
z7+H@qP+h3VnN?GTek5_~t9}R(!DfP0r~F9j0;pPW_M*MZ+~57#8P`Yd(1r-!-)5nF
zBqjf(y#@M%CSXzm3_WH*6MNCb+DVkuP$VJKr&B(|ZH6+du3e%d6p6<+p?@3wJ)3kl
z0Y7ZZZc(@)^nRyu3=pp?hc`l|Hlqyc4fo@_%zpbU0}VcMZ{Xq+K94G`*>oRhKqURj
ztQe<UlI_j+yd>&U-#?_BaQv0pDk~gC5B+bBhzj}k(VNF+Ro=|(NE%-%H_l;{udx3H
z=p;UGQwgUjrbnOuDpjil?k36o0}g+$*VnVDz?Z(9d}6NMU!Nlzh{Aq*V$W=rToMA-
z91&<yWVhC(W8VJUankT<*Psw5x4CMGt@waD{e+Wv{?KV%c%J=l_>zrPuZA#JR!oE1
zWerj6XvKWdWZiC?3rPInPTEJ_e(*vnky4%-?OnZA$(lvhS2Poy4AeVI-l7WXCF6o`
zY;8>ggAV5#A(YVRObYK?1NnkfFZBiSzG!;b&sDz&Uy7<0&sW21TvJhqPFEQYzes!O
zEyk@WWm?Pvt@X=pK!YyzgQaD-Ww`!pe;TA6<u~vV7)IOm<X=u!ah9!pG|ZZOgVih~
zw3N}xNR!uf9*xo^=l0Ts_J8{VUFtZPuy23*X9;GEzz`0ICB+p5Q^kcTBz4pe@kmTh
zcYNHHgk6TzhU;cDNJ;|g(?h!wmq|FttEws})3lZI+=TTxWL^3II4b|fSkxqaSM->l
z`04>NqDH`^zXkW{!hs{0(&vr%#1}~cR=kGR<(>-iAR+}sI$4qU&I674X1<sLg&6i9
z+I0(I9gn!JgwfIoD=7&}>dce><gWdw{;akW-9#G`9i9a`c3g)JIQ`(Y*#V>0RP5|A
zC++vqVCaC{PFM7`A#&K=r(k{}F~ew**md?5J>oVOnN-RDMB_yUBEOlB!W#d2hnuz%
z4W>4H3~jum{v9?GbJ5++3ae4^z99}!s~Y+m!y-kkSp24fPR(${LY{AdlBs9IHa0%q
ze?+t4oP_@hOB52;e5m5TqzfudO(3oVnEB5s3pplzU^%t8ZL&4B?21@QJY2l*J0l<<
zN%loGju2+tI6kT7+1H*K!&`xT8g{U&c*16;3$AitXFLxdh`x6%=Rrrr4iJ@?!c6d;
zt6M}f0Ka&l&*BNM$(9}Tr2rws0x#REml&|G@xq|Bw4)MrQ!s5z=sprUaFQ#H1x6Cy
zzyB5Ip!h<K1ewy6_c(nx4((h7#z<XYgK*hPg3lF3ZqqunOVp3U7@J!-C4RD`{CS)V
z2T1S~ZSUnyx2*>f=W*mLLP%~}&8L_a>mp2h{^>v@=(?ok)*QP%`SjzfcuKV%m3I$w
zN6}@#6d{oNf-VreXckcFfytzJ;Qj3t1=Gcq$TUfHqaP>v=WZwb;4s^79og>RzY;+v
zmk1S(xk$W`iT|*9^Y2AaHVmH+C6u<RS0NPX{81aW+)&w-TuUs92SJyD$|9&)G2dS9
zc;w7mG4Q)sc7kb1>S5GWL{84Gx)WPRwxZmek}ajC!jg3yUn*1p&9pU$PP*RvfZwAb
za5*k64`v~u2m|cToInkKH-h$$EBZY@H=sR+f%RYf?9W%-r;==HKAx1aVFT3EYDE<|
zkt)R!dN=2BSEUjta=8^BF}Zep2CO)3N3M>{S9tvV;ppLQJTn^)wPfPL=Ghy6WF;Vz
z&ldZDR16Rc7{tLAg~S?v2VDwL7uSPn{XI&Lp;6W@D<0pS!ZfCzJQ)%JIcKqVgz6^q
zXoghSV1lI<D5zYott#xZ?k{dst`DXXKVR{^{IH{<xzA9p0{^-BO%aQk6ZiY@m-FrA
zzJe_g*phy)K9jD0Q2$K;_qmw?m(b@@*(g5;&&Lk4zi66d^1@ygl>0uKFB=_0MTjbQ
zSP=EW=n1qE0Q$EArA$STc1r56O`atePALDdv-i&aA?tQu3)L^m5VTRW(%O{S`-}X)
z1`ar49dEI!wWGjL7d%~8r|H@DS{M<I!*3X*^x6=9E-J);Z0$rDP%mh;y4rnKC{&9X
z|H57}XD9zit1-jG>`WUTvYIFPKU|$<P}FU>hiR7X?vySGDWzFJy1To(yOs{YK#)cx
zq@=q+Bqfy&Q953_&a>y!nKJ{!7oB03|2}cwzw5efIABr^Q#5`Lv2jM(Oysk-A^38F
zgJ~x34Le<05YN1E`S#!V9{fI*Ar_qfI#F0VHMl<J?4(2^cgg-Er!{Y`(#u1%lPxi^
zOxh+ZW#4@d3J9!;3^wzZ*+q1;ik`$4Feq!#;yFs=%WW63@m^3EEY~SLa&p@7Ww*e^
zVkifh)9Y*PN0CP}fGZ7yM^(+r>LQoY+zbu$P?6jxcu)waRl57k-z&}Ps?&L=3`Y}F
zd&-phs6(vz$_9FN3tnKVm1EbcJ(PlnUsIC+u%)tas<>zO+;c7TYHrR)>;e&_<_2|?
z?YdKqHhO-wP-*H*#xmR=kH#X(X%<K)=B0%1=-p7;7_z*aP~K!w*JisdUzDJeCdT+D
znp<lof)v@O0s=pt)p?-~To`_>KS#1jexSc0AxUYhyqa!zrz9X~yj#<&`EJ+SK?Ar<
zHcVJ{S5#3S=6uz6(K5u-qcw#``(h1hTL@vmm<qTV<O#n2R`mTdynz(9>TuS5OIIP1
zUgr)=mgY04jg#MS45wlGbQ6&ub{z@w8t*|=%}Q|ICrDf`JWa%2zp-}bj3wXd)e{f4
z<*CrG1rboND*OBS*(sT=WwPt*+4SMM5<iKbV>37C;Q3F0!PWtkCI3@UP7`40;Gr;s
zy8oLcAfW)i$;M&1n+R-ud^v8<JCpDBBa;|&(4`W~tF_TTDS-}@f^LLkH)MiC_%2LG
zO=+j^{xCvWOVamtPd9ecdOtQ0?C#Ma`#wW36w-;q)E6d1YNuY(57zop=6JAbNIVjX
zUs<D0ud(ou&e`cF(f%aI&Rbcr!2pU{MO>hS4SD&4o%ZF488&~4hr3w>F{M28T_s<v
zTy4l9t3vV}UB%ikfSUhK7Y`Hc&PX)IaB~Xe9Eg6$+(&=hq&{11rxg{!Vg~k(Ve^*w
z?f9oD=h)oA*-@egaBQ|(>k6I2j$1q41$6k%US-YKBzE6DL=r0dLcJD(AMuIOt<4XS
zCRu?4Mijyy6D${+pYcvcZt5qfLk2wzhHf`9lyW9hx&`cIF;f}K=Wbr;Ka!XK`Yc`K
zPonqJwndkUF_KaLAGZ}O@#}*GFjvUPaF@rk%Q)r4Z7!r^GmaY%(M|_|=pVkTCE^F7
z3@3qSQN0G<iu8~epN{18e&wC6n@bzou)oqgU4I$jkwPo$v`+^1&d<|9ApMKLR1P*c
zit{Oy(Kecg_H<fPoJCeutl{()Z2WFUo&2ObmwM+j!Lv6IflVtx-uyd-Es-2MjzhCQ
z`j?*AlNyS*#=2<G36&AOi1Z5?q~FqVn)D1Cf7P7+Pd@P)o=?PBi=-()Ktt5Uhn*g;
zQ?IqVIKLoh=1R|^{vJzKN~cLfUf`TY+~hiLZ>~lWMQ>3@4pWDesa&CbbJc%-rp)uB
z#9}YB#s49}0fP}Fsbel%;?7vi(YMc8Ajcv|DqbI%CO(*J^mJ|w@dsbJd<wjE1o;L}
zlXh?8H=n?Hi=ni^Tvqe|c9I{%ozElZt+B&9)GKMaPPU`-V&8Xy+Sf`7*Dg(*8uu-i
zp$yQ~Bo<A9I5}mibY3_fNUoHGOb_d-nMF*kis;79f8tsVH6FS9D?EE9kltfnn})nf
z5aZYVRcl)VRa)vodQ4&GzxGNjvvIlf!$qgAy1L%wFnx33uOK#)t>j7$jwybx{}??A
z3T8)r0cD>5;-y}E#f)8(R9?=@Ct7xdtR8fOFvhc%l>$l<6uh&$<L>8H0}xO^fJ1HG
zyE%p9igNgn!!T~pJuV1Jr76+KuNB@;sonuuRxUYwAwe9OHr%g5-;8$;3MX<2apR-m
zJK&EWuVrLpMkXhB0i9Db^KItl$}-Vexo`0pgcS)M&t#&E@)R9Px9;-SC$5njs}827
z-<NNW{fclLd9yg#o>wnVSNsKYC}t?e{DBe886QoGXED0$ntyPv$_Vg6Xv0NU8o8x0
z-wxIDyF?Hb-w|i1TyCtkQum@nF{8<mOTLt7e`NmCrVsh&wL@@MQSrZmBzU0?C?y5~
z(<5iIJ(~A`h4x&_!!@>MkGDuD>zcq^qwOR#<gOX?*rOkE#Zz0w=f13gg=VQc{ME+B
z2Eg7Cbafr)t0=%*Q&suTM~%U)IhhSJaygl3V7Y*sU!(&OWqV9$DTRKnDpg-=)9$Bv
zYk)*pj&Kqyi_ksHno_3d`Xdh-jByBNV-=^)`fakhe1h@E7aG2pL3k$hItU3=GJ#p8
zNe>y)y-wTF{wR0R#wW+D`CmVi%^$1-5x<`L%V05*I|q+{riz`y)6~fhO}aJ4Un7Za
zUO9C?++BGC&bJ}p)XFUQP{<7^GJ-x3mEMGc)bT)rKX=)68A@~ae{@QRPc*UtjNlk%
zCW~RPdjQ{GJUsZoOB@&;J_mLXU?`IN2~Ni=n!Cvx7q=f?J&PXycm~b^sd;byt~76G
zhC&((Z#v<rt<+cDi1a06*>WcTbyqZx_l1z<hjk<#RB53|TSWRk)5mjGxua~9LwcW+
zrwF}!1PG6;t`G`h-$Xlwc8D0vN+^&hBD2*UL?S~6z7uf`*VlNuUX#rIL*UlX@l4v0
zc&m^qO&UE#{J~TOtEE-@r6z`=n%A4CzPfyo9PmIF2R#%-in8RTAN8j$+Jj^!9N>n*
zAa)bEhRxz0s^Uk=aYA<$#INt@Sh_po%1{n`p_7%c_N|{%EZ?;g+$0`_;5$iDlJ)h;
z1}Hpk_=qh56-CM~RI!Vs*mFODyJl?cOc>h17<<2y&{SivFAnRtF3f&S82j?tXlGvK
zPlt5xLYf}SZQ!Mg4BrxrTo5rrm$PAHf}{G^Mu3DsH}9g-*u3YQ;O9h4)BDpqw=*mW
zIr&#|cS#cra-;oOzcQx2%2pEFS5~WLc)FWY2Pn|!l#ob!i{mFVB2k)ExqLH=HLtar
zlA!N<`8n@`D(PQ1F)|W<Sg$0x^-TGqARRw*)D-Q4Wny)p+@GtvKw$5HYkyjR&?gZ5
z1QHNM8t@@AIoB?D+h)$%vQ-*$(b_m4M5_prwh1$+$Mi6iy(Z9m*;yl;-+jBKML&Pz
zu4Rt~_dLN1W<XQp(ThrEQTNwA?}EZmZ)DL7zueJ`LV<G()3|y2XNN6U5gR_W%TV0~
z-euS&-_ZTV``1X8$iqaQqq4eqoT`_bzfSfG(Fr2THjQ3I!_J4JdS3^A_-xc_sY&eF
zZriacd$#7*{9~4PzRp{bAQNJoG2$+oZtS)nu@taf&Lo`$J{)MPMR>PULL?mL5-5Tp
zQ8;v@D7ce&(`^1D!(EzD_~pi{mL~)sfuBOa?gu#>Ki?jGqi*!1`j%wAW<}9Yk6m-R
z*XQqhs0UPaE#CV9+v}Nm{N7gtM3a5{q8o`AbB`;}$sOJ?+4Eq$P@X7+98!(FxoJ@(
z--=jYQ0ICc{umPWcw<i~>;h<RS%=k*n>z9RT))Ib4mQqn9hLIu<jj``9DfcM-XVc-
z8#$_YJ)7S$wD!<Tto0?+O}r8)VhqnaNfo?dV<fMhQ^|6sfA8WD1w$wiN(JIF7abP-
z@9blkN~3<D#4lQRb;NErd1UqDkF0p6>F>)ot<Pe?N0Bg;8Mb(r2131z+Z{io${jrL
za~5J_;uvA?J&5eNS^J5QQ5d6`l?pVA!#_<419^=?8m+2~3=+I~uKJNF{+t#^_2>1@
z@Hp3_HC_&yL#^)^V%*yDm!R=L)&9Bz6&-(hU4dfO$`@}{84hxPBW*McM~b#`1HyQF
zxf6m0kL)x5S00rhCgy7=m8`%790Fw^&i5l%^bM#Z0C^0Yg-Kg6-2GPMNZCEUsNkt7
zFSH>o9g&Dhr}h~yA}Zur6nUxzq32A5eBqU+<A$l+p_!Sj6o}1su0cVop+GDz61`<E
z4rGZcoV32}r$fWK-pqyD6|-^t8@Fz`o*@O4cZF9As@OL*UCp#W$+zElm>=A6CmMfB
z>(+uyOyjMw@MBkQcS+=Rj747&<r~I04XtgNER?njXrb-?8$6NApiVdd4L4ma`}|ae
zBlC(}1%49SJ7NZn0-u$w<bPWHejDOvM?BMZxS^CtnX^~3YKWk!jYRvnHGog50vG`8
zlR1=oc|&M#hMdXcUyMY7z59jo=g6BrVu2tdztj~f4*V*>4RTX1fFov~I-@0G_<=<z
zcPFyovy^5j0wFpHf;@4q6*g*InP8ey^<MFJ+GNj^gw|7TP)KHi;Lq%ExPTCYleJ)?
ziI0*u$WssQ>cUbnD(oZ>w0<%|q1Y_qmDoyz8pPSN-(bZG&NQj)Umde2Q46?Qgu2~L
zlv_DHavEQ$l!c;W^?!AbAG89f&32lB|G{GKh-UF(ldCvg$O(hKsqePO>ek@l+e)A9
zb&anU^075RveL{Y=ejTSA7WV1u@lN<Jcb#BbRyH_H7XGe=u8tq{tIjyT2rSh&UsHf
z;4vVUXONW4gG*4ZuWxz(yEGqi5w;LicBs(X;hYl0(Ztsfs@`KLqj?l62+Wi7>%uqw
zn1twYY^L<M>(8|7s)(5aKVHF_JS5r|=*e9TKKj%M*Q~fnZstTqy*3f)KtA6+DyZ#|
zfBl+oH*c-v_H-Ph*4LZXw)B%Mi=KRnsW>J-`9Qdh1YBG>wCv66$UgexaYMz(?<jP}
znno=|h7DZYZEUlx1eMtm6+B=cW{n1DWZO&<fEBLmT3b+_wfuoa4x8Fv>pbxf{SPHY
zJ>L0g(OhuDw6wGYj+~OOaSNicpcWAHpR(=Hf85t0k&Mb;aQgqvaDQlR5MblJpobvJ
zZe3MgKX61L9R7MXbNMUM0!!hdZRE*^`H04uAf-SjOXK81wg7b%#$t?L*W9J(khY)N
zf;)bUW#dta8PzKXe#?cmMS_}za4k{xc%qZT`v>Ufrehj9o_9%~KZRso0Sk-qUGtv_
zQq1{(#Y@f1*`VcS;Wl+dSH0jQ<aN<__?|x$3l&u@Dw-R+r2wPOQ^?y1e?VC&dtdqj
zy<`j69+2B4zTzT;PO36-aKLE9F(H`3KkKX+Gc=_^zwCS=&IF~v^9`G_^76r&zJf^z
z+o3qCWj$J0?Dp~@D9$qEL@3+e^!_44Drf@<<B$vx-kI|_;Fw@kw(E=wzEYNp#+W<q
zpDCc(0PR<N(yO!-bT|*sW%w=rKIB;@TTwYB_wNA$0?$pO*Knc%oG_T5i^t=chvgno
zxNk0JL=su-Ds1<?22qq4pV~inN}0<&6R`F+ZKFNb7mFc3`V69aA-T)As*1Zd<=V(#
zu`vMYa?o2Hw&1f6{`iz(t@-%~X!i76Hy!56+L&6Um0#d!rG}F7#40f5xL!+=%wbB>
zIN^hvPEf88HuN#8F6-fCqfvKaqxZ><E7QhTcLHq1cjTh9Mb=TR8-G%{ZmjWe<KM+O
zX8d>b`y|E}(hY1#K!T<M;K=vg7KrS&`-{&-iUo#I_#Cdi9x=;*G*O?VEpS7SHa0#3
zH;Upev8FxRz@Z;lZ6<4jE}Q@_$Ri^wJ32D54lE2cmF&Kch&gLSkY>q5;fz=ARA0Hf
znaIZ%jm#r@!=Tmk_!-F$QJ%zLEt%WLV?M)(--%*_N7O;v*2|KW<l~<xW=n{AU*qFK
z&&1BjfK+tievM&3!@RZ9zE{-3W-&lNz&X}Z%_*Te&>eqV5Q?$;i>gnPxN-jA_8wWy
z?}%-{=B>k9TG<gxQ8KvDB}Xs;Uf2Wo@iQQFv8Xff__`9{p~ummFNN=$ZJ76j0zU5b
z_4Vn@3F5~cM22tSM<#IGQgkslF?kItbAIi4@88QV^(-wd0ZFCu_d=Jy%gZLfy9W$j
zLPh?Mm24A?AYTpA&9#CDKCd&xR4><;Ix(gF)+bhC_a|$lbs>8}IG2ybl`?LKi7Fj7
zPmPFc>i$Gt7N}8ZWwUzB%R)HbR!J3p8Dy^_SW1yWt&v~;^TujJPCe?IX&oHx5T>f}
z+jSH1Ja8)9Txq6~Mq%e79N+!QQYmr5H$VF;Nt))gjrhYoxCDvq9!AWzk4AW{5>1`z
zZEniKV72p365u|#Tgf}GsYl8-k@br&m&)x(&ZZ>gVt)aCBLe-b_X4(<pnX-bXRmr3
zU2RKe68;cL?^GZSO&hBfoaH?HcYpWgr(7S9Jqo)mLtnqf1`m57;2;~De1TyIJPHWg
zySvXpas?PvCV`M+IPKLVXd%F*S=e+hb3vfz4EPf)1S#QKE$gA=FO@1UoKKESRgQv1
zz)PZ#jwrx=N8mqp2W4n{w#>eB1F~kE6Bo~yZ1-f}&eBHDWR|T44!F~yN6B5^MBcU8
z`QpBMYN2@7A02t;KqZ!k`>E#ia7pf8M>CA(@v@LyTbdap20%7M?xNMxH3!*9s{Olp
zrYiV!6mU2sCK`OHsuW9-5<vM+K?k*3pjAVPzWgFk$meTnEWqxScU2YArJhq7xVFW0
z>j^0swr8%>GeyULD+-O@%P~&G#QJko&~c!VH4n@-RVEp#N=r^dDednj^2DD5v7!^N
z07B-t+C7XHnWLeV0_500!qX3>f2s6sWwEya{7{~fvIEyT06Cx~U}k*&&BRi<wG_@b
z0hIzcaac9~25~qBa7`%&{j8AlagmObJ}lr_h0AB*94A~Ti;O2B1>8&`D?UWPBx9@l
zzP8L!m&K=%2wEieahmuQ>YDmMgFR6Z%gOwa!70=HkSQqEhc{lbgZ+&pe8R$ev6QP}
zHGu<#cKt(BSFZUPq-oBaC~TD`s}&_IuM1VNl`L9Xrgubae~$Tf;SIf8pLq+Js;<+V
z2?ywEe<xFy=fgm88oN#BctxlhRpBEA0><^x6@`JO29(W;?7UK*#8xoO1FNcM9x*$-
zD|i)g%+jf|Q5Q|l2#5sPXI{L|%iIfr3P?s4SzK^XH`!Qt-N$*Z{)C(Kxp~+r&GX%k
zA2AqP6tBW$s(rsBVgsBHCs0QJ!`#gzeEB)Pf=l~k@g!pw_E}!Mcn=<7BWHSGzFToq
z{3%GvWgrD0KXwidU|$Augh(j^55T=5z-6CtaR9jM|0Ee)fj#nP{3ASX1e&;RD`Kz9
z!jD3#kIP9)M*_t;eDhK8QGgBRW9w1+{f(tVP9m=}H|^+%@HwhOR16}W!+Mt0lI5OC
zPA!#OM%2AmuA2E=Mp{Iod94N)Gf|CO-*Wjr0qR}HhkcH2Oo@&a8A$7pUO9P!ZWDeL
zaS?`90e)67+mwy~W86SN8%_C<mh!GzS;*+_!0PSGLPgDv*q*NyjMfw&LPWqVe4?wt
zTvh;K#m*7o>tNpWxn|=!!{-uXyZ&PdPM7pXTrqkkvR`L$2`KPS*b6Xm%hVn+BawL}
zy6}ogw>+7{lTs(TBf-{hYikSFyCDE)*?f)p<cj~=IPho%4ZMUq{xX?8!71;}yLa&b
zCm(eNs^+9Yt;R8gblcV(sP+G&pNAPz{ogVf$@0qygI+^}Kq7m=p^dVGne!KIX?*^{
z&&5v9HTW5+-=Z^=V<z$^uYc;7<|8Dq9CQ-Wqs)lNjVsr|cVlDxHkloB^~&o3TU+K_
zDp<Anb)t2ot3=?CR8)@XXj%T9(B{3drXX&$cD85r2ij{ue(^;aZIo;`zk?ik{3Gzq
z>sSiU5Dz3@t>U(T5StKk3HO!C$PGrbLL$9(DwH*9V}bDvv#Tp5vo)WA9C><(Cgr9j
z>Zq`13Pt)nqs-M}nW-P@Hho$0lC0U-?)12Qsp0ecDg5V>cY+dM9fWD}8WO;}22xLO
zZ<ExzgCp1$z*TJim!2lES;(n8@b(6Pbq#nPfq+ae`vDsRqj&dw8npj_X4U)85BLh1
z_4e&d$tVN6t5RmIQMIo56Ip5<N_0Z`*fW5R$?!my*9w{*Ap60;3GQO@L@?Ed7&o8m
zYgnlD074}lXrRQ*!~i=JnL|QAAUx4s%l<rUHEQ^9*g|ls*csdOHY7&qi)iEUJ{siE
zmx$ye+4?G9yy`sGlDkuqCf1Acq*!L*QdP#fZHD_+apbuNBz+<eIXbs%W(`hdjn1#X
za9|Z05?i7Y!A<l=9D^vGW?XsbKW~cS*$r804<;q<>!aqJwxg1Ah-4V#KLRrw=`2Ef
zQyNxYng&;?l?ErQ{sk0P{^`H+o6c*Mb5PEqZTtfS?e}Lt?cVaN8h`LEst@+(_fI`i
z1N28;>G^RX=qz)MxvX_{It|yl{ubuo;Ba!?_iHx<RzDDdf_s_a{qEoH-Ffg@8*o7)
z%WJlPAPWX2rWKIJ!QGv3u{lt0zDrLx67YVqHu!kKa5Fs!1QCom(I7M?{Vk5Bu9Kl2
z@$_n9yWZ6qvPSuOs0rDDMf~3)zENZA0UwRe-=%g2N5_nV-{uNFulaDa`Z69+>XFkx
zHvDZ2|JvrSE|f_wk-ji>DP=5tA&PkNq)Fn~f<uW}rJwk&esG%gl(B%J0)MuIDsJi*
zGLp_xbclt%uW?!ArXT)tIZ^b_0@DSn#3T;rYiy>M48(JW)X+BZg9jPi-hx1wc5~rv
z&ulG~l}&#|zWBmq!Qc2{P^Py^c7y&mWnU{?ZQ4}QQXZvBb<Ri0BB$9C0hR80!~OB>
zueaLxEwIAcbNU()g#z<gxl@lRB?-wOD7QPiQc+X?0M1L3tvir>*g9-p`+C{Ay=?;u
z)pOe2>kZV_asoqc3)KF>)1+!P@sB2y4WII;Xk|TJ2J648Osp=z=pK9<|9-w6(AXrB
zWe2*r3o>U>j~*o-AAFFcVtBbRl28ariRpgP4z;hMs|}5(<Yeg?$7=B3kxQGLRa0NP
zNaj%DLuLhWtWBQZqqb{~Nd}bQpQ+*`sCGnoP&l1iCb4psmAut)F;1|67@(*kdTU9^
z>k*>P%rGjch;IiIMKeLbK1uybGGVi(XA6~`i+Yq2eB66xN>~6oS%Vq-_T1ka0Hmj6
zyU{3*eWBZ-9#}ZEhH=D{rq-$z6>}Uhl@(8qG<6f<Nn)zZ2f$1O%#YQqeV?_%r$C9A
zf|tpzMo|EZEfbd0{x0(auNB3E?;L+hw}lALyENXear@&>gv?47lShPSXVl3ZmFP_c
z#ps;>9n3|98nU>oy-ll-aq5TMUa7MPekD2T+-J)Zr|6|;gfXsmRuTxsGRn%5qOGXZ
z6*Bzz>ChD}q>tll%-xEIAkBQ18GmxP7H7YUiBDgCC>mq_qDzl%N_zRQW;bQw6?2LS
znwckw=k8i8yBJCM4dUxM)e770c7N_+^IUUXcVs2mm$gKbt8*xgU;bI7CE^0hGMirY
zk6sIWO<kb9Rt`#9+`6~%0cZ?}AHa>EHD;>$FGGEVR)Z4DMy^4QO72AfD+_F$cLGsf
z`t$|+dokRrU*?OPjkftd=DLy3q7g55vl{Hx`MHg0I#U%K%40t&_^<@Ki`Flwcp(Q%
z>FWOk%f+K!h_bw_Rm!2*8O}fQ(*leWaJ$N+jPZQzu@f2+84_mw_qYTg8Wd^3E$)=Z
zW`Pz!AE1`;ll@jh_n&J8T}rV1*5&qB&?cEp6evk|DCas4s@0+^iXqq0rXWK)V|r)0
zlmmi$AA(gN82mr)as@Y?h-7A`MU%b|A=mpKoB18DjsuNK>&>5!38|?Hn#6Fd28et2
zt(5cLND02@6FHzBDH1Nv##5{nJc_dr)E0P9Qr-<s?1kifA(wf&RN#fc!m%_jj@F|5
zCn>QvlmL~F0z6m5r9O0{DPopT)tdRI<H2+k)&A-aY$Sx~94Zr-k!9fO=r^!Z<$&=;
z&$~p}#_&($g4q_Z+1j71_rgzRjd9t?pblLUFywtbhQ&W9+_zjHv?@t8H7jLhvnaY?
zsdVTu&?+PI!z9dhU%QC8s#x3n%bT@>6e^KSU_t)x-_O&V5Z#Ry;g39+`rY{OLTmV=
z8yr`-k$9i^6LJ(Fn!~uY+RBwa&5bcoAr8B8XI6AdDhQaCBTC6xRB{f1`jxqi+7pNH
z;|^*f5yo705k$b9WJp>bFQ)s=S*jLNT7)Oc4${(2j|`J6@R7vvC{W(OIz7c@lE&D<
z91!K*mI|kV0hIjzG#%8`aKZ2eTM$?601jrwIh*qxH>Ru34K@~e`bd#r&UuBygsvlb
zKqZ5dE#WU$1EJP;Oodc<dEw7}&wjfKd;s!u>Bd)nz!nKCk^fc1e8yk;yzUcN_uKVZ
z6-RJ-^t6_(%FE9uAJH?go{xrFLjYyF&ifrnNlLb-s`bV2DW<X2aZjEfd((ZX524?B
z&MBwN!hZUJC-AeeB~~)8gZj(&cOhsXr9h7Xj2gsG_kx)$F>n=zxlxn@3JR#^<rI`j
zLjHNMOH0;lWmb#6esIH!y?KdW;Mm~McBL%&#n}W1+Tr32X66{+F8f)xHSJr$(ZnUi
z&Mi}(WlSJQV!kR202d2K53fLA<b6^AjXUbs_it<|<T0paN9ynMp$FooK#P}l<L$rh
zbpuPm+1MO2N`G$<{P*jgC!<l+3#)@_Hq9;<%*RY*hyvo^4#r42lwK#1x&ju%f&P~%
zB$mZwu<=qi%avrCQ==(_5&)0MMw)(XxqA3c2c#$%#EJ~Sg(BsdrSs3yF^ojkf%toY
z$aUpNa-b{0xN#ZZpHIwwha3YASx{%<z`i+1rx`C0x)H*^L9TcUw2xW14n{V{b({OC
z??t+?QM`0g&C<4XHvGnHTsRAsHwrDm41~<IP9Q~pE}ch?-~4+#0ui}oD^+~Y>iS9I
z?kfBIa3Ne%lYvPINj{bYRQ#Ea_B0Pi2F(2;HuLj`0OG9$CEh?DCYasRNWlb|iPeJ?
zheW)!F4gQn%4{#TMu|!xrGa1zysouR^4hw5i7<teor9iESAsa0O^2wh;wlaXV@?gy
zaaYQm&MRgNf$xOZTgBBcsLiGXDMb=&uDSCZ7hcxQ7ZU3$NvST{48EO5tZ`xxZ$OcY
zL>sj{_?<tvSm)nhHHf(iEU!ksiy(%#v{VX;TA7*s70ik*)x2QyKt4)tgk*Pl2j>^Q
zC(5DQFTNRWF?m@QR`+(C3J<}ek)ma@LfGdh(`46s|8FiE=wJe;pfqrknqpTYAFP_j
zjO`5CV1|USzL8rELk|qZ)x!O5>%ehio?IAm+6+;cx?f*uNGw4ovr==4^?u-`z77z4
zoN<vWrcl+c@S{?c*VWL)9izpgCm)Z=^A^9!Uo+KBsGUEmC$l!-8WBhwLyS<*GT<~#
z9ILhzKodLu!Dh3Q$ORmUB}3$o2aN~8S9TIhR2gz7O<j?(apI%P+R5|~W)a9n13R9T
zAUl^yOS)dQ)BN>nL;HJ2^RJ(4r()SKza76H+0kzrvf@Wcy*Rp!1D-r4G8-$$Wz^uc
z-y4DlQo?Q{w&&*Pe7Xwlfs&Z#y9Win`!Bej8RD5Ke*RicgzVMQ0h{<TY*+coztEAM
zLe3>#(gGmi8g+p@MIR!*W+C1W#)37akNk~7?*b9vl9|Y3rMS<ST&M))t1Gf&km5!R
zlHnl%Q_{schX`g1=V?08@GDD!_IPgGYGntFH@)*-2-4hl+P8Oi>)u=LK8TOWcCga3
zLTzE`f|kGOUXjC04UX2d@+@Ihm9JQWj?KxFBUo!)g89U&OHpP71wo}}voEsQ`}kS~
z1Fwd{X<y&sPjdD4jws&;W<eZ4ZxmS?p{xX<Z2wB(PXSEoy`%mjH%C>u-wazK(h(*_
ze19bseqK;KEu$A{<|kU!pLWwOYkCp%5MW(?=Qb)Qk51>rtlXHyPWfXb_QQ43mF_KP
z4AWSq1F&T!J76iYFQpMvs?27qTR1MK+)o~tS^cJiDKJ^?E=zmF_W}U!(!C3ro1TL4
zg9f!4N)LHn#KSV22Vn)VKEWj%R;oI>s=|U7c-FRRi4OQsC7gq@4zpM0>V}iIzITX-
zN_*;e_~@pR`5Q#zS?{Ts5(v?&-!UtfFgFjl)kvxrcYMp;w8D$~?v#M8Eq-~(S-Esg
zM~OFnEMAz|funT@EsCh1B__~KNW^87kV&(c{Spqe@mvF>Hk-E}#p>OBapmPnN(7S&
zT-#%uF8hdAVC%U08e+#xC`$6$AfFqt-V}{5Yn3izmCl))J*kK@UW1DLy%)W#3^aNM
z(%vVsR6hri4ipv0V;?yCpTX!vQNsgYM%l<=VoLmt7B^G)ckiEm?oEIl;n1_{&I}08
zO_K}cs9q^(vvYb>v4=fZFEC0HQ^y)<%Dg-CXV=QnU=%_QN6j-@bXa5ez;1h9oQ|{a
z%}-exQNhFbv6}4XtVll8mu|y|C(!7eL$<tv(sDzH?iZjP**@Npewsnzn&4z7pw#)^
zj1EhY^i2p;mS+cPg<^1vj;l2|Kt2-tx@j<OJTP;7Kg4C8nXpzvNJckQ`mVW3z)wu<
z0%Ij&mSM-1mN^<a#b5GLEpAE!EA(#8Y=Zznn*R3JJD}du&Zr}nHNd_^wB{Q^8MrU{
z8j^XhK3wo3k>EAsyxj_50AlxiBFSGkVJaWTwF<cC$}$~~a{|~g5do#K8&=NwsRuuN
zq|QanNwx2<V-`;U86d91h7o5m(N}-q$eG>ZTWP!~J!jRv-r6cZ^1|Lo7+Ad~eXKXB
za(=aKRh&)e&Vd$Aae9M6-iR2Ha^^MJtAcr-6c>fjOpg2L(x(#6^i4CIx|FeE^I56o
zi_l3|o$-R(Gl7v%XSS|oE^I=dTG9r#0BSWBY{6Jz*Dc4jI=}b_jABRP*PUrIfpq9+
z4#dFKAwBc-$1+j!_VjBnz1K7g_urXDL~uZExz^w!eVEkw99HyXH;<_H`5nd2=KANw
z2ug(i7FiO)9zNz?HbY14o}Ha3$=x54%U_6LfM~_M!ivYSK^%hw0Z?VNH2o;vULe!V
z9R%7aroD>KNDpw|4C|(e2;L~RgT}YbXC}s^1j{X084I0ew(o%|z}od!5`1e1hcxqf
zQsv$7Was|Z2gnNACl>)aHJ}uF)58O_FDa#-{yIG|@h@ySb1)xe=#&k4;{Ga>l*L#H
zVFl=75a6I96YQDw-3Ol{<21T{gFFzgp%U@O$W8Ha7sXBceZ4~!<;9w2EJDHwscijm
zNB+p!MK9P&fJ605-EbzKH=Jyu@fBlEd4)6Mt{q(c^RvfuSO|W16tboC_1Pl>BPa$P
zD1Urv&9hCq(i3wKvU_ODtL4KoZU0Oi93i$|aH?U-Li}*XMe>`J!ALfN{6*^2rv?2+
ztbYNby@x!~)od7lU+ESF<X+d$Gf?)hDAZ4gWrcj9APr}z)(=;06G)5+w%3zq4{Uhq
ztFz)Le~D5IvzYR5eIIFm>?F8kO47>$rF^XKkb3Y?f<T|JkSfgk3E#v`QjZ-SqLQaD
zKID+px|pJxc~did4{Os6fu`%FvL?nPVE)Rffw`wcuXTCn!|<4`h0N2?7Dt~ohB8~}
z9BQ3DOR3QKNPWC!YYeUzem=-viKi{LK&m_hNZwK@Ne@Fkc#x83MN3W0`d&wa?=U>;
z#|^zlRZ*1vf^z236OQwi>>fgtO}U)}J^cr_+Gp!}!u5hKDB`bo88;N4Hx*1x!KSJf
zVBCud#8miVIwpg2Ozj*`y2{?j?%ODg4T3@^7R~vuNTmfd08k>R+nX~>Zax_wa2%a-
zFvcM#NbW#N^&PcP`}F9yLQ5e^(qF8s0aP6nLFwX<t&lz*_&Kby+T4?LV{ELvLJ_Ip
zwrA!7X#n<3JLFw2@oyC4F~n?!S&niGSYowqtG?o{i-HLKXYy$20hhN^xRnnD6_mms
z|0Pr6g^z~ihL6kfp?SSPk@C&jGP&~>Le`>+vSHpV!r1;pT&M0DOTdf7;UJ*K|2A~~
zlg~HWAcv1s-};I3Q~i%g`5bgvoCxHZb?%`75Y~|*{J`k|E}UOhw2!2gq3-gaQtGxC
z$s0bx;;m20+7MI5<}!(AeYhG&{=L5CA+?>~{f6{vVh%DRJsDW=II?!}MQ~{>QvIl>
zZWBG`J2jCi^4(1{k>vJhPn()h5%dR+yA^n+ml{Y>vSCR_!?`-o>Ez+B4e#*9Ah{|b
zsBR6|2m?3tbw^lHQS{$moxLAS-?K63U@75{w-jgFNm4tXpF=4{XxP~pHM)-1*}~`;
zXm}JLlG_n)qxup$53PZF#R~RX;agl(1!F;`t;1vugZjDmiT4kc037$vQl`EjJg&4a
zqw!n%+SQYo+Jjz~5?vZWeh!+JxP5t4CucL!4<MbbLPr=w@8aEBtVNX>FNpErObER~
zE$XWvttJ4y;*#37%1Pk=U;zsS|Dce9Z?@+B`OmD0kJK+#XQpr76&y?LMWp6sLU0;u
zZM|t(izEGL2vdz(k9{n_e*(@m;-@plOx9v2U_~2PJ^a44SMw#T>O*(N=9pc>6tX6Q
z{3QXAb?ISXr_;A);XUOu<nz;aJDPP5c$hCc60FQ2SkXuQ>_hZL8ahS%hn1?y<SW(R
zBe}`!ib(8<>@I87`RH~3_?1Iqux=JFd3wLlyv3+-1W{Sf<}f+vZq7o{tx5K1M7--0
zG0vc-N0k4G+%%}zA}W)G5b~LE1o%}eU?zR>%>xA;CNT1fDMv7+Na?5cLGA-Ouy@J^
z5XNbU8<N!7<g^jYl+dA3dPx*LFFju?TyLoJ5keubY0Z_W!yrh>`FUJnCZ}NJe9Wv&
z-baLSD2xpsk!8r~Nh)X2gB76}^B`J!u(n%ZnUnXw(AIeaWYsI@qfrJ^N>3pRV3QKM
ze7~Jc$cY+Im+bWkqD)7KwyIXE)Ry=SWzA9^AEnSc7DVt1Q#d!|C>-k)y-)$s9C3Om
zdjfsrA8pU-d}-(|19d}rN9u%OgqzfamEc@j@NdSBrH_0pg2Uu<2>wF?f^z`yMkFt$
zqIH2j>Th}e2>X~ff5e6PAwnNpVdc+O&GIXXFmJ^f$~iLw45<s<bpv5kRb(ulwt%wr
zY!*h@uMADSr=6<5#Z2X^>O_8_hwI5Zqhj7;ymuiD`QXg{{t_YQKpvk#&um;iPAMzK
zPnK;-wxbCrNh{J$kPsU3A9^OHE#K!-YgoktOarr?D13Jph=16C00ui_6o8B{{vx+D
zV$@M`7|-3uiW-ZTi5vh)G>ii;M~}n%xw2hH3<oE$$*{8_g*!6jV3xI@4WvC%I&U2X
zT)7XP4i8_mARBW<pC<N_3-Xc|d@G3}853}PPwELYO<-UpJAFhP42fJ2bx@n&BwmLS
z1!RpTI0ldPm6S1i?VfwIoEIv+fdZxzY^M-SFj~`L?F-^*ogN2ijr@5)<zA|DjzWos
zRKxkNEt8Ka_T2=%IWpISSJ8~xXZvI2X3ySU{~5pz2U;APFL>V3e$1<P&l8Ig*kA0G
zH_oV&6bZ!udum3$FK-TS!81D;HZ{}U-+SXJoDsuBkzgj41~XKoA!JoqPhzw)9t3kY
zQkLiDePWlvC#_N!g)ml|UQ$>=g&aPTRj48d2^M`$H00<7%H*YxgOROyw4BFoWks^$
zzeuF6J4^K6&1A^<WUe+A0}_N(a)e$JkP>=vKx6L5``g^qcM_AdL7gl<2dC-CA=ZD_
zxfd`{rKALQ&>WlI`9m=o-VQqRBR_1*I$ee#fc0+p6I&#@n)G{yF+b9NGFgbhBabE~
zOmTj2hmlehcDfJzDmPB?TQV#K9k|$Grk`ArakzPz`0}CeinO1m;jhjl3!@&dq~D7h
z_c=n8Qct@#Kg;t^(#~FpRx-ve!3%`Prm|nB$T__E5po2R6eJX0I1luALZV>ahm(*@
zFjkNX<IqpYfn(=<)#5VOML77*5+8YNWMbhuv?l*CTmmNUjqK?kmP${7kyb14ck_=f
zn#xbX)sSETUdbzBcN`aa&L_JLT9RpCdMdi(S13iAg+yleS*7<`IT<>x6pzw^k&Nuu
zBg0_`5JuopWcha2sOVD9`#D+tm5IEjC2ADJS??HNiaX1XiWy(PgvV#_Bm7|QG8(%e
zWw;??R7s@XKW%(OX*pt{#4*cx^*i>|<%DsGNZ}0gDS4wvi(@EhHNJ+!0>dc~TVX0-
z5s*El0TB@cBt9!_g+gT{?-t?aPcVV20?IgggmjFTGZArqLZa@^6SDw1FeU>x<r!pq
zhsI(#2Hk*fpNv&4C^t+5M{%){+<g_7>&s!CaJ8}G)hW>6BbuW{lxa0xYr{uV?D<+d
z&nE#FR3CoOS)9vCPAK@pgNtV2%>xJL(~lyMPj(SD?K1rwrtOYB0-VS<Kq>uit@{;d
zKr@1o0APX;6#|%71q6){3=B(9i29KulmwV*?-kz#n)C6SwM2f7?7viIpHKGDPl5dg
zpOmH?^1c&sTt6}~vb5K~?)NOw=zif8lZFk=*kM0mOJN}9j@k1}1nFM8DPoBa`%>Ec
z&ziqvB>Ur!ANK<*C)lmjTo8+5f`_%_(#YCbUU~edlMoRln{1iw`r>dqzjwb%c)sY>
zqs8)KyTkUck#7tsj@Bkowa-zF+mq_@HtM$$D`OJJ&pqAVM|KCi0&g9x<2r0PSkHTX
zE5qR>9c+E2on-Eq>G4us?N`kvn9AUviUjbBqWjL;#8kBhxCmNTx6Xj;{kR<|ubGLh
z6p+`BX{$@d&N;ELvB7}Q;1pyl;0f60w%>`tyrGuOM9zu-&L+tVSK=(4!+5|H0rDbB
z0BrTWX-&Pqs$n{g_=p_$hg^s^s(=B5QZ4?+Q4zh?>%tg4C!Fu7!J|)SZXCkns5n*J
z6D&t>ulQs1^~<ucF1^-^jXjS|aHTZj$6)RMXp%E<fUWP}O)RpOSm4lG7#k!fk;}l)
zy^YY`!8zv{T2)jL8$5FMI_QkT&AHG+T@i6voc%HM+2(gT7|orev@7q+>Z?-^5GPQ0
zB)bC4zRVkBrErq{-A&%*Byz%pKj*rlw*)%N#$=nk;t{>HAg(0jPSKxe`(dO$ku8d=
z63`5)(2>&{UvL0rzSz7lC6*7mJWvAl?Fv8Q-rnAp6*qVw7#?B-Lv3%)cHrY+UWVR!
z95yc0J4_cbr3Mbbio(Ga@CBTPPln3z<<tou#Z_NO$(f5r?G?9`<i}0pxA5kaX|<i0
z*606=_MV$`sHYk~9%^-BmG71?HpSKeT6#~yi3=>GLmLs`)-7^*G>-A<qOKc{jp4X4
z(YZu$HTkK*hqhNyY>97EpH-IC<_hLb3V=CrU?6LtS<o55Q2rWun!3fiIwmR^Ui-9M
zg{4IIFNX6c{{a+c^vMdl>wzNsLSvvOoYlkiT1dn5^jOzbo}ipPH=v_LB+YVU&eb8^
z>X|T4OxJFi=QNA2d9bPCu8s!q>!P3D;eE*lkY+AWwWAomh0lR;07-7M8qgt=n$8D9
zJSAw=8z1429}unr&ISxjiyrma+xN)Dk;4un4mW%`?pI@79fF%!2AtPQqn6)XXwr{8
z+>PZPcde&wT9uhFrP6ge4ELvX!sj&K?o|V)#7q1PWA=DPC5}6h`_YcXylp;QZ#ldP
z$Bi;WImP*h4Mnj7X=h@~%jEok+nr|^SQwU<xuW0P7ow_$yKGu_(1cO(<VliYtmyJz
z1>&wdpk++V&*m9cmt9BX4YL;lud!eV4(nI*2PGU1Pg0-27^5~ncd2-_gpqo!Vw1X5
zwa_v%3jE?}STDXo=A6@@6OM#jj<BmT7-|J3g>e}-VMl-zC*g$%#SxHR^Eu8i15<n?
zIxRqkAFCZI=tM=3r%Cm;T9hzTSRTK+Q`Cn)!)d-*{CLP9IQx3x2nTqUw^ejpKBWE0
zAozh8O<%<5P>@iNZL<NS52cp&2q4hNxNFa(6ZaO=VI=dzQs9&OaYuhi#l$P1$6R$3
zZxH`&$;rjl>HiNBg-XTYbU?{TDa}<UFaO42R<3w-l2B;K;<x~bk$xulxHc>H(AUoY
zwsZGt_yD&`3h46R-v!csvi}ggyX-&lWBEAuyh-|ib0wY;yX3E;sjtMl{FRhsS59`G
zORG9I2;v45mQB~uTUvepI>?%Jl)Jd~c=fAD-4{lR$VgUX==k@FqPN&QSQ*)&8d80h
zdEYVVPEY5{C;uqc2U@f5FOV?A6zCsw{qmRadQrn7N;ODJ=UP8UAE6~uG~O7LGO*h8
z%GY^UE}9hIB4Y1#s;|vGy8#ym>Dbjf5mwgmk9JJ-VB{oxjwgKF9FJ)m9DfCq5Yy(I
zY=QnA9yoxHi2}2tPr(f3I?(h5!_K-`t-+jDEv=%$Y`=Dzz?e_i-6%4H#_SIZG$>b9
z3I;AMimNi;1Mz6*gT;BC=QDiZyM~C;R&cUVHp=TrbSP7{b4Noko>63T@|$(U9IQ=9
zt4x$__M=bjS6FkBFtE&>O0qZl4Kh-h?ph{dah6wMdd%S<U*GpP9{kcxD1u;r**pk&
z_JHi(bFkL0%NvOwb3R-Y-i%pYSIa~wNnPj=F2w%H7Nv>1U~x5R>9+tI=a)gtg%=OW
zP1w|#@$`EKq&Xyme<b!ystrRUuFAeL5YJ}xv<VHG_?o(l`HOa>7qK)7jqa|h52APX
zL(m8~44(fYSO{ch-=dXf<LJ6*C||p#p`RwqguZ<0eF_LAdSfmyt{?^G4#MY(X%M5n
z8<PhFeu5k$l)ZH$jtZ)h^yma1GiulwBn0w4EQh~nIwk+|>PnOl6doLo&er5bTcxyd
zaJ>2u#a+pH`jBR#w+(%y*I!HeVtT?WBR;RW58ZO!Wy&%XV?;;_m5{CYG${1oSt6Oy
zbJVEX)XLX0U4~aQKshe&D5v@5<`%Wh?159uSZuLQwunvB9I9btP?=mZzWFNC<Ijr2
zzr&I5K3mBR>*ul7+5RM`cH@h^iF@)sm0?M-o@&aVM_#y9iN}A~P;FDf*0RtIvmeKf
zDa+tH2qIKmiPZQQmg|c#_Lh_obI+h^+Waca9A>L>HC?BDOYr6*bFvFv@lgt?UB9rj
zsc*XMm64X+LBO8@_3|}7tj@2AEvd3~JGq<b{6ufGp$!D-VD$3z7j5m`4KVu;Pn?&V
zJLNerk=uNjJAmns5iB>8S{4?H5{(o@y8pf}#vIA0E!oZ#I@R8L*D)C9tEbHQ%Q9Ot
zUrTPFj$s4ySBMBk5;k2`2o^m{T#TM%&=h|wi?cpSkr5X;Tgj+viAz9gX<FByxct_?
zsW8Dff)xc~D7K~S0uG}t`qa$tjJHSI_nxNjZ{~#eI>Jg2A=0X<5gfMkJEPU7oTCQ(
z2$4mQD1umDZS#Wt)$RghDC8h7dGB>XHbzDvLhr-=bZj=%K#`hok2{8*dWI^@f|_AU
z210!pPc4Kl?^}95JRn8%=7}XICE~t@Ac4HfSbJbqa*p})i@z%xNRROvG)E5MDc?8a
zU<H3)X0eo#5U8`uOozRYZ9`NsqD|O+m5A*Y<L)orU*i+025o^jx>YI2{|t&CP%^pF
zWX5aDQh_ovw7ijY*3YD`_)B_E{aFj|C7m^TN$prAA;unCqKur;NMcnldnJC;^<uyS
z+V)gkUJ|c)KrD%_siv+5aimO)m?xjKIz81vx`DF3LgljXr3ePa^@6J6e=W%#OFZ|-
zkCfrPPC7{nNiCI2rDiZ%42BFSMqc|jDcz6q2t^5_;X-8*xkx23HD&I9ez&5M$H^%J
zuCdhQZa?frUcBPQhzC6-Y2}i}<}!1BD4wnRO2W|<6N*8@*{SP&PLBIy96lySzZ-qv
zL1_Gq94-ra@<5QsH^(;vna3z?v@2dSv9YrwY9<OKnPruDgCC@BNy4RH5PIqyTElKB
zXgZHA=87_)1T?ukA|lIvqGQ5wAD*UPGg`*J4rL0Jc#<1CPG)JaCNkdC#dD0jATJ88
zdO~M&m{B(i;1|!5j!AZqGkE8w^o<j{;28B=c`XXJn%$w}<m2wUq6OV_;}7cQp+fUe
z$-IZBsCkM5?-V2^naR};xx#;E|0@&WzbUoE9APgg>PX#D!O|p-ER%Pi`rVw};Ob1#
zc}IOP!NA1eWY-Pz-+2Zp#D8_IOc=iw6FiV&An<Ez?Vj7o3TtmZRQ}=2ahfC{CJ`?@
z{%vM-@JM64X}g6z%Nod{4inLfKg*c5Zvs~^?C%p!ll7&se>d>iS7USkgwsS$%Sn|m
z)UnzV^6=7|bbUO#I$|;ThMa_khB?D0g}&gZUr@5m?-3AlpPrIf^J)0c@3hiTb*-AP
zQ;*LL4J_4}dk%(;iG&M14wF6@uKnH=dk<8SXrHw{`coueP?`qY74vpR7K!LdLlqI{
zSxg3UPCT;ejqrslO!cj=(Esf}A*d`)o{;Xtx*x-c<r5Oh5)ww$hwp{+knSJ(+T5u1
z9{TjjoX(#1mUy*ZUVMrkZswM8@LQIxVfxz7w<(6*51H3KzrZS>KA3qXi}t9xwodDX
z6MuBBpS5s)X<YfG{G_lVY6Vqyb{?(e@y}zfP!ouBHW%9XGpweq(Yv|J`*5wmQaE*D
zHAZn0&04P4`Bf9#xG4z7S29cLt}+5{$8q#$p;Vz*U!r+g4ys{x)y|5L{yCRzlY8Td
zunV_&efL3USNV#rJhfiK@(g^__ND>Rmhs+m#bueY=drhQKVc4P-?0=KDyV2q2~DAi
z-#*0**t01>GP2_PPtUjYV>mAv`w^?Sf9GcjoPwPdLt9D>k=tnKrvTHZ6;vUI9H|AI
zTs11!GrT{IBkB2V*>uPQj0H9$1#W}!E$F7GI0T9UgfGuhR2ol0pPi=W3F+v~4~NB8
zR`g9&4<6j^g`T0|SpR$rIsbFezxyO~L`vpi*K9;M-wm}`!_|;>NHJW%v22dNCMe;l
z`{trD^Wlw^ruoMA(Anx+VkV9FhJA9mq+-SKr8D%OJ6-<J`=4wf=&RD&#fSab1H@H_
zan{$8;Gbq`{Np)ry(L&XCfvL;jbiA%*i%kEd$5qa_H3e>D=);q^po^V>)2UbeNPuj
z@So=Cv}^!bcZ#47SmmZRZ{KNFc?Rf*y>1ENy}W;HY0D$w40N|TQ|Tv8*Nzx3^ZYl5
z{Qk1nNqKC~lckM3Fo@@aVf=rpWQ;}^mv_@6i3&uPe2h@Z5Cf?UY{iDMlythMRQYvo
zCas+5S|o>&TvE)xh!tz?Z}#MouT8q&q))~dmCsUq3DKU9{Kz^nz;=!M)D$}#b`@mZ
zx;9ifp!^9+<5GP)@Qq8SExM@N?6t)e8bZb9r;@=Ma|X+)?D+Bc<QV)M5}`&M$L0>%
z!MiIWj+I#MIu`goukyX*?^9T;|37;iZF;EK@yi*WP)zxr0*0l7Rb90nI)yo69Yz~p
z2X&*g1*N}|q(U1fxoKMplu-6luArnn-uyJoN`&JxM~J*M;|b4$pjc;)^7oc_rOv-~
zME>B!U>Czq{~6-H;WLrap_)R;*m)OQazQz@-BP=lKo>&tc$k!KSzJg{krE@V@9B_-
zKbR5mHcQoG?K1A*pF>H8P$iMT!RPHpsuP|tJX8;{dZ_11wb&mDB!tMZO&KK2NnZ1B
zW$0IXiWKxIa;!A;bd(woWG2%0J$%V!R?htSkzST)P{zp8SK+2-UegvsMX$mr2L~HY
z%uz6f3l@tjiTlmfFOuy&FDT-zlI;4?{b~ws_sJ4wv(`uGpfm5+;6Em0&oT55<h`4v
z3=U?iA9LGm^9v2Af<BR>39a4ZT&3^V+WD1uzk9nZu_4&?zUjkCdxZvZG-|NGX;~v&
z&WFos?%NUYt~1U4RGhvDDIv&%q-?RkZO$Gi97~EGCog9yd{A<mlIx#q>dRWNx3I(a
zmJZIBW!I&Z>5=q$s1^R>Jw<<+Q#%?*0~7k)<)&Q%-iF6$5EG)A%ZEIjs!kB(WY2o0
z5%Mg_gC`PU;*)*4;H}!;d`P7IMlc&`ByrUB6ruu!dL8!_k5k1S6w9?Fc-@<r*)vh3
z?Si>D427MAG$~5_<(_!J{#b04vGvLRrQEpymvH7}Cz<{+mb7NAt-tvDk!LK0Gp2Q3
z^a$kz3;KoNYjG3u?y%^u`#VChTJ*vw_Xn2{D*Menkx$@{X`v)Z&Q9`$-6^0B{iFS;
z2f~UrCJuqA;Sxp+QiefTd^#E@*>Q<=ut?~w-MRKD&t1gkyBcXUS$bF;0Uaf#3pvgu
z%@1Az#x{~K!QWpa^54i?VUCOrO`Y<5qTGMtPPi5BNGjW;4dU+ef)Yr}vsnK19l+}n
z3-TsYoG|t9Ay=4S!a8v`kFiP{UTq_LJjD9);m)f+Ra+L-tT3H&QTpQ#mg@mp#@}QS
zXl)V8CmcqxZ|fEJ)3DK}xG?RL|HWj5xphDAtcD(kh6S$?pYhP9YPR91C=5G9moqIf
zOKCQ;{&ts%Ff&(a@p<-k+huP)OuofI-6a-jeXW&x?d=YHm8nX|FA^4X^75fhr92*j
z1Qcx&WEeF3wvsGSv9v8s+jDyAtqLYEWzN<W<ZcNx`78Em@JRs;k9`Ms+U%4dK~PFX
znbmW<diF;cLFa0cMh^_=KxpP)H?(~n<s||OuNa}DT+T~oDdkQ_j0;&jx@e;L%-ram
zZ(|uj?=$sl1{>B_WU1HC`LXf!o^kUAM-}bzvTB*fYM}W2+wqxf$nt4J4qB6Hzx<-3
zduTCRo^(1JbiMv?^<pdec~SH>-is(yRr7T28oCgX>_uKL`QGx%(#@^-(NG8JHwPz+
z4LYcW%J(11w40AO;X9mGXhcqpIrATX0*=f16{YT@V^u=dH<{GTtfD7D(QJuvSp`O1
z%;-({i!DUlpP%L2T?hHetu>U9_}dQjtd@yjMAr;KZaw{|QY345jl6Jmf9Cg{1lN}C
zz6rbVX7fDS>iC+VbUyTYSaOMt^7LUSx{SVG;YatzUNC<j<<*<`8uL4rPpHR?Tt$vq
zIs;O1*u?QP_QzPYI~R&dP*f%kBQ#9he}&#F&yp#YqK-*JhTikrXTEvUzZ~d|mpIIy
z$#?CvDTXJnfD3#8^g1oHPpN|r?5z$;?9dfh#joCmqRUzR$gHfcK+d`6W<!u!xwK!@
z4(v<U!lAMy^?a9HmAcjkH+f1*NyF)c4A^aQb{mhI%9oO%!m(+1gWVT@AFLmloce>y
zFl|O)alD+Esj2-5uMWMmCc^)RrmKLevJ19=fOL0vr$61D(%mWD-JvwnA=2I54U!_#
z4FXcq-Te;lt;bq-EtGq|`^B7@IWv3qCY!J32G6<C$B&v9HBfr|^xu{h_uXB!wdsXX
zP#Y94%m8XT20qGPoh!#e4yS8;Y2cYM<Lml&V336$P$QGYj7N_@r%yuNo*8W&5mZrI
zjGo9@y*4kp7BRr3;GC^gu>K`ouiV-QC)otEUtd!eU)f8?sW8F};~e{;CJ8jj#f(Fs
zu%7J&_Q(rKeoW^hONdF7qe5Wd!w&8YOnGDFB_7{3{}!*qTwzEGV0EwP>FszzJm4Ou
zIjpv>b2%X9D|S{mt+IQ(bKjihQ@u8zQ<Y};Y<#ZxOSm;B^Zh?^CSm;#{~lYz^}l?D
zK4<Yq?dpzl8YB@wd=2$HSJ(F+O7OU!p^M#<6OU1nj8S?HMpyXw^2iRejYWoqeY?+{
z#CFvlHtwn5Xys#o-EF@fOhR7WudMDl!=?^Ge&8e2RlCj6v0{I^$E2w3#9?@?wT$3r
zWf7_TBE|ZX?~$o#Q|A1?=T1#K<%NW-aG@LxBHmFK`Q1&?ekEW(qU}<&oT%gB%L#M5
zR1Kjhh)KsH?keW-h)s--YB-eyVgH>><2F&^L>r;61M27ZT-;vyym-7!&|;Wy3Y+Wo
zXs@Y`&ez1aS>KEWP3H)=L$%(Kp-)Ayi<Blh-8)b;4Qi`i20TTpsAEX4MfolwCo6O%
z{ar(b<7_|Fe!i}-qUSBld_6l?N(G*u9C&__K5XAeb!^kIy&!_X?PAGrmqJ~QyhLKG
z9ygfaEJjG+iVJr_JZoDHV1f*_L`NzX)LIS@AsP#jJ~%wQcNQFi`fJ4<F)N>)doXIr
z{`p#aMlT`Y+Nh3-lq7Qd=mUY}qY$3&LarGaQqa5uS>W?wcwu7%S6QM|$lvbkpn=qx
zt=h!>S9m}-@ta?0puvkFVHpDyJ<5z;-7#OK3+-^5);r-aDwhmK(#}wtVM`h~3*J>{
z@2c^?5{KAG6ttU8#7GI?|9JR<JKMjP2hk33{=BL~+*tTKCmYjL0`ACgj?88%OotBk
zrD~mgILyp|2s~_DHkfT_vLey3-@9c`oAgLMxbmu8)>^{h6}xT19#rL)g5WXBC5I^n
z2Y2a}rN2;p$6SMv9BkM&IMDP9u7$^<$~!Pc5b_R2A(k?&;}=np6}k=BRO-l-ac7_Z
zdLvymUNKaXZ*<0Vs>ByPAIb1M(UJN)nq(lI2s=*3p1#O&?XzS50vUPrD4nkUt6Zh9
z6jlD#HLP<7vGh?DWG=O3lHbWQg!+U}?896`ug}l67!`deP%}GYdcw8-CN$?HL;ZY!
zTU*%l!>BfMGB`%!S9E?hKXKy~6A^LPP?+i_M&zLnO|U_R)wGtu=hQ!mv(Y5d_hkVz
zbDt079QZm(juir#exT6M^iX|-<b{Mr)Kq@?@ows?4U8ByKfUiE_u|Txw7ksg1|&he
zPny%n4NXFwSg6xI%hy`c!G9F$ozw)gA%!xBzb@`<&*vM-_Ct#@-5qfGJYo$e5|;_a
zvVv=g+F&YAZrMCXY)&$H;5(WKS)b1*8VLlB;zK-S=e`B@b!@vPhmm)V_Vgy+Q0;5=
z^bR2x0jvmd4Y!C*d2}7FTfRoc>z`%L>QO6oPmHh{ns)3^36gX%Rx&r^|EAeYE3oEL
z#15#H$SBD_hZ2@P9XX2)u1qF2$hQ5>qq(PmD8m(kn(l4cunn1$(r;;~-MIozpL8BG
zKKiYxZ>d0|TpJ~$Kn+!Ev7N~C5L*zYlacnM3cLU`+F}<JKHT6#G@=4V84H!0@nswa
zTApt0KexD>Rfj!pMt6{bMLxSUTewNVmEy|!5pqWoaN>r-W&0bh1$r`jt_gl_z4)Z=
zcaD^BfMMs|RlvQxO$6bm_fJLelJ{zvVm_cHMx|?FyABX)+t|!#M4`iD{2@N>5d95=
zeb!o{rk{8G?$%%hngY}TPI9B5Tj=L8W;hzI=M^&%Gc~TBBKK29kGKXkx};>v)S0$8
zLLp?creGY=W`$1O?lt*g^~6-y&XmR+y2GMeX2um`4u7#n%W!nV%6D^Nw(tjcBXV_Q
zt<%f8<|C)R8Z*=BKW|qJB!MwkH52OdX2T?$2MUU|pMEAlkAVB64MNjDUYU2Bd^fZ^
zPvIz1<e|@S@H@IfB%GKrR%48V`G3z3j`X_|uXMCFSs5fs|E_mKbv-~ZJNjj%pRVOT
zj9Ny_s*|hZeS<&ucT-hEHR6Svi%QM>G8^e3OvPZ4nyi0=6M8=TP{)Dg-Kfp^YxIDg
z;XGmemtw6Kt*#iKjhLFS@=i(I#(HN^E_HTQb3>`}r7T<2K2f@Cr(smXx;BQi!rc!Y
zKAcxoGxPb0P{Ee>H~I}(RU@xlz`;QfWl6=eT`>qC7E>Tcnl>Z~+z3_f+Xr50>$-61
zBo(96X~%L<@GM`&d63zW>DLibEt(5L5s#4TiI&TGjroR*BB&7fQHh2AO<lrzUf;Rn
z8W*P<cw!9WICHp5AMih{+qtnb4M#(ewNwH&mHGE23j$PLWCRW!7@klFY#k!QzB<m?
zh#nb@zqF82I~g-Ztx8_--foik^NZ2#ts(f?3Q0?wnEZkE(3a+;QkIiTl6F}w;-zvY
zK&z^PIr?MxquL;aYG7q=nexls9i`*ovp&{Qx%qA^5S3;$=&=0j@p?CD1d646Z&!<D
zOT_Hx6?zBx4TFWTcdh+a#n*;(#;ciUlP`v6lU>h&wU9w3`iM;Uj(Nq-%+&L|vsVnG
z<*6WVt}0i3tVGH_RBv5r`JXo(fdW4glxjplDq%~@iTKU4Yrdizm9xRHv8=shhDh4D
zfo`p}xV3m<o7ic!!|W1f>3S1vUa+B-!cvE)nf+>X>y5Yw9bHte>gcC9cQgW%e1*xp
z$&B9*>CIoQ(BeYZk1du<2;oEXQyf>SmM1tDZUu`|duTuJ$lsoyzdwYBEtrGUWMbT2
zCVWUpKDyhW1*;Pkl!D;Jw6~nD`A#i+6vb}YXA3`}mC>2Dp2FxJWR$6|#221Y3STj6
zLgCR>)*1ELJYt80zEgnRN56{*2=+9sz$pwyhq?{A1PfT`;F^iPtyTKZ;x>E7Y9F@N
zt;KcubMm*jq!pIv*psegmSyzS6AU?y_ta=ZJ^tj*$Kko{iFGZcVuj|iU$&m&k=)A~
ze7qsQy#IuOLg>d`6C@*HYQHzt?WzZJxPbGY`J(f-<C6?jFdSetpuqZIy0C7bJb0k-
zh|l@3_x$lAh(d2E2WA=bkA^+_t0)6#SE+i1)6kM3X~Pw)_S>J^k1yNr+#OQ+9BKAe
z#0mX*SD*fygrVCsHn&bmAJu*QgxB|33XWcVo|mOq!<d|z5(ZB-zZFA<*q2Zi(1q!5
zlqh~n$?IQ%NbG&sR}Xw9m!kT}v{jQ={Vg@w{fyV2g~ofWgKBS(eOEXL?XxkN7*F9A
zsstP?r9%Jb&|WoPs3;1OcflfQB4pE(tF`ABw5Ja=G_MM2P%a2lhZS@$J)`F}8A@6I
z%u_3{i_(b;$NOL&@zz|C`+dt>`Ey{KB|J`mD|popEeU|=<Bh0#&>oL?#r4j{4-w}R
z3n~xN8X;raL|MyapRB{M@<+M(>1VEwaH7(Z$o99sVA<l%2Gq8rW$f`|Z@$#FoVX5L
zmUsJwMctA}5co8@=A+51ib(a&T_Wit!^c86XDQNlEr5!cPY%q<Bbze13`LlzCL*Kr
z_+HQNN@y-ehCx#mvrjvh%Ot$tb|jMGjcNa548FMVsnp~=SXY6uzfc;#28lt26&KXz
zxmRgrB^8i)4hbopY3=UL26Wr;-@iwW?m&^@A=Nvs3r$Z?gL!X(P@Gq{18%<<$9_K!
z!lQ;9MIC+8$`23{j+uyB_lxzFK{tY3U)!EOrZTT8*(*R)D8S&88}lsA?R#GR#t*Z+
zK`fPko1KftV(BHQrZSh65*?0*$}lJxwV(LQk}i7vXQA}|!SRkeQ2rMTYMwu<yuq!e
z^3z2MC<+ij%;76NG_n3A@mV$5o%?++TsVu`ygEl6i}J-8wF46DqVr`vk>N6;aTdBi
z?n=MksLJ>i+uATyL?(S?d=($P*t0I5wJ@=7xosvRI~9P6Q_&Z{;qA6}*Poz83nhu3
zDO`=XQ9B;|Ou`Y}*DG4Jw|wozc%Pp3<YJ8jMNzb!q$IR4#=UM1*pQ!2JbMZN88jgH
z!|P|8bk8cu2}o~k<stF<3keiZgGGV>S91Kf+so7FSA9J9lqq3oXbG_-GcgDp+E1z3
zSdMV$^)-}#&kKlDvkh`F&%$Mhd|Bx!1IOOKKD>AAyklyK`)Q&6sQT}?8M8Nd?rb^-
zztuxa`C&>t1O+Pys#^guQXMzZdzV(>S2?JXiLD;gm=QxD_k1h24(y%V6`O4x_;T~U
z^JhFMd42^^Borv@51X`xucGlgJ};Yc*)vJas6QZCDg46@G(B6noh-7T38Va8siIYD
zWstVRh-19RS7xm^WH235NX3s59(sp)B(UJ$BvFn~SJ*68H%#gwz`Qy*h6$@`s2RR$
z!cJNH@nfu5gG?5-m=n+}aoNrC#DJ5yO}8A+l#Rw=_5y<8)-%P(K;_ADB$0ZreH<{G
z_D$xA17&N^SiWMH5c9@=VFmqcOp%dveCE=cCjWKrX@gx?VyQ%}wgy)veWBm2gF5wc
zh#WUnyZYWFHKAixoNta-tRAU%ti$fAOwIaOv84y+R=L9l(dj8q&5P)Aa;M~d1~&$n
zKwrcPkrb_y#6+Yl(t|9mL6jua*Eo*#`9Q$%fPj7VAV<+uq;S+y*xWZ40AW&wjs{N`
z7Hrz`caMj{W4ylT@$NQ*QK!2xUJ}vcazg8y^7I9i3#pM}xNObT<5V1-&9f~I^I9B-
z8<wQ9?&ha?Ej?KU*DR;=GbaBW|2X8_v)lef?oVj$wmkxy-cb}*Qy<o_P(k-bq(M6v
zVqkF4q)FF|c*o=9k8B*tltfjvLC2e#pjaMS7Qd$yoG6%Ie`DRr_^P6#GXkW5(Z!VY
z^}m*Loi0l>XF8Pb_VgAjy~Zfi6!yF-G%pSP)zC%5agKm(X%)dx{{!tz$u@~4>sKFr
zql4~4=X$qd?|3oyn_3`X3p#N*ZmX+>t02{KT7FatR}+WVfs*V`VQ9;wQC8F_U0h^x
zK0|Wy?0=p&x>2`U326t#fns_Xc2MHnB9N3w%{auh#Z3;TJrX9;FYlNZ`JfEk6VVNa
zmv%jdmUx%>sfZGojC5tFOpWU#))8xHe@0RJ{_H8o)wOAC{2BZSc}Ko+kGH&BS!1W5
z{g;y5fzoNhv`ncv|EJe><W)7|vAA<hwTB74TH)r($cFWSf$c@$rl6)$WD0^7?YKVw
zuu663A(EQmsMKU^YHF$*?gk9@ap7|nT9ieF?d|PLHDb*U^jlwwcxQQND4`qSng3R{
zOP6V0r0sS_`*0#_grVa_gw<-T2BnTT-NT`BHW(Q{dOq@g6B{xzHY@JIW}H{g#Smri
z21SZqGy-z;xZYebTX}8g56hR2Rw!YRx-|{=b^#$-9XI)9pA!@L<PvqA?}Pa39K4cb
z$RY%5-tB7%g-@o_T-q<`|7zsr+-qimquQas^~~k`#_`*?wK^)dgkUG0j+LAYYdS%7
zXzdV(IX#t--R(;VH2*m~EtfZ&axDb|3Gxd;p3GJGJLsFrp-%`Akt1=%OK>v&{(@kJ
zUtL2(MNdz@#<1&duOJSO!wL;BVg37z$U+7rvW1^EP~T`b-jo>Ex_r6uDFLf<$4NV4
zkpJTiI!d_Gpgb=PKmq9d*Q{b-7(G}H_1T~q-CT-ml97$L=*7BualCkO8o5VCpp`%c
zOY8^4`q*;sP`WKVElq+k)(i{JH;jwg^2EJy^N{!mSd!8Zh*~_pPt}kT%lh(uW3c*N
z!>ThIuqd5%Fcp$9QqE<{dYHK$N2^y|cOGI**o!Xhd5Z?&U17+!(Qxm@>*hj>;hLEY
z#JxKZrhSkGg`Z7yv3z*{wxftb!_(oUfXs{N$N<IdG-YhQ3cFIzAA+75-htmTdxktU
z+&O;|<G=LO%*ZfpNm6ch_WLOgLdZAC`ZP{k9N>z3UY}0^Th(_#-<valJeaD`Vg<?p
zU&_nB67#1HlLltq9Iv<zVrvn(3m7cK`Bd^M8<F;H!ROh$e<p^QvtcoT<V|qj4q2#M
z3lKuwq{-z-W>N_eT~r&_ol1!qjO@+T9s5+Zg@H76b(}0!uon@=2p4i9t$89*!%E)j
zNleiImskEXBB0hXDS3dGWA@5}XL^P6xNWL;p7~Ry5w81JZi<koGY!?}4a8P8qv3`N
zS09)Pk%oqeX;k#;5~UoRE8@vUsWfq9=lf4t_#BnQ**2_Iy3~p39-Gl*e9OobCTAL6
z-5CT1kY({a4;DvNKG|os-_xF0C+(oOfo{Q8{b>3>>OVe*Rc|i^l8%RiiZ<|4Dz+9z
zJcsv4#K0;q$4Va3)l^N)$IX?Q4E>A-ODx`S#!z<D7~6-&po-9EBM|14i-s*59E?R!
zmOz|rWe_hWN&wr&t4Kkf_3ph*!OE9%v4Cf-MLAq_95jZEu&VYD2jbUX<?b3m!5l~~
zF8Xj9pDw4Hcr*pWb=%Qx6+IIt0^w|<{gA}#&8JBGt~bKw)F@$(Dc4Qu_2g@<2vt?;
zevy6gtC^9%hcF3ovX=vV@L~XN4v9vmt@xb2Plto}ozfw9wbwSVnHLv_pSd~^-!je>
z4%_}m>WbiUgQPW=(L3N}8GbvC(^J%L8u%Q0M>2SG8oHqExxtOo2wxG0S?ROWX`9Dq
zKa@1qWDGit!p$$TQc~&HF<0$*LyevNH+#z+)##9Ho#iAIwV%11(fabO!POVrFh|tp
z-p)m{3%MKux{qN>zJ-nARs5hZ<``)xU#>5JRBc^fmZ~y|=xAuao43%;H%WO((d8s7
zG>;GNj#8mH&&`Bfurp@-g*84hl7|n8y!udQ*6*2Mv6rTc3_AE11=OivkTT(=RDL%e
z)m<|Sjd%=Y_v>Ab{Tg;gQb=&m!_pAw5gYYk#V37ryd*l{p6I_kW-MILbP8^cL|`VZ
zsH9}B|L*`oxNszq0xPQGAo8bA`F|_<j9lvYHXn+6$w~(rrUS71{aF)9b$d;RYwqn7
z7*Y)m_vb0UM6>LN={KbM%ZtnCZQWtupQ*N5(G;ad!YSm+r_>+pX@1X#`$a6MQ)Mu0
zb*cvK!8Bx4o7fuLZ?zzlVz2jW${k+Dtt^R9Um~+#@uHR93Vy%P;07ydHPrG)2HwlE
z6(;@8-=^Twcn&60bGpx-gVF=H9aK1CcBZD&eqd`rr7C#H`)j?8loCs<4x(Hvv$ELT
z3r1DSvFyTEA&~;(B9pwZwpNeCc@Y-6qmr^#9#wPSEaRAoL+nma^xp&{8Tr(YpX6qd
zhrEdooUWoWScn7eBtPn(iOZ%hYB_FjcneWdq7-8i#g&~+xdoD6E-h49ywdV0S^9So
z8%+G!#Y)|N&+HL?{Nd1r;A?0U4-1e+%YVv>;U@A?(ZZYSci^DNMd@@u3KTlr;Bm8C
zkAs~-VuaVPvg~u*LXxQKt)M-w6c*$Vi{Ui~w3n(&-bPhs2r*I0j{$pn2u(tGVlkoD
zBvDck6!NH8_@dMp$gdjEQG>t!)mToNtYG|kSJOM^&y-gW1=`p{2@Bzq4W|@+59p`+
zd<^GO^j!Pm2%=W6TylmTk(tRv<D}e~36|T^n1nt4=~4S?@X&{4S+{($yQ@kEl5vm#
z7+?eXQ;|q>uG;Q0{a{1_GHOP)FW&a-4IklMV+`%b(~%>&`BN`$Ieh^H*jcdCl<h06
z7=WASJnmzlR{aJS4Ws9P+E>&OKW(+C*UOf#>WoCs49nmwg<X!GD>qW)OiqrCINuKC
zhqz8%>x`q5)I+v^Gy=h7*cxsi<vAPH7Yz}np6@TFEHiyOHTo*VfbIrTF|ub!08j`m
z=y%dT=MoD&TzXt0jo%=nLr_IkeI8WfSw`(!a_0aA&*_VyK;?&j_ILpkKaT0e^8pMf
zeF=Zil-6X?osNt4{fmoG!m>>W)U~aJ>7A`?+gU=+hi4iZ8sT@KDnJFAeYKz5^8#il
zu%d%GM6{U-4e!Xt$4k^6Zon+>Ed@%MJ9gKPfBB30&<NZOGrV6TOkcB6C34L~5n%LM
z6Hl@O%^LK|JZ=Z|XhoXh@s^an{YAne+~d^n!s(7ISQUwi>iqa>hq)0!-OLuRI+;D#
zc)=_|0Ly&EJV1!O7fzR!oV$qfFBDq7*#!E_dwWN;gv}nKSmu)V);&I9tt7j-t*VoU
zw&=)3)1M?T7Vh0ZRn*3=qWsDTbJDiLtbzld{fx!`+W@;KS6TljQ&X3NxAj5J!MMNB
zBepM0i?-YYIrv>-6Tc<N?Z!N+rFzJy@$zga2#Sp=wNaqZM%?E}#+&Wo7aD<e!f_+w
zIHjHT&(nAb!;DL|K1khV`6WUJ`R97ipXD!^=22|Q`Qztdb5gQ(;UZ)h@?dt}C;kM#
zs4;3*wa1-VJL2CDZV7o=%H`#oNf~-u3tX&lMEug|`K0$~2vYu>-N@^zs-DEWsH}bS
z^@4ftVBoMtC^(RPUwUZ*P|8U4;Ri{Qq$+*$B0+EI2Uh%+!~r>|xhiLk3@49NE@)7M
z$jxt0ObLF1M$klC?2b<i)E_=HsxhnGC`ds}FBC=)Ei8|B@+L1cb0`cvbUu1Er!^tw
zh=!WZmGq)q{gP7<LhZpDrThu*^a6G2iY%+1FTc20*+@TxMHC&j>4s>29*O+oMCwZU
z4DCE3>>XnQABJk8EXgH8uvRs#dLv9815Zb+n1YOSj69XeT}_>Z+gHJf)Vavn^PZB6
zEqc6{gD(LnlA_wwd7D+UHIGpt=-lWG>(Z^}@>ez<gi%%2oPX}RaW@Qjum#boE*RaN
zuG=BzzJGnYTHvF%$h2Ftrvq2P%9RQk6P@1Um*snO2=NKoPfkofH7v1NCS^ac1;v@(
zrAvJq3qhsAawsgG1*j7i%J}#5;{NV4XqwavSlQPN`~DlJ?2SB{tn2d53E{@tmf`js
zMX4-E0D7dukAn#hOF%Fr_ql`1)Nws&_B>9d$C+iVHpNlkn{quZW}Pi=RJm~zokSlx
z1(??I-A`)$-ud5<$>^cd3243_8AfhKfG#d-pG)Jf#?OW9n9kP%-<Ag*)$XSXrtBHW
zO{v9yy?Fm8r%z$y!c~gH*7ws`u*mVxSJS$!1AY_;%a;GFy)O@EfBRxnV)cUMJ7RXq
zzi|p4Il%5<#gj&U4WWO&bv(1z4>E;0C@*{Z8$eXmzOm;qZnFFSraJFw9TNZvLZDuV
z<<A|BqLf8BjxPBrVKo(~&r{2i6|!Vn?JU)#`5>SqRq%LPIq~sX4DTOJMaZc(=~VdZ
zc=kzHJi5c^v_ta6vSf-Er{Lw?p;p!SlDtP#D*ssLU1cd~&#}=>6>F3iJKZc+|7`y-
ztq*z8u@=Tzg$-fj$1r|kNqqT7dwM0&??w=yJ$qZMSw5JOapbS){NTG1Ax<cNMrKS7
zGo-wBpZy((Y*}-e55)w1Uh)ihI?kseI1eK$hHSwaH3e%_8iQq>_}WBUl+Kur0NV%l
z2WQ6K>^+lulXYuy;I8RcAES><vB@N~)ObE=pe~I@zl87l0TBr+AKPp7nN=(%&SPiA
z=xCTUS(_OGLm}*J88EU5qV8r(Vlm$Tc3+imT1JSkOm9!E8(~#mI7i9dPfd_0^50Eu
zw41u2olsaUpEzSmk3zpNv7IB<t1z6cc}|QdA<CnZ9!JI^6BlCpsx3wyq})VVvj+-t
z>B#@MucY(aRh8||@v?28Y@iT{Tk0qHAm(kh=a|i&b_8YRJO;)WFd9sYQmG`eek2y(
zqaomR>jD%4xB4Ik9o}rTxyPd7u$fYUK`ZL2rlRyoX1HNy(4->e0ehsyXTk(F>;$(>
zFgKyinx|^}MN%rmHD%!Xlg6NCyQ5b^NXYP+m%F+ohG@JQGv&y?cLBQReeF#~%1?n>
zrgwh1pG}*R*tT^kI0r+t#v~RAlHL)@ed7qjlXtoiz%}7rL`N5=)o&slTiq14%lMv_
zmT8mj*$?;w2y(=xmplt>xVWr#Yx{-LBWf`d5Hh%uFRp*v1GhNB+I`E5M=XT_kM}bk
z5<qxUh$@m1Ip}~?xCy3#_njn^Yvyeo-H9gGkikF+;~&49Ww&1?g3`YZXh))%Y_)SF
z9!n*}L@lag&9ZpD1SJwNHp1l^D;G6VOdWiB;-%lV>To)Iv?6^;4G!%oI;bLK7wg{2
z!J<EB@h0piHL!+P_yGr7s&~!~FJ2?<N^1s<Y(CkWJ{3E_j#=Au62{!ZX#Gu*eE%!I
zsrbDoOWxUt%8{=h+an3-XV2^1VaIP^SxrQNt~E;H(4R=^21L%BWRXvBc@l#nGgv4`
zHPm>sbt?sGk8{7<1fIqXhi-{_V{l`JmaOsGdt!lm4d1wi$Q|Hyy5S)_o%TeB!h(OS
z>RNI<j**0gj8w;JUXq#_VUJe#CoNfc{qCsWvlCQ=8ugVdK7`q!na)hb9M4&1u_B(2
zJ|S!1Q!^z$Wl+&S!_Q$b;gymyXG}R^kH1p}-0?gz|D#A4y8X3_$R;MF-!<N!Iu+dR
z{-3fP%G*B@=V{WF;pk$&bp5?<my>&O25RMcqF-bNA?1u7Ih&>VOzfA0JhF!||MtG<
z>a#F@XC^A;a!mnii~%$<d7RyeC8L=Mm$)2h)0z2V+8zFStkkV=e*QrQh99T^?iY#s
zj~6Dl&<&5Z?0d&lT`_{}LA8W1s_uzU?GOB$1_GQO0%V(c<k<{S(0Tc0k`NyIgJoK#
z+Oyh*;c_7aobSnF77-*UYVm}YC2%l>&n7~ig_3Jg!!ei(C?wqm)q1XoKaUu|%M+HB
zIV&8PJd=r$cb`(vGQVQsM@c*rz_vbo6|du~-MIlzHQ4DBc{)GAg<q)?2gaX{Fy)bE
zt1%{W5ty!GjMp1UcF3jfs!Y^%18MgrQVq5J1eHO2;wV}C0Aw4nXtS0jkebn&A^cP_
zK8x4HI;sTs!zyDnnFH4{XC+0@*TZ$i9#gv~K(MCkhM?y`W9DK*sx}*xkZr`|iFmV|
zZ+PAzy^}YU?<)$GVKkTELNC!mQ^!Erk8x10bf;dU>sJg?m*-7K?~SKOw<{@aHOCgB
zF0yXB4HKf+Gg0@1RaM!D4>)YM_;jNx44AZ}xuJR699Rg}V`gW~Zlecod!Q+`<3MsR
zBlxue;M#LI;UP|cDV05)kyA8LTSej*>R-Iyd^Gf0#Ebfm_kbrZ1l0oxOQi=>LT@AB
z{@#Lt;h;`i9m$~H#Stp<FsI};mbbBWOD$y1d9n2^lS3^IdV<!q7;wD|nOSUZLb8Ti
z_WTGWG<80vIq|<Kj{*_HJiyTnv6ub94*-plXRDsDCO&(j1r4Xio;IrLe(j7(&TGN=
z(*n7QOQyk?YiUlvXs2Z=6C4j~vSD0zsoAU}ZH6u?mda77I(*=kugmeFyobEvAK^b5
zfe4mLI3y$v42n>y!O+^~1Mym~_jr6Uq4ikwOD2TB8CG^8+TqBS8oQyRB{x9RhWJrV
zf)1p<N_~GTMB)Y<wl^+Tle@eT6AoNL5JVk0KY<>2m%9>ESycv-$>_0S-(TC(tV<uJ
zt+lz`kDyltkz2eh?XEzEvriKup}Sccbtt#G9GFmRt8=7}iJ6o`swNUZ;n0%x^3DZ5
z6zb#m3m!o9CV)x1HbNOE5VcCu6$yJ=RdDxO4N?wX9REUH<C-Sc%PpGC)om#|LzxAK
zRi4+>xXGK}W|xK|Lz_Y>=e-d((<1r5*fexA!fqP*!br$_+$nJ5PJ`PUQm47sbLVey
z4!2mSzkY$<g81@Oljy!t1*X98+G8<x9rh+5+tILW4HBZl7~DmS>*$CUPl3##gxF`~
zLMrV>nEunq8|fBe|4$U^@>heLpu^0-K$04k(G8W5{sEx+iD}5hret~!V2JL&z#Zv?
zaux`}zpII`_oQwm9vIduyM$Fvwj%Wo7R^_IgT>41*E>y5OMCJIJ1(e2KtT@pDj8v5
zI=l|60%wY5*en7xf*9B2P7RqOf8SvS3udSsy?@LJudAO9pmvwz!=ctCa(-T>;Bz}x
zT4)%mJ!(ps4P&gc3SS6PMJdZ<L?@!FgrRcoVxg+@;um~<)?1cJRjw(FFU8%Qv@O#+
zrBWruM)qBqIkQiHAEFXJfUogW9O?o|{YbM`hw1}S;Zat{8IvC(60x2ls_vgVg65;@
zVGeFxQ>n(A6ps7@a^w((Slt)Ny1@;SfOjVw+(aI~9Ts>lGg17m)_gkG$fDx6!~aeH
z;wNQpc^Av|E_Q?lJaEh}^kmY#aPyQ*9s?$G$f&zO65MjSP}c2qEekN|y5vsw+14;o
zW3b1=&DRL^rX8#<GQD5jKuF}<JdZ4-*#I=MeVH{}7sm?-O-+zY>%N{;Q1k`Q2$D>3
zV)=9w&2!s(UVMG~WaR2!_2^?QBIyvis}Rrab@95oNAZ!bKjbSu+~l<#)n!V3FQSC1
zr8DF^4vhS@OCo7K1v&A=9gn<lWp5<4Psp7Ba9!%2@|gzG1&bLpv?Nho3c?s+t0Kpp
z-0@+-%DsXYumB27GUI#h51M>bl#B_2gr2!IKGCz&OFYXtvcFn~mABO5Y(IjF-U0|;
zakeKC(D_P#h*{5F+>#;kUQt=8@DU&(@xM|r419K{QOx`aIA5Ws`&JysCFLUL2(WYd
zHqrRa`P=KTWE+ojO@bH=<8#%wre>69VFT;XyaXWQb4%AdilG%?BO2tz&_xZD-{q01
z$Gns@_3eM9*<8AONOZdqR1dg;-i+6fOq#C~HM;s7ln3|6huv~sXu#pnxKH=P_k^ZK
zB*(rE{KZbVJa~!zwG2~xy5V|vdW;BCCvs*-A56}1{{dd?^pq=G!-1vakDnj8O-E0b
zm752A1}@`58)tc@rLKZDluTw<!!aDCjJ~I3A^<c>NR$#uk<IcY9J5a#@{xAffgDZs
zWhu?B{h!tZ;NZSbPphu^-x>gRZbT7pZ{m%Jm##2nKARS2o?jj+J+AmN@1(yaG<F`a
z>MNV<6tK#b($$F1Brn=*%19vXWtKQiNoJK+t7&(w_UicraqXKZ5~h2fNsv1=lKye(
z_(oLb6_uPz-7I0n1%Ocr=_xqJK8ZW^n+DfhP?Mn;Rc4anGZo^{RVrUYl4qG8zdk@3
zB{(hDpmZowMsdczb=`&kPQUWJ3KN*H5B3jiW0@kzylZdy+}<>UHou3g3K2FD?zvN<
z0KL8b`of(=t7Op|1ljp~vjRpV2V;v0c}B)%%`b;ADm9TKKP%kH=-=Yb&O4j`d19R{
zv^(<lSFuyPXxNB-;sR?_vVjh4x)C35=JSa+3{<d3x(~cO7HDP)eECLFUUU6^B+(HH
zVpWY%9+&Te%YzDHZlO*6dJ`hwJe*$@Wc6M`TnrjR`cK5;SAICGN)cB}<I#hUxlV#j
zj6!iXvANDVA=F)*8`G;(>R9qyq6C^2iB5xSn9m+*!1dr7BgjySiNal<d2HLbt(P=6
zr@nEM0+#D<BT2NbKnB*df%AB#itcmtkmoyi{^oCA;@#l=Wi6gf1uBBRW;X)TQkn_p
zHaoYyWR%``@+H68mduEV_w*{0W<nU`&`q2HLN88~vhJ`Ofv>QFXkTfLH8fNo{#9)Y
zb)r$TppUJ7z9U~yy`SojfJKjpg}lLOvoe^VzR&bUzk${loQPR6b(~AOn_(Gl*b_v9
zg^M6%%WfRa8R~I`Px#fP?thLY<*!hg`VAUj0BH9Vc=$9lgl<l%`!;G57ijRujJ`Uq
zn1+l6gEm})US8w~eeM!MgbCj4{QTzaqY74NV<U&UMiIDZ87eWreh%N24a_v4cSk+o
z3R+c4Z(J86NtQ5xxniMm=U;}yz~S`;81-X!+5TA&o0Min%RZvGbu)!>C}Sr@yw<9I
zplO+Oy8QDj;n-<#1;y9l+dx<5&ZA)He88ePvFL_r(z73dhJpr+yE{2o;1laI%$HmS
zIy~cGnSGC|=_)$!67=e2rQ<)}@()AOT2_1AH%J$)kj#{dFo)m6AQ2@itUP^*wai<x
za_$!fNaZlkIZJI-S6Td9ywpM_5P(6QxY*_)Nx=?Bq_F&qcI+9o&fk{zw5u!Zz(!bX
zYA?;wM0?~N-J=#Ir1Wl{`5(UPLb&fB>lRI~3bI>ijsXnuNuaAzU`+sKu}J~H;ihBY
zBcT}FINu>tb@#&orq9^+C8>lvjcShx*ZiHtA4-RRF-wH4mKRh&N9)I%bh?MfO(2ok
zOqgma%W`yi*;Ian+y}<Z>Om3l16Y^_4^@@@-E!N<0gO!wjYpNuKN12v+pr7q1vR2D
zzh>v{nscH>-IyXp(4lO{aFp{UF&JkBJaob^RI;GuE8RxEJX{;0mo2Ur_O$2sdafjC
z>u!cSVNb^cXw5H$mTW1_8*&t<GR(z>^uN_b(4IA9t-jdfB)jo!dk9!s2x3h`6fOx<
zxe#-!pkcOnwqSZEXf?t+>=wXdcLsJ)>KgZE!SH$>;j%ZDQ=|0+#K8dMYV)uA<MN#_
zNdxg);+AE>NS3DzmoYX3M7!m8$FrO}Ry&_AMuD7TLbLsf%P4aTo6nV%ZnMC}*x^~&
zM`2J&NOGGuPaIGGsG-5>DR76D6L1BIDs)GNNvk+u^StphZ+e;!iGGS;6&kr|RHJXI
zo=7vLDr1z6{(9y^b^3Im(wn8*=tJr15`G0~Ml6-=y3%kj64&yCGy>zEs88cf5~JVG
z4|EMpU;t8lwp<6!x^`zvb!bI?KXcN}orbOSKa53HE6F+uWx~V<oAPHnZLL2NA~ydz
zC62jk$O#N>jFmWFvTV<oyuTfF^pIH}`Kx!{LP)VeVPtkmaW28~GlVH#|M~VdHTg9c
z71WQXtDpXxZ{&JCw-z#rNGEtD_t&HnYfv#85l7{=8>2^KSpJ+0E0zXRtng70XG~KB
zkPYGiX;efUroTns3^%!Z?veA9BY>Tm)oIHpD116y>r#L=2Ym2v^v8H%KY*lK?K1IC
znfTrl4ZHx<o6hdxWX4*zr4J6>_ZC+&Axgr}$$8@8Bn_)NOAZ_r#PruzWidmEhYFdD
zQ2lJBUbhYch{6)u<f<?~o-+#26*;bGr7eyBa(<vZd;=Dho>B(}(#JM3`C-G@%}J?N
zSPVV{KtJ<Ecdtse5K3=YEsz2f{*+8Poc2X~?oV}zUMq~WPk0JhLNBX~1etrt9u-r8
zpd1KFCRtZTg3-(+gfGO=byT~54}483HRl5J17?VRl_-iy<qzU?Q9u5GOXA#jy+gg{
zQY=YH#-hdE5mM*WospM<DuiyylfyrbpeT4DR1^64QIQ4Qt@?(Bw%avdNH2gmNE7ns
zD=#<YxSC=^dZ%IUvt_+dO&llmKqm%A;&mnpHCL4Ef5y`!FDxSW&URF>v2!PfXPHFM
z?EW<Hl^<2`3e7Uh9S3C7PbykG(gpr{wEn7|^~J5!(-`4uto>*N?ly=a91$T<ji;Hi
z?TjK*vD9^?HI(GkbW&-4Z4u2Ik}84X?0b>)g$Az}LoKlC;id)#Arx>AU-ssqKREb?
zA1nn@D%V2kRb*3Tu?LSw!yx`UmeBYyiJZwoXwXjV=bx~gXcaJo{u>=(&mCK&(XkC6
z7lmyM{W5w)VJYjzzb+y{!HmLdoeW-(YLcfHJB@p1oq0h4=vF}~Ll^(xR8=X78TS-U
z>+=Nl`ITN^vf%$OLJgp&`9+in<vp1DFAjWt@e&}hn8;S>@HhsDpF;bUVtX)vZTq{;
z!E;Eh<8RI)P8aT&jxzKQf4$PXJ@Wd~`-57E=KA^%G)mdOD|H&YHX~`%*i8CNpxl{C
zrQ&A}jA$10;?~m@Z3-dc(#L&PBxm=}x35Wi_bxBss~;g&ckMAO2oL8=(2&2+K}Y=u
z-_`vu_dk91$u4Xvl#cXl2U5M4E)0VM73avk?IwjE4%iV_&4otyL{havnjJPrZG5T5
zyuNbKv0k(Qbv9k@jsddmi_77%tXEj6Uc&3FpH?r@mHC}e<JD#XSAGNk22+-`^|_-U
z`wm3vNDb$o9~*751^|?_wx^aLykAUHkwWgw^{GQMUz->pe*fdgO;5{&0$v>p2WN=r
zACPS@dAd8lI$h5Ze!ZW7hJtkHIOa-BOpJZ-qr{+Sb!6n?X(cvG2$A>5UC<bMP=)xN
zBb>A1&;gXE(zxsi0BlOYp#$3zAy%_XW}P2f@(8k*xSEWz5x;3hXFH}R>;jcWMd(%`
zM9OYOG7PDb_qguUEtzsHR-R3(QV&ds*wXBya^v2rmQ?Rr`TGy1@eRjHSC^#589Q!;
zlY36OwBrOTb&;<nDgt7XdZqVwo>weuVp@6#`vs&_-L{Sp2bzIWJ0*e8)C`R>ld7%&
z$xhko_)cub3AHET?&TZNZRfx(X`JO$4?LC~G^CiCED`5M+Gt#>z9ydsAC)y53%+H`
z7B0uzZd0)VJ03`*odE?-E^|w1$9iHux4&~*MuNq=jO8%4QqtQuHP%RFDd>}`E-G7z
z!FybbmlMJhm`VabOijPHVM0jW=HPrmEzu|KZ?MAR3M#;o3l@I;tM+HMRHwC4wB6{H
z=*Kiv`EFO_qy!T>s`w^}w;&jmk8AqmOED)27mYlZ@)`XMjjt9T6!t~oHkWV7uj@<7
z((j2lbA%O)7~M3B3(RPH=C)e(pHUk?_g3v<8TpQjH6|)Gq#lgsyzHrhq4<Le%Y5!@
zsn=-qW9I42g_<vUl$9RpmKV+wh(XlMald7AHVmG)+DnL}cW0ch+s<fXtU-~<#=04x
zO=QZqB8#EoeV}h^52rf$&#JQ_Wtgm&3JtID`b-N-OCx|t#|I#tmhJry-=Xt-D97Bk
z@BxAvl*I!_Tad|{(QkAfYC#q8Aen=sAohUM<bnE_TJ!rE+U9$GoWu4KYGNjZ{6$PF
z?-}8@rT~}$h&*)TD`yOS;G7YG6sO}Ue2;(eSLd4dTs#ojW~PM`v=aViYtV6-R@BKT
z;Aa<|s`gm$1!q;5?#x57<ul?A{*$Qd2$=39isI~IrpFa>=^Ofwot?Ld!xr+MUPfHn
z)pGLU=GNKq5oLAv$C3%0jSC8Zj936d{Wn>}oo`0dHAX!-UhJ|vjl!xen~Yd`ZaM2U
zRJA-O7UgsL>V42D65ZK)ALm287n9WBG&LyH<N2)0zQ5K9_yFt<Ht9>O@dh7ON7BH5
z4$|6NraG4x)C7GtG5n?*(t`xJ$VG`8Uo>Q};DFwX@j`v^COLJ~s_gXAZKm=$8^CSs
zoJYs{CQCsNlEgI(Yh%E^Abd~j2dKM|Pu~oNz*PTtAWEr(i`ol!4}YSAf9y++6%s^1
z9nD8goWcebNkJu=lQS(NKV9PsZ>#w}xs}SU&A{6EnbAsCM>+u6CvBSBvA^T!zrLLg
z_>(0WFr#T^y+ks&FA;`hJ}DBZ)GK(jAdM8+Wa0UP^RKEo&w(vF6zmO(xU(A`fqU{W
z3rzyXbj>5-3Hidywl_OC-A|X%2El$gfM3LAnAo${Zt26z!$ME?B$J0pT^_Y1Wy}t;
z@EMGA!Uv93acu=j?D{$ua(Y>h&GtCjmKi(>jZke>z31Tof{p2{sbgBkDK{5vHzj6}
zpME&GgE+0fMURr|>or1O*!fDiQexHE`T%3cy9$4#8%!-2H=xpz2Sz3Ro+*Cx5l<lu
zphq0?rG*4b*HnTwgz5B8R$UQ-ekaS{X+J{XF=dpLc7Zl?vE@WI?G>Mx3a~XpP=G;a
zSc`@%Y*zdvkz;4Uij000jt=#3HCF5&UYcYW<4W2TK9om?!rf;o{cHIP^)hv7HI}~M
z_>Z}*O!4tl0OFp#Yov@#^1E4m&;O1qo_)i<y!#Gw<DyI3wLQ`0MI7$<jL!f>A)57F
zTt>3;a<Gf=2EoJY<$m`KeexYB4LRJRt~bU3mIv!?fivdrm<CZ5Cl5mLify8cNA)^K
zN8}ND=f=}`oQhqyN-Np#fD1%!XE+X2%j5rAtv}Yn(X7wks|URJQlx}x>owC}Zw}Wr
z9eCdbt&h$(?_N`0a+8`scfLAPn}d6D;U7>3wRw5px}7Lb$H1`pmS9cGD>?mFKncbG
zVjizD(go)ESijGE3o}D=U4LFTnK=0DxB}HUkV=KrULH9?rK%yb&j{|?Qrz3@T8zw&
zp$kcRoXgiw;|joh?@zbAC^eLzd$b|T0F)LB;T-tNsVY70r-0Wul77=VG?ef_*uwxO
zsH)q+f!kJyywVWlFHT#B=nU?J|5QDV0~-gu=^mVncYQK5a5s;*_?E?6z5|)6h&S$!
zz`Jd=2s9`S4Zs9avYbpRF~i|z;CSPTprVJ|ACW?3GtQB7`Z}eekz&jyqp$ymzU_ro
zQ-sfLFKEtY2heKKVj&PO;DMl@?|wKxtTDa4n^#x;Qxr?1DBp>;=}|8!DI+@Xu+D7R
zs3d<tA?{jOcn*$yvB?$}h+Qq(HCy=hV~QzjX-P{;!gzaok3CqF&>_IUz!a90Y`O7e
z9A}DMS+oF<us#c}tgP&qkD<E(FgC&P6nPs5Z6+L`74;J+Q3S|Uv&LS&TksPuTHo>#
zDinh&O#TBHy9I9lxB_YzjuqD8e7fH+7AUs!RlD6Cr#2TuGP1MQYaf-dLiHTGQr_gW
zD=RB41WOpTs<#W`gf@X(w9Py>pQK6}kndSA3KRrHGqyL)FZ)V?`JSh1Kq<9-X_+CU
zckA3~!NBu;*|uiz<oy`Bqzd>}Zl^YVVfT!e)|(#zb&T;+1f#i6K(dGpOY&Y8|D5)n
z20C4R-t{qgT(tPFdT!ho+kfgBQv~TVq^Fj9=6U5db}4qtKBmf8)HgbzYKNA0<R3|8
z{@Y2nt^VQQ_@ZIR?{RdOWaE|fvAXj@4oD3tHQ@`(I5@E28hUM%ac!l&+2PNhKkqg$
zGa!Kw)_u)g<?^vdHK2V_vmat-_d5H#7?+ha#u|V)IsksQ-}6>~`gmy|o^6iQpIcrY
zm7h-`Cnq;am^lUTkLsG5*`|m2K+lJPkugIe4CF0x^70u+Q(^rk;1?xpXcaQF-1_=N
zV>UR6ti+U*6fA6P6-~|gLUwIyYwM%VBn1E|q^736^XsOG(*n3c1B2D=I<<QptIQst
zsnkzmn>jNyVYihD3l*$pVXL8m4c@E(rEOzlQ(9hb<5InFSd^c?X`X@Ce|oE?t4rW(
zRFwRNPt@1<Tl55(=27j$$jA^IQm3Kg(c{Uv7fMFY!@~njPEro0D`G|PSO)ihfSl>!
zYX8fi=|)S>GqdeN^$u90P=7lw$G5b4_}r<5KzE0==jj8`c7z0b!fmad2W&}6Nr8vG
z0+8YAwm6IdlZYDL+DMQl`LT5#pPBhPnO;ps=5C`{ffl2m%?p5y?-qJq2H(;XAfKvJ
zX9|NV{KV$}a78X8lr5V?lUL3F^vZa^K^KobNCKan0R#MbKQ50bD1ux7h$Tj(L}^Wp
z<>o--)zMPY+wKZ{9*v%QLEjbGGRL3Gb$2|5R~>94{8&1U>XR&Jd$o-G2Z62w<j;|j
z4M00({QPhYRLupij#s{ggg~EkUMd0^Y9)bl=*OGmeuoyo>Xw|;-PHs{E)2bQun!In
z;#EHYrZ+D8Wh~o4C!l=m0yaDKE<1ABfiK<=VW_0z3f!w#q*Bk`cSj8qZ<2ulJvk=*
zVeD>u$`0-OG(Oh{p)>p2fHaPllQZQFrLK^{m1j)=&ijl$dfW#b?e6av)VB{W0+49n
z{iqQH9=j<t5Zs{zyhB@YflriSCX*mZSiC*56qoZBcnxr_)UFfD_FVIJ0hD||Fi%yE
zeg|7ipIi%o66NLPcO-EQs78PNQvB}UUOie~&TwBM{4CoI_Wt0FIer(i_kD=0KSjvM
z96QgY+Khid-OuYejsU#;Ph~o))3#lb`<KVl95>+C;^*e(TmYZc1$dwds&c%JYXaU=
zlDH$1(KsXN=>+se$wvOyDVc00pOIj{TucPMG(3$Oy|}0zBZZ39mzK(a^8zk|?zH>S
z0Ur)}dl1)%2{;Aj7Kc>^){dhOI);WUHDGbJ0l2k}wl>{x3?Z$pt*ugd6_5*@KD_=V
zLq#zhN6Ok_zd{4}tQg+@j?1J?R#8z=l&m_Q&MChJ$`~Ke;C}9A`zi_n@jKc%MfxIe
zE+4QEWa*25;qvc*LJ=is&FQvz9P8?+t5dOd-<j&@>M}Tjhr<g}M;#p<`nLn$o~{Q@
z+gH`pd{dQc1CQSusPc{$$$$F=tVIh5%5mTL8+H4&bsm&g6i)0WId<O{cLF&irS>fF
zexQc!z3zWIGRoSckAsap0&2!O0guNm^~J?fK%Z*R<@M$H0x-}efZHE=b?@f_wm2@J
zj4;gi<ZK<>7CI;5<HH9;a&`bA7m$58G2Wjp+G|&gBM}6t11&G_cK{jA`tS&}Q0eL6
z!~4;r+v(`&{x;4i_FFVeb@_TWJFLoyib8^S4Vu<X_5=ok|L~c%#}<oEON$tFuO;&H
z@ZbW{-C!Jm_hTAcFxXE1AKqOMNMU7qEL$J_X}c_3wyx7Q?HOMHWRjrJ0YuySz|ny}
z2U0KLV-{wI^Ho505tQ0~j*rujlLrGhZ>GAQnwpw(oBq*S7onsIp!tJg)@{r+>Iv`z
zSN{jxg5>06Wo2a%@W^YI?10>92r<7$raC|QvaSyh+<U{r0@pcRZNLu|3?b-qMMg>a
zrLd5y*=|XU4l5tXpHR`#CVl_zJ-r?n1jakww7yPSw@{>F2;Nw^@CCfsz)`l|>NFs<
zn%<tQy6sIuhWGyf8p?}6W81Fd%p7>Y-)~+3h^)n|92Zwt#{$!RxH*nqIi_J`j0TPm
zT!J__K_?y}AbIs-`TCNfr;WV*@1F~(C_|kAhm``1284by-<U8Hj*O5#DM%F!&KuEr
zcR?JF#XX}*_vhp=IXQZ>=6}1pmn-WpIB(M@Y?k9hc3gT%{*P*(4iQ;9gOU@0d1M04
zNSYTadg}CIPVe7rU)3<KRkgK$Zp8_Y0};Z=mE$2G=-K-Y72@o1{WY7{^`CK*Zdpae
zJ2H&O%=!tt;aK9f$CD0kAVsLtXr*%0K@}1)ww~u&e)vEiP1O$8`2JcKzhl=m&Nrg~
z-rMy+VTd=6Z4_`%-PYDtcdAZ>5gD47^HJzwkCMx7vG;8Zr2f;W*X#4Gibm6jXP5H}
z!S1#B@3J;1>$h;VxoKKnQL)r)4~isi<lvzLyQe8W0x3wz?2s{A#KPg=);UP5fW0Cb
z1wL~Y6%_&J^ckr7O&7`I{u|Gd2MG(1tSbQPzGPrtnJ#2_D-S#Z5i-z6;eCV9tE<1+
zhl|Sx@burzKKa|X|7_=(+C5JLRPDb?Pzby}T^=npguf^80+@4NUS3dk`<23|Q*REh
zD%b9||GB&zunc}}vRweVrPXTtwQQ3+$b9|;zCP<a4`L$`@t_4oMMb5(Yt79K8W@lO
z2Of{YW~5_e1VJAkkW?NQE-zVWxw~_MOpAR3<Rrhp{}+x-$o2Q!Yj3%U(V-<H6cM-H
zc7Y`-I@)@@`&qWd9oSi`J_`_?198=Sz1trP7gtSBZxk@2f+Fi1?<Eiv8q$OW`*ogu
zVGl4fG7OAQ0jw$TXK+ICILx|iuKzwFARvH5IN$g7<V{u__#$wgi|y`UzLKO%Wi_%_
zKbWJ|a)QJ#;QkX>`NG0EplY4=RuKRZ3>^u!4S0%xX1VU+&&i2^i7phJCPhf^zMIP!
z=fc^nnv(MyOR4!!G!TL$f!!>1djIFS>W&LCG&3+kRWh&!Vo*E?w|G746L0nKo`B~x
zK6eJd3{j#)Diuo<nNRl?CP8au1~>tho~Mf>8#4t3<hSsDe}DfLM&B;Rwz@40Y{M^Z
zdtQmjp_6D7C_pR+AG!ol23W@Zm}7wJv@Plw5b*{c=s*7lx9+=5#FMUD!(YFCdF`ZV
zDQl#htai8#;adV*$LDb(0}UN+V6FPzJ1M}Szl}?rgE07xo*Ew?|Bq8YGN=??fNvlC
zH*xIomb7I1-{=UvKDoZd+v#Z(NAWY)OQ5<__h}S|Lt!gb&p~vC<E?WE0TQWJAf*KS
zRi^Jr+aK^fAb%d#q@ztbI<f=ROyF9Xbm~L8{U4{B?e(&K4(N-Oa>i={pKU>5xb?Vs
z<xOPp{-ooy^=eu^@P@R@g5)Rp0Fn3@hxpB;eYW_kkXuD*0;QnG2i~iPm}ahk`SgOS
zFcY#6tL9UG${4+Xm=(G)Z^evD&wb_Y9)BN>6{YLU7rfa1`7Ixl;vmrD0zZ)<Z+kWx
zT2wU=fDZ#>U5IbLpyW-FQWR04!98t%Sc};rMvVb2F%GT&C5-{L@SWF{#<Ay_>pzR%
zcT-t(xXii(GfKiS>!-W@sKNrKG1Xvt&`fbsx6Oq*PVi<iu0Ai}t-SUoX*ztZkC~Jd
zOixDB1FVtZO=2DFao#l7-$pZ*mv!H)vex7sh{+(U)vVTEqPlDYqjwq=nkwq*5@6wQ
zWyGLUN`Z$5`GyN%)U15-9B;`P@Pf5Aa}0eOY#=ED@8=wOsIIQI*!h)0#N(7J3n@xZ
z9U2;1_c;yJNZ%ZRk<kQ*H99R021nZ9xZp$S>FL@(I^WK~3uM`Et_AG&UpkX)OkJ>k
z3!o(#5xp6T?{O0n0Y?NFu8GF~&*BVNV%C4-JYU{kOAVH0O}g}8wf|0IN5W$Z0+yMS
zk}?QdAXY2Q%;2C;H!VFoZ$e^UP=$<7{<MLnC`kT|!L6WUVR<{(M!Tg%u(cu9(4c2!
zwcP%hkQwl>|E6Y~%B%<cLN(1jvZ~yh4YIUtV}zj){{nY@4Ct-BH4N4|y+M2%`tTcD
z@`ptOyZtf;*bTfD&A)?{=IrdO@3Hg=%n1dOfyvLlKCz)V(g~oWrnFI+@x*FcE%Gry
z1#tbTsGz^$Ox}Q1{o$y?Ai~#~NxvC%ry;9+Dib*~GcpKykEcBopg#O_baV?S{BDEf
zCpzVSO1t)#rou41ZniE=X_HM*$jTBd)MXbJ-ImKXD6<l^p>r)OWlhr%CX--uBfBuX
z5auR^j8k%GpiD<n7G!Q`9<r#|kRlz+NFrD(XdA<xulf%<Kb;>A=e*xJFVFM5-*?{c
z+32p=P5>3pB~xU?t~TwW{V*DqM7y;j_xq-Ggf<bR>(cAx*y)+v1MtWAe8G}yF+TKq
zHUFSxF#A?|VV6F#P^wauz=wH7()xSY2^mJi<W1t44`}-4r_Agd8{*2cGR$<taZ?R?
zc2E=qJ1}LD8PIkw>DkolOfHXpr+@{lX>~FMPB6s2+z*99oYEy3Y8If1OZTA<I2r+9
zz8K#^QPk98!J15Nk_plS8Y$(;-QrE1y_(0!u$~^l?AT>um+kfW&SA&l;V`!COs0!>
z0raTjwyA9c1}VajD-^+r__xjZ`JDhy6!awAEMrDro`}0LMS8NbQjR-=(m?rL{r!pN
z61sy?_vRqG1mjM=Q0OrWs|h*+sGt#=Ey)=YL6tuoPavk++AN~0&{T>1GU3BoW3kru
z6cmDDvDiB^^O*>E5TFKy2T*=Q<Kba=LfW8IsjO7pney_TjdO?Y<f>kRuo>ceoudmJ
zQz+IxJUbeM?4AeEts0cVy!7bbo-=MH*DP&+uEtxP1atpdur-ji7qnSnp$StwnzSDu
zpISN?<{4YGCn}r&*$S7w2GPD2ExT+&AGzS0p4J>N?of`Q>_W$+8e#wYon^!jZTcRG
z$dR`AKY$De`q^5O=Y2bLdI)PfUjxJBN=zfzU_VX37K|80;ldH^Z!LftJdbjOxM1-z
zBy(3Kh0x=+z@xT!3;7lfdH1V=+Ymn$RnEq}Ec#A8U~z_J9e{TOU==ozR=YiY)>U@|
z=~*#apNg7iO6P6{9=*ZPx+Q)96np9qIBxZrD}AIYpA3c6Xv)LET=~zF@KRf%UYtmn
P{2Ux|*@+^jw8r=w*#FAQ

diff --git a/master/_modules/autosklearn/estimators.html b/master/_modules/autosklearn/estimators.html
index aeb99d5845..f2af275f42 100644
--- a/master/_modules/autosklearn/estimators.html
+++ b/master/_modules/autosklearn/estimators.html
@@ -4,7 +4,7 @@
   <head>
     <meta charset="utf-8" />
     <meta name="viewport" content="width=device-width, initial-scale=1.0" />
-    <title>autosklearn.estimators &#8212; AutoSklearn 0.14.2 documentation</title>
+    <title>autosklearn.estimators &#8212; AutoSklearn 0.14.4 documentation</title>
     <link rel="stylesheet" type="text/css" href="../../_static/pygments.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/bootstrap-sphinx.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/sg_gallery.css" />
@@ -54,7 +54,7 @@
         </button>
         <a class="navbar-brand" href="../../index.html">
           auto-sklearn</a>
-        <span class="navbar-text navbar-version pull-left"><b>0.14.2</b></span>
+        <span class="navbar-text navbar-version pull-left"><b>0.14.4</b></span>
       </div>
 
         <div class="collapse navbar-collapse nav-collapse">
@@ -147,8 +147,8 @@ <h1>Source code for autosklearn.estimators</h1><div class="highlight"><pre>
         <span class="n">max_models_on_disc</span><span class="o">=</span><span class="mi">50</span><span class="p">,</span>
         <span class="n">seed</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span>
         <span class="n">memory_limit</span><span class="o">=</span><span class="mi">3072</span><span class="p">,</span>
-        <span class="n">include</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
-        <span class="n">exclude</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">include</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">]]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
+        <span class="n">exclude</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="n">List</span><span class="p">[</span><span class="nb">str</span><span class="p">]]]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">,</span>
         <span class="n">resampling_strategy</span><span class="o">=</span><span class="s1">&#39;holdout&#39;</span><span class="p">,</span>
         <span class="n">resampling_strategy_arguments</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
         <span class="n">tmp_folder</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
@@ -189,7 +189,7 @@ <h1>Source code for autosklearn.estimators</h1><div class="highlight"><pre>
 <span class="sd">        ensemble_size : int, optional (default=50)</span>
 <span class="sd">            Number of models added to the ensemble built by *Ensemble</span>
 <span class="sd">            selection from libraries of models*. Models are drawn with</span>
-<span class="sd">            replacement.</span>
+<span class="sd">            replacement. If set to ``0`` no ensemble is fit.</span>
 
 <span class="sd">        ensemble_nbest : int, optional (default=50)</span>
 <span class="sd">            Only consider the ``ensemble_nbest`` models when building an</span>
@@ -209,21 +209,64 @@ <h1>Source code for autosklearn.estimators</h1><div class="highlight"><pre>
 <span class="sd">        memory_limit : int, optional (3072)</span>
 <span class="sd">            Memory limit in MB for the machine learning algorithm.</span>
 <span class="sd">            `auto-sklearn` will stop fitting the machine learning algorithm if</span>
-<span class="sd">            it tries to allocate more than `memory_limit` MB.</span>
-<span class="sd">            If None is provided, no memory limit is set.</span>
-<span class="sd">            In case of multi-processing, `memory_limit` will be per job.</span>
-<span class="sd">            This memory limit also applies to the ensemble creation process.</span>
-
-<span class="sd">        include : dict, optional (None)</span>
-<span class="sd">            If None, all possible algorithms are used. Otherwise specifies</span>
-<span class="sd">            set of algorithms for each added component is used. Include and </span>
-<span class="sd">            exclude are incompatible if used together on the same component</span>
-
-<span class="sd">        exclude : dict, optional (None)</span>
-<span class="sd">            If None, all possible algorithms are used. Otherwise specifies</span>
-<span class="sd">            set of algorithms for each added component is not used.</span>
-<span class="sd">            Incompatible with include. Include and exclude are incompatible</span>
-<span class="sd">            if used together on the same component</span>
+<span class="sd">            it tries to allocate more than ``memory_limit`` MB.</span>
+
+<span class="sd">            **Important notes:**</span>
+
+<span class="sd">            * If ``None`` is provided, no memory limit is set.</span>
+<span class="sd">            * In case of multi-processing, ``memory_limit`` will be *per job*, so the total usage is</span>
+<span class="sd">              ``n_jobs x memory_limit``.</span>
+<span class="sd">            * The memory limit also applies to the ensemble creation process.</span>
+
+<span class="sd">        include : Optional[Dict[str, List[str]]] = None</span>
+<span class="sd">            If None, all possible algorithms are used.</span>
+
+<span class="sd">            Otherwise, specifies a step and the components that are included in search.</span>
+<span class="sd">            See ``/pipeline/components/&lt;step&gt;/*`` for available components.</span>
+
+<span class="sd">            Incompatible with parameter ``exclude``.</span>
+
+<span class="sd">            **Possible Steps**:</span>
+
+<span class="sd">            * ``&quot;data_preprocessor&quot;``</span>
+<span class="sd">            * ``&quot;balancing&quot;``</span>
+<span class="sd">            * ``&quot;feature_preprocessor&quot;``</span>
+<span class="sd">            * ``&quot;classifier&quot;`` - Only for when when using ``AutoSklearnClasssifier``</span>
+<span class="sd">            * ``&quot;regressor&quot;`` - Only for when when using ``AutoSklearnRegressor``</span>
+
+<span class="sd">            **Example**:</span>
+
+<span class="sd">            .. code-block:: python</span>
+
+<span class="sd">                include = {</span>
+<span class="sd">                    &#39;classifier&#39;: [&quot;random_forest&quot;],</span>
+<span class="sd">                    &#39;feature_preprocessor&#39;: [&quot;no_preprocessing&quot;]</span>
+<span class="sd">                }</span>
+
+<span class="sd">        exclude : Optional[Dict[str, List[str]]] = None</span>
+<span class="sd">            If None, all possible algorithms are used.</span>
+
+<span class="sd">            Otherwise, specifies a step and the components that are excluded from search.</span>
+<span class="sd">            See ``/pipeline/components/&lt;step&gt;/*`` for available components.</span>
+
+<span class="sd">            Incompatible with parameter ``include``.</span>
+
+<span class="sd">            **Possible Steps**:</span>
+
+<span class="sd">            * ``&quot;data_preprocessor&quot;``</span>
+<span class="sd">            * ``&quot;balancing&quot;``</span>
+<span class="sd">            * ``&quot;feature_preprocessor&quot;``</span>
+<span class="sd">            * ``&quot;classifier&quot;`` - Only for when when using ``AutoSklearnClasssifier``</span>
+<span class="sd">            * ``&quot;regressor&quot;`` - Only for when when using ``AutoSklearnRegressor``</span>
+
+<span class="sd">            **Example**:</span>
+
+<span class="sd">            .. code-block:: python</span>
+
+<span class="sd">                exclude = {</span>
+<span class="sd">                    &#39;classifier&#39;: [&quot;random_forest&quot;],</span>
+<span class="sd">                    &#39;feature_preprocessor&#39;: [&quot;no_preprocessing&quot;]</span>
+<span class="sd">                }</span>
 
 <span class="sd">        resampling_strategy : string or object, optional (&#39;holdout&#39;)</span>
 <span class="sd">            how to to handle overfitting, might need &#39;resampling_strategy_arguments&#39;</span>
@@ -258,10 +301,10 @@ <h1>Source code for autosklearn.estimators</h1><div class="highlight"><pre>
 <span class="sd">            * &#39;cv-iterative-fit&#39;: {&#39;folds&#39;: int}</span>
 <span class="sd">            * &#39;partial-cv&#39;: {&#39;folds&#39;: int, &#39;shuffle&#39;: bool}</span>
 <span class="sd">            * BaseCrossValidator or _RepeatedSplits or BaseShuffleSplit object: all arguments</span>
-<span class="sd">                required by chosen class as specified in scikit-learn documentation.</span>
-<span class="sd">                If arguments are not provided, scikit-learn defaults are used.</span>
-<span class="sd">                If no defaults are available, an exception is raised.</span>
-<span class="sd">                Refer to the &#39;n_splits&#39; argument as &#39;folds&#39;.</span>
+<span class="sd">              required by chosen class as specified in scikit-learn documentation.</span>
+<span class="sd">              If arguments are not provided, scikit-learn defaults are used.</span>
+<span class="sd">              If no defaults are available, an exception is raised.</span>
+<span class="sd">              Refer to the &#39;n_splits&#39; argument as &#39;folds&#39;.</span>
 
 <span class="sd">        tmp_folder : string, optional (None)</span>
 <span class="sd">            folder to store configuration output and log files, if ``None``</span>
@@ -273,13 +316,15 @@ <h1>Source code for autosklearn.estimators</h1><div class="highlight"><pre>
 
 <span class="sd">        n_jobs : int, optional, experimental</span>
 <span class="sd">            The number of jobs to run in parallel for ``fit()``. ``-1`` means</span>
-<span class="sd">            using all processors. By default, Auto-sklearn uses a single core</span>
-<span class="sd">            for fitting the machine learning model and a single core for fitting</span>
-<span class="sd">            an ensemble. Ensemble building is not affected by ``n_jobs`` but</span>
-<span class="sd">            can be controlled by the number of models in the ensemble. In</span>
-<span class="sd">            contrast to most scikit-learn models, ``n_jobs`` given in the</span>
-<span class="sd">            constructor is not applied to the ``predict()`` method. If</span>
-<span class="sd">            ``dask_client`` is None, a new dask client is created.</span>
+<span class="sd">            using all processors. </span>
+<span class="sd">            </span>
+<span class="sd">            **Important notes**: </span>
+<span class="sd">            </span>
+<span class="sd">            * By default, Auto-sklearn uses one core. </span>
+<span class="sd">            * Ensemble building is not affected by ``n_jobs`` but can be controlled by the number </span>
+<span class="sd">              of models in the ensemble.</span>
+<span class="sd">            * ``predict()`` is not affected by ``n_jobs`` (in contrast to most scikit-learn models)</span>
+<span class="sd">            * If ``dask_client`` is ``None``, a new dask client is created.</span>
 
 <span class="sd">        dask_client : dask.distributed.Client, optional</span>
 <span class="sd">            User-created dask client, can be used to start a dask cluster and then</span>
@@ -295,7 +340,7 @@ <h1>Source code for autosklearn.estimators</h1><div class="highlight"><pre>
 <span class="sd">            * ``&#39;y_optimization&#39;`` : do not save the predictions for the</span>
 <span class="sd">              optimization/validation set, which would later on be used to build</span>
 <span class="sd">              an ensemble.</span>
-<span class="sd">            * ``&#39;model&#39;`` : do not save any model files</span>
+<span class="sd">            * ``model`` : do not save any model files</span>
 
 <span class="sd">        smac_scenario_args : dict, optional (None)</span>
 <span class="sd">            Additional arguments inserted into the scenario of SMAC. See the</span>
@@ -341,13 +386,13 @@ <h1>Source code for autosklearn.estimators</h1><div class="highlight"><pre>
 <span class="sd">        Attributes</span>
 <span class="sd">        ----------</span>
 
-<span class="sd">        cv_results\_ : dict of numpy (masked) ndarrays</span>
+<span class="sd">        cv_results_ : dict of numpy (masked) ndarrays</span>
 <span class="sd">            A dict with keys as column headers and values as columns, that can be</span>
 <span class="sd">            imported into a pandas ``DataFrame``.</span>
 
 <span class="sd">            Not all keys returned by scikit-learn are supported yet.</span>
 
-<span class="sd">        performance_over_time\_ : pandas.core.frame.DataFrame</span>
+<span class="sd">        performance_over_time_ : pandas.core.frame.DataFrame</span>
 <span class="sd">            A ``DataFrame`` containing the models performance over time data. Can be</span>
 <span class="sd">            used for plotting directly. Please refer to the example</span>
 <span class="sd">            :ref:`Train and Test Inputs &lt;sphx_glr_examples_40_advanced_example_pandas_train_test.py&gt;`.</span>
@@ -605,13 +650,74 @@ <h1>Source code for autosklearn.estimators</h1><div class="highlight"><pre>
         <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">automl_</span><span class="o">.</span><span class="n">score</span><span class="p">(</span><span class="n">X</span><span class="p">,</span> <span class="n">y</span><span class="p">)</span>
 
     <span class="k">def</span> <span class="nf">show_models</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
-        <span class="sd">&quot;&quot;&quot;Return a representation of the final ensemble found by auto-sklearn.</span>
+        <span class="sd">&quot;&quot;&quot; Returns a dictionary containing dictionaries of ensemble models.</span>
+
+<span class="sd">        Each model in the ensemble can be accessed by giving its ``model_id`` as key.</span>
+
+<span class="sd">        A model dictionary contains the following:</span>
+
+<span class="sd">        * ``&quot;model_id&quot;`` - The id given to a model by ``autosklearn``.</span>
+<span class="sd">        * ``&quot;rank&quot;`` - The rank of the model based on it&#39;s ``&quot;cost&quot;``.</span>
+<span class="sd">        * ``&quot;cost&quot;`` - The loss of the model on the validation set.</span>
+<span class="sd">        * ``&quot;ensemble_weight&quot;`` - The weight given to the model in the ensemble.</span>
+<span class="sd">        * ``&quot;voting_model&quot;`` - The ``cv_voting_ensemble`` model (for &#39;cv&#39; resampling).</span>
+<span class="sd">        * ``&quot;estimators&quot;`` - List of models (dicts) in ``cv_voting_ensemble`` (for &#39;cv&#39; resampling).</span>
+<span class="sd">        * ``&quot;data_preprocessor&quot;`` - The preprocessor used on the data.</span>
+<span class="sd">        * ``&quot;balancing&quot;`` - The balancing used on the data (for classification).</span>
+<span class="sd">        * ``&quot;feature_preprocessor&quot;`` - The preprocessor for features types.</span>
+<span class="sd">        * ``&quot;classifier&quot;`` or ``&quot;regressor&quot;`` - The autosklearn wrapped classifier or regressor.</span>
+<span class="sd">        * ``&quot;sklearn_classifier&quot;`` or ``&quot;sklearn_regressor&quot;`` - The sklearn classifier or regressor.</span>
+
+<span class="sd">        **Example**</span>
+
+<span class="sd">        .. code-block:: python</span>
+
+<span class="sd">            import sklearn.datasets</span>
+<span class="sd">            import sklearn.metrics</span>
+<span class="sd">            import autosklearn.regression</span>
+
+<span class="sd">            X, y = sklearn.datasets.load_diabetes(return_X_y=True)</span>
+
+<span class="sd">            automl = autosklearn.regression.AutoSklearnRegressor(</span>
+<span class="sd">                time_left_for_this_task=120</span>
+<span class="sd">                )</span>
+<span class="sd">            automl.fit(X_train, y_train, dataset_name=&#39;diabetes&#39;)</span>
+
+<span class="sd">            ensemble_dict = automl.show_models()</span>
+<span class="sd">            print(ensemble_dict)</span>
+
+<span class="sd">        Output:</span>
+
+<span class="sd">        .. code-block:: text</span>
+
+<span class="sd">            {</span>
+<span class="sd">                25: {&#39;model_id&#39;: 25.0,</span>
+<span class="sd">                     &#39;rank&#39;: 1,</span>
+<span class="sd">                     &#39;cost&#39;: 0.43667876507897496,</span>
+<span class="sd">                     &#39;ensemble_weight&#39;: 0.38,</span>
+<span class="sd">                     &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing....&gt;,</span>
+<span class="sd">                     &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components....&gt;,</span>
+<span class="sd">                     &#39;regressor&#39;: &lt;autosklearn.pipeline.components.regression....&gt;,</span>
+<span class="sd">                     &#39;sklearn_regressor&#39;: SGDRegressor(alpha=0.0006517033225329654,...)</span>
+<span class="sd">                    },</span>
+<span class="sd">                6: {&#39;model_id&#39;: 6.0,</span>
+<span class="sd">                    &#39;rank&#39;: 2,</span>
+<span class="sd">                    &#39;cost&#39;: 0.4550418898836528,</span>
+<span class="sd">                    &#39;ensemble_weight&#39;: 0.3,</span>
+<span class="sd">                    &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing....&gt;,</span>
+<span class="sd">                    &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components....&gt;,</span>
+<span class="sd">                    &#39;regressor&#39;: &lt;autosklearn.pipeline.components.regression....&gt;,</span>
+<span class="sd">                    &#39;sklearn_regressor&#39;: ARDRegression(alpha_1=0.0003701926442639788,...)</span>
+<span class="sd">                    }...</span>
+<span class="sd">            }</span>
 
 <span class="sd">        Returns</span>
 <span class="sd">        -------</span>
-<span class="sd">        str</span>
+<span class="sd">        Dict(int, Any) : dictionary of length = number of models in the ensemble</span>
+<span class="sd">            A dictionary of models in the ensemble, where ``model_id`` is the key.</span>
 
 <span class="sd">        &quot;&quot;&quot;</span>
+
         <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">automl_</span><span class="o">.</span><span class="n">show_models</span><span class="p">()</span>
 
     <span class="k">def</span> <span class="nf">get_models_with_weights</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
@@ -672,7 +778,7 @@ <h1>Source code for autosklearn.estimators</h1><div class="highlight"><pre>
 <span class="sd">        Gives an overview of all models trained during the search process along</span>
 <span class="sd">        with various statistics about their training.</span>
 
-<span class="sd">        The availble statistics are:</span>
+<span class="sd">        The available statistics are:</span>
 
 <span class="sd">        **Simple**:</span>
 
@@ -1239,7 +1345,7 @@ <h1>Source code for autosklearn.estimators</h1><div class="highlight"><pre>
       
     </p>
     <p>
-        &copy; Copyright 2014-2021, Machine Learning Professorship Freiburg.<br/>
+        &copy; Copyright 2014-2022, Machine Learning Professorship Freiburg.<br/>
       Created using <a href="http://sphinx-doc.org/">Sphinx</a> 4.2.0.<br/>
     </p>
   </div>
diff --git a/master/_modules/autosklearn/experimental/askl2.html b/master/_modules/autosklearn/experimental/askl2.html
index 6d483574b5..aa69a03f10 100644
--- a/master/_modules/autosklearn/experimental/askl2.html
+++ b/master/_modules/autosklearn/experimental/askl2.html
@@ -4,7 +4,7 @@
   <head>
     <meta charset="utf-8" />
     <meta name="viewport" content="width=device-width, initial-scale=1.0" />
-    <title>autosklearn.experimental.askl2 &#8212; AutoSklearn 0.14.2 documentation</title>
+    <title>autosklearn.experimental.askl2 &#8212; AutoSklearn 0.14.4 documentation</title>
     <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css" />
     <link rel="stylesheet" type="text/css" href="../../../_static/bootstrap-sphinx.css" />
     <link rel="stylesheet" type="text/css" href="../../../_static/sg_gallery.css" />
@@ -54,7 +54,7 @@
         </button>
         <a class="navbar-brand" href="../../../index.html">
           auto-sklearn</a>
-        <span class="navbar-text navbar-version pull-left"><b>0.14.2</b></span>
+        <span class="navbar-text navbar-version pull-left"><b>0.14.4</b></span>
       </div>
 
         <div class="collapse navbar-collapse nav-collapse">
@@ -331,7 +331,7 @@ <h1>Source code for autosklearn.experimental.askl2</h1><div class="highlight"><p
 <span class="sd">        ensemble_size : int, optional (default=50)</span>
 <span class="sd">            Number of models added to the ensemble built by *Ensemble</span>
 <span class="sd">            selection from libraries of models*. Models are drawn with</span>
-<span class="sd">            replacement.</span>
+<span class="sd">            replacement. If set to ``0`` no ensemble is fit.</span>
 
 <span class="sd">        ensemble_nbest : int, optional (default=50)</span>
 <span class="sd">            Only consider the ``ensemble_nbest`` models when building an</span>
@@ -351,10 +351,14 @@ <h1>Source code for autosklearn.experimental.askl2</h1><div class="highlight"><p
 <span class="sd">        memory_limit : int, optional (3072)</span>
 <span class="sd">            Memory limit in MB for the machine learning algorithm.</span>
 <span class="sd">            `auto-sklearn` will stop fitting the machine learning algorithm if</span>
-<span class="sd">            it tries to allocate more than `memory_limit` MB.</span>
-<span class="sd">            If None is provided, no memory limit is set.</span>
-<span class="sd">            In case of multi-processing, `memory_limit` will be per job.</span>
-<span class="sd">            This memory limit also applies to the ensemble creation process.</span>
+<span class="sd">            it tries to allocate more than ``memory_limit`` MB.</span>
+<span class="sd">            </span>
+<span class="sd">            **Important notes:** </span>
+<span class="sd">            </span>
+<span class="sd">            * If ``None`` is provided, no memory limit is set.</span>
+<span class="sd">            * In case of multi-processing, ``memory_limit`` will be *per job*, so the total usage is </span>
+<span class="sd">              ``n_jobs x memory_limit``.</span>
+<span class="sd">            * The memory limit also applies to the ensemble creation process.</span>
 
 <span class="sd">        tmp_folder : string, optional (None)</span>
 <span class="sd">            folder to store configuration output and log files, if ``None``</span>
@@ -366,13 +370,15 @@ <h1>Source code for autosklearn.experimental.askl2</h1><div class="highlight"><p
 
 <span class="sd">        n_jobs : int, optional, experimental</span>
 <span class="sd">            The number of jobs to run in parallel for ``fit()``. ``-1`` means</span>
-<span class="sd">            using all processors. By default, Auto-sklearn uses a single core</span>
-<span class="sd">            for fitting the machine learning model and a single core for fitting</span>
-<span class="sd">            an ensemble. Ensemble building is not affected by ``n_jobs`` but</span>
-<span class="sd">            can be controlled by the number of models in the ensemble. In</span>
-<span class="sd">            contrast to most scikit-learn models, ``n_jobs`` given in the</span>
-<span class="sd">            constructor is not applied to the ``predict()`` method. If</span>
-<span class="sd">            ``dask_client`` is None, a new dask client is created.</span>
+<span class="sd">            using all processors. </span>
+<span class="sd">            </span>
+<span class="sd">            **Important notes**: </span>
+<span class="sd">            </span>
+<span class="sd">            * By default, Auto-sklearn uses one core. </span>
+<span class="sd">            * Ensemble building is not affected by ``n_jobs`` but can be controlled by the number </span>
+<span class="sd">              of models in the ensemble.</span>
+<span class="sd">            * ``predict()`` is not affected by ``n_jobs`` (in contrast to most scikit-learn models)</span>
+<span class="sd">            * If ``dask_client`` is ``None``, a new dask client is created.</span>
 
 <span class="sd">        dask_client : dask.distributed.Client, optional</span>
 <span class="sd">            User-created dask client, can be used to start a dask cluster and then</span>
@@ -388,7 +394,7 @@ <h1>Source code for autosklearn.experimental.askl2</h1><div class="highlight"><p
 <span class="sd">            * ``&#39;y_optimization&#39;`` : do not save the predictions for the</span>
 <span class="sd">              optimization/validation set, which would later on be used to build</span>
 <span class="sd">              an ensemble.</span>
-<span class="sd">            * ``&#39;model&#39;`` : do not save any model files</span>
+<span class="sd">            * ``model`` : do not save any model files</span>
 
 <span class="sd">        smac_scenario_args : dict, optional (None)</span>
 <span class="sd">            Additional arguments inserted into the scenario of SMAC. See the</span>
@@ -588,7 +594,7 @@ <h1>Source code for autosklearn.experimental.askl2</h1><div class="highlight"><p
       
     </p>
     <p>
-        &copy; Copyright 2014-2021, Machine Learning Professorship Freiburg.<br/>
+        &copy; Copyright 2014-2022, Machine Learning Professorship Freiburg.<br/>
       Created using <a href="http://sphinx-doc.org/">Sphinx</a> 4.2.0.<br/>
     </p>
   </div>
diff --git a/master/_modules/autosklearn/metrics.html b/master/_modules/autosklearn/metrics.html
index 35c8615d1c..f61c30f5fd 100644
--- a/master/_modules/autosklearn/metrics.html
+++ b/master/_modules/autosklearn/metrics.html
@@ -4,7 +4,7 @@
   <head>
     <meta charset="utf-8" />
     <meta name="viewport" content="width=device-width, initial-scale=1.0" />
-    <title>autosklearn.metrics &#8212; AutoSklearn 0.14.2 documentation</title>
+    <title>autosklearn.metrics &#8212; AutoSklearn 0.14.4 documentation</title>
     <link rel="stylesheet" type="text/css" href="../../_static/pygments.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/bootstrap-sphinx.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/sg_gallery.css" />
@@ -54,7 +54,7 @@
         </button>
         <a class="navbar-brand" href="../../index.html">
           auto-sklearn</a>
-        <span class="navbar-text navbar-version pull-left"><b>0.14.2</b></span>
+        <span class="navbar-text navbar-version pull-left"><b>0.14.4</b></span>
       </div>
 
         <div class="collapse navbar-collapse nav-collapse">
@@ -113,6 +113,7 @@
   <h1>Source code for autosklearn.metrics</h1><div class="highlight"><pre>
 <span></span><span class="kn">from</span> <span class="nn">abc</span> <span class="kn">import</span> <span class="n">ABCMeta</span><span class="p">,</span> <span class="n">abstractmethod</span>
 <span class="kn">from</span> <span class="nn">functools</span> <span class="kn">import</span> <span class="n">partial</span>
+<span class="kn">from</span> <span class="nn">itertools</span> <span class="kn">import</span> <span class="n">product</span>
 <span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Any</span><span class="p">,</span> <span class="n">Callable</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Union</span><span class="p">,</span> <span class="n">cast</span>
 
 <span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
@@ -391,16 +392,14 @@ <h1>Source code for autosklearn.metrics</h1><div class="highlight"><pre>
                                     <span class="n">optimum</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span>
                                     <span class="n">worst_possible_result</span><span class="o">=</span><span class="n">MAXINT</span><span class="p">,</span>
                                     <span class="n">greater_is_better</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
-<span class="n">r2</span> <span class="o">=</span> <span class="n">make_scorer</span><span class="p">(</span><span class="s1">&#39;r2&#39;</span><span class="p">,</span>
-                 <span class="n">sklearn</span><span class="o">.</span><span class="n">metrics</span><span class="o">.</span><span class="n">r2_score</span><span class="p">)</span>
+
+<span class="n">r2</span> <span class="o">=</span> <span class="n">make_scorer</span><span class="p">(</span><span class="s1">&#39;r2&#39;</span><span class="p">,</span> <span class="n">sklearn</span><span class="o">.</span><span class="n">metrics</span><span class="o">.</span><span class="n">r2_score</span><span class="p">)</span>
 
 <span class="c1"># Standard Classification Scores</span>
 <span class="n">accuracy</span> <span class="o">=</span> <span class="n">make_scorer</span><span class="p">(</span><span class="s1">&#39;accuracy&#39;</span><span class="p">,</span>
                        <span class="n">sklearn</span><span class="o">.</span><span class="n">metrics</span><span class="o">.</span><span class="n">accuracy_score</span><span class="p">)</span>
 <span class="n">balanced_accuracy</span> <span class="o">=</span> <span class="n">make_scorer</span><span class="p">(</span><span class="s1">&#39;balanced_accuracy&#39;</span><span class="p">,</span>
                                 <span class="n">sklearn</span><span class="o">.</span><span class="n">metrics</span><span class="o">.</span><span class="n">balanced_accuracy_score</span><span class="p">)</span>
-<span class="n">f1</span> <span class="o">=</span> <span class="n">make_scorer</span><span class="p">(</span><span class="s1">&#39;f1&#39;</span><span class="p">,</span>
-                 <span class="n">sklearn</span><span class="o">.</span><span class="n">metrics</span><span class="o">.</span><span class="n">f1_score</span><span class="p">)</span>
 
 <span class="c1"># Score functions that need decision values</span>
 <span class="n">roc_auc</span> <span class="o">=</span> <span class="n">make_scorer</span><span class="p">(</span><span class="s1">&#39;roc_auc&#39;</span><span class="p">,</span>
@@ -410,10 +409,20 @@ <h1>Source code for autosklearn.metrics</h1><div class="highlight"><pre>
 <span class="n">average_precision</span> <span class="o">=</span> <span class="n">make_scorer</span><span class="p">(</span><span class="s1">&#39;average_precision&#39;</span><span class="p">,</span>
                                 <span class="n">sklearn</span><span class="o">.</span><span class="n">metrics</span><span class="o">.</span><span class="n">average_precision_score</span><span class="p">,</span>
                                 <span class="n">needs_threshold</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
-<span class="n">precision</span> <span class="o">=</span> <span class="n">make_scorer</span><span class="p">(</span><span class="s1">&#39;precision&#39;</span><span class="p">,</span>
-                        <span class="n">sklearn</span><span class="o">.</span><span class="n">metrics</span><span class="o">.</span><span class="n">precision_score</span><span class="p">)</span>
-<span class="n">recall</span> <span class="o">=</span> <span class="n">make_scorer</span><span class="p">(</span><span class="s1">&#39;recall&#39;</span><span class="p">,</span>
-                     <span class="n">sklearn</span><span class="o">.</span><span class="n">metrics</span><span class="o">.</span><span class="n">recall_score</span><span class="p">)</span>
+
+<span class="c1"># NOTE: zero_division</span>
+<span class="c1">#</span>
+<span class="c1">#   Specified as the explicit default, see sklearn docs:</span>
+<span class="c1">#   https://scikit-learn.org/stable/modules/generated/sklearn.metrics.precision_score.html#sklearn-metrics-precision-score</span>
+<span class="n">precision</span> <span class="o">=</span> <span class="n">make_scorer</span><span class="p">(</span>
+    <span class="s1">&#39;precision&#39;</span><span class="p">,</span> <span class="n">partial</span><span class="p">(</span><span class="n">sklearn</span><span class="o">.</span><span class="n">metrics</span><span class="o">.</span><span class="n">precision_score</span><span class="p">,</span> <span class="n">zero_division</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
+<span class="p">)</span>
+<span class="n">recall</span> <span class="o">=</span> <span class="n">make_scorer</span><span class="p">(</span>
+    <span class="s1">&#39;recall&#39;</span><span class="p">,</span> <span class="n">partial</span><span class="p">(</span><span class="n">sklearn</span><span class="o">.</span><span class="n">metrics</span><span class="o">.</span><span class="n">recall_score</span><span class="p">,</span> <span class="n">zero_division</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
+<span class="p">)</span>
+<span class="n">f1</span> <span class="o">=</span> <span class="n">make_scorer</span><span class="p">(</span>
+    <span class="s1">&#39;f1&#39;</span><span class="p">,</span> <span class="n">partial</span><span class="p">(</span><span class="n">sklearn</span><span class="o">.</span><span class="n">metrics</span><span class="o">.</span><span class="n">f1_score</span><span class="p">,</span> <span class="n">zero_division</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
+<span class="p">)</span>
 
 <span class="c1"># Score function for probabilistic classification</span>
 <span class="n">log_loss</span> <span class="o">=</span> <span class="n">make_scorer</span><span class="p">(</span><span class="s1">&#39;log_loss&#39;</span><span class="p">,</span>
@@ -425,29 +434,39 @@ <h1>Source code for autosklearn.metrics</h1><div class="highlight"><pre>
 <span class="c1"># TODO what about mathews correlation coefficient etc?</span>
 
 
-<span class="n">REGRESSION_METRICS</span> <span class="o">=</span> <span class="nb">dict</span><span class="p">()</span>
-<span class="k">for</span> <span class="n">scorer</span> <span class="ow">in</span> <span class="p">[</span><span class="n">mean_absolute_error</span><span class="p">,</span> <span class="n">mean_squared_error</span><span class="p">,</span> <span class="n">root_mean_squared_error</span><span class="p">,</span>
-               <span class="n">mean_squared_log_error</span><span class="p">,</span> <span class="n">median_absolute_error</span><span class="p">,</span> <span class="n">r2</span><span class="p">]:</span>
-    <span class="n">REGRESSION_METRICS</span><span class="p">[</span><span class="n">scorer</span><span class="o">.</span><span class="n">name</span><span class="p">]</span> <span class="o">=</span> <span class="n">scorer</span>
-
-<span class="n">CLASSIFICATION_METRICS</span> <span class="o">=</span> <span class="nb">dict</span><span class="p">()</span>
+<span class="n">REGRESSION_METRICS</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="n">scorer</span><span class="o">.</span><span class="n">name</span><span class="p">:</span> <span class="n">scorer</span>
+    <span class="k">for</span> <span class="n">scorer</span> <span class="ow">in</span> <span class="p">[</span>
+        <span class="n">mean_absolute_error</span><span class="p">,</span> <span class="n">mean_squared_error</span><span class="p">,</span> <span class="n">root_mean_squared_error</span><span class="p">,</span>
+        <span class="n">mean_squared_log_error</span><span class="p">,</span> <span class="n">median_absolute_error</span><span class="p">,</span> <span class="n">r2</span>
+    <span class="p">]</span>
+<span class="p">}</span>
 
-<span class="k">for</span> <span class="n">scorer</span> <span class="ow">in</span> <span class="p">[</span><span class="n">accuracy</span><span class="p">,</span> <span class="n">balanced_accuracy</span><span class="p">,</span> <span class="n">roc_auc</span><span class="p">,</span> <span class="n">average_precision</span><span class="p">,</span>
-               <span class="n">log_loss</span><span class="p">]:</span>
-    <span class="n">CLASSIFICATION_METRICS</span><span class="p">[</span><span class="n">scorer</span><span class="o">.</span><span class="n">name</span><span class="p">]</span> <span class="o">=</span> <span class="n">scorer</span>
+<span class="n">CLASSIFICATION_METRICS</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="n">scorer</span><span class="o">.</span><span class="n">name</span><span class="p">:</span> <span class="n">scorer</span>
+    <span class="k">for</span> <span class="n">scorer</span> <span class="ow">in</span> <span class="p">[</span>
+        <span class="n">accuracy</span><span class="p">,</span> <span class="n">balanced_accuracy</span><span class="p">,</span> <span class="n">roc_auc</span><span class="p">,</span> <span class="n">average_precision</span><span class="p">,</span> <span class="n">log_loss</span>
+    <span class="p">]</span>
+<span class="p">}</span>
 
-<span class="k">for</span> <span class="n">name</span><span class="p">,</span> <span class="n">metric</span> <span class="ow">in</span> <span class="p">[(</span><span class="s1">&#39;precision&#39;</span><span class="p">,</span> <span class="n">sklearn</span><span class="o">.</span><span class="n">metrics</span><span class="o">.</span><span class="n">precision_score</span><span class="p">),</span>
-                     <span class="p">(</span><span class="s1">&#39;recall&#39;</span><span class="p">,</span> <span class="n">sklearn</span><span class="o">.</span><span class="n">metrics</span><span class="o">.</span><span class="n">recall_score</span><span class="p">),</span>
-                     <span class="p">(</span><span class="s1">&#39;f1&#39;</span><span class="p">,</span> <span class="n">sklearn</span><span class="o">.</span><span class="n">metrics</span><span class="o">.</span><span class="n">f1_score</span><span class="p">)]:</span>
-    <span class="nb">globals</span><span class="p">()[</span><span class="n">name</span><span class="p">]</span> <span class="o">=</span> <span class="n">make_scorer</span><span class="p">(</span><span class="n">name</span><span class="p">,</span> <span class="n">metric</span><span class="p">)</span>
-    <span class="n">CLASSIFICATION_METRICS</span><span class="p">[</span><span class="n">name</span><span class="p">]</span> <span class="o">=</span> <span class="nb">globals</span><span class="p">()[</span><span class="n">name</span><span class="p">]</span>
-    <span class="k">for</span> <span class="n">average</span> <span class="ow">in</span> <span class="p">[</span><span class="s1">&#39;macro&#39;</span><span class="p">,</span> <span class="s1">&#39;micro&#39;</span><span class="p">,</span> <span class="s1">&#39;samples&#39;</span><span class="p">,</span> <span class="s1">&#39;weighted&#39;</span><span class="p">]:</span>
-        <span class="n">qualified_name</span> <span class="o">=</span> <span class="s1">&#39;</span><span class="si">{0}</span><span class="s1">_</span><span class="si">{1}</span><span class="s1">&#39;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="n">name</span><span class="p">,</span> <span class="n">average</span><span class="p">)</span>
-        <span class="nb">globals</span><span class="p">()[</span><span class="n">qualified_name</span><span class="p">]</span> <span class="o">=</span> <span class="n">make_scorer</span><span class="p">(</span><span class="n">qualified_name</span><span class="p">,</span>
-                                                <span class="n">partial</span><span class="p">(</span><span class="n">metric</span><span class="p">,</span>
-                                                        <span class="n">pos_label</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
-                                                        <span class="n">average</span><span class="o">=</span><span class="n">average</span><span class="p">))</span>
-        <span class="n">CLASSIFICATION_METRICS</span><span class="p">[</span><span class="n">qualified_name</span><span class="p">]</span> <span class="o">=</span> <span class="nb">globals</span><span class="p">()[</span><span class="n">qualified_name</span><span class="p">]</span>
+<span class="c1"># NOTE: zero_division</span>
+<span class="c1">#</span>
+<span class="c1">#   Specified as the explicit default, see sklearn docs:</span>
+<span class="c1">#   https://scikit-learn.org/stable/modules/generated/sklearn.metrics.precision_score.html#sklearn-metrics-precision-score</span>
+<span class="k">for</span> <span class="p">(</span><span class="n">base_name</span><span class="p">,</span> <span class="n">sklearn_metric</span><span class="p">),</span> <span class="n">average</span> <span class="ow">in</span> <span class="n">product</span><span class="p">(</span>
+    <span class="p">[</span>
+        <span class="p">(</span><span class="s1">&#39;precision&#39;</span><span class="p">,</span> <span class="n">sklearn</span><span class="o">.</span><span class="n">metrics</span><span class="o">.</span><span class="n">precision_score</span><span class="p">),</span>
+        <span class="p">(</span><span class="s1">&#39;recall&#39;</span><span class="p">,</span> <span class="n">sklearn</span><span class="o">.</span><span class="n">metrics</span><span class="o">.</span><span class="n">recall_score</span><span class="p">),</span>
+        <span class="p">(</span><span class="s1">&#39;f1&#39;</span><span class="p">,</span> <span class="n">sklearn</span><span class="o">.</span><span class="n">metrics</span><span class="o">.</span><span class="n">f1_score</span><span class="p">),</span>
+    <span class="p">],</span>
+    <span class="p">[</span><span class="s1">&#39;macro&#39;</span><span class="p">,</span> <span class="s1">&#39;micro&#39;</span><span class="p">,</span> <span class="s1">&#39;samples&#39;</span><span class="p">,</span> <span class="s1">&#39;weighted&#39;</span><span class="p">]</span>
+<span class="p">):</span>
+    <span class="n">name</span> <span class="o">=</span> <span class="sa">f</span><span class="s1">&#39;</span><span class="si">{</span><span class="n">base_name</span><span class="si">}</span><span class="s1">_</span><span class="si">{</span><span class="n">average</span><span class="si">}</span><span class="s1">&#39;</span>
+    <span class="n">scorer</span> <span class="o">=</span> <span class="n">make_scorer</span><span class="p">(</span>
+        <span class="n">name</span><span class="p">,</span> <span class="n">partial</span><span class="p">(</span><span class="n">sklearn_metric</span><span class="p">,</span> <span class="n">pos_label</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">average</span><span class="o">=</span><span class="n">average</span><span class="p">,</span> <span class="n">zero_division</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
+    <span class="p">)</span>
+    <span class="nb">globals</span><span class="p">()[</span><span class="n">name</span><span class="p">]</span> <span class="o">=</span> <span class="n">scorer</span>  <span class="c1"># Adds scorer to the module scope</span>
+    <span class="n">CLASSIFICATION_METRICS</span><span class="p">[</span><span class="n">name</span><span class="p">]</span> <span class="o">=</span> <span class="n">scorer</span>
 
 
 <span class="k">def</span> <span class="nf">calculate_score</span><span class="p">(</span>
@@ -682,7 +701,7 @@ <h1>Source code for autosklearn.metrics</h1><div class="highlight"><pre>
       
     </p>
     <p>
-        &copy; Copyright 2014-2021, Machine Learning Professorship Freiburg.<br/>
+        &copy; Copyright 2014-2022, Machine Learning Professorship Freiburg.<br/>
       Created using <a href="http://sphinx-doc.org/">Sphinx</a> 4.2.0.<br/>
     </p>
   </div>
diff --git a/master/_modules/autosklearn/pipeline/components/base.html b/master/_modules/autosklearn/pipeline/components/base.html
index 654a09d1bc..99ec451b90 100644
--- a/master/_modules/autosklearn/pipeline/components/base.html
+++ b/master/_modules/autosklearn/pipeline/components/base.html
@@ -4,7 +4,7 @@
   <head>
     <meta charset="utf-8" />
     <meta name="viewport" content="width=device-width, initial-scale=1.0" />
-    <title>autosklearn.pipeline.components.base &#8212; AutoSklearn 0.14.2 documentation</title>
+    <title>autosklearn.pipeline.components.base &#8212; AutoSklearn 0.14.4 documentation</title>
     <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/bootstrap-sphinx.css" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/sg_gallery.css" />
@@ -54,7 +54,7 @@
         </button>
         <a class="navbar-brand" href="../../../../index.html">
           auto-sklearn</a>
-        <span class="navbar-text navbar-version pull-left"><b>0.14.2</b></span>
+        <span class="navbar-text navbar-version pull-left"><b>0.14.4</b></span>
       </div>
 
         <div class="collapse navbar-collapse nav-collapse">
@@ -260,13 +260,16 @@ <h1>Source code for autosklearn.pipeline.components.base</h1><div class="highlig
 
 
 <span class="k">class</span> <span class="nc">IterativeComponent</span><span class="p">(</span><span class="n">AutoSklearnComponent</span><span class="p">):</span>
+
     <span class="k">def</span> <span class="nf">fit</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">X</span><span class="p">,</span> <span class="n">y</span><span class="p">,</span> <span class="n">sample_weight</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span>
         <span class="bp">self</span><span class="o">.</span><span class="n">iterative_fit</span><span class="p">(</span><span class="n">X</span><span class="p">,</span> <span class="n">y</span><span class="p">,</span> <span class="n">n_iter</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span> <span class="n">refit</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+
         <span class="n">iteration</span> <span class="o">=</span> <span class="mi">2</span>
         <span class="k">while</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">configuration_fully_fitted</span><span class="p">():</span>
             <span class="n">n_iter</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="mi">2</span> <span class="o">**</span> <span class="n">iteration</span> <span class="o">/</span> <span class="mi">2</span><span class="p">)</span>
             <span class="bp">self</span><span class="o">.</span><span class="n">iterative_fit</span><span class="p">(</span><span class="n">X</span><span class="p">,</span> <span class="n">y</span><span class="p">,</span> <span class="n">n_iter</span><span class="o">=</span><span class="n">n_iter</span><span class="p">,</span> <span class="n">refit</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
             <span class="n">iteration</span> <span class="o">+=</span> <span class="mi">1</span>
+
         <span class="k">return</span> <span class="bp">self</span>
 
     <span class="nd">@staticmethod</span>
@@ -278,15 +281,16 @@ <h1>Source code for autosklearn.pipeline.components.base</h1><div class="highlig
 
 
 <span class="k">class</span> <span class="nc">IterativeComponentWithSampleWeight</span><span class="p">(</span><span class="n">AutoSklearnComponent</span><span class="p">):</span>
+
     <span class="k">def</span> <span class="nf">fit</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">X</span><span class="p">,</span> <span class="n">y</span><span class="p">,</span> <span class="n">sample_weight</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span>
-        <span class="bp">self</span><span class="o">.</span><span class="n">iterative_fit</span><span class="p">(</span>
-            <span class="n">X</span><span class="p">,</span> <span class="n">y</span><span class="p">,</span> <span class="n">n_iter</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span> <span class="n">refit</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span> <span class="n">sample_weight</span><span class="o">=</span><span class="n">sample_weight</span>
-        <span class="p">)</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">iterative_fit</span><span class="p">(</span><span class="n">X</span><span class="p">,</span> <span class="n">y</span><span class="p">,</span> <span class="n">n_iter</span><span class="o">=</span><span class="mi">2</span><span class="p">,</span> <span class="n">refit</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span> <span class="n">sample_weight</span><span class="o">=</span><span class="n">sample_weight</span><span class="p">)</span>
+
         <span class="n">iteration</span> <span class="o">=</span> <span class="mi">2</span>
         <span class="k">while</span> <span class="ow">not</span> <span class="bp">self</span><span class="o">.</span><span class="n">configuration_fully_fitted</span><span class="p">():</span>
             <span class="n">n_iter</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="mi">2</span> <span class="o">**</span> <span class="n">iteration</span> <span class="o">/</span> <span class="mi">2</span><span class="p">)</span>
-            <span class="bp">self</span><span class="o">.</span><span class="n">iterative_fit</span><span class="p">(</span><span class="n">X</span><span class="p">,</span> <span class="n">y</span><span class="p">,</span> <span class="n">n_iter</span><span class="o">=</span><span class="n">n_iter</span><span class="p">,</span> <span class="n">sample_weight</span><span class="o">=</span><span class="n">sample_weight</span><span class="p">)</span>
+            <span class="bp">self</span><span class="o">.</span><span class="n">iterative_fit</span><span class="p">(</span><span class="n">X</span><span class="p">,</span> <span class="n">y</span><span class="p">,</span> <span class="n">n_iter</span><span class="o">=</span><span class="n">n_iter</span><span class="p">,</span> <span class="n">refit</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="n">sample_weight</span><span class="o">=</span><span class="n">sample_weight</span><span class="p">)</span>
             <span class="n">iteration</span> <span class="o">+=</span> <span class="mi">1</span>
+
         <span class="k">return</span> <span class="bp">self</span>
 
     <span class="nd">@staticmethod</span>
@@ -558,7 +562,7 @@ <h1>Source code for autosklearn.pipeline.components.base</h1><div class="highlig
       
     </p>
     <p>
-        &copy; Copyright 2014-2021, Machine Learning Professorship Freiburg.<br/>
+        &copy; Copyright 2014-2022, Machine Learning Professorship Freiburg.<br/>
       Created using <a href="http://sphinx-doc.org/">Sphinx</a> 4.2.0.<br/>
     </p>
   </div>
diff --git a/master/_modules/autosklearn/pipeline/components/classification.html b/master/_modules/autosklearn/pipeline/components/classification.html
index 6d12006196..c1f15db362 100644
--- a/master/_modules/autosklearn/pipeline/components/classification.html
+++ b/master/_modules/autosklearn/pipeline/components/classification.html
@@ -4,7 +4,7 @@
   <head>
     <meta charset="utf-8" />
     <meta name="viewport" content="width=device-width, initial-scale=1.0" />
-    <title>autosklearn.pipeline.components.classification &#8212; AutoSklearn 0.14.2 documentation</title>
+    <title>autosklearn.pipeline.components.classification &#8212; AutoSklearn 0.14.4 documentation</title>
     <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/bootstrap-sphinx.css" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/sg_gallery.css" />
@@ -54,7 +54,7 @@
         </button>
         <a class="navbar-brand" href="../../../../index.html">
           auto-sklearn</a>
-        <span class="navbar-text navbar-version pull-left"><b>0.14.2</b></span>
+        <span class="navbar-text navbar-version pull-left"><b>0.14.4</b></span>
       </div>
 
         <div class="collapse navbar-collapse nav-collapse">
@@ -279,7 +279,7 @@ <h1>Source code for autosklearn.pipeline.components.classification</h1><div clas
       
     </p>
     <p>
-        &copy; Copyright 2014-2021, Machine Learning Professorship Freiburg.<br/>
+        &copy; Copyright 2014-2022, Machine Learning Professorship Freiburg.<br/>
       Created using <a href="http://sphinx-doc.org/">Sphinx</a> 4.2.0.<br/>
     </p>
   </div>
diff --git a/master/_modules/autosklearn/pipeline/components/feature_preprocessing.html b/master/_modules/autosklearn/pipeline/components/feature_preprocessing.html
index 7b644ab6e6..3bf95c5f58 100644
--- a/master/_modules/autosklearn/pipeline/components/feature_preprocessing.html
+++ b/master/_modules/autosklearn/pipeline/components/feature_preprocessing.html
@@ -4,7 +4,7 @@
   <head>
     <meta charset="utf-8" />
     <meta name="viewport" content="width=device-width, initial-scale=1.0" />
-    <title>autosklearn.pipeline.components.feature_preprocessing &#8212; AutoSklearn 0.14.2 documentation</title>
+    <title>autosklearn.pipeline.components.feature_preprocessing &#8212; AutoSklearn 0.14.4 documentation</title>
     <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/bootstrap-sphinx.css" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/sg_gallery.css" />
@@ -54,7 +54,7 @@
         </button>
         <a class="navbar-brand" href="../../../../index.html">
           auto-sklearn</a>
-        <span class="navbar-text navbar-version pull-left"><b>0.14.2</b></span>
+        <span class="navbar-text navbar-version pull-left"><b>0.14.4</b></span>
       </div>
 
         <div class="collapse navbar-collapse nav-collapse">
@@ -257,7 +257,7 @@ <h1>Source code for autosklearn.pipeline.components.feature_preprocessing</h1><d
       
     </p>
     <p>
-        &copy; Copyright 2014-2021, Machine Learning Professorship Freiburg.<br/>
+        &copy; Copyright 2014-2022, Machine Learning Professorship Freiburg.<br/>
       Created using <a href="http://sphinx-doc.org/">Sphinx</a> 4.2.0.<br/>
     </p>
   </div>
diff --git a/master/_modules/autosklearn/pipeline/components/regression.html b/master/_modules/autosklearn/pipeline/components/regression.html
index 231db7fe19..3d8232d636 100644
--- a/master/_modules/autosklearn/pipeline/components/regression.html
+++ b/master/_modules/autosklearn/pipeline/components/regression.html
@@ -4,7 +4,7 @@
   <head>
     <meta charset="utf-8" />
     <meta name="viewport" content="width=device-width, initial-scale=1.0" />
-    <title>autosklearn.pipeline.components.regression &#8212; AutoSklearn 0.14.2 documentation</title>
+    <title>autosklearn.pipeline.components.regression &#8212; AutoSklearn 0.14.4 documentation</title>
     <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/bootstrap-sphinx.css" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/sg_gallery.css" />
@@ -54,7 +54,7 @@
         </button>
         <a class="navbar-brand" href="../../../../index.html">
           auto-sklearn</a>
-        <span class="navbar-text navbar-version pull-left"><b>0.14.2</b></span>
+        <span class="navbar-text navbar-version pull-left"><b>0.14.4</b></span>
       </div>
 
         <div class="collapse navbar-collapse nav-collapse">
@@ -265,7 +265,7 @@ <h1>Source code for autosklearn.pipeline.components.regression</h1><div class="h
       
     </p>
     <p>
-        &copy; Copyright 2014-2021, Machine Learning Professorship Freiburg.<br/>
+        &copy; Copyright 2014-2022, Machine Learning Professorship Freiburg.<br/>
       Created using <a href="http://sphinx-doc.org/">Sphinx</a> 4.2.0.<br/>
     </p>
   </div>
diff --git a/master/_modules/index.html b/master/_modules/index.html
index 51c6cd4873..2bbb002bf4 100644
--- a/master/_modules/index.html
+++ b/master/_modules/index.html
@@ -4,7 +4,7 @@
   <head>
     <meta charset="utf-8" />
     <meta name="viewport" content="width=device-width, initial-scale=1.0" />
-    <title>Overview: module code &#8212; AutoSklearn 0.14.2 documentation</title>
+    <title>Overview: module code &#8212; AutoSklearn 0.14.4 documentation</title>
     <link rel="stylesheet" type="text/css" href="../_static/pygments.css" />
     <link rel="stylesheet" type="text/css" href="../_static/bootstrap-sphinx.css" />
     <link rel="stylesheet" type="text/css" href="../_static/sg_gallery.css" />
@@ -54,7 +54,7 @@
         </button>
         <a class="navbar-brand" href="../index.html">
           auto-sklearn</a>
-        <span class="navbar-text navbar-version pull-left"><b>0.14.2</b></span>
+        <span class="navbar-text navbar-version pull-left"><b>0.14.4</b></span>
       </div>
 
         <div class="collapse navbar-collapse nav-collapse">
@@ -134,7 +134,7 @@ <h1>All modules for which code is available</h1>
       
     </p>
     <p>
-        &copy; Copyright 2014-2021, Machine Learning Professorship Freiburg.<br/>
+        &copy; Copyright 2014-2022, Machine Learning Professorship Freiburg.<br/>
       Created using <a href="http://sphinx-doc.org/">Sphinx</a> 4.2.0.<br/>
     </p>
   </div>
diff --git a/master/_sources/examples/20_basic/example_classification.rst.txt b/master/_sources/examples/20_basic/example_classification.rst.txt
index 8486a1e513..691387a3d8 100644
--- a/master/_sources/examples/20_basic/example_classification.rst.txt
+++ b/master/_sources/examples/20_basic/example_classification.rst.txt
@@ -25,10 +25,12 @@ Classification
 The following example shows how to fit a simple classification model with
 *auto-sklearn*.
 
-.. GENERATED FROM PYTHON SOURCE LINES 10-16
+.. GENERATED FROM PYTHON SOURCE LINES 10-18
 
 .. code-block:: default
 
+    from pprint import pprint
+
     import sklearn.datasets
     import sklearn.metrics
 
@@ -42,12 +44,12 @@ The following example shows how to fit a simple classification model with
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 17-19
+.. GENERATED FROM PYTHON SOURCE LINES 19-21
 
 Data Loading
 ============
 
-.. GENERATED FROM PYTHON SOURCE LINES 19-24
+.. GENERATED FROM PYTHON SOURCE LINES 21-26
 
 .. code-block:: default
 
@@ -63,12 +65,12 @@ Data Loading
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 25-27
+.. GENERATED FROM PYTHON SOURCE LINES 27-29
 
 Build and fit a classifier
 ==========================
 
-.. GENERATED FROM PYTHON SOURCE LINES 27-35
+.. GENERATED FROM PYTHON SOURCE LINES 29-37
 
 .. code-block:: default
 
@@ -90,18 +92,22 @@ Build and fit a classifier
 
  .. code-block:: none
 
+    /home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:68: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+      self.metafeatures = self.metafeatures.append(metafeatures)
+    /home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:72: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+      self.algorithm_runs[metric].append(runs)
 
     AutoSklearnClassifier(per_run_time_limit=30, time_left_for_this_task=120,
                           tmp_folder='/tmp/autosklearn_classification_example_tmp')
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 36-38
+.. GENERATED FROM PYTHON SOURCE LINES 38-40
 
 View the models found by auto-sklearn
 =====================================
 
-.. GENERATED FROM PYTHON SOURCE LINES 38-41
+.. GENERATED FROM PYTHON SOURCE LINES 40-43
 
 .. code-block:: default
 
@@ -120,40 +126,51 @@ View the models found by auto-sklearn
 
               rank  ensemble_weight                type      cost  duration
     model_id                                                               
-    34           1             0.08         extra_trees  0.014184  1.673455
-    7            2             0.06         extra_trees  0.014184  1.503079
-    29           3             0.06         extra_trees  0.021277  1.718799
-    16           4             0.02   gradient_boosting  0.021277  1.020501
-    3            5             0.08                 mlp  0.028369  1.030776
-    22           6             0.04   gradient_boosting  0.028369  1.163809
-    26           7             0.02         extra_trees  0.028369  2.200900
-    6            8             0.02                 mlp  0.028369  1.111869
-    10           9             0.04       random_forest  0.028369  1.950961
-    5           10             0.04       random_forest  0.035461  2.045974
-    9           11             0.06         extra_trees  0.042553  1.824072
-    27          12             0.04         extra_trees  0.042553  2.006197
-    30          13             0.02       liblinear_svc  0.042553  0.949611
-    32          14             0.12         extra_trees  0.049645  1.795890
-    40          15             0.12          libsvm_svc  0.049645  0.821530
-    33          16             0.02       random_forest  0.056738  1.952558
-    20          17             0.02  passive_aggressive  0.078014  0.691132
-    37          18             0.12          libsvm_svc  0.085106  0.968546
-    35          19             0.02         extra_trees  0.085106  1.519158
-
-
-
-
-.. GENERATED FROM PYTHON SOURCE LINES 42-44
+    39           1             0.06         extra_trees  0.014184  1.416798
+    7            2             0.04         extra_trees  0.014184  1.423290
+    21           3             0.04         extra_trees  0.021277  1.313003
+    2            4             0.04       random_forest  0.028369  1.564060
+    32           5             0.04       liblinear_svc  0.028369  0.748310
+    29           6             0.02         extra_trees  0.028369  1.389265
+    27           7             0.04         extra_trees  0.028369  8.149227
+    26           8             0.04         extra_trees  0.028369  2.153994
+    13           9             0.02   gradient_boosting  0.028369  1.305902
+    19          10             0.02         extra_trees  0.028369  2.652099
+    11          11             0.04       random_forest  0.028369  1.933282
+    6           12             0.08                 mlp  0.028369  0.973582
+    10          13             0.08       random_forest  0.028369  1.716530
+    5           14             0.04       random_forest  0.035461  1.797078
+    31          15             0.02       liblinear_svc  0.035461  0.807887
+    12          16             0.04   gradient_boosting  0.035461  1.177767
+    8           17             0.02       random_forest  0.035461  1.783555
+    18          18             0.04       random_forest  0.035461  1.805679
+    17          19             0.02   gradient_boosting  0.035461  1.504512
+    9           20             0.02         extra_trees  0.042553  1.689027
+    15          21             0.02                 mlp  0.049645  3.394691
+    23          22             0.04                 mlp  0.049645  1.976257
+    4           23             0.04                 mlp  0.056738  1.186914
+    28          24             0.02                 lda  0.063830  0.863830
+    34          25             0.02       random_forest  0.063830  1.685210
+    24          26             0.02       random_forest  0.070922  1.560245
+    41          27             0.02       decision_tree  0.070922  0.741800
+    37          28             0.02         extra_trees  0.085106  1.440288
+    43          29             0.02         extra_trees  0.092199  1.604374
+    42          30             0.02  passive_aggressive  0.120567  0.747741
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 44-46
 
 Print the final ensemble constructed by auto-sklearn
 ====================================================
 
-.. GENERATED FROM PYTHON SOURCE LINES 44-47
+.. GENERATED FROM PYTHON SOURCE LINES 46-49
 
 .. code-block:: default
 
 
-    print(automl.show_models())
+    pprint(automl.show_models(), indent=4)
 
 
 
@@ -165,169 +182,354 @@ Print the final ensemble constructed by auto-sklearn
 
  .. code-block:: none
 
-    [(0.120000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'extra_trees', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'feature_agglomeration', 'classifier:extra_trees:bootstrap': 'True', 'classifier:extra_trees:criterion': 'entropy', 'classifier:extra_trees:max_depth': 'None', 'classifier:extra_trees:max_features': 0.3871420537981852, 'classifier:extra_trees:max_leaf_nodes': 'None', 'classifier:extra_trees:min_impurity_decrease': 0.0, 'classifier:extra_trees:min_samples_leaf': 5, 'classifier:extra_trees:min_samples_split': 14, 'classifier:extra_trees:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'robust_scaler', 'feature_preprocessor:feature_agglomeration:affinity': 'manhattan', 'feature_preprocessor:feature_agglomeration:linkage': 'complete', 'feature_preprocessor:feature_agglomeration:n_clusters': 355, 'feature_preprocessor:feature_agglomeration:pooling_func': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max': 0.726182961545702, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min': 0.02260942170772907},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.120000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'libsvm_svc', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'feature_agglomeration', 'classifier:libsvm_svc:C': 0.07563955664363409, 'classifier:libsvm_svc:gamma': 1.9836355993007913, 'classifier:libsvm_svc:kernel': 'rbf', 'classifier:libsvm_svc:max_iter': -1, 'classifier:libsvm_svc:shrinking': 'True', 'classifier:libsvm_svc:tol': 0.005003173341372008, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'minmax', 'feature_preprocessor:feature_agglomeration:affinity': 'cosine', 'feature_preprocessor:feature_agglomeration:linkage': 'complete', 'feature_preprocessor:feature_agglomeration:n_clusters': 58, 'feature_preprocessor:feature_agglomeration:pooling_func': 'mean', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.15890408760884822},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.120000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'libsvm_svc', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'select_percentile_classification', 'classifier:libsvm_svc:C': 5121.989374401993, 'classifier:libsvm_svc:gamma': 0.0024386428371511353, 'classifier:libsvm_svc:kernel': 'rbf', 'classifier:libsvm_svc:max_iter': -1, 'classifier:libsvm_svc:shrinking': 'True', 'classifier:libsvm_svc:tol': 0.0706216214952021, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'median', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'robust_scaler', 'feature_preprocessor:select_percentile_classification:percentile': 90.1749860017522, 'feature_preprocessor:select_percentile_classification:score_func': 'chi2', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.00017888891850981892, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max': 0.7393440007191528, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min': 0.04986510953238},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.080000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'extra_trees', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'polynomial', 'classifier:extra_trees:bootstrap': 'False', 'classifier:extra_trees:criterion': 'gini', 'classifier:extra_trees:max_depth': 'None', 'classifier:extra_trees:max_features': 0.4791448484072812, 'classifier:extra_trees:max_leaf_nodes': 'None', 'classifier:extra_trees:min_impurity_decrease': 0.0, 'classifier:extra_trees:min_samples_leaf': 1, 'classifier:extra_trees:min_samples_split': 6, 'classifier:extra_trees:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'none', 'feature_preprocessor:polynomial:degree': 3, 'feature_preprocessor:polynomial:include_bias': 'True', 'feature_preprocessor:polynomial:interaction_only': 'True', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.17184142431776436},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.080000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'mlp', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'feature_agglomeration', 'classifier:mlp:activation': 'tanh', 'classifier:mlp:alpha': 0.0001363185819149026, 'classifier:mlp:batch_size': 'auto', 'classifier:mlp:beta_1': 0.9, 'classifier:mlp:beta_2': 0.999, 'classifier:mlp:early_stopping': 'valid', 'classifier:mlp:epsilon': 1e-08, 'classifier:mlp:hidden_layer_depth': 3, 'classifier:mlp:learning_rate_init': 0.00018009776276177523, 'classifier:mlp:n_iter_no_change': 32, 'classifier:mlp:num_nodes_per_layer': 115, 'classifier:mlp:shuffle': 'True', 'classifier:mlp:solver': 'adam', 'classifier:mlp:tol': 0.0001, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'median', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize', 'feature_preprocessor:feature_agglomeration:affinity': 'euclidean', 'feature_preprocessor:feature_agglomeration:linkage': 'ward', 'feature_preprocessor:feature_agglomeration:n_clusters': 182, 'feature_preprocessor:feature_agglomeration:pooling_func': 'mean', 'classifier:mlp:validation_fraction': 0.1},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.060000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'extra_trees', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'polynomial', 'classifier:extra_trees:bootstrap': 'False', 'classifier:extra_trees:criterion': 'gini', 'classifier:extra_trees:max_depth': 'None', 'classifier:extra_trees:max_features': 0.5707983257382487, 'classifier:extra_trees:max_leaf_nodes': 'None', 'classifier:extra_trees:min_impurity_decrease': 0.0, 'classifier:extra_trees:min_samples_leaf': 3, 'classifier:extra_trees:min_samples_split': 11, 'classifier:extra_trees:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'median', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'none', 'feature_preprocessor:polynomial:degree': 2, 'feature_preprocessor:polynomial:include_bias': 'False', 'feature_preprocessor:polynomial:interaction_only': 'False'},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.060000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'extra_trees', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:extra_trees:bootstrap': 'False', 'classifier:extra_trees:criterion': 'gini', 'classifier:extra_trees:max_depth': 'None', 'classifier:extra_trees:max_features': 0.661377361743419, 'classifier:extra_trees:max_leaf_nodes': 'None', 'classifier:extra_trees:min_impurity_decrease': 0.0, 'classifier:extra_trees:min_samples_leaf': 2, 'classifier:extra_trees:min_samples_split': 13, 'classifier:extra_trees:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'none'},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.060000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'extra_trees', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'extra_trees_preproc_for_classification', 'classifier:extra_trees:bootstrap': 'False', 'classifier:extra_trees:criterion': 'gini', 'classifier:extra_trees:max_depth': 'None', 'classifier:extra_trees:max_features': 0.9292309396985746, 'classifier:extra_trees:max_leaf_nodes': 'None', 'classifier:extra_trees:min_impurity_decrease': 0.0, 'classifier:extra_trees:min_samples_leaf': 1, 'classifier:extra_trees:min_samples_split': 10, 'classifier:extra_trees:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'median', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'normalize', 'feature_preprocessor:extra_trees_preproc_for_classification:bootstrap': 'False', 'feature_preprocessor:extra_trees_preproc_for_classification:criterion': 'entropy', 'feature_preprocessor:extra_trees_preproc_for_classification:max_depth': 'None', 'feature_preprocessor:extra_trees_preproc_for_classification:max_features': 0.9929881254946676, 'feature_preprocessor:extra_trees_preproc_for_classification:max_leaf_nodes': 'None', 'feature_preprocessor:extra_trees_preproc_for_classification:min_impurity_decrease': 0.0, 'feature_preprocessor:extra_trees_preproc_for_classification:min_samples_leaf': 1, 'feature_preprocessor:extra_trees_preproc_for_classification:min_samples_split': 2, 'feature_preprocessor:extra_trees_preproc_for_classification:min_weight_fraction_leaf': 0.0, 'feature_preprocessor:extra_trees_preproc_for_classification:n_estimators': 100, 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.032719158639429445},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.040000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'random_forest', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'select_percentile_classification', 'classifier:random_forest:bootstrap': 'True', 'classifier:random_forest:criterion': 'entropy', 'classifier:random_forest:max_depth': 'None', 'classifier:random_forest:max_features': 0.49138075723513286, 'classifier:random_forest:max_leaf_nodes': 'None', 'classifier:random_forest:min_impurity_decrease': 0.0, 'classifier:random_forest:min_samples_leaf': 1, 'classifier:random_forest:min_samples_split': 6, 'classifier:random_forest:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'minmax', 'feature_preprocessor:select_percentile_classification:percentile': 56.97947373958566, 'feature_preprocessor:select_percentile_classification:score_func': 'mutual_info', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.42693600390988135},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.040000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'gradient_boosting', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'extra_trees_preproc_for_classification', 'classifier:gradient_boosting:early_stop': 'valid', 'classifier:gradient_boosting:l2_regularization': 8.057778875694463e-05, 'classifier:gradient_boosting:learning_rate': 0.09179220974965213, 'classifier:gradient_boosting:loss': 'auto', 'classifier:gradient_boosting:max_bins': 255, 'classifier:gradient_boosting:max_depth': 'None', 'classifier:gradient_boosting:max_leaf_nodes': 200, 'classifier:gradient_boosting:min_samples_leaf': 20, 'classifier:gradient_boosting:scoring': 'loss', 'classifier:gradient_boosting:tol': 1e-07, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'quantile_transformer', 'feature_preprocessor:extra_trees_preproc_for_classification:bootstrap': 'False', 'feature_preprocessor:extra_trees_preproc_for_classification:criterion': 'gini', 'feature_preprocessor:extra_trees_preproc_for_classification:max_depth': 'None', 'feature_preprocessor:extra_trees_preproc_for_classification:max_features': 0.9984367650965825, 'feature_preprocessor:extra_trees_preproc_for_classification:max_leaf_nodes': 'None', 'feature_preprocessor:extra_trees_preproc_for_classification:min_impurity_decrease': 0.0, 'feature_preprocessor:extra_trees_preproc_for_classification:min_samples_leaf': 13, 'feature_preprocessor:extra_trees_preproc_for_classification:min_samples_split': 18, 'feature_preprocessor:extra_trees_preproc_for_classification:min_weight_fraction_leaf': 0.0, 'feature_preprocessor:extra_trees_preproc_for_classification:n_estimators': 100, 'classifier:gradient_boosting:n_iter_no_change': 18, 'classifier:gradient_boosting:validation_fraction': 0.14295295806077554, 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.008542188583124829, 'data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles': 1102, 'data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution': 'uniform'},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.040000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'random_forest', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'extra_trees_preproc_for_classification', 'classifier:random_forest:bootstrap': 'True', 'classifier:random_forest:criterion': 'entropy', 'classifier:random_forest:max_depth': 'None', 'classifier:random_forest:max_features': 0.4775492074518431, 'classifier:random_forest:max_leaf_nodes': 'None', 'classifier:random_forest:min_impurity_decrease': 0.0, 'classifier:random_forest:min_samples_leaf': 2, 'classifier:random_forest:min_samples_split': 2, 'classifier:random_forest:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'robust_scaler', 'feature_preprocessor:extra_trees_preproc_for_classification:bootstrap': 'False', 'feature_preprocessor:extra_trees_preproc_for_classification:criterion': 'entropy', 'feature_preprocessor:extra_trees_preproc_for_classification:max_depth': 'None', 'feature_preprocessor:extra_trees_preproc_for_classification:max_features': 0.5662900693317384, 'feature_preprocessor:extra_trees_preproc_for_classification:max_leaf_nodes': 'None', 'feature_preprocessor:extra_trees_preproc_for_classification:min_impurity_decrease': 0.0, 'feature_preprocessor:extra_trees_preproc_for_classification:min_samples_leaf': 1, 'feature_preprocessor:extra_trees_preproc_for_classification:min_samples_split': 7, 'feature_preprocessor:extra_trees_preproc_for_classification:min_weight_fraction_leaf': 0.0, 'feature_preprocessor:extra_trees_preproc_for_classification:n_estimators': 100, 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.015996368052062886, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max': 0.7845396961078424, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min': 0.25545052141264185},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.040000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'extra_trees', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'polynomial', 'classifier:extra_trees:bootstrap': 'True', 'classifier:extra_trees:criterion': 'entropy', 'classifier:extra_trees:max_depth': 'None', 'classifier:extra_trees:max_features': 0.3793256905143867, 'classifier:extra_trees:max_leaf_nodes': 'None', 'classifier:extra_trees:min_impurity_decrease': 0.0, 'classifier:extra_trees:min_samples_leaf': 20, 'classifier:extra_trees:min_samples_split': 5, 'classifier:extra_trees:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'none', 'feature_preprocessor:polynomial:degree': 3, 'feature_preprocessor:polynomial:include_bias': 'False', 'feature_preprocessor:polynomial:interaction_only': 'True', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.00010083678209087315},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.020000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'gradient_boosting', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'select_percentile_classification', 'classifier:gradient_boosting:early_stop': 'train', 'classifier:gradient_boosting:l2_regularization': 3.387912939529945e-10, 'classifier:gradient_boosting:learning_rate': 0.30755227194768237, 'classifier:gradient_boosting:loss': 'auto', 'classifier:gradient_boosting:max_bins': 255, 'classifier:gradient_boosting:max_depth': 'None', 'classifier:gradient_boosting:max_leaf_nodes': 60, 'classifier:gradient_boosting:min_samples_leaf': 39, 'classifier:gradient_boosting:scoring': 'loss', 'classifier:gradient_boosting:tol': 1e-07, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'none', 'feature_preprocessor:select_percentile_classification:percentile': 93.39844669585806, 'feature_preprocessor:select_percentile_classification:score_func': 'f_classif', 'classifier:gradient_boosting:n_iter_no_change': 18, 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.010000000000000004},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.020000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'mlp', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'fast_ica', 'classifier:mlp:activation': 'relu', 'classifier:mlp:alpha': 0.0017940473175767063, 'classifier:mlp:batch_size': 'auto', 'classifier:mlp:beta_1': 0.9, 'classifier:mlp:beta_2': 0.999, 'classifier:mlp:early_stopping': 'valid', 'classifier:mlp:epsilon': 1e-08, 'classifier:mlp:hidden_layer_depth': 2, 'classifier:mlp:learning_rate_init': 0.0004684917334431039, 'classifier:mlp:n_iter_no_change': 32, 'classifier:mlp:num_nodes_per_layer': 101, 'classifier:mlp:shuffle': 'True', 'classifier:mlp:solver': 'adam', 'classifier:mlp:tol': 0.0001, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize', 'feature_preprocessor:fast_ica:algorithm': 'parallel', 'feature_preprocessor:fast_ica:fun': 'exp', 'feature_preprocessor:fast_ica:whiten': 'False', 'classifier:mlp:validation_fraction': 0.1},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.020000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'extra_trees', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'polynomial', 'classifier:extra_trees:bootstrap': 'False', 'classifier:extra_trees:criterion': 'entropy', 'classifier:extra_trees:max_depth': 'None', 'classifier:extra_trees:max_features': 0.971348315581384, 'classifier:extra_trees:max_leaf_nodes': 'None', 'classifier:extra_trees:min_impurity_decrease': 0.0, 'classifier:extra_trees:min_samples_leaf': 2, 'classifier:extra_trees:min_samples_split': 19, 'classifier:extra_trees:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'normalize', 'feature_preprocessor:polynomial:degree': 2, 'feature_preprocessor:polynomial:include_bias': 'False', 'feature_preprocessor:polynomial:interaction_only': 'False'},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.020000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'liblinear_svc', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'select_percentile_classification', 'classifier:liblinear_svc:C': 140.23434217313954, 'classifier:liblinear_svc:dual': 'False', 'classifier:liblinear_svc:fit_intercept': 'True', 'classifier:liblinear_svc:intercept_scaling': 1, 'classifier:liblinear_svc:loss': 'squared_hinge', 'classifier:liblinear_svc:multi_class': 'ovr', 'classifier:liblinear_svc:penalty': 'l2', 'classifier:liblinear_svc:tol': 0.0002944604673080368, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'quantile_transformer', 'feature_preprocessor:select_percentile_classification:percentile': 29.807816607300627, 'feature_preprocessor:select_percentile_classification:score_func': 'chi2', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.034330895903523305, 'data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles': 1450, 'data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution': 'uniform'},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.020000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'random_forest', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'select_rates_classification', 'classifier:random_forest:bootstrap': 'True', 'classifier:random_forest:criterion': 'gini', 'classifier:random_forest:max_depth': 'None', 'classifier:random_forest:max_features': 0.5561389303694175, 'classifier:random_forest:max_leaf_nodes': 'None', 'classifier:random_forest:min_impurity_decrease': 0.0, 'classifier:random_forest:min_samples_leaf': 16, 'classifier:random_forest:min_samples_split': 4, 'classifier:random_forest:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'quantile_transformer', 'feature_preprocessor:select_rates_classification:alpha': 0.48507733455436836, 'feature_preprocessor:select_rates_classification:score_func': 'f_classif', 'data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles': 1567, 'data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution': 'normal', 'feature_preprocessor:select_rates_classification:mode': 'fpr'},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.020000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'passive_aggressive', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:passive_aggressive:C': 0.14268277711454813, 'classifier:passive_aggressive:average': 'False', 'classifier:passive_aggressive:fit_intercept': 'True', 'classifier:passive_aggressive:loss': 'hinge', 'classifier:passive_aggressive:tol': 0.0002600768160857831, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'none', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.0070580904199417415},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.020000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'extra_trees', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'pca', 'classifier:extra_trees:bootstrap': 'False', 'classifier:extra_trees:criterion': 'gini', 'classifier:extra_trees:max_depth': 'None', 'classifier:extra_trees:max_features': 0.5137138356508442, 'classifier:extra_trees:max_leaf_nodes': 'None', 'classifier:extra_trees:min_impurity_decrease': 0.0, 'classifier:extra_trees:min_samples_leaf': 9, 'classifier:extra_trees:min_samples_split': 15, 'classifier:extra_trees:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'none', 'feature_preprocessor:pca:keep_variance': 0.9694955791292086, 'feature_preprocessor:pca:whiten': 'True', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.00014078218950356456},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    ]
-
-
-
-
-.. GENERATED FROM PYTHON SOURCE LINES 48-50
+    {   2: {   'balancing': Balancing(random_state=1),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc790129ac0>,
+               'cost': 0.028368794326241176,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7903bd040>,
+               'ensemble_weight': 0.04,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc79023cd90>,
+               'model_id': 2,
+               'rank': 4,
+               'sklearn_classifier': RandomForestClassifier(max_features=5, n_estimators=512, n_jobs=1,
+                           random_state=1, warm_start=True)},
+        4: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78b6bc040>,
+               'cost': 0.05673758865248224,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78b85f6d0>,
+               'ensemble_weight': 0.04,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78b714ca0>,
+               'model_id': 4,
+               'rank': 23,
+               'sklearn_classifier': MLPClassifier(activation='tanh', alpha=0.00021148999718383549, beta_1=0.999,
+                  beta_2=0.9, hidden_layer_sizes=(113, 113, 113),
+                  learning_rate_init=0.0007452270241186694, max_iter=64,
+                  n_iter_no_change=32, random_state=1, validation_fraction=0.0,
+                  verbose=0, warm_start=True)},
+        5: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78c414be0>,
+               'cost': 0.03546099290780147,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78c864820>,
+               'ensemble_weight': 0.04,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c4ed730>,
+               'model_id': 5,
+               'rank': 14,
+               'sklearn_classifier': RandomForestClassifier(criterion='entropy', max_features=3, min_samples_leaf=2,
+                           n_estimators=512, n_jobs=1, random_state=1,
+                           warm_start=True)},
+        6: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc790040550>,
+               'cost': 0.028368794326241176,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc79015a340>,
+               'ensemble_weight': 0.08,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc7900401f0>,
+               'model_id': 6,
+               'rank': 5,
+               'sklearn_classifier': MLPClassifier(alpha=0.0017940473175767063, beta_1=0.999, beta_2=0.9,
+                  early_stopping=True, hidden_layer_sizes=(101, 101),
+                  learning_rate_init=0.0004684917334431039, max_iter=32,
+                  n_iter_no_change=32, random_state=1, verbose=0, warm_start=True)},
+        7: {   'balancing': Balancing(random_state=1),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc79068b250>,
+               'cost': 0.014184397163120588,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc793f99a30>,
+               'ensemble_weight': 0.04,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc79068b370>,
+               'model_id': 7,
+               'rank': 1,
+               'sklearn_classifier': ExtraTreesClassifier(max_features=34, min_samples_leaf=3, min_samples_split=11,
+                         n_estimators=512, n_jobs=1, random_state=1,
+                         warm_start=True)},
+        8: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78c1f30a0>,
+               'cost': 0.03546099290780147,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78c70c0a0>,
+               'ensemble_weight': 0.02,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c3b75e0>,
+               'model_id': 8,
+               'rank': 15,
+               'sklearn_classifier': RandomForestClassifier(max_features=2, min_samples_leaf=2, n_estimators=512,
+                           n_jobs=1, random_state=1, warm_start=True)},
+        9: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78b820eb0>,
+               'cost': 0.04255319148936165,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78be25d90>,
+               'ensemble_weight': 0.02,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78b9c7250>,
+               'model_id': 9,
+               'rank': 20,
+               'sklearn_classifier': ExtraTreesClassifier(max_features=9, min_samples_split=10, n_estimators=512,
+                         n_jobs=1, random_state=1, warm_start=True)},
+        10: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78fe73d60>,
+                'cost': 0.028368794326241176,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc79016ebe0>,
+                'ensemble_weight': 0.08,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78fe73970>,
+                'model_id': 10,
+                'rank': 6,
+                'sklearn_classifier': RandomForestClassifier(criterion='entropy', max_features=4, min_samples_split=6,
+                           n_estimators=512, n_jobs=1, random_state=1,
+                           warm_start=True)},
+        11: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78fd0ae80>,
+                'cost': 0.028368794326241176,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7901eaf70>,
+                'ensemble_weight': 0.04,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78fd0ac70>,
+                'model_id': 11,
+                'rank': 7,
+                'sklearn_classifier': RandomForestClassifier(criterion='entropy', max_features=23, min_samples_leaf=7,
+                           n_estimators=512, n_jobs=1, random_state=1,
+                           warm_start=True)},
+        12: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78c1f3c70>,
+                'cost': 0.03546099290780147,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78c5fa5b0>,
+                'ensemble_weight': 0.04,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c1f37f0>,
+                'model_id': 12,
+                'rank': 16,
+                'sklearn_classifier': HistGradientBoostingClassifier(early_stopping=True,
+                                   l2_regularization=0.005326508887463406,
+                                   learning_rate=0.060800813211425456, max_iter=512,
+                                   max_leaf_nodes=6, min_samples_leaf=5,
+                                   n_iter_no_change=5, random_state=1,
+                                   validation_fraction=None, warm_start=True)},
+        13: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78fa69a30>,
+                'cost': 0.028368794326241176,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78feca6a0>,
+                'ensemble_weight': 0.02,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78fa69730>,
+                'model_id': 13,
+                'rank': 8,
+                'sklearn_classifier': HistGradientBoostingClassifier(early_stopping=False,
+                                   l2_regularization=1.0647401999412075e-10,
+                                   learning_rate=0.08291320147381159, max_iter=512,
+                                   max_leaf_nodes=39, n_iter_no_change=0,
+                                   random_state=1, validation_fraction=None,
+                                   warm_start=True)},
+        15: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78b836a90>,
+                'cost': 0.049645390070921946,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78ba74a90>,
+                'ensemble_weight': 0.02,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78b836370>,
+                'model_id': 15,
+                'rank': 21,
+                'sklearn_classifier': MLPClassifier(alpha=4.2841884333778574e-06, beta_1=0.999, beta_2=0.9,
+                  hidden_layer_sizes=(263, 263, 263),
+                  learning_rate_init=0.0011804284312897009, max_iter=128,
+                  n_iter_no_change=32, random_state=1, validation_fraction=0.0,
+                  verbose=0, warm_start=True)},
+        17: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78be46b20>,
+                'cost': 0.03546099290780147,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78c917c10>,
+                'ensemble_weight': 0.02,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78bc74550>,
+                'model_id': 17,
+                'rank': 17,
+                'sklearn_classifier': HistGradientBoostingClassifier(early_stopping=True,
+                                   l2_regularization=0.4635442279519353,
+                                   learning_rate=0.09809681787962342, max_iter=512,
+                                   max_leaf_nodes=328, min_samples_leaf=2,
+                                   n_iter_no_change=2, random_state=1,
+                                   validation_fraction=None, warm_start=True)},
+        18: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78ba1bee0>,
+                'cost': 0.03546099290780147,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78c345ee0>,
+                'ensemble_weight': 0.04,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78be46c10>,
+                'model_id': 18,
+                'rank': 18,
+                'sklearn_classifier': RandomForestClassifier(criterion='entropy', max_features=3, n_estimators=512,
+                           n_jobs=1, random_state=1, warm_start=True)},
+        19: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78cb3ca30>,
+                'cost': 0.028368794326241176,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78fe1ea00>,
+                'ensemble_weight': 0.02,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78cb3c880>,
+                'model_id': 19,
+                'rank': 9,
+                'sklearn_classifier': ExtraTreesClassifier(criterion='entropy', max_features=448, min_samples_leaf=2,
+                         min_samples_split=20, n_estimators=512, n_jobs=1,
+                         random_state=1, warm_start=True)},
+        21: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc790288640>,
+                'cost': 0.021276595744680882,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7902d5a30>,
+                'ensemble_weight': 0.04,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc790279550>,
+                'model_id': 21,
+                'rank': 3,
+                'sklearn_classifier': ExtraTreesClassifier(criterion='entropy', max_features=4, min_samples_leaf=2,
+                         min_samples_split=15, n_estimators=512, n_jobs=1,
+                         random_state=1, warm_start=True)},
+        23: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78b75a580>,
+                'cost': 0.049645390070921946,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78ba3ddf0>,
+                'ensemble_weight': 0.04,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78b75a130>,
+                'model_id': 23,
+                'rank': 22,
+                'sklearn_classifier': MLPClassifier(alpha=0.02847755502162456, beta_1=0.999, beta_2=0.9,
+                  hidden_layer_sizes=(123, 123),
+                  learning_rate_init=0.000421568792103947, max_iter=256,
+                  n_iter_no_change=32, random_state=1, validation_fraction=0.0,
+                  verbose=0, warm_start=True)},
+        24: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78b44ee50>,
+                'cost': 0.07092198581560283,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78b58aa90>,
+                'ensemble_weight': 0.02,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc790288bb0>,
+                'model_id': 24,
+                'rank': 26,
+                'sklearn_classifier': RandomForestClassifier(criterion='entropy', max_features=16, n_estimators=512,
+                           n_jobs=1, random_state=1, warm_start=True)},
+        26: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78ca98610>,
+                'cost': 0.028368794326241176,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78fbb9b50>,
+                'ensemble_weight': 0.04,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78ca98190>,
+                'model_id': 26,
+                'rank': 10,
+                'sklearn_classifier': ExtraTreesClassifier(criterion='entropy', max_features=414, min_samples_leaf=2,
+                         min_samples_split=19, n_estimators=512, n_jobs=1,
+                         random_state=1, warm_start=True)},
+        27: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78c944760>,
+                'cost': 0.028368794326241176,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78cb34910>,
+                'ensemble_weight': 0.04,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c93bdc0>,
+                'model_id': 27,
+                'rank': 11,
+                'sklearn_classifier': ExtraTreesClassifier(bootstrap=True, criterion='entropy', max_features=4126,
+                         min_samples_leaf=7, min_samples_split=13, n_estimators=512,
+                         n_jobs=1, random_state=1, warm_start=True)},
+        28: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78b545430>,
+                'cost': 0.06382978723404253,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78b79d940>,
+                'ensemble_weight': 0.02,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78b5451f0>,
+                'model_id': 28,
+                'rank': 24,
+                'sklearn_classifier': LinearDiscriminantAnalysis(tol=0.032840833359764676)},
+        29: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78c883580>,
+                'cost': 0.028368794326241176,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78cae7460>,
+                'ensemble_weight': 0.02,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c883070>,
+                'model_id': 29,
+                'rank': 12,
+                'sklearn_classifier': ExtraTreesClassifier(criterion='entropy', max_features=8, min_samples_leaf=11,
+                         min_samples_split=5, n_estimators=512, n_jobs=1,
+                         random_state=1, warm_start=True)},
+        31: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78ba251f0>,
+                'cost': 0.03546099290780147,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78bd04670>,
+                'ensemble_weight': 0.02,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78ba1bdc0>,
+                'model_id': 31,
+                'rank': 19,
+                'sklearn_classifier': LinearSVC(class_weight='balanced', dual=False, intercept_scaling=1.0,
+              random_state=1, tol=0.0001350687148423338)},
+        32: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78c667580>,
+                'cost': 0.028368794326241176,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78ca8e9a0>,
+                'ensemble_weight': 0.04,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c6cef10>,
+                'model_id': 32,
+                'rank': 13,
+                'sklearn_classifier': LinearSVC(C=832.5744883902069, dual=False, intercept_scaling=1.0,
+              random_state=1, tol=0.015340961719445064)},
+        34: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78f7b95e0>,
+                'cost': 0.06382978723404253,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78b753850>,
+                'ensemble_weight': 0.02,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c3e6490>,
+                'model_id': 34,
+                'rank': 25,
+                'sklearn_classifier': RandomForestClassifier(bootstrap=False, criterion='entropy', max_features=54,
+                           min_samples_leaf=14, min_samples_split=3,
+                           n_estimators=512, n_jobs=1, random_state=1,
+                           warm_start=True)},
+        37: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78acf7d00>,
+                'cost': 0.08510638297872342,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc790382190>,
+                'ensemble_weight': 0.02,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78ad54a00>,
+                'model_id': 37,
+                'rank': 28,
+                'sklearn_classifier': ExtraTreesClassifier(criterion='entropy', max_features=1, min_samples_leaf=10,
+                         min_samples_split=18, n_estimators=512, n_jobs=1,
+                         random_state=1, warm_start=True)},
+        39: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc790362c40>,
+                'cost': 0.014184397163120588,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7902d58b0>,
+                'ensemble_weight': 0.06,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc7901c2250>,
+                'model_id': 39,
+                'rank': 2,
+                'sklearn_classifier': ExtraTreesClassifier(criterion='entropy', max_features=32, min_samples_leaf=8,
+                         min_samples_split=8, n_estimators=512, n_jobs=1,
+                         random_state=1, warm_start=True)},
+        41: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78adce8b0>,
+                'cost': 0.07092198581560283,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78bff5f70>,
+                'ensemble_weight': 0.02,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78adc2d30>,
+                'model_id': 41,
+                'rank': 27,
+                'sklearn_classifier': DecisionTreeClassifier(max_depth=1, min_samples_leaf=15, min_samples_split=5,
+                           random_state=1)},
+        42: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78a8d7a90>,
+                'cost': 0.12056737588652477,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78ad91f70>,
+                'ensemble_weight': 0.02,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78a8d7790>,
+                'model_id': 42,
+                'rank': 30,
+                'sklearn_classifier': PassiveAggressiveClassifier(C=8.738643048253599e-05, average=True,
+                                loss='squared_hinge', max_iter=32, random_state=1,
+                                tol=0.0010303157708058924, warm_start=True)},
+        43: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78ab3f340>,
+                'cost': 0.09219858156028371,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78adfc730>,
+                'ensemble_weight': 0.02,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78aba4cd0>,
+                'model_id': 43,
+                'rank': 29,
+                'sklearn_classifier': ExtraTreesClassifier(bootstrap=True, criterion='entropy', max_features=12,
+                         min_samples_leaf=19, min_samples_split=17,
+                         n_estimators=512, n_jobs=1, random_state=1,
+                         warm_start=True)}}
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 50-52
 
 Get the Score of the final ensemble
 ===================================
 
-.. GENERATED FROM PYTHON SOURCE LINES 50-54
+.. GENERATED FROM PYTHON SOURCE LINES 52-56
 
 .. code-block:: default
 
@@ -345,7 +547,7 @@ Get the Score of the final ensemble
 
  .. code-block:: none
 
-    Accuracy score: 0.9440559440559441
+    Accuracy score: 0.958041958041958
 
 
 
@@ -353,7 +555,7 @@ Get the Score of the final ensemble
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 2 minutes  0.650 seconds)
+   **Total running time of the script:** ( 2 minutes  1.747 seconds)
 
 
 .. _sphx_glr_download_examples_20_basic_example_classification.py:
diff --git a/master/_sources/examples/20_basic/example_multilabel_classification.rst.txt b/master/_sources/examples/20_basic/example_multilabel_classification.rst.txt
index 5ee036eae5..206dc4f5f8 100644
--- a/master/_sources/examples/20_basic/example_multilabel_classification.rst.txt
+++ b/master/_sources/examples/20_basic/example_multilabel_classification.rst.txt
@@ -26,11 +26,12 @@ This examples shows how to format the targets for a multilabel classification
 problem. Details on multilabel classification can be found
 `here <https://scikit-learn.org/stable/modules/multiclass.html>`_.
 
-.. GENERATED FROM PYTHON SOURCE LINES 10-19
+.. GENERATED FROM PYTHON SOURCE LINES 10-20
 
 .. code-block:: default
 
     import numpy as np
+    from pprint import pprint
 
     import sklearn.datasets
     import sklearn.metrics
@@ -46,12 +47,12 @@ problem. Details on multilabel classification can be found
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 20-22
+.. GENERATED FROM PYTHON SOURCE LINES 21-23
 
 Data Loading
 ============
 
-.. GENERATED FROM PYTHON SOURCE LINES 22-43
+.. GENERATED FROM PYTHON SOURCE LINES 23-44
 
 .. code-block:: default
 
@@ -66,7 +67,7 @@ Data Loading
     # More information on: https://scikit-learn.org/stable/modules/multiclass.html
     y[y == 'TRUE'] = 1
     y[y == 'FALSE'] = 0
-    y = y.astype(np.int)
+    y = y.astype(int)
 
     # Using type of target is a good way to make sure your data
     # is properly formatted
@@ -86,20 +87,17 @@ Data Loading
 
  .. code-block:: none
 
-    /home/runner/work/auto-sklearn/auto-sklearn/examples/20_basic/example_multilabel_classification.py:33: DeprecationWarning: `np.int` is a deprecated alias for the builtin `int`. To silence this warning, use `int` by itself. Doing this will not modify any behavior and is safe. When replacing `np.int`, you may wish to use e.g. `np.int64` or `np.int32` to specify the precision. If you wish to review your current use, check the release note link for additional information.
-    Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations
-      y = y.astype(np.int)
     type_of_target=multilabel-indicator
 
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 44-46
+.. GENERATED FROM PYTHON SOURCE LINES 45-47
 
 Building the classifier
 =======================
 
-.. GENERATED FROM PYTHON SOURCE LINES 46-57
+.. GENERATED FROM PYTHON SOURCE LINES 47-58
 
 .. code-block:: default
 
@@ -132,12 +130,12 @@ Building the classifier
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 58-60
+.. GENERATED FROM PYTHON SOURCE LINES 59-61
 
 View the models found by auto-sklearn
 =====================================
 
-.. GENERATED FROM PYTHON SOURCE LINES 60-64
+.. GENERATED FROM PYTHON SOURCE LINES 61-65
 
 .. code-block:: default
 
@@ -157,22 +155,22 @@ View the models found by auto-sklearn
 
               rank  ensemble_weight           type      cost  duration
     model_id                                                          
-    2            1              1.0  random_forest  0.447294  3.636021
+    2            1              1.0  random_forest  0.447294  3.264884
 
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 65-67
+.. GENERATED FROM PYTHON SOURCE LINES 66-68
 
 Print the final ensemble constructed by auto-sklearn
 ====================================================
 
-.. GENERATED FROM PYTHON SOURCE LINES 67-70
+.. GENERATED FROM PYTHON SOURCE LINES 68-71
 
 .. code-block:: default
 
 
-    print(automl.show_models())
+    pprint(automl.show_models(), indent=4)
 
 
 
@@ -184,25 +182,26 @@ Print the final ensemble constructed by auto-sklearn
 
  .. code-block:: none
 
-    [(1.000000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'random_forest', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:random_forest:bootstrap': 'True', 'classifier:random_forest:criterion': 'gini', 'classifier:random_forest:max_depth': 'None', 'classifier:random_forest:max_features': 0.5, 'classifier:random_forest:max_leaf_nodes': 'None', 'classifier:random_forest:min_impurity_decrease': 0.0, 'classifier:random_forest:min_samples_leaf': 1, 'classifier:random_forest:min_samples_split': 2, 'classifier:random_forest:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.01},
-    dataset_properties={
-      'task': 3,
-      'sparse': False,
-      'multilabel': True,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    ]
+    {   2: {   'balancing': Balancing(random_state=1),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78ca21f10>,
+               'cost': 0.4472941828699525,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78a8d7610>,
+               'ensemble_weight': 1.0,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78ca21670>,
+               'model_id': 2,
+               'rank': 1,
+               'sklearn_classifier': RandomForestClassifier(max_features=15, n_estimators=512, n_jobs=1,
+                           random_state=1, warm_start=True)}}
 
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 71-73
+.. GENERATED FROM PYTHON SOURCE LINES 72-74
 
 Print statistics about the auto-sklearn run
 ===========================================
 
-.. GENERATED FROM PYTHON SOURCE LINES 73-78
+.. GENERATED FROM PYTHON SOURCE LINES 74-79
 
 .. code-block:: default
 
@@ -235,12 +234,12 @@ Print statistics about the auto-sklearn run
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 79-81
+.. GENERATED FROM PYTHON SOURCE LINES 80-82
 
 Get the Score of the final ensemble
 ===================================
 
-.. GENERATED FROM PYTHON SOURCE LINES 81-84
+.. GENERATED FROM PYTHON SOURCE LINES 82-85
 
 .. code-block:: default
 
@@ -265,7 +264,7 @@ Get the Score of the final ensemble
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 0 minutes  15.973 seconds)
+   **Total running time of the script:** ( 0 minutes  12.175 seconds)
 
 
 .. _sphx_glr_download_examples_20_basic_example_multilabel_classification.py:
diff --git a/master/_sources/examples/20_basic/example_multioutput_regression.rst.txt b/master/_sources/examples/20_basic/example_multioutput_regression.rst.txt
index 96d89e5a7e..0ec41f9fc1 100644
--- a/master/_sources/examples/20_basic/example_multioutput_regression.rst.txt
+++ b/master/_sources/examples/20_basic/example_multioutput_regression.rst.txt
@@ -25,11 +25,12 @@ Multi-output Regression
 The following example shows how to fit a multioutput regression model with
 *auto-sklearn*.
 
-.. GENERATED FROM PYTHON SOURCE LINES 10-19
+.. GENERATED FROM PYTHON SOURCE LINES 10-20
 
 .. code-block:: default
 
     import numpy as numpy
+    from pprint import pprint
 
     from sklearn.datasets import make_regression
     from sklearn.metrics import r2_score
@@ -45,12 +46,12 @@ The following example shows how to fit a multioutput regression model with
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 20-22
+.. GENERATED FROM PYTHON SOURCE LINES 21-23
 
 Data Loading
 ============
 
-.. GENERATED FROM PYTHON SOURCE LINES 22-27
+.. GENERATED FROM PYTHON SOURCE LINES 23-28
 
 .. code-block:: default
 
@@ -66,12 +67,12 @@ Data Loading
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 28-30
+.. GENERATED FROM PYTHON SOURCE LINES 29-31
 
 Build and fit a regressor
 =========================
 
-.. GENERATED FROM PYTHON SOURCE LINES 30-38
+.. GENERATED FROM PYTHON SOURCE LINES 31-39
 
 .. code-block:: default
 
@@ -99,12 +100,12 @@ Build and fit a regressor
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 39-41
+.. GENERATED FROM PYTHON SOURCE LINES 40-42
 
 View the models found by auto-sklearn
 =====================================
 
-.. GENERATED FROM PYTHON SOURCE LINES 41-45
+.. GENERATED FROM PYTHON SOURCE LINES 42-46
 
 .. code-block:: default
 
@@ -122,26 +123,26 @@ View the models found by auto-sklearn
 
  .. code-block:: none
 
-              rank  ensemble_weight                 type      cost  duration
-    model_id                                                                
-    2            1             0.62        random_forest  0.157218  2.282688
-    5            2             0.18  k_nearest_neighbors  0.207332  0.459119
-    25           3             0.20  k_nearest_neighbors  0.318861  0.454935
+              rank  ensemble_weight           type      cost  duration
+    model_id                                                          
+    22           1             0.32    extra_trees  0.136510  5.840576
+    2            2             0.42  random_forest  0.142814  2.232426
+    21           3             0.26  random_forest  0.224804  1.676522
 
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 46-48
+.. GENERATED FROM PYTHON SOURCE LINES 47-49
 
 Print the final ensemble constructed by auto-sklearn
 ====================================================
 
-.. GENERATED FROM PYTHON SOURCE LINES 48-51
+.. GENERATED FROM PYTHON SOURCE LINES 49-52
 
 .. code-block:: default
 
 
-    print(automl.show_models())
+    pprint(automl.show_models(), indent=4)
 
 
 
@@ -153,38 +154,46 @@ Print the final ensemble constructed by auto-sklearn
 
  .. code-block:: none
 
-    [(0.620000, SimpleRegressionPipeline({'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'regressor:__choice__': 'random_forest', 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize', 'regressor:random_forest:bootstrap': 'True', 'regressor:random_forest:criterion': 'mse', 'regressor:random_forest:max_depth': 'None', 'regressor:random_forest:max_features': 1.0, 'regressor:random_forest:max_leaf_nodes': 'None', 'regressor:random_forest:min_impurity_decrease': 0.0, 'regressor:random_forest:min_samples_leaf': 1, 'regressor:random_forest:min_samples_split': 2, 'regressor:random_forest:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.01},
-    dataset_properties={
-      'task': 5,
-      'sparse': False,
-      'multioutput': True,
-      'target_type': 'regression',
-      'signed': False})),
-    (0.200000, SimpleRegressionPipeline({'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'polynomial', 'regressor:__choice__': 'k_nearest_neighbors', 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'minmax', 'feature_preprocessor:polynomial:degree': 3, 'feature_preprocessor:polynomial:include_bias': 'False', 'feature_preprocessor:polynomial:interaction_only': 'True', 'regressor:k_nearest_neighbors:n_neighbors': 1, 'regressor:k_nearest_neighbors:p': 2, 'regressor:k_nearest_neighbors:weights': 'distance'},
-    dataset_properties={
-      'task': 5,
-      'sparse': False,
-      'multioutput': True,
-      'target_type': 'regression',
-      'signed': False})),
-    (0.180000, SimpleRegressionPipeline({'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'pca', 'regressor:__choice__': 'k_nearest_neighbors', 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'normalize', 'feature_preprocessor:pca:keep_variance': 0.8772654126094648, 'feature_preprocessor:pca:whiten': 'True', 'regressor:k_nearest_neighbors:n_neighbors': 14, 'regressor:k_nearest_neighbors:p': 1, 'regressor:k_nearest_neighbors:weights': 'uniform'},
-    dataset_properties={
-      'task': 5,
-      'sparse': False,
-      'multioutput': True,
-      'target_type': 'regression',
-      'signed': False})),
-    ]
-
-
-
-
-.. GENERATED FROM PYTHON SOURCE LINES 52-54
+    {   2: {   'cost': 0.14281372797350056,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78f7b94c0>,
+               'ensemble_weight': 0.42,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c920640>,
+               'model_id': 2,
+               'rank': 2,
+               'regressor': <autosklearn.pipeline.components.regression.RegressorChoice object at 0x7fc78c920430>,
+               'sklearn_regressor': RandomForestRegressor(max_features=1.0, n_estimators=512, n_jobs=1,
+                          random_state=1, warm_start=True)},
+        21: {   'cost': 0.22480405782181234,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78f8d10a0>,
+                'ensemble_weight': 0.26,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78ca212e0>,
+                'model_id': 21,
+                'rank': 3,
+                'regressor': <autosklearn.pipeline.components.regression.RegressorChoice object at 0x7fc78ca210d0>,
+                'sklearn_regressor': RandomForestRegressor(criterion='friedman_mse', max_features=0.9689529284463153,
+                          min_samples_leaf=5, min_samples_split=4, n_estimators=512,
+                          n_jobs=1, random_state=1, warm_start=True)},
+        22: {   'cost': 0.13651009800609504,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78f7b9640>,
+                'ensemble_weight': 0.32,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78f7b94f0>,
+                'model_id': 22,
+                'rank': 1,
+                'regressor': <autosklearn.pipeline.components.regression.RegressorChoice object at 0x7fc78f7b9f10>,
+                'sklearn_regressor': ExtraTreesRegressor(bootstrap=True, criterion='mae',
+                        max_features=0.9195228400065838, min_samples_leaf=2,
+                        min_samples_split=7, n_estimators=512, n_jobs=1,
+                        random_state=1, warm_start=True)}}
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 53-55
 
 Get the Score of the final ensemble
 ===================================
 
-.. GENERATED FROM PYTHON SOURCE LINES 54-58
+.. GENERATED FROM PYTHON SOURCE LINES 55-59
 
 .. code-block:: default
 
@@ -202,17 +211,17 @@ Get the Score of the final ensemble
 
  .. code-block:: none
 
-    R2 score: 0.8683277328634489
+    R2 score: 0.8858495026618883
 
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 59-61
+.. GENERATED FROM PYTHON SOURCE LINES 60-62
 
 Get the configuration space
 ===========================
 
-.. GENERATED FROM PYTHON SOURCE LINES 61-64
+.. GENERATED FROM PYTHON SOURCE LINES 62-65
 
 .. code-block:: default
 
@@ -425,7 +434,7 @@ Get the configuration space
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 1 minutes  55.310 seconds)
+   **Total running time of the script:** ( 1 minutes  56.218 seconds)
 
 
 .. _sphx_glr_download_examples_20_basic_example_multioutput_regression.py:
diff --git a/master/_sources/examples/20_basic/example_regression.rst.txt b/master/_sources/examples/20_basic/example_regression.rst.txt
index 506b24b321..9ee56987e5 100644
--- a/master/_sources/examples/20_basic/example_regression.rst.txt
+++ b/master/_sources/examples/20_basic/example_regression.rst.txt
@@ -25,10 +25,12 @@ Regression
 The following example shows how to fit a simple regression model with
 *auto-sklearn*.
 
-.. GENERATED FROM PYTHON SOURCE LINES 10-16
+.. GENERATED FROM PYTHON SOURCE LINES 10-18
 
 .. code-block:: default
 
+    from pprint import pprint
+
     import sklearn.datasets
     import sklearn.metrics
 
@@ -42,12 +44,12 @@ The following example shows how to fit a simple regression model with
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 17-19
+.. GENERATED FROM PYTHON SOURCE LINES 19-21
 
 Data Loading
 ============
 
-.. GENERATED FROM PYTHON SOURCE LINES 19-25
+.. GENERATED FROM PYTHON SOURCE LINES 21-27
 
 .. code-block:: default
 
@@ -64,12 +66,12 @@ Data Loading
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 26-28
+.. GENERATED FROM PYTHON SOURCE LINES 28-30
 
 Build and fit a regressor
 =========================
 
-.. GENERATED FROM PYTHON SOURCE LINES 28-36
+.. GENERATED FROM PYTHON SOURCE LINES 30-38
 
 .. code-block:: default
 
@@ -91,18 +93,22 @@ Build and fit a regressor
 
  .. code-block:: none
 
+    /home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:68: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+      self.metafeatures = self.metafeatures.append(metafeatures)
+    /home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:72: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+      self.algorithm_runs[metric].append(runs)
 
     AutoSklearnRegressor(per_run_time_limit=30, time_left_for_this_task=120,
                          tmp_folder='/tmp/autosklearn_regression_example_tmp')
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 37-39
+.. GENERATED FROM PYTHON SOURCE LINES 39-41
 
 View the models found by auto-sklearn
 =====================================
 
-.. GENERATED FROM PYTHON SOURCE LINES 39-42
+.. GENERATED FROM PYTHON SOURCE LINES 41-44
 
 .. code-block:: default
 
@@ -121,26 +127,25 @@ View the models found by auto-sklearn
 
               rank  ensemble_weight               type      cost  duration
     model_id                                                              
-    25           1             0.46                sgd  0.436679  0.628683
-    6            2             0.32     ard_regression  0.455042  0.645808
-    27           3             0.14     ard_regression  0.462249  0.639516
-    11           4             0.02      random_forest  0.507400  8.241306
-    7            5             0.06  gradient_boosting  0.518673  1.216493
+    25           1             0.44                sgd  0.436679  0.608678
+    6            2             0.34     ard_regression  0.455042  0.619489
+    39           3             0.18     ard_regression  0.474807  0.617622
+    7            4             0.04  gradient_boosting  0.518673  1.104580
 
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 43-45
+.. GENERATED FROM PYTHON SOURCE LINES 45-47
 
 Print the final ensemble constructed by auto-sklearn
 ====================================================
 
-.. GENERATED FROM PYTHON SOURCE LINES 45-48
+.. GENERATED FROM PYTHON SOURCE LINES 47-50
 
 .. code-block:: default
 
 
-    print(automl.show_models())
+    pprint(automl.show_models(), indent=4)
 
 
 
@@ -152,47 +157,57 @@ Print the final ensemble constructed by auto-sklearn
 
  .. code-block:: none
 
-    [(0.460000, SimpleRegressionPipeline({'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'polynomial', 'regressor:__choice__': 'sgd', 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'minmax', 'feature_preprocessor:polynomial:degree': 3, 'feature_preprocessor:polynomial:include_bias': 'False', 'feature_preprocessor:polynomial:interaction_only': 'False', 'regressor:sgd:alpha': 0.0006517033225329654, 'regressor:sgd:average': 'False', 'regressor:sgd:fit_intercept': 'True', 'regressor:sgd:learning_rate': 'invscaling', 'regressor:sgd:loss': 'epsilon_insensitive', 'regressor:sgd:penalty': 'elasticnet', 'regressor:sgd:tol': 0.002431731981071206, 'regressor:sgd:epsilon': 0.012150149892783745, 'regressor:sgd:eta0': 0.016444224834275295, 'regressor:sgd:l1_ratio': 1.7462342366289323e-09, 'regressor:sgd:power_t': 0.21521743568582094},
-    dataset_properties={
-      'task': 4,
-      'sparse': False,
-      'multioutput': False,
-      'target_type': 'regression',
-      'signed': False})),
-    (0.320000, SimpleRegressionPipeline({'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'polynomial', 'regressor:__choice__': 'ard_regression', 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize', 'feature_preprocessor:polynomial:degree': 2, 'feature_preprocessor:polynomial:include_bias': 'True', 'feature_preprocessor:polynomial:interaction_only': 'False', 'regressor:ard_regression:alpha_1': 0.0003701926442639788, 'regressor:ard_regression:alpha_2': 2.2118001735899097e-07, 'regressor:ard_regression:fit_intercept': 'True', 'regressor:ard_regression:lambda_1': 1.2037591637980971e-06, 'regressor:ard_regression:lambda_2': 4.358378124977852e-09, 'regressor:ard_regression:n_iter': 300, 'regressor:ard_regression:threshold_lambda': 1136.5286041327277, 'regressor:ard_regression:tol': 0.021944240404849075},
-    dataset_properties={
-      'task': 4,
-      'sparse': False,
-      'multioutput': False,
-      'target_type': 'regression',
-      'signed': False})),
-    (0.140000, SimpleRegressionPipeline({'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'select_rates_regression', 'regressor:__choice__': 'ard_regression', 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'robust_scaler', 'feature_preprocessor:select_rates_regression:alpha': 0.3621762718897781, 'feature_preprocessor:select_rates_regression:mode': 'fwe', 'feature_preprocessor:select_rates_regression:score_func': 'f_regression', 'regressor:ard_regression:alpha_1': 2.7664515192592053e-05, 'regressor:ard_regression:alpha_2': 9.504988116581138e-07, 'regressor:ard_regression:fit_intercept': 'True', 'regressor:ard_regression:lambda_1': 6.50650698230178e-09, 'regressor:ard_regression:lambda_2': 4.238533890074848e-07, 'regressor:ard_regression:n_iter': 300, 'regressor:ard_regression:threshold_lambda': 78251.58542976103, 'regressor:ard_regression:tol': 0.0007301343236220855, 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.0001745391328519669, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max': 0.8057830372269097, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min': 0.24982831110057324},
-    dataset_properties={
-      'task': 4,
-      'sparse': False,
-      'multioutput': False,
-      'target_type': 'regression',
-      'signed': False})),
-    (0.060000, SimpleRegressionPipeline({'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'regressor:__choice__': 'gradient_boosting', 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'quantile_transformer', 'regressor:gradient_boosting:early_stop': 'off', 'regressor:gradient_boosting:l2_regularization': 1.8428972335335263e-10, 'regressor:gradient_boosting:learning_rate': 0.012607824914758717, 'regressor:gradient_boosting:loss': 'least_squares', 'regressor:gradient_boosting:max_bins': 255, 'regressor:gradient_boosting:max_depth': 'None', 'regressor:gradient_boosting:max_leaf_nodes': 10, 'regressor:gradient_boosting:min_samples_leaf': 8, 'regressor:gradient_boosting:scoring': 'loss', 'regressor:gradient_boosting:tol': 1e-07, 'data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles': 952, 'data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution': 'uniform'},
-    dataset_properties={
-      'task': 4,
-      'sparse': False,
-      'multioutput': False,
-      'target_type': 'regression',
-      'signed': False})),
-    (0.020000, SimpleRegressionPipeline({'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'polynomial', 'regressor:__choice__': 'random_forest', 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'minmax', 'feature_preprocessor:polynomial:degree': 2, 'feature_preprocessor:polynomial:include_bias': 'True', 'feature_preprocessor:polynomial:interaction_only': 'True', 'regressor:random_forest:bootstrap': 'False', 'regressor:random_forest:criterion': 'mae', 'regressor:random_forest:max_depth': 'None', 'regressor:random_forest:max_features': 0.6277363920171745, 'regressor:random_forest:max_leaf_nodes': 'None', 'regressor:random_forest:min_impurity_decrease': 0.0, 'regressor:random_forest:min_samples_leaf': 6, 'regressor:random_forest:min_samples_split': 15, 'regressor:random_forest:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.0010413452644415357},
-    dataset_properties={
-      'task': 4,
-      'sparse': False,
-      'multioutput': False,
-      'target_type': 'regression',
-      'signed': False})),
-    ]
-
-
-
-
-.. GENERATED FROM PYTHON SOURCE LINES 49-55
+    {   6: {   'cost': 0.4550418898836528,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78a826a00>,
+               'ensemble_weight': 0.34,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78b503820>,
+               'model_id': 6,
+               'rank': 2,
+               'regressor': <autosklearn.pipeline.components.regression.RegressorChoice object at 0x7fc78b503880>,
+               'sklearn_regressor': ARDRegression(alpha_1=0.0003701926442639788, alpha_2=2.2118001735899097e-07,
+                  copy_X=False, lambda_1=1.2037591637980971e-06,
+                  lambda_2=4.358378124977852e-09,
+                  threshold_lambda=1136.5286041327277, tol=0.021944240404849075)},
+        7: {   'cost': 0.5186726734789994,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78f913a00>,
+               'ensemble_weight': 0.04,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78b48c700>,
+               'model_id': 7,
+               'rank': 4,
+               'regressor': <autosklearn.pipeline.components.regression.RegressorChoice object at 0x7fc78a5d3580>,
+               'sklearn_regressor': HistGradientBoostingRegressor(l2_regularization=1.8428972335335263e-10,
+                                  learning_rate=0.012607824914758717, max_iter=512,
+                                  max_leaf_nodes=10, min_samples_leaf=8,
+                                  n_iter_no_change=0, random_state=1,
+                                  validation_fraction=None, warm_start=True)},
+        25: {   'cost': 0.43667876507897496,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78c94d400>,
+                'ensemble_weight': 0.44,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78f913df0>,
+                'model_id': 25,
+                'rank': 1,
+                'regressor': <autosklearn.pipeline.components.regression.RegressorChoice object at 0x7fc78f80b160>,
+                'sklearn_regressor': SGDRegressor(alpha=0.0006517033225329654, epsilon=0.012150149892783745,
+                 eta0=0.016444224834275295, l1_ratio=1.7462342366289323e-09,
+                 loss='epsilon_insensitive', max_iter=16, penalty='elasticnet',
+                 power_t=0.21521743568582094, random_state=1,
+                 tol=0.002431731981071206, warm_start=True)},
+        39: {   'cost': 0.4748068089650166,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78f80b610>,
+                'ensemble_weight': 0.18,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c482280>,
+                'model_id': 39,
+                'rank': 3,
+                'regressor': <autosklearn.pipeline.components.regression.RegressorChoice object at 0x7fc78c4826a0>,
+                'sklearn_regressor': ARDRegression(alpha_1=0.0005012365297609799, alpha_2=3.025360750168211e-08,
+                  copy_X=False, lambda_1=4.9749646614525684e-05,
+                  lambda_2=3.2368037115065363e-10,
+                  threshold_lambda=18669.665899307194, tol=0.0012624032013298571)}}
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 51-57
 
 Get the Score of the final ensemble
 ===================================
@@ -201,7 +216,7 @@ is the `R2 score <https://scikit-learn.org/stable/modules/model_evaluation.html#
 The values range between -inf and 1 with 1 being the best possible value. A dummy estimator
 predicting the data mean has an R2 score of 0.
 
-.. GENERATED FROM PYTHON SOURCE LINES 55-61
+.. GENERATED FROM PYTHON SOURCE LINES 57-63
 
 .. code-block:: default
 
@@ -221,13 +236,13 @@ predicting the data mean has an R2 score of 0.
 
  .. code-block:: none
 
-    Train R2 score: 0.5944780427522034
-    Test R2 score: 0.3959585042866587
+    Train R2 score: 0.5855373845454157
+    Test R2 score: 0.39879073225079487
 
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 62-69
+.. GENERATED FROM PYTHON SOURCE LINES 64-71
 
 Plot the predictions
 ====================
@@ -237,7 +252,7 @@ predictions. Points below the diagonal were overestimated by the model (predicte
 than the true value), points above the diagonal were underestimated (predicted value is lower than
 the true value).
 
-.. GENERATED FROM PYTHON SOURCE LINES 69-80
+.. GENERATED FROM PYTHON SOURCE LINES 71-82
 
 .. code-block:: default
 
@@ -267,7 +282,7 @@ the true value).
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 1 minutes  58.195 seconds)
+   **Total running time of the script:** ( 1 minutes  59.382 seconds)
 
 
 .. _sphx_glr_download_examples_20_basic_example_regression.py:
diff --git a/master/_sources/examples/20_basic/sg_execution_times.rst.txt b/master/_sources/examples/20_basic/sg_execution_times.rst.txt
index 7f4de685e2..bef5aeab93 100644
--- a/master/_sources/examples/20_basic/sg_execution_times.rst.txt
+++ b/master/_sources/examples/20_basic/sg_execution_times.rst.txt
@@ -5,14 +5,14 @@
 
 Computation times
 =================
-**06:10.128** total execution time for **examples_20_basic** files:
+**06:09.522** total execution time for **examples_20_basic** files:
 
 +-------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_examples_20_basic_example_classification.py` (``example_classification.py``)                       | 02:00.650 | 0.0 MB |
+| :ref:`sphx_glr_examples_20_basic_example_classification.py` (``example_classification.py``)                       | 02:01.747 | 0.0 MB |
 +-------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_examples_20_basic_example_regression.py` (``example_regression.py``)                               | 01:58.195 | 0.0 MB |
+| :ref:`sphx_glr_examples_20_basic_example_regression.py` (``example_regression.py``)                               | 01:59.382 | 0.0 MB |
 +-------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_examples_20_basic_example_multioutput_regression.py` (``example_multioutput_regression.py``)       | 01:55.310 | 0.0 MB |
+| :ref:`sphx_glr_examples_20_basic_example_multioutput_regression.py` (``example_multioutput_regression.py``)       | 01:56.218 | 0.0 MB |
 +-------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_examples_20_basic_example_multilabel_classification.py` (``example_multilabel_classification.py``) | 00:15.973 | 0.0 MB |
+| :ref:`sphx_glr_examples_20_basic_example_multilabel_classification.py` (``example_multilabel_classification.py``) | 00:12.175 | 0.0 MB |
 +-------------------------------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/master/_sources/examples/40_advanced/example_calc_multiple_metrics.rst.txt b/master/_sources/examples/40_advanced/example_calc_multiple_metrics.rst.txt
index 053a89690a..56c45162c9 100644
--- a/master/_sources/examples/40_advanced/example_calc_multiple_metrics.rst.txt
+++ b/master/_sources/examples/40_advanced/example_calc_multiple_metrics.rst.txt
@@ -116,6 +116,10 @@ Build and fit a classifier
 
  .. code-block:: none
 
+    /home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:68: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+      self.metafeatures = self.metafeatures.append(metafeatures)
+    /home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:72: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+      self.algorithm_runs[metric].append(runs)
 
     AutoSklearnClassifier(per_run_time_limit=30,
                           scoring_functions=[balanced_accuracy, precision, recall,
@@ -150,47 +154,51 @@ Get the Score of the final ensemble
 
  .. code-block:: none
 
-    Accuracy score 0.9440559440559441
+    Accuracy score 0.958041958041958
     ################################################################################
     Metric results
      rank_test_scores param_classifier:__choice__  mean_test_score  metric_balanced_accuracy  metric_precision  metric_recall  metric_f1  metric_custom_error
-                    7               random_forest         0.971631                  0.969533          0.977528       0.977528   0.977528             0.028369
-                    7                         mlp         0.971631                  0.961538          0.956989       1.000000   0.978022             0.028369
-                   29                         mlp         0.943262                  0.935069          0.945055       0.966292   0.955556             0.056738
-                   17               random_forest         0.964539                  0.959918          0.966667       0.977528   0.972067             0.035461
-                    7                         mlp         0.971631                  0.961538          0.956989       1.000000   0.978022             0.028369
+                    5               random_forest         0.971631                  0.969533          0.977528       0.977528   0.977528             0.028369
+                    5                         mlp         0.971631                  0.961538          0.956989       1.000000   0.978022             0.028369
+                   31                         mlp         0.943262                  0.935069          0.945055       0.966292   0.955556             0.056738
+                   19               random_forest         0.964539                  0.959918          0.966667       0.977528   0.972067             0.035461
+                    5                         mlp         0.971631                  0.961538          0.956989       1.000000   0.978022             0.028369
                     1                 extra_trees         0.985816                  0.984767          0.988764       0.988764   0.988764             0.014184
-                   17               random_forest         0.964539                  0.963915          0.977273       0.966292   0.971751             0.035461
-                   22                 extra_trees         0.957447                  0.954300          0.966292       0.966292   0.966292             0.042553
-                    7               random_forest         0.971631                  0.969533          0.977528       0.977528   0.977528             0.028369
-                    7               random_forest         0.971631                  0.969533          0.977528       0.977528   0.977528             0.028369
-                   17           gradient_boosting         0.964539                  0.963915          0.977273       0.966292   0.971751             0.035461
-                    7           gradient_boosting         0.971631                  0.965536          0.967033       0.988764   0.977778             0.028369
-                    7                         mlp         0.971631                  0.965536          0.967033       0.988764   0.977778             0.028369
-                   26                         mlp         0.950355                  0.948682          0.965909       0.955056   0.960452             0.049645
+                   19               random_forest         0.964539                  0.963915          0.977273       0.966292   0.971751             0.035461
+                   25                 extra_trees         0.957447                  0.954300          0.966292       0.966292   0.966292             0.042553
+                    5               random_forest         0.971631                  0.969533          0.977528       0.977528   0.977528             0.028369
+                    5               random_forest         0.971631                  0.969533          0.977528       0.977528   0.977528             0.028369
+                   19           gradient_boosting         0.964539                  0.963915          0.977273       0.966292   0.971751             0.035461
+                    5           gradient_boosting         0.971631                  0.965536          0.967033       0.988764   0.977778             0.028369
+                    5                         mlp         0.971631                  0.965536          0.967033       0.988764   0.977778             0.028369
+                   28                         mlp         0.950355                  0.948682          0.965909       0.955056   0.960452             0.049645
                     3           gradient_boosting         0.978723                  0.975151          0.977778       0.988764   0.983240             0.021277
-                   17           gradient_boosting         0.964539                  0.959918          0.966667       0.977528   0.972067             0.035461
-                   17               random_forest         0.964539                  0.959918          0.966667       0.977528   0.972067             0.035461
-                    7                 extra_trees         0.971631                  0.969533          0.977528       0.977528   0.977528             0.028369
-                   34          passive_aggressive         0.921986                  0.894231          0.890000       1.000000   0.941799             0.078014
+                   19           gradient_boosting         0.964539                  0.959918          0.966667       0.977528   0.972067             0.035461
+                   19               random_forest         0.964539                  0.959918          0.966667       0.977528   0.972067             0.035461
+                    5                 extra_trees         0.971631                  0.969533          0.977528       0.977528   0.977528             0.028369
+                   37          passive_aggressive         0.921986                  0.894231          0.890000       1.000000   0.941799             0.078014
                     3                 extra_trees         0.978723                  0.975151          0.977778       0.988764   0.983240             0.021277
-                    7           gradient_boosting         0.971631                  0.965536          0.967033       0.988764   0.977778             0.028369
-                   26                         mlp         0.950355                  0.940687          0.945652       0.977528   0.961326             0.049645
-                   32               random_forest         0.929078                  0.923833          0.943820       0.943820   0.943820             0.070922
-                   22                    adaboost         0.957447                  0.950303          0.956044       0.977528   0.966667             0.042553
-                    7                 extra_trees         0.971631                  0.965536          0.967033       0.988764   0.977778             0.028369
-                   22                 extra_trees         0.957447                  0.954300          0.966292       0.966292   0.966292             0.042553
-                   32                bernoulli_nb         0.929078                  0.923833          0.943820       0.943820   0.943820             0.070922
-                    3                 extra_trees         0.978723                  0.979149          0.988636       0.977528   0.983051             0.021277
-                   22               liblinear_svc         0.957447                  0.954300          0.966292       0.966292   0.966292             0.042553
-                   29                 gaussian_nb         0.943262                  0.935069          0.945055       0.966292   0.955556             0.056738
-                   26                 extra_trees         0.950355                  0.936690          0.936170       0.988764   0.961749             0.049645
-                   29               random_forest         0.943262                  0.943064          0.965517       0.943820   0.954545             0.056738
+                    5           gradient_boosting         0.971631                  0.965536          0.967033       0.988764   0.977778             0.028369
+                   28                         mlp         0.950355                  0.940687          0.945652       0.977528   0.961326             0.049645
+                   35               random_forest         0.929078                  0.923833          0.943820       0.943820   0.943820             0.070922
+                   25                    adaboost         0.957447                  0.950303          0.956044       0.977528   0.966667             0.042553
+                    5                 extra_trees         0.971631                  0.965536          0.967033       0.988764   0.977778             0.028369
+                    5                 extra_trees         0.971631                  0.969533          0.977528       0.977528   0.977528             0.028369
+                   33                         lda         0.936170                  0.925454          0.934783       0.966292   0.950276             0.063830
+                    5                 extra_trees         0.971631                  0.965536          0.967033       0.988764   0.977778             0.028369
+                    5                         lda         0.971631                  0.961538          0.956989       1.000000   0.978022             0.028369
+                   19               liblinear_svc         0.964539                  0.959918          0.966667       0.977528   0.972067             0.035461
+                    5               liblinear_svc         0.971631                  0.965536          0.967033       0.988764   0.977778             0.028369
+                   31                    adaboost         0.943262                  0.943064          0.965517       0.943820   0.954545             0.056738
+                   33               random_forest         0.936170                  0.937446          0.965116       0.932584   0.948571             0.063830
+                   28                    adaboost         0.950355                  0.952679          0.976744       0.943820   0.960000             0.049645
+                   40                 extra_trees         0.631206                  0.500000          0.631206       1.000000   0.773913             0.368794
+                   38                 extra_trees         0.914894                  0.884615          0.881188       1.000000   0.936842             0.085106
+                   41                         sgd         0.368794                  0.500000          0.000000       0.000000   0.000000             0.631206
                     1                 extra_trees         0.985816                  0.984767          0.988764       0.988764   0.988764             0.014184
-                   35                 extra_trees         0.914894                  0.884615          0.881188       1.000000   0.936842             0.085106
-                    3                 extra_trees         0.978723                  0.975151          0.977778       0.988764   0.983240             0.021277
-                   35                  libsvm_svc         0.914894                  0.924589          0.975309       0.887640   0.929412             0.085106
-                   37                         sgd         0.829787                  0.845182          0.933333       0.786517   0.853659             0.170213
+                   25                         mlp         0.957447                  0.950303          0.956044       0.977528   0.966667             0.042553
+                   35               decision_tree         0.929078                  0.919836          0.934066       0.955056   0.944444             0.070922
+                   39          passive_aggressive         0.879433                  0.868518          0.900000       0.910112   0.905028             0.120567
 
 
 
@@ -198,7 +206,7 @@ Get the Score of the final ensemble
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 2 minutes  8.166 seconds)
+   **Total running time of the script:** ( 1 minutes  59.942 seconds)
 
 
 .. _sphx_glr_download_examples_40_advanced_example_calc_multiple_metrics.py:
diff --git a/master/_sources/examples/40_advanced/example_debug_logging.rst.txt b/master/_sources/examples/40_advanced/example_debug_logging.rst.txt
index ba5d36873c..60da9daf6c 100644
--- a/master/_sources/examples/40_advanced/example_debug_logging.rst.txt
+++ b/master/_sources/examples/40_advanced/example_debug_logging.rst.txt
@@ -174,13 +174,13 @@ Build and fit a classifier
 
  .. code-block:: none
 
-    2021-11-09 19:52:24,412 - Client-EnsembleBuilder - INFO - DummyFuture: ([{'Timestamp': Timestamp('2021-11-09 19:52:24.397684'), 'ensemble_optimization_score': 0.9886363636363636, 'ensemble_test_score': 0.9899874843554443}], 50, None, None, None)/SingleThreadedClient() Started Ensemble builder job at 2021.11.09-19.52.24 for iteration 0.
-    2021-11-09 19:52:28,964 - Client-EnsembleBuilder - INFO - DummyFuture: ([{'Timestamp': Timestamp('2021-11-09 19:52:28.945787'), 'ensemble_optimization_score': 0.9911616161616161, 'ensemble_test_score': 0.9899874843554443}], 50, None, None, None)/SingleThreadedClient() Started Ensemble builder job at 2021.11.09-19.52.28 for iteration 1.
+    2022-01-25 16:22:18,915 - Client-EnsembleBuilder - INFO - DummyFuture: ([{'Timestamp': Timestamp('2022-01-25 16:22:18.897617'), 'ensemble_optimization_score': 0.9886363636363636, 'ensemble_test_score': 0.9899874843554443}], 50, None, None, None)/SingleThreadedClient() Started Ensemble builder job at 2022.01.25-16.22.18 for iteration 0.
+    2022-01-25 16:22:23,231 - Client-EnsembleBuilder - INFO - DummyFuture: ([{'Timestamp': Timestamp('2022-01-25 16:22:23.213965'), 'ensemble_optimization_score': 0.9911616161616161, 'ensemble_test_score': 0.9899874843554443}], 50, None, None, None)/SingleThreadedClient() Started Ensemble builder job at 2022.01.25-16.22.23 for iteration 1.
+    tmp_folder/distributed.log
     tmp_folder/.auto-sklearn
-    tmp_folder/AutoML(1):893634d8-4196-11ec-8a32-1b4c952643c3.log
-    tmp_folder/space.json
     tmp_folder/smac3-output
-    tmp_folder/distributed.log
+    tmp_folder/AutoML(1):f52bd18f-7dfa-11ec-875f-3137a7553b01.log
+    tmp_folder/space.json
 
 
 
@@ -188,7 +188,7 @@ Build and fit a classifier
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 0 minutes  22.911 seconds)
+   **Total running time of the script:** ( 0 minutes  14.975 seconds)
 
 
 .. _sphx_glr_download_examples_40_advanced_example_debug_logging.py:
diff --git a/master/_sources/examples/40_advanced/example_feature_types.rst.txt b/master/_sources/examples/40_advanced/example_feature_types.rst.txt
index e46143bed8..ee7d6e482f 100644
--- a/master/_sources/examples/40_advanced/example_feature_types.rst.txt
+++ b/master/_sources/examples/40_advanced/example_feature_types.rst.txt
@@ -161,7 +161,7 @@ Get the Score of the final ensemble
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 0 minutes  10.620 seconds)
+   **Total running time of the script:** ( 0 minutes  8.763 seconds)
 
 
 .. _sphx_glr_download_examples_40_advanced_example_feature_types.py:
diff --git a/master/_sources/examples/40_advanced/example_get_pipeline_components.rst.txt b/master/_sources/examples/40_advanced/example_get_pipeline_components.rst.txt
index 921f6bb393..7d4b41464a 100644
--- a/master/_sources/examples/40_advanced/example_get_pipeline_components.rst.txt
+++ b/master/_sources/examples/40_advanced/example_get_pipeline_components.rst.txt
@@ -32,10 +32,12 @@ Auto-sklearn is a wrapper on top of
 the sklearn models. This example illustrates how to interact
 with the sklearn components directly, in this case a PCA preprocessor.
 
-.. GENERATED FROM PYTHON SOURCE LINES 17-23
+.. GENERATED FROM PYTHON SOURCE LINES 17-25
 
 .. code-block:: default
 
+    from pprint import pprint
+
     import sklearn.datasets
     import sklearn.metrics
 
@@ -49,12 +51,12 @@ with the sklearn components directly, in this case a PCA preprocessor.
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 24-26
+.. GENERATED FROM PYTHON SOURCE LINES 26-28
 
 Data Loading
 ============
 
-.. GENERATED FROM PYTHON SOURCE LINES 26-31
+.. GENERATED FROM PYTHON SOURCE LINES 28-33
 
 .. code-block:: default
 
@@ -70,12 +72,12 @@ Data Loading
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 32-34
+.. GENERATED FROM PYTHON SOURCE LINES 34-36
 
 Build and fit the classifier
 ============================
 
-.. GENERATED FROM PYTHON SOURCE LINES 34-47
+.. GENERATED FROM PYTHON SOURCE LINES 36-49
 
 .. code-block:: default
 
@@ -102,18 +104,22 @@ Build and fit the classifier
 
  .. code-block:: none
 
+    /home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:68: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+      self.metafeatures = self.metafeatures.append(metafeatures)
+    /home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:72: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+      self.algorithm_runs[metric].append(runs)
 
     AutoSklearnClassifier(include={'feature_preprocessor': ['pca']},
                           per_run_time_limit=10, time_left_for_this_task=30)
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 48-50
+.. GENERATED FROM PYTHON SOURCE LINES 50-52
 
 Predict using the model
 =======================
 
-.. GENERATED FROM PYTHON SOURCE LINES 50-57
+.. GENERATED FROM PYTHON SOURCE LINES 52-59
 
 .. code-block:: default
 
@@ -134,12 +140,12 @@ Predict using the model
 
  .. code-block:: none
 
-    Accuracy score:0.951048951048951
+    Accuracy score:0.958041958041958
 
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 58-67
+.. GENERATED FROM PYTHON SOURCE LINES 60-76
 
 Report the models found by Auto-Sklearn
 =======================================
@@ -148,15 +154,22 @@ Auto-sklearn uses
 `Ensemble Selection <https://www.cs.cornell.edu/~alexn/papers/shotgun.icml04.revised.rev2.pdf>`_
 to construct ensembles in a post-hoc fashion. The ensemble is a linear
 weighting of all models constructed during the hyperparameter optimization.
-This prints the final ensemble. It is a list of tuples, each tuple being
-the model weight in the ensemble and the model itself.
-
-.. GENERATED FROM PYTHON SOURCE LINES 67-70
+This prints the final ensemble. It is a dictionary where ``model_id`` of
+each model is a key, and value is a dictionary containing information
+of that model. A model's dict contains its ``'model_id'``, ``'rank'``,
+``'cost'``, ``'ensemble_weight'``, and the model itself. The model is
+given by the ``'data_preprocessor'``, ``'feature_preprocessor'``,
+``'regressor'/'classifier'`` and ``'sklearn_regressor'/'sklearn_classifier'``
+entries. But for the ``'cv'`` resampling strategy, the same for each cv
+model is stored in the ``'estimators'`` list in the dict, along with the
+``'voting_model'``.
+
+.. GENERATED FROM PYTHON SOURCE LINES 76-79
 
 .. code-block:: default
 
 
-    print(automl.show_models())
+    pprint(automl.show_models(), indent=4)
 
 
 
@@ -168,76 +181,110 @@ the model weight in the ensemble and the model itself.
 
  .. code-block:: none
 
-    [(0.340000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'sgd', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'pca', 'classifier:sgd:alpha': 1.0333981795670597e-07, 'classifier:sgd:average': 'False', 'classifier:sgd:fit_intercept': 'True', 'classifier:sgd:learning_rate': 'constant', 'classifier:sgd:loss': 'modified_huber', 'classifier:sgd:penalty': 'l2', 'classifier:sgd:tol': 2.5846457091661748e-05, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'minmax', 'feature_preprocessor:pca:keep_variance': 0.9755563101648113, 'feature_preprocessor:pca:whiten': 'True', 'classifier:sgd:epsilon': 0.03457277874230573, 'classifier:sgd:eta0': 0.004635634934629576, 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.00015788887334393744},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.160000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'mlp', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'pca', 'classifier:mlp:activation': 'tanh', 'classifier:mlp:alpha': 1.103855734598575e-05, 'classifier:mlp:batch_size': 'auto', 'classifier:mlp:beta_1': 0.9, 'classifier:mlp:beta_2': 0.999, 'classifier:mlp:early_stopping': 'valid', 'classifier:mlp:epsilon': 1e-08, 'classifier:mlp:hidden_layer_depth': 3, 'classifier:mlp:learning_rate_init': 0.00014375616988222174, 'classifier:mlp:n_iter_no_change': 32, 'classifier:mlp:num_nodes_per_layer': 229, 'classifier:mlp:shuffle': 'True', 'classifier:mlp:solver': 'adam', 'classifier:mlp:tol': 0.0001, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'quantile_transformer', 'feature_preprocessor:pca:keep_variance': 0.7895711479212801, 'feature_preprocessor:pca:whiten': 'True', 'classifier:mlp:validation_fraction': 0.1, 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.002766772136115771, 'data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles': 180, 'data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution': 'uniform'},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.140000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'liblinear_svc', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'pca', 'classifier:liblinear_svc:C': 10.369811497206404, 'classifier:liblinear_svc:dual': 'False', 'classifier:liblinear_svc:fit_intercept': 'True', 'classifier:liblinear_svc:intercept_scaling': 1, 'classifier:liblinear_svc:loss': 'squared_hinge', 'classifier:liblinear_svc:multi_class': 'ovr', 'classifier:liblinear_svc:penalty': 'l2', 'classifier:liblinear_svc:tol': 0.0015130257264171173, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'power_transformer', 'feature_preprocessor:pca:keep_variance': 0.6661824659281315, 'feature_preprocessor:pca:whiten': 'False'},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.120000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'random_forest', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'pca', 'classifier:random_forest:bootstrap': 'True', 'classifier:random_forest:criterion': 'gini', 'classifier:random_forest:max_depth': 'None', 'classifier:random_forest:max_features': 0.9331254454871041, 'classifier:random_forest:max_leaf_nodes': 'None', 'classifier:random_forest:min_impurity_decrease': 0.0, 'classifier:random_forest:min_samples_leaf': 2, 'classifier:random_forest:min_samples_split': 20, 'classifier:random_forest:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'none', 'feature_preprocessor:pca:keep_variance': 0.9967857433838874, 'feature_preprocessor:pca:whiten': 'False', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.005069923784737444},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.120000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'random_forest', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'pca', 'classifier:random_forest:bootstrap': 'True', 'classifier:random_forest:criterion': 'gini', 'classifier:random_forest:max_depth': 'None', 'classifier:random_forest:max_features': 0.5, 'classifier:random_forest:max_leaf_nodes': 'None', 'classifier:random_forest:min_impurity_decrease': 0.0, 'classifier:random_forest:min_samples_leaf': 1, 'classifier:random_forest:min_samples_split': 2, 'classifier:random_forest:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize', 'feature_preprocessor:pca:keep_variance': 0.9999, 'feature_preprocessor:pca:whiten': 'False', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.01},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.040000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'decision_tree', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'pca', 'classifier:decision_tree:criterion': 'gini', 'classifier:decision_tree:max_depth_factor': 0.14069368736662313, 'classifier:decision_tree:max_features': 1.0, 'classifier:decision_tree:max_leaf_nodes': 'None', 'classifier:decision_tree:min_impurity_decrease': 0.0, 'classifier:decision_tree:min_samples_leaf': 5, 'classifier:decision_tree:min_samples_split': 19, 'classifier:decision_tree:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'quantile_transformer', 'feature_preprocessor:pca:keep_variance': 0.7772788812704434, 'feature_preprocessor:pca:whiten': 'True', 'data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles': 526, 'data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution': 'uniform'},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.040000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'k_nearest_neighbors', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'pca', 'classifier:k_nearest_neighbors:n_neighbors': 4, 'classifier:k_nearest_neighbors:p': 2, 'classifier:k_nearest_neighbors:weights': 'distance', 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'normalize', 'feature_preprocessor:pca:keep_variance': 0.8047274080856589, 'feature_preprocessor:pca:whiten': 'False', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.008015420020402715},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.040000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'libsvm_svc', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'pca', 'classifier:libsvm_svc:C': 100.5905006626969, 'classifier:libsvm_svc:gamma': 0.011333066835975528, 'classifier:libsvm_svc:kernel': 'poly', 'classifier:libsvm_svc:max_iter': -1, 'classifier:libsvm_svc:shrinking': 'True', 'classifier:libsvm_svc:tol': 0.012391313886912093, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'minmax', 'feature_preprocessor:pca:keep_variance': 0.9290439925152777, 'feature_preprocessor:pca:whiten': 'False', 'classifier:libsvm_svc:coef0': 0.08087614244138486, 'classifier:libsvm_svc:degree': 3, 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.010000000000000004},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    ]
-
-
-
-
-.. GENERATED FROM PYTHON SOURCE LINES 71-76
+    {   2: {   'balancing': Balancing(random_state=1),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78b929ac0>,
+               'cost': 0.07801418439716312,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78ca86640>,
+               'ensemble_weight': 0.04,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc790382760>,
+               'model_id': 2,
+               'rank': 7,
+               'sklearn_classifier': RandomForestClassifier(max_features=5, n_estimators=512, n_jobs=1,
+                           random_state=1, warm_start=True)},
+        3: {   'balancing': Balancing(random_state=1),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc786b361f0>,
+               'cost': 0.07092198581560283,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78a826520>,
+               'ensemble_weight': 0.18,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc786b36910>,
+               'model_id': 3,
+               'rank': 6,
+               'sklearn_classifier': RandomForestClassifier(max_features=1, min_samples_leaf=2, min_samples_split=20,
+                           n_estimators=512, n_jobs=1, random_state=1,
+                           warm_start=True)},
+        4: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc7901c5820>,
+               'cost': 0.028368794326241176,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7901c53d0>,
+               'ensemble_weight': 0.08,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc79032af10>,
+               'model_id': 4,
+               'rank': 1,
+               'sklearn_classifier': MLPClassifier(activation='tanh', alpha=1.103855734598575e-05, beta_1=0.999,
+                  beta_2=0.9, early_stopping=True,
+                  hidden_layer_sizes=(229, 229, 229),
+                  learning_rate_init=0.00014375616988222174, max_iter=32,
+                  n_iter_no_change=32, random_state=1, verbose=0, warm_start=True)},
+        5: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78c9441c0>,
+               'cost': 0.1063829787234043,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78a5eb790>,
+               'ensemble_weight': 0.04,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78f8d1970>,
+               'model_id': 5,
+               'rank': 8,
+               'sklearn_classifier': KNeighborsClassifier(n_neighbors=4, weights='distance')},
+        6: {   'balancing': Balancing(random_state=1),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78c90c6d0>,
+               'cost': 0.11347517730496459,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78b289160>,
+               'ensemble_weight': 0.16,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c90c400>,
+               'model_id': 6,
+               'rank': 9,
+               'sklearn_classifier': SVC(C=100.5905006626969, cache_size=1799.9375, coef0=0.08087614244138486,
+        gamma=0.011333066835975528, kernel='poly', max_iter=-1.0, random_state=1,
+        tol=0.012391313886912093)},
+        7: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78a7e35b0>,
+               'cost': 0.03546099290780147,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7903465e0>,
+               'ensemble_weight': 0.16,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc790169d30>,
+               'model_id': 7,
+               'rank': 3,
+               'sklearn_classifier': LinearSVC(C=10.369811497206404, class_weight='balanced', dual=False,
+              intercept_scaling=1.0, random_state=1, tol=0.0015130257264171173)},
+        8: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78bcf7dc0>,
+               'cost': 0.05673758865248224,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78a892370>,
+               'ensemble_weight': 0.02,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78bcf7910>,
+               'model_id': 8,
+               'rank': 4,
+               'sklearn_classifier': DecisionTreeClassifier(class_weight='balanced', max_depth=1, min_samples_leaf=5,
+                           min_samples_split=19, random_state=1)},
+        9: {   'balancing': Balancing(random_state=1),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78a5d3d60>,
+               'cost': 0.028368794326241176,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78a593c70>,
+               'ensemble_weight': 0.26,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc793fbc9d0>,
+               'model_id': 9,
+               'rank': 2,
+               'sklearn_classifier': SGDClassifier(alpha=1.0333981795670635e-07, epsilon=0.03457277874230573,
+                  eta0=0.004635634934629583, learning_rate='constant',
+                  loss='modified_huber', max_iter=128, random_state=1,
+                  tol=2.5846457091661748e-05, warm_start=True)},
+        10: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc786687c10>,
+                'cost': 0.05673758865248224,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78c94dc10>,
+                'ensemble_weight': 0.06,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc7903bdbb0>,
+                'model_id': 10,
+                'rank': 5,
+                'sklearn_classifier': SGDClassifier(alpha=0.0006046872822704108, epsilon=0.0006085470273398056,
+                  eta0=0.0003474820560182061, l1_ratio=4.5783995454757295e-09,
+                  learning_rate='invscaling', loss='modified_huber', max_iter=1024,
+                  penalty='elasticnet', power_t=0.6064541566557635, random_state=1,
+                  tol=1.7244397014112777e-05, warm_start=True)}}
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 80-85
 
 Report statistics about the search
 ==================================
@@ -245,7 +292,7 @@ Report statistics about the search
 Print statistics about the auto-sklearn run such as number of
 iterations, number of models failed with a time out etc.
 
-.. GENERATED FROM PYTHON SOURCE LINES 76-78
+.. GENERATED FROM PYTHON SOURCE LINES 85-87
 
 .. code-block:: default
 
@@ -265,8 +312,8 @@ iterations, number of models failed with a time out etc.
       Dataset name: breast_cancer
       Metric: accuracy
       Best validation score: 0.971631
-      Number of target algorithm runs: 8
-      Number of successful target algorithm runs: 8
+      Number of target algorithm runs: 9
+      Number of successful target algorithm runs: 9
       Number of crashed target algorithm runs: 0
       Number of target algorithms that exceeded the time limit: 0
       Number of target algorithms that exceeded the memory limit: 0
@@ -275,7 +322,7 @@ iterations, number of models failed with a time out etc.
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 79-86
+.. GENERATED FROM PYTHON SOURCE LINES 88-95
 
 Detailed statistics about the search - part 1
 =============================================
@@ -285,7 +332,7 @@ optimization procedurce, which are stored in a so-called
 `run history <https://automl.github.io/SMAC3/master/apidoc/smac.
 runhistory.runhistory.html#smac.runhistory.runhistory.RunHistory>`_.
 
-.. GENERATED FROM PYTHON SOURCE LINES 86-89
+.. GENERATED FROM PYTHON SOURCE LINES 95-98
 
 .. code-block:: default
 
@@ -302,16 +349,16 @@ runhistory.runhistory.html#smac.runhistory.runhistory.RunHistory>`_.
 
  .. code-block:: none
 
-    <smac.runhistory.runhistory.RunHistory object at 0x7f9379a40f10>
+    <smac.runhistory.runhistory.RunHistory object at 0x7fc78ab3f670>
 
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 90-91
+.. GENERATED FROM PYTHON SOURCE LINES 99-100
 
 Runs are stored inside an ``OrderedDict`` called ``data``:
 
-.. GENERATED FROM PYTHON SOURCE LINES 91-94
+.. GENERATED FROM PYTHON SOURCE LINES 100-103
 
 .. code-block:: default
 
@@ -328,16 +375,16 @@ Runs are stored inside an ``OrderedDict`` called ``data``:
 
  .. code-block:: none
 
-    9
+    10
 
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 95-96
+.. GENERATED FROM PYTHON SOURCE LINES 104-105
 
 Let's iterative over all entries
 
-.. GENERATED FROM PYTHON SOURCE LINES 96-102
+.. GENERATED FROM PYTHON SOURCE LINES 105-111
 
 .. code-block:: default
 
@@ -359,40 +406,43 @@ Let's iterative over all entries
 
     #########
     RunKey(config_id=1, instance_id='{"task_id": "breast_cancer"}', seed=0, budget=0.0)
-    RunValue(cost=0.07801418439716312, time=1.9545562267303467, status=<StatusType.SUCCESS: 1>, starttime=1636488092.7797232, endtime=1636488094.7563245, additional_info={'duration': 1.8572731018066406, 'num_run': 2, 'train_loss': 0.0, 'configuration_origin': 'Initial design'})
+    RunValue(cost=0.07801418439716312, time=1.5860536098480225, status=<StatusType.SUCCESS: 1>, starttime=1643128262.388968, endtime=1643128263.9952888, additional_info={'duration': 1.4848618507385254, 'num_run': 2, 'train_loss': 0.0, 'configuration_origin': 'Initial design'})
     #########
     RunKey(config_id=2, instance_id='{"task_id": "breast_cancer"}', seed=0, budget=0.0)
-    RunValue(cost=0.07092198581560283, time=1.6941378116607666, status=<StatusType.SUCCESS: 1>, starttime=1636488094.8219044, endtime=1636488096.537016, additional_info={'duration': 1.6041510105133057, 'num_run': 3, 'train_loss': 0.06315789473684208, 'configuration_origin': 'Initial design'})
+    RunValue(cost=0.07092198581560283, time=1.3667099475860596, status=<StatusType.SUCCESS: 1>, starttime=1643128264.0590467, endtime=1643128265.4495397, additional_info={'duration': 1.2673461437225342, 'num_run': 3, 'train_loss': 0.06315789473684208, 'configuration_origin': 'Initial design'})
     #########
     RunKey(config_id=3, instance_id='{"task_id": "breast_cancer"}', seed=0, budget=0.0)
-    RunValue(cost=0.028368794326241176, time=1.6140191555023193, status=<StatusType.SUCCESS: 1>, starttime=1636488096.6401973, endtime=1636488098.2758095, additional_info={'duration': 1.5358099937438965, 'num_run': 4, 'train_loss': 0.04210526315789476, 'configuration_origin': 'Initial design'})
+    RunValue(cost=0.028368794326241176, time=1.1778380870819092, status=<StatusType.SUCCESS: 1>, starttime=1643128265.5568986, endtime=1643128266.7576518, additional_info={'duration': 1.1020641326904297, 'num_run': 4, 'train_loss': 0.04210526315789476, 'configuration_origin': 'Initial design'})
     #########
     RunKey(config_id=4, instance_id='{"task_id": "breast_cancer"}', seed=0, budget=0.0)
-    RunValue(cost=0.1063829787234043, time=0.6702926158905029, status=<StatusType.SUCCESS: 1>, starttime=1636488098.4037113, endtime=1636488099.0938587, additional_info={'duration': 0.6066544055938721, 'num_run': 5, 'train_loss': 0.0, 'configuration_origin': 'Initial design'})
+    RunValue(cost=0.1063829787234043, time=0.6564128398895264, status=<StatusType.SUCCESS: 1>, starttime=1643128266.868573, endtime=1643128267.5475628, additional_info={'duration': 0.5852375030517578, 'num_run': 5, 'train_loss': 0.0, 'configuration_origin': 'Initial design'})
     #########
     RunKey(config_id=5, instance_id='{"task_id": "breast_cancer"}', seed=0, budget=0.0)
-    RunValue(cost=0.11347517730496459, time=0.5629768371582031, status=<StatusType.SUCCESS: 1>, starttime=1636488099.2329674, endtime=1636488099.8163307, additional_info={'duration': 0.5027651786804199, 'num_run': 6, 'train_loss': 0.09122807017543855, 'configuration_origin': 'Initial design'})
+    RunValue(cost=0.11347517730496459, time=0.5501728057861328, status=<StatusType.SUCCESS: 1>, starttime=1643128267.676605, endtime=1643128268.2494493, additional_info={'duration': 0.4777560234069824, 'num_run': 6, 'train_loss': 0.09122807017543855, 'configuration_origin': 'Initial design'})
     #########
     RunKey(config_id=6, instance_id='{"task_id": "breast_cancer"}', seed=0, budget=0.0)
-    RunValue(cost=0.03546099290780147, time=0.9639365673065186, status=<StatusType.SUCCESS: 1>, starttime=1636488102.2641013, endtime=1636488103.2531207, additional_info={'duration': 0.8704240322113037, 'num_run': 7, 'train_loss': 0.04561403508771933, 'configuration_origin': 'Random Search (sorted)'})
+    RunValue(cost=0.03546099290780147, time=0.7203934192657471, status=<StatusType.SUCCESS: 1>, starttime=1643128270.32824, endtime=1643128271.0703733, additional_info={'duration': 0.6468265056610107, 'num_run': 7, 'train_loss': 0.04561403508771933, 'configuration_origin': 'Random Search (sorted)'})
     #########
     RunKey(config_id=7, instance_id='{"task_id": "breast_cancer"}', seed=0, budget=0.0)
-    RunValue(cost=0.05673758865248224, time=0.8888447284698486, status=<StatusType.SUCCESS: 1>, starttime=1636488105.9350605, endtime=1636488106.851598, additional_info={'duration': 0.7990398406982422, 'num_run': 8, 'train_loss': 0.06315789473684208, 'configuration_origin': 'Random Search (sorted)'})
+    RunValue(cost=0.05673758865248224, time=0.6808576583862305, status=<StatusType.SUCCESS: 1>, starttime=1643128273.2836225, endtime=1643128273.9862375, additional_info={'duration': 0.6066226959228516, 'num_run': 8, 'train_loss': 0.06315789473684208, 'configuration_origin': 'Random Search (sorted)'})
     #########
     RunKey(config_id=8, instance_id='{"task_id": "breast_cancer"}', seed=0, budget=0.0)
-    RunValue(cost=0.028368794326241176, time=0.7385828495025635, status=<StatusType.SUCCESS: 1>, starttime=1636488107.1076975, endtime=1636488107.8694334, additional_info={'duration': 0.6549544334411621, 'num_run': 9, 'train_loss': 0.007017543859649145, 'configuration_origin': 'Random Search'})
+    RunValue(cost=0.028368794326241176, time=0.5615131855010986, status=<StatusType.SUCCESS: 1>, starttime=1643128274.1657884, endtime=1643128274.7493172, additional_info={'duration': 0.48226284980773926, 'num_run': 9, 'train_loss': 0.007017543859649145, 'configuration_origin': 'Random Search'})
     #########
     RunKey(config_id=9, instance_id='{"task_id": "breast_cancer"}', seed=0, budget=0.0)
-    RunValue(cost=1.0, time=0.0, status=<StatusType.STOP: 8>, starttime=1636488110.8355012, endtime=1636488110.8355012, additional_info={})
+    RunValue(cost=0.05673758865248224, time=0.6582047939300537, status=<StatusType.SUCCESS: 1>, starttime=1643128277.5765781, endtime=1643128278.254898, additional_info={'duration': 0.5808424949645996, 'num_run': 10, 'train_loss': 0.06666666666666665, 'configuration_origin': 'Random Search (sorted)'})
+    #########
+    RunKey(config_id=10, instance_id='{"task_id": "breast_cancer"}', seed=0, budget=0.0)
+    RunValue(cost=1.0, time=0.0, status=<StatusType.STOP: 8>, starttime=1643128280.7154517, endtime=1643128280.715452, additional_info={})
 
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 103-104
+.. GENERATED FROM PYTHON SOURCE LINES 112-113
 
 and have a detailed look at one entry:
 
-.. GENERATED FROM PYTHON SOURCE LINES 104-108
+.. GENERATED FROM PYTHON SOURCE LINES 113-117
 
 .. code-block:: default
 
@@ -407,11 +457,11 @@ and have a detailed look at one entry:
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 109-110
+.. GENERATED FROM PYTHON SOURCE LINES 118-119
 
 The ``run_key`` contains all information describing a run:
 
-.. GENERATED FROM PYTHON SOURCE LINES 110-116
+.. GENERATED FROM PYTHON SOURCE LINES 119-125
 
 .. code-block:: default
 
@@ -439,11 +489,11 @@ The ``run_key`` contains all information describing a run:
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 117-118
+.. GENERATED FROM PYTHON SOURCE LINES 126-127
 
 and the configuration can be looked up in the run history as well:
 
-.. GENERATED FROM PYTHON SOURCE LINES 118-121
+.. GENERATED FROM PYTHON SOURCE LINES 127-130
 
 .. code-block:: default
 
@@ -486,7 +536,7 @@ and the configuration can be looked up in the run history as well:
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 122-128
+.. GENERATED FROM PYTHON SOURCE LINES 131-137
 
 The only other important entry is the budget in case you are using
 auto-sklearn with
@@ -495,11 +545,11 @@ The remaining parts of the key can be ignored for auto-sklearn and are
 only there because the underlying optimizer, SMAC, can handle more general
 problems, too.
 
-.. GENERATED FROM PYTHON SOURCE LINES 130-131
+.. GENERATED FROM PYTHON SOURCE LINES 139-140
 
 The ``run_value`` contains all output from running the configuration:
 
-.. GENERATED FROM PYTHON SOURCE LINES 131-139
+.. GENERATED FROM PYTHON SOURCE LINES 140-148
 
 .. code-block:: default
 
@@ -522,16 +572,16 @@ The ``run_value`` contains all output from running the configuration:
  .. code-block:: none
 
     Cost: 0.07801418439716312
-    Time: 1.9545562267303467
+    Time: 1.5860536098480225
     Status: StatusType.SUCCESS
-    Additional information: {'duration': 1.8572731018066406, 'num_run': 2, 'train_loss': 0.0, 'configuration_origin': 'Initial design'}
-    Start time: 1636488092.7797232
-    End time 1636488094.7563245
+    Additional information: {'duration': 1.4848618507385254, 'num_run': 2, 'train_loss': 0.0, 'configuration_origin': 'Initial design'}
+    Start time: 1643128262.388968
+    End time 1643128263.9952888
 
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 140-146
+.. GENERATED FROM PYTHON SOURCE LINES 149-155
 
 Cost is basically the same as a loss. In case the metric to optimize for
 should be maximized, it is internally transformed into a minimization
@@ -540,13 +590,13 @@ was successful, while the additional information's most interesting entry
 is the internal training loss. Furthermore, there is detailed information
 on the runtime available.
 
-.. GENERATED FROM PYTHON SOURCE LINES 148-151
+.. GENERATED FROM PYTHON SOURCE LINES 157-160
 
 As an example, let's find the best configuration evaluated. As
 Auto-sklearn solves a minimization problem internally, we need to look
 for the entry with the lowest loss:
 
-.. GENERATED FROM PYTHON SOURCE LINES 151-163
+.. GENERATED FROM PYTHON SOURCE LINES 160-172
 
 .. code-block:: default
 
@@ -607,7 +657,7 @@ for the entry with the lowest loss:
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 164-171
+.. GENERATED FROM PYTHON SOURCE LINES 173-180
 
 Detailed statistics about the search - part 2
 =============================================
@@ -617,7 +667,7 @@ same data as
 `cv_results_ <https://scikit-learn.org/stable/modules/generated/sklearn.
 model_selection.GridSearchCV.html>`_.
 
-.. GENERATED FROM PYTHON SOURCE LINES 171-174
+.. GENERATED FROM PYTHON SOURCE LINES 180-183
 
 .. code-block:: default
 
@@ -635,332 +685,450 @@ model_selection.GridSearchCV.html>`_.
  .. code-block:: none
 
     {'mean_test_score': array([0.92198582, 0.92907801, 0.97163121, 0.89361702, 0.88652482,
-           0.96453901, 0.94326241, 0.97163121]), 'mean_fit_time': array([1.95455623, 1.69413781, 1.61401916, 0.67029262, 0.56297684,
-           0.96393657, 0.88884473, 0.73858285]), 'params': [{'balancing:strategy': 'none', 'classifier:__choice__': 'random_forest', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'pca', 'classifier:random_forest:bootstrap': 'True', 'classifier:random_forest:criterion': 'gini', 'classifier:random_forest:max_depth': 'None', 'classifier:random_forest:max_features': 0.5, 'classifier:random_forest:max_leaf_nodes': 'None', 'classifier:random_forest:min_impurity_decrease': 0.0, 'classifier:random_forest:min_samples_leaf': 1, 'classifier:random_forest:min_samples_split': 2, 'classifier:random_forest:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize', 'feature_preprocessor:pca:keep_variance': 0.9999, 'feature_preprocessor:pca:whiten': 'False', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.01}, {'balancing:strategy': 'none', 'classifier:__choice__': 'random_forest', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'pca', 'classifier:random_forest:bootstrap': 'True', 'classifier:random_forest:criterion': 'gini', 'classifier:random_forest:max_depth': 'None', 'classifier:random_forest:max_features': 0.9331254454871041, 'classifier:random_forest:max_leaf_nodes': 'None', 'classifier:random_forest:min_impurity_decrease': 0.0, 'classifier:random_forest:min_samples_leaf': 2, 'classifier:random_forest:min_samples_split': 20, 'classifier:random_forest:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'none', 'feature_preprocessor:pca:keep_variance': 0.9967857433838874, 'feature_preprocessor:pca:whiten': 'False', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.005069923784737444}, {'balancing:strategy': 'weighting', 'classifier:__choice__': 'mlp', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'pca', 'classifier:mlp:activation': 'tanh', 'classifier:mlp:alpha': 1.103855734598575e-05, 'classifier:mlp:batch_size': 'auto', 'classifier:mlp:beta_1': 0.9, 'classifier:mlp:beta_2': 0.999, 'classifier:mlp:early_stopping': 'valid', 'classifier:mlp:epsilon': 1e-08, 'classifier:mlp:hidden_layer_depth': 3, 'classifier:mlp:learning_rate_init': 0.00014375616988222174, 'classifier:mlp:n_iter_no_change': 32, 'classifier:mlp:num_nodes_per_layer': 229, 'classifier:mlp:shuffle': 'True', 'classifier:mlp:solver': 'adam', 'classifier:mlp:tol': 0.0001, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'quantile_transformer', 'feature_preprocessor:pca:keep_variance': 0.7895711479212801, 'feature_preprocessor:pca:whiten': 'True', 'classifier:mlp:validation_fraction': 0.1, 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.002766772136115771, 'data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles': 180, 'data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution': 'uniform'}, {'balancing:strategy': 'weighting', 'classifier:__choice__': 'k_nearest_neighbors', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'pca', 'classifier:k_nearest_neighbors:n_neighbors': 4, 'classifier:k_nearest_neighbors:p': 2, 'classifier:k_nearest_neighbors:weights': 'distance', 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'normalize', 'feature_preprocessor:pca:keep_variance': 0.8047274080856589, 'feature_preprocessor:pca:whiten': 'False', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.008015420020402715}, {'balancing:strategy': 'none', 'classifier:__choice__': 'libsvm_svc', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'pca', 'classifier:libsvm_svc:C': 100.5905006626969, 'classifier:libsvm_svc:gamma': 0.011333066835975528, 'classifier:libsvm_svc:kernel': 'poly', 'classifier:libsvm_svc:max_iter': -1, 'classifier:libsvm_svc:shrinking': 'True', 'classifier:libsvm_svc:tol': 0.012391313886912093, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'minmax', 'feature_preprocessor:pca:keep_variance': 0.9290439925152777, 'feature_preprocessor:pca:whiten': 'False', 'classifier:libsvm_svc:coef0': 0.08087614244138486, 'classifier:libsvm_svc:degree': 3, 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.010000000000000004}, {'balancing:strategy': 'weighting', 'classifier:__choice__': 'liblinear_svc', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'pca', 'classifier:liblinear_svc:C': 10.369811497206404, 'classifier:liblinear_svc:dual': 'False', 'classifier:liblinear_svc:fit_intercept': 'True', 'classifier:liblinear_svc:intercept_scaling': 1, 'classifier:liblinear_svc:loss': 'squared_hinge', 'classifier:liblinear_svc:multi_class': 'ovr', 'classifier:liblinear_svc:penalty': 'l2', 'classifier:liblinear_svc:tol': 0.0015130257264171173, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'power_transformer', 'feature_preprocessor:pca:keep_variance': 0.6661824659281315, 'feature_preprocessor:pca:whiten': 'False'}, {'balancing:strategy': 'weighting', 'classifier:__choice__': 'decision_tree', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'pca', 'classifier:decision_tree:criterion': 'gini', 'classifier:decision_tree:max_depth_factor': 0.14069368736662313, 'classifier:decision_tree:max_features': 1.0, 'classifier:decision_tree:max_leaf_nodes': 'None', 'classifier:decision_tree:min_impurity_decrease': 0.0, 'classifier:decision_tree:min_samples_leaf': 5, 'classifier:decision_tree:min_samples_split': 19, 'classifier:decision_tree:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'quantile_transformer', 'feature_preprocessor:pca:keep_variance': 0.7772788812704434, 'feature_preprocessor:pca:whiten': 'True', 'data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles': 526, 'data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution': 'uniform'}, {'balancing:strategy': 'none', 'classifier:__choice__': 'sgd', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'pca', 'classifier:sgd:alpha': 1.0333981795670597e-07, 'classifier:sgd:average': 'False', 'classifier:sgd:fit_intercept': 'True', 'classifier:sgd:learning_rate': 'constant', 'classifier:sgd:loss': 'modified_huber', 'classifier:sgd:penalty': 'l2', 'classifier:sgd:tol': 2.5846457091661748e-05, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'minmax', 'feature_preprocessor:pca:keep_variance': 0.9755563101648113, 'feature_preprocessor:pca:whiten': 'True', 'classifier:sgd:epsilon': 0.03457277874230573, 'classifier:sgd:eta0': 0.004635634934629576, 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.00015788887334393744}], 'rank_test_scores': array([6, 5, 1, 7, 8, 3, 4, 1]), 'status': ['Success', 'Success', 'Success', 'Success', 'Success', 'Success', 'Success', 'Success'], 'budgets': [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0], 'param_balancing:strategy': masked_array(data=['none', 'none', 'weighting', 'weighting', 'none',
-                       'weighting', 'weighting', 'none'],
-                 mask=[False, False, False, False, False, False, False, False],
+           0.96453901, 0.94326241, 0.97163121, 0.94326241]), 'mean_fit_time': array([1.58605361, 1.36670995, 1.17783809, 0.65641284, 0.55017281,
+           0.72039342, 0.68085766, 0.56151319, 0.65820479]), 'params': [{'balancing:strategy': 'none', 'classifier:__choice__': 'random_forest', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'pca', 'classifier:random_forest:bootstrap': 'True', 'classifier:random_forest:criterion': 'gini', 'classifier:random_forest:max_depth': 'None', 'classifier:random_forest:max_features': 0.5, 'classifier:random_forest:max_leaf_nodes': 'None', 'classifier:random_forest:min_impurity_decrease': 0.0, 'classifier:random_forest:min_samples_leaf': 1, 'classifier:random_forest:min_samples_split': 2, 'classifier:random_forest:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize', 'feature_preprocessor:pca:keep_variance': 0.9999, 'feature_preprocessor:pca:whiten': 'False', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.01}, {'balancing:strategy': 'none', 'classifier:__choice__': 'random_forest', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'pca', 'classifier:random_forest:bootstrap': 'True', 'classifier:random_forest:criterion': 'gini', 'classifier:random_forest:max_depth': 'None', 'classifier:random_forest:max_features': 0.9331254454871041, 'classifier:random_forest:max_leaf_nodes': 'None', 'classifier:random_forest:min_impurity_decrease': 0.0, 'classifier:random_forest:min_samples_leaf': 2, 'classifier:random_forest:min_samples_split': 20, 'classifier:random_forest:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'none', 'feature_preprocessor:pca:keep_variance': 0.9967857433838874, 'feature_preprocessor:pca:whiten': 'False', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.005069923784737444}, {'balancing:strategy': 'weighting', 'classifier:__choice__': 'mlp', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'pca', 'classifier:mlp:activation': 'tanh', 'classifier:mlp:alpha': 1.103855734598575e-05, 'classifier:mlp:batch_size': 'auto', 'classifier:mlp:beta_1': 0.9, 'classifier:mlp:beta_2': 0.999, 'classifier:mlp:early_stopping': 'valid', 'classifier:mlp:epsilon': 1e-08, 'classifier:mlp:hidden_layer_depth': 3, 'classifier:mlp:learning_rate_init': 0.00014375616988222174, 'classifier:mlp:n_iter_no_change': 32, 'classifier:mlp:num_nodes_per_layer': 229, 'classifier:mlp:shuffle': 'True', 'classifier:mlp:solver': 'adam', 'classifier:mlp:tol': 0.0001, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'quantile_transformer', 'feature_preprocessor:pca:keep_variance': 0.7895711479212801, 'feature_preprocessor:pca:whiten': 'True', 'classifier:mlp:validation_fraction': 0.1, 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.002766772136115771, 'data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles': 180, 'data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution': 'uniform'}, {'balancing:strategy': 'weighting', 'classifier:__choice__': 'k_nearest_neighbors', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'pca', 'classifier:k_nearest_neighbors:n_neighbors': 4, 'classifier:k_nearest_neighbors:p': 2, 'classifier:k_nearest_neighbors:weights': 'distance', 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'normalize', 'feature_preprocessor:pca:keep_variance': 0.8047274080856589, 'feature_preprocessor:pca:whiten': 'False', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.008015420020402715}, {'balancing:strategy': 'none', 'classifier:__choice__': 'libsvm_svc', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'pca', 'classifier:libsvm_svc:C': 100.5905006626969, 'classifier:libsvm_svc:gamma': 0.011333066835975528, 'classifier:libsvm_svc:kernel': 'poly', 'classifier:libsvm_svc:max_iter': -1, 'classifier:libsvm_svc:shrinking': 'True', 'classifier:libsvm_svc:tol': 0.012391313886912093, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'minmax', 'feature_preprocessor:pca:keep_variance': 0.9290439925152777, 'feature_preprocessor:pca:whiten': 'False', 'classifier:libsvm_svc:coef0': 0.08087614244138486, 'classifier:libsvm_svc:degree': 3, 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.010000000000000004}, {'balancing:strategy': 'weighting', 'classifier:__choice__': 'liblinear_svc', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'pca', 'classifier:liblinear_svc:C': 10.369811497206404, 'classifier:liblinear_svc:dual': 'False', 'classifier:liblinear_svc:fit_intercept': 'True', 'classifier:liblinear_svc:intercept_scaling': 1, 'classifier:liblinear_svc:loss': 'squared_hinge', 'classifier:liblinear_svc:multi_class': 'ovr', 'classifier:liblinear_svc:penalty': 'l2', 'classifier:liblinear_svc:tol': 0.0015130257264171173, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'power_transformer', 'feature_preprocessor:pca:keep_variance': 0.6661824659281315, 'feature_preprocessor:pca:whiten': 'False'}, {'balancing:strategy': 'weighting', 'classifier:__choice__': 'decision_tree', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'pca', 'classifier:decision_tree:criterion': 'gini', 'classifier:decision_tree:max_depth_factor': 0.14069368736662313, 'classifier:decision_tree:max_features': 1.0, 'classifier:decision_tree:max_leaf_nodes': 'None', 'classifier:decision_tree:min_impurity_decrease': 0.0, 'classifier:decision_tree:min_samples_leaf': 5, 'classifier:decision_tree:min_samples_split': 19, 'classifier:decision_tree:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'quantile_transformer', 'feature_preprocessor:pca:keep_variance': 0.7772788812704434, 'feature_preprocessor:pca:whiten': 'True', 'data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles': 526, 'data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution': 'uniform'}, {'balancing:strategy': 'none', 'classifier:__choice__': 'sgd', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'pca', 'classifier:sgd:alpha': 1.0333981795670635e-07, 'classifier:sgd:average': 'False', 'classifier:sgd:fit_intercept': 'True', 'classifier:sgd:learning_rate': 'constant', 'classifier:sgd:loss': 'modified_huber', 'classifier:sgd:penalty': 'l2', 'classifier:sgd:tol': 2.5846457091661748e-05, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'minmax', 'feature_preprocessor:pca:keep_variance': 0.9755563101648113, 'feature_preprocessor:pca:whiten': 'True', 'classifier:sgd:epsilon': 0.03457277874230573, 'classifier:sgd:eta0': 0.004635634934629583, 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.00015788887334393714}, {'balancing:strategy': 'none', 'classifier:__choice__': 'sgd', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'pca', 'classifier:sgd:alpha': 0.0006046872822704108, 'classifier:sgd:average': 'False', 'classifier:sgd:fit_intercept': 'True', 'classifier:sgd:learning_rate': 'invscaling', 'classifier:sgd:loss': 'modified_huber', 'classifier:sgd:penalty': 'elasticnet', 'classifier:sgd:tol': 1.7244397014112777e-05, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'power_transformer', 'feature_preprocessor:pca:keep_variance': 0.9941379232526211, 'feature_preprocessor:pca:whiten': 'False', 'classifier:sgd:epsilon': 0.0006085470273398056, 'classifier:sgd:eta0': 0.0003474820560182061, 'classifier:sgd:l1_ratio': 4.5783995454757295e-09, 'classifier:sgd:power_t': 0.6064541566557635}], 'rank_test_scores': array([7, 6, 1, 8, 9, 3, 4, 1, 4]), 'status': ['Success', 'Success', 'Success', 'Success', 'Success', 'Success', 'Success', 'Success', 'Success'], 'budgets': [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0], 'param_balancing:strategy': masked_array(data=['none', 'none', 'weighting', 'weighting', 'none',
+                       'weighting', 'weighting', 'none', 'none'],
+                 mask=[False, False, False, False, False, False, False, False,
+                       False],
            fill_value='N/A',
                 dtype='<U9'), 'param_classifier:__choice__': masked_array(data=['random_forest', 'random_forest', 'mlp',
                        'k_nearest_neighbors', 'libsvm_svc', 'liblinear_svc',
-                       'decision_tree', 'sgd'],
-                 mask=[False, False, False, False, False, False, False, False],
+                       'decision_tree', 'sgd', 'sgd'],
+                 mask=[False, False, False, False, False, False, False, False,
+                       False],
            fill_value='N/A',
                 dtype='<U19'), 'param_data_preprocessor:__choice__': masked_array(data=['feature_type', 'feature_type', 'feature_type',
                        'feature_type', 'feature_type', 'feature_type',
-                       'feature_type', 'feature_type'],
-                 mask=[False, False, False, False, False, False, False, False],
+                       'feature_type', 'feature_type', 'feature_type'],
+                 mask=[False, False, False, False, False, False, False, False,
+                       False],
            fill_value='N/A',
-                dtype='<U12'), 'param_feature_preprocessor:__choice__': masked_array(data=['pca', 'pca', 'pca', 'pca', 'pca', 'pca', 'pca', 'pca'],
-                 mask=[False, False, False, False, False, False, False, False],
+                dtype='<U12'), 'param_feature_preprocessor:__choice__': masked_array(data=['pca', 'pca', 'pca', 'pca', 'pca', 'pca', 'pca', 'pca',
+                       'pca'],
+                 mask=[False, False, False, False, False, False, False, False,
+                       False],
            fill_value='N/A',
-                dtype='<U3'), 'param_classifier:adaboost:algorithm': masked_array(data=[--, --, --, --, --, --, --, --],
-                 mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+                dtype='<U3'), 'param_classifier:adaboost:algorithm': masked_array(data=[--, --, --, --, --, --, --, --, --],
+                 mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                        True],
            fill_value=1e+20,
-                dtype=float64), 'param_classifier:adaboost:learning_rate': masked_array(data=[--, --, --, --, --, --, --, --],
-                 mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+                dtype=float64), 'param_classifier:adaboost:learning_rate': masked_array(data=[--, --, --, --, --, --, --, --, --],
+                 mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                        True],
            fill_value=1e+20,
-                dtype=float64), 'param_classifier:adaboost:max_depth': masked_array(data=[--, --, --, --, --, --, --, --],
-                 mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+                dtype=float64), 'param_classifier:adaboost:max_depth': masked_array(data=[--, --, --, --, --, --, --, --, --],
+                 mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                        True],
            fill_value=1e+20,
-                dtype=float64), 'param_classifier:adaboost:n_estimators': masked_array(data=[--, --, --, --, --, --, --, --],
-                 mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+                dtype=float64), 'param_classifier:adaboost:n_estimators': masked_array(data=[--, --, --, --, --, --, --, --, --],
+                 mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                        True],
            fill_value=1e+20,
-                dtype=float64), 'param_classifier:bernoulli_nb:alpha': masked_array(data=[--, --, --, --, --, --, --, --],
-                 mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+                dtype=float64), 'param_classifier:bernoulli_nb:alpha': masked_array(data=[--, --, --, --, --, --, --, --, --],
+                 mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                        True],
            fill_value=1e+20,
-                dtype=float64), 'param_classifier:bernoulli_nb:fit_prior': masked_array(data=[--, --, --, --, --, --, --, --],
-                 mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+                dtype=float64), 'param_classifier:bernoulli_nb:fit_prior': masked_array(data=[--, --, --, --, --, --, --, --, --],
+                 mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                        True],
            fill_value=1e+20,
-                dtype=float64), 'param_classifier:decision_tree:criterion': masked_array(data=[--, --, --, --, --, --, 'gini', --],
-                 mask=[ True,  True,  True,  True,  True,  True, False,  True],
+                dtype=float64), 'param_classifier:decision_tree:criterion': masked_array(data=[--, --, --, --, --, --, 'gini', --, --],
+                 mask=[ True,  True,  True,  True,  True,  True, False,  True,
+                        True],
            fill_value='N/A',
-                dtype='<U32'), 'param_classifier:decision_tree:max_depth_factor': masked_array(data=[--, --, --, --, --, --, 0.14069368736662313, --],
-                 mask=[ True,  True,  True,  True,  True,  True, False,  True],
-           fill_value=1e+20), 'param_classifier:decision_tree:max_features': masked_array(data=[--, --, --, --, --, --, 1.0, --],
-                 mask=[ True,  True,  True,  True,  True,  True, False,  True],
-           fill_value=1e+20), 'param_classifier:decision_tree:max_leaf_nodes': masked_array(data=[--, --, --, --, --, --, 'None', --],
-                 mask=[ True,  True,  True,  True,  True,  True, False,  True],
+                dtype='<U32'), 'param_classifier:decision_tree:max_depth_factor': masked_array(data=[--, --, --, --, --, --, 0.14069368736662313, --, --],
+                 mask=[ True,  True,  True,  True,  True,  True, False,  True,
+                        True],
+           fill_value=1e+20), 'param_classifier:decision_tree:max_features': masked_array(data=[--, --, --, --, --, --, 1.0, --, --],
+                 mask=[ True,  True,  True,  True,  True,  True, False,  True,
+                        True],
+           fill_value=1e+20), 'param_classifier:decision_tree:max_leaf_nodes': masked_array(data=[--, --, --, --, --, --, 'None', --, --],
+                 mask=[ True,  True,  True,  True,  True,  True, False,  True,
+                        True],
            fill_value='N/A',
-                dtype='<U32'), 'param_classifier:decision_tree:min_impurity_decrease': masked_array(data=[--, --, --, --, --, --, 0.0, --],
-                 mask=[ True,  True,  True,  True,  True,  True, False,  True],
-           fill_value=1e+20), 'param_classifier:decision_tree:min_samples_leaf': masked_array(data=[--, --, --, --, --, --, 5.0, --],
-                 mask=[ True,  True,  True,  True,  True,  True, False,  True],
-           fill_value=1e+20), 'param_classifier:decision_tree:min_samples_split': masked_array(data=[--, --, --, --, --, --, 19.0, --],
-                 mask=[ True,  True,  True,  True,  True,  True, False,  True],
-           fill_value=1e+20), 'param_classifier:decision_tree:min_weight_fraction_leaf': masked_array(data=[--, --, --, --, --, --, 0.0, --],
-                 mask=[ True,  True,  True,  True,  True,  True, False,  True],
-           fill_value=1e+20), 'param_classifier:extra_trees:bootstrap': masked_array(data=[--, --, --, --, --, --, --, --],
-                 mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+                dtype='<U32'), 'param_classifier:decision_tree:min_impurity_decrease': masked_array(data=[--, --, --, --, --, --, 0.0, --, --],
+                 mask=[ True,  True,  True,  True,  True,  True, False,  True,
+                        True],
+           fill_value=1e+20), 'param_classifier:decision_tree:min_samples_leaf': masked_array(data=[--, --, --, --, --, --, 5.0, --, --],
+                 mask=[ True,  True,  True,  True,  True,  True, False,  True,
+                        True],
+           fill_value=1e+20), 'param_classifier:decision_tree:min_samples_split': masked_array(data=[--, --, --, --, --, --, 19.0, --, --],
+                 mask=[ True,  True,  True,  True,  True,  True, False,  True,
+                        True],
+           fill_value=1e+20), 'param_classifier:decision_tree:min_weight_fraction_leaf': masked_array(data=[--, --, --, --, --, --, 0.0, --, --],
+                 mask=[ True,  True,  True,  True,  True,  True, False,  True,
+                        True],
+           fill_value=1e+20), 'param_classifier:extra_trees:bootstrap': masked_array(data=[--, --, --, --, --, --, --, --, --],
+                 mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                        True],
            fill_value=1e+20,
-                dtype=float64), 'param_classifier:extra_trees:criterion': masked_array(data=[--, --, --, --, --, --, --, --],
-                 mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+                dtype=float64), 'param_classifier:extra_trees:criterion': masked_array(data=[--, --, --, --, --, --, --, --, --],
+                 mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                        True],
            fill_value=1e+20,
-                dtype=float64), 'param_classifier:extra_trees:max_depth': masked_array(data=[--, --, --, --, --, --, --, --],
-                 mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+                dtype=float64), 'param_classifier:extra_trees:max_depth': masked_array(data=[--, --, --, --, --, --, --, --, --],
+                 mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                        True],
            fill_value=1e+20,
-                dtype=float64), 'param_classifier:extra_trees:max_features': masked_array(data=[--, --, --, --, --, --, --, --],
-                 mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+                dtype=float64), 'param_classifier:extra_trees:max_features': masked_array(data=[--, --, --, --, --, --, --, --, --],
+                 mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                        True],
            fill_value=1e+20,
-                dtype=float64), 'param_classifier:extra_trees:max_leaf_nodes': masked_array(data=[--, --, --, --, --, --, --, --],
-                 mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+                dtype=float64), 'param_classifier:extra_trees:max_leaf_nodes': masked_array(data=[--, --, --, --, --, --, --, --, --],
+                 mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                        True],
            fill_value=1e+20,
-                dtype=float64), 'param_classifier:extra_trees:min_impurity_decrease': masked_array(data=[--, --, --, --, --, --, --, --],
-                 mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+                dtype=float64), 'param_classifier:extra_trees:min_impurity_decrease': masked_array(data=[--, --, --, --, --, --, --, --, --],
+                 mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                        True],
            fill_value=1e+20,
-                dtype=float64), 'param_classifier:extra_trees:min_samples_leaf': masked_array(data=[--, --, --, --, --, --, --, --],
-                 mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+                dtype=float64), 'param_classifier:extra_trees:min_samples_leaf': masked_array(data=[--, --, --, --, --, --, --, --, --],
+                 mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                        True],
            fill_value=1e+20,
-                dtype=float64), 'param_classifier:extra_trees:min_samples_split': masked_array(data=[--, --, --, --, --, --, --, --],
-                 mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+                dtype=float64), 'param_classifier:extra_trees:min_samples_split': masked_array(data=[--, --, --, --, --, --, --, --, --],
+                 mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                        True],
            fill_value=1e+20,
-                dtype=float64), 'param_classifier:extra_trees:min_weight_fraction_leaf': masked_array(data=[--, --, --, --, --, --, --, --],
-                 mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+                dtype=float64), 'param_classifier:extra_trees:min_weight_fraction_leaf': masked_array(data=[--, --, --, --, --, --, --, --, --],
+                 mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                        True],
            fill_value=1e+20,
-                dtype=float64), 'param_classifier:gradient_boosting:early_stop': masked_array(data=[--, --, --, --, --, --, --, --],
-                 mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+                dtype=float64), 'param_classifier:gradient_boosting:early_stop': masked_array(data=[--, --, --, --, --, --, --, --, --],
+                 mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                        True],
            fill_value=1e+20,
-                dtype=float64), 'param_classifier:gradient_boosting:l2_regularization': masked_array(data=[--, --, --, --, --, --, --, --],
-                 mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+                dtype=float64), 'param_classifier:gradient_boosting:l2_regularization': masked_array(data=[--, --, --, --, --, --, --, --, --],
+                 mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                        True],
            fill_value=1e+20,
-                dtype=float64), 'param_classifier:gradient_boosting:learning_rate': masked_array(data=[--, --, --, --, --, --, --, --],
-                 mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+                dtype=float64), 'param_classifier:gradient_boosting:learning_rate': masked_array(data=[--, --, --, --, --, --, --, --, --],
+                 mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                        True],
            fill_value=1e+20,
-                dtype=float64), 'param_classifier:gradient_boosting:loss': masked_array(data=[--, --, --, --, --, --, --, --],
-                 mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+                dtype=float64), 'param_classifier:gradient_boosting:loss': masked_array(data=[--, --, --, --, --, --, --, --, --],
+                 mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                        True],
            fill_value=1e+20,
-                dtype=float64), 'param_classifier:gradient_boosting:max_bins': masked_array(data=[--, --, --, --, --, --, --, --],
-                 mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+                dtype=float64), 'param_classifier:gradient_boosting:max_bins': masked_array(data=[--, --, --, --, --, --, --, --, --],
+                 mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                        True],
            fill_value=1e+20,
-                dtype=float64), 'param_classifier:gradient_boosting:max_depth': masked_array(data=[--, --, --, --, --, --, --, --],
-                 mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+                dtype=float64), 'param_classifier:gradient_boosting:max_depth': masked_array(data=[--, --, --, --, --, --, --, --, --],
+                 mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                        True],
            fill_value=1e+20,
-                dtype=float64), 'param_classifier:gradient_boosting:max_leaf_nodes': masked_array(data=[--, --, --, --, --, --, --, --],
-                 mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+                dtype=float64), 'param_classifier:gradient_boosting:max_leaf_nodes': masked_array(data=[--, --, --, --, --, --, --, --, --],
+                 mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                        True],
            fill_value=1e+20,
-                dtype=float64), 'param_classifier:gradient_boosting:min_samples_leaf': masked_array(data=[--, --, --, --, --, --, --, --],
-                 mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+                dtype=float64), 'param_classifier:gradient_boosting:min_samples_leaf': masked_array(data=[--, --, --, --, --, --, --, --, --],
+                 mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                        True],
            fill_value=1e+20,
-                dtype=float64), 'param_classifier:gradient_boosting:scoring': masked_array(data=[--, --, --, --, --, --, --, --],
-                 mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+                dtype=float64), 'param_classifier:gradient_boosting:scoring': masked_array(data=[--, --, --, --, --, --, --, --, --],
+                 mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                        True],
            fill_value=1e+20,
-                dtype=float64), 'param_classifier:gradient_boosting:tol': masked_array(data=[--, --, --, --, --, --, --, --],
-                 mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+                dtype=float64), 'param_classifier:gradient_boosting:tol': masked_array(data=[--, --, --, --, --, --, --, --, --],
+                 mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                        True],
            fill_value=1e+20,
-                dtype=float64), 'param_classifier:k_nearest_neighbors:n_neighbors': masked_array(data=[--, --, --, 4.0, --, --, --, --],
-                 mask=[ True,  True,  True, False,  True,  True,  True,  True],
-           fill_value=1e+20), 'param_classifier:k_nearest_neighbors:p': masked_array(data=[--, --, --, 2.0, --, --, --, --],
-                 mask=[ True,  True,  True, False,  True,  True,  True,  True],
-           fill_value=1e+20), 'param_classifier:k_nearest_neighbors:weights': masked_array(data=[--, --, --, 'distance', --, --, --, --],
-                 mask=[ True,  True,  True, False,  True,  True,  True,  True],
+                dtype=float64), 'param_classifier:k_nearest_neighbors:n_neighbors': masked_array(data=[--, --, --, 4.0, --, --, --, --, --],
+                 mask=[ True,  True,  True, False,  True,  True,  True,  True,
+                        True],
+           fill_value=1e+20), 'param_classifier:k_nearest_neighbors:p': masked_array(data=[--, --, --, 2.0, --, --, --, --, --],
+                 mask=[ True,  True,  True, False,  True,  True,  True,  True,
+                        True],
+           fill_value=1e+20), 'param_classifier:k_nearest_neighbors:weights': masked_array(data=[--, --, --, 'distance', --, --, --, --, --],
+                 mask=[ True,  True,  True, False,  True,  True,  True,  True,
+                        True],
            fill_value='N/A',
-                dtype='<U32'), 'param_classifier:lda:shrinkage': masked_array(data=[--, --, --, --, --, --, --, --],
-                 mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+                dtype='<U32'), 'param_classifier:lda:shrinkage': masked_array(data=[--, --, --, --, --, --, --, --, --],
+                 mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                        True],
            fill_value=1e+20,
-                dtype=float64), 'param_classifier:lda:tol': masked_array(data=[--, --, --, --, --, --, --, --],
-                 mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+                dtype=float64), 'param_classifier:lda:tol': masked_array(data=[--, --, --, --, --, --, --, --, --],
+                 mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                        True],
            fill_value=1e+20,
-                dtype=float64), 'param_classifier:liblinear_svc:C': masked_array(data=[--, --, --, --, --, 10.369811497206404, --, --],
-                 mask=[ True,  True,  True,  True,  True, False,  True,  True],
-           fill_value=1e+20), 'param_classifier:liblinear_svc:dual': masked_array(data=[--, --, --, --, --, 'False', --, --],
-                 mask=[ True,  True,  True,  True,  True, False,  True,  True],
+                dtype=float64), 'param_classifier:liblinear_svc:C': masked_array(data=[--, --, --, --, --, 10.369811497206404, --, --, --],
+                 mask=[ True,  True,  True,  True,  True, False,  True,  True,
+                        True],
+           fill_value=1e+20), 'param_classifier:liblinear_svc:dual': masked_array(data=[--, --, --, --, --, 'False', --, --, --],
+                 mask=[ True,  True,  True,  True,  True, False,  True,  True,
+                        True],
            fill_value='N/A',
-                dtype='<U32'), 'param_classifier:liblinear_svc:fit_intercept': masked_array(data=[--, --, --, --, --, 'True', --, --],
-                 mask=[ True,  True,  True,  True,  True, False,  True,  True],
+                dtype='<U32'), 'param_classifier:liblinear_svc:fit_intercept': masked_array(data=[--, --, --, --, --, 'True', --, --, --],
+                 mask=[ True,  True,  True,  True,  True, False,  True,  True,
+                        True],
            fill_value='N/A',
-                dtype='<U32'), 'param_classifier:liblinear_svc:intercept_scaling': masked_array(data=[--, --, --, --, --, 1.0, --, --],
-                 mask=[ True,  True,  True,  True,  True, False,  True,  True],
-           fill_value=1e+20), 'param_classifier:liblinear_svc:loss': masked_array(data=[--, --, --, --, --, 'squared_hinge', --, --],
-                 mask=[ True,  True,  True,  True,  True, False,  True,  True],
+                dtype='<U32'), 'param_classifier:liblinear_svc:intercept_scaling': masked_array(data=[--, --, --, --, --, 1.0, --, --, --],
+                 mask=[ True,  True,  True,  True,  True, False,  True,  True,
+                        True],
+           fill_value=1e+20), 'param_classifier:liblinear_svc:loss': masked_array(data=[--, --, --, --, --, 'squared_hinge', --, --, --],
+                 mask=[ True,  True,  True,  True,  True, False,  True,  True,
+                        True],
            fill_value='N/A',
-                dtype='<U32'), 'param_classifier:liblinear_svc:multi_class': masked_array(data=[--, --, --, --, --, 'ovr', --, --],
-                 mask=[ True,  True,  True,  True,  True, False,  True,  True],
+                dtype='<U32'), 'param_classifier:liblinear_svc:multi_class': masked_array(data=[--, --, --, --, --, 'ovr', --, --, --],
+                 mask=[ True,  True,  True,  True,  True, False,  True,  True,
+                        True],
            fill_value='N/A',
-                dtype='<U32'), 'param_classifier:liblinear_svc:penalty': masked_array(data=[--, --, --, --, --, 'l2', --, --],
-                 mask=[ True,  True,  True,  True,  True, False,  True,  True],
+                dtype='<U32'), 'param_classifier:liblinear_svc:penalty': masked_array(data=[--, --, --, --, --, 'l2', --, --, --],
+                 mask=[ True,  True,  True,  True,  True, False,  True,  True,
+                        True],
            fill_value='N/A',
-                dtype='<U32'), 'param_classifier:liblinear_svc:tol': masked_array(data=[--, --, --, --, --, 0.0015130257264171173, --, --],
-                 mask=[ True,  True,  True,  True,  True, False,  True,  True],
-           fill_value=1e+20), 'param_classifier:libsvm_svc:C': masked_array(data=[--, --, --, --, 100.5905006626969, --, --, --],
-                 mask=[ True,  True,  True,  True, False,  True,  True,  True],
-           fill_value=1e+20), 'param_classifier:libsvm_svc:gamma': masked_array(data=[--, --, --, --, 0.011333066835975528, --, --, --],
-                 mask=[ True,  True,  True,  True, False,  True,  True,  True],
-           fill_value=1e+20), 'param_classifier:libsvm_svc:kernel': masked_array(data=[--, --, --, --, 'poly', --, --, --],
-                 mask=[ True,  True,  True,  True, False,  True,  True,  True],
+                dtype='<U32'), 'param_classifier:liblinear_svc:tol': masked_array(data=[--, --, --, --, --, 0.0015130257264171173, --, --, --],
+                 mask=[ True,  True,  True,  True,  True, False,  True,  True,
+                        True],
+           fill_value=1e+20), 'param_classifier:libsvm_svc:C': masked_array(data=[--, --, --, --, 100.5905006626969, --, --, --, --],
+                 mask=[ True,  True,  True,  True, False,  True,  True,  True,
+                        True],
+           fill_value=1e+20), 'param_classifier:libsvm_svc:gamma': masked_array(data=[--, --, --, --, 0.011333066835975528, --, --, --, --],
+                 mask=[ True,  True,  True,  True, False,  True,  True,  True,
+                        True],
+           fill_value=1e+20), 'param_classifier:libsvm_svc:kernel': masked_array(data=[--, --, --, --, 'poly', --, --, --, --],
+                 mask=[ True,  True,  True,  True, False,  True,  True,  True,
+                        True],
            fill_value='N/A',
-                dtype='<U32'), 'param_classifier:libsvm_svc:max_iter': masked_array(data=[--, --, --, --, -1.0, --, --, --],
-                 mask=[ True,  True,  True,  True, False,  True,  True,  True],
-           fill_value=1e+20), 'param_classifier:libsvm_svc:shrinking': masked_array(data=[--, --, --, --, 'True', --, --, --],
-                 mask=[ True,  True,  True,  True, False,  True,  True,  True],
+                dtype='<U32'), 'param_classifier:libsvm_svc:max_iter': masked_array(data=[--, --, --, --, -1.0, --, --, --, --],
+                 mask=[ True,  True,  True,  True, False,  True,  True,  True,
+                        True],
+           fill_value=1e+20), 'param_classifier:libsvm_svc:shrinking': masked_array(data=[--, --, --, --, 'True', --, --, --, --],
+                 mask=[ True,  True,  True,  True, False,  True,  True,  True,
+                        True],
            fill_value='N/A',
-                dtype='<U32'), 'param_classifier:libsvm_svc:tol': masked_array(data=[--, --, --, --, 0.012391313886912093, --, --, --],
-                 mask=[ True,  True,  True,  True, False,  True,  True,  True],
-           fill_value=1e+20), 'param_classifier:mlp:activation': masked_array(data=[--, --, 'tanh', --, --, --, --, --],
-                 mask=[ True,  True, False,  True,  True,  True,  True,  True],
+                dtype='<U32'), 'param_classifier:libsvm_svc:tol': masked_array(data=[--, --, --, --, 0.012391313886912093, --, --, --, --],
+                 mask=[ True,  True,  True,  True, False,  True,  True,  True,
+                        True],
+           fill_value=1e+20), 'param_classifier:mlp:activation': masked_array(data=[--, --, 'tanh', --, --, --, --, --, --],
+                 mask=[ True,  True, False,  True,  True,  True,  True,  True,
+                        True],
            fill_value='N/A',
-                dtype='<U32'), 'param_classifier:mlp:alpha': masked_array(data=[--, --, 1.103855734598575e-05, --, --, --, --, --],
-                 mask=[ True,  True, False,  True,  True,  True,  True,  True],
-           fill_value=1e+20), 'param_classifier:mlp:batch_size': masked_array(data=[--, --, 'auto', --, --, --, --, --],
-                 mask=[ True,  True, False,  True,  True,  True,  True,  True],
+                dtype='<U32'), 'param_classifier:mlp:alpha': masked_array(data=[--, --, 1.103855734598575e-05, --, --, --, --, --, --],
+                 mask=[ True,  True, False,  True,  True,  True,  True,  True,
+                        True],
+           fill_value=1e+20), 'param_classifier:mlp:batch_size': masked_array(data=[--, --, 'auto', --, --, --, --, --, --],
+                 mask=[ True,  True, False,  True,  True,  True,  True,  True,
+                        True],
            fill_value='N/A',
-                dtype='<U32'), 'param_classifier:mlp:beta_1': masked_array(data=[--, --, 0.9, --, --, --, --, --],
-                 mask=[ True,  True, False,  True,  True,  True,  True,  True],
-           fill_value=1e+20), 'param_classifier:mlp:beta_2': masked_array(data=[--, --, 0.999, --, --, --, --, --],
-                 mask=[ True,  True, False,  True,  True,  True,  True,  True],
-           fill_value=1e+20), 'param_classifier:mlp:early_stopping': masked_array(data=[--, --, 'valid', --, --, --, --, --],
-                 mask=[ True,  True, False,  True,  True,  True,  True,  True],
+                dtype='<U32'), 'param_classifier:mlp:beta_1': masked_array(data=[--, --, 0.9, --, --, --, --, --, --],
+                 mask=[ True,  True, False,  True,  True,  True,  True,  True,
+                        True],
+           fill_value=1e+20), 'param_classifier:mlp:beta_2': masked_array(data=[--, --, 0.999, --, --, --, --, --, --],
+                 mask=[ True,  True, False,  True,  True,  True,  True,  True,
+                        True],
+           fill_value=1e+20), 'param_classifier:mlp:early_stopping': masked_array(data=[--, --, 'valid', --, --, --, --, --, --],
+                 mask=[ True,  True, False,  True,  True,  True,  True,  True,
+                        True],
            fill_value='N/A',
-                dtype='<U32'), 'param_classifier:mlp:epsilon': masked_array(data=[--, --, 1e-08, --, --, --, --, --],
-                 mask=[ True,  True, False,  True,  True,  True,  True,  True],
-           fill_value=1e+20), 'param_classifier:mlp:hidden_layer_depth': masked_array(data=[--, --, 3.0, --, --, --, --, --],
-                 mask=[ True,  True, False,  True,  True,  True,  True,  True],
-           fill_value=1e+20), 'param_classifier:mlp:learning_rate_init': masked_array(data=[--, --, 0.00014375616988222174, --, --, --, --, --],
-                 mask=[ True,  True, False,  True,  True,  True,  True,  True],
-           fill_value=1e+20), 'param_classifier:mlp:n_iter_no_change': masked_array(data=[--, --, 32.0, --, --, --, --, --],
-                 mask=[ True,  True, False,  True,  True,  True,  True,  True],
-           fill_value=1e+20), 'param_classifier:mlp:num_nodes_per_layer': masked_array(data=[--, --, 229.0, --, --, --, --, --],
-                 mask=[ True,  True, False,  True,  True,  True,  True,  True],
-           fill_value=1e+20), 'param_classifier:mlp:shuffle': masked_array(data=[--, --, 'True', --, --, --, --, --],
-                 mask=[ True,  True, False,  True,  True,  True,  True,  True],
+                dtype='<U32'), 'param_classifier:mlp:epsilon': masked_array(data=[--, --, 1e-08, --, --, --, --, --, --],
+                 mask=[ True,  True, False,  True,  True,  True,  True,  True,
+                        True],
+           fill_value=1e+20), 'param_classifier:mlp:hidden_layer_depth': masked_array(data=[--, --, 3.0, --, --, --, --, --, --],
+                 mask=[ True,  True, False,  True,  True,  True,  True,  True,
+                        True],
+           fill_value=1e+20), 'param_classifier:mlp:learning_rate_init': masked_array(data=[--, --, 0.00014375616988222174, --, --, --, --, --, --],
+                 mask=[ True,  True, False,  True,  True,  True,  True,  True,
+                        True],
+           fill_value=1e+20), 'param_classifier:mlp:n_iter_no_change': masked_array(data=[--, --, 32.0, --, --, --, --, --, --],
+                 mask=[ True,  True, False,  True,  True,  True,  True,  True,
+                        True],
+           fill_value=1e+20), 'param_classifier:mlp:num_nodes_per_layer': masked_array(data=[--, --, 229.0, --, --, --, --, --, --],
+                 mask=[ True,  True, False,  True,  True,  True,  True,  True,
+                        True],
+           fill_value=1e+20), 'param_classifier:mlp:shuffle': masked_array(data=[--, --, 'True', --, --, --, --, --, --],
+                 mask=[ True,  True, False,  True,  True,  True,  True,  True,
+                        True],
            fill_value='N/A',
-                dtype='<U32'), 'param_classifier:mlp:solver': masked_array(data=[--, --, 'adam', --, --, --, --, --],
-                 mask=[ True,  True, False,  True,  True,  True,  True,  True],
+                dtype='<U32'), 'param_classifier:mlp:solver': masked_array(data=[--, --, 'adam', --, --, --, --, --, --],
+                 mask=[ True,  True, False,  True,  True,  True,  True,  True,
+                        True],
            fill_value='N/A',
-                dtype='<U32'), 'param_classifier:mlp:tol': masked_array(data=[--, --, 0.0001, --, --, --, --, --],
-                 mask=[ True,  True, False,  True,  True,  True,  True,  True],
-           fill_value=1e+20), 'param_classifier:passive_aggressive:C': masked_array(data=[--, --, --, --, --, --, --, --],
-                 mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+                dtype='<U32'), 'param_classifier:mlp:tol': masked_array(data=[--, --, 0.0001, --, --, --, --, --, --],
+                 mask=[ True,  True, False,  True,  True,  True,  True,  True,
+                        True],
+           fill_value=1e+20), 'param_classifier:passive_aggressive:C': masked_array(data=[--, --, --, --, --, --, --, --, --],
+                 mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                        True],
            fill_value=1e+20,
-                dtype=float64), 'param_classifier:passive_aggressive:average': masked_array(data=[--, --, --, --, --, --, --, --],
-                 mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+                dtype=float64), 'param_classifier:passive_aggressive:average': masked_array(data=[--, --, --, --, --, --, --, --, --],
+                 mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                        True],
            fill_value=1e+20,
-                dtype=float64), 'param_classifier:passive_aggressive:fit_intercept': masked_array(data=[--, --, --, --, --, --, --, --],
-                 mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+                dtype=float64), 'param_classifier:passive_aggressive:fit_intercept': masked_array(data=[--, --, --, --, --, --, --, --, --],
+                 mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                        True],
            fill_value=1e+20,
-                dtype=float64), 'param_classifier:passive_aggressive:loss': masked_array(data=[--, --, --, --, --, --, --, --],
-                 mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+                dtype=float64), 'param_classifier:passive_aggressive:loss': masked_array(data=[--, --, --, --, --, --, --, --, --],
+                 mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                        True],
            fill_value=1e+20,
-                dtype=float64), 'param_classifier:passive_aggressive:tol': masked_array(data=[--, --, --, --, --, --, --, --],
-                 mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+                dtype=float64), 'param_classifier:passive_aggressive:tol': masked_array(data=[--, --, --, --, --, --, --, --, --],
+                 mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                        True],
            fill_value=1e+20,
-                dtype=float64), 'param_classifier:qda:reg_param': masked_array(data=[--, --, --, --, --, --, --, --],
-                 mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+                dtype=float64), 'param_classifier:qda:reg_param': masked_array(data=[--, --, --, --, --, --, --, --, --],
+                 mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                        True],
            fill_value=1e+20,
-                dtype=float64), 'param_classifier:random_forest:bootstrap': masked_array(data=['True', 'True', --, --, --, --, --, --],
-                 mask=[False, False,  True,  True,  True,  True,  True,  True],
+                dtype=float64), 'param_classifier:random_forest:bootstrap': masked_array(data=['True', 'True', --, --, --, --, --, --, --],
+                 mask=[False, False,  True,  True,  True,  True,  True,  True,
+                        True],
            fill_value='N/A',
-                dtype='<U32'), 'param_classifier:random_forest:criterion': masked_array(data=['gini', 'gini', --, --, --, --, --, --],
-                 mask=[False, False,  True,  True,  True,  True,  True,  True],
+                dtype='<U32'), 'param_classifier:random_forest:criterion': masked_array(data=['gini', 'gini', --, --, --, --, --, --, --],
+                 mask=[False, False,  True,  True,  True,  True,  True,  True,
+                        True],
            fill_value='N/A',
-                dtype='<U32'), 'param_classifier:random_forest:max_depth': masked_array(data=['None', 'None', --, --, --, --, --, --],
-                 mask=[False, False,  True,  True,  True,  True,  True,  True],
+                dtype='<U32'), 'param_classifier:random_forest:max_depth': masked_array(data=['None', 'None', --, --, --, --, --, --, --],
+                 mask=[False, False,  True,  True,  True,  True,  True,  True,
+                        True],
            fill_value='N/A',
-                dtype='<U32'), 'param_classifier:random_forest:max_features': masked_array(data=[0.5, 0.9331254454871041, --, --, --, --, --, --],
-                 mask=[False, False,  True,  True,  True,  True,  True,  True],
-           fill_value=1e+20), 'param_classifier:random_forest:max_leaf_nodes': masked_array(data=['None', 'None', --, --, --, --, --, --],
-                 mask=[False, False,  True,  True,  True,  True,  True,  True],
+                dtype='<U32'), 'param_classifier:random_forest:max_features': masked_array(data=[0.5, 0.9331254454871041, --, --, --, --, --, --, --],
+                 mask=[False, False,  True,  True,  True,  True,  True,  True,
+                        True],
+           fill_value=1e+20), 'param_classifier:random_forest:max_leaf_nodes': masked_array(data=['None', 'None', --, --, --, --, --, --, --],
+                 mask=[False, False,  True,  True,  True,  True,  True,  True,
+                        True],
            fill_value='N/A',
-                dtype='<U32'), 'param_classifier:random_forest:min_impurity_decrease': masked_array(data=[0.0, 0.0, --, --, --, --, --, --],
-                 mask=[False, False,  True,  True,  True,  True,  True,  True],
-           fill_value=1e+20), 'param_classifier:random_forest:min_samples_leaf': masked_array(data=[1.0, 2.0, --, --, --, --, --, --],
-                 mask=[False, False,  True,  True,  True,  True,  True,  True],
-           fill_value=1e+20), 'param_classifier:random_forest:min_samples_split': masked_array(data=[2.0, 20.0, --, --, --, --, --, --],
-                 mask=[False, False,  True,  True,  True,  True,  True,  True],
-           fill_value=1e+20), 'param_classifier:random_forest:min_weight_fraction_leaf': masked_array(data=[0.0, 0.0, --, --, --, --, --, --],
-                 mask=[False, False,  True,  True,  True,  True,  True,  True],
-           fill_value=1e+20), 'param_classifier:sgd:alpha': masked_array(data=[--, --, --, --, --, --, --, 1.0333981795670597e-07],
-                 mask=[ True,  True,  True,  True,  True,  True,  True, False],
-           fill_value=1e+20), 'param_classifier:sgd:average': masked_array(data=[--, --, --, --, --, --, --, 'False'],
-                 mask=[ True,  True,  True,  True,  True,  True,  True, False],
+                dtype='<U32'), 'param_classifier:random_forest:min_impurity_decrease': masked_array(data=[0.0, 0.0, --, --, --, --, --, --, --],
+                 mask=[False, False,  True,  True,  True,  True,  True,  True,
+                        True],
+           fill_value=1e+20), 'param_classifier:random_forest:min_samples_leaf': masked_array(data=[1.0, 2.0, --, --, --, --, --, --, --],
+                 mask=[False, False,  True,  True,  True,  True,  True,  True,
+                        True],
+           fill_value=1e+20), 'param_classifier:random_forest:min_samples_split': masked_array(data=[2.0, 20.0, --, --, --, --, --, --, --],
+                 mask=[False, False,  True,  True,  True,  True,  True,  True,
+                        True],
+           fill_value=1e+20), 'param_classifier:random_forest:min_weight_fraction_leaf': masked_array(data=[0.0, 0.0, --, --, --, --, --, --, --],
+                 mask=[False, False,  True,  True,  True,  True,  True,  True,
+                        True],
+           fill_value=1e+20), 'param_classifier:sgd:alpha': masked_array(data=[--, --, --, --, --, --, --, 1.0333981795670635e-07,
+                       0.0006046872822704108],
+                 mask=[ True,  True,  True,  True,  True,  True,  True, False,
+                       False],
+           fill_value=1e+20), 'param_classifier:sgd:average': masked_array(data=[--, --, --, --, --, --, --, 'False', 'False'],
+                 mask=[ True,  True,  True,  True,  True,  True,  True, False,
+                       False],
            fill_value='N/A',
-                dtype='<U32'), 'param_classifier:sgd:fit_intercept': masked_array(data=[--, --, --, --, --, --, --, 'True'],
-                 mask=[ True,  True,  True,  True,  True,  True,  True, False],
+                dtype='<U32'), 'param_classifier:sgd:fit_intercept': masked_array(data=[--, --, --, --, --, --, --, 'True', 'True'],
+                 mask=[ True,  True,  True,  True,  True,  True,  True, False,
+                       False],
            fill_value='N/A',
-                dtype='<U32'), 'param_classifier:sgd:learning_rate': masked_array(data=[--, --, --, --, --, --, --, 'constant'],
-                 mask=[ True,  True,  True,  True,  True,  True,  True, False],
+                dtype='<U32'), 'param_classifier:sgd:learning_rate': masked_array(data=[--, --, --, --, --, --, --, 'constant', 'invscaling'],
+                 mask=[ True,  True,  True,  True,  True,  True,  True, False,
+                       False],
            fill_value='N/A',
-                dtype='<U32'), 'param_classifier:sgd:loss': masked_array(data=[--, --, --, --, --, --, --, 'modified_huber'],
-                 mask=[ True,  True,  True,  True,  True,  True,  True, False],
+                dtype='<U32'), 'param_classifier:sgd:loss': masked_array(data=[--, --, --, --, --, --, --, 'modified_huber',
+                       'modified_huber'],
+                 mask=[ True,  True,  True,  True,  True,  True,  True, False,
+                       False],
            fill_value='N/A',
-                dtype='<U32'), 'param_classifier:sgd:penalty': masked_array(data=[--, --, --, --, --, --, --, 'l2'],
-                 mask=[ True,  True,  True,  True,  True,  True,  True, False],
+                dtype='<U32'), 'param_classifier:sgd:penalty': masked_array(data=[--, --, --, --, --, --, --, 'l2', 'elasticnet'],
+                 mask=[ True,  True,  True,  True,  True,  True,  True, False,
+                       False],
            fill_value='N/A',
-                dtype='<U32'), 'param_classifier:sgd:tol': masked_array(data=[--, --, --, --, --, --, --, 2.5846457091661748e-05],
-                 mask=[ True,  True,  True,  True,  True,  True,  True, False],
+                dtype='<U32'), 'param_classifier:sgd:tol': masked_array(data=[--, --, --, --, --, --, --, 2.5846457091661748e-05,
+                       1.7244397014112777e-05],
+                 mask=[ True,  True,  True,  True,  True,  True,  True, False,
+                       False],
            fill_value=1e+20), 'param_data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': masked_array(data=['one_hot_encoding', 'no_encoding', 'no_encoding',
                        'no_encoding', 'one_hot_encoding', 'no_encoding',
-                       'no_encoding', 'encoding'],
-                 mask=[False, False, False, False, False, False, False, False],
+                       'no_encoding', 'encoding', 'no_encoding'],
+                 mask=[False, False, False, False, False, False, False, False,
+                       False],
            fill_value='N/A',
                 dtype='<U16'), 'param_data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': masked_array(data=['minority_coalescer', 'minority_coalescer',
                        'minority_coalescer', 'minority_coalescer',
                        'minority_coalescer', 'no_coalescense',
-                       'no_coalescense', 'minority_coalescer'],
-                 mask=[False, False, False, False, False, False, False, False],
+                       'no_coalescense', 'minority_coalescer',
+                       'no_coalescense'],
+                 mask=[False, False, False, False, False, False, False, False,
+                       False],
            fill_value='N/A',
                 dtype='<U18'), 'param_data_preprocessor:feature_type:numerical_transformer:imputation:strategy': masked_array(data=['mean', 'mean', 'most_frequent', 'mean', 'mean',
-                       'mean', 'mean', 'most_frequent'],
-                 mask=[False, False, False, False, False, False, False, False],
+                       'mean', 'mean', 'most_frequent', 'mean'],
+                 mask=[False, False, False, False, False, False, False, False,
+                       False],
            fill_value='N/A',
                 dtype='<U13'), 'param_data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': masked_array(data=['standardize', 'none', 'quantile_transformer',
                        'normalize', 'minmax', 'power_transformer',
-                       'quantile_transformer', 'minmax'],
-                 mask=[False, False, False, False, False, False, False, False],
+                       'quantile_transformer', 'minmax', 'power_transformer'],
+                 mask=[False, False, False, False, False, False, False, False,
+                       False],
            fill_value='N/A',
                 dtype='<U20'), 'param_feature_preprocessor:pca:keep_variance': masked_array(data=[0.9999, 0.9967857433838874, 0.7895711479212801,
                        0.8047274080856589, 0.9290439925152777,
                        0.6661824659281315, 0.7772788812704434,
-                       0.9755563101648113],
-                 mask=[False, False, False, False, False, False, False, False],
+                       0.9755563101648113, 0.9941379232526211],
+                 mask=[False, False, False, False, False, False, False, False,
+                       False],
            fill_value=1e+20), 'param_feature_preprocessor:pca:whiten': masked_array(data=['False', 'False', 'True', 'False', 'False', 'False',
-                       'True', 'True'],
-                 mask=[False, False, False, False, False, False, False, False],
+                       'True', 'True', 'False'],
+                 mask=[False, False, False, False, False, False, False, False,
+                       False],
            fill_value='N/A',
-                dtype='<U5'), 'param_classifier:gradient_boosting:n_iter_no_change': masked_array(data=[--, --, --, --, --, --, --, --],
-                 mask=[ True,  True,  True,  True,  True,  True,  True,  True],
-           fill_value=1e+20,
-                dtype=float64), 'param_classifier:gradient_boosting:validation_fraction': masked_array(data=[--, --, --, --, --, --, --, --],
-                 mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+                dtype='<U5'), 'param_classifier:gradient_boosting:n_iter_no_change': masked_array(data=[--, --, --, --, --, --, --, --, --],
+                 mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                        True],
            fill_value=1e+20,
-                dtype=float64), 'param_classifier:lda:shrinkage_factor': masked_array(data=[--, --, --, --, --, --, --, --],
-                 mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+                dtype=float64), 'param_classifier:gradient_boosting:validation_fraction': masked_array(data=[--, --, --, --, --, --, --, --, --],
+                 mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                        True],
            fill_value=1e+20,
-                dtype=float64), 'param_classifier:libsvm_svc:coef0': masked_array(data=[--, --, --, --, 0.08087614244138486, --, --, --],
-                 mask=[ True,  True,  True,  True, False,  True,  True,  True],
-           fill_value=1e+20), 'param_classifier:libsvm_svc:degree': masked_array(data=[--, --, --, --, 3.0, --, --, --],
-                 mask=[ True,  True,  True,  True, False,  True,  True,  True],
-           fill_value=1e+20), 'param_classifier:mlp:validation_fraction': masked_array(data=[--, --, 0.1, --, --, --, --, --],
-                 mask=[ True,  True, False,  True,  True,  True,  True,  True],
-           fill_value=1e+20), 'param_classifier:sgd:epsilon': masked_array(data=[--, --, --, --, --, --, --, 0.03457277874230573],
-                 mask=[ True,  True,  True,  True,  True,  True,  True, False],
-           fill_value=1e+20), 'param_classifier:sgd:eta0': masked_array(data=[--, --, --, --, --, --, --, 0.004635634934629576],
-                 mask=[ True,  True,  True,  True,  True,  True,  True, False],
-           fill_value=1e+20), 'param_classifier:sgd:l1_ratio': masked_array(data=[--, --, --, --, --, --, --, --],
-                 mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+                dtype=float64), 'param_classifier:lda:shrinkage_factor': masked_array(data=[--, --, --, --, --, --, --, --, --],
+                 mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                        True],
            fill_value=1e+20,
-                dtype=float64), 'param_classifier:sgd:power_t': masked_array(data=[--, --, --, --, --, --, --, --],
-                 mask=[ True,  True,  True,  True,  True,  True,  True,  True],
-           fill_value=1e+20,
-                dtype=float64), 'param_data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': masked_array(data=[0.01, 0.005069923784737444, 0.002766772136115771,
+                dtype=float64), 'param_classifier:libsvm_svc:coef0': masked_array(data=[--, --, --, --, 0.08087614244138486, --, --, --, --],
+                 mask=[ True,  True,  True,  True, False,  True,  True,  True,
+                        True],
+           fill_value=1e+20), 'param_classifier:libsvm_svc:degree': masked_array(data=[--, --, --, --, 3.0, --, --, --, --],
+                 mask=[ True,  True,  True,  True, False,  True,  True,  True,
+                        True],
+           fill_value=1e+20), 'param_classifier:mlp:validation_fraction': masked_array(data=[--, --, 0.1, --, --, --, --, --, --],
+                 mask=[ True,  True, False,  True,  True,  True,  True,  True,
+                        True],
+           fill_value=1e+20), 'param_classifier:sgd:epsilon': masked_array(data=[--, --, --, --, --, --, --, 0.03457277874230573,
+                       0.0006085470273398056],
+                 mask=[ True,  True,  True,  True,  True,  True,  True, False,
+                       False],
+           fill_value=1e+20), 'param_classifier:sgd:eta0': masked_array(data=[--, --, --, --, --, --, --, 0.004635634934629583,
+                       0.0003474820560182061],
+                 mask=[ True,  True,  True,  True,  True,  True,  True, False,
+                       False],
+           fill_value=1e+20), 'param_classifier:sgd:l1_ratio': masked_array(data=[--, --, --, --, --, --, --, --, 4.5783995454757295e-09],
+                 mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                       False],
+           fill_value=1e+20), 'param_classifier:sgd:power_t': masked_array(data=[--, --, --, --, --, --, --, --, 0.6064541566557635],
+                 mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                       False],
+           fill_value=1e+20), 'param_data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': masked_array(data=[0.01, 0.005069923784737444, 0.002766772136115771,
                        0.008015420020402715, 0.010000000000000004, --, --,
-                       0.00015788887334393744],
-                 mask=[False, False, False, False, False,  True,  True, False],
-           fill_value=1e+20), 'param_data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles': masked_array(data=[--, --, 180.0, --, --, --, 526.0, --],
-                 mask=[ True,  True, False,  True,  True,  True, False,  True],
-           fill_value=1e+20), 'param_data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution': masked_array(data=[--, --, 'uniform', --, --, --, 'uniform', --],
-                 mask=[ True,  True, False,  True,  True,  True, False,  True],
+                       0.00015788887334393714, --],
+                 mask=[False, False, False, False, False,  True,  True, False,
+                        True],
+           fill_value=1e+20), 'param_data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles': masked_array(data=[--, --, 180.0, --, --, --, 526.0, --, --],
+                 mask=[ True,  True, False,  True,  True,  True, False,  True,
+                        True],
+           fill_value=1e+20), 'param_data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution': masked_array(data=[--, --, 'uniform', --, --, --, 'uniform', --, --],
+                 mask=[ True,  True, False,  True,  True,  True, False,  True,
+                        True],
            fill_value='N/A',
-                dtype='<U32'), 'param_data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max': masked_array(data=[--, --, --, --, --, --, --, --],
-                 mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+                dtype='<U32'), 'param_data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max': masked_array(data=[--, --, --, --, --, --, --, --, --],
+                 mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                        True],
            fill_value=1e+20,
-                dtype=float64), 'param_data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min': masked_array(data=[--, --, --, --, --, --, --, --],
-                 mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+                dtype=float64), 'param_data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min': masked_array(data=[--, --, --, --, --, --, --, --, --],
+                 mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                        True],
            fill_value=1e+20,
                 dtype=float64)}
 
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 175-180
+.. GENERATED FROM PYTHON SOURCE LINES 184-189
 
 Inspect the components of the best model
 ========================================
@@ -968,7 +1136,7 @@ Inspect the components of the best model
 Iterate over the components of the model and print
 The explained variance ratio per stage
 
-.. GENERATED FROM PYTHON SOURCE LINES 180-193
+.. GENERATED FROM PYTHON SOURCE LINES 189-202
 
 .. code-block:: default
 
@@ -998,20 +1166,23 @@ The explained variance ratio per stage
     The 0th pipeline has a explained variance of [0.49503611 0.16649281 0.09111888 0.07213284 0.04865917 0.03208923
      0.01851537 0.01223987 0.0104432  0.01006215 0.00894849 0.0076293
      0.00684163]
-    The 1th pipeline has a explained variance of [0.45954164 0.18010277 0.09814953 0.06334655]
+    The 1th pipeline has a explained variance of [0.98080571 0.01684553]
     The 2th pipeline has a explained variance of [0.44685232 0.17122959 0.1085501 ]
-    The 3th pipeline has a explained variance of [0.98080571 0.01684553]
-    The 4th pipeline has a explained variance of [4.32956881e-01 1.79057296e-01 1.11737571e-01 6.80724345e-02
+    The 3th pipeline has a explained variance of [0.49503611 0.16649281 0.09111888 0.07213284 0.04865917 0.03208923
+     0.01851537 0.01223987]
+    The 4th pipeline has a explained variance of [0.45954164 0.18010277 0.09814953 0.06334655]
+    The 5th pipeline has a explained variance of [0.44685232 0.17122959 0.1085501  0.07063243 0.06437417 0.03986668
+     0.02462169 0.01589975 0.01257378 0.00948519 0.00858029 0.0067563
+     0.0059319  0.00431052 0.0030875  0.00172583]
+    The 6th pipeline has a explained variance of [4.32956881e-01 1.79057296e-01 1.11737571e-01 6.80724345e-02
      5.94611519e-02 3.70629898e-02 2.38430977e-02 1.49326086e-02
      1.37641366e-02 1.13704890e-02 1.03737258e-02 8.74116751e-03
      7.57629717e-03 4.86528503e-03 3.32225143e-03 2.55773043e-03
      2.20759805e-03 1.88675402e-03 1.36245140e-03 1.03409213e-03
      8.39749085e-04 7.91287172e-04 6.75655689e-04 5.42961621e-04
      5.02641737e-04 2.07827509e-04 1.74597367e-04]
-    The 5th pipeline has a explained variance of [0.4595393  0.18012072 0.09809101 0.06332899]
-    The 6th pipeline has a explained variance of [0.76699224 0.17152095]
-    The 7th pipeline has a explained variance of [0.49503611 0.16649281 0.09111888 0.07213284 0.04865917 0.03208923
-     0.01851537 0.01223987]
+    The 7th pipeline has a explained variance of [0.76699224 0.17152095]
+    The 8th pipeline has a explained variance of [0.4595393  0.18012072 0.09809101 0.06332899]
 
 
 
@@ -1019,7 +1190,7 @@ The explained variance ratio per stage
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 0 minutes  34.468 seconds)
+   **Total running time of the script:** ( 0 minutes  24.928 seconds)
 
 
 .. _sphx_glr_download_examples_40_advanced_example_get_pipeline_components.py:
diff --git a/master/_sources/examples/40_advanced/example_inspect_predictions.rst.txt b/master/_sources/examples/40_advanced/example_inspect_predictions.rst.txt
index 03e9931e33..bc4369165d 100644
--- a/master/_sources/examples/40_advanced/example_inspect_predictions.rst.txt
+++ b/master/_sources/examples/40_advanced/example_inspect_predictions.rst.txt
@@ -90,8 +90,12 @@ accelerometer and gyroscope data collected by a phone. For more information see
 
  .. code-block:: none
 
-    Train score 0.9954285714285714
-    Test score 0.9826666666666667
+    /home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:68: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+      self.metafeatures = self.metafeatures.append(metafeatures)
+    /home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:72: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+      self.algorithm_runs[metric].append(runs)
+    Train score 0.9957142857142857
+    Test score 0.984
 
 
 
@@ -144,12 +148,12 @@ in the `scikit-learn docs <https://scikit-learn.org/stable/modules/permutation_i
 
  .. code-block:: none
 
-    gyro_y    : 0.000 +/- 0.001
-    gyro_x    : 0.025 +/- 0.003
-    gyro_z    : 0.037 +/- 0.003
-    acceleration_x: 0.056 +/- 0.006
-    acceleration_z: 0.126 +/- 0.006
-    acceleration_y: 0.280 +/- 0.006
+    gyro_y    : 0.001 +/- 0.002
+    gyro_x    : 0.023 +/- 0.002
+    gyro_z    : 0.036 +/- 0.003
+    acceleration_x: 0.051 +/- 0.006
+    acceleration_z: 0.144 +/- 0.007
+    acceleration_y: 0.272 +/- 0.006
 
 
 
@@ -233,7 +237,7 @@ these features. Again, we'll look at acceleration_y and acceleration_z.
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 4 minutes  2.065 seconds)
+   **Total running time of the script:** ( 4 minutes  6.567 seconds)
 
 
 .. _sphx_glr_download_examples_40_advanced_example_inspect_predictions.py:
diff --git a/master/_sources/examples/40_advanced/example_interpretable_models.rst.txt b/master/_sources/examples/40_advanced/example_interpretable_models.rst.txt
index 964719d51e..31658251b1 100644
--- a/master/_sources/examples/40_advanced/example_interpretable_models.rst.txt
+++ b/master/_sources/examples/40_advanced/example_interpretable_models.rst.txt
@@ -25,10 +25,12 @@ Interpretable models
 The following example shows how to inspect the models which *auto-sklearn*
 optimizes over and how to restrict them to an interpretable subset.
 
-.. GENERATED FROM PYTHON SOURCE LINES 10-14
+.. GENERATED FROM PYTHON SOURCE LINES 10-16
 
 .. code-block:: default
 
+    from pprint import pprint
+
     import autosklearn.classification
     import sklearn.datasets
     import sklearn.metrics
@@ -40,7 +42,7 @@ optimizes over and how to restrict them to an interpretable subset.
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 15-21
+.. GENERATED FROM PYTHON SOURCE LINES 17-23
 
 Show available classification models
 ====================================
@@ -49,7 +51,7 @@ We will first list all classifiers Auto-sklearn chooses from. A similar
 call is available for preprocessors (see below) and regression (not shown)
 as well.
 
-.. GENERATED FROM PYTHON SOURCE LINES 21-26
+.. GENERATED FROM PYTHON SOURCE LINES 23-28
 
 .. code-block:: default
 
@@ -88,12 +90,12 @@ as well.
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 27-29
+.. GENERATED FROM PYTHON SOURCE LINES 29-31
 
 Show available preprocessors
 ============================
 
-.. GENERATED FROM PYTHON SOURCE LINES 29-35
+.. GENERATED FROM PYTHON SOURCE LINES 31-37
 
 .. code-block:: default
 
@@ -135,12 +137,12 @@ Show available preprocessors
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 36-38
+.. GENERATED FROM PYTHON SOURCE LINES 38-40
 
 Data Loading
 ============
 
-.. GENERATED FROM PYTHON SOURCE LINES 38-43
+.. GENERATED FROM PYTHON SOURCE LINES 40-45
 
 .. code-block:: default
 
@@ -156,7 +158,7 @@ Data Loading
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 44-52
+.. GENERATED FROM PYTHON SOURCE LINES 46-54
 
 Build and fit a classifier
 ==========================
@@ -167,7 +169,7 @@ single best model in the end. However, we would like to note that the
 choice of which models is deemed interpretable is very much up to the user
 and can change from use case to use case.
 
-.. GENERATED FROM PYTHON SOURCE LINES 52-69
+.. GENERATED FROM PYTHON SOURCE LINES 54-71
 
 .. code-block:: default
 
@@ -198,6 +200,10 @@ and can change from use case to use case.
 
  .. code-block:: none
 
+    /home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:68: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+      self.metafeatures = self.metafeatures.append(metafeatures)
+    /home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:72: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+      self.algorithm_runs[metric].append(runs)
 
     AutoSklearnClassifier(ensemble_size=1,
                           include={'classifier': ['decision_tree', 'lda', 'sgd'],
@@ -209,17 +215,17 @@ and can change from use case to use case.
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 70-72
+.. GENERATED FROM PYTHON SOURCE LINES 72-74
 
 Print the final ensemble constructed by auto-sklearn
 ====================================================
 
-.. GENERATED FROM PYTHON SOURCE LINES 72-75
+.. GENERATED FROM PYTHON SOURCE LINES 74-77
 
 .. code-block:: default
 
 
-    print(automl.show_models())
+    pprint(automl.show_models(), indent=4)
 
 
 
@@ -231,25 +237,28 @@ Print the final ensemble constructed by auto-sklearn
 
  .. code-block:: none
 
-    [(1.000000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'sgd', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'polynomial', 'classifier:sgd:alpha': 0.0003336259810364362, 'classifier:sgd:average': 'False', 'classifier:sgd:fit_intercept': 'True', 'classifier:sgd:learning_rate': 'invscaling', 'classifier:sgd:loss': 'log', 'classifier:sgd:penalty': 'l2', 'classifier:sgd:tol': 0.006915066463246923, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'median', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize', 'feature_preprocessor:polynomial:degree': 3, 'feature_preprocessor:polynomial:include_bias': 'True', 'feature_preprocessor:polynomial:interaction_only': 'False', 'classifier:sgd:eta0': 0.010000000000000014, 'classifier:sgd:power_t': 0.5, 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.012199160249499144},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    ]
+    {   63: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78c482160>,
+                'cost': 0.007092198581560294,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78f80b910>,
+                'ensemble_weight': 1.0,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c482eb0>,
+                'model_id': 63,
+                'rank': 1,
+                'sklearn_classifier': SGDClassifier(alpha=9.999999999999991e-05, epsilon=1.3700860264543197e-05,
+                  eta0=0.010000000000000014, learning_rate='invscaling',
+                  loss='modified_huber', max_iter=64, random_state=1,
+                  tol=0.0004668987145850572, warm_start=True)}}
 
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 76-78
+.. GENERATED FROM PYTHON SOURCE LINES 78-80
 
 Get the Score of the final ensemble
 ===================================
 
-.. GENERATED FROM PYTHON SOURCE LINES 78-81
+.. GENERATED FROM PYTHON SOURCE LINES 80-83
 
 .. code-block:: default
 
@@ -266,7 +275,7 @@ Get the Score of the final ensemble
 
  .. code-block:: none
 
-    Accuracy score: 0.958041958041958
+    Accuracy score: 0.965034965034965
 
 
 
@@ -274,7 +283,7 @@ Get the Score of the final ensemble
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 1 minutes  55.571 seconds)
+   **Total running time of the script:** ( 1 minutes  54.044 seconds)
 
 
 .. _sphx_glr_download_examples_40_advanced_example_interpretable_models.py:
diff --git a/master/_sources/examples/40_advanced/example_metrics.rst.txt b/master/_sources/examples/40_advanced/example_metrics.rst.txt
index bcbe6232d9..42273f6376 100644
--- a/master/_sources/examples/40_advanced/example_metrics.rst.txt
+++ b/master/_sources/examples/40_advanced/example_metrics.rst.txt
@@ -141,17 +141,14 @@ Print a list of available metrics
             *roc_auc
             *average_precision
             *log_loss
-            *precision
             *precision_macro
             *precision_micro
             *precision_samples
             *precision_weighted
-            *recall
             *recall_macro
             *recall_micro
             *recall_samples
             *recall_weighted
-            *f1
             *f1_macro
             *f1_micro
             *f1_samples
@@ -204,7 +201,11 @@ First example: Use predefined accuracy metric
 
     ################################################################################
     Use predefined accuracy metric
-    Accuracy score 0.944 using accuracy
+    /home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:68: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+      self.metafeatures = self.metafeatures.append(metafeatures)
+    /home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:72: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+      self.algorithm_runs[metric].append(runs)
+    Accuracy score 0.958 using accuracy
 
 
 
@@ -304,17 +305,19 @@ Third example: Use own error metric
 
     ################################################################################
     Use self defined error metric
-    [WARNING] [2021-11-09 20:04:18,731:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
-    [WARNING] [2021-11-09 20:04:19,737:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
-    [WARNING] [2021-11-09 20:04:23,184:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
-    [WARNING] [2021-11-09 20:04:24,140:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
-    [WARNING] [2021-11-09 20:04:25,232:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
-    [WARNING] [2021-11-09 20:04:29,419:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
-    [WARNING] [2021-11-09 20:04:31,278:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
-    [WARNING] [2021-11-09 20:04:35,411:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
-    [WARNING] [2021-11-09 20:04:40,045:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
-    [WARNING] [2021-11-09 20:04:45,068:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
-    [WARNING] [2021-11-09 20:04:50,071:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+    [WARNING] [2022-01-25 16:33:33,542:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+    [WARNING] [2022-01-25 16:33:34,650:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+    [WARNING] [2022-01-25 16:33:37,944:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+    [WARNING] [2022-01-25 16:33:38,788:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+    [WARNING] [2022-01-25 16:33:39,764:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+    [WARNING] [2022-01-25 16:33:43,502:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+    [WARNING] [2022-01-25 16:33:45,472:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+    [WARNING] [2022-01-25 16:33:49,255:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+    [WARNING] [2022-01-25 16:33:53,483:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+    [WARNING] [2022-01-25 16:33:58,566:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+    [WARNING] [2022-01-25 16:34:03,041:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+    [WARNING] [2022-01-25 16:34:07,836:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+    [WARNING] [2022-01-25 16:34:08,936:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
     Error score -0.063 using error
 
 
@@ -416,20 +419,21 @@ Fifth example: Use own accuracy metric with additional argument
 
     ################################################################################
     Use self defined error with additional argument
-    [WARNING] [2021-11-09 20:06:23,413:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
-    [WARNING] [2021-11-09 20:06:27,269:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
-    [WARNING] [2021-11-09 20:06:34,117:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
-    [WARNING] [2021-11-09 20:06:35,067:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
-    [WARNING] [2021-11-09 20:06:38,388:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
-    [WARNING] [2021-11-09 20:06:39,226:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
-    [WARNING] [2021-11-09 20:06:40,191:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
-    [WARNING] [2021-11-09 20:06:45,538:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
-    [WARNING] [2021-11-09 20:06:49,773:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
-    [WARNING] [2021-11-09 20:06:54,348:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
-    [WARNING] [2021-11-09 20:06:56,606:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
-    [WARNING] [2021-11-09 20:07:01,814:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
-    [WARNING] [2021-11-09 20:07:06,326:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
-    [WARNING] [2021-11-09 20:07:07,428:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+    [WARNING] [2022-01-25 16:35:15,090:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+    [WARNING] [2022-01-25 16:35:18,533:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+    [WARNING] [2022-01-25 16:35:24,273:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+    [WARNING] [2022-01-25 16:35:25,186:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+    [WARNING] [2022-01-25 16:35:28,418:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+    [WARNING] [2022-01-25 16:35:29,368:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+    [WARNING] [2022-01-25 16:35:30,457:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+    [WARNING] [2022-01-25 16:35:35,650:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+    [WARNING] [2022-01-25 16:35:39,374:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+    [WARNING] [2022-01-25 16:35:43,590:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+    [WARNING] [2022-01-25 16:35:45,754:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+    [WARNING] [2022-01-25 16:35:50,626:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+    [WARNING] [2022-01-25 16:35:54,807:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+    [WARNING] [2022-01-25 16:35:55,786:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+    [WARNING] [2022-01-25 16:36:00,907:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
     Error score 0.615 using error_add
 
 
@@ -438,7 +442,7 @@ Fifth example: Use own accuracy metric with additional argument
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 5 minutes  19.475 seconds)
+   **Total running time of the script:** ( 4 minutes  39.637 seconds)
 
 
 .. _sphx_glr_download_examples_40_advanced_example_metrics.py:
diff --git a/master/_sources/examples/40_advanced/example_pandas_train_test.rst.txt b/master/_sources/examples/40_advanced/example_pandas_train_test.rst.txt
index 670cb7c1d9..14051320cf 100644
--- a/master/_sources/examples/40_advanced/example_pandas_train_test.rst.txt
+++ b/master/_sources/examples/40_advanced/example_pandas_train_test.rst.txt
@@ -170,6 +170,10 @@ Build and fit a classifier
 
  .. code-block:: none
 
+    /home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:68: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+      self.metafeatures = self.metafeatures.append(metafeatures)
+    /home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:72: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+      self.algorithm_runs[metric].append(runs)
 
     AutoSklearnClassifier(per_run_time_limit=30, time_left_for_this_task=120)
 
@@ -198,7 +202,7 @@ Get the Score of the final ensemble
 
  .. code-block:: none
 
-    Accuracy score 0.8579710144927536
+    Accuracy score 0.8492753623188406
 
 
 
@@ -239,7 +243,7 @@ be directly used for plotting
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 2 minutes  1.039 seconds)
+   **Total running time of the script:** ( 1 minutes  55.852 seconds)
 
 
 .. _sphx_glr_download_examples_40_advanced_example_pandas_train_test.py:
diff --git a/master/_sources/examples/40_advanced/example_resampling.rst.txt b/master/_sources/examples/40_advanced/example_resampling.rst.txt
index fedc608636..e5c5f9ba30 100644
--- a/master/_sources/examples/40_advanced/example_resampling.rst.txt
+++ b/master/_sources/examples/40_advanced/example_resampling.rst.txt
@@ -101,6 +101,10 @@ Holdout
 
  .. code-block:: none
 
+    /home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:68: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+      self.metafeatures = self.metafeatures.append(metafeatures)
+    /home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:72: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+      self.algorithm_runs[metric].append(runs)
 
     AutoSklearnClassifier(per_run_time_limit=30,
                           resampling_strategy_arguments={'train_size': 0.67},
@@ -133,7 +137,7 @@ Get the Score of the final ensemble
 
  .. code-block:: none
 
-    Accuracy score holdout:  0.9440559440559441
+    Accuracy score holdout:  0.958041958041958
 
 
 
@@ -176,6 +180,10 @@ Cross-validation
 
  .. code-block:: none
 
+    /home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:68: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+      self.metafeatures = self.metafeatures.append(metafeatures)
+    /home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:72: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+      self.algorithm_runs[metric].append(runs)
     Before re-fit
     Accuracy score CV 0.965034965034965
 
@@ -210,7 +218,7 @@ final ensemble on the whole dataset.
  .. code-block:: none
 
     After re-fit
-    Accuracy score CV 0.958041958041958
+    Accuracy score CV 0.965034965034965
 
 
 
@@ -232,13 +240,14 @@ Below is an example of using a predefined split. We split the training
 data by the first feature. In practice, one would use a splitting according
 to the use case at hand.
 
-.. GENERATED FROM PYTHON SOURCE LINES 100-114
+.. GENERATED FROM PYTHON SOURCE LINES 100-117
 
 .. code-block:: default
 
 
+    selected_indices = (X_train[:, 0] < np.mean(X_train[:, 0])).astype(int)
     resampling_strategy = sklearn.model_selection.PredefinedSplit(
-        test_fold=np.where(X_train[:, 0] < np.mean(X_train[:, 0]))[0]
+        test_fold=selected_indices
     )
 
     automl = autosklearn.classification.AutoSklearnClassifier(
@@ -250,6 +259,8 @@ to the use case at hand.
     )
     automl.fit(X_train, y_train, dataset_name='breast_cancer')
 
+    print(automl.sprint_statistics())
+
 
 
 
@@ -260,20 +271,30 @@ to the use case at hand.
 
  .. code-block:: none
 
+    /home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:68: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+      self.metafeatures = self.metafeatures.append(metafeatures)
+    /home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:72: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+      self.algorithm_runs[metric].append(runs)
+    auto-sklearn results:
+      Dataset name: breast_cancer
+      Metric: accuracy
+      Best validation score: 0.964789
+      Number of target algorithm runs: 31
+      Number of successful target algorithm runs: 31
+      Number of crashed target algorithm runs: 0
+      Number of target algorithms that exceeded the time limit: 0
+      Number of target algorithms that exceeded the memory limit: 0
+
 
-    AutoSklearnClassifier(per_run_time_limit=30,
-                          resampling_strategy=PredefinedSplit(test_fold=array([  3,   4, ..., 424, 425])),
-                          time_left_for_this_task=120,
-                          tmp_folder='/tmp/autosklearn_resampling_example_tmp')
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 115-117
+.. GENERATED FROM PYTHON SOURCE LINES 118-120
 
 For custom resampling strategies (i.e. resampling strategies that are not
 defined as strings by Auto-sklearn) it is necessary to perform a refit:
 
-.. GENERATED FROM PYTHON SOURCE LINES 117-119
+.. GENERATED FROM PYTHON SOURCE LINES 120-122
 
 .. code-block:: default
 
@@ -291,13 +312,13 @@ defined as strings by Auto-sklearn) it is necessary to perform a refit:
 
 
     AutoSklearnClassifier(per_run_time_limit=30,
-                          resampling_strategy=PredefinedSplit(test_fold=array([  3,   4, ..., 424, 425])),
+                          resampling_strategy=PredefinedSplit(test_fold=array([0, 0, ..., 1, 1])),
                           time_left_for_this_task=120,
                           tmp_folder='/tmp/autosklearn_resampling_example_tmp')
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 120-125
+.. GENERATED FROM PYTHON SOURCE LINES 123-128
 
 Get the Score of the final ensemble (again)
 ===========================================
@@ -305,7 +326,7 @@ Get the Score of the final ensemble (again)
 Obviously, this score is pretty bad as we "destroyed" the dataset by
 splitting it on the first feature.
 
-.. GENERATED FROM PYTHON SOURCE LINES 125-127
+.. GENERATED FROM PYTHON SOURCE LINES 128-130
 
 .. code-block:: default
 
@@ -321,7 +342,7 @@ splitting it on the first feature.
 
  .. code-block:: none
 
-    Accuracy score custom split 0.38461538461538464
+    Accuracy score custom split 0.958041958041958
 
 
 
@@ -329,7 +350,7 @@ splitting it on the first feature.
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 6 minutes  22.500 seconds)
+   **Total running time of the script:** ( 6 minutes  27.008 seconds)
 
 
 .. _sphx_glr_download_examples_40_advanced_example_resampling.py:
diff --git a/master/_sources/examples/40_advanced/example_single_configuration.rst.txt b/master/_sources/examples/40_advanced/example_single_configuration.rst.txt
index f844023b85..86fc49afdb 100644
--- a/master/_sources/examples/40_advanced/example_single_configuration.rst.txt
+++ b/master/_sources/examples/40_advanced/example_single_configuration.rst.txt
@@ -161,61 +161,73 @@ Fit an user provided configuration
 
  .. code-block:: none
 
-    {'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7f937c5a9ca0>, 'balancing': Balancing(random_state=1, strategy='weighting'), 'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7f937b0d0e50>, 'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7f937b0d0a00>}
+    {'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78ffee460>, 'balancing': Balancing(random_state=1), 'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78ab79cd0>, 'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78aae48b0>}
     RunInfo(config=Configuration:
-      balancing:strategy, Value: 'weighting'
+      balancing:strategy, Value: 'none'
       classifier:__choice__, Value: 'random_forest'
-      classifier:random_forest:bootstrap, Value: 'False'
-      classifier:random_forest:criterion, Value: 'gini'
+      classifier:random_forest:bootstrap, Value: 'True'
+      classifier:random_forest:criterion, Value: 'entropy'
       classifier:random_forest:max_depth, Constant: 'None'
-      classifier:random_forest:max_features, Value: 0.40665293991189455
+      classifier:random_forest:max_features, Value: 0.7056146361355262
       classifier:random_forest:max_leaf_nodes, Constant: 'None'
       classifier:random_forest:min_impurity_decrease, Constant: 0.0
-      classifier:random_forest:min_samples_leaf, Value: 13
+      classifier:random_forest:min_samples_leaf, Value: 11
       classifier:random_forest:min_samples_split, Value: 11
       classifier:random_forest:min_weight_fraction_leaf, Constant: 0.0
       data_preprocessor:__choice__, Value: 'feature_type'
-      data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__, Value: 'no_encoding'
-      data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__, Value: 'minority_coalescer'
-      data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction, Value: 0.0012559712452161484
-      data_preprocessor:feature_type:numerical_transformer:imputation:strategy, Value: 'median'
-      data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__, Value: 'quantile_transformer'
-      data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles, Value: 204
-      data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution, Value: 'normal'
-      feature_preprocessor:__choice__, Value: 'polynomial'
-      feature_preprocessor:polynomial:degree, Value: 2
-      feature_preprocessor:polynomial:include_bias, Value: 'False'
-      feature_preprocessor:polynomial:interaction_only, Value: 'True'
+      data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__, Value: 'one_hot_encoding'
+      data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__, Value: 'no_coalescense'
+      data_preprocessor:feature_type:numerical_transformer:imputation:strategy, Value: 'mean'
+      data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__, Value: 'none'
+      feature_preprocessor:__choice__, Value: 'extra_trees_preproc_for_classification'
+      feature_preprocessor:extra_trees_preproc_for_classification:bootstrap, Value: 'False'
+      feature_preprocessor:extra_trees_preproc_for_classification:criterion, Value: 'gini'
+      feature_preprocessor:extra_trees_preproc_for_classification:max_depth, Constant: 'None'
+      feature_preprocessor:extra_trees_preproc_for_classification:max_features, Value: 0.41746352744644133
+      feature_preprocessor:extra_trees_preproc_for_classification:max_leaf_nodes, Constant: 'None'
+      feature_preprocessor:extra_trees_preproc_for_classification:min_impurity_decrease, Constant: 0.0
+      feature_preprocessor:extra_trees_preproc_for_classification:min_samples_leaf, Value: 5
+      feature_preprocessor:extra_trees_preproc_for_classification:min_samples_split, Value: 2
+      feature_preprocessor:extra_trees_preproc_for_classification:min_weight_fraction_leaf, Constant: 0.0
+      feature_preprocessor:extra_trees_preproc_for_classification:n_estimators, Constant: 100
     , instance=None, instance_specific=None, seed=1, cutoff=60, capped=False, budget=0.0, source_id=0)
-    RunValue(cost=0.012322274881516604, time=2.914306402206421, status=<StatusType.SUCCESS: 1>, starttime=1636487580.5942185, endtime=1636487583.5312755, additional_info={'duration': 2.8303675651550293, 'num_run': 2, 'train_loss': 0.012143858010275621, 'configuration_origin': None})
+    RunValue(cost=0.03033175355450235, time=2.2301056385040283, status=<StatusType.SUCCESS: 1>, starttime=1643127754.4012537, endtime=1643127756.6555753, additional_info={'duration': 2.1325058937072754, 'num_run': 2, 'train_loss': 0.047641289117234975, 'configuration_origin': None})
     Passed Configuration: Configuration:
-      balancing:strategy, Value: 'weighting'
+      balancing:strategy, Value: 'none'
       classifier:__choice__, Value: 'random_forest'
-      classifier:random_forest:bootstrap, Value: 'False'
-      classifier:random_forest:criterion, Value: 'gini'
+      classifier:random_forest:bootstrap, Value: 'True'
+      classifier:random_forest:criterion, Value: 'entropy'
       classifier:random_forest:max_depth, Constant: 'None'
-      classifier:random_forest:max_features, Value: 0.40665293991189455
+      classifier:random_forest:max_features, Value: 0.7056146361355262
       classifier:random_forest:max_leaf_nodes, Constant: 'None'
       classifier:random_forest:min_impurity_decrease, Constant: 0.0
-      classifier:random_forest:min_samples_leaf, Value: 13
+      classifier:random_forest:min_samples_leaf, Value: 11
       classifier:random_forest:min_samples_split, Value: 11
       classifier:random_forest:min_weight_fraction_leaf, Constant: 0.0
       data_preprocessor:__choice__, Value: 'feature_type'
-      data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__, Value: 'no_encoding'
-      data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__, Value: 'minority_coalescer'
-      data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction, Value: 0.0012559712452161484
-      data_preprocessor:feature_type:numerical_transformer:imputation:strategy, Value: 'median'
-      data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__, Value: 'quantile_transformer'
-      data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles, Value: 204
-      data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution, Value: 'normal'
-      feature_preprocessor:__choice__, Value: 'polynomial'
-      feature_preprocessor:polynomial:degree, Value: 2
-      feature_preprocessor:polynomial:include_bias, Value: 'False'
-      feature_preprocessor:polynomial:interaction_only, Value: 'True'
-
-    Random Forest: RandomForestClassifier(bootstrap=False, max_features=14, min_samples_leaf=13,
+      data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__, Value: 'one_hot_encoding'
+      data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__, Value: 'no_coalescense'
+      data_preprocessor:feature_type:numerical_transformer:imputation:strategy, Value: 'mean'
+      data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__, Value: 'none'
+      feature_preprocessor:__choice__, Value: 'extra_trees_preproc_for_classification'
+      feature_preprocessor:extra_trees_preproc_for_classification:bootstrap, Value: 'False'
+      feature_preprocessor:extra_trees_preproc_for_classification:criterion, Value: 'gini'
+      feature_preprocessor:extra_trees_preproc_for_classification:max_depth, Constant: 'None'
+      feature_preprocessor:extra_trees_preproc_for_classification:max_features, Value: 0.41746352744644133
+      feature_preprocessor:extra_trees_preproc_for_classification:max_leaf_nodes, Constant: 'None'
+      feature_preprocessor:extra_trees_preproc_for_classification:min_impurity_decrease, Constant: 0.0
+      feature_preprocessor:extra_trees_preproc_for_classification:min_samples_leaf, Value: 5
+      feature_preprocessor:extra_trees_preproc_for_classification:min_samples_split, Value: 2
+      feature_preprocessor:extra_trees_preproc_for_classification:min_weight_fraction_leaf, Constant: 0.0
+      feature_preprocessor:extra_trees_preproc_for_classification:n_estimators, Constant: 100
+
+    Random Forest: RandomForestClassifier(criterion='entropy', max_features=6, min_samples_leaf=11,
                            min_samples_split=11, n_estimators=512, n_jobs=1,
                            random_state=1, warm_start=True)
+    /home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:68: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+      self.metafeatures = self.metafeatures.append(metafeatures)
+    /home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:72: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+      self.algorithm_runs[metric].append(runs)
 
 
 
@@ -223,7 +235,7 @@ Fit an user provided configuration
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 2 minutes  24.877 seconds)
+   **Total running time of the script:** ( 2 minutes  5.409 seconds)
 
 
 .. _sphx_glr_download_examples_40_advanced_example_single_configuration.py:
diff --git a/master/_sources/examples/40_advanced/sg_execution_times.rst.txt b/master/_sources/examples/40_advanced/sg_execution_times.rst.txt
index 9363953a69..553b5ff9ae 100644
--- a/master/_sources/examples/40_advanced/sg_execution_times.rst.txt
+++ b/master/_sources/examples/40_advanced/sg_execution_times.rst.txt
@@ -5,26 +5,26 @@
 
 Computation times
 =================
-**25:21.692** total execution time for **examples_40_advanced** files:
+**23:57.125** total execution time for **examples_40_advanced** files:
 
 +------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_examples_40_advanced_example_resampling.py` (``example_resampling.py``)                           | 06:22.500 | 0.0 MB |
+| :ref:`sphx_glr_examples_40_advanced_example_resampling.py` (``example_resampling.py``)                           | 06:27.008 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_examples_40_advanced_example_metrics.py` (``example_metrics.py``)                                 | 05:19.475 | 0.0 MB |
+| :ref:`sphx_glr_examples_40_advanced_example_metrics.py` (``example_metrics.py``)                                 | 04:39.637 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_examples_40_advanced_example_inspect_predictions.py` (``example_inspect_predictions.py``)         | 04:02.065 | 0.0 MB |
+| :ref:`sphx_glr_examples_40_advanced_example_inspect_predictions.py` (``example_inspect_predictions.py``)         | 04:06.567 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_examples_40_advanced_example_single_configuration.py` (``example_single_configuration.py``)       | 02:24.877 | 0.0 MB |
+| :ref:`sphx_glr_examples_40_advanced_example_single_configuration.py` (``example_single_configuration.py``)       | 02:05.409 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_examples_40_advanced_example_calc_multiple_metrics.py` (``example_calc_multiple_metrics.py``)     | 02:08.166 | 0.0 MB |
+| :ref:`sphx_glr_examples_40_advanced_example_calc_multiple_metrics.py` (``example_calc_multiple_metrics.py``)     | 01:59.942 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_examples_40_advanced_example_pandas_train_test.py` (``example_pandas_train_test.py``)             | 02:01.039 | 0.0 MB |
+| :ref:`sphx_glr_examples_40_advanced_example_pandas_train_test.py` (``example_pandas_train_test.py``)             | 01:55.852 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_examples_40_advanced_example_interpretable_models.py` (``example_interpretable_models.py``)       | 01:55.571 | 0.0 MB |
+| :ref:`sphx_glr_examples_40_advanced_example_interpretable_models.py` (``example_interpretable_models.py``)       | 01:54.044 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_examples_40_advanced_example_get_pipeline_components.py` (``example_get_pipeline_components.py``) | 00:34.468 | 0.0 MB |
+| :ref:`sphx_glr_examples_40_advanced_example_get_pipeline_components.py` (``example_get_pipeline_components.py``) | 00:24.928 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_examples_40_advanced_example_debug_logging.py` (``example_debug_logging.py``)                     | 00:22.911 | 0.0 MB |
+| :ref:`sphx_glr_examples_40_advanced_example_debug_logging.py` (``example_debug_logging.py``)                     | 00:14.975 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_examples_40_advanced_example_feature_types.py` (``example_feature_types.py``)                     | 00:10.620 | 0.0 MB |
+| :ref:`sphx_glr_examples_40_advanced_example_feature_types.py` (``example_feature_types.py``)                     | 00:08.763 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/master/_sources/examples/60_search/example_parallel_manual_spawning_cli.rst.txt b/master/_sources/examples/60_search/example_parallel_manual_spawning_cli.rst.txt
index 40a9b6f91c..0bd4e89192 100644
--- a/master/_sources/examples/60_search/example_parallel_manual_spawning_cli.rst.txt
+++ b/master/_sources/examples/60_search/example_parallel_manual_spawning_cli.rst.txt
@@ -29,11 +29,17 @@ for parallel optimization.
 This example shows how to start the dask scheduler and spawn
 workers for *Auto-sklearn* manually from the command line. Use this example
 as a starting point to parallelize *Auto-sklearn* across multiple
-machines. If you want to start everything manually from within Python
-please see :ref:`sphx_glr_examples_60_search_example_parallel_manual_spawning_python.py`.
+machines.
+
 To run *Auto-sklearn* in parallel on a single machine check out the example
 :ref:`sphx_glr_examples_60_search_example_parallel_n_jobs.py`.
 
+If you want to start everything manually from within Python
+please see ``:ref:sphx_glr_examples_60_search_example_parallel_manual_spawning_python.py``.
+
+**NOTE:** Above example is disabled due to issue https://github.com/dask/distributed/issues/5627
+
+
 You can learn more about the dask command line interface from
 https://docs.dask.org/en/latest/setup/cli.html.
 
@@ -64,12 +70,12 @@ be started from the command line and not from within a Python file via
 the ``subprocess`` module as done here (for the sake of having a self-contained
 example).
 
-.. GENERATED FROM PYTHON SOURCE LINES 51-53
+.. GENERATED FROM PYTHON SOURCE LINES 57-59
 
 Import statements
 =================
 
-.. GENERATED FROM PYTHON SOURCE LINES 53-70
+.. GENERATED FROM PYTHON SOURCE LINES 59-76
 
 .. code-block:: default
 
@@ -97,7 +103,7 @@ Import statements
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 71-79
+.. GENERATED FROM PYTHON SOURCE LINES 77-85
 
 0. Setup client-scheduler communication
 =======================================
@@ -108,7 +114,7 @@ relevant information in a file for which we provided the name and
 location. This filename is also given to the worker so they can find all
 relevant information to connect to the scheduler.
 
-.. GENERATED FROM PYTHON SOURCE LINES 79-83
+.. GENERATED FROM PYTHON SOURCE LINES 85-89
 
 .. code-block:: default
 
@@ -123,7 +129,7 @@ relevant information to connect to the scheduler.
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 84-95
+.. GENERATED FROM PYTHON SOURCE LINES 90-101
 
 1. Start scheduler
 ==================
@@ -137,7 +143,7 @@ Starting the scheduler is done with the following bash command:
 We will now execute this bash command from within Python to have a
 self-contained example:
 
-.. GENERATED FROM PYTHON SOURCE LINES 95-118
+.. GENERATED FROM PYTHON SOURCE LINES 101-124
 
 .. code-block:: default
 
@@ -171,7 +177,7 @@ self-contained example:
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 119-137
+.. GENERATED FROM PYTHON SOURCE LINES 125-143
 
 2. Start two workers
 ====================
@@ -192,7 +198,7 @@ compatible with Auto-sklearn creating new processes in the workers itself.
 We disable dask's memory management by passing ``--memory-limit`` as
 Auto-sklearn does the memory management itself.
 
-.. GENERATED FROM PYTHON SOURCE LINES 137-161
+.. GENERATED FROM PYTHON SOURCE LINES 143-167
 
 .. code-block:: default
 
@@ -227,7 +233,7 @@ Auto-sklearn does the memory management itself.
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 162-167
+.. GENERATED FROM PYTHON SOURCE LINES 168-173
 
 3. Creating a client in Python
 ==============================
@@ -235,7 +241,7 @@ Auto-sklearn does the memory management itself.
 Finally we create a dask cluster which also connects to the scheduler via
 the information in the file created by the scheduler.
 
-.. GENERATED FROM PYTHON SOURCE LINES 167-170
+.. GENERATED FROM PYTHON SOURCE LINES 173-176
 
 .. code-block:: default
 
@@ -249,12 +255,12 @@ the information in the file created by the scheduler.
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 171-173
+.. GENERATED FROM PYTHON SOURCE LINES 177-179
 
 Start Auto-sklearn
 ~~~~~~~~~~~~~~~~~~
 
-.. GENERATED FROM PYTHON SOURCE LINES 173-205
+.. GENERATED FROM PYTHON SOURCE LINES 179-211
 
 .. code-block:: default
 
@@ -300,22 +306,26 @@ Start Auto-sklearn
 
  .. code-block:: none
 
+    /home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:68: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+      self.metafeatures = self.metafeatures.append(metafeatures)
+    /home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:72: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+      self.algorithm_runs[metric].append(runs)
     auto-sklearn results:
-      Dataset name: 4433d19e-4199-11ec-8a32-1b4c952643c3
+      Dataset name: 74d6c3db-7dfd-11ec-875f-3137a7553b01
       Metric: accuracy
       Best validation score: 0.992908
-      Number of target algorithm runs: 12
-      Number of successful target algorithm runs: 11
+      Number of target algorithm runs: 14
+      Number of successful target algorithm runs: 13
       Number of crashed target algorithm runs: 0
       Number of target algorithms that exceeded the time limit: 1
       Number of target algorithms that exceeded the memory limit: 0
 
-    Accuracy score 0.965034965034965
+    Accuracy score 0.958041958041958
 
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 206-212
+.. GENERATED FROM PYTHON SOURCE LINES 212-218
 
 Wait until all workers are closed
 =================================
@@ -324,7 +334,7 @@ This is only necessary if the workers are started from within this python
 script. In a real application one would start them directly from the command
 line.
 
-.. GENERATED FROM PYTHON SOURCE LINES 212-216
+.. GENERATED FROM PYTHON SOURCE LINES 218-222
 
 .. code-block:: default
 
@@ -342,7 +352,7 @@ line.
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 0 minutes  39.363 seconds)
+   **Total running time of the script:** ( 0 minutes  39.167 seconds)
 
 
 .. _sphx_glr_download_examples_60_search_example_parallel_manual_spawning_cli.py:
diff --git a/master/_sources/examples/60_search/example_parallel_manual_spawning_python.rst.txt b/master/_sources/examples/60_search/example_parallel_manual_spawning_python.rst.txt
deleted file mode 100644
index c40c361ac9..0000000000
--- a/master/_sources/examples/60_search/example_parallel_manual_spawning_python.rst.txt
+++ /dev/null
@@ -1,268 +0,0 @@
-
-.. DO NOT EDIT.
-.. THIS FILE WAS AUTOMATICALLY GENERATED BY SPHINX-GALLERY.
-.. TO MAKE CHANGES, EDIT THE SOURCE PYTHON FILE:
-.. "examples/60_search/example_parallel_manual_spawning_python.py"
-.. LINE NUMBERS ARE GIVEN BELOW.
-
-.. only:: html
-
-    .. note::
-        :class: sphx-glr-download-link-note
-
-        Click :ref:`here <sphx_glr_download_examples_60_search_example_parallel_manual_spawning_python.py>`
-        to download the full example code or to run this example in your browser via Binder
-
-.. rst-class:: sphx-glr-example-title
-
-.. _sphx_glr_examples_60_search_example_parallel_manual_spawning_python.py:
-
-
-===================================================
-Parallel Usage: Spawning workers from within Python
-===================================================
-
-*Auto-sklearn* uses
-`dask.distributed <https://distributed.dask.org/en/latest/index.html>`_
-for parallel optimization.
-
-This example shows how to start the dask scheduler and spawn
-workers for *Auto-sklearn* manually within Python. Use this example
-as a starting point to parallelize *Auto-sklearn* across multiple
-machines. If you want to start everything manually from the command line
-please see :ref:`sphx_glr_examples_60_search_example_parallel_manual_spawning_cli.py`.
-To run *Auto-sklearn* in parallel on a single machine check out the example
-:ref:`sphx_glr_examples_60_search_example_parallel_n_jobs.py`.
-
-When manually passing a dask client to Auto-sklearn, all logic
-must be guarded by ``if __name__ == "__main__":`` statements! We use
-multiple such statements to properly render this example as a notebook
-and also allow execution via the command line.
-
-Background
-==========
-
-To run Auto-sklearn distributed on multiple machines we need to set
-up three components:
-
-1. **Auto-sklearn and a dask client**. This will manage all workload, find new
-   configurations to evaluate and submit jobs via a dask client. As this
-   runs Bayesian optimization it should be executed on its own CPU.
-2. **The dask workers**. They will do the actual work of running machine
-   learning algorithms and require their own CPU each.
-3. **The scheduler**. It manages the communication between the dask client
-   and the different dask workers. As the client and all workers connect
-   to the scheduler it must be started first. This is a light-weight job
-   and does not require its own CPU.
-
-We will now start these three components in reverse order: scheduler,
-workers and client. Also, in a real setup, the scheduler and the workers should
-be started from the command line and not from within a Python file via
-the ``subprocess`` module as done here (for the sake of having a self-contained
-example).
-
-.. GENERATED FROM PYTHON SOURCE LINES 46-62
-
-.. code-block:: default
-
-
-    import asyncio
-    import multiprocessing
-    import time
-
-    import dask
-    import dask.distributed
-    import sklearn.datasets
-    import sklearn.metrics
-
-    from autosklearn.classification import AutoSklearnClassifier
-    from autosklearn.constants import MULTICLASS_CLASSIFICATION
-
-    tmp_folder = '/tmp/autosklearn_parallel_2_example_tmp'
-
-
-
-
-
-
-
-
-
-.. GENERATED FROM PYTHON SOURCE LINES 63-69
-
-Define function to start worker
-===============================
-
-Define the function to start a dask worker from python. This
-is a bit cumbersome and should ideally be done from the command line.
-We do it here only for illustrational purpose.
-
-.. GENERATED FROM PYTHON SOURCE LINES 69-89
-
-.. code-block:: default
-
-
-    # Check the dask docs at
-    # https://docs.dask.org/en/latest/setup/python-advanced.html for further
-    # information.
-
-    def start_python_worker(scheduler_address):
-        dask.config.set({'distributed.worker.daemon': False})
-
-        async def do_work():
-            async with dask.distributed.Nanny(
-                scheduler_ip=scheduler_address,
-                nthreads=1,
-                lifetime=35,  # automatically shut down the worker so this loop ends
-                memory_limit=0,  # Disable memory management as it is done by Auto-sklearn itself
-            ) as worker:
-                await worker.finished()
-
-        asyncio.get_event_loop().run_until_complete(do_work())
-
-
-
-
-
-
-
-
-
-.. GENERATED FROM PYTHON SOURCE LINES 90-98
-
-Start Auto-sklearn
-==================
-
-We are now ready to start *auto-sklearn* and all dask related processes.
-
-To use auto-sklearn in parallel we must guard the code with
-``if __name__ == '__main__'``. We then start a dask cluster as a context,
-which means that it is automatically stopped once all computation is done.
-
-.. GENERATED FROM PYTHON SOURCE LINES 98-154
-
-.. code-block:: default
-
-    if __name__ == '__main__':
-        X, y = sklearn.datasets.load_breast_cancer(return_X_y=True)
-        X_train, X_test, y_train, y_test = \
-            sklearn.model_selection.train_test_split(X, y, random_state=1)
-
-        # 1. Create a dask scheduler (LocalCluster)
-        with dask.distributed.LocalCluster(
-            n_workers=0, processes=True, threads_per_worker=1,
-        ) as cluster:
-
-            # 2. Start the workers
-            # now we start the two workers, one from within Python, the other
-            # via the command line.
-            worker_processes = []
-            for _ in range(2):
-                process_python_worker = multiprocessing.Process(
-                    target=start_python_worker,
-                    args=(cluster.scheduler_address, ),
-                )
-                process_python_worker.start()
-                worker_processes.append(process_python_worker)
-
-            # Wait a second for workers to become available
-            time.sleep(1)
-
-            # 3. Start the client
-            with dask.distributed.Client(address=cluster.scheduler_address) as client:
-                automl = AutoSklearnClassifier(
-                    delete_tmp_folder_after_terminate=False,
-                    time_left_for_this_task=30,
-                    per_run_time_limit=10,
-                    memory_limit=1024,
-                    tmp_folder=tmp_folder,
-                    seed=777,
-                    # n_jobs is ignored internally as we pass a dask client.
-                    n_jobs=1,
-                    # Pass a dask client which connects to the previously constructed cluster.
-                    dask_client=client,
-                )
-                automl.fit(X_train, y_train)
-
-                automl.fit_ensemble(
-                    y_train,
-                    task=MULTICLASS_CLASSIFICATION,
-                    dataset_name='digits',
-                    ensemble_size=20,
-                    ensemble_nbest=50,
-                )
-
-            predictions = automl.predict(X_test)
-            print(automl.sprint_statistics())
-            print("Accuracy score", sklearn.metrics.accuracy_score(y_test, predictions))
-
-            # Wait until all workers are closed
-            for process in worker_processes:
-                process_python_worker.join()
-
-
-
-
-.. rst-class:: sphx-glr-script-out
-
- Out:
-
- .. code-block:: none
-
-    [ERROR] [2021-11-09 20:12:34,562:asyncio] _GatheringFuture exception was never retrieved
-    future: <_GatheringFuture finished exception=CancelledError()>
-    asyncio.exceptions.CancelledError
-    auto-sklearn results:
-      Dataset name: 5b2e548d-4199-11ec-8a32-1b4c952643c3
-      Metric: accuracy
-      Best validation score: 0.992908
-      Number of target algorithm runs: 14
-      Number of successful target algorithm runs: 12
-      Number of crashed target algorithm runs: 0
-      Number of target algorithms that exceeded the time limit: 2
-      Number of target algorithms that exceeded the memory limit: 0
-
-    Accuracy score 0.965034965034965
-
-
-
-
-
-.. rst-class:: sphx-glr-timing
-
-   **Total running time of the script:** ( 0 minutes  35.796 seconds)
-
-
-.. _sphx_glr_download_examples_60_search_example_parallel_manual_spawning_python.py:
-
-
-.. only :: html
-
- .. container:: sphx-glr-footer
-    :class: sphx-glr-footer-example
-
-
-  .. container:: binder-badge
-
-    .. image:: images/binder_badge_logo.svg
-      :target: https://mybinder.org/v2/gh/automl/auto-sklearn/master?urlpath=lab/tree/notebooks/examples/60_search/example_parallel_manual_spawning_python.ipynb
-      :alt: Launch binder
-      :width: 150 px
-
-
-  .. container:: sphx-glr-download sphx-glr-download-python
-
-     :download:`Download Python source code: example_parallel_manual_spawning_python.py <example_parallel_manual_spawning_python.py>`
-
-
-
-  .. container:: sphx-glr-download sphx-glr-download-jupyter
-
-     :download:`Download Jupyter notebook: example_parallel_manual_spawning_python.ipynb <example_parallel_manual_spawning_python.ipynb>`
-
-
-.. only:: html
-
- .. rst-class:: sphx-glr-signature
-
-    `Gallery generated by Sphinx-Gallery <https://sphinx-gallery.github.io>`_
diff --git a/master/_sources/examples/60_search/example_parallel_n_jobs.rst.txt b/master/_sources/examples/60_search/example_parallel_n_jobs.rst.txt
index 738d6622e1..7a592fcf5c 100644
--- a/master/_sources/examples/60_search/example_parallel_n_jobs.rst.txt
+++ b/master/_sources/examples/60_search/example_parallel_n_jobs.rst.txt
@@ -109,14 +109,18 @@ To use ``n_jobs_`` we must guard the code
 
  .. code-block:: none
 
+    /home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:68: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+      self.metafeatures = self.metafeatures.append(metafeatures)
+    /home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:72: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+      self.algorithm_runs[metric].append(runs)
     auto-sklearn results:
       Dataset name: breast_cancer
       Metric: accuracy
       Best validation score: 0.985816
-      Number of target algorithm runs: 49
-      Number of successful target algorithm runs: 48
-      Number of crashed target algorithm runs: 0
-      Number of target algorithms that exceeded the time limit: 1
+      Number of target algorithm runs: 52
+      Number of successful target algorithm runs: 49
+      Number of crashed target algorithm runs: 1
+      Number of target algorithms that exceeded the time limit: 2
       Number of target algorithms that exceeded the memory limit: 0
 
 
@@ -126,7 +130,7 @@ To use ``n_jobs_`` we must guard the code
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 2 minutes  24.442 seconds)
+   **Total running time of the script:** ( 1 minutes  57.601 seconds)
 
 
 .. _sphx_glr_download_examples_60_search_example_parallel_n_jobs.py:
diff --git a/master/_sources/examples/60_search/example_random_search.rst.txt b/master/_sources/examples/60_search/example_random_search.rst.txt
index cfda9b3e32..5363902a85 100644
--- a/master/_sources/examples/60_search/example_random_search.rst.txt
+++ b/master/_sources/examples/60_search/example_random_search.rst.txt
@@ -31,10 +31,11 @@ SMAC, as demonstrated in the example below. Furthermore, the example also demons
 as yet another alternative optimizatino strategy.
 Both examples are intended to show how the optimization strategy in *auto-sklearn* can be adapted.
 
-.. GENERATED FROM PYTHON SOURCE LINES 15-26
+.. GENERATED FROM PYTHON SOURCE LINES 15-27
 
 .. code-block:: default
 
+    from pprint import pprint
 
     import sklearn.model_selection
     import sklearn.datasets
@@ -53,12 +54,12 @@ Both examples are intended to show how the optimization strategy in *auto-sklear
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 27-29
+.. GENERATED FROM PYTHON SOURCE LINES 28-30
 
 Data Loading
 ============
 
-.. GENERATED FROM PYTHON SOURCE LINES 29-35
+.. GENERATED FROM PYTHON SOURCE LINES 30-36
 
 .. code-block:: default
 
@@ -75,12 +76,12 @@ Data Loading
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 36-38
+.. GENERATED FROM PYTHON SOURCE LINES 37-39
 
 Fit a classifier using ROAR
 ===========================
 
-.. GENERATED FROM PYTHON SOURCE LINES 38-86
+.. GENERATED FROM PYTHON SOURCE LINES 39-87
 
 .. code-block:: default
 
@@ -124,7 +125,7 @@ Fit a classifier using ROAR
     print('#' * 80)
     print('Results for ROAR.')
     # Print the final ensemble constructed by auto-sklearn via ROAR.
-    print(automl.show_models())
+    pprint(automl.show_models(), indent=4)
     predictions = automl.predict(X_test)
     # Print statistics about the auto-sklearn run such as number of
     # iterations, number of models failed with a time out.
@@ -142,214 +143,293 @@ Fit a classifier using ROAR
 
  .. code-block:: none
 
+    [ERROR] [2022-01-25 16:41:15,309:asyncio] _GatheringFuture exception was never retrieved
+    future: <_GatheringFuture finished exception=CancelledError()>
+    asyncio.exceptions.CancelledError
     ################################################################################
     Results for ROAR.
-    [(0.120000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'random_forest', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'select_rates_classification', 'classifier:random_forest:bootstrap': 'False', 'classifier:random_forest:criterion': 'entropy', 'classifier:random_forest:max_depth': 'None', 'classifier:random_forest:max_features': 0.13366756044911932, 'classifier:random_forest:max_leaf_nodes': 'None', 'classifier:random_forest:min_impurity_decrease': 0.0, 'classifier:random_forest:min_samples_leaf': 10, 'classifier:random_forest:min_samples_split': 18, 'classifier:random_forest:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'robust_scaler', 'feature_preprocessor:select_rates_classification:alpha': 0.06835899987255477, 'feature_preprocessor:select_rates_classification:score_func': 'chi2', 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max': 0.791639810538127, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min': 0.2949722712387337, 'feature_preprocessor:select_rates_classification:mode': 'fwe'},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.100000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'qda', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:qda:reg_param': 0.45597150689539434, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize'},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.100000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'adaboost', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'fast_ica', 'classifier:adaboost:algorithm': 'SAMME', 'classifier:adaboost:learning_rate': 1.6587261797143242, 'classifier:adaboost:max_depth': 8, 'classifier:adaboost:n_estimators': 368, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'median', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'normalize', 'feature_preprocessor:fast_ica:algorithm': 'parallel', 'feature_preprocessor:fast_ica:fun': 'cube', 'feature_preprocessor:fast_ica:whiten': 'True', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.00031132126574883666, 'feature_preprocessor:fast_ica:n_components': 530},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.080000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'passive_aggressive', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'extra_trees_preproc_for_classification', 'classifier:passive_aggressive:C': 1.4622203211010016e-05, 'classifier:passive_aggressive:average': 'True', 'classifier:passive_aggressive:fit_intercept': 'True', 'classifier:passive_aggressive:loss': 'hinge', 'classifier:passive_aggressive:tol': 0.0017964371726429177, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'power_transformer', 'feature_preprocessor:extra_trees_preproc_for_classification:bootstrap': 'False', 'feature_preprocessor:extra_trees_preproc_for_classification:criterion': 'entropy', 'feature_preprocessor:extra_trees_preproc_for_classification:max_depth': 'None', 'feature_preprocessor:extra_trees_preproc_for_classification:max_features': 0.10392737384710127, 'feature_preprocessor:extra_trees_preproc_for_classification:max_leaf_nodes': 'None', 'feature_preprocessor:extra_trees_preproc_for_classification:min_impurity_decrease': 0.0, 'feature_preprocessor:extra_trees_preproc_for_classification:min_samples_leaf': 11, 'feature_preprocessor:extra_trees_preproc_for_classification:min_samples_split': 9, 'feature_preprocessor:extra_trees_preproc_for_classification:min_weight_fraction_leaf': 0.0, 'feature_preprocessor:extra_trees_preproc_for_classification:n_estimators': 100},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.060000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'lda', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'nystroem_sampler', 'classifier:lda:shrinkage': 'manual', 'classifier:lda:tol': 0.0001384744065375479, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize', 'feature_preprocessor:nystroem_sampler:kernel': 'poly', 'feature_preprocessor:nystroem_sampler:n_components': 414, 'classifier:lda:shrinkage_factor': 0.2192999307296778, 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.0010364937923091088, 'feature_preprocessor:nystroem_sampler:coef0': 0.9586161969293623, 'feature_preprocessor:nystroem_sampler:degree': 4, 'feature_preprocessor:nystroem_sampler:gamma': 4.85086661900421e-05},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.060000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'lda', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'select_rates_classification', 'classifier:lda:shrinkage': 'None', 'classifier:lda:tol': 0.018038793247120707, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'power_transformer', 'feature_preprocessor:select_rates_classification:alpha': 0.3720251865973017, 'feature_preprocessor:select_rates_classification:score_func': 'mutual_info_classif'},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.040000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'random_forest', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:random_forest:bootstrap': 'True', 'classifier:random_forest:criterion': 'gini', 'classifier:random_forest:max_depth': 'None', 'classifier:random_forest:max_features': 0.5, 'classifier:random_forest:max_leaf_nodes': 'None', 'classifier:random_forest:min_impurity_decrease': 0.0, 'classifier:random_forest:min_samples_leaf': 1, 'classifier:random_forest:min_samples_split': 2, 'classifier:random_forest:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.01},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.040000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'qda', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:qda:reg_param': 0.01667385805274091, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'median', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'robust_scaler', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.24966112905354662, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max': 0.8911872287672372, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min': 0.27951285357994865},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.040000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'qda', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'select_percentile_classification', 'classifier:qda:reg_param': 0.23910875621465733, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'robust_scaler', 'feature_preprocessor:select_percentile_classification:percentile': 87.86214879612422, 'feature_preprocessor:select_percentile_classification:score_func': 'mutual_info', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.0015543757334360092, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max': 0.7530273530198002, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min': 0.0792482460648931},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.040000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'liblinear_svc', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'fast_ica', 'classifier:liblinear_svc:C': 64.64437157661465, 'classifier:liblinear_svc:dual': 'False', 'classifier:liblinear_svc:fit_intercept': 'True', 'classifier:liblinear_svc:intercept_scaling': 1, 'classifier:liblinear_svc:loss': 'squared_hinge', 'classifier:liblinear_svc:multi_class': 'ovr', 'classifier:liblinear_svc:penalty': 'l2', 'classifier:liblinear_svc:tol': 0.013399889539803811, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'none', 'feature_preprocessor:fast_ica:algorithm': 'deflation', 'feature_preprocessor:fast_ica:fun': 'exp', 'feature_preprocessor:fast_ica:whiten': 'True', 'feature_preprocessor:fast_ica:n_components': 43},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.040000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'mlp', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'random_trees_embedding', 'classifier:mlp:activation': 'relu', 'classifier:mlp:alpha': 1.040063255316952e-07, 'classifier:mlp:batch_size': 'auto', 'classifier:mlp:beta_1': 0.9, 'classifier:mlp:beta_2': 0.999, 'classifier:mlp:early_stopping': 'valid', 'classifier:mlp:epsilon': 1e-08, 'classifier:mlp:hidden_layer_depth': 1, 'classifier:mlp:learning_rate_init': 0.0018475934154961247, 'classifier:mlp:n_iter_no_change': 32, 'classifier:mlp:num_nodes_per_layer': 35, 'classifier:mlp:shuffle': 'True', 'classifier:mlp:solver': 'adam', 'classifier:mlp:tol': 0.0001, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'none', 'feature_preprocessor:random_trees_embedding:bootstrap': 'True', 'feature_preprocessor:random_trees_embedding:max_depth': 9, 'feature_preprocessor:random_trees_embedding:max_leaf_nodes': 'None', 'feature_preprocessor:random_trees_embedding:min_samples_leaf': 6, 'feature_preprocessor:random_trees_embedding:min_samples_split': 17, 'feature_preprocessor:random_trees_embedding:min_weight_fraction_leaf': 1.0, 'feature_preprocessor:random_trees_embedding:n_estimators': 32, 'classifier:mlp:validation_fraction': 0.1},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.040000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'decision_tree', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'pca', 'classifier:decision_tree:criterion': 'entropy', 'classifier:decision_tree:max_depth_factor': 1.7056696489196201, 'classifier:decision_tree:max_features': 1.0, 'classifier:decision_tree:max_leaf_nodes': 'None', 'classifier:decision_tree:min_impurity_decrease': 0.0, 'classifier:decision_tree:min_samples_leaf': 18, 'classifier:decision_tree:min_samples_split': 8, 'classifier:decision_tree:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'median', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'robust_scaler', 'feature_preprocessor:pca:keep_variance': 0.927211764000264, 'feature_preprocessor:pca:whiten': 'False', 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max': 0.9968675728600899, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min': 0.29662315421901003},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.040000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'passive_aggressive', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'kernel_pca', 'classifier:passive_aggressive:C': 0.114391364050916, 'classifier:passive_aggressive:average': 'False', 'classifier:passive_aggressive:fit_intercept': 'True', 'classifier:passive_aggressive:loss': 'hinge', 'classifier:passive_aggressive:tol': 1.3440590734654458e-05, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'normalize', 'feature_preprocessor:kernel_pca:kernel': 'sigmoid', 'feature_preprocessor:kernel_pca:n_components': 1117, 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.023878029144121318, 'feature_preprocessor:kernel_pca:coef0': -0.34481806774236556},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.020000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'gradient_boosting', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:gradient_boosting:early_stop': 'valid', 'classifier:gradient_boosting:l2_regularization': 0.009193751229756878, 'classifier:gradient_boosting:learning_rate': 0.6836522210129492, 'classifier:gradient_boosting:loss': 'auto', 'classifier:gradient_boosting:max_bins': 255, 'classifier:gradient_boosting:max_depth': 'None', 'classifier:gradient_boosting:max_leaf_nodes': 828, 'classifier:gradient_boosting:min_samples_leaf': 3, 'classifier:gradient_boosting:scoring': 'loss', 'classifier:gradient_boosting:tol': 1e-07, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'normalize', 'classifier:gradient_boosting:n_iter_no_change': 6, 'classifier:gradient_boosting:validation_fraction': 0.28817393948552533, 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.00010868227587711292},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.020000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'liblinear_svc', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'pca', 'classifier:liblinear_svc:C': 2.285306370007781, 'classifier:liblinear_svc:dual': 'False', 'classifier:liblinear_svc:fit_intercept': 'True', 'classifier:liblinear_svc:intercept_scaling': 1, 'classifier:liblinear_svc:loss': 'squared_hinge', 'classifier:liblinear_svc:multi_class': 'ovr', 'classifier:liblinear_svc:penalty': 'l2', 'classifier:liblinear_svc:tol': 0.005821344021856017, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'quantile_transformer', 'feature_preprocessor:pca:keep_variance': 0.7978382221839412, 'feature_preprocessor:pca:whiten': 'True', 'data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles': 1644, 'data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution': 'normal'},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.020000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'gradient_boosting', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'select_rates_classification', 'classifier:gradient_boosting:early_stop': 'train', 'classifier:gradient_boosting:l2_regularization': 4.520111487139148e-06, 'classifier:gradient_boosting:learning_rate': 0.12585871697357337, 'classifier:gradient_boosting:loss': 'auto', 'classifier:gradient_boosting:max_bins': 255, 'classifier:gradient_boosting:max_depth': 'None', 'classifier:gradient_boosting:max_leaf_nodes': 1894, 'classifier:gradient_boosting:min_samples_leaf': 2, 'classifier:gradient_boosting:scoring': 'loss', 'classifier:gradient_boosting:tol': 1e-07, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'median', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'robust_scaler', 'feature_preprocessor:select_rates_classification:alpha': 0.06707365823758892, 'feature_preprocessor:select_rates_classification:score_func': 'chi2', 'classifier:gradient_boosting:n_iter_no_change': 20, 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.0014466154794086487, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max': 0.7104048919054863, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min': 0.2345103479872152, 'feature_preprocessor:select_rates_classification:mode': 'fpr'},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.020000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'lda', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'kernel_pca', 'classifier:lda:shrinkage': 'auto', 'classifier:lda:tol': 0.014034416932373166, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'median', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'power_transformer', 'feature_preprocessor:kernel_pca:kernel': 'cosine', 'feature_preprocessor:kernel_pca:n_components': 285, 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.0074025063474275335},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.020000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'lda', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'liblinear_svc_preprocessor', 'classifier:lda:shrinkage': 'None', 'classifier:lda:tol': 0.035543606049804645, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'none', 'feature_preprocessor:liblinear_svc_preprocessor:C': 4.279105226196228, 'feature_preprocessor:liblinear_svc_preprocessor:dual': 'False', 'feature_preprocessor:liblinear_svc_preprocessor:fit_intercept': 'True', 'feature_preprocessor:liblinear_svc_preprocessor:intercept_scaling': 1, 'feature_preprocessor:liblinear_svc_preprocessor:loss': 'squared_hinge', 'feature_preprocessor:liblinear_svc_preprocessor:multi_class': 'ovr', 'feature_preprocessor:liblinear_svc_preprocessor:penalty': 'l1', 'feature_preprocessor:liblinear_svc_preprocessor:tol': 4.410651487451908e-05},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.020000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'mlp', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'select_rates_classification', 'classifier:mlp:activation': 'tanh', 'classifier:mlp:alpha': 4.421745686388309e-05, 'classifier:mlp:batch_size': 'auto', 'classifier:mlp:beta_1': 0.9, 'classifier:mlp:beta_2': 0.999, 'classifier:mlp:early_stopping': 'train', 'classifier:mlp:epsilon': 1e-08, 'classifier:mlp:hidden_layer_depth': 3, 'classifier:mlp:learning_rate_init': 0.0005485701552075972, 'classifier:mlp:n_iter_no_change': 32, 'classifier:mlp:num_nodes_per_layer': 36, 'classifier:mlp:shuffle': 'True', 'classifier:mlp:solver': 'adam', 'classifier:mlp:tol': 0.0001, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'minmax', 'feature_preprocessor:select_rates_classification:alpha': 0.44455023733729254, 'feature_preprocessor:select_rates_classification:score_func': 'mutual_info_classif'},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.020000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'mlp', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'extra_trees_preproc_for_classification', 'classifier:mlp:activation': 'relu', 'classifier:mlp:alpha': 1.4882838781863879e-06, 'classifier:mlp:batch_size': 'auto', 'classifier:mlp:beta_1': 0.9, 'classifier:mlp:beta_2': 0.999, 'classifier:mlp:early_stopping': 'train', 'classifier:mlp:epsilon': 1e-08, 'classifier:mlp:hidden_layer_depth': 2, 'classifier:mlp:learning_rate_init': 0.06290764761689918, 'classifier:mlp:n_iter_no_change': 32, 'classifier:mlp:num_nodes_per_layer': 23, 'classifier:mlp:shuffle': 'True', 'classifier:mlp:solver': 'adam', 'classifier:mlp:tol': 0.0001, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'median', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'quantile_transformer', 'feature_preprocessor:extra_trees_preproc_for_classification:bootstrap': 'False', 'feature_preprocessor:extra_trees_preproc_for_classification:criterion': 'gini', 'feature_preprocessor:extra_trees_preproc_for_classification:max_depth': 'None', 'feature_preprocessor:extra_trees_preproc_for_classification:max_features': 0.9565048776795676, 'feature_preprocessor:extra_trees_preproc_for_classification:max_leaf_nodes': 'None', 'feature_preprocessor:extra_trees_preproc_for_classification:min_impurity_decrease': 0.0, 'feature_preprocessor:extra_trees_preproc_for_classification:min_samples_leaf': 7, 'feature_preprocessor:extra_trees_preproc_for_classification:min_samples_split': 13, 'feature_preprocessor:extra_trees_preproc_for_classification:min_weight_fraction_leaf': 0.0, 'feature_preprocessor:extra_trees_preproc_for_classification:n_estimators': 100, 'data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles': 273, 'data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution': 'normal'},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.020000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'decision_tree', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'random_trees_embedding', 'classifier:decision_tree:criterion': 'gini', 'classifier:decision_tree:max_depth_factor': 0.6228726042266908, 'classifier:decision_tree:max_features': 1.0, 'classifier:decision_tree:max_leaf_nodes': 'None', 'classifier:decision_tree:min_impurity_decrease': 0.0, 'classifier:decision_tree:min_samples_leaf': 16, 'classifier:decision_tree:min_samples_split': 19, 'classifier:decision_tree:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'quantile_transformer', 'feature_preprocessor:random_trees_embedding:bootstrap': 'False', 'feature_preprocessor:random_trees_embedding:max_depth': 2, 'feature_preprocessor:random_trees_embedding:max_leaf_nodes': 'None', 'feature_preprocessor:random_trees_embedding:min_samples_leaf': 17, 'feature_preprocessor:random_trees_embedding:min_samples_split': 12, 'feature_preprocessor:random_trees_embedding:min_weight_fraction_leaf': 1.0, 'feature_preprocessor:random_trees_embedding:n_estimators': 38, 'data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles': 606, 'data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution': 'uniform'},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.020000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'decision_tree', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'pca', 'classifier:decision_tree:criterion': 'entropy', 'classifier:decision_tree:max_depth_factor': 1.1387664107898294, 'classifier:decision_tree:max_features': 1.0, 'classifier:decision_tree:max_leaf_nodes': 'None', 'classifier:decision_tree:min_impurity_decrease': 0.0, 'classifier:decision_tree:min_samples_leaf': 8, 'classifier:decision_tree:min_samples_split': 9, 'classifier:decision_tree:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'none', 'feature_preprocessor:pca:keep_variance': 0.9503612101678187, 'feature_preprocessor:pca:whiten': 'True'},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.020000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'bernoulli_nb', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'select_percentile_classification', 'classifier:bernoulli_nb:alpha': 8.843515731149308, 'classifier:bernoulli_nb:fit_prior': 'True', 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'robust_scaler', 'feature_preprocessor:select_percentile_classification:percentile': 50.58615414999356, 'feature_preprocessor:select_percentile_classification:score_func': 'chi2', 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max': 0.7852218490061461, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min': 0.15162626175424537},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    ]
+    {   2: {   'balancing': Balancing(random_state=1),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc787e05a30>,
+               'cost': 0.028368794326241176,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7861841f0>,
+               'ensemble_weight': 0.04,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78a86d100>,
+               'model_id': 2,
+               'rank': 3,
+               'sklearn_classifier': RandomForestClassifier(max_features=5, n_estimators=512, n_jobs=1,
+                           random_state=1, warm_start=True)},
+        4: {   'balancing': Balancing(random_state=1),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc787302940>,
+               'cost': 0.05673758865248224,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7876620a0>,
+               'ensemble_weight': 0.02,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc786554580>,
+               'model_id': 4,
+               'rank': 15,
+               'sklearn_classifier': MLPClassifier(alpha=1.0400632553169558e-07, beta_1=0.999, beta_2=0.9,
+                  early_stopping=True, hidden_layer_sizes=(35,),
+                  learning_rate_init=0.0018475934154961215, max_iter=32,
+                  n_iter_no_change=32, random_state=1, verbose=0, warm_start=True)},
+        5: {   'balancing': Balancing(random_state=1),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc787374fd0>,
+               'cost': 0.11347517730496459,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78706ef70>,
+               'ensemble_weight': 0.02,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc787374400>,
+               'model_id': 5,
+               'rank': 26,
+               'sklearn_classifier': BernoulliNB(alpha=0.010232681621243814)},
+        6: {   'balancing': Balancing(random_state=1),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc787748400>,
+               'cost': 0.049645390070921946,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78582ee20>,
+               'ensemble_weight': 0.06,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc787748250>,
+               'model_id': 6,
+               'rank': 11,
+               'sklearn_classifier': LinearDiscriminantAnalysis(tol=1.3202262453017803e-05)},
+        7: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc787aa5220>,
+               'cost': 0.099290780141844,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78af89d60>,
+               'ensemble_weight': 0.02,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc787a8deb0>,
+               'model_id': 7,
+               'rank': 24,
+               'sklearn_classifier': BernoulliNB(alpha=8.843515731149308)},
+        9: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc787793460>,
+               'cost': 0.07801418439716312,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78ff456d0>,
+               'ensemble_weight': 0.02,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc787788ca0>,
+               'model_id': 9,
+               'rank': 20,
+               'sklearn_classifier': BernoulliNB(alpha=0.04662763358613446, fit_prior=False)},
+        10: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc787e05310>,
+                'cost': 0.007092198581560294,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78509f9a0>,
+                'ensemble_weight': 0.08,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78bd13160>,
+                'model_id': 10,
+                'rank': 1,
+                'sklearn_classifier': QuadraticDiscriminantAnalysis(reg_param=0.45597150689539434)},
+        11: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc79068b160>,
+                'cost': 0.028368794326241176,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7850a8850>,
+                'ensemble_weight': 0.06,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78a791250>,
+                'model_id': 11,
+                'rank': 4,
+                'sklearn_classifier': QuadraticDiscriminantAnalysis(reg_param=0.0995499108996053)},
+        12: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc787ba29a0>,
+                'cost': 0.049645390070921946,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78580d130>,
+                'ensemble_weight': 0.04,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc787ba2040>,
+                'model_id': 12,
+                'rank': 12,
+                'sklearn_classifier': LinearDiscriminantAnalysis(shrinkage='auto', solver='lsqr',
+                               tol=0.014034416932373166)},
+        13: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78fa980a0>,
+                'cost': 0.021276595744680882,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc787888340>,
+                'ensemble_weight': 0.06,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78fa98670>,
+                'model_id': 13,
+                'rank': 2,
+                'sklearn_classifier': LinearDiscriminantAnalysis(shrinkage=0.2192999307296778, solver='lsqr',
+                               tol=0.0001384744065375479)},
+        14: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc787e47580>,
+                'cost': 0.05673758865248224,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc784bbe130>,
+                'ensemble_weight': 0.04,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc787e47400>,
+                'model_id': 14,
+                'rank': 16,
+                'sklearn_classifier': LinearDiscriminantAnalysis(tol=0.035543606049804645)},
+        16: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78af79460>,
+                'cost': 0.08510638297872342,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc79068b340>,
+                'ensemble_weight': 0.04,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78af90700>,
+                'model_id': 16,
+                'rank': 21,
+                'sklearn_classifier': DecisionTreeClassifier(criterion='entropy', max_depth=1, min_samples_leaf=8,
+                           min_samples_split=9, random_state=1)},
+        17: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78b9260a0>,
+                'cost': 0.04255319148936165,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7865df100>,
+                'ensemble_weight': 0.04,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78b926c40>,
+                'model_id': 17,
+                'rank': 10,
+                'sklearn_classifier': LinearSVC(C=64.64437157661465, dual=False, intercept_scaling=1.0,
+              random_state=1, tol=0.013399889539803811)},
+        18: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc786bff640>,
+                'cost': 0.099290780141844,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc786aaaf10>,
+                'ensemble_weight': 0.02,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc786bd1b50>,
+                'model_id': 18,
+                'rank': 25,
+                'sklearn_classifier': PassiveAggressiveClassifier(C=0.114391364050916, max_iter=128, random_state=1,
+                                tol=1.3440590734654458e-05, warm_start=True)},
+        19: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78f986610>,
+                'cost': 0.03546099290780147,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78f84d1f0>,
+                'ensemble_weight': 0.02,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc786854070>,
+                'model_id': 19,
+                'rank': 8,
+                'sklearn_classifier': QuadraticDiscriminantAnalysis(reg_param=0.01667385805274091)},
+        21: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78f69fd90>,
+                'cost': 0.028368794326241176,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78c43e280>,
+                'ensemble_weight': 0.06,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c94d1f0>,
+                'model_id': 21,
+                'rank': 5,
+                'sklearn_classifier': LinearSVC(C=2.285306370007781, dual=False, intercept_scaling=1.0,
+              random_state=1, tol=0.005821344021856017)},
+        24: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc7903bddf0>,
+                'cost': 0.06382978723404253,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc786581c40>,
+                'ensemble_weight': 0.02,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c9bf490>,
+                'model_id': 24,
+                'rank': 17,
+                'sklearn_classifier': SGDClassifier(alpha=5.189696963506304e-05, eta0=0.007608275318572964,
+                  learning_rate='constant', loss='perceptron', max_iter=64,
+                  random_state=1, tol=0.0005024616279807072, warm_start=True)},
+        25: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc786149370>,
+                'cost': 0.08510638297872342,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78770be20>,
+                'ensemble_weight': 0.02,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc786c845e0>,
+                'model_id': 25,
+                'rank': 22,
+                'sklearn_classifier': PassiveAggressiveClassifier(C=1.4622203211010016e-05, average=True,
+                                max_iter=128, random_state=1,
+                                tol=0.0017964371726429177, warm_start=True)},
+        26: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78ff433d0>,
+                'cost': 0.06382978723404253,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc787e52070>,
+                'ensemble_weight': 0.06,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78ff43730>,
+                'model_id': 26,
+                'rank': 18,
+                'sklearn_classifier': AdaBoostClassifier(algorithm='SAMME',
+                       base_estimator=DecisionTreeClassifier(max_depth=8),
+                       learning_rate=1.6587261797143242, n_estimators=368,
+                       random_state=1)},
+        27: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc786149910>,
+                'cost': 0.09219858156028371,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78face370>,
+                'ensemble_weight': 0.02,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc7861490d0>,
+                'model_id': 27,
+                'rank': 23,
+                'sklearn_classifier': DecisionTreeClassifier(criterion='entropy', max_depth=4, min_samples_leaf=18,
+                           min_samples_split=8, random_state=1)},
+        30: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78c9bff70>,
+                'cost': 0.028368794326241176,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78a25e7c0>,
+                'ensemble_weight': 0.06,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78a4be100>,
+                'model_id': 30,
+                'rank': 6,
+                'sklearn_classifier': RandomForestClassifier(bootstrap=False, criterion='entropy', max_features=1,
+                           min_samples_leaf=10, min_samples_split=18,
+                           n_estimators=512, n_jobs=1, random_state=1,
+                           warm_start=True)},
+        31: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc7877cdaf0>,
+                'cost': 0.049645390070921946,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78733d670>,
+                'ensemble_weight': 0.02,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc7877cd460>,
+                'model_id': 31,
+                'rank': 13,
+                'sklearn_classifier': LinearDiscriminantAnalysis(tol=0.018038793247120707)},
+        33: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc786796cd0>,
+                'cost': 0.03546099290780147,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc785b99af0>,
+                'ensemble_weight': 0.04,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c5839d0>,
+                'model_id': 33,
+                'rank': 9,
+                'sklearn_classifier': HistGradientBoostingClassifier(early_stopping=True,
+                                   l2_regularization=4.5201114871391565e-06,
+                                   learning_rate=0.12585871697357337, max_iter=256,
+                                   max_leaf_nodes=1894, min_samples_leaf=2,
+                                   n_iter_no_change=20, random_state=1,
+                                   validation_fraction=None, warm_start=True)},
+        35: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78b257790>,
+                'cost': 0.049645390070921946,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78776fac0>,
+                'ensemble_weight': 0.02,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc784bd90d0>,
+                'model_id': 35,
+                'rank': 14,
+                'sklearn_classifier': KNeighborsClassifier(n_neighbors=2)},
+        36: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc786806670>,
+                'cost': 0.07092198581560283,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78c752cd0>,
+                'ensemble_weight': 0.06,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78a86d9a0>,
+                'model_id': 36,
+                'rank': 19,
+                'sklearn_classifier': GaussianNB()},
+        37: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78aac0730>,
+                'cost': 0.028368794326241176,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78f83e5e0>,
+                'ensemble_weight': 0.04,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78b6561c0>,
+                'model_id': 37,
+                'rank': 7,
+                'sklearn_classifier': LinearDiscriminantAnalysis(shrinkage=0.32361629219618826, solver='lsqr',
+                               tol=3.142032801716559e-05)}}
     auto-sklearn results:
       Dataset name: breast_cancer
       Metric: accuracy
       Best validation score: 0.992908
-      Number of target algorithm runs: 33
-      Number of successful target algorithm runs: 31
+      Number of target algorithm runs: 36
+      Number of successful target algorithm runs: 35
       Number of crashed target algorithm runs: 1
-      Number of target algorithms that exceeded the time limit: 1
+      Number of target algorithms that exceeded the time limit: 0
       Number of target algorithms that exceeded the memory limit: 0
 
-    Accuracy score 0.9440559440559441
+    Accuracy score 0.958041958041958
 
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 87-89
+.. GENERATED FROM PYTHON SOURCE LINES 88-90
 
 Fit a classifier using Random Search
 ====================================
 
-.. GENERATED FROM PYTHON SOURCE LINES 89-140
+.. GENERATED FROM PYTHON SOURCE LINES 90-141
 
 .. code-block:: default
 
@@ -396,7 +476,7 @@ Fit a classifier using Random Search
     print('Results for random search.')
 
     # Print the final ensemble constructed by auto-sklearn via random search.
-    print(automl.show_models())
+    pprint(automl.show_models(), indent=4)
 
     # Print statistics about the auto-sklearn run such as number of
     # iterations, number of models failed with a time out.
@@ -416,199 +496,304 @@ Fit a classifier using Random Search
 
     ################################################################################
     Results for random search.
-    [(0.120000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'random_forest', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'select_rates_classification', 'classifier:random_forest:bootstrap': 'False', 'classifier:random_forest:criterion': 'entropy', 'classifier:random_forest:max_depth': 'None', 'classifier:random_forest:max_features': 0.13366756044911932, 'classifier:random_forest:max_leaf_nodes': 'None', 'classifier:random_forest:min_impurity_decrease': 0.0, 'classifier:random_forest:min_samples_leaf': 10, 'classifier:random_forest:min_samples_split': 18, 'classifier:random_forest:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'robust_scaler', 'feature_preprocessor:select_rates_classification:alpha': 0.06835899987255477, 'feature_preprocessor:select_rates_classification:score_func': 'chi2', 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max': 0.791639810538127, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min': 0.2949722712387337, 'feature_preprocessor:select_rates_classification:mode': 'fwe'},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.100000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'qda', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:qda:reg_param': 0.45597150689539434, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize'},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.100000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'adaboost', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'fast_ica', 'classifier:adaboost:algorithm': 'SAMME', 'classifier:adaboost:learning_rate': 1.6587261797143242, 'classifier:adaboost:max_depth': 8, 'classifier:adaboost:n_estimators': 368, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'median', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'normalize', 'feature_preprocessor:fast_ica:algorithm': 'parallel', 'feature_preprocessor:fast_ica:fun': 'cube', 'feature_preprocessor:fast_ica:whiten': 'True', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.00031132126574883666, 'feature_preprocessor:fast_ica:n_components': 530},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.080000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'passive_aggressive', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'extra_trees_preproc_for_classification', 'classifier:passive_aggressive:C': 1.4622203211010016e-05, 'classifier:passive_aggressive:average': 'True', 'classifier:passive_aggressive:fit_intercept': 'True', 'classifier:passive_aggressive:loss': 'hinge', 'classifier:passive_aggressive:tol': 0.0017964371726429177, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'power_transformer', 'feature_preprocessor:extra_trees_preproc_for_classification:bootstrap': 'False', 'feature_preprocessor:extra_trees_preproc_for_classification:criterion': 'entropy', 'feature_preprocessor:extra_trees_preproc_for_classification:max_depth': 'None', 'feature_preprocessor:extra_trees_preproc_for_classification:max_features': 0.10392737384710127, 'feature_preprocessor:extra_trees_preproc_for_classification:max_leaf_nodes': 'None', 'feature_preprocessor:extra_trees_preproc_for_classification:min_impurity_decrease': 0.0, 'feature_preprocessor:extra_trees_preproc_for_classification:min_samples_leaf': 11, 'feature_preprocessor:extra_trees_preproc_for_classification:min_samples_split': 9, 'feature_preprocessor:extra_trees_preproc_for_classification:min_weight_fraction_leaf': 0.0, 'feature_preprocessor:extra_trees_preproc_for_classification:n_estimators': 100},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.060000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'lda', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'nystroem_sampler', 'classifier:lda:shrinkage': 'manual', 'classifier:lda:tol': 0.0001384744065375479, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize', 'feature_preprocessor:nystroem_sampler:kernel': 'poly', 'feature_preprocessor:nystroem_sampler:n_components': 414, 'classifier:lda:shrinkage_factor': 0.2192999307296778, 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.0010364937923091088, 'feature_preprocessor:nystroem_sampler:coef0': 0.9586161969293623, 'feature_preprocessor:nystroem_sampler:degree': 4, 'feature_preprocessor:nystroem_sampler:gamma': 4.85086661900421e-05},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.060000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'lda', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'select_rates_classification', 'classifier:lda:shrinkage': 'None', 'classifier:lda:tol': 0.018038793247120707, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'power_transformer', 'feature_preprocessor:select_rates_classification:alpha': 0.3720251865973017, 'feature_preprocessor:select_rates_classification:score_func': 'mutual_info_classif'},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.040000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'random_forest', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:random_forest:bootstrap': 'True', 'classifier:random_forest:criterion': 'gini', 'classifier:random_forest:max_depth': 'None', 'classifier:random_forest:max_features': 0.5, 'classifier:random_forest:max_leaf_nodes': 'None', 'classifier:random_forest:min_impurity_decrease': 0.0, 'classifier:random_forest:min_samples_leaf': 1, 'classifier:random_forest:min_samples_split': 2, 'classifier:random_forest:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.01},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.040000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'qda', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:qda:reg_param': 0.01667385805274091, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'median', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'robust_scaler', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.24966112905354662, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max': 0.8911872287672372, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min': 0.27951285357994865},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.040000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'qda', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'select_percentile_classification', 'classifier:qda:reg_param': 0.23910875621465733, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'robust_scaler', 'feature_preprocessor:select_percentile_classification:percentile': 87.86214879612422, 'feature_preprocessor:select_percentile_classification:score_func': 'mutual_info', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.0015543757334360092, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max': 0.7530273530198002, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min': 0.0792482460648931},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.040000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'liblinear_svc', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'fast_ica', 'classifier:liblinear_svc:C': 64.64437157661465, 'classifier:liblinear_svc:dual': 'False', 'classifier:liblinear_svc:fit_intercept': 'True', 'classifier:liblinear_svc:intercept_scaling': 1, 'classifier:liblinear_svc:loss': 'squared_hinge', 'classifier:liblinear_svc:multi_class': 'ovr', 'classifier:liblinear_svc:penalty': 'l2', 'classifier:liblinear_svc:tol': 0.013399889539803811, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'none', 'feature_preprocessor:fast_ica:algorithm': 'deflation', 'feature_preprocessor:fast_ica:fun': 'exp', 'feature_preprocessor:fast_ica:whiten': 'True', 'feature_preprocessor:fast_ica:n_components': 43},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.040000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'mlp', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'random_trees_embedding', 'classifier:mlp:activation': 'relu', 'classifier:mlp:alpha': 1.040063255316952e-07, 'classifier:mlp:batch_size': 'auto', 'classifier:mlp:beta_1': 0.9, 'classifier:mlp:beta_2': 0.999, 'classifier:mlp:early_stopping': 'valid', 'classifier:mlp:epsilon': 1e-08, 'classifier:mlp:hidden_layer_depth': 1, 'classifier:mlp:learning_rate_init': 0.0018475934154961247, 'classifier:mlp:n_iter_no_change': 32, 'classifier:mlp:num_nodes_per_layer': 35, 'classifier:mlp:shuffle': 'True', 'classifier:mlp:solver': 'adam', 'classifier:mlp:tol': 0.0001, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'none', 'feature_preprocessor:random_trees_embedding:bootstrap': 'True', 'feature_preprocessor:random_trees_embedding:max_depth': 9, 'feature_preprocessor:random_trees_embedding:max_leaf_nodes': 'None', 'feature_preprocessor:random_trees_embedding:min_samples_leaf': 6, 'feature_preprocessor:random_trees_embedding:min_samples_split': 17, 'feature_preprocessor:random_trees_embedding:min_weight_fraction_leaf': 1.0, 'feature_preprocessor:random_trees_embedding:n_estimators': 32, 'classifier:mlp:validation_fraction': 0.1},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.040000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'decision_tree', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'pca', 'classifier:decision_tree:criterion': 'entropy', 'classifier:decision_tree:max_depth_factor': 1.7056696489196201, 'classifier:decision_tree:max_features': 1.0, 'classifier:decision_tree:max_leaf_nodes': 'None', 'classifier:decision_tree:min_impurity_decrease': 0.0, 'classifier:decision_tree:min_samples_leaf': 18, 'classifier:decision_tree:min_samples_split': 8, 'classifier:decision_tree:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'median', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'robust_scaler', 'feature_preprocessor:pca:keep_variance': 0.927211764000264, 'feature_preprocessor:pca:whiten': 'False', 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max': 0.9968675728600899, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min': 0.29662315421901003},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.040000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'passive_aggressive', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'kernel_pca', 'classifier:passive_aggressive:C': 0.114391364050916, 'classifier:passive_aggressive:average': 'False', 'classifier:passive_aggressive:fit_intercept': 'True', 'classifier:passive_aggressive:loss': 'hinge', 'classifier:passive_aggressive:tol': 1.3440590734654458e-05, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'normalize', 'feature_preprocessor:kernel_pca:kernel': 'sigmoid', 'feature_preprocessor:kernel_pca:n_components': 1117, 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.023878029144121318, 'feature_preprocessor:kernel_pca:coef0': -0.34481806774236556},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.020000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'gradient_boosting', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:gradient_boosting:early_stop': 'valid', 'classifier:gradient_boosting:l2_regularization': 0.009193751229756878, 'classifier:gradient_boosting:learning_rate': 0.6836522210129492, 'classifier:gradient_boosting:loss': 'auto', 'classifier:gradient_boosting:max_bins': 255, 'classifier:gradient_boosting:max_depth': 'None', 'classifier:gradient_boosting:max_leaf_nodes': 828, 'classifier:gradient_boosting:min_samples_leaf': 3, 'classifier:gradient_boosting:scoring': 'loss', 'classifier:gradient_boosting:tol': 1e-07, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'normalize', 'classifier:gradient_boosting:n_iter_no_change': 6, 'classifier:gradient_boosting:validation_fraction': 0.28817393948552533, 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.00010868227587711292},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.020000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'liblinear_svc', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'pca', 'classifier:liblinear_svc:C': 2.285306370007781, 'classifier:liblinear_svc:dual': 'False', 'classifier:liblinear_svc:fit_intercept': 'True', 'classifier:liblinear_svc:intercept_scaling': 1, 'classifier:liblinear_svc:loss': 'squared_hinge', 'classifier:liblinear_svc:multi_class': 'ovr', 'classifier:liblinear_svc:penalty': 'l2', 'classifier:liblinear_svc:tol': 0.005821344021856017, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'quantile_transformer', 'feature_preprocessor:pca:keep_variance': 0.7978382221839412, 'feature_preprocessor:pca:whiten': 'True', 'data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles': 1644, 'data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution': 'normal'},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.020000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'gradient_boosting', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'select_rates_classification', 'classifier:gradient_boosting:early_stop': 'train', 'classifier:gradient_boosting:l2_regularization': 4.520111487139148e-06, 'classifier:gradient_boosting:learning_rate': 0.12585871697357337, 'classifier:gradient_boosting:loss': 'auto', 'classifier:gradient_boosting:max_bins': 255, 'classifier:gradient_boosting:max_depth': 'None', 'classifier:gradient_boosting:max_leaf_nodes': 1894, 'classifier:gradient_boosting:min_samples_leaf': 2, 'classifier:gradient_boosting:scoring': 'loss', 'classifier:gradient_boosting:tol': 1e-07, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'median', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'robust_scaler', 'feature_preprocessor:select_rates_classification:alpha': 0.06707365823758892, 'feature_preprocessor:select_rates_classification:score_func': 'chi2', 'classifier:gradient_boosting:n_iter_no_change': 20, 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.0014466154794086487, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max': 0.7104048919054863, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min': 0.2345103479872152, 'feature_preprocessor:select_rates_classification:mode': 'fpr'},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.020000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'lda', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'kernel_pca', 'classifier:lda:shrinkage': 'auto', 'classifier:lda:tol': 0.014034416932373166, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'median', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'power_transformer', 'feature_preprocessor:kernel_pca:kernel': 'cosine', 'feature_preprocessor:kernel_pca:n_components': 285, 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.0074025063474275335},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.020000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'lda', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'liblinear_svc_preprocessor', 'classifier:lda:shrinkage': 'None', 'classifier:lda:tol': 0.035543606049804645, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'none', 'feature_preprocessor:liblinear_svc_preprocessor:C': 4.279105226196228, 'feature_preprocessor:liblinear_svc_preprocessor:dual': 'False', 'feature_preprocessor:liblinear_svc_preprocessor:fit_intercept': 'True', 'feature_preprocessor:liblinear_svc_preprocessor:intercept_scaling': 1, 'feature_preprocessor:liblinear_svc_preprocessor:loss': 'squared_hinge', 'feature_preprocessor:liblinear_svc_preprocessor:multi_class': 'ovr', 'feature_preprocessor:liblinear_svc_preprocessor:penalty': 'l1', 'feature_preprocessor:liblinear_svc_preprocessor:tol': 4.410651487451908e-05},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.020000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'mlp', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'select_rates_classification', 'classifier:mlp:activation': 'tanh', 'classifier:mlp:alpha': 4.421745686388309e-05, 'classifier:mlp:batch_size': 'auto', 'classifier:mlp:beta_1': 0.9, 'classifier:mlp:beta_2': 0.999, 'classifier:mlp:early_stopping': 'train', 'classifier:mlp:epsilon': 1e-08, 'classifier:mlp:hidden_layer_depth': 3, 'classifier:mlp:learning_rate_init': 0.0005485701552075972, 'classifier:mlp:n_iter_no_change': 32, 'classifier:mlp:num_nodes_per_layer': 36, 'classifier:mlp:shuffle': 'True', 'classifier:mlp:solver': 'adam', 'classifier:mlp:tol': 0.0001, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'minmax', 'feature_preprocessor:select_rates_classification:alpha': 0.44455023733729254, 'feature_preprocessor:select_rates_classification:score_func': 'mutual_info_classif'},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.020000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'mlp', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'extra_trees_preproc_for_classification', 'classifier:mlp:activation': 'relu', 'classifier:mlp:alpha': 1.4882838781863879e-06, 'classifier:mlp:batch_size': 'auto', 'classifier:mlp:beta_1': 0.9, 'classifier:mlp:beta_2': 0.999, 'classifier:mlp:early_stopping': 'train', 'classifier:mlp:epsilon': 1e-08, 'classifier:mlp:hidden_layer_depth': 2, 'classifier:mlp:learning_rate_init': 0.06290764761689918, 'classifier:mlp:n_iter_no_change': 32, 'classifier:mlp:num_nodes_per_layer': 23, 'classifier:mlp:shuffle': 'True', 'classifier:mlp:solver': 'adam', 'classifier:mlp:tol': 0.0001, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'median', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'quantile_transformer', 'feature_preprocessor:extra_trees_preproc_for_classification:bootstrap': 'False', 'feature_preprocessor:extra_trees_preproc_for_classification:criterion': 'gini', 'feature_preprocessor:extra_trees_preproc_for_classification:max_depth': 'None', 'feature_preprocessor:extra_trees_preproc_for_classification:max_features': 0.9565048776795676, 'feature_preprocessor:extra_trees_preproc_for_classification:max_leaf_nodes': 'None', 'feature_preprocessor:extra_trees_preproc_for_classification:min_impurity_decrease': 0.0, 'feature_preprocessor:extra_trees_preproc_for_classification:min_samples_leaf': 7, 'feature_preprocessor:extra_trees_preproc_for_classification:min_samples_split': 13, 'feature_preprocessor:extra_trees_preproc_for_classification:min_weight_fraction_leaf': 0.0, 'feature_preprocessor:extra_trees_preproc_for_classification:n_estimators': 100, 'data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles': 273, 'data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution': 'normal'},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.020000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'decision_tree', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'random_trees_embedding', 'classifier:decision_tree:criterion': 'gini', 'classifier:decision_tree:max_depth_factor': 0.6228726042266908, 'classifier:decision_tree:max_features': 1.0, 'classifier:decision_tree:max_leaf_nodes': 'None', 'classifier:decision_tree:min_impurity_decrease': 0.0, 'classifier:decision_tree:min_samples_leaf': 16, 'classifier:decision_tree:min_samples_split': 19, 'classifier:decision_tree:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'quantile_transformer', 'feature_preprocessor:random_trees_embedding:bootstrap': 'False', 'feature_preprocessor:random_trees_embedding:max_depth': 2, 'feature_preprocessor:random_trees_embedding:max_leaf_nodes': 'None', 'feature_preprocessor:random_trees_embedding:min_samples_leaf': 17, 'feature_preprocessor:random_trees_embedding:min_samples_split': 12, 'feature_preprocessor:random_trees_embedding:min_weight_fraction_leaf': 1.0, 'feature_preprocessor:random_trees_embedding:n_estimators': 38, 'data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles': 606, 'data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution': 'uniform'},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.020000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'decision_tree', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'pca', 'classifier:decision_tree:criterion': 'entropy', 'classifier:decision_tree:max_depth_factor': 1.1387664107898294, 'classifier:decision_tree:max_features': 1.0, 'classifier:decision_tree:max_leaf_nodes': 'None', 'classifier:decision_tree:min_impurity_decrease': 0.0, 'classifier:decision_tree:min_samples_leaf': 8, 'classifier:decision_tree:min_samples_split': 9, 'classifier:decision_tree:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'none', 'feature_preprocessor:pca:keep_variance': 0.9503612101678187, 'feature_preprocessor:pca:whiten': 'True'},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.020000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'bernoulli_nb', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'select_percentile_classification', 'classifier:bernoulli_nb:alpha': 8.843515731149308, 'classifier:bernoulli_nb:fit_prior': 'True', 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'robust_scaler', 'feature_preprocessor:select_percentile_classification:percentile': 50.58615414999356, 'feature_preprocessor:select_percentile_classification:score_func': 'chi2', 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max': 0.7852218490061461, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min': 0.15162626175424537},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    ]
+    {   3: {   'balancing': Balancing(random_state=1),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc787842790>,
+               'cost': 0.05673758865248224,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78fa984f0>,
+               'ensemble_weight': 0.02,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc7876ba3d0>,
+               'model_id': 3,
+               'rank': 16,
+               'sklearn_classifier': MLPClassifier(alpha=1.0400632553169558e-07, beta_1=0.999, beta_2=0.9,
+                  early_stopping=True, hidden_layer_sizes=(35,),
+                  learning_rate_init=0.0018475934154961215, max_iter=32,
+                  n_iter_no_change=32, random_state=1, verbose=0, warm_start=True)},
+        4: {   'balancing': Balancing(random_state=1),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc785c0eb80>,
+               'cost': 0.028368794326241176,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7877aefa0>,
+               'ensemble_weight': 0.04,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc785c0eb50>,
+               'model_id': 4,
+               'rank': 3,
+               'sklearn_classifier': RandomForestClassifier(max_features=5, n_estimators=512, n_jobs=1,
+                           random_state=1, warm_start=True)},
+        5: {   'balancing': Balancing(random_state=1),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc785b6ac70>,
+               'cost': 0.11347517730496459,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78679fe50>,
+               'ensemble_weight': 0.02,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc785b6a0a0>,
+               'model_id': 5,
+               'rank': 28,
+               'sklearn_classifier': BernoulliNB(alpha=0.010232681621243814)},
+        6: {   'balancing': Balancing(random_state=1),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78c9c7700>,
+               'cost': 0.049645390070921946,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78a7f2c70>,
+               'ensemble_weight': 0.08,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78a605820>,
+               'model_id': 6,
+               'rank': 12,
+               'sklearn_classifier': LinearDiscriminantAnalysis(tol=1.3202262453017803e-05)},
+        7: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78700b100>,
+               'cost': 0.099290780141844,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7877938b0>,
+               'ensemble_weight': 0.02,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc787772ca0>,
+               'model_id': 7,
+               'rank': 26,
+               'sklearn_classifier': BernoulliNB(alpha=8.843515731149308)},
+        8: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78662d580>,
+               'cost': 0.08510638297872342,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc784a023a0>,
+               'ensemble_weight': 0.02,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78b94c490>,
+               'model_id': 8,
+               'rank': 22,
+               'sklearn_classifier': SGDClassifier(alpha=3.816570529036241e-06, eta0=0.043879389796003546,
+                  l1_ratio=0.003142456770095119, learning_rate='constant',
+                  loss='squared_hinge', max_iter=32, penalty='elasticnet',
+                  random_state=1, tol=0.000896391087588558, warm_start=True)},
+        9: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78657f370>,
+               'cost': 0.07801418439716312,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc787ae89d0>,
+               'ensemble_weight': 0.02,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78657f1c0>,
+               'model_id': 9,
+               'rank': 21,
+               'sklearn_classifier': BernoulliNB(alpha=0.04662763358613446, fit_prior=False)},
+        10: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78af797f0>,
+                'cost': 0.007092198581560294,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc787788bb0>,
+                'ensemble_weight': 0.06,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78af79550>,
+                'model_id': 10,
+                'rank': 1,
+                'sklearn_classifier': QuadraticDiscriminantAnalysis(reg_param=0.45597150689539434)},
+        11: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78c8f8bb0>,
+                'cost': 0.028368794326241176,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78779a040>,
+                'ensemble_weight': 0.04,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c8f8a60>,
+                'model_id': 11,
+                'rank': 4,
+                'sklearn_classifier': QuadraticDiscriminantAnalysis(reg_param=0.0995499108996053)},
+        12: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78a25eb20>,
+                'cost': 0.049645390070921946,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78698af70>,
+                'ensemble_weight': 0.02,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c43e5b0>,
+                'model_id': 12,
+                'rank': 13,
+                'sklearn_classifier': LinearDiscriminantAnalysis(shrinkage='auto', solver='lsqr',
+                               tol=0.014034416932373166)},
+        13: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78fad68e0>,
+                'cost': 0.021276595744680882,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78faf4670>,
+                'ensemble_weight': 0.06,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78fad64f0>,
+                'model_id': 13,
+                'rank': 2,
+                'sklearn_classifier': LinearDiscriminantAnalysis(shrinkage=0.2192999307296778, solver='lsqr',
+                               tol=0.0001384744065375479)},
+        14: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78c5d43d0>,
+                'cost': 0.05673758865248224,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc786184070>,
+                'ensemble_weight': 0.08,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c5d4250>,
+                'model_id': 14,
+                'rank': 17,
+                'sklearn_classifier': LinearDiscriminantAnalysis(tol=0.035543606049804645)},
+        16: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78faf4e50>,
+                'cost': 0.08510638297872342,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc784ec5c40>,
+                'ensemble_weight': 0.02,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78fae4fd0>,
+                'model_id': 16,
+                'rank': 23,
+                'sklearn_classifier': DecisionTreeClassifier(criterion='entropy', max_depth=1, min_samples_leaf=8,
+                           min_samples_split=9, random_state=1)},
+        17: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78c917340>,
+                'cost': 0.04255319148936165,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78f83e280>,
+                'ensemble_weight': 0.02,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c9171f0>,
+                'model_id': 17,
+                'rank': 11,
+                'sklearn_classifier': LinearSVC(C=64.64437157661465, dual=False, intercept_scaling=1.0,
+              random_state=1, tol=0.013399889539803811)},
+        18: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc784f7fa00>,
+                'cost': 0.099290780141844,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78a4c8340>,
+                'ensemble_weight': 0.04,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc784f7f190>,
+                'model_id': 18,
+                'rank': 27,
+                'sklearn_classifier': PassiveAggressiveClassifier(C=0.114391364050916, max_iter=128, random_state=1,
+                                tol=1.3440590734654458e-05, warm_start=True)},
+        19: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78ca21940>,
+                'cost': 0.03546099290780147,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7860954f0>,
+                'ensemble_weight': 0.02,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc785b994f0>,
+                'model_id': 19,
+                'rank': 8,
+                'sklearn_classifier': QuadraticDiscriminantAnalysis(reg_param=0.01667385805274091)},
+        21: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc7871ff3d0>,
+                'cost': 0.028368794326241176,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78770aca0>,
+                'ensemble_weight': 0.06,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78fb2b040>,
+                'model_id': 21,
+                'rank': 5,
+                'sklearn_classifier': LinearSVC(C=2.285306370007781, dual=False, intercept_scaling=1.0,
+              random_state=1, tol=0.005821344021856017)},
+        24: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc784b36370>,
+                'cost': 0.06382978723404253,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc786bb7a90>,
+                'ensemble_weight': 0.02,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc786625e20>,
+                'model_id': 24,
+                'rank': 18,
+                'sklearn_classifier': SGDClassifier(alpha=5.189696963506304e-05, eta0=0.007608275318572964,
+                  learning_rate='constant', loss='perceptron', max_iter=64,
+                  random_state=1, tol=0.0005024616279807072, warm_start=True)},
+        27: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc785e313d0>,
+                'cost': 0.09219858156028371,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78b498ac0>,
+                'ensemble_weight': 0.06,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc784bf14f0>,
+                'model_id': 27,
+                'rank': 25,
+                'sklearn_classifier': DecisionTreeClassifier(criterion='entropy', max_depth=4, min_samples_leaf=18,
+                           min_samples_split=8, random_state=1)},
+        28: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc784b39ee0>,
+                'cost': 0.06382978723404253,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78c5a1e80>,
+                'ensemble_weight': 0.02,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc784b361f0>,
+                'model_id': 28,
+                'rank': 19,
+                'sklearn_classifier': DecisionTreeClassifier(max_depth=1, min_samples_leaf=16, min_samples_split=19,
+                           random_state=1)},
+        29: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc7880f9eb0>,
+                'cost': 0.03546099290780147,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78687a1f0>,
+                'ensemble_weight': 0.04,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78b503dc0>,
+                'model_id': 29,
+                'rank': 9,
+                'sklearn_classifier': QuadraticDiscriminantAnalysis(reg_param=0.23910875621465733)},
+        30: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78662deb0>,
+                'cost': 0.028368794326241176,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc785e49970>,
+                'ensemble_weight': 0.04,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78582e1f0>,
+                'model_id': 30,
+                'rank': 6,
+                'sklearn_classifier': RandomForestClassifier(bootstrap=False, criterion='entropy', max_features=1,
+                           min_samples_leaf=10, min_samples_split=18,
+                           n_estimators=512, n_jobs=1, random_state=1,
+                           warm_start=True)},
+        31: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc787888d00>,
+                'cost': 0.049645390070921946,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78a791d60>,
+                'ensemble_weight': 0.02,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc787888340>,
+                'model_id': 31,
+                'rank': 14,
+                'sklearn_classifier': LinearDiscriminantAnalysis(tol=0.018038793247120707)},
+        33: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc790171700>,
+                'cost': 0.03546099290780147,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78fad6d60>,
+                'ensemble_weight': 0.02,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc790179040>,
+                'model_id': 33,
+                'rank': 10,
+                'sklearn_classifier': HistGradientBoostingClassifier(early_stopping=True,
+                                   l2_regularization=4.5201114871391565e-06,
+                                   learning_rate=0.12585871697357337, max_iter=256,
+                                   max_leaf_nodes=1894, min_samples_leaf=2,
+                                   n_iter_no_change=20, random_state=1,
+                                   validation_fraction=None, warm_start=True)},
+        34: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78b3ce220>,
+                'cost': 0.24822695035460995,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc787011e50>,
+                'ensemble_weight': 0.02,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78b3d35e0>,
+                'model_id': 34,
+                'rank': 29,
+                'sklearn_classifier': ExtraTreesClassifier(criterion='entropy', max_features=8, min_samples_leaf=15,
+                         min_samples_split=17, n_estimators=512, n_jobs=1,
+                         random_state=1, warm_start=True)},
+        35: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc787e35f70>,
+                'cost': 0.049645390070921946,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc790367be0>,
+                'ensemble_weight': 0.04,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78509f550>,
+                'model_id': 35,
+                'rank': 15,
+                'sklearn_classifier': KNeighborsClassifier(n_neighbors=2)},
+        36: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78fa97970>,
+                'cost': 0.07092198581560283,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc787282b20>,
+                'ensemble_weight': 0.02,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78fa97670>,
+                'model_id': 36,
+                'rank': 20,
+                'sklearn_classifier': GaussianNB()},
+        37: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc786887dc0>,
+                'cost': 0.028368794326241176,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78580d250>,
+                'ensemble_weight': 0.04,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78b2743a0>,
+                'model_id': 37,
+                'rank': 7,
+                'sklearn_classifier': LinearDiscriminantAnalysis(shrinkage=0.32361629219618826, solver='lsqr',
+                               tol=3.142032801716559e-05)},
+        38: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78770b160>,
+                'cost': 0.08510638297872342,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc786567e80>,
+                'ensemble_weight': 0.02,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78770ba60>,
+                'model_id': 38,
+                'rank': 24,
+                'sklearn_classifier': DecisionTreeClassifier(criterion='entropy', max_depth=1, min_samples_leaf=13,
+                           min_samples_split=5, random_state=1)}}
     auto-sklearn results:
       Dataset name: breast_cancer
       Metric: accuracy
       Best validation score: 0.992908
-      Number of target algorithm runs: 33
-      Number of successful target algorithm runs: 31
+      Number of target algorithm runs: 37
+      Number of successful target algorithm runs: 36
       Number of crashed target algorithm runs: 1
-      Number of target algorithms that exceeded the time limit: 1
+      Number of target algorithms that exceeded the time limit: 0
       Number of target algorithms that exceeded the memory limit: 0
 
     Accuracy score 0.9440559440559441
@@ -619,7 +804,7 @@ Fit a classifier using Random Search
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 1 minutes  52.038 seconds)
+   **Total running time of the script:** ( 1 minutes  53.718 seconds)
 
 
 .. _sphx_glr_download_examples_60_search_example_random_search.py:
diff --git a/master/_sources/examples/60_search/example_sequential.rst.txt b/master/_sources/examples/60_search/example_sequential.rst.txt
index 3864b1cb78..75faf45ba4 100644
--- a/master/_sources/examples/60_search/example_sequential.rst.txt
+++ b/master/_sources/examples/60_search/example_sequential.rst.txt
@@ -27,10 +27,11 @@ ensembles in parallel. However, it is also possible to run the two processes
 sequentially. The example below shows how to first fit the models and build the
 ensembles afterwards.
 
-.. GENERATED FROM PYTHON SOURCE LINES 11-19
+.. GENERATED FROM PYTHON SOURCE LINES 11-20
 
 .. code-block:: default
 
+    from pprint import pprint
 
     import sklearn.model_selection
     import sklearn.datasets
@@ -46,12 +47,12 @@ ensembles afterwards.
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 20-22
+.. GENERATED FROM PYTHON SOURCE LINES 21-23
 
 Data Loading
 ======================================
 
-.. GENERATED FROM PYTHON SOURCE LINES 22-27
+.. GENERATED FROM PYTHON SOURCE LINES 23-28
 
 .. code-block:: default
 
@@ -67,12 +68,12 @@ Data Loading
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 28-30
+.. GENERATED FROM PYTHON SOURCE LINES 29-31
 
 Build and fit the classifier
 ======================================
 
-.. GENERATED FROM PYTHON SOURCE LINES 30-47
+.. GENERATED FROM PYTHON SOURCE LINES 31-48
 
 .. code-block:: default
 
@@ -103,6 +104,10 @@ Build and fit the classifier
 
  .. code-block:: none
 
+    /home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:68: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+      self.metafeatures = self.metafeatures.append(metafeatures)
+    /home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:72: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+      self.algorithm_runs[metric].append(runs)
 
     AutoSklearnClassifier(delete_tmp_folder_after_terminate=False, ensemble_size=0,
                           per_run_time_limit=30, time_left_for_this_task=120,
@@ -110,17 +115,17 @@ Build and fit the classifier
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 48-50
+.. GENERATED FROM PYTHON SOURCE LINES 49-51
 
 Print the final ensemble constructed by auto-sklearn
 ====================================================
 
-.. GENERATED FROM PYTHON SOURCE LINES 50-53
+.. GENERATED FROM PYTHON SOURCE LINES 51-54
 
 .. code-block:: default
 
 
-    print(automl.show_models())
+    pprint(automl.show_models(), indent=4)
 
 
 
@@ -132,193 +137,306 @@ Print the final ensemble constructed by auto-sklearn
 
  .. code-block:: none
 
-    [(0.100000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'libsvm_svc', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'feature_agglomeration', 'classifier:libsvm_svc:C': 0.07563955664363409, 'classifier:libsvm_svc:gamma': 1.9836355993007913, 'classifier:libsvm_svc:kernel': 'rbf', 'classifier:libsvm_svc:max_iter': -1, 'classifier:libsvm_svc:shrinking': 'True', 'classifier:libsvm_svc:tol': 0.005003173341372008, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'minmax', 'feature_preprocessor:feature_agglomeration:affinity': 'cosine', 'feature_preprocessor:feature_agglomeration:linkage': 'complete', 'feature_preprocessor:feature_agglomeration:n_clusters': 58, 'feature_preprocessor:feature_agglomeration:pooling_func': 'mean', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.15890408760884822},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.080000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'mlp', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'feature_agglomeration', 'classifier:mlp:activation': 'tanh', 'classifier:mlp:alpha': 0.0001363185819149026, 'classifier:mlp:batch_size': 'auto', 'classifier:mlp:beta_1': 0.9, 'classifier:mlp:beta_2': 0.999, 'classifier:mlp:early_stopping': 'valid', 'classifier:mlp:epsilon': 1e-08, 'classifier:mlp:hidden_layer_depth': 3, 'classifier:mlp:learning_rate_init': 0.00018009776276177523, 'classifier:mlp:n_iter_no_change': 32, 'classifier:mlp:num_nodes_per_layer': 115, 'classifier:mlp:shuffle': 'True', 'classifier:mlp:solver': 'adam', 'classifier:mlp:tol': 0.0001, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'median', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize', 'feature_preprocessor:feature_agglomeration:affinity': 'euclidean', 'feature_preprocessor:feature_agglomeration:linkage': 'ward', 'feature_preprocessor:feature_agglomeration:n_clusters': 182, 'feature_preprocessor:feature_agglomeration:pooling_func': 'mean', 'classifier:mlp:validation_fraction': 0.1},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.080000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'gradient_boosting', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'extra_trees_preproc_for_classification', 'classifier:gradient_boosting:early_stop': 'valid', 'classifier:gradient_boosting:l2_regularization': 8.057778875694463e-05, 'classifier:gradient_boosting:learning_rate': 0.09179220974965213, 'classifier:gradient_boosting:loss': 'auto', 'classifier:gradient_boosting:max_bins': 255, 'classifier:gradient_boosting:max_depth': 'None', 'classifier:gradient_boosting:max_leaf_nodes': 200, 'classifier:gradient_boosting:min_samples_leaf': 20, 'classifier:gradient_boosting:scoring': 'loss', 'classifier:gradient_boosting:tol': 1e-07, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'quantile_transformer', 'feature_preprocessor:extra_trees_preproc_for_classification:bootstrap': 'False', 'feature_preprocessor:extra_trees_preproc_for_classification:criterion': 'gini', 'feature_preprocessor:extra_trees_preproc_for_classification:max_depth': 'None', 'feature_preprocessor:extra_trees_preproc_for_classification:max_features': 0.9984367650965825, 'feature_preprocessor:extra_trees_preproc_for_classification:max_leaf_nodes': 'None', 'feature_preprocessor:extra_trees_preproc_for_classification:min_impurity_decrease': 0.0, 'feature_preprocessor:extra_trees_preproc_for_classification:min_samples_leaf': 13, 'feature_preprocessor:extra_trees_preproc_for_classification:min_samples_split': 18, 'feature_preprocessor:extra_trees_preproc_for_classification:min_weight_fraction_leaf': 0.0, 'feature_preprocessor:extra_trees_preproc_for_classification:n_estimators': 100, 'classifier:gradient_boosting:n_iter_no_change': 18, 'classifier:gradient_boosting:validation_fraction': 0.14295295806077554, 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.008542188583124829, 'data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles': 1102, 'data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution': 'uniform'},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.080000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'extra_trees', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'feature_agglomeration', 'classifier:extra_trees:bootstrap': 'True', 'classifier:extra_trees:criterion': 'entropy', 'classifier:extra_trees:max_depth': 'None', 'classifier:extra_trees:max_features': 0.3871420537981852, 'classifier:extra_trees:max_leaf_nodes': 'None', 'classifier:extra_trees:min_impurity_decrease': 0.0, 'classifier:extra_trees:min_samples_leaf': 5, 'classifier:extra_trees:min_samples_split': 14, 'classifier:extra_trees:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'robust_scaler', 'feature_preprocessor:feature_agglomeration:affinity': 'manhattan', 'feature_preprocessor:feature_agglomeration:linkage': 'complete', 'feature_preprocessor:feature_agglomeration:n_clusters': 355, 'feature_preprocessor:feature_agglomeration:pooling_func': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max': 0.726182961545702, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min': 0.02260942170772907},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.060000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'extra_trees', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'polynomial', 'classifier:extra_trees:bootstrap': 'False', 'classifier:extra_trees:criterion': 'gini', 'classifier:extra_trees:max_depth': 'None', 'classifier:extra_trees:max_features': 0.5707983257382487, 'classifier:extra_trees:max_leaf_nodes': 'None', 'classifier:extra_trees:min_impurity_decrease': 0.0, 'classifier:extra_trees:min_samples_leaf': 3, 'classifier:extra_trees:min_samples_split': 11, 'classifier:extra_trees:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'median', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'none', 'feature_preprocessor:polynomial:degree': 2, 'feature_preprocessor:polynomial:include_bias': 'False', 'feature_preprocessor:polynomial:interaction_only': 'False'},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.060000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'extra_trees', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'polynomial', 'classifier:extra_trees:bootstrap': 'False', 'classifier:extra_trees:criterion': 'gini', 'classifier:extra_trees:max_depth': 'None', 'classifier:extra_trees:max_features': 0.4791448484072812, 'classifier:extra_trees:max_leaf_nodes': 'None', 'classifier:extra_trees:min_impurity_decrease': 0.0, 'classifier:extra_trees:min_samples_leaf': 1, 'classifier:extra_trees:min_samples_split': 6, 'classifier:extra_trees:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'none', 'feature_preprocessor:polynomial:degree': 3, 'feature_preprocessor:polynomial:include_bias': 'True', 'feature_preprocessor:polynomial:interaction_only': 'True', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.17184142431776436},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.060000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'gradient_boosting', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'liblinear_svc_preprocessor', 'classifier:gradient_boosting:early_stop': 'train', 'classifier:gradient_boosting:l2_regularization': 0.005326508887463406, 'classifier:gradient_boosting:learning_rate': 0.060800813211425456, 'classifier:gradient_boosting:loss': 'auto', 'classifier:gradient_boosting:max_bins': 255, 'classifier:gradient_boosting:max_depth': 'None', 'classifier:gradient_boosting:max_leaf_nodes': 6, 'classifier:gradient_boosting:min_samples_leaf': 5, 'classifier:gradient_boosting:scoring': 'loss', 'classifier:gradient_boosting:tol': 1e-07, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'median', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'minmax', 'feature_preprocessor:liblinear_svc_preprocessor:C': 13.550960330919455, 'feature_preprocessor:liblinear_svc_preprocessor:dual': 'False', 'feature_preprocessor:liblinear_svc_preprocessor:fit_intercept': 'True', 'feature_preprocessor:liblinear_svc_preprocessor:intercept_scaling': 1, 'feature_preprocessor:liblinear_svc_preprocessor:loss': 'squared_hinge', 'feature_preprocessor:liblinear_svc_preprocessor:multi_class': 'ovr', 'feature_preprocessor:liblinear_svc_preprocessor:penalty': 'l1', 'feature_preprocessor:liblinear_svc_preprocessor:tol': 1.2958033930435781e-05, 'classifier:gradient_boosting:n_iter_no_change': 5},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.060000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'extra_trees', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'polynomial', 'classifier:extra_trees:bootstrap': 'True', 'classifier:extra_trees:criterion': 'entropy', 'classifier:extra_trees:max_depth': 'None', 'classifier:extra_trees:max_features': 0.3793256905143867, 'classifier:extra_trees:max_leaf_nodes': 'None', 'classifier:extra_trees:min_impurity_decrease': 0.0, 'classifier:extra_trees:min_samples_leaf': 20, 'classifier:extra_trees:min_samples_split': 5, 'classifier:extra_trees:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'none', 'feature_preprocessor:polynomial:degree': 3, 'feature_preprocessor:polynomial:include_bias': 'False', 'feature_preprocessor:polynomial:interaction_only': 'True', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.00010083678209087315},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.040000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'gradient_boosting', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'select_percentile_classification', 'classifier:gradient_boosting:early_stop': 'train', 'classifier:gradient_boosting:l2_regularization': 3.387912939529945e-10, 'classifier:gradient_boosting:learning_rate': 0.30755227194768237, 'classifier:gradient_boosting:loss': 'auto', 'classifier:gradient_boosting:max_bins': 255, 'classifier:gradient_boosting:max_depth': 'None', 'classifier:gradient_boosting:max_leaf_nodes': 60, 'classifier:gradient_boosting:min_samples_leaf': 39, 'classifier:gradient_boosting:scoring': 'loss', 'classifier:gradient_boosting:tol': 1e-07, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'none', 'feature_preprocessor:select_percentile_classification:percentile': 93.39844669585806, 'feature_preprocessor:select_percentile_classification:score_func': 'f_classif', 'classifier:gradient_boosting:n_iter_no_change': 18, 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.010000000000000004},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.040000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'extra_trees', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:extra_trees:bootstrap': 'False', 'classifier:extra_trees:criterion': 'gini', 'classifier:extra_trees:max_depth': 'None', 'classifier:extra_trees:max_features': 0.661377361743419, 'classifier:extra_trees:max_leaf_nodes': 'None', 'classifier:extra_trees:min_impurity_decrease': 0.0, 'classifier:extra_trees:min_samples_leaf': 2, 'classifier:extra_trees:min_samples_split': 13, 'classifier:extra_trees:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'none'},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.040000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'extra_trees', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'polynomial', 'classifier:extra_trees:bootstrap': 'False', 'classifier:extra_trees:criterion': 'entropy', 'classifier:extra_trees:max_depth': 'None', 'classifier:extra_trees:max_features': 0.993803313878608, 'classifier:extra_trees:max_leaf_nodes': 'None', 'classifier:extra_trees:min_impurity_decrease': 0.0, 'classifier:extra_trees:min_samples_leaf': 2, 'classifier:extra_trees:min_samples_split': 20, 'classifier:extra_trees:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'median', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'robust_scaler', 'feature_preprocessor:polynomial:degree': 2, 'feature_preprocessor:polynomial:include_bias': 'True', 'feature_preprocessor:polynomial:interaction_only': 'True', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.41826215858914706, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max': 0.7305615609807856, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min': 0.25595970768123566},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.040000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'gradient_boosting', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'extra_trees_preproc_for_classification', 'classifier:gradient_boosting:early_stop': 'train', 'classifier:gradient_boosting:l2_regularization': 0.4635442279519353, 'classifier:gradient_boosting:learning_rate': 0.09809681787962342, 'classifier:gradient_boosting:loss': 'auto', 'classifier:gradient_boosting:max_bins': 255, 'classifier:gradient_boosting:max_depth': 'None', 'classifier:gradient_boosting:max_leaf_nodes': 328, 'classifier:gradient_boosting:min_samples_leaf': 2, 'classifier:gradient_boosting:scoring': 'loss', 'classifier:gradient_boosting:tol': 1e-07, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'median', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize', 'feature_preprocessor:extra_trees_preproc_for_classification:bootstrap': 'False', 'feature_preprocessor:extra_trees_preproc_for_classification:criterion': 'entropy', 'feature_preprocessor:extra_trees_preproc_for_classification:max_depth': 'None', 'feature_preprocessor:extra_trees_preproc_for_classification:max_features': 0.989729761503726, 'feature_preprocessor:extra_trees_preproc_for_classification:max_leaf_nodes': 'None', 'feature_preprocessor:extra_trees_preproc_for_classification:min_impurity_decrease': 0.0, 'feature_preprocessor:extra_trees_preproc_for_classification:min_samples_leaf': 19, 'feature_preprocessor:extra_trees_preproc_for_classification:min_samples_split': 5, 'feature_preprocessor:extra_trees_preproc_for_classification:min_weight_fraction_leaf': 0.0, 'feature_preprocessor:extra_trees_preproc_for_classification:n_estimators': 100, 'classifier:gradient_boosting:n_iter_no_change': 2, 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.0208475757765491},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.040000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'extra_trees', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'extra_trees_preproc_for_classification', 'classifier:extra_trees:bootstrap': 'False', 'classifier:extra_trees:criterion': 'gini', 'classifier:extra_trees:max_depth': 'None', 'classifier:extra_trees:max_features': 0.9292309396985746, 'classifier:extra_trees:max_leaf_nodes': 'None', 'classifier:extra_trees:min_impurity_decrease': 0.0, 'classifier:extra_trees:min_samples_leaf': 1, 'classifier:extra_trees:min_samples_split': 10, 'classifier:extra_trees:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'median', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'normalize', 'feature_preprocessor:extra_trees_preproc_for_classification:bootstrap': 'False', 'feature_preprocessor:extra_trees_preproc_for_classification:criterion': 'entropy', 'feature_preprocessor:extra_trees_preproc_for_classification:max_depth': 'None', 'feature_preprocessor:extra_trees_preproc_for_classification:max_features': 0.9929881254946676, 'feature_preprocessor:extra_trees_preproc_for_classification:max_leaf_nodes': 'None', 'feature_preprocessor:extra_trees_preproc_for_classification:min_impurity_decrease': 0.0, 'feature_preprocessor:extra_trees_preproc_for_classification:min_samples_leaf': 1, 'feature_preprocessor:extra_trees_preproc_for_classification:min_samples_split': 2, 'feature_preprocessor:extra_trees_preproc_for_classification:min_weight_fraction_leaf': 0.0, 'feature_preprocessor:extra_trees_preproc_for_classification:n_estimators': 100, 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.032719158639429445},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.040000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'random_forest', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'select_rates_classification', 'classifier:random_forest:bootstrap': 'True', 'classifier:random_forest:criterion': 'gini', 'classifier:random_forest:max_depth': 'None', 'classifier:random_forest:max_features': 0.5561389303694175, 'classifier:random_forest:max_leaf_nodes': 'None', 'classifier:random_forest:min_impurity_decrease': 0.0, 'classifier:random_forest:min_samples_leaf': 16, 'classifier:random_forest:min_samples_split': 4, 'classifier:random_forest:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'quantile_transformer', 'feature_preprocessor:select_rates_classification:alpha': 0.48507733455436836, 'feature_preprocessor:select_rates_classification:score_func': 'f_classif', 'data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles': 1567, 'data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution': 'normal', 'feature_preprocessor:select_rates_classification:mode': 'fpr'},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.040000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'libsvm_svc', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'select_percentile_classification', 'classifier:libsvm_svc:C': 5121.989374401993, 'classifier:libsvm_svc:gamma': 0.0024386428371511353, 'classifier:libsvm_svc:kernel': 'rbf', 'classifier:libsvm_svc:max_iter': -1, 'classifier:libsvm_svc:shrinking': 'True', 'classifier:libsvm_svc:tol': 0.0706216214952021, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'median', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'robust_scaler', 'feature_preprocessor:select_percentile_classification:percentile': 90.1749860017522, 'feature_preprocessor:select_percentile_classification:score_func': 'chi2', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.00017888891850981892, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max': 0.7393440007191528, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min': 0.04986510953238},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.020000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'mlp', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'fast_ica', 'classifier:mlp:activation': 'relu', 'classifier:mlp:alpha': 0.0017940473175767063, 'classifier:mlp:batch_size': 'auto', 'classifier:mlp:beta_1': 0.9, 'classifier:mlp:beta_2': 0.999, 'classifier:mlp:early_stopping': 'valid', 'classifier:mlp:epsilon': 1e-08, 'classifier:mlp:hidden_layer_depth': 2, 'classifier:mlp:learning_rate_init': 0.0004684917334431039, 'classifier:mlp:n_iter_no_change': 32, 'classifier:mlp:num_nodes_per_layer': 101, 'classifier:mlp:shuffle': 'True', 'classifier:mlp:solver': 'adam', 'classifier:mlp:tol': 0.0001, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize', 'feature_preprocessor:fast_ica:algorithm': 'parallel', 'feature_preprocessor:fast_ica:fun': 'exp', 'feature_preprocessor:fast_ica:whiten': 'False', 'classifier:mlp:validation_fraction': 0.1},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.020000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'mlp', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'extra_trees_preproc_for_classification', 'classifier:mlp:activation': 'tanh', 'classifier:mlp:alpha': 2.5550223982458062e-06, 'classifier:mlp:batch_size': 'auto', 'classifier:mlp:beta_1': 0.9, 'classifier:mlp:beta_2': 0.999, 'classifier:mlp:early_stopping': 'train', 'classifier:mlp:epsilon': 1e-08, 'classifier:mlp:hidden_layer_depth': 3, 'classifier:mlp:learning_rate_init': 0.00027271287919467994, 'classifier:mlp:n_iter_no_change': 32, 'classifier:mlp:num_nodes_per_layer': 54, 'classifier:mlp:shuffle': 'True', 'classifier:mlp:solver': 'adam', 'classifier:mlp:tol': 0.0001, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize', 'feature_preprocessor:extra_trees_preproc_for_classification:bootstrap': 'False', 'feature_preprocessor:extra_trees_preproc_for_classification:criterion': 'entropy', 'feature_preprocessor:extra_trees_preproc_for_classification:max_depth': 'None', 'feature_preprocessor:extra_trees_preproc_for_classification:max_features': 0.7171678618990129, 'feature_preprocessor:extra_trees_preproc_for_classification:max_leaf_nodes': 'None', 'feature_preprocessor:extra_trees_preproc_for_classification:min_impurity_decrease': 0.0, 'feature_preprocessor:extra_trees_preproc_for_classification:min_samples_leaf': 4, 'feature_preprocessor:extra_trees_preproc_for_classification:min_samples_split': 17, 'feature_preprocessor:extra_trees_preproc_for_classification:min_weight_fraction_leaf': 0.0, 'feature_preprocessor:extra_trees_preproc_for_classification:n_estimators': 100, 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.017116661677715188},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.020000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'liblinear_svc', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'select_percentile_classification', 'classifier:liblinear_svc:C': 140.23434217313954, 'classifier:liblinear_svc:dual': 'False', 'classifier:liblinear_svc:fit_intercept': 'True', 'classifier:liblinear_svc:intercept_scaling': 1, 'classifier:liblinear_svc:loss': 'squared_hinge', 'classifier:liblinear_svc:multi_class': 'ovr', 'classifier:liblinear_svc:penalty': 'l2', 'classifier:liblinear_svc:tol': 0.0002944604673080368, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'quantile_transformer', 'feature_preprocessor:select_percentile_classification:percentile': 29.807816607300627, 'feature_preprocessor:select_percentile_classification:score_func': 'chi2', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.034330895903523305, 'data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles': 1450, 'data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution': 'uniform'},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.020000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'mlp', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'feature_agglomeration', 'classifier:mlp:activation': 'relu', 'classifier:mlp:alpha': 4.2841884333778574e-06, 'classifier:mlp:batch_size': 'auto', 'classifier:mlp:beta_1': 0.9, 'classifier:mlp:beta_2': 0.999, 'classifier:mlp:early_stopping': 'train', 'classifier:mlp:epsilon': 1e-08, 'classifier:mlp:hidden_layer_depth': 3, 'classifier:mlp:learning_rate_init': 0.0011804284312897009, 'classifier:mlp:n_iter_no_change': 32, 'classifier:mlp:num_nodes_per_layer': 263, 'classifier:mlp:shuffle': 'True', 'classifier:mlp:solver': 'adam', 'classifier:mlp:tol': 0.0001, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'minmax', 'feature_preprocessor:feature_agglomeration:affinity': 'manhattan', 'feature_preprocessor:feature_agglomeration:linkage': 'average', 'feature_preprocessor:feature_agglomeration:n_clusters': 338, 'feature_preprocessor:feature_agglomeration:pooling_func': 'mean', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.004071801722749603},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.020000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'mlp', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'fast_ica', 'classifier:mlp:activation': 'relu', 'classifier:mlp:alpha': 0.02847755502162456, 'classifier:mlp:batch_size': 'auto', 'classifier:mlp:beta_1': 0.9, 'classifier:mlp:beta_2': 0.999, 'classifier:mlp:early_stopping': 'train', 'classifier:mlp:epsilon': 1e-08, 'classifier:mlp:hidden_layer_depth': 2, 'classifier:mlp:learning_rate_init': 0.000421568792103947, 'classifier:mlp:n_iter_no_change': 32, 'classifier:mlp:num_nodes_per_layer': 123, 'classifier:mlp:shuffle': 'True', 'classifier:mlp:solver': 'adam', 'classifier:mlp:tol': 0.0001, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'median', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'minmax', 'feature_preprocessor:fast_ica:algorithm': 'parallel', 'feature_preprocessor:fast_ica:fun': 'exp', 'feature_preprocessor:fast_ica:whiten': 'False'},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.020000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'bernoulli_nb', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'select_rates_classification', 'classifier:bernoulli_nb:alpha': 39.87397441278958, 'classifier:bernoulli_nb:fit_prior': 'False', 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'median', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize', 'feature_preprocessor:select_rates_classification:alpha': 0.3094962987325228, 'feature_preprocessor:select_rates_classification:score_func': 'mutual_info_classif', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.09580337973953734},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.020000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'passive_aggressive', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:passive_aggressive:C': 0.14268277711454813, 'classifier:passive_aggressive:average': 'False', 'classifier:passive_aggressive:fit_intercept': 'True', 'classifier:passive_aggressive:loss': 'hinge', 'classifier:passive_aggressive:tol': 0.0002600768160857831, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'none', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.0070580904199417415},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    ]
-
-
-
-
-.. GENERATED FROM PYTHON SOURCE LINES 54-56
+    {   2: {   'balancing': Balancing(random_state=1),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc7901c5dc0>,
+               'cost': 0.028368794326241176,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78a5f4f70>,
+               'ensemble_weight': 0.020000000000000004,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc790049a90>,
+               'model_id': 2,
+               'rank': 4,
+               'sklearn_classifier': RandomForestClassifier(max_features=5, n_estimators=512, n_jobs=1,
+                           random_state=1, warm_start=True)},
+        4: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc786184580>,
+               'cost': 0.05673758865248224,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7879e2c10>,
+               'ensemble_weight': 0.04000000000000001,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78fb05610>,
+               'model_id': 4,
+               'rank': 21,
+               'sklearn_classifier': MLPClassifier(activation='tanh', alpha=0.00021148999718383549, beta_1=0.999,
+                  beta_2=0.9, hidden_layer_sizes=(113, 113, 113),
+                  learning_rate_init=0.0007452270241186694, max_iter=64,
+                  n_iter_no_change=32, random_state=1, validation_fraction=0.0,
+                  verbose=0, warm_start=True)},
+        5: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc786918190>,
+               'cost': 0.03546099290780147,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc785e4f4f0>,
+               'ensemble_weight': 0.020000000000000004,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc787a9e310>,
+               'model_id': 5,
+               'rank': 14,
+               'sklearn_classifier': RandomForestClassifier(criterion='entropy', max_features=3, min_samples_leaf=2,
+                           n_estimators=512, n_jobs=1, random_state=1,
+                           warm_start=True)},
+        6: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78c944ca0>,
+               'cost': 0.028368794326241176,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7902bd4f0>,
+               'ensemble_weight': 0.10000000000000002,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c944160>,
+               'model_id': 6,
+               'rank': 5,
+               'sklearn_classifier': MLPClassifier(alpha=0.0017940473175767063, beta_1=0.999, beta_2=0.9,
+                  early_stopping=True, hidden_layer_sizes=(101, 101),
+                  learning_rate_init=0.0004684917334431039, max_iter=32,
+                  n_iter_no_change=32, random_state=1, verbose=0, warm_start=True)},
+        7: {   'balancing': Balancing(random_state=1),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78a7d9bb0>,
+               'cost': 0.014184397163120588,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78ca46460>,
+               'ensemble_weight': 0.04000000000000001,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78ff934c0>,
+               'model_id': 7,
+               'rank': 1,
+               'sklearn_classifier': ExtraTreesClassifier(max_features=34, min_samples_leaf=3, min_samples_split=11,
+                         n_estimators=512, n_jobs=1, random_state=1,
+                         warm_start=True)},
+        8: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc787005400>,
+               'cost': 0.03546099290780147,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc787e71580>,
+               'ensemble_weight': 0.060000000000000005,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc786918580>,
+               'model_id': 8,
+               'rank': 15,
+               'sklearn_classifier': RandomForestClassifier(max_features=2, min_samples_leaf=2, n_estimators=512,
+                           n_jobs=1, random_state=1, warm_start=True)},
+        9: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78fc1f100>,
+               'cost': 0.04255319148936165,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7857b5a30>,
+               'ensemble_weight': 0.060000000000000005,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78bae9040>,
+               'model_id': 9,
+               'rank': 19,
+               'sklearn_classifier': ExtraTreesClassifier(max_features=9, min_samples_split=10, n_estimators=512,
+                         n_jobs=1, random_state=1, warm_start=True)},
+        10: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78f986940>,
+                'cost': 0.028368794326241176,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78c482100>,
+                'ensemble_weight': 0.060000000000000005,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78bcec7c0>,
+                'model_id': 10,
+                'rank': 6,
+                'sklearn_classifier': RandomForestClassifier(criterion='entropy', max_features=4, min_samples_split=6,
+                           n_estimators=512, n_jobs=1, random_state=1,
+                           warm_start=True)},
+        11: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78ab3f070>,
+                'cost': 0.028368794326241176,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78be43c70>,
+                'ensemble_weight': 0.04000000000000001,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78ab3f550>,
+                'model_id': 11,
+                'rank': 7,
+                'sklearn_classifier': RandomForestClassifier(criterion='entropy', max_features=23, min_samples_leaf=7,
+                           n_estimators=512, n_jobs=1, random_state=1,
+                           warm_start=True)},
+        13: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78b2344c0>,
+                'cost': 0.028368794326241176,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc786095eb0>,
+                'ensemble_weight': 0.020000000000000004,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78b2451f0>,
+                'model_id': 13,
+                'rank': 8,
+                'sklearn_classifier': HistGradientBoostingClassifier(early_stopping=False,
+                                   l2_regularization=1.0647401999412075e-10,
+                                   learning_rate=0.08291320147381159, max_iter=512,
+                                   max_leaf_nodes=39, n_iter_no_change=0,
+                                   random_state=1, validation_fraction=None,
+                                   warm_start=True)},
+        15: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78c482310>,
+                'cost': 0.049645390070921946,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc785872910>,
+                'ensemble_weight': 0.020000000000000004,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78fc1fc40>,
+                'model_id': 15,
+                'rank': 20,
+                'sklearn_classifier': MLPClassifier(alpha=4.2841884333778574e-06, beta_1=0.999, beta_2=0.9,
+                  hidden_layer_sizes=(263, 263, 263),
+                  learning_rate_init=0.0011804284312897009, max_iter=128,
+                  n_iter_no_change=32, random_state=1, validation_fraction=0.0,
+                  verbose=0, warm_start=True)},
+        17: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc785b7d6a0>,
+                'cost': 0.03546099290780147,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7870329a0>,
+                'ensemble_weight': 0.020000000000000004,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc7870056d0>,
+                'model_id': 17,
+                'rank': 16,
+                'sklearn_classifier': HistGradientBoostingClassifier(early_stopping=True,
+                                   l2_regularization=0.4635442279519353,
+                                   learning_rate=0.09809681787962342, max_iter=512,
+                                   max_leaf_nodes=328, min_samples_leaf=2,
+                                   n_iter_no_change=2, random_state=1,
+                                   validation_fraction=None, warm_start=True)},
+        18: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78a38fd30>,
+                'cost': 0.03546099290780147,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78790c850>,
+                'ensemble_weight': 0.04000000000000001,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc785b7d5e0>,
+                'model_id': 18,
+                'rank': 17,
+                'sklearn_classifier': RandomForestClassifier(criterion='entropy', max_features=3, n_estimators=512,
+                           n_jobs=1, random_state=1, warm_start=True)},
+        19: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78438b220>,
+                'cost': 0.028368794326241176,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc786c78730>,
+                'ensemble_weight': 0.020000000000000004,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78438b070>,
+                'model_id': 19,
+                'rank': 9,
+                'sklearn_classifier': ExtraTreesClassifier(criterion='entropy', max_features=448, min_samples_leaf=2,
+                         min_samples_split=20, n_estimators=512, n_jobs=1,
+                         random_state=1, warm_start=True)},
+        21: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78a791d30>,
+                'cost': 0.021276595744680882,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78a593e50>,
+                'ensemble_weight': 0.04000000000000001,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc79068ba00>,
+                'model_id': 21,
+                'rank': 3,
+                'sklearn_classifier': ExtraTreesClassifier(criterion='entropy', max_features=4, min_samples_leaf=2,
+                         min_samples_split=15, n_estimators=512, n_jobs=1,
+                         random_state=1, warm_start=True)},
+        26: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc7857f8c40>,
+                'cost': 0.028368794326241176,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78ab77d60>,
+                'ensemble_weight': 0.04000000000000001,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc7857f88e0>,
+                'model_id': 26,
+                'rank': 10,
+                'sklearn_classifier': ExtraTreesClassifier(criterion='entropy', max_features=414, min_samples_leaf=2,
+                         min_samples_split=19, n_estimators=512, n_jobs=1,
+                         random_state=1, warm_start=True)},
+        27: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc784e47520>,
+                'cost': 0.028368794326241176,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc786b74100>,
+                'ensemble_weight': 0.020000000000000004,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc784e47160>,
+                'model_id': 27,
+                'rank': 11,
+                'sklearn_classifier': ExtraTreesClassifier(bootstrap=True, criterion='entropy', max_features=4126,
+                         min_samples_leaf=7, min_samples_split=13, n_estimators=512,
+                         n_jobs=1, random_state=1, warm_start=True)},
+        28: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78b4f2310>,
+                'cost': 0.06382978723404253,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78666a6d0>,
+                'ensemble_weight': 0.020000000000000004,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc790366ac0>,
+                'model_id': 28,
+                'rank': 22,
+                'sklearn_classifier': LinearDiscriminantAnalysis(tol=0.032840833359764676)},
+        29: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78c53a9d0>,
+                'cost': 0.028368794326241176,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc787368bb0>,
+                'ensemble_weight': 0.020000000000000004,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c53a5e0>,
+                'model_id': 29,
+                'rank': 12,
+                'sklearn_classifier': ExtraTreesClassifier(criterion='entropy', max_features=8, min_samples_leaf=11,
+                         min_samples_split=5, n_estimators=512, n_jobs=1,
+                         random_state=1, warm_start=True)},
+        31: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78a378070>,
+                'cost': 0.03546099290780147,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc785fb2160>,
+                'ensemble_weight': 0.04000000000000001,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78a38fc40>,
+                'model_id': 31,
+                'rank': 18,
+                'sklearn_classifier': LinearSVC(class_weight='balanced', dual=False, intercept_scaling=1.0,
+              random_state=1, tol=0.0001350687148423338)},
+        32: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc7850a2c70>,
+                'cost': 0.028368794326241176,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc784f51070>,
+                'ensemble_weight': 0.04000000000000001,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc7850a2760>,
+                'model_id': 32,
+                'rank': 13,
+                'sklearn_classifier': LinearSVC(C=832.5744883902069, dual=False, intercept_scaling=1.0,
+              random_state=1, tol=0.015340961719445064)},
+        34: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc784c102e0>,
+                'cost': 0.06382978723404253,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc786abb970>,
+                'ensemble_weight': 0.020000000000000004,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc784b3b400>,
+                'model_id': 34,
+                'rank': 23,
+                'sklearn_classifier': RandomForestClassifier(bootstrap=False, criterion='entropy', max_features=54,
+                           min_samples_leaf=14, min_samples_split=3,
+                           n_estimators=512, n_jobs=1, random_state=1,
+                           warm_start=True)},
+        36: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc786a9d310>,
+                'cost': 0.36879432624113473,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc787e1a9a0>,
+                'ensemble_weight': 0.08000000000000002,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc786a909a0>,
+                'model_id': 36,
+                'rank': 26,
+                'sklearn_classifier': ExtraTreesClassifier(bootstrap=True, max_features=4, min_samples_leaf=13,
+                         min_samples_split=11, n_estimators=512, n_jobs=1,
+                         random_state=1, warm_start=True)},
+        39: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78f69f220>,
+                'cost': 0.014184397163120588,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78c384be0>,
+                'ensemble_weight': 0.060000000000000005,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc790366ee0>,
+                'model_id': 39,
+                'rank': 2,
+                'sklearn_classifier': ExtraTreesClassifier(criterion='entropy', max_features=32, min_samples_leaf=8,
+                         min_samples_split=8, n_estimators=512, n_jobs=1,
+                         random_state=1, warm_start=True)},
+        41: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78612c5b0>,
+                'cost': 0.07092198581560283,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78febd7c0>,
+                'ensemble_weight': 0.020000000000000004,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78612c280>,
+                'model_id': 41,
+                'rank': 24,
+                'sklearn_classifier': DecisionTreeClassifier(max_depth=1, min_samples_leaf=15, min_samples_split=5,
+                           random_state=1)},
+        43: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78bc70c10>,
+                'cost': 0.09219858156028371,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7873bf130>,
+                'ensemble_weight': 0.04000000000000001,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78b800b20>,
+                'model_id': 43,
+                'rank': 25,
+                'sklearn_classifier': ExtraTreesClassifier(bootstrap=True, criterion='entropy', max_features=12,
+                         min_samples_leaf=19, min_samples_split=17,
+                         n_estimators=512, n_jobs=1, random_state=1,
+                         warm_start=True)}}
+
+
+
+
+.. GENERATED FROM PYTHON SOURCE LINES 55-57
 
 Get the Score of the final ensemble
 ===================================
 
-.. GENERATED FROM PYTHON SOURCE LINES 56-60
+.. GENERATED FROM PYTHON SOURCE LINES 57-61
 
 .. code-block:: default
 
@@ -340,13 +458,13 @@ Get the Score of the final ensemble
       Dataset name: breast_cancer
       Metric: accuracy
       Best validation score: 0.985816
-      Number of target algorithm runs: 41
-      Number of successful target algorithm runs: 41
+      Number of target algorithm runs: 44
+      Number of successful target algorithm runs: 43
       Number of crashed target algorithm runs: 0
-      Number of target algorithms that exceeded the time limit: 0
+      Number of target algorithms that exceeded the time limit: 1
       Number of target algorithms that exceeded the memory limit: 0
 
-    Accuracy score 0.951048951048951
+    Accuracy score 0.958041958041958
 
 
 
@@ -354,7 +472,7 @@ Get the Score of the final ensemble
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 2 minutes  4.152 seconds)
+   **Total running time of the script:** ( 2 minutes  1.878 seconds)
 
 
 .. _sphx_glr_download_examples_60_search_example_sequential.py:
diff --git a/master/_sources/examples/60_search/example_successive_halving.rst.txt b/master/_sources/examples/60_search/example_successive_halving.rst.txt
index 58b04b1f23..f0e9a8c852 100644
--- a/master/_sources/examples/60_search/example_successive_halving.rst.txt
+++ b/master/_sources/examples/60_search/example_successive_halving.rst.txt
@@ -36,7 +36,7 @@ To get the BOHB algorithm, simply import Hyperband and use it as the intensifica
 
 .. code-block:: default
 
-
+    from pprint import pprint
 
     import sklearn.model_selection
     import sklearn.datasets
@@ -171,7 +171,7 @@ Build and fit a classifier
     )
     automl.fit(X_train, y_train, dataset_name='breast_cancer')
 
-    print(automl.show_models())
+    pprint(automl.show_models(), indent=4)
     predictions = automl.predict(X_test)
     # Print statistics about the auto-sklearn run such as number of
     # iterations, number of models failed with a time out.
@@ -188,172 +188,239 @@ Build and fit a classifier
 
  .. code-block:: none
 
+    /home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:68: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+      self.metafeatures = self.metafeatures.append(metafeatures)
+    /home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:72: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+      self.algorithm_runs[metric].append(runs)
+    /opt/hostedtoolcache/Python/3.8.12/x64/lib/python3.8/site-packages/smac/intensification/parallel_scheduling.py:152: UserWarning: SuccessiveHalving is intended to be used with more than 1 worker but num_workers=1
+      warnings.warn("{} is intended to be used "
     /opt/hostedtoolcache/Python/3.8.12/x64/lib/python3.8/site-packages/smac/intensification/parallel_scheduling.py:152: UserWarning: SuccessiveHalving is intended to be used with more than 1 worker but num_workers=1
       warnings.warn("{} is intended to be used "
-    [(0.120000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'sgd', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:sgd:alpha': 9.410144741041167e-05, 'classifier:sgd:average': 'True', 'classifier:sgd:fit_intercept': 'True', 'classifier:sgd:learning_rate': 'constant', 'classifier:sgd:loss': 'log', 'classifier:sgd:penalty': 'l1', 'classifier:sgd:tol': 0.05082904256838701, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'normalize', 'classifier:sgd:eta0': 0.0018055343233337954},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.100000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'sgd', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:sgd:alpha': 0.0002346515712987664, 'classifier:sgd:average': 'True', 'classifier:sgd:fit_intercept': 'True', 'classifier:sgd:learning_rate': 'optimal', 'classifier:sgd:loss': 'log', 'classifier:sgd:penalty': 'l1', 'classifier:sgd:tol': 1.3716748930467322e-05, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize'},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.080000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'gradient_boosting', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:gradient_boosting:early_stop': 'off', 'classifier:gradient_boosting:l2_regularization': 1.0945814167023392e-10, 'classifier:gradient_boosting:learning_rate': 0.11042628136263043, 'classifier:gradient_boosting:loss': 'auto', 'classifier:gradient_boosting:max_bins': 255, 'classifier:gradient_boosting:max_depth': 'None', 'classifier:gradient_boosting:max_leaf_nodes': 30, 'classifier:gradient_boosting:min_samples_leaf': 22, 'classifier:gradient_boosting:scoring': 'loss', 'classifier:gradient_boosting:tol': 1e-07, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.05141281638752715},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.080000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'random_forest', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:random_forest:bootstrap': 'True', 'classifier:random_forest:criterion': 'entropy', 'classifier:random_forest:max_depth': 'None', 'classifier:random_forest:max_features': 0.926283631486858, 'classifier:random_forest:max_leaf_nodes': 'None', 'classifier:random_forest:min_impurity_decrease': 0.0, 'classifier:random_forest:min_samples_leaf': 7, 'classifier:random_forest:min_samples_split': 2, 'classifier:random_forest:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.16265262021972576},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.060000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'random_forest', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:random_forest:bootstrap': 'False', 'classifier:random_forest:criterion': 'entropy', 'classifier:random_forest:max_depth': 'None', 'classifier:random_forest:max_features': 0.41808321658160696, 'classifier:random_forest:max_leaf_nodes': 'None', 'classifier:random_forest:min_impurity_decrease': 0.0, 'classifier:random_forest:min_samples_leaf': 1, 'classifier:random_forest:min_samples_split': 4, 'classifier:random_forest:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'median', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.011283688651384545},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.060000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'extra_trees', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:extra_trees:bootstrap': 'False', 'classifier:extra_trees:criterion': 'entropy', 'classifier:extra_trees:max_depth': 'None', 'classifier:extra_trees:max_features': 0.6128603428070196, 'classifier:extra_trees:max_leaf_nodes': 'None', 'classifier:extra_trees:min_impurity_decrease': 0.0, 'classifier:extra_trees:min_samples_leaf': 1, 'classifier:extra_trees:min_samples_split': 3, 'classifier:extra_trees:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'robust_scaler', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.08125391652261632, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max': 0.8601586365248128, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min': 0.052862156055921525},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.060000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'random_forest', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:random_forest:bootstrap': 'False', 'classifier:random_forest:criterion': 'entropy', 'classifier:random_forest:max_depth': 'None', 'classifier:random_forest:max_features': 0.41808321658160696, 'classifier:random_forest:max_leaf_nodes': 'None', 'classifier:random_forest:min_impurity_decrease': 0.0, 'classifier:random_forest:min_samples_leaf': 1, 'classifier:random_forest:min_samples_split': 4, 'classifier:random_forest:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'median', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.011283688651384545},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.060000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'gradient_boosting', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:gradient_boosting:early_stop': 'off', 'classifier:gradient_boosting:l2_regularization': 1e-10, 'classifier:gradient_boosting:learning_rate': 0.16262682406125173, 'classifier:gradient_boosting:loss': 'auto', 'classifier:gradient_boosting:max_bins': 255, 'classifier:gradient_boosting:max_depth': 'None', 'classifier:gradient_boosting:max_leaf_nodes': 66, 'classifier:gradient_boosting:min_samples_leaf': 20, 'classifier:gradient_boosting:scoring': 'loss', 'classifier:gradient_boosting:tol': 1e-07, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'robust_scaler', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.005428587241449129, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max': 0.75, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min': 0.23746960178084334},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.060000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'extra_trees', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:extra_trees:bootstrap': 'True', 'classifier:extra_trees:criterion': 'gini', 'classifier:extra_trees:max_depth': 'None', 'classifier:extra_trees:max_features': 0.37705188916038523, 'classifier:extra_trees:max_leaf_nodes': 'None', 'classifier:extra_trees:min_impurity_decrease': 0.0, 'classifier:extra_trees:min_samples_leaf': 2, 'classifier:extra_trees:min_samples_split': 3, 'classifier:extra_trees:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'quantile_transformer', 'data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles': 521, 'data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution': 'uniform'},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.060000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'random_forest', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:random_forest:bootstrap': 'True', 'classifier:random_forest:criterion': 'entropy', 'classifier:random_forest:max_depth': 'None', 'classifier:random_forest:max_features': 0.09884140378258977, 'classifier:random_forest:max_leaf_nodes': 'None', 'classifier:random_forest:min_impurity_decrease': 0.0, 'classifier:random_forest:min_samples_leaf': 6, 'classifier:random_forest:min_samples_split': 13, 'classifier:random_forest:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.022834561782884507},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.040000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'gradient_boosting', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:gradient_boosting:early_stop': 'off', 'classifier:gradient_boosting:l2_regularization': 1.0945814167023392e-10, 'classifier:gradient_boosting:learning_rate': 0.11042628136263043, 'classifier:gradient_boosting:loss': 'auto', 'classifier:gradient_boosting:max_bins': 255, 'classifier:gradient_boosting:max_depth': 'None', 'classifier:gradient_boosting:max_leaf_nodes': 30, 'classifier:gradient_boosting:min_samples_leaf': 22, 'classifier:gradient_boosting:scoring': 'loss', 'classifier:gradient_boosting:tol': 1e-07, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.05141281638752715},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.040000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'random_forest', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:random_forest:bootstrap': 'False', 'classifier:random_forest:criterion': 'entropy', 'classifier:random_forest:max_depth': 'None', 'classifier:random_forest:max_features': 0.7708651296941559, 'classifier:random_forest:max_leaf_nodes': 'None', 'classifier:random_forest:min_impurity_decrease': 0.0, 'classifier:random_forest:min_samples_leaf': 13, 'classifier:random_forest:min_samples_split': 2, 'classifier:random_forest:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'quantile_transformer', 'data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles': 931, 'data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution': 'normal'},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.040000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'random_forest', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:random_forest:bootstrap': 'False', 'classifier:random_forest:criterion': 'entropy', 'classifier:random_forest:max_depth': 'None', 'classifier:random_forest:max_features': 0.7323115919225983, 'classifier:random_forest:max_leaf_nodes': 'None', 'classifier:random_forest:min_impurity_decrease': 0.0, 'classifier:random_forest:min_samples_leaf': 15, 'classifier:random_forest:min_samples_split': 6, 'classifier:random_forest:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'robust_scaler', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.011901034843417571, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max': 0.7818500358383581, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min': 0.20068746139723115},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.040000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'random_forest', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:random_forest:bootstrap': 'True', 'classifier:random_forest:criterion': 'entropy', 'classifier:random_forest:max_depth': 'None', 'classifier:random_forest:max_features': 0.6779841015398226, 'classifier:random_forest:max_leaf_nodes': 'None', 'classifier:random_forest:min_impurity_decrease': 0.0, 'classifier:random_forest:min_samples_leaf': 14, 'classifier:random_forest:min_samples_split': 14, 'classifier:random_forest:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'robust_scaler', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.03961232028373377, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max': 0.9772109830437746, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min': 0.13300503334706695},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.020000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'random_forest', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:random_forest:bootstrap': 'True', 'classifier:random_forest:criterion': 'gini', 'classifier:random_forest:max_depth': 'None', 'classifier:random_forest:max_features': 0.5, 'classifier:random_forest:max_leaf_nodes': 'None', 'classifier:random_forest:min_impurity_decrease': 0.0, 'classifier:random_forest:min_samples_leaf': 1, 'classifier:random_forest:min_samples_split': 2, 'classifier:random_forest:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.01},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.020000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'random_forest', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:random_forest:bootstrap': 'False', 'classifier:random_forest:criterion': 'entropy', 'classifier:random_forest:max_depth': 'None', 'classifier:random_forest:max_features': 0.41808321658160696, 'classifier:random_forest:max_leaf_nodes': 'None', 'classifier:random_forest:min_impurity_decrease': 0.0, 'classifier:random_forest:min_samples_leaf': 1, 'classifier:random_forest:min_samples_split': 4, 'classifier:random_forest:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'median', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.011283688651384545},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.020000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'random_forest', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:random_forest:bootstrap': 'False', 'classifier:random_forest:criterion': 'entropy', 'classifier:random_forest:max_depth': 'None', 'classifier:random_forest:max_features': 0.41808321658160696, 'classifier:random_forest:max_leaf_nodes': 'None', 'classifier:random_forest:min_impurity_decrease': 0.0, 'classifier:random_forest:min_samples_leaf': 1, 'classifier:random_forest:min_samples_split': 4, 'classifier:random_forest:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'median', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.011283688651384545},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.020000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'gradient_boosting', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:gradient_boosting:early_stop': 'train', 'classifier:gradient_boosting:l2_regularization': 3.609412172481434e-10, 'classifier:gradient_boosting:learning_rate': 0.05972079854295879, 'classifier:gradient_boosting:loss': 'auto', 'classifier:gradient_boosting:max_bins': 255, 'classifier:gradient_boosting:max_depth': 'None', 'classifier:gradient_boosting:max_leaf_nodes': 4, 'classifier:gradient_boosting:min_samples_leaf': 2, 'classifier:gradient_boosting:scoring': 'loss', 'classifier:gradient_boosting:tol': 1e-07, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'minmax', 'classifier:gradient_boosting:n_iter_no_change': 14},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.020000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'gradient_boosting', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:gradient_boosting:early_stop': 'off', 'classifier:gradient_boosting:l2_regularization': 1.0945814167023392e-10, 'classifier:gradient_boosting:learning_rate': 0.11042628136263043, 'classifier:gradient_boosting:loss': 'auto', 'classifier:gradient_boosting:max_bins': 255, 'classifier:gradient_boosting:max_depth': 'None', 'classifier:gradient_boosting:max_leaf_nodes': 30, 'classifier:gradient_boosting:min_samples_leaf': 22, 'classifier:gradient_boosting:scoring': 'loss', 'classifier:gradient_boosting:tol': 1e-07, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.05141281638752715},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    ]
+    {   2: {   'balancing': Balancing(random_state=1),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc784bf1d00>,
+               'cost': 0.021276595744680882,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78febd310>,
+               'ensemble_weight': 0.02,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc784bf19d0>,
+               'model_id': 2,
+               'rank': 5,
+               'sklearn_classifier': RandomForestClassifier(max_features=5, n_estimators=64, n_jobs=1,
+                           random_state=1, warm_start=True)},
+        3: {   'balancing': Balancing(random_state=1),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78c43e580>,
+               'cost': 0.028368794326241176,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc786095a90>,
+               'ensemble_weight': 0.04,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78a9ea4f0>,
+               'model_id': 3,
+               'rank': 9,
+               'sklearn_classifier': RandomForestClassifier(criterion='entropy', max_features=23, min_samples_leaf=7,
+                           n_estimators=64, n_jobs=1, random_state=1,
+                           warm_start=True)},
+        4: {   'balancing': Balancing(random_state=1),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78c057850>,
+               'cost': 0.07801418439716312,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7869195e0>,
+               'ensemble_weight': 0.08,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c057550>,
+               'model_id': 4,
+               'rank': 21,
+               'sklearn_classifier': PassiveAggressiveClassifier(C=0.14268277711454813, max_iter=128, random_state=1,
+                                tol=0.0002600768160857831, warm_start=True)},
+        5: {   'balancing': Balancing(random_state=1),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc7871ffbe0>,
+               'cost': 0.028368794326241176,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78b915d00>,
+               'ensemble_weight': 0.04,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc7871ff9a0>,
+               'model_id': 5,
+               'rank': 10,
+               'sklearn_classifier': HistGradientBoostingClassifier(early_stopping=False, l2_regularization=1e-10,
+                                   learning_rate=0.16262682406125173, max_iter=64,
+                                   max_leaf_nodes=66, n_iter_no_change=0,
+                                   random_state=1, validation_fraction=None,
+                                   warm_start=True)},
+        7: {   'balancing': Balancing(random_state=1),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc784d70430>,
+               'cost': 0.03546099290780147,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7867e4640>,
+               'ensemble_weight': 0.04,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc784d702b0>,
+               'model_id': 7,
+               'rank': 13,
+               'sklearn_classifier': SGDClassifier(alpha=0.0002346515712987664, average=True, eta0=0.01, loss='log',
+                  max_iter=128, penalty='l1', random_state=1,
+                  tol=1.3716748930467322e-05, warm_start=True)},
+        8: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78509fd30>,
+               'cost': 0.014184397163120588,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc790682160>,
+               'ensemble_weight': 0.06,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78509f910>,
+               'model_id': 8,
+               'rank': 1,
+               'sklearn_classifier': RandomForestClassifier(bootstrap=False, criterion='entropy', max_features=4,
+                           min_samples_split=4, n_estimators=64, n_jobs=1,
+                           random_state=1, warm_start=True)},
+        9: {   'balancing': Balancing(random_state=1),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc786abba00>,
+               'cost': 0.021276595744680882,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78a86deb0>,
+               'ensemble_weight': 0.04,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc786abb100>,
+               'model_id': 9,
+               'rank': 7,
+               'sklearn_classifier': ExtraTreesClassifier(criterion='entropy', max_features=8, min_samples_split=3,
+                         n_estimators=512, n_jobs=1, random_state=1,
+                         warm_start=True)},
+        10: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78582e940>,
+                'cost': 0.028368794326241176,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78b656a90>,
+                'ensemble_weight': 0.04,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78582e3d0>,
+                'model_id': 10,
+                'rank': 11,
+                'sklearn_classifier': HistGradientBoostingClassifier(early_stopping=True,
+                                   l2_regularization=5.027708640006448e-08,
+                                   learning_rate=0.09750328007832798, max_iter=64,
+                                   max_leaf_nodes=1234, min_samples_leaf=25,
+                                   n_iter_no_change=1, random_state=1,
+                                   validation_fraction=0.08300813783286698,
+                                   warm_start=True)},
+        11: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78fc1c340>,
+                'cost': 0.014184397163120588,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc787888970>,
+                'ensemble_weight': 0.0,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78fc1c760>,
+                'model_id': 11,
+                'rank': 8,
+                'sklearn_classifier': HistGradientBoostingClassifier(early_stopping=False,
+                                   l2_regularization=1.0945814167023392e-10,
+                                   learning_rate=0.11042628136263043, max_iter=64,
+                                   max_leaf_nodes=30, min_samples_leaf=22,
+                                   n_iter_no_change=0, random_state=1,
+                                   validation_fraction=None, warm_start=True)},
+        12: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc7878b19d0>,
+                'cost': 0.04255319148936165,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78a7d53a0>,
+                'ensemble_weight': 0.1,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc7878b1850>,
+                'model_id': 12,
+                'rank': 16,
+                'sklearn_classifier': RandomForestClassifier(criterion='entropy', max_features=1, min_samples_leaf=6,
+                           min_samples_split=13, n_estimators=64, n_jobs=1,
+                           random_state=1, warm_start=True)},
+        13: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc787a63dc0>,
+                'cost': 0.03546099290780147,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc785869040>,
+                'ensemble_weight': 0.02,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc787a63bb0>,
+                'model_id': 13,
+                'rank': 14,
+                'sklearn_classifier': HistGradientBoostingClassifier(early_stopping=True,
+                                   l2_regularization=2.506856350040198e-06,
+                                   learning_rate=0.04634380160611007, max_iter=64,
+                                   max_leaf_nodes=11, min_samples_leaf=41,
+                                   n_iter_no_change=17, random_state=1,
+                                   validation_fraction=None, warm_start=True)},
+        14: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc7871e5400>,
+                'cost': 0.03546099290780147,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78b418430>,
+                'ensemble_weight': 0.04,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc7871e5280>,
+                'model_id': 14,
+                'rank': 15,
+                'sklearn_classifier': ExtraTreesClassifier(bootstrap=True, max_features=3, min_samples_leaf=2,
+                         min_samples_split=3, n_estimators=64, n_jobs=1,
+                         random_state=1, warm_start=True)},
+        16: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78a9cc160>,
+                'cost': 0.049645390070921946,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7871ccb50>,
+                'ensemble_weight': 0.06,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc7869b8e80>,
+                'model_id': 16,
+                'rank': 18,
+                'sklearn_classifier': RandomForestClassifier(bootstrap=False, criterion='entropy', max_features=12,
+                           min_samples_leaf=15, min_samples_split=6,
+                           n_estimators=64, n_jobs=1, random_state=1,
+                           warm_start=True)},
+        17: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78a370be0>,
+                'cost': 0.099290780141844,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78c0e9b50>,
+                'ensemble_weight': 0.06,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78a3702e0>,
+                'model_id': 17,
+                'rank': 22,
+                'sklearn_classifier': SGDClassifier(alpha=9.410144741041167e-05, average=True,
+                  eta0=0.0018055343233337954, learning_rate='constant', loss='log',
+                  max_iter=128, penalty='l1', random_state=1,
+                  tol=0.05082904256838701, warm_start=True)},
+        18: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc7845473a0>,
+                'cost': 0.05673758865248224,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7864a4700>,
+                'ensemble_weight': 0.06,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc784547220>,
+                'model_id': 18,
+                'rank': 20,
+                'sklearn_classifier': RandomForestClassifier(criterion='entropy', max_features=10,
+                           min_samples_leaf=14, min_samples_split=14,
+                           n_estimators=64, n_jobs=1, random_state=1,
+                           warm_start=True)},
+        19: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc7848717c0>,
+                'cost': 0.028368794326241176,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78b498c70>,
+                'ensemble_weight': 0.04,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78af79730>,
+                'model_id': 19,
+                'rank': 12,
+                'sklearn_classifier': RandomForestClassifier(bootstrap=False, criterion='entropy', max_features=13,
+                           min_samples_leaf=13, n_estimators=64, n_jobs=1,
+                           random_state=1, warm_start=True)},
+        20: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc7868a13a0>,
+                'cost': 0.049645390070921946,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78758e4c0>,
+                'ensemble_weight': 0.02,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc7868a1220>,
+                'model_id': 20,
+                'rank': 19,
+                'sklearn_classifier': HistGradientBoostingClassifier(early_stopping=False,
+                                   l2_regularization=5.2497721880674565e-06,
+                                   learning_rate=0.03162215674470446, max_iter=64,
+                                   max_leaf_nodes=74, min_samples_leaf=1,
+                                   n_iter_no_change=0, random_state=1,
+                                   validation_fraction=None, warm_start=True)},
+        21: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc7864b81f0>,
+                'cost': 0.04255319148936165,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78bad4130>,
+                'ensemble_weight': 0.02,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc7864b8130>,
+                'model_id': 21,
+                'rank': 17,
+                'sklearn_classifier': PassiveAggressiveClassifier(C=0.05491437530740365, average=True,
+                                loss='squared_hinge', max_iter=128, random_state=1,
+                                tol=0.003640536588885903, warm_start=True)}}
     auto-sklearn results:
       Dataset name: breast_cancer
       Metric: accuracy
       Best validation score: 0.985816
-      Number of target algorithm runs: 24
-      Number of successful target algorithm runs: 24
+      Number of target algorithm runs: 26
+      Number of successful target algorithm runs: 26
       Number of crashed target algorithm runs: 0
       Number of target algorithms that exceeded the time limit: 0
       Number of target algorithms that exceeded the memory limit: 0
 
-    Accuracy score 0.9440559440559441
+    Accuracy score 0.951048951048951
 
 
 
@@ -390,7 +457,7 @@ We can also use cross-validation with successive halving
     automl.fit(X_train, y_train, dataset_name='breast_cancer')
 
     # Print the final ensemble constructed by auto-sklearn.
-    print(automl.show_models())
+    pprint(automl.show_models(), indent=4)
     automl.refit(X_train, y_train)
     predictions = automl.predict(X_test)
     # Print statistics about the auto-sklearn run such as number of
@@ -408,73 +475,355 @@ We can also use cross-validation with successive halving
 
  .. code-block:: none
 
+    /home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:68: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+      self.metafeatures = self.metafeatures.append(metafeatures)
+    /home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:72: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+      self.algorithm_runs[metric].append(runs)
     /opt/hostedtoolcache/Python/3.8.12/x64/lib/python3.8/site-packages/smac/intensification/parallel_scheduling.py:152: UserWarning: SuccessiveHalving is intended to be used with more than 1 worker but num_workers=1
       warnings.warn("{} is intended to be used "
-    [(0.280000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'passive_aggressive', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:passive_aggressive:C': 0.14268277711454813, 'classifier:passive_aggressive:average': 'False', 'classifier:passive_aggressive:fit_intercept': 'True', 'classifier:passive_aggressive:loss': 'hinge', 'classifier:passive_aggressive:tol': 0.0002600768160857831, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'none', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.0070580904199417415},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.180000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'extra_trees', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:extra_trees:bootstrap': 'False', 'classifier:extra_trees:criterion': 'entropy', 'classifier:extra_trees:max_depth': 'None', 'classifier:extra_trees:max_features': 0.6128603428070196, 'classifier:extra_trees:max_leaf_nodes': 'None', 'classifier:extra_trees:min_impurity_decrease': 0.0, 'classifier:extra_trees:min_samples_leaf': 1, 'classifier:extra_trees:min_samples_split': 3, 'classifier:extra_trees:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'robust_scaler', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.08125391652261632, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max': 0.8601586365248128, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min': 0.052862156055921525},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.180000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'extra_trees', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:extra_trees:bootstrap': 'False', 'classifier:extra_trees:criterion': 'entropy', 'classifier:extra_trees:max_depth': 'None', 'classifier:extra_trees:max_features': 0.6128603428070196, 'classifier:extra_trees:max_leaf_nodes': 'None', 'classifier:extra_trees:min_impurity_decrease': 0.0, 'classifier:extra_trees:min_samples_leaf': 1, 'classifier:extra_trees:min_samples_split': 3, 'classifier:extra_trees:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'robust_scaler', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.08125391652261632, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max': 0.8601586365248128, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min': 0.052862156055921525},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.160000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'extra_trees', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:extra_trees:bootstrap': 'False', 'classifier:extra_trees:criterion': 'entropy', 'classifier:extra_trees:max_depth': 'None', 'classifier:extra_trees:max_features': 0.6128603428070196, 'classifier:extra_trees:max_leaf_nodes': 'None', 'classifier:extra_trees:min_impurity_decrease': 0.0, 'classifier:extra_trees:min_samples_leaf': 1, 'classifier:extra_trees:min_samples_split': 3, 'classifier:extra_trees:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'robust_scaler', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.08125391652261632, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max': 0.8601586365248128, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min': 0.052862156055921525},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.100000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'sgd', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:sgd:alpha': 0.0002346515712987664, 'classifier:sgd:average': 'True', 'classifier:sgd:fit_intercept': 'True', 'classifier:sgd:learning_rate': 'optimal', 'classifier:sgd:loss': 'log', 'classifier:sgd:penalty': 'l1', 'classifier:sgd:tol': 1.3716748930467322e-05, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize'},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.080000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'random_forest', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:random_forest:bootstrap': 'True', 'classifier:random_forest:criterion': 'gini', 'classifier:random_forest:max_depth': 'None', 'classifier:random_forest:max_features': 0.5, 'classifier:random_forest:max_leaf_nodes': 'None', 'classifier:random_forest:min_impurity_decrease': 0.0, 'classifier:random_forest:min_samples_leaf': 1, 'classifier:random_forest:min_samples_split': 2, 'classifier:random_forest:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.01},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.020000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'random_forest', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:random_forest:bootstrap': 'True', 'classifier:random_forest:criterion': 'entropy', 'classifier:random_forest:max_depth': 'None', 'classifier:random_forest:max_features': 0.926283631486858, 'classifier:random_forest:max_leaf_nodes': 'None', 'classifier:random_forest:min_impurity_decrease': 0.0, 'classifier:random_forest:min_samples_leaf': 7, 'classifier:random_forest:min_samples_split': 2, 'classifier:random_forest:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.16265262021972576},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    ]
+    {   2: {   'cost': 0.046948356807511755,
+               'ensemble_weight': 0.14,
+               'estimators': [   {   'balancing': Balancing(random_state=1),
+                                     'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78baa93d0>,
+                                     'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc786564c10>,
+                                     'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc7878b0160>,
+                                     'sklearn_classifier': RandomForestClassifier(max_features=5, n_estimators=64, n_jobs=1,
+                           random_state=1, warm_start=True)},
+                                 {   'balancing': Balancing(random_state=1),
+                                     'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc787c7a3d0>,
+                                     'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc786edd8b0>,
+                                     'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc787c7a280>,
+                                     'sklearn_classifier': RandomForestClassifier(max_features=5, n_estimators=64, n_jobs=1,
+                           random_state=1, warm_start=True)},
+                                 {   'balancing': Balancing(random_state=1),
+                                     'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78753bcd0>,
+                                     'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78715d9d0>,
+                                     'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78753bb20>,
+                                     'sklearn_classifier': RandomForestClassifier(max_features=5, n_estimators=64, n_jobs=1,
+                           random_state=1, warm_start=True)},
+                                 {   'balancing': Balancing(random_state=1),
+                                     'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78beb7c40>,
+                                     'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc785684b50>,
+                                     'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78beb7b50>,
+                                     'sklearn_classifier': RandomForestClassifier(max_features=5, n_estimators=64, n_jobs=1,
+                           random_state=1, warm_start=True)},
+                                 {   'balancing': Balancing(random_state=1),
+                                     'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc7872bce20>,
+                                     'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78728de20>,
+                                     'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc7872bcd30>,
+                                     'sklearn_classifier': RandomForestClassifier(max_features=5, n_estimators=64, n_jobs=1,
+                           random_state=1, warm_start=True)}],
+               'model_id': 2,
+               'rank': 8,
+               'voting_model': VotingClassifier(estimators=None, voting='soft')},
+        3: {   'cost': 0.05164319248826292,
+               'ensemble_weight': 0.04,
+               'estimators': [   {   'balancing': Balancing(random_state=1),
+                                     'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc787c79550>,
+                                     'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7878b5580>,
+                                     'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc786ed8610>,
+                                     'sklearn_classifier': RandomForestClassifier(criterion='entropy', max_features=23, min_samples_leaf=7,
+                           n_estimators=64, n_jobs=1, random_state=1,
+                           warm_start=True)},
+                                 {   'balancing': Balancing(random_state=1),
+                                     'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78c5a0e80>,
+                                     'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78c0a7940>,
+                                     'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c5a0cd0>,
+                                     'sklearn_classifier': RandomForestClassifier(criterion='entropy', max_features=23, min_samples_leaf=7,
+                           n_estimators=64, n_jobs=1, random_state=1,
+                           warm_start=True)},
+                                 {   'balancing': Balancing(random_state=1),
+                                     'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc7876f1ac0>,
+                                     'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7876d8b80>,
+                                     'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc7876f1be0>,
+                                     'sklearn_classifier': RandomForestClassifier(criterion='entropy', max_features=23, min_samples_leaf=7,
+                           n_estimators=64, n_jobs=1, random_state=1,
+                           warm_start=True)},
+                                 {   'balancing': Balancing(random_state=1),
+                                     'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc787711040>,
+                                     'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc787742160>,
+                                     'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc7877111c0>,
+                                     'sklearn_classifier': RandomForestClassifier(criterion='entropy', max_features=23, min_samples_leaf=7,
+                           n_estimators=64, n_jobs=1, random_state=1,
+                           warm_start=True)},
+                                 {   'balancing': Balancing(random_state=1),
+                                     'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78662fb20>,
+                                     'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc786633bb0>,
+                                     'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78662fc40>,
+                                     'sklearn_classifier': RandomForestClassifier(criterion='entropy', max_features=23, min_samples_leaf=7,
+                           n_estimators=64, n_jobs=1, random_state=1,
+                           warm_start=True)}],
+               'model_id': 3,
+               'rank': 10,
+               'voting_model': VotingClassifier(estimators=None, voting='soft')},
+        4: {   'cost': 0.08215962441314555,
+               'ensemble_weight': 0.2,
+               'estimators': [   {   'balancing': Balancing(random_state=1),
+                                     'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc787157910>,
+                                     'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7877de790>,
+                                     'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78714dfd0>,
+                                     'sklearn_classifier': PassiveAggressiveClassifier(C=0.14268277711454813, max_iter=128, random_state=1,
+                                tol=0.0002600768160857831, warm_start=True)},
+                                 {   'balancing': Balancing(random_state=1),
+                                     'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc7876d02b0>,
+                                     'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78c4f2e80>,
+                                     'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc7876d0310>,
+                                     'sklearn_classifier': PassiveAggressiveClassifier(C=0.14268277711454813, max_iter=128, random_state=1,
+                                tol=0.0002600768160857831, warm_start=True)},
+                                 {   'balancing': Balancing(random_state=1),
+                                     'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78c53cb80>,
+                                     'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc787711670>,
+                                     'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c53c5e0>,
+                                     'sklearn_classifier': PassiveAggressiveClassifier(C=0.14268277711454813, max_iter=128, random_state=1,
+                                tol=0.0002600768160857831, warm_start=True)},
+                                 {   'balancing': Balancing(random_state=1),
+                                     'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc7843c0100>,
+                                     'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7843a12e0>,
+                                     'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc7843b7d60>,
+                                     'sklearn_classifier': PassiveAggressiveClassifier(C=0.14268277711454813, max_iter=128, random_state=1,
+                                tol=0.0002600768160857831, warm_start=True)},
+                                 {   'balancing': Balancing(random_state=1),
+                                     'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc785b7e4c0>,
+                                     'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc787ec5b20>,
+                                     'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc785b7e400>,
+                                     'sklearn_classifier': PassiveAggressiveClassifier(C=0.14268277711454813, max_iter=128, random_state=1,
+                                tol=0.0002600768160857831, warm_start=True)}],
+               'model_id': 4,
+               'rank': 11,
+               'voting_model': VotingClassifier(estimators=None, voting='soft')},
+        5: {   'cost': 0.035211267605633804,
+               'ensemble_weight': 0.02,
+               'estimators': [   {   'balancing': Balancing(random_state=1),
+                                     'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc7903592e0>,
+                                     'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc787e052e0>,
+                                     'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78ba37610>,
+                                     'sklearn_classifier': HistGradientBoostingClassifier(early_stopping=False, l2_regularization=1e-10,
+                                   learning_rate=0.16262682406125173, max_iter=64,
+                                   max_leaf_nodes=66, n_iter_no_change=0,
+                                   random_state=1, validation_fraction=None,
+                                   warm_start=True)},
+                                 {   'balancing': Balancing(random_state=1),
+                                     'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc786184130>,
+                                     'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78a956c40>,
+                                     'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78fdee8b0>,
+                                     'sklearn_classifier': HistGradientBoostingClassifier(early_stopping=False, l2_regularization=1e-10,
+                                   learning_rate=0.16262682406125173, max_iter=64,
+                                   max_leaf_nodes=66, n_iter_no_change=0,
+                                   random_state=1, validation_fraction=None,
+                                   warm_start=True)},
+                                 {   'balancing': Balancing(random_state=1),
+                                     'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78656e760>,
+                                     'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7878b0700>,
+                                     'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78656e670>,
+                                     'sklearn_classifier': HistGradientBoostingClassifier(early_stopping=False, l2_regularization=1e-10,
+                                   learning_rate=0.16262682406125173, max_iter=64,
+                                   max_leaf_nodes=66, n_iter_no_change=0,
+                                   random_state=1, validation_fraction=None,
+                                   warm_start=True)},
+                                 {   'balancing': Balancing(random_state=1),
+                                     'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc7850b5e50>,
+                                     'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc785087190>,
+                                     'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc7850b5d30>,
+                                     'sklearn_classifier': HistGradientBoostingClassifier(early_stopping=False, l2_regularization=1e-10,
+                                   learning_rate=0.16262682406125173, max_iter=64,
+                                   max_leaf_nodes=66, n_iter_no_change=0,
+                                   random_state=1, validation_fraction=None,
+                                   warm_start=True)},
+                                 {   'balancing': Balancing(random_state=1),
+                                     'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78690f400>,
+                                     'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7875b5640>,
+                                     'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78690ffd0>,
+                                     'sklearn_classifier': HistGradientBoostingClassifier(early_stopping=False, l2_regularization=1e-10,
+                                   learning_rate=0.16262682406125173, max_iter=64,
+                                   max_leaf_nodes=66, n_iter_no_change=0,
+                                   random_state=1, validation_fraction=None,
+                                   warm_start=True)}],
+               'model_id': 5,
+               'rank': 5,
+               'voting_model': VotingClassifier(estimators=None, voting='soft')},
+        6: {   'cost': 0.04694835680751174,
+               'ensemble_weight': 0.02,
+               'estimators': [   {   'balancing': Balancing(random_state=1),
+                                     'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc786fe3af0>,
+                                     'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78c1023d0>,
+                                     'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc786fe30a0>,
+                                     'sklearn_classifier': HistGradientBoostingClassifier(early_stopping=True,
+                                   l2_regularization=3.609412172481434e-10,
+                                   learning_rate=0.05972079854295879, max_iter=64,
+                                   max_leaf_nodes=4, min_samples_leaf=2,
+                                   n_iter_no_change=14, random_state=1,
+                                   validation_fraction=None, warm_start=True)},
+                                 {   'balancing': Balancing(random_state=1),
+                                     'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc7876a55b0>,
+                                     'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78753b430>,
+                                     'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc7876a5520>,
+                                     'sklearn_classifier': HistGradientBoostingClassifier(early_stopping=True,
+                                   l2_regularization=3.609412172481434e-10,
+                                   learning_rate=0.05972079854295879, max_iter=64,
+                                   max_leaf_nodes=4, min_samples_leaf=2,
+                                   n_iter_no_change=14, random_state=1,
+                                   validation_fraction=None, warm_start=True)},
+                                 {   'balancing': Balancing(random_state=1),
+                                     'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc785820a90>,
+                                     'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78728db80>,
+                                     'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc785820970>,
+                                     'sklearn_classifier': HistGradientBoostingClassifier(early_stopping=True,
+                                   l2_regularization=3.609412172481434e-10,
+                                   learning_rate=0.05972079854295879, max_iter=64,
+                                   max_leaf_nodes=4, min_samples_leaf=2,
+                                   n_iter_no_change=14, random_state=1,
+                                   validation_fraction=None, warm_start=True)},
+                                 {   'balancing': Balancing(random_state=1),
+                                     'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78c5ab9a0>,
+                                     'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78c099940>,
+                                     'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c5ab8b0>,
+                                     'sklearn_classifier': HistGradientBoostingClassifier(early_stopping=True,
+                                   l2_regularization=3.609412172481434e-10,
+                                   learning_rate=0.05972079854295879, max_iter=64,
+                                   max_leaf_nodes=4, min_samples_leaf=2,
+                                   n_iter_no_change=14, random_state=1,
+                                   validation_fraction=None, warm_start=True)},
+                                 {   'balancing': Balancing(random_state=1),
+                                     'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc785eda430>,
+                                     'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc784850e80>,
+                                     'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc785eda250>,
+                                     'sklearn_classifier': HistGradientBoostingClassifier(early_stopping=True,
+                                   l2_regularization=3.609412172481434e-10,
+                                   learning_rate=0.05972079854295879, max_iter=64,
+                                   max_leaf_nodes=4, min_samples_leaf=2,
+                                   n_iter_no_change=14, random_state=1,
+                                   validation_fraction=None, warm_start=True)}],
+               'model_id': 6,
+               'rank': 9,
+               'voting_model': VotingClassifier(estimators=None, voting='soft')},
+        7: {   'cost': 0.035211267605633784,
+               'ensemble_weight': 0.16,
+               'estimators': [   {   'balancing': Balancing(random_state=1),
+                                     'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc7868546d0>,
+                                     'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78a593d90>,
+                                     'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c9447c0>,
+                                     'sklearn_classifier': SGDClassifier(alpha=0.0002346515712987664, average=True, eta0=0.01, loss='log',
+                  max_iter=128, penalty='l1', random_state=1,
+                  tol=1.3716748930467322e-05, warm_start=True)},
+                                 {   'balancing': Balancing(random_state=1),
+                                     'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc786899d90>,
+                                     'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc784bd3a00>,
+                                     'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc786899b50>,
+                                     'sklearn_classifier': SGDClassifier(alpha=0.0002346515712987664, average=True, eta0=0.01, loss='log',
+                  max_iter=128, penalty='l1', random_state=1,
+                  tol=1.3716748930467322e-05, warm_start=True)},
+                                 {   'balancing': Balancing(random_state=1),
+                                     'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc786fe3550>,
+                                     'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7875c2c40>,
+                                     'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc786fe33d0>,
+                                     'sklearn_classifier': SGDClassifier(alpha=0.0002346515712987664, average=True, eta0=0.01, loss='log',
+                  max_iter=128, penalty='l1', random_state=1,
+                  tol=1.3716748930467322e-05, warm_start=True)},
+                                 {   'balancing': Balancing(random_state=1),
+                                     'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc787a7a910>,
+                                     'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc787a7e670>,
+                                     'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc787a7a7f0>,
+                                     'sklearn_classifier': SGDClassifier(alpha=0.0002346515712987664, average=True, eta0=0.01, loss='log',
+                  max_iter=128, penalty='l1', random_state=1,
+                  tol=1.3716748930467322e-05, warm_start=True)},
+                                 {   'balancing': Balancing(random_state=1),
+                                     'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc786b94460>,
+                                     'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc786b860d0>,
+                                     'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc786b94340>,
+                                     'sklearn_classifier': SGDClassifier(alpha=0.0002346515712987664, average=True, eta0=0.01, loss='log',
+                  max_iter=128, penalty='l1', random_state=1,
+                  tol=1.3716748930467322e-05, warm_start=True)}],
+               'model_id': 7,
+               'rank': 6,
+               'voting_model': VotingClassifier(estimators=None, voting='soft')},
+        8: {   'cost': 0.039906103286385,
+               'ensemble_weight': 0.08,
+               'estimators': [   {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                                     'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78509fe80>,
+                                     'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78b4c3280>,
+                                     'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78ff43b20>,
+                                     'sklearn_classifier': RandomForestClassifier(bootstrap=False, criterion='entropy', max_features=4,
+                           min_samples_split=4, n_estimators=64, n_jobs=1,
+                           random_state=1, warm_start=True)},
+                                 {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                                     'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc787a7a5b0>,
+                                     'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78b73b580>,
+                                     'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc787a7afa0>,
+                                     'sklearn_classifier': RandomForestClassifier(bootstrap=False, criterion='entropy', max_features=4,
+                           min_samples_split=4, n_estimators=64, n_jobs=1,
+                           random_state=1, warm_start=True)},
+                                 {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                                     'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc786ed1c10>,
+                                     'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78b7d9340>,
+                                     'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc786ed1af0>,
+                                     'sklearn_classifier': RandomForestClassifier(bootstrap=False, criterion='entropy', max_features=4,
+                           min_samples_split=4, n_estimators=64, n_jobs=1,
+                           random_state=1, warm_start=True)},
+                                 {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                                     'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78764e8b0>,
+                                     'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc787ba92b0>,
+                                     'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78764e7c0>,
+                                     'sklearn_classifier': RandomForestClassifier(bootstrap=False, criterion='entropy', max_features=4,
+                           min_samples_split=4, n_estimators=64, n_jobs=1,
+                           random_state=1, warm_start=True)},
+                                 {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                                     'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc784c08610>,
+                                     'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78715dfd0>,
+                                     'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc784c08520>,
+                                     'sklearn_classifier': RandomForestClassifier(bootstrap=False, criterion='entropy', max_features=4,
+                           min_samples_split=4, n_estimators=64, n_jobs=1,
+                           random_state=1, warm_start=True)}],
+               'model_id': 8,
+               'rank': 7,
+               'voting_model': VotingClassifier(estimators=None, voting='soft')},
+        9: {   'cost': 0.032863849765258205,
+               'ensemble_weight': 0.04,
+               'estimators': [   {   'balancing': Balancing(random_state=1),
+                                     'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78ca86700>,
+                                     'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7873f3460>,
+                                     'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78ca86b50>,
+                                     'sklearn_classifier': ExtraTreesClassifier(criterion='entropy', max_features=8, min_samples_split=3,
+                         n_estimators=512, n_jobs=1, random_state=1,
+                         warm_start=True)},
+                                 {   'balancing': Balancing(random_state=1),
+                                     'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78c93b3a0>,
+                                     'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78c8b7970>,
+                                     'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c93bf10>,
+                                     'sklearn_classifier': ExtraTreesClassifier(criterion='entropy', max_features=8, min_samples_split=3,
+                         n_estimators=512, n_jobs=1, random_state=1,
+                         warm_start=True)},
+                                 {   'balancing': Balancing(random_state=1),
+                                     'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc787c50580>,
+                                     'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78b7d9970>,
+                                     'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc787c50460>,
+                                     'sklearn_classifier': ExtraTreesClassifier(criterion='entropy', max_features=8, min_samples_split=3,
+                         n_estimators=512, n_jobs=1, random_state=1,
+                         warm_start=True)},
+                                 {   'balancing': Balancing(random_state=1),
+                                     'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78b21c880>,
+                                     'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78beb3ac0>,
+                                     'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78b21c7c0>,
+                                     'sklearn_classifier': ExtraTreesClassifier(criterion='entropy', max_features=8, min_samples_split=3,
+                         n_estimators=512, n_jobs=1, random_state=1,
+                         warm_start=True)},
+                                 {   'balancing': Balancing(random_state=1),
+                                     'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc787e609a0>,
+                                     'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc785edabe0>,
+                                     'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc787e608e0>,
+                                     'sklearn_classifier': ExtraTreesClassifier(criterion='entropy', max_features=8, min_samples_split=3,
+                         n_estimators=512, n_jobs=1, random_state=1,
+                         warm_start=True)}],
+               'model_id': 9,
+               'rank': 4,
+               'voting_model': VotingClassifier(estimators=None, voting='soft')}}
     auto-sklearn results:
       Dataset name: breast_cancer
       Metric: accuracy
       Best validation score: 0.971831
       Number of target algorithm runs: 11
-      Number of successful target algorithm runs: 10
+      Number of successful target algorithm runs: 11
       Number of crashed target algorithm runs: 0
-      Number of target algorithms that exceeded the time limit: 1
+      Number of target algorithms that exceeded the time limit: 0
       Number of target algorithms that exceeded the memory limit: 0
 
     Accuracy score 0.965034965034965
@@ -514,7 +863,7 @@ Use an iterative fit cross-validation with successive halving
     automl.fit(X_train, y_train, dataset_name='breast_cancer')
 
     # Print the final ensemble constructed by auto-sklearn.
-    print(automl.show_models())
+    pprint(automl.show_models(), indent=4)
     automl.refit(X_train, y_train)
     predictions = automl.predict(X_test)
     # Print statistics about the auto-sklearn run such as number of
@@ -532,65 +881,75 @@ Use an iterative fit cross-validation with successive halving
 
  .. code-block:: none
 
+    /home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:68: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+      self.metafeatures = self.metafeatures.append(metafeatures)
+    /home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:72: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+      self.algorithm_runs[metric].append(runs)
     /opt/hostedtoolcache/Python/3.8.12/x64/lib/python3.8/site-packages/smac/intensification/parallel_scheduling.py:152: UserWarning: SuccessiveHalving is intended to be used with more than 1 worker but num_workers=1
       warnings.warn("{} is intended to be used "
-    [(0.240000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'sgd', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:sgd:alpha': 0.0002346515712987664, 'classifier:sgd:average': 'True', 'classifier:sgd:fit_intercept': 'True', 'classifier:sgd:learning_rate': 'optimal', 'classifier:sgd:loss': 'log', 'classifier:sgd:penalty': 'l1', 'classifier:sgd:tol': 1.3716748930467322e-05, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize'},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.180000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'random_forest', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:random_forest:bootstrap': 'False', 'classifier:random_forest:criterion': 'entropy', 'classifier:random_forest:max_depth': 'None', 'classifier:random_forest:max_features': 0.41808321658160696, 'classifier:random_forest:max_leaf_nodes': 'None', 'classifier:random_forest:min_impurity_decrease': 0.0, 'classifier:random_forest:min_samples_leaf': 1, 'classifier:random_forest:min_samples_split': 4, 'classifier:random_forest:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'median', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.011283688651384545},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.160000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'passive_aggressive', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:passive_aggressive:C': 0.14268277711454813, 'classifier:passive_aggressive:average': 'False', 'classifier:passive_aggressive:fit_intercept': 'True', 'classifier:passive_aggressive:loss': 'hinge', 'classifier:passive_aggressive:tol': 0.0002600768160857831, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'none', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.0070580904199417415},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.140000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'gradient_boosting', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:gradient_boosting:early_stop': 'train', 'classifier:gradient_boosting:l2_regularization': 3.609412172481434e-10, 'classifier:gradient_boosting:learning_rate': 0.05972079854295879, 'classifier:gradient_boosting:loss': 'auto', 'classifier:gradient_boosting:max_bins': 255, 'classifier:gradient_boosting:max_depth': 'None', 'classifier:gradient_boosting:max_leaf_nodes': 4, 'classifier:gradient_boosting:min_samples_leaf': 2, 'classifier:gradient_boosting:scoring': 'loss', 'classifier:gradient_boosting:tol': 1e-07, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'minmax', 'classifier:gradient_boosting:n_iter_no_change': 14},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.120000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'random_forest', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:random_forest:bootstrap': 'True', 'classifier:random_forest:criterion': 'gini', 'classifier:random_forest:max_depth': 'None', 'classifier:random_forest:max_features': 0.5, 'classifier:random_forest:max_leaf_nodes': 'None', 'classifier:random_forest:min_impurity_decrease': 0.0, 'classifier:random_forest:min_samples_leaf': 1, 'classifier:random_forest:min_samples_split': 2, 'classifier:random_forest:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.01},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.100000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'random_forest', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:random_forest:bootstrap': 'True', 'classifier:random_forest:criterion': 'entropy', 'classifier:random_forest:max_depth': 'None', 'classifier:random_forest:max_features': 0.926283631486858, 'classifier:random_forest:max_leaf_nodes': 'None', 'classifier:random_forest:min_impurity_decrease': 0.0, 'classifier:random_forest:min_samples_leaf': 7, 'classifier:random_forest:min_samples_split': 2, 'classifier:random_forest:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.16265262021972576},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.060000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'gradient_boosting', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:gradient_boosting:early_stop': 'off', 'classifier:gradient_boosting:l2_regularization': 1e-10, 'classifier:gradient_boosting:learning_rate': 0.16262682406125173, 'classifier:gradient_boosting:loss': 'auto', 'classifier:gradient_boosting:max_bins': 255, 'classifier:gradient_boosting:max_depth': 'None', 'classifier:gradient_boosting:max_leaf_nodes': 66, 'classifier:gradient_boosting:min_samples_leaf': 20, 'classifier:gradient_boosting:scoring': 'loss', 'classifier:gradient_boosting:tol': 1e-07, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'robust_scaler', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.005428587241449129, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max': 0.75, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min': 0.23746960178084334},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    ]
+    {   2: {   'balancing': Balancing(random_state=1),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78648be80>,
+               'cost': 0.046948356807511755,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78753bfa0>,
+               'ensemble_weight': 0.12,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78648b5b0>,
+               'model_id': 2,
+               'rank': 4,
+               'sklearn_classifier': None},
+        3: {   'balancing': Balancing(random_state=1),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc7871573a0>,
+               'cost': 0.05164319248826292,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc785814430>,
+               'ensemble_weight': 0.1,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc7871571c0>,
+               'model_id': 3,
+               'rank': 6,
+               'sklearn_classifier': None},
+        4: {   'balancing': Balancing(random_state=1),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc786ff0ca0>,
+               'cost': 0.11267605633802817,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7875b5970>,
+               'ensemble_weight': 0.16,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc786ff0970>,
+               'model_id': 4,
+               'rank': 7,
+               'sklearn_classifier': None},
+        5: {   'balancing': Balancing(random_state=1),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc786912b20>,
+               'cost': 0.035211267605633804,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc787bb7550>,
+               'ensemble_weight': 0.06,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc786912460>,
+               'model_id': 5,
+               'rank': 2,
+               'sklearn_classifier': None},
+        6: {   'balancing': Balancing(random_state=1),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc785087220>,
+               'cost': 0.04694835680751174,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78715d850>,
+               'ensemble_weight': 0.14,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc785087df0>,
+               'model_id': 6,
+               'rank': 5,
+               'sklearn_classifier': None},
+        7: {   'balancing': Balancing(random_state=1),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc784c0adf0>,
+               'cost': 0.03286384976525822,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78715d9a0>,
+               'ensemble_weight': 0.24,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc787bb7430>,
+               'model_id': 7,
+               'rank': 1,
+               'sklearn_classifier': None},
+        8: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78715d310>,
+               'cost': 0.039906103286385,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc784c0a520>,
+               'ensemble_weight': 0.18,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78715d3a0>,
+               'model_id': 8,
+               'rank': 3,
+               'sklearn_classifier': None}}
     auto-sklearn results:
       Dataset name: breast_cancer
       Metric: accuracy
@@ -635,7 +994,7 @@ Next, we see the use of subsampling as a budget in Auto-sklearn
     automl.fit(X_train, y_train, dataset_name='breast_cancer')
 
     # Print the final ensemble constructed by auto-sklearn.
-    print(automl.show_models())
+    pprint(automl.show_models(), indent=4)
     predictions = automl.predict(X_test)
     # Print statistics about the auto-sklearn run such as number of
     # iterations, number of models failed with a time out.
@@ -652,100 +1011,47 @@ Next, we see the use of subsampling as a budget in Auto-sklearn
 
  .. code-block:: none
 
+    /home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:68: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+      self.metafeatures = self.metafeatures.append(metafeatures)
+    /home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:72: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+      self.algorithm_runs[metric].append(runs)
     /opt/hostedtoolcache/Python/3.8.12/x64/lib/python3.8/site-packages/smac/intensification/parallel_scheduling.py:152: UserWarning: SuccessiveHalving is intended to be used with more than 1 worker but num_workers=1
       warnings.warn("{} is intended to be used "
-    [(0.340000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'mlp', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'fast_ica', 'classifier:mlp:activation': 'relu', 'classifier:mlp:alpha': 0.0017940473175767063, 'classifier:mlp:batch_size': 'auto', 'classifier:mlp:beta_1': 0.9, 'classifier:mlp:beta_2': 0.999, 'classifier:mlp:early_stopping': 'valid', 'classifier:mlp:epsilon': 1e-08, 'classifier:mlp:hidden_layer_depth': 2, 'classifier:mlp:learning_rate_init': 0.0004684917334431039, 'classifier:mlp:n_iter_no_change': 32, 'classifier:mlp:num_nodes_per_layer': 101, 'classifier:mlp:shuffle': 'True', 'classifier:mlp:solver': 'adam', 'classifier:mlp:tol': 0.0001, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize', 'feature_preprocessor:fast_ica:algorithm': 'parallel', 'feature_preprocessor:fast_ica:fun': 'exp', 'feature_preprocessor:fast_ica:whiten': 'False', 'classifier:mlp:validation_fraction': 0.1},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.260000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'mlp', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'feature_agglomeration', 'classifier:mlp:activation': 'tanh', 'classifier:mlp:alpha': 0.0001363185819149026, 'classifier:mlp:batch_size': 'auto', 'classifier:mlp:beta_1': 0.9, 'classifier:mlp:beta_2': 0.999, 'classifier:mlp:early_stopping': 'valid', 'classifier:mlp:epsilon': 1e-08, 'classifier:mlp:hidden_layer_depth': 3, 'classifier:mlp:learning_rate_init': 0.00018009776276177523, 'classifier:mlp:n_iter_no_change': 32, 'classifier:mlp:num_nodes_per_layer': 115, 'classifier:mlp:shuffle': 'True', 'classifier:mlp:solver': 'adam', 'classifier:mlp:tol': 0.0001, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'median', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize', 'feature_preprocessor:feature_agglomeration:affinity': 'euclidean', 'feature_preprocessor:feature_agglomeration:linkage': 'ward', 'feature_preprocessor:feature_agglomeration:n_clusters': 182, 'feature_preprocessor:feature_agglomeration:pooling_func': 'mean', 'classifier:mlp:validation_fraction': 0.1},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.120000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'random_forest', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'select_percentile_classification', 'classifier:random_forest:bootstrap': 'True', 'classifier:random_forest:criterion': 'entropy', 'classifier:random_forest:max_depth': 'None', 'classifier:random_forest:max_features': 0.49138075723513286, 'classifier:random_forest:max_leaf_nodes': 'None', 'classifier:random_forest:min_impurity_decrease': 0.0, 'classifier:random_forest:min_samples_leaf': 1, 'classifier:random_forest:min_samples_split': 6, 'classifier:random_forest:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'minmax', 'feature_preprocessor:select_percentile_classification:percentile': 56.97947373958566, 'feature_preprocessor:select_percentile_classification:score_func': 'mutual_info', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.42693600390988135},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.060000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'random_forest', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:random_forest:bootstrap': 'True', 'classifier:random_forest:criterion': 'gini', 'classifier:random_forest:max_depth': 'None', 'classifier:random_forest:max_features': 0.5, 'classifier:random_forest:max_leaf_nodes': 'None', 'classifier:random_forest:min_impurity_decrease': 0.0, 'classifier:random_forest:min_samples_leaf': 1, 'classifier:random_forest:min_samples_split': 2, 'classifier:random_forest:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.01},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.040000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'random_forest', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:random_forest:bootstrap': 'True', 'classifier:random_forest:criterion': 'gini', 'classifier:random_forest:max_depth': 'None', 'classifier:random_forest:max_features': 0.5, 'classifier:random_forest:max_leaf_nodes': 'None', 'classifier:random_forest:min_impurity_decrease': 0.0, 'classifier:random_forest:min_samples_leaf': 1, 'classifier:random_forest:min_samples_split': 2, 'classifier:random_forest:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.01},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.040000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'random_forest', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:random_forest:bootstrap': 'True', 'classifier:random_forest:criterion': 'gini', 'classifier:random_forest:max_depth': 'None', 'classifier:random_forest:max_features': 0.5, 'classifier:random_forest:max_leaf_nodes': 'None', 'classifier:random_forest:min_impurity_decrease': 0.0, 'classifier:random_forest:min_samples_leaf': 1, 'classifier:random_forest:min_samples_split': 2, 'classifier:random_forest:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.01},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.040000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'extra_trees', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'polynomial', 'classifier:extra_trees:bootstrap': 'False', 'classifier:extra_trees:criterion': 'gini', 'classifier:extra_trees:max_depth': 'None', 'classifier:extra_trees:max_features': 0.5707983257382487, 'classifier:extra_trees:max_leaf_nodes': 'None', 'classifier:extra_trees:min_impurity_decrease': 0.0, 'classifier:extra_trees:min_samples_leaf': 3, 'classifier:extra_trees:min_samples_split': 11, 'classifier:extra_trees:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'median', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'none', 'feature_preprocessor:polynomial:degree': 2, 'feature_preprocessor:polynomial:include_bias': 'False', 'feature_preprocessor:polynomial:interaction_only': 'False'},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.040000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'random_forest', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'extra_trees_preproc_for_classification', 'classifier:random_forest:bootstrap': 'True', 'classifier:random_forest:criterion': 'gini', 'classifier:random_forest:max_depth': 'None', 'classifier:random_forest:max_features': 0.43999367631975456, 'classifier:random_forest:max_leaf_nodes': 'None', 'classifier:random_forest:min_impurity_decrease': 0.0, 'classifier:random_forest:min_samples_leaf': 2, 'classifier:random_forest:min_samples_split': 2, 'classifier:random_forest:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize', 'feature_preprocessor:extra_trees_preproc_for_classification:bootstrap': 'True', 'feature_preprocessor:extra_trees_preproc_for_classification:criterion': 'entropy', 'feature_preprocessor:extra_trees_preproc_for_classification:max_depth': 'None', 'feature_preprocessor:extra_trees_preproc_for_classification:max_features': 0.8134515743047006, 'feature_preprocessor:extra_trees_preproc_for_classification:max_leaf_nodes': 'None', 'feature_preprocessor:extra_trees_preproc_for_classification:min_impurity_decrease': 0.0, 'feature_preprocessor:extra_trees_preproc_for_classification:min_samples_leaf': 9, 'feature_preprocessor:extra_trees_preproc_for_classification:min_samples_split': 20, 'feature_preprocessor:extra_trees_preproc_for_classification:min_weight_fraction_leaf': 0.0, 'feature_preprocessor:extra_trees_preproc_for_classification:n_estimators': 100},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.040000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'extra_trees', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'extra_trees_preproc_for_classification', 'classifier:extra_trees:bootstrap': 'False', 'classifier:extra_trees:criterion': 'gini', 'classifier:extra_trees:max_depth': 'None', 'classifier:extra_trees:max_features': 0.9292309396985746, 'classifier:extra_trees:max_leaf_nodes': 'None', 'classifier:extra_trees:min_impurity_decrease': 0.0, 'classifier:extra_trees:min_samples_leaf': 1, 'classifier:extra_trees:min_samples_split': 10, 'classifier:extra_trees:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'median', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'normalize', 'feature_preprocessor:extra_trees_preproc_for_classification:bootstrap': 'False', 'feature_preprocessor:extra_trees_preproc_for_classification:criterion': 'entropy', 'feature_preprocessor:extra_trees_preproc_for_classification:max_depth': 'None', 'feature_preprocessor:extra_trees_preproc_for_classification:max_features': 0.9929881254946676, 'feature_preprocessor:extra_trees_preproc_for_classification:max_leaf_nodes': 'None', 'feature_preprocessor:extra_trees_preproc_for_classification:min_impurity_decrease': 0.0, 'feature_preprocessor:extra_trees_preproc_for_classification:min_samples_leaf': 1, 'feature_preprocessor:extra_trees_preproc_for_classification:min_samples_split': 2, 'feature_preprocessor:extra_trees_preproc_for_classification:min_weight_fraction_leaf': 0.0, 'feature_preprocessor:extra_trees_preproc_for_classification:n_estimators': 100, 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.032719158639429445},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.020000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'random_forest', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:random_forest:bootstrap': 'True', 'classifier:random_forest:criterion': 'gini', 'classifier:random_forest:max_depth': 'None', 'classifier:random_forest:max_features': 0.5, 'classifier:random_forest:max_leaf_nodes': 'None', 'classifier:random_forest:min_impurity_decrease': 0.0, 'classifier:random_forest:min_samples_leaf': 1, 'classifier:random_forest:min_samples_split': 2, 'classifier:random_forest:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.01},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    ]
+    {   3: {   'balancing': Balancing(random_state=1),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78b503fd0>,
+               'cost': 0.021276595744680882,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc790367be0>,
+               'ensemble_weight': 0.4,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78a6db2b0>,
+               'model_id': 3,
+               'rank': 2,
+               'sklearn_classifier': MLPClassifier(activation='tanh', alpha=0.0001363185819149026, beta_1=0.999,
+                  beta_2=0.9, early_stopping=True,
+                  hidden_layer_sizes=(115, 115, 115),
+                  learning_rate_init=0.00018009776276177523, max_iter=32,
+                  n_iter_no_change=32, random_state=1, verbose=0, warm_start=True)},
+        7: {   'balancing': Balancing(random_state=1),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78ca21b20>,
+               'cost': 0.014184397163120588,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78c5ada90>,
+               'ensemble_weight': 0.04,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78a956520>,
+               'model_id': 7,
+               'rank': 3,
+               'sklearn_classifier': ExtraTreesClassifier(max_features=34, min_samples_leaf=3, min_samples_split=11,
+                         n_estimators=512, n_jobs=1, random_state=1,
+                         warm_start=True)}}
     auto-sklearn results:
       Dataset name: breast_cancer
       Metric: accuracy
-      Best validation score: 0.978723
-      Number of target algorithm runs: 14
-      Number of successful target algorithm runs: 14
+      Best validation score: 0.985816
+      Number of target algorithm runs: 17
+      Number of successful target algorithm runs: 17
       Number of crashed target algorithm runs: 0
       Number of target algorithms that exceeded the time limit: 0
       Number of target algorithms that exceeded the memory limit: 0
 
-    Accuracy score 0.951048951048951
+    Accuracy score 0.958041958041958
 
 
 
@@ -786,7 +1092,7 @@ subsamples otherwise
     automl.fit(X_train, y_train, dataset_name='breast_cancer')
 
     # Print the final ensemble constructed by auto-sklearn.
-    print(automl.show_models())
+    pprint(automl.show_models(), indent=4)
     predictions = automl.predict(X_test)
     # Print statistics about the auto-sklearn run such as number of
     # iterations, number of models failed with a time out.
@@ -802,111 +1108,112 @@ subsamples otherwise
 
  .. code-block:: none
 
+    /home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:68: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+      self.metafeatures = self.metafeatures.append(metafeatures)
+    /home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:72: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+      self.algorithm_runs[metric].append(runs)
     /opt/hostedtoolcache/Python/3.8.12/x64/lib/python3.8/site-packages/smac/intensification/parallel_scheduling.py:152: UserWarning: SuccessiveHalving is intended to be used with more than 1 worker but num_workers=1
       warnings.warn("{} is intended to be used "
-    [(0.280000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'gradient_boosting', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'select_percentile_classification', 'classifier:gradient_boosting:early_stop': 'train', 'classifier:gradient_boosting:l2_regularization': 3.387912939529945e-10, 'classifier:gradient_boosting:learning_rate': 0.30755227194768237, 'classifier:gradient_boosting:loss': 'auto', 'classifier:gradient_boosting:max_bins': 255, 'classifier:gradient_boosting:max_depth': 'None', 'classifier:gradient_boosting:max_leaf_nodes': 60, 'classifier:gradient_boosting:min_samples_leaf': 39, 'classifier:gradient_boosting:scoring': 'loss', 'classifier:gradient_boosting:tol': 1e-07, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'none', 'feature_preprocessor:select_percentile_classification:percentile': 93.39844669585806, 'feature_preprocessor:select_percentile_classification:score_func': 'f_classif', 'classifier:gradient_boosting:n_iter_no_change': 18, 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.010000000000000004},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.140000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'extra_trees', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'polynomial', 'classifier:extra_trees:bootstrap': 'False', 'classifier:extra_trees:criterion': 'gini', 'classifier:extra_trees:max_depth': 'None', 'classifier:extra_trees:max_features': 0.5707983257382487, 'classifier:extra_trees:max_leaf_nodes': 'None', 'classifier:extra_trees:min_impurity_decrease': 0.0, 'classifier:extra_trees:min_samples_leaf': 3, 'classifier:extra_trees:min_samples_split': 11, 'classifier:extra_trees:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'median', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'none', 'feature_preprocessor:polynomial:degree': 2, 'feature_preprocessor:polynomial:include_bias': 'False', 'feature_preprocessor:polynomial:interaction_only': 'False'},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.120000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'extra_trees', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'polynomial', 'classifier:extra_trees:bootstrap': 'False', 'classifier:extra_trees:criterion': 'gini', 'classifier:extra_trees:max_depth': 'None', 'classifier:extra_trees:max_features': 0.5707983257382487, 'classifier:extra_trees:max_leaf_nodes': 'None', 'classifier:extra_trees:min_impurity_decrease': 0.0, 'classifier:extra_trees:min_samples_leaf': 3, 'classifier:extra_trees:min_samples_split': 11, 'classifier:extra_trees:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'median', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'none', 'feature_preprocessor:polynomial:degree': 2, 'feature_preprocessor:polynomial:include_bias': 'False', 'feature_preprocessor:polynomial:interaction_only': 'False'},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.100000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'extra_trees', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'polynomial', 'classifier:extra_trees:bootstrap': 'False', 'classifier:extra_trees:criterion': 'gini', 'classifier:extra_trees:max_depth': 'None', 'classifier:extra_trees:max_features': 0.5707983257382487, 'classifier:extra_trees:max_leaf_nodes': 'None', 'classifier:extra_trees:min_impurity_decrease': 0.0, 'classifier:extra_trees:min_samples_leaf': 3, 'classifier:extra_trees:min_samples_split': 11, 'classifier:extra_trees:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'median', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'none', 'feature_preprocessor:polynomial:degree': 2, 'feature_preprocessor:polynomial:include_bias': 'False', 'feature_preprocessor:polynomial:interaction_only': 'False'},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.100000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'random_forest', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:random_forest:bootstrap': 'True', 'classifier:random_forest:criterion': 'gini', 'classifier:random_forest:max_depth': 'None', 'classifier:random_forest:max_features': 0.5, 'classifier:random_forest:max_leaf_nodes': 'None', 'classifier:random_forest:min_impurity_decrease': 0.0, 'classifier:random_forest:min_samples_leaf': 1, 'classifier:random_forest:min_samples_split': 2, 'classifier:random_forest:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.01},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.080000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'extra_trees', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'polynomial', 'classifier:extra_trees:bootstrap': 'False', 'classifier:extra_trees:criterion': 'gini', 'classifier:extra_trees:max_depth': 'None', 'classifier:extra_trees:max_features': 0.5707983257382487, 'classifier:extra_trees:max_leaf_nodes': 'None', 'classifier:extra_trees:min_impurity_decrease': 0.0, 'classifier:extra_trees:min_samples_leaf': 3, 'classifier:extra_trees:min_samples_split': 11, 'classifier:extra_trees:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'median', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'none', 'feature_preprocessor:polynomial:degree': 2, 'feature_preprocessor:polynomial:include_bias': 'False', 'feature_preprocessor:polynomial:interaction_only': 'False'},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.060000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'extra_trees', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'extra_trees_preproc_for_classification', 'classifier:extra_trees:bootstrap': 'False', 'classifier:extra_trees:criterion': 'gini', 'classifier:extra_trees:max_depth': 'None', 'classifier:extra_trees:max_features': 0.9292309396985746, 'classifier:extra_trees:max_leaf_nodes': 'None', 'classifier:extra_trees:min_impurity_decrease': 0.0, 'classifier:extra_trees:min_samples_leaf': 1, 'classifier:extra_trees:min_samples_split': 10, 'classifier:extra_trees:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'median', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'normalize', 'feature_preprocessor:extra_trees_preproc_for_classification:bootstrap': 'False', 'feature_preprocessor:extra_trees_preproc_for_classification:criterion': 'entropy', 'feature_preprocessor:extra_trees_preproc_for_classification:max_depth': 'None', 'feature_preprocessor:extra_trees_preproc_for_classification:max_features': 0.9929881254946676, 'feature_preprocessor:extra_trees_preproc_for_classification:max_leaf_nodes': 'None', 'feature_preprocessor:extra_trees_preproc_for_classification:min_impurity_decrease': 0.0, 'feature_preprocessor:extra_trees_preproc_for_classification:min_samples_leaf': 1, 'feature_preprocessor:extra_trees_preproc_for_classification:min_samples_split': 2, 'feature_preprocessor:extra_trees_preproc_for_classification:min_weight_fraction_leaf': 0.0, 'feature_preprocessor:extra_trees_preproc_for_classification:n_estimators': 100, 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.032719158639429445},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.040000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'gradient_boosting', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'liblinear_svc_preprocessor', 'classifier:gradient_boosting:early_stop': 'train', 'classifier:gradient_boosting:l2_regularization': 0.005326508887463406, 'classifier:gradient_boosting:learning_rate': 0.060800813211425456, 'classifier:gradient_boosting:loss': 'auto', 'classifier:gradient_boosting:max_bins': 255, 'classifier:gradient_boosting:max_depth': 'None', 'classifier:gradient_boosting:max_leaf_nodes': 6, 'classifier:gradient_boosting:min_samples_leaf': 5, 'classifier:gradient_boosting:scoring': 'loss', 'classifier:gradient_boosting:tol': 1e-07, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'median', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'minmax', 'feature_preprocessor:liblinear_svc_preprocessor:C': 13.550960330919455, 'feature_preprocessor:liblinear_svc_preprocessor:dual': 'False', 'feature_preprocessor:liblinear_svc_preprocessor:fit_intercept': 'True', 'feature_preprocessor:liblinear_svc_preprocessor:intercept_scaling': 1, 'feature_preprocessor:liblinear_svc_preprocessor:loss': 'squared_hinge', 'feature_preprocessor:liblinear_svc_preprocessor:multi_class': 'ovr', 'feature_preprocessor:liblinear_svc_preprocessor:penalty': 'l1', 'feature_preprocessor:liblinear_svc_preprocessor:tol': 1.2958033930435781e-05, 'classifier:gradient_boosting:n_iter_no_change': 5},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.020000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'extra_trees', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'select_rates_classification', 'classifier:extra_trees:bootstrap': 'False', 'classifier:extra_trees:criterion': 'entropy', 'classifier:extra_trees:max_depth': 'None', 'classifier:extra_trees:max_features': 0.562561668029056, 'classifier:extra_trees:max_leaf_nodes': 'None', 'classifier:extra_trees:min_impurity_decrease': 0.0, 'classifier:extra_trees:min_samples_leaf': 2, 'classifier:extra_trees:min_samples_split': 15, 'classifier:extra_trees:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'median', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'none', 'feature_preprocessor:select_rates_classification:alpha': 0.32106218519214325, 'feature_preprocessor:select_rates_classification:score_func': 'chi2', 'feature_preprocessor:select_rates_classification:mode': 'fwe'},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.020000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'extra_trees', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'polynomial', 'classifier:extra_trees:bootstrap': 'False', 'classifier:extra_trees:criterion': 'entropy', 'classifier:extra_trees:max_depth': 'None', 'classifier:extra_trees:max_features': 0.993803313878608, 'classifier:extra_trees:max_leaf_nodes': 'None', 'classifier:extra_trees:min_impurity_decrease': 0.0, 'classifier:extra_trees:min_samples_leaf': 2, 'classifier:extra_trees:min_samples_split': 20, 'classifier:extra_trees:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'median', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'robust_scaler', 'feature_preprocessor:polynomial:degree': 2, 'feature_preprocessor:polynomial:include_bias': 'True', 'feature_preprocessor:polynomial:interaction_only': 'True', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.41826215858914706, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max': 0.7305615609807856, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min': 0.25595970768123566},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.020000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'gradient_boosting', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'extra_trees_preproc_for_classification', 'classifier:gradient_boosting:early_stop': 'valid', 'classifier:gradient_boosting:l2_regularization': 8.057778875694463e-05, 'classifier:gradient_boosting:learning_rate': 0.09179220974965213, 'classifier:gradient_boosting:loss': 'auto', 'classifier:gradient_boosting:max_bins': 255, 'classifier:gradient_boosting:max_depth': 'None', 'classifier:gradient_boosting:max_leaf_nodes': 200, 'classifier:gradient_boosting:min_samples_leaf': 20, 'classifier:gradient_boosting:scoring': 'loss', 'classifier:gradient_boosting:tol': 1e-07, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'most_frequent', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'quantile_transformer', 'feature_preprocessor:extra_trees_preproc_for_classification:bootstrap': 'False', 'feature_preprocessor:extra_trees_preproc_for_classification:criterion': 'gini', 'feature_preprocessor:extra_trees_preproc_for_classification:max_depth': 'None', 'feature_preprocessor:extra_trees_preproc_for_classification:max_features': 0.9984367650965825, 'feature_preprocessor:extra_trees_preproc_for_classification:max_leaf_nodes': 'None', 'feature_preprocessor:extra_trees_preproc_for_classification:min_impurity_decrease': 0.0, 'feature_preprocessor:extra_trees_preproc_for_classification:min_samples_leaf': 13, 'feature_preprocessor:extra_trees_preproc_for_classification:min_samples_split': 18, 'feature_preprocessor:extra_trees_preproc_for_classification:min_weight_fraction_leaf': 0.0, 'feature_preprocessor:extra_trees_preproc_for_classification:n_estimators': 100, 'classifier:gradient_boosting:n_iter_no_change': 18, 'classifier:gradient_boosting:validation_fraction': 0.14295295806077554, 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.008542188583124829, 'data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles': 1102, 'data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution': 'uniform'},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.020000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'random_forest', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'random_trees_embedding', 'classifier:random_forest:bootstrap': 'True', 'classifier:random_forest:criterion': 'entropy', 'classifier:random_forest:max_depth': 'None', 'classifier:random_forest:max_features': 0.5, 'classifier:random_forest:max_leaf_nodes': 'None', 'classifier:random_forest:min_impurity_decrease': 0.0, 'classifier:random_forest:min_samples_leaf': 1, 'classifier:random_forest:min_samples_split': 2, 'classifier:random_forest:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'no_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'no_coalescense', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'median', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'quantile_transformer', 'feature_preprocessor:random_trees_embedding:bootstrap': 'True', 'feature_preprocessor:random_trees_embedding:max_depth': 5, 'feature_preprocessor:random_trees_embedding:max_leaf_nodes': 'None', 'feature_preprocessor:random_trees_embedding:min_samples_leaf': 1, 'feature_preprocessor:random_trees_embedding:min_samples_split': 2, 'feature_preprocessor:random_trees_embedding:min_weight_fraction_leaf': 1.0, 'feature_preprocessor:random_trees_embedding:n_estimators': 10, 'data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles': 937, 'data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution': 'uniform'},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    ]
+    {   2: {   'balancing': Balancing(random_state=1),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78fc1f700>,
+               'cost': 0.021276595744680882,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78a6050d0>,
+               'ensemble_weight': 0.06,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c944df0>,
+               'model_id': 2,
+               'rank': 5,
+               'sklearn_classifier': RandomForestClassifier(max_features=5, n_estimators=64, n_jobs=1,
+                           random_state=1, warm_start=True)},
+        4: {   'balancing': Balancing(random_state=1),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc784bf13a0>,
+               'cost': 0.014184397163120588,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc79068bb80>,
+               'ensemble_weight': 0.14,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc784bf10a0>,
+               'model_id': 4,
+               'rank': 4,
+               'sklearn_classifier': ExtraTreesClassifier(max_features=34, min_samples_leaf=3, min_samples_split=11,
+                         n_estimators=512, n_jobs=1, random_state=1,
+                         warm_start=True)},
+        6: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc784871e50>,
+               'cost': 0.04255319148936165,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78580dac0>,
+               'ensemble_weight': 0.12,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c0ad400>,
+               'model_id': 6,
+               'rank': 10,
+               'sklearn_classifier': ExtraTreesClassifier(max_features=9, min_samples_split=10, n_estimators=64,
+                         n_jobs=1, random_state=1, warm_start=True)},
+        11: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc787e35400>,
+                'cost': 0.021276595744680882,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78af3f1c0>,
+                'ensemble_weight': 0.08,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc787e35280>,
+                'model_id': 11,
+                'rank': 7,
+                'sklearn_classifier': HistGradientBoostingClassifier(early_stopping=True,
+                                   l2_regularization=3.387912939529945e-10,
+                                   learning_rate=0.30755227194768237, max_iter=128,
+                                   max_leaf_nodes=60, min_samples_leaf=39,
+                                   n_iter_no_change=18, random_state=1,
+                                   validation_fraction=None, warm_start=True)},
+        13: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc79039a790>,
+                'cost': 0.04255319148936165,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78a4bed00>,
+                'ensemble_weight': 0.02,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78f8a2b80>,
+                'model_id': 13,
+                'rank': 11,
+                'sklearn_classifier': RandomForestClassifier(criterion='entropy', max_features=3, n_estimators=64,
+                           n_jobs=1, random_state=1, warm_start=True)},
+        16: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78582ed30>,
+                'cost': 0.028368794326241176,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78a7e35b0>,
+                'ensemble_weight': 0.06,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c482ca0>,
+                'model_id': 16,
+                'rank': 8,
+                'sklearn_classifier': HistGradientBoostingClassifier(early_stopping=True,
+                                   l2_regularization=8.057778875694463e-05,
+                                   learning_rate=0.09179220974965213, max_iter=64,
+                                   max_leaf_nodes=200, n_iter_no_change=18,
+                                   random_state=1,
+                                   validation_fraction=0.14295295806077554,
+                                   warm_start=True)},
+        18: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc79039a070>,
+                'cost': 0.04255319148936165,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78bc2d7c0>,
+                'ensemble_weight': 0.02,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc79039a820>,
+                'model_id': 18,
+                'rank': 12,
+                'sklearn_classifier': ExtraTreesClassifier(criterion='entropy', max_features=414, min_samples_leaf=2,
+                         min_samples_split=19, n_estimators=64, n_jobs=1,
+                         random_state=1, warm_start=True)},
+        19: {   'balancing': Balancing(random_state=1),
+                'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78ba373a0>,
+                'cost': 0.028368794326241176,
+                'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7850a8be0>,
+                'ensemble_weight': 0.02,
+                'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78582efa0>,
+                'model_id': 19,
+                'rank': 9,
+                'sklearn_classifier': HistGradientBoostingClassifier(early_stopping=False, l2_regularization=1e-10,
+                                   learning_rate=0.16262682406125173, max_iter=64,
+                                   max_leaf_nodes=66, n_iter_no_change=0,
+                                   random_state=1, validation_fraction=None,
+                                   warm_start=True)}}
     auto-sklearn results:
       Dataset name: breast_cancer
       Metric: accuracy
       Best validation score: 0.985816
-      Number of target algorithm runs: 19
-      Number of successful target algorithm runs: 19
+      Number of target algorithm runs: 24
+      Number of successful target algorithm runs: 24
       Number of crashed target algorithm runs: 0
       Number of target algorithms that exceeded the time limit: 0
       Number of target algorithms that exceeded the memory limit: 0
@@ -919,7 +1226,7 @@ subsamples otherwise
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 3 minutes  12.940 seconds)
+   **Total running time of the script:** ( 2 minutes  59.204 seconds)
 
 
 .. _sphx_glr_download_examples_60_search_example_successive_halving.py:
diff --git a/master/_sources/examples/60_search/sg_execution_times.rst.txt b/master/_sources/examples/60_search/sg_execution_times.rst.txt
index 7714e0b994..6cc6e34c27 100644
--- a/master/_sources/examples/60_search/sg_execution_times.rst.txt
+++ b/master/_sources/examples/60_search/sg_execution_times.rst.txt
@@ -5,18 +5,16 @@
 
 Computation times
 =================
-**10:48.731** total execution time for **examples_60_search** files:
+**09:31.567** total execution time for **examples_60_search** files:
 
-+--------------------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_examples_60_search_example_successive_halving.py` (``example_successive_halving.py``)                           | 03:12.940 | 0.0 MB |
-+--------------------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_examples_60_search_example_parallel_n_jobs.py` (``example_parallel_n_jobs.py``)                                 | 02:24.442 | 0.0 MB |
-+--------------------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_examples_60_search_example_sequential.py` (``example_sequential.py``)                                           | 02:04.152 | 0.0 MB |
-+--------------------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_examples_60_search_example_random_search.py` (``example_random_search.py``)                                     | 01:52.038 | 0.0 MB |
-+--------------------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_examples_60_search_example_parallel_manual_spawning_cli.py` (``example_parallel_manual_spawning_cli.py``)       | 00:39.363 | 0.0 MB |
-+--------------------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_examples_60_search_example_parallel_manual_spawning_python.py` (``example_parallel_manual_spawning_python.py``) | 00:35.796 | 0.0 MB |
-+--------------------------------------------------------------------------------------------------------------------------------+-----------+--------+
++--------------------------------------------------------------------------------------------------------------------------+-----------+--------+
+| :ref:`sphx_glr_examples_60_search_example_successive_halving.py` (``example_successive_halving.py``)                     | 02:59.204 | 0.0 MB |
++--------------------------------------------------------------------------------------------------------------------------+-----------+--------+
+| :ref:`sphx_glr_examples_60_search_example_sequential.py` (``example_sequential.py``)                                     | 02:01.878 | 0.0 MB |
++--------------------------------------------------------------------------------------------------------------------------+-----------+--------+
+| :ref:`sphx_glr_examples_60_search_example_parallel_n_jobs.py` (``example_parallel_n_jobs.py``)                           | 01:57.601 | 0.0 MB |
++--------------------------------------------------------------------------------------------------------------------------+-----------+--------+
+| :ref:`sphx_glr_examples_60_search_example_random_search.py` (``example_random_search.py``)                               | 01:53.718 | 0.0 MB |
++--------------------------------------------------------------------------------------------------------------------------+-----------+--------+
+| :ref:`sphx_glr_examples_60_search_example_parallel_manual_spawning_cli.py` (``example_parallel_manual_spawning_cli.py``) | 00:39.167 | 0.0 MB |
++--------------------------------------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/master/_sources/examples/80_extending/example_extending_classification.rst.txt b/master/_sources/examples/80_extending/example_extending_classification.rst.txt
index 679c058bdb..cf5fa5edcc 100644
--- a/master/_sources/examples/80_extending/example_extending_classification.rst.txt
+++ b/master/_sources/examples/80_extending/example_extending_classification.rst.txt
@@ -25,10 +25,11 @@ Extending Auto-Sklearn with Classification Component
 The following example demonstrates how to create a new classification
 component for using in auto-sklearn.
 
-.. GENERATED FROM PYTHON SOURCE LINES 9-26
+.. GENERATED FROM PYTHON SOURCE LINES 9-27
 
 .. code-block:: default
 
+    from pprint import pprint
 
     from ConfigSpace.configuration_space import ConfigurationSpace
     from ConfigSpace.hyperparameters import CategoricalHyperparameter, \
@@ -53,12 +54,12 @@ component for using in auto-sklearn.
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 27-29
+.. GENERATED FROM PYTHON SOURCE LINES 28-30
 
 Create MLP classifier component for auto-sklearn
 ================================================
 
-.. GENERATED FROM PYTHON SOURCE LINES 29-122
+.. GENERATED FROM PYTHON SOURCE LINES 30-123
 
 .. code-block:: default
 
@@ -177,12 +178,12 @@ Create MLP classifier component for auto-sklearn
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 123-125
+.. GENERATED FROM PYTHON SOURCE LINES 124-126
 
 Data Loading
 ============
 
-.. GENERATED FROM PYTHON SOURCE LINES 125-129
+.. GENERATED FROM PYTHON SOURCE LINES 126-130
 
 .. code-block:: default
 
@@ -197,12 +198,12 @@ Data Loading
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 130-132
+.. GENERATED FROM PYTHON SOURCE LINES 131-133
 
 Fit MLP classifier to the data
 ==============================
 
-.. GENERATED FROM PYTHON SOURCE LINES 132-146
+.. GENERATED FROM PYTHON SOURCE LINES 133-147
 
 .. code-block:: default
 
@@ -239,19 +240,19 @@ Fit MLP classifier to the data
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 147-149
+.. GENERATED FROM PYTHON SOURCE LINES 148-150
 
 Print test accuracy and statistics
 ==================================
 
-.. GENERATED FROM PYTHON SOURCE LINES 149-153
+.. GENERATED FROM PYTHON SOURCE LINES 150-154
 
 .. code-block:: default
 
 
     y_pred = clf.predict(X_test)
     print("accuracy: ", sklearn.metrics.accuracy_score(y_pred, y_test))
-    print(clf.show_models())
+    pprint(clf.show_models(), indent=4)
 
 
 
@@ -262,24 +263,26 @@ Print test accuracy and statistics
 
  .. code-block:: none
 
-    accuracy:  0.951048951048951
-    [(0.580000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'MLPClassifier', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:MLPClassifier:activation': 'relu', 'classifier:MLPClassifier:alpha': 0.0001, 'classifier:MLPClassifier:hidden_layer_depth': 1, 'classifier:MLPClassifier:num_nodes_per_layer': 32, 'classifier:MLPClassifier:solver': 'adam', 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.01},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.420000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'MLPClassifier', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'select_rates_classification', 'classifier:MLPClassifier:activation': 'relu', 'classifier:MLPClassifier:alpha': 0.054648061621061846, 'classifier:MLPClassifier:hidden_layer_depth': 1, 'classifier:MLPClassifier:num_nodes_per_layer': 187, 'classifier:MLPClassifier:solver': 'adam', 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'robust_scaler', 'feature_preprocessor:select_rates_classification:alpha': 0.17677801820619737, 'feature_preprocessor:select_rates_classification:score_func': 'mutual_info_classif', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.06846183238648647, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max': 0.8465036896999549, 'data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min': 0.14555017892884228},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    ]
+    accuracy:  0.986013986013986
+    {   2: {   'balancing': Balancing(random_state=1),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78f92cc70>,
+               'cost': 0.049645390070921946,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc786854e20>,
+               'ensemble_weight': 0.52,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78a9eab80>,
+               'model_id': 2,
+               'rank': 1,
+               'sklearn_classifier': MLPClassifier(hidden_layer_sizes=(32,), random_state=1)},
+        5: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78b2896a0>,
+               'cost': 0.1063829787234043,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc786854430>,
+               'ensemble_weight': 0.48,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c93bf10>,
+               'model_id': 5,
+               'rank': 2,
+               'sklearn_classifier': MLPClassifier(alpha=0.054648061621061846, hidden_layer_sizes=(187,),
+                  random_state=1)}}
 
 
 
@@ -287,7 +290,7 @@ Print test accuracy and statistics
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 0 minutes  18.369 seconds)
+   **Total running time of the script:** ( 0 minutes  12.500 seconds)
 
 
 .. _sphx_glr_download_examples_80_extending_example_extending_classification.py:
diff --git a/master/_sources/examples/80_extending/example_extending_data_preprocessor.rst.txt b/master/_sources/examples/80_extending/example_extending_data_preprocessor.rst.txt
index bfef02d2b0..d8ad7c8a70 100644
--- a/master/_sources/examples/80_extending/example_extending_data_preprocessor.rst.txt
+++ b/master/_sources/examples/80_extending/example_extending_data_preprocessor.rst.txt
@@ -24,10 +24,11 @@ Extending Auto-Sklearn with Data Preprocessor Component
 
 The following example demonstrates how to turn off data preprocessing step in auto-skearn.
 
-.. GENERATED FROM PYTHON SOURCE LINES 8-19
+.. GENERATED FROM PYTHON SOURCE LINES 8-20
 
 .. code-block:: default
 
+    from pprint import pprint
 
     import autosklearn.classification
     import autosklearn.pipeline.components.data_preprocessing
@@ -46,12 +47,12 @@ The following example demonstrates how to turn off data preprocessing step in au
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 20-22
+.. GENERATED FROM PYTHON SOURCE LINES 21-23
 
 Create NoPreprocessing component for auto-sklearn
 =================================================
 
-.. GENERATED FROM PYTHON SOURCE LINES 22-59
+.. GENERATED FROM PYTHON SOURCE LINES 23-60
 
 .. code-block:: default
 
@@ -99,12 +100,12 @@ Create NoPreprocessing component for auto-sklearn
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 60-62
+.. GENERATED FROM PYTHON SOURCE LINES 61-63
 
 Create dataset
 ==============
 
-.. GENERATED FROM PYTHON SOURCE LINES 62-66
+.. GENERATED FROM PYTHON SOURCE LINES 63-67
 
 .. code-block:: default
 
@@ -119,12 +120,12 @@ Create dataset
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 67-69
+.. GENERATED FROM PYTHON SOURCE LINES 68-70
 
 Fit the model without performing data preprocessing
 ===================================================
 
-.. GENERATED FROM PYTHON SOURCE LINES 69-86
+.. GENERATED FROM PYTHON SOURCE LINES 70-87
 
 .. code-block:: default
 
@@ -156,12 +157,12 @@ Fit the model without performing data preprocessing
  .. code-block:: none
 
     auto-sklearn results:
-      Dataset name: 2613be54-419a-11ec-8a32-1b4c952643c3
+      Dataset name: 3a1d2162-7dfe-11ec-875f-3137a7553b01
       Metric: accuracy
-      Best validation score: 0.936170
+      Best validation score: 0.964539
       Number of target algorithm runs: 5
-      Number of successful target algorithm runs: 5
-      Number of crashed target algorithm runs: 0
+      Number of successful target algorithm runs: 4
+      Number of crashed target algorithm runs: 1
       Number of target algorithms that exceeded the time limit: 0
       Number of target algorithms that exceeded the memory limit: 0
 
@@ -169,19 +170,19 @@ Fit the model without performing data preprocessing
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 87-89
+.. GENERATED FROM PYTHON SOURCE LINES 88-90
 
 Print prediction score and statistics
 =====================================
 
-.. GENERATED FROM PYTHON SOURCE LINES 89-93
+.. GENERATED FROM PYTHON SOURCE LINES 90-94
 
 .. code-block:: default
 
 
     y_pred = clf.predict(X_test)
     print("accuracy: ", sklearn.metrics.accuracy_score(y_pred, y_test))
-    print(clf.show_models())
+    pprint(clf.show_models(), indent=4)
 
 
 
@@ -193,31 +194,37 @@ Print prediction score and statistics
  .. code-block:: none
 
     accuracy:  0.9370629370629371
-    [(0.540000, SimpleClassificationPipeline({'balancing:strategy': 'weighting', 'classifier:__choice__': 'passive_aggressive', 'data_preprocessor:__choice__': 'NoPreprocessing', 'feature_preprocessor:__choice__': 'kernel_pca', 'classifier:passive_aggressive:C': 7.328568999403367e-05, 'classifier:passive_aggressive:average': 'False', 'classifier:passive_aggressive:fit_intercept': 'True', 'classifier:passive_aggressive:loss': 'hinge', 'classifier:passive_aggressive:tol': 0.02258105767523384, 'feature_preprocessor:kernel_pca:kernel': 'cosine', 'feature_preprocessor:kernel_pca:n_components': 503},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.280000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'random_forest', 'data_preprocessor:__choice__': 'NoPreprocessing', 'feature_preprocessor:__choice__': 'no_preprocessing', 'classifier:random_forest:bootstrap': 'True', 'classifier:random_forest:criterion': 'gini', 'classifier:random_forest:max_depth': 'None', 'classifier:random_forest:max_features': 0.5, 'classifier:random_forest:max_leaf_nodes': 'None', 'classifier:random_forest:min_impurity_decrease': 0.0, 'classifier:random_forest:min_samples_leaf': 1, 'classifier:random_forest:min_samples_split': 2, 'classifier:random_forest:min_weight_fraction_leaf': 0.0},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    (0.180000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'lda', 'data_preprocessor:__choice__': 'NoPreprocessing', 'feature_preprocessor:__choice__': 'select_rates_classification', 'classifier:lda:shrinkage': 'auto', 'classifier:lda:tol': 0.00010495766342828794, 'feature_preprocessor:select_rates_classification:alpha': 0.4886183647550176, 'feature_preprocessor:select_rates_classification:score_func': 'mutual_info_classif'},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    ]
+    {   2: {   'balancing': Balancing(random_state=1),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78f84daf0>,
+               'cost': 0.03546099290780147,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78fdee5e0>,
+               'ensemble_weight': 0.44,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78fdeec10>,
+               'model_id': 2,
+               'rank': 1,
+               'sklearn_classifier': RandomForestClassifier(max_features=5, n_estimators=512, n_jobs=1,
+                           random_state=1, warm_start=True)},
+        4: {   'balancing': Balancing(random_state=1, strategy='weighting'),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78c4f9700>,
+               'cost': 0.36170212765957444,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7867d8340>,
+               'ensemble_weight': 0.4,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c59d910>,
+               'model_id': 4,
+               'rank': 3,
+               'sklearn_classifier': PassiveAggressiveClassifier(C=7.328568999403367e-05, max_iter=16,
+                                random_state=1, tol=0.02258105767523384,
+                                warm_start=True)},
+        6: {   'balancing': Balancing(random_state=1),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc79015a340>,
+               'cost': 0.11347517730496459,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78f84d070>,
+               'ensemble_weight': 0.16,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c98fe20>,
+               'model_id': 6,
+               'rank': 2,
+               'sklearn_classifier': LinearDiscriminantAnalysis(shrinkage='auto', solver='lsqr',
+                               tol=0.00010495766342828794)}}
 
 
 
@@ -225,7 +232,7 @@ Print prediction score and statistics
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 0 minutes  23.312 seconds)
+   **Total running time of the script:** ( 0 minutes  15.803 seconds)
 
 
 .. _sphx_glr_download_examples_80_extending_example_extending_data_preprocessor.py:
diff --git a/master/_sources/examples/80_extending/example_extending_preprocessor.rst.txt b/master/_sources/examples/80_extending/example_extending_preprocessor.rst.txt
index 20cda0c1c6..2541fb4d64 100644
--- a/master/_sources/examples/80_extending/example_extending_preprocessor.rst.txt
+++ b/master/_sources/examples/80_extending/example_extending_preprocessor.rst.txt
@@ -26,10 +26,11 @@ The following example demonstrates how to create a wrapper around the linear
 discriminant analysis (LDA) algorithm from sklearn and use it as a preprocessor
 in auto-sklearn.
 
-.. GENERATED FROM PYTHON SOURCE LINES 10-28
+.. GENERATED FROM PYTHON SOURCE LINES 10-29
 
 .. code-block:: default
 
+    from pprint import pprint
 
     from ConfigSpace.configuration_space import ConfigurationSpace
     from ConfigSpace.hyperparameters import UniformFloatHyperparameter, CategoricalHyperparameter
@@ -55,12 +56,12 @@ in auto-sklearn.
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 29-31
+.. GENERATED FROM PYTHON SOURCE LINES 30-32
 
 Create LDA component for auto-sklearn
 =====================================
 
-.. GENERATED FROM PYTHON SOURCE LINES 31-97
+.. GENERATED FROM PYTHON SOURCE LINES 32-98
 
 .. code-block:: default
 
@@ -137,12 +138,12 @@ Create LDA component for auto-sklearn
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 98-100
+.. GENERATED FROM PYTHON SOURCE LINES 99-101
 
 Create dataset
 ==============
 
-.. GENERATED FROM PYTHON SOURCE LINES 100-104
+.. GENERATED FROM PYTHON SOURCE LINES 101-105
 
 .. code-block:: default
 
@@ -157,12 +158,12 @@ Create dataset
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 105-107
+.. GENERATED FROM PYTHON SOURCE LINES 106-108
 
 Configuration space
 ===================
 
-.. GENERATED FROM PYTHON SOURCE LINES 107-111
+.. GENERATED FROM PYTHON SOURCE LINES 108-112
 
 .. code-block:: default
 
@@ -192,12 +193,12 @@ Configuration space
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 112-114
+.. GENERATED FROM PYTHON SOURCE LINES 113-115
 
 Fit the model using LDA as preprocessor
 =======================================
 
-.. GENERATED FROM PYTHON SOURCE LINES 114-127
+.. GENERATED FROM PYTHON SOURCE LINES 115-128
 
 .. code-block:: default
 
@@ -233,19 +234,19 @@ Fit the model using LDA as preprocessor
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 128-130
+.. GENERATED FROM PYTHON SOURCE LINES 129-131
 
 Print prediction score and statistics
 =====================================
 
-.. GENERATED FROM PYTHON SOURCE LINES 130-134
+.. GENERATED FROM PYTHON SOURCE LINES 131-135
 
 .. code-block:: default
 
 
     y_pred = clf.predict(X_test)
     print("accuracy: ", sklearn.metrics.accuracy_score(y_pred, y_test))
-    print(clf.show_models())
+    pprint(clf.show_models(), indent=4)
 
 
 
@@ -256,16 +257,17 @@ Print prediction score and statistics
 
  .. code-block:: none
 
-    accuracy:  0.9370629370629371
-    [(1.000000, SimpleClassificationPipeline({'balancing:strategy': 'none', 'classifier:__choice__': 'random_forest', 'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'LDA', 'classifier:random_forest:bootstrap': 'True', 'classifier:random_forest:criterion': 'gini', 'classifier:random_forest:max_depth': 'None', 'classifier:random_forest:max_features': 0.5, 'classifier:random_forest:max_leaf_nodes': 'None', 'classifier:random_forest:min_impurity_decrease': 0.0, 'classifier:random_forest:min_samples_leaf': 1, 'classifier:random_forest:min_samples_split': 2, 'classifier:random_forest:min_weight_fraction_leaf': 0.0, 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize', 'feature_preprocessor:LDA:solver': 'svd', 'feature_preprocessor:LDA:tol': 0.0001, 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.01},
-    dataset_properties={
-      'task': 1,
-      'sparse': False,
-      'multilabel': False,
-      'multiclass': False,
-      'target_type': 'classification',
-      'signed': False})),
-    ]
+    accuracy:  0.958041958041958
+    {   2: {   'balancing': Balancing(random_state=1),
+               'classifier': <autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78c384d90>,
+               'cost': 0.021276595744680882,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78af3f910>,
+               'ensemble_weight': 1.0,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c8a0250>,
+               'model_id': 2,
+               'rank': 1,
+               'sklearn_classifier': RandomForestClassifier(max_features=1, n_estimators=512, n_jobs=1,
+                           random_state=1, warm_start=True)}}
 
 
 
@@ -273,7 +275,7 @@ Print prediction score and statistics
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 0 minutes  18.761 seconds)
+   **Total running time of the script:** ( 0 minutes  17.499 seconds)
 
 
 .. _sphx_glr_download_examples_80_extending_example_extending_preprocessor.py:
diff --git a/master/_sources/examples/80_extending/example_extending_regression.rst.txt b/master/_sources/examples/80_extending/example_extending_regression.rst.txt
index 1a8b939316..ef60c99e44 100644
--- a/master/_sources/examples/80_extending/example_extending_regression.rst.txt
+++ b/master/_sources/examples/80_extending/example_extending_regression.rst.txt
@@ -25,10 +25,11 @@ Extending Auto-Sklearn with Regression Component
 The following example demonstrates how to create a new regression
 component for using in auto-sklearn.
 
-.. GENERATED FROM PYTHON SOURCE LINES 9-26
+.. GENERATED FROM PYTHON SOURCE LINES 9-27
 
 .. code-block:: default
 
+    from pprint import pprint
 
     from ConfigSpace.configuration_space import ConfigurationSpace
     from ConfigSpace.hyperparameters import UniformFloatHyperparameter, \
@@ -53,12 +54,12 @@ component for using in auto-sklearn.
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 27-29
+.. GENERATED FROM PYTHON SOURCE LINES 28-30
 
 Implement kernel ridge regression component for auto-sklearn
 ============================================================
 
-.. GENERATED FROM PYTHON SOURCE LINES 29-111
+.. GENERATED FROM PYTHON SOURCE LINES 30-112
 
 .. code-block:: default
 
@@ -169,12 +170,12 @@ Implement kernel ridge regression component for auto-sklearn
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 112-114
+.. GENERATED FROM PYTHON SOURCE LINES 113-115
 
 Generate data
 =============
 
-.. GENERATED FROM PYTHON SOURCE LINES 114-118
+.. GENERATED FROM PYTHON SOURCE LINES 115-119
 
 .. code-block:: default
 
@@ -189,12 +190,12 @@ Generate data
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 119-121
+.. GENERATED FROM PYTHON SOURCE LINES 120-122
 
 Fit the model using KRR
 =======================
 
-.. GENERATED FROM PYTHON SOURCE LINES 121-135
+.. GENERATED FROM PYTHON SOURCE LINES 122-136
 
 .. code-block:: default
 
@@ -231,18 +232,18 @@ Fit the model using KRR
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 136-138
+.. GENERATED FROM PYTHON SOURCE LINES 137-139
 
 Print prediction score and statistics
 =====================================
 
-.. GENERATED FROM PYTHON SOURCE LINES 138-141
+.. GENERATED FROM PYTHON SOURCE LINES 139-142
 
 .. code-block:: default
 
     y_pred = reg.predict(X_test)
     print("r2 score: ", sklearn.metrics.r2_score(y_pred, y_test))
-    print(reg.show_models())
+    pprint(reg.show_models(), indent=4)
 
 
 
@@ -253,15 +254,15 @@ Print prediction score and statistics
 
  .. code-block:: none
 
-    r2 score:  0.1280839244790386
-    [(1.000000, SimpleRegressionPipeline({'data_preprocessor:__choice__': 'feature_type', 'feature_preprocessor:__choice__': 'no_preprocessing', 'regressor:__choice__': 'KernelRidgeRegression', 'data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__': 'one_hot_encoding', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__': 'minority_coalescer', 'data_preprocessor:feature_type:numerical_transformer:imputation:strategy': 'mean', 'data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__': 'standardize', 'regressor:KernelRidgeRegression:alpha': 1.0, 'regressor:KernelRidgeRegression:gamma': 0.1, 'regressor:KernelRidgeRegression:kernel': 'polynomial', 'data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction': 0.01, 'regressor:KernelRidgeRegression:coef0': 1.0, 'regressor:KernelRidgeRegression:degree': 3},
-    dataset_properties={
-      'task': 4,
-      'sparse': False,
-      'multioutput': False,
-      'target_type': 'regression',
-      'signed': False})),
-    ]
+    r2 score:  0.11103398852619628
+    {   2: {   'cost': 0.65199313632313,
+               'data_preprocessor': <autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7863feee0>,
+               'ensemble_weight': 1.0,
+               'feature_preprocessor': <autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78f6db9a0>,
+               'model_id': 2,
+               'rank': 1,
+               'regressor': <autosklearn.pipeline.components.regression.RegressorChoice object at 0x7fc78a826580>,
+               'sklearn_regressor': KernelRidge(alpha=1.0, coef0=1.0, gamma=0.1, kernel='polynomial')}}
 
 
 
@@ -269,7 +270,7 @@ Print prediction score and statistics
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 0 minutes  11.876 seconds)
+   **Total running time of the script:** ( 0 minutes  11.558 seconds)
 
 
 .. _sphx_glr_download_examples_80_extending_example_extending_regression.py:
diff --git a/master/_sources/examples/80_extending/example_restrict_number_of_hyperparameters.rst.txt b/master/_sources/examples/80_extending/example_restrict_number_of_hyperparameters.rst.txt
index d246c450d7..8e121e0ccc 100644
--- a/master/_sources/examples/80_extending/example_restrict_number_of_hyperparameters.rst.txt
+++ b/master/_sources/examples/80_extending/example_restrict_number_of_hyperparameters.rst.txt
@@ -628,7 +628,7 @@ Print the configuration space
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 0 minutes  7.799 seconds)
+   **Total running time of the script:** ( 0 minutes  6.778 seconds)
 
 
 .. _sphx_glr_download_examples_80_extending_example_restrict_number_of_hyperparameters.py:
diff --git a/master/_sources/examples/80_extending/sg_execution_times.rst.txt b/master/_sources/examples/80_extending/sg_execution_times.rst.txt
index e4ab7fd0bf..ad4f34c370 100644
--- a/master/_sources/examples/80_extending/sg_execution_times.rst.txt
+++ b/master/_sources/examples/80_extending/sg_execution_times.rst.txt
@@ -5,16 +5,16 @@
 
 Computation times
 =================
-**01:20.118** total execution time for **examples_80_extending** files:
+**01:04.137** total execution time for **examples_80_extending** files:
 
 +-----------------------------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_examples_80_extending_example_extending_data_preprocessor.py` (``example_extending_data_preprocessor.py``)               | 00:23.312 | 0.0 MB |
+| :ref:`sphx_glr_examples_80_extending_example_extending_preprocessor.py` (``example_extending_preprocessor.py``)                         | 00:17.499 | 0.0 MB |
 +-----------------------------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_examples_80_extending_example_extending_preprocessor.py` (``example_extending_preprocessor.py``)                         | 00:18.761 | 0.0 MB |
+| :ref:`sphx_glr_examples_80_extending_example_extending_data_preprocessor.py` (``example_extending_data_preprocessor.py``)               | 00:15.803 | 0.0 MB |
 +-----------------------------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_examples_80_extending_example_extending_classification.py` (``example_extending_classification.py``)                     | 00:18.369 | 0.0 MB |
+| :ref:`sphx_glr_examples_80_extending_example_extending_classification.py` (``example_extending_classification.py``)                     | 00:12.500 | 0.0 MB |
 +-----------------------------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_examples_80_extending_example_extending_regression.py` (``example_extending_regression.py``)                             | 00:11.876 | 0.0 MB |
+| :ref:`sphx_glr_examples_80_extending_example_extending_regression.py` (``example_extending_regression.py``)                             | 00:11.558 | 0.0 MB |
 +-----------------------------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_examples_80_extending_example_restrict_number_of_hyperparameters.py` (``example_restrict_number_of_hyperparameters.py``) | 00:07.799 | 0.0 MB |
+| :ref:`sphx_glr_examples_80_extending_example_restrict_number_of_hyperparameters.py` (``example_restrict_number_of_hyperparameters.py``) | 00:06.778 | 0.0 MB |
 +-----------------------------------------------------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/master/_sources/examples/index.rst.txt b/master/_sources/examples/index.rst.txt
index 93499569f1..352fdef141 100644
--- a/master/_sources/examples/index.rst.txt
+++ b/master/_sources/examples/index.rst.txt
@@ -425,27 +425,6 @@ Examples of changing the search procedure of Auto-sklearn.
 
    /examples/60_search/example_parallel_manual_spawning_cli
 
-.. raw:: html
-
-    <div class="sphx-glr-thumbcontainer" tooltip="*Auto-sklearn* uses `dask.distributed &lt;https://distributed.dask.org/en/latest/index.html&gt;`_ for...">
-
-.. only:: html
-
- .. figure:: /examples/60_search/images/thumb/sphx_glr_example_parallel_manual_spawning_python_thumb.png
-     :alt: Parallel Usage: Spawning workers from within Python
-
-     :ref:`sphx_glr_examples_60_search_example_parallel_manual_spawning_python.py`
-
-.. raw:: html
-
-    </div>
-
-
-.. toctree::
-   :hidden:
-
-   /examples/60_search/example_parallel_manual_spawning_python
-
 .. raw:: html
 
     <div class="sphx-glr-thumbcontainer" tooltip="A crucial feature of *auto-sklearn* is automatically optimizing the hyperparameters through SMA...">
diff --git a/master/_sources/faq.rst.txt b/master/_sources/faq.rst.txt
index d562eadc06..439e5c9be3 100644
--- a/master/_sources/faq.rst.txt
+++ b/master/_sources/faq.rst.txt
@@ -6,267 +6,518 @@
 FAQ
 ===
 
-Issues
-======
+General
+=======
 
-Auto-sklearn is extremely memory hungry in a sequential setting
----------------------------------------------------------------
+.. collapse:: <b>Where can I find examples on how to use auto-sklearn?</b>
 
-Auto-sklearn can appear very memory hungry (i.e. requiring a lot of memory for small datasets) due
-to the use of ``fork`` for creating new processes when running in sequential manner (if this
-happens in a parallel setting or if you pass your own dask client this is due to a different
-issue, see the other issues below).
+    We provide examples on using *auto-sklearn* for multiple use cases ranging from
+    simple classification to advanced uses such as feature importance, parallel runs
+    and customization. They can be found in the :ref:`sphx_glr_examples`.
 
-Let's go into some more detail and discuss how to fix it:
-Auto-sklearn executes each machine learning algorithm in its own process to be able to apply a
-memory limit and a time limit. To start such a process, Python gives three options: ``fork``,
-``forkserver`` and ``spawn``. The default ``fork`` copies the whole process memory into the
-subprocess. If the main process already uses 1.5GB of main memory and we apply a 3GB memory
-limit to Auto-sklearn, executing a machine learning pipeline is limited to use at most 1.5GB.
-We would have loved to use ``forkserver`` or ``spawn`` as the default option instead, which both
-copy only relevant data into the subprocess and thereby alleaviate the issue of eating up a lot
-of your main memory
-(and also do not suffer from potential deadlocks as ``fork`` does, see
-`here <https://pythonspeed.com/articles/python-multiprocessing/>`_),
-but they have the downside that code must be guarded by ``if __name__ == "__main__"`` or executed
-in a notebook, and we decided that we do not want to require this by default.
+.. collapse:: <b>What type of tasks can auto-sklearn tackle?</b>
 
-There are now two possible solutions:
+    *auto-sklearn* can accept targets for the following tasks (more details on `Sklearn algorithms <https://scikit-learn.org/stable/modules/multiclass.html>`_):
 
-1. Use Auto-sklearn in parallel: if you use Auto-sklean in parallel, it defaults to ``forkserver``
-   as the parallelization mechanism itself requires Auto-sklearn the code to be guarded. Please
-   find more information on how to do this in the following two examples:
+    * Binary Classification
+    * Multiclass Classification
+    * Multilabel Classification
+    * Regression
+    * Multioutput Regression
 
-   1. :ref:`sphx_glr_examples_60_search_example_parallel_n_jobs.py`
-   2. :ref:`sphx_glr_examples_60_search_example_parallel_manual_spawning_cli.py`
+    You can provide feature and target training pairs (X_train/y_train) to *auto-sklearn* to fit an
+    ensemble of pipelines as described in the next section. This X_train/y_train dataset must belong
+    to one of the supported formats: np.ndarray, pd.DataFrame, scipy.sparse.csr_matrix and python lists.
+    Optionally, you can measure the ability of this fitted model to generalize to unseen data by
+    providing an optional testing pair (X_test/Y_test). For further details, please refer to the
+    Example :ref:`sphx_glr_examples_40_advanced_example_pandas_train_test.py`.
+    Supported formats for these training and testing pairs are: np.ndarray,
+    pd.DataFrame, scipy.sparse.csr_matrix and python lists.
 
-   .. note::
+    If your data contains categorical values (in the features or targets), autosklearn will automatically encode your
+    data using a `sklearn.preprocessing.LabelEncoder <https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.LabelEncoder.html>`_
+    for unidimensional data and a `sklearn.preprocessing.OrdinalEncoder <https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.OrdinalEncoder.html>`_
+    for multidimensional data.
 
-       This requires all code to be guarded by ``if __name__ == "__main__"``.
+    Regarding the features, there are two methods to guide *auto-sklearn* to properly encode categorical columns:
 
-2. Pass a `dask client <https://distributed.dask.org/en/latest/client.html>`_. If the user passes
-   a dask client, Auto-sklearn can no longer assume that it runs in sequential mode and will use
-   a ``forkserver`` to start new processes.
+    * Providing a X_train/X_test numpy array with the optional flag feat_type. For further details, you
+      can check the Example :ref:`sphx_glr_examples_40_advanced_example_feature_types.py`.
+    * You can provide a pandas DataFrame, with properly formatted columns. If a column has numerical
+      dtype, *auto-sklearn* will not encode it and it will be passed directly to scikit-learn. If the
+      column has a categorical/boolean class, it will be encoded. If the column is of any other type
+      (Object or Timeseries), an error will be raised. For further details on how to properly encode
+      your data, you can check the Pandas Example
+      `Working with categorical data <https://pandas.pydata.org/pandas-docs/stable/user_guide/categorical.html>`_).
+      If you are working with time series, it is recommended that you follow this approach
+      `Working with time data <https://stats.stackexchange.com/questions/311494/>`_.
 
-   .. note::
+    Regarding the targets (y_train/y_test), if the task involves a classification problem, such features will be
+    automatically encoded. It is recommended to provide both y_train and y_test during fit, so that a common encoding
+    is created between these splits (if only y_train is provided during fit, the categorical encoder will not be able
+    to handle new classes that are exclusive to y_test). If the task is regression, no encoding happens on the
+    targets.
 
-       This requires all code to be guarded by ``if __name__ == "__main__"``.
+.. collapse:: <b>Where can I find slides and notebooks from talks and tutorials?</b>
 
-We therefore suggest using one of the above settings by default.
+    We provide resources for talks, tutorials and presentations on *auto-sklearn* under `auto-sklearn-talks <https://github.com/automl/auto-sklearn-talks>`_
 
-Auto-sklearn is extremely memory hungry in a parallel setting
--------------------------------------------------------------
+.. collapse:: <b>How should I cite auto-sklearn in a scientific publication?</b>
 
-When running Auto-sklearn in a parallel setting it starts new processes for evaluating machine
-learning models using the ``forkserver`` mechanism. Code that is in the main script and that is
-not guarded by ``if __name__ == "__main__"`` will be executed for each subprocess. If, for example,
-you are loading your dataset outside of the guarded code, your dataset will be loaded for each
-evaluation of a machine learning algorithm and thus blocking your RAM.
+    If you've used auto-sklearn in scientific publications, we would appreciate citations.
 
-We therefore suggest moving all code inside functions or the main block.
+    .. code-block::
 
-Auto-sklearn crashes with a segmentation fault
-----------------------------------------------
+        @inproceedings{feurer-neurips15a,
+            title     = {Efficient and Robust Automated Machine Learning},
+            author    = {Feurer, Matthias and Klein, Aaron and Eggensperger, Katharina  Springenberg, Jost and Blum, Manuel and Hutter, Frank},
+            booktitle = {Advances in Neural Information Processing Systems 28 (2015)},
+            pages     = {2962--2970},
+            year      = {2015}
+        }
 
-Please make sure that you have read and followed the :ref:`installation` section! In case
-everything is set up correctly, this is most likely due to the dependency
-`pyrfr <https://github.com/automl/random_forest_run>`_ not being compiled correctly. If this is the
-case please execute:
+    Or this, if you've used auto-sklearn 2.0 in your work:
 
-.. code:: python
+    .. code-block::
 
-    import pyrfr.regression as reg
-    data = reg.default_data_container(64)
+        @article{feurer-arxiv20a,
+            title     = {Auto-Sklearn 2.0: Hands-free AutoML via Meta-Learning},
+            author    = {Feurer, Matthias and Eggensperger, Katharina and Falkner, Stefan and Lindauer, Marius and Hutter, Frank},
+            booktitle = {arXiv:2007.04074 [cs.LG]},
+            year      = {2020}
+        }
 
-If this fails, the pyrfr dependency is most likely not compiled correctly. We advice you to do the
-following:
+.. collapse:: <b>I want to contribute. What can I do?</b>
 
-1. Check if you can use a pre-compiled version of the pyrfr to avoid compiling it yourself. We
-   provide pre-compiled versions of the pyrfr on `pypi <https://pypi.org/project/pyrfr/#files>`_.
-2. Check if the dependencies specified under :ref:`installation` are correctly installed,
-   especially that you have ``swig`` and a ``C++`` compiler.
-3. If you are not yet using Conda, consider using it; it simplifies installation of the correct
-   dependencies.
-4. Install correct build dependencies before installing the pyrfr, you can check the following
-   github issues for suggestions: `1025 <https://github.com/automl/auto-sklearn/issues/1025>`_,
-   `856 <https://github.com/automl/auto-sklearn/issues/856>`_
+    This sounds great. Please have a look at our `contribution guide <https://github.com/automl/auto-sklearn/blob/master/CONTRIBUTING.md>`_
 
-Log files and output
-====================
+.. collapse:: <b>I have a question which is not answered here. What should I do?</b>
 
-Where does Auto-sklearn output files by default?
-------------------------------------------------
+    Thanks a lot. We regularly update this section with questions from our issue tracker. So please use the
+    `issue tracker <https://github.com/automl/auto-sklearn/issues>`_
 
-*Auto-sklearn* heavily uses the hard drive to store temporary data, models and log files which can
-be used to inspect the behavior of Auto-sklearn. Each run of Auto-sklearn requires
-its own directory. If not provided by the user, *Auto-sklearn* requests a temporary directory from
-Python, which by default is located under ``/tmp`` and starts with ``autosklearn_tmp_`` followed
-by a random string. By default, this directory is deleted when the *Auto-sklearn* object is
-destroyed. If you want to keep these files you can pass the argument
-``delete_tmp_folder_after_terminate=True`` to the *Auto-sklearn* object.
+Resource Management
+===================
 
-The :class:`autosklearn.classification.AutoSklearnClassifier` and all other *auto-sklearn*
-estimators accept the argument ``tmp_directory`` which change where such output is written to.
+.. collapse:: <b>How should I set the time and memory limits?</b>
 
-There's an additional argument ``output_directory`` which can be passed to *Auto-sklearn* and it
-controls where test predictions of the ensemble are stored if the test set is passed to ``fit()``.
+    While *auto-sklearn* alleviates manual hyperparameter tuning, the user still
+    has to set memory and time limits. For most datasets a memory limit of 3GB or
+    6GB as found on most modern computers is sufficient. For the time limits it
+    is harder to give clear guidelines. If possible, a good default is a total
+    time limit of one day, and a time limit of 30 minutes for a single run.
 
-Auto-sklearn eats up all my disk space
---------------------------------------
+    Further guidelines can be found in
+    `auto-sklearn/issues/142 <https://github.com/automl/auto-sklearn/issues/142>`_.
 
-*Auto-sklearn* heavily uses the hard drive to store temporary data, models and log files which can
-be used to inspect the behavior of Auto-sklearn. By default, *Auto-sklearn* stores 50
-models and their predictions on the validation data (which is a subset of the training data in
-case of holdout and the full training data in case of cross-validation) on the hard drive.
-Redundant models and their predictions (i.e. when we have more than 50 models) are removed
-everytime the ensemble builder finishes an iteration, which means that the number of models stored
-on disk can temporarily be higher if a model is output while the ensemble builder is running.
+.. collapse:: <b>How many CPU cores does auto-sklearn use by default?</b>
 
-One can therefore change the number of models that will be stored on disk by passing an integer
-for the argument ``max_models_on_disc`` to *Auto-sklearn*, for example reduce the number of models
-stored on disk if you have space issues.
+    By default, *auto-sklearn* uses **one core**. See also :ref:`parallel` on how to configure this.
 
-As the number of models is only an indicator of the disk space used it is also possible to pass
-the memory in MB the models are allowed to use as a ``float`` (also via the ``max_models_on_disc``
-arguments). As above, this is rather a guideline on how much memory is used as redundant models
-are only removed from disk when the ensemble builder finishes an iteration.
+.. collapse:: <b>How can I run auto-sklearn in parallel?</b>
 
-.. note::
+    Nevertheless, *auto-sklearn* also supports parallel Bayesian optimization via the use of
+    `Dask.distributed  <https://distributed.dask.org/>`_. By providing the arguments ``n_jobs``
+    to the estimator construction, one can control the number of cores available to *auto-sklearn*
+    (As shown in the Example :ref:`sphx_glr_examples_60_search_example_parallel_n_jobs.py`).
+    Distributed processes are also supported by providing a custom client object to *auto-sklearn* like
+    in the Example: :ref:`sphx_glr_examples_60_search_example_parallel_manual_spawning_cli.py`. When
+    multiple cores are
+    available, *auto-sklearn* will create a worker per core, and use the available workers to both search
+    for better machine learning models as well as building an ensemble with them until the time resource
+    is exhausted.
 
-    Especially when running in parallel it can happen that multiple models are constructed during
-    one run of the ensemble builder and thus *Auto-sklearn* can exceed the given limit.
+    **Note:** *auto-sklearn* requires all workers to have access to a shared file system for storing training data and models.
 
-.. note::
+    *auto-sklearn* employs `threadpoolctl <https://github.com/joblib/threadpoolctl/>`_ to control the number of threads employed by scientific libraries like numpy or scikit-learn. This is done exclusively during the building procedure of models, not during inference. In particular, *auto-sklearn* allows each pipeline to use at most 1 thread during training. At predicting and scoring time this limitation is not enforced by *auto-sklearn*. You can control the number of resources
+    employed by the pipelines by setting the following variables in your environment, prior to running *auto-sklearn*:
 
-   These limits do only apply to models and their predictions, but not to other files stored in
-   the temporary directory such as the log files.
+    .. code-block:: shell-session
 
-Available machine learning models
-=================================
+        $ export OPENBLAS_NUM_THREADS=1
+        $ export MKL_NUM_THREADS=1
+        $ export OMP_NUM_THREADS=1
 
-Will non-scikit-learn models be added to Auto-sklearn?
-------------------------------------------------------
+    For further information about how scikit-learn handles multiprocessing, please check the `Parallelism, resource management, and configuration <https://scikit-learn.org/stable/computing/parallelism.html>`_ documentation from the library.
 
-The short answer: no.
 
-The long answer answer is a bit more nuanced: maintaining Auto-sklearn requires a lot of time and
-effort, which would grow even larger when depending on more libraries. Also, adding more
-libraries would require us to generate meta-data more often. Lastly, having more choices does not
-guarantee a better performance for most users as having more choices demands a longer search for
-good models and can lead to more overfitting.
+.. collapse:: <b>Auto-sklearn is extremely memory hungry in a sequential setting</b>
 
-Nevertheless, everyone can still add their favorite model to Auto-sklearn's search space by
-following the `examples on how to extend Auto-sklearn
-<https://automl.github.io/auto-sklearn/master/examples/index.html#extension-examples>`_.
+    Auto-sklearn can appear very memory hungry (i.e. requiring a lot of memory for small datasets) due
+    to the use of ``fork`` for creating new processes when running in sequential manner (if this
+    happens in a parallel setting or if you pass your own dask client this is due to a different
+    issue, see the other issues below).
 
-If there is interest in creating a Auto-sklearn-contrib repository with 3rd-party models please
-open an issue for that.
+    Let's go into some more detail and discuss how to fix it:
+    Auto-sklearn executes each machine learning algorithm in its own process to be able to apply a
+    memory limit and a time limit. To start such a process, Python gives three options: ``fork``,
+    ``forkserver`` and ``spawn``. The default ``fork`` copies the whole process memory into the
+    subprocess. If the main process already uses 1.5GB of main memory and we apply a 3GB memory
+    limit to Auto-sklearn, executing a machine learning pipeline is limited to use at most 1.5GB.
+    We would have loved to use ``forkserver`` or ``spawn`` as the default option instead, which both
+    copy only relevant data into the subprocess and thereby alleaviate the issue of eating up a lot
+    of your main memory
+    (and also do not suffer from potential deadlocks as ``fork`` does, see
+    `here <https://pythonspeed.com/articles/python-multiprocessing/>`_),
+    but they have the downside that code must be guarded by ``if __name__ == "__main__"`` or executed
+    in a notebook, and we decided that we do not want to require this by default.
 
-Can the preprocessing be disabled
----------------------------------
+    There are now two possible solutions:
 
-Feature preprocessing can be disabled as discussed in the example
-:ref:`restricting_the_searchspace`. Other preprocessing steps such as one hot encoding, missing
-feature imputation and normalization cannot yet be disabled, but we're working on that.
+    1. Use Auto-sklearn in parallel: if you use Auto-sklean in parallel, it defaults to ``forkserver``
+       as the parallelization mechanism itself requires Auto-sklearn the code to be guarded. Please
+       find more information on how to do this in the following two examples:
 
-Usage
-=====
+       1. :ref:`sphx_glr_examples_60_search_example_parallel_n_jobs.py`
+       2. :ref:`sphx_glr_examples_60_search_example_parallel_manual_spawning_cli.py`
+
+       .. note::
+
+           This requires all code to be guarded by ``if __name__ == "__main__"``.
+
+    2. Pass a `dask client <https://distributed.dask.org/en/latest/client.html>`_. If the user passes
+       a dask client, Auto-sklearn can no longer assume that it runs in sequential mode and will use
+       a ``forkserver`` to start new processes.
+
+       .. note::
+
+           This requires all code to be guarded by ``if __name__ == "__main__"``.
+
+    We therefore suggest using one of the above settings by default.
+
+.. collapse:: <b>Auto-sklearn is extremely memory hungry in a parallel setting</b>
+
+    When running Auto-sklearn in a parallel setting it starts new processes for evaluating machine
+    learning models using the ``forkserver`` mechanism. Code that is in the main script and that is
+    not guarded by ``if __name__ == "__main__"`` will be executed for each subprocess. If, for example,
+    you are loading your dataset outside of the guarded code, your dataset will be loaded for each
+    evaluation of a machine learning algorithm and thus blocking your RAM.
+
+    We therefore suggest moving all code inside functions or the main block.
+
+.. collapse:: <b>Auto-sklearn crashes with a segmentation fault</b>
+
+    Please make sure that you have read and followed the :ref:`installation` section! In case
+    everything is set up correctly, this is most likely due to the dependency
+    `pyrfr <https://github.com/automl/random_forest_run>`_ not being compiled correctly. If this is the
+    case please execute:
+
+    .. code:: python
+
+        import pyrfr.regression as reg
+        data = reg.default_data_container(64)
+
+    If this fails, the pyrfr dependency is most likely not compiled correctly. We advice you to do the
+    following:
+
+    1. Check if you can use a pre-compiled version of the pyrfr to avoid compiling it yourself. We
+       provide pre-compiled versions of the pyrfr on `pypi <https://pypi.org/project/pyrfr/#files>`_.
+    2. Check if the dependencies specified under :ref:`installation` are correctly installed,
+       especially that you have ``swig`` and a ``C++`` compiler.
+    3. If you are not yet using Conda, consider using it; it simplifies installation of the correct
+       dependencies.
+    4. Install correct build dependencies before installing the pyrfr, you can check the following
+       github issues for suggestions: `1025 <https://github.com/automl/auto-sklearn/issues/1025>`_,
+       `856 <https://github.com/automl/auto-sklearn/issues/856>`_
+
+Results, Log Files and Output
+=============================
+
+.. collapse:: <b>How can I get an overview of the run statistics?</b>
+
+    ``sprint_statistics()`` is a method that prints the name of the  dataset, the metric used, and the best validation score
+    obtained by running *auto-sklearn*. It additionally prints the number of both successful and unsuccessful
+    algorithm runs.
+
+.. collapse:: <b>What was the performance over time?</b>
+
+    ``performance_over_time_``  returns a DataFrame containing the models performance over time data, which can
+    be used for plotting directly (Here is an example: :ref:`sphx_glr_examples_40_advanced_example_pandas_train_test.py`).
+
+    .. code:: python
+
+        automl.performance_over_time_.plot(
+                x='Timestamp',
+                kind='line',
+                legend=True,
+                title='Auto-sklearn accuracy over time',
+                grid=True,
+            )
+            plt.show()
+
+.. collapse:: <b>Which models were evaluated?</b>
+
+    You can see all models evaluated using :meth:`automl.leaderboard(ensemble_only=False) <autosklearn.classification.AutoSklearnClassifier.leaderboard>`.
+
+.. collapse:: <b>Which models are in the final ensemble?</b>
+
+    Use either :meth:`automl.leaderboard(ensemble_only=True) <autosklearn.classification.AutoSklearnClassifier.leaderboard>` or ``automl.show_models()``
+
+.. collapse:: <b>Is there more data I can look at?</b>
+
+    ``cv_results_`` returns a dict with keys as column headers and values as columns, that can be imported into
+    a pandas DataFrame, e.g. ``df = pd.DataFrame(automl.cv_results_)``
+
+.. collapse:: <b>Where does Auto-sklearn output files by default?</b>
+
+    *Auto-sklearn* heavily uses the hard drive to store temporary data, models and log files which can
+    be used to inspect the behavior of Auto-sklearn. Each run of Auto-sklearn requires
+    its own directory. If not provided by the user, *Auto-sklearn* requests a temporary directory from
+    Python, which by default is located under ``/tmp`` and starts with ``autosklearn_tmp_`` followed
+    by a random string. By default, this directory is deleted when the *Auto-sklearn* object is
+    finished fitting. If you want to keep these files you can pass the argument
+    ``delete_tmp_folder_after_terminate=True`` to the *Auto-sklearn* object.
+
+    The :class:`autosklearn.classification.AutoSklearnClassifier` and all other *auto-sklearn*
+    estimators accept the argument ``tmp_folder`` which change where such output is written to.
+
+    There's an additional argument ``output_directory`` which can be passed to *Auto-sklearn* and it
+    controls where test predictions of the ensemble are stored if the test set is passed to ``fit()``.
+
+.. collapse:: <b>Auto-sklearn's logfiles eat up all my disk space. What can I do?</b>
+
+    *Auto-sklearn* heavily uses the hard drive to store temporary data, models and log files which can
+    be used to inspect the behavior of Auto-sklearn. By default, *Auto-sklearn* stores 50
+    models and their predictions on the validation data (which is a subset of the training data in
+    case of holdout and the full training data in case of cross-validation) on the hard drive.
+    Redundant models and their predictions (i.e. when we have more than 50 models) are removed
+    everytime the ensemble builder finishes an iteration, which means that the number of models stored
+    on disk can temporarily be higher if a model is output while the ensemble builder is running.
+
+    One can therefore change the number of models that will be stored on disk by passing an integer
+    for the argument ``max_models_on_disc`` to *Auto-sklearn*, for example reduce the number of models
+    stored on disk if you have space issues.
+
+    As the number of models is only an indicator of the disk space used it is also possible to pass
+    the memory in MB the models are allowed to use as a ``float`` (also via the ``max_models_on_disc``
+    arguments). As above, this is rather a guideline on how much memory is used as redundant models
+    are only removed from disk when the ensemble builder finishes an iteration.
+
+    .. note::
+
+        Especially when running in parallel it can happen that multiple models are constructed during
+        one run of the ensemble builder and thus *Auto-sklearn* can exceed the given limit.
+
+    .. note::
+
+       These limits do only apply to models and their predictions, but not to other files stored in
+       the temporary directory such as the log files.
+
+The Search Space
+================
 
-Only use interpretable models
------------------------------
+.. collapse:: <b>How can I restrict the searchspace?</b>
 
-Auto-sklearn can be restricted to only use interpretable models and preprocessing algorithms.
-Please see the Section :ref:`restricting_the_searchspace` to learn how to restrict the models
-which are searched over or see the Example
-:ref:`sphx_glr_examples_40_advanced_example_interpretable_models.py`.
+ The following shows an example of how to exclude all preprocessing methods and restrict the configuration space to
+ only random forests.
 
-We don't provide a judgement which of the models are interpretable as this is very much up to the
-specific use case, but would like to note that decision trees and linear models usually most
-interpretable.
+    .. code:: python
 
-Limiting the number of model evaluations
-----------------------------------------
+        import autosklearn.classification
+        automl = autosklearn.classification.AutoSklearnClassifier(
+            include = {
+                'classifier': ["random_forest"],
+                'feature_preprocessor': ["no_preprocessing"]
+            },
+            exclude=None
+        )
+        automl.fit(X_train, y_train)
+        predictions = automl.predict(X_test)
 
-In certain cases, for example for debugging, it can be helpful to limit the number of
-model evaluations. We do not provide this as an argument in the API as we believe that it
-should NOT be used in practice, but that the user should rather provide time limits.
-An example on how to add the number of models to try as an additional stopping condition
-can be found `in this github issue <https://github.com/automl/auto-sklearn/issues/451#issuecomment-376445607>`_.
-Please note that Auto-sklearn will stop when either the time limit or the number of
-models termination condition is reached.
+    **Note:** The strings used to identify estimators and preprocessors are the filenames without *.py*.
 
-Ensemble contains only a dummy model
-------------------------------------
+    For a full list please have a look at the source code (in `autosklearn/pipeline/components/`):
 
-This is a symptom of the problem that all runs started by Auto-sklearn failed. Usually, the issue
-is that the runtime or memory limit were too tight. Please check the output of
-``sprint_statistics`` to see the distribution of why runs failed. If there are mostly crashed
-runs, please check the log file for further details. If there are mostly runs that exceed the
-memory or time limit, please increase the respective limit and rerun the optimization.
+      * `Classifiers <https://github.com/automl/auto-sklearn/tree/master/autosklearn/pipeline/components/classification>`_
+      * `Regressors <https://github.com/automl/auto-sklearn/tree/master/autosklearn/pipeline/components/regression>`_
+      * `Preprocessors <https://github.com/automl/auto-sklearn/tree/master/autosklearn/pipeline/components/feature_preprocessing>`_
 
-Parallel processing and oversubscription
-----------------------------------------
+    We do also provide an example on how to restrict the classifiers to search over
+    :ref:`sphx_glr_examples_40_advanced_example_interpretable_models.py`.
 
-Auto-sklearn wraps scikit-learn and therefore inherits its parallelism implementation. In short,
-scikit-learn uses two modes of parallelizing computations:
+.. collapse:: <b>How can I turn off data preprocessing?</b>
 
-1. By using joblib to distribute independent function calls on multiple cores.
-2. By using lower level libraries such as OpenMP and numpy to distribute more fine-grained
-   computation.
+    Data preprocessing includes One-Hot encoding of categorical features, imputation
+    of missing values and the normalization of features or samples. These ensure that
+    the data the gets to the sklearn models is well formed and can be used for
+    training models.
 
-This means that Auto-sklearn can use more resources than expected by the user. For technical
-reasons we can only control the 1st way of parallel execution, but not the 2nd. Thus, the user
-needs to make sure that the lower level parallelization libraries only use as many cores as
-allocated (on a laptop or workstation running a single copy of Auto-sklearn it can be fine to not
-adjust this, but when using a compute cluster it is necessary to align the parallelism setting
-with the number of requested CPUs). This can be done by setting the following environment
-variables: ``MKL_NUM_THREADS``, ``OPENBLAS_NUM_THREADS``, ``BLIS_NUM_THREADS`` and
-``OMP_NUM_THREADS``.
+    While this is necessary in general, if you'd like to disable this step, please
+    refer to this :ref:`example <sphx_glr_examples_80_extending_example_extending_data_preprocessor.py>`.
 
-More details can be found in the `scikit-learn docs <https://scikit-learn.org/stable/computing/parallelism.html?highlight=joblib#parallelism>`_.
+.. collapse:: <b>How can I turn off feature preprocessing?</b>
+
+    Feature preprocessing is a single transformer which implements for example feature
+    selection or transformation of features into a different space (i.e. PCA).
+
+    This can be turned off by setting
+    ``include={'feature_preprocessor'=["no_preprocessing"]}`` as shown in the example above.
+
+.. collapse:: <b>Will non-scikit-learn models be added to Auto-sklearn?</b>
+
+    The short answer: no.
+
+    The long answer answer is a bit more nuanced: maintaining Auto-sklearn requires a lot of time and
+    effort, which would grow even larger when depending on more libraries. Also, adding more
+    libraries would require us to generate meta-data more often. Lastly, having more choices does not
+    guarantee a better performance for most users as having more choices demands a longer search for
+    good models and can lead to more overfitting.
+
+    Nevertheless, everyone can still add their favorite model to Auto-sklearn's search space by
+    following the `examples on how to extend Auto-sklearn
+    <https://automl.github.io/auto-sklearn/master/examples/index.html#extension-examples>`_.
+
+    If there is interest in creating a Auto-sklearn-contrib repository with 3rd-party models please
+    open an issue for that.
+
+.. collapse:: <b>How can I only search for interpretable models</b>
+
+    Auto-sklearn can be restricted to only use interpretable models and preprocessing algorithms.
+    Please see the Section :ref:`space` to learn how to restrict the models
+    which are searched over or see the Example
+    :ref:`sphx_glr_examples_40_advanced_example_interpretable_models.py`.
+
+    We don't provide a judgement which of the models are interpretable as this is very much up to the
+    specific use case, but would like to note that decision trees and linear models usually most
+    interpretable.
+
+Ensembling
+==========
+
+.. collapse:: <b>What can I configure wrt the ensemble building process?</b>
+
+    The following hyperparameters control how the ensemble is constructed:
+
+    * ``ensemble_size`` determines the maximal size of the ensemble. If it is set to zero, no ensemble will be constructed.
+    * ``ensemble_nbest`` allows the user to directly specify the number of models considered for the ensemble.  This hyperparameter can be an integer *n*, such that only the best *n* models are used in the final ensemble. If a float between 0.0 and 1.0 is provided, ``ensemble_nbest`` would be interpreted as a fraction suggesting the percentage of models to use in the ensemble building process (namely, if ensemble_nbest is a float, library pruning is implemented as described in `Caruana et al. (2006) <https://dl.acm.org/doi/10.1109/ICDM.2006.76>`_).
+    * ``max_models_on_disc`` defines the maximum number of models that are kept on the disc, as a mechanism to control the amount of disc space consumed by *auto-sklearn*. Throughout the automl process, different individual models are optimized, and their predictions (and other metadata) is stored on disc. The user can set the upper bound on how many models are acceptable to keep on disc, yet this variable takes priority in the definition of the number of models used by the ensemble builder (that is, the minimum of ``ensemble_size``, ``ensemble_nbest`` and ``max_models_on_disc`` determines the maximal amount of models used in the ensemble). If set to None, this feature is disabled.
+
+.. collapse:: <b>Which models are in the final ensemble?</b>
+
+    The results obtained from the final ensemble can be printed by calling ``show_models()`` or  ``leaderboard()``.
+    The *auto-sklearn* ensemble is composed of scikit-learn models that can be inspected as exemplified
+    in the Example :ref:`sphx_glr_examples_40_advanced_example_get_pipeline_components.py`.
+
+.. collapse:: <b>Can I fit an ensemble also only post-hoc?</b>
+
+    It is possible to build ensembles post-hoc. An example on how to do this (first searching for individual models, and then building an ensemble from them) can be seen in :ref:`sphx_glr_examples_60_search_example_sequential.py`.
+
+Configuring the Search Procedure
+================================
+
+.. collapse:: <b>Can I change the resampling strategy?</b>
+
+    Examples for using holdout and cross-validation can be found in :ref:`example <sphx_glr_examples_40_advanced_example_resampling.py>`
+
+.. collapse:: <b>Can I use a custom metric</b>
+
+    Examples for using a custom metric can be found in :ref:`example <sphx_glr_examples_40_advanced_example_metrics.py>`
 
 Meta-Learning
 =============
 
-Which datasets are used for meta-learning?
-------------------------------------------
+.. collapse:: <b>Which datasets are used for meta-learning?</b>
+
+    We updated the list of datasets used for meta-learning several times and this list now differs
+    significantly from the original 140 datasets we used in 2015 when the paper and the package were
+    released. An up-to-date list of `OpenML task IDs <https://docs.openml.org/#tasks>`_ can be found
+    on `github <https://github.com/automl/auto-sklearn/blob/master/scripts/update_metadata_util.py>`_.
+
+.. collapse:: <b>How can datasets from the meta-data be excluded?</b>
+
+    For *Auto-sklearn 1.0* one can pass the dataset name via the ``fit()`` function. If a dataset
+    with the same name is within the meta-data, that datasets will not be used.
+
+    For *Auto-sklearn 2.0* it is not possible to do so because of the method used to construct the
+    meta-data.
+
+.. collapse:: <b>Which meta-features are used for meta-learning?</b>
 
-We updated the list of datasets used for meta-learning several times and this list now differs
-significantly from the original 140 datasets we used in 2015 when the paper and the package were
-released. An up-to-date list of `OpenML task IDs <https://docs.openml.org/#tasks>`_ can be found
-on `github <https://github.com/automl/auto-sklearn/blob/master/scripts/update_metadata_util.py>`_.
+    We do not have a user guide on meta-features but they are all pretty simple and can be found
+    `in the source code <https://github.com/automl/auto-sklearn/blob/master/autosklearn/metalearning/metafeatures/metafeatures.py>`_.
 
-How can datasets from the meta-data be excluded?
-------------------------------------------------
+.. collapse:: <b>How is the meta-data generated for Auto-sklearn 1.0?</b>
 
-For *Auto-sklearn 1.0* one can pass the dataset name via the ``fit()`` function. If a dataset
-with the same name is within the meta-data, that datasets will not be used.
+    We currently generate meta-data the following way. First, for each of the datasets mentioned
+    above, we run Auto-sklearn without meta-learning for a total of two days on multiple metrics (for
+    classification these are accuracy, balanced accuracy, log loss and the area under the curce).
+    Second, for each run we then have a look at each models that improved the score, i.e. the
+    trajectory of the best known model at a time, and refit it on the whole training data. Third, for
+    each of these models we then compute all scores we're interested in, these also include other
+    ones such F1 and precision. Finally, for each combination of dataset and metric we store the best
+    model we know of.
+
+.. collapse:: <b>How is the meta-data generated for Auto-sklearn 2.0?</b>
+
+    Please check `our paper <https://arxiv.org/abs/2007.04074>`_ for details.
+
+
+Issues and Debugging
+====================
 
-For *Auto-sklearn 2.0* it is not possible to do so because of the method used to construct the
-meta-data.
+.. collapse:: <b>How can I limit the number of model evaluations for debugging?</b>
+
+    In certain cases, for example for debugging, it can be helpful to limit the number of
+    model evaluations. We do not provide this as an argument in the API as we believe that it
+    should NOT be used in practice, but that the user should rather provide time limits.
+    An example on how to add the number of models to try as an additional stopping condition
+    can be found `in this github issue <https://github.com/automl/auto-sklearn/issues/451#issuecomment-376445607>`_.
+    Please note that Auto-sklearn will stop when either the time limit or the number of
+    models termination condition is reached.
+
+.. collapse:: <b>Why does the final ensemble contains only a dummy model?</b>
+
+    This is a symptom of the problem that all runs started by Auto-sklearn failed. Usually, the issue
+    is that the runtime or memory limit were too tight. Please check the output of
+    ``sprint_statistics()`` to see the distribution of why runs failed. If there are mostly crashed
+    runs, please check the log file for further details. If there are mostly runs that exceed the
+    memory or time limit, please increase the respective limit and rerun the optimization.
+
+.. collapse:: <b>Auto-sklearn does not use the specified amount of resources?</b>
+
+    Auto-sklearn wraps scikit-learn and therefore inherits its parallelism implementation. In short,
+    scikit-learn uses two modes of parallelizing computations:
+
+    1. By using joblib to distribute independent function calls on multiple cores.
+    2. By using lower level libraries such as OpenMP and numpy to distribute more fine-grained
+       computation.
+
+    This means that Auto-sklearn can use more resources than expected by the user. For technical
+    reasons we can only control the 1st way of parallel execution, but not the 2nd. Thus, the user
+    needs to make sure that the lower level parallelization libraries only use as many cores as
+    allocated (on a laptop or workstation running a single copy of Auto-sklearn it can be fine to not
+    adjust this, but when using a compute cluster it is necessary to align the parallelism setting
+    with the number of requested CPUs). This can be done by setting the following environment
+    variables: ``MKL_NUM_THREADS``, ``OPENBLAS_NUM_THREADS``, ``BLIS_NUM_THREADS`` and
+    ``OMP_NUM_THREADS``.
+
+    More details can be found in the `scikit-learn docs <https://scikit-learn.org/stable/computing/parallelism.html?highlight=joblib#parallelism>`_.
+
+Other
+=====
 
-Which meta-features are used for meta-learning?
------------------------------------------------
+.. collapse:: <b>Model persistence</b>
 
-We do not have a user guide on meta-features but they are all pretty simple and can be found
-`in the source code <https://github.com/automl/auto-sklearn/blob/master/autosklearn/metalearning/metafeatures/metafeatures.py>`_.
+    *auto-sklearn* is mostly a wrapper around scikit-learn. Therefore, it is
+    possible to follow the
+    `persistence Example <https://scikit-learn.org/stable/modules/model_persistence.html>`_
+    from scikit-learn.
 
-How is the meta-data generated?
--------------------------------
+.. collapse:: <b>Vanilla auto-sklearn</b>
 
-Auto-sklearn 1.0
-~~~~~~~~~~~~~~~~
+    In order to obtain *vanilla auto-sklearn* as used in `Efficient and Robust Automated Machine Learning
+    <https://papers.neurips.cc/paper/5872-efficient-and-robust-automated-machine-learning>`_
+    set ``ensemble_size=1`` and ``initial_configurations_via_metalearning=0``:
 
-We currently generate meta-data the following way. First, for each of the datasets mentioned
-above, we run Auto-sklearn without meta-learning for a total of two days on multiple metrics (for
-classification these are accuracy, balanced accuracy, log loss and the area under the curce).
-Second, for each run we then have a look at each models that improved the score, i.e. the
-trajectory of the best known model at a time, and refit it on the whole training data. Third, for
-each of these models we then compute all scores we're interested in, these also include other
-ones such F1 and precision. Finally, for each combination of dataset and metric we store the best
-model we know of.
+    .. code:: python
 
-Auto-sklearn 2.0
-~~~~~~~~~~~~~~~~
+        import autosklearn.classification
+        automl = autosklearn.classification.AutoSklearnClassifier(
+            ensemble_size=1,
+            initial_configurations_via_metalearning=0
+        )
 
-Please check `our paper <https://arxiv.org/abs/2007.04074>`_ for details.
+    An ensemble of size one will result in always choosing the current best model
+    according to its performance on the validation set. Setting the initial
+    configurations found by meta-learning to zero makes *auto-sklearn* use the
+    regular SMAC algorithm for suggesting new hyperparameter configurations.
diff --git a/master/_sources/index.rst.txt b/master/_sources/index.rst.txt
index c82cdb0eae..e0690ac8e7 100644
--- a/master/_sources/index.rst.txt
+++ b/master/_sources/index.rst.txt
@@ -22,7 +22,7 @@ replacement for a scikit-learn estimator:
 hyperparameter tuning. It leverages recent advantages in *Bayesian
 optimization*, *meta-learning* and *ensemble construction*. Learn more about
 the technology behind *auto-sklearn* by reading our paper published at
-`NIPS 2015 <https://papers.neurips.cc/paper/5872-efficient-and-robust-automated-machine-learning.pdf>`_
+`NeurIPS 2015 <https://papers.neurips.cc/paper/5872-efficient-and-robust-automated-machine-learning.pdf>`_
 .
 
 .. topic:: NEW: Auto-sklearn 2.0
@@ -38,6 +38,11 @@ the technology behind *auto-sklearn* by reading our paper published at
 
 A paper describing our advances is available on `arXiv <https://arxiv.org/abs/2007.04074>`_.
 
+.. topic:: NEW: Material from tutorials and presentations
+
+    We provide slides and notebooks from talks and tutorials here: `auto-sklearn-talks <https://github.com/automl/auto-sklearn-talks>`_
+
+
 Example
 *******
 
diff --git a/master/_sources/installation.rst.txt b/master/_sources/installation.rst.txt
index dff3cecd36..544dd9fff5 100644
--- a/master/_sources/installation.rst.txt
+++ b/master/_sources/installation.rst.txt
@@ -21,8 +21,8 @@ need:
 
 * SWIG (`get SWIG here <http://www.swig.org/survey.html>`_).
 
-For an explanation of missing Microsoft Windows and MAC OSX support please
-check the Section `Windows/OSX compatibility`_.
+For an explanation of missing Microsoft Windows and macOS support please
+check the Section `Windows/macOS compatibility`_.
 
 Installing auto-sklearn
 =======================
@@ -100,9 +100,33 @@ to read in more details check
 for more information about Conda forge check
 `conda-forge documentations <https://conda-forge.org/docs/>`_.
 
+Source Installation
+===================
+
+You can install auto-sklearn directly form source by following the below:
+
+.. code:: bash
 
-Windows/OSX compatibility
-=========================
+    git clone --recurse-submodules git@github.com:automl/auto-sklearn.git
+    cd auto-sklearn
+
+    # Install it in editable mode with all optional dependencies
+    pip install -e ".[test,doc,examples]"
+
+We use submodules so you will have to make sure the submodule is initialized if you
+missed the `--recurse-submodules` option.
+
+.. code:: bash
+
+    git clone git@github.com:automl/auto-sklearn.git
+    cd auto-sklearn
+    git submodule update --init --recursive
+
+    pip install -e ".[test,doc,examples]"
+
+
+Windows/macOS compatibility
+===========================
 
 Windows
 ~~~~~~~
@@ -120,15 +144,15 @@ Possible solutions:
 * docker image
 
 
-Mac OSX
-~~~~~~~
+macOS
+~~~~~
 
-We currently do not know if *auto-sklearn* works on OSX. There are at least two
-issues holding us back from actively supporting OSX:
+We currently do not know if *auto-sklearn* works on macOS. There are at least two
+issues holding us back from actively supporting macOS:
 
 * The ``resource`` module cannot enforce a memory limit on a Python process
   (see `SMAC3/issues/115 <https://github.com/automl/SMAC3/issues/115>`_).
-* Not all dependencies we are using are set up to work on OSX.
+* Not all dependencies we are using are set up to work on macOS.
 
 In case you're having issues installing the `pyrfr package <https://github.com/automl/random_forest_run>`_, check out
 `this installation suggestion on github <https://github.com/automl/auto-sklearn/issues/360#issuecomment-335150470>`_.
diff --git a/master/_sources/manual.rst.txt b/master/_sources/manual.rst.txt
index 252626666d..2a3df6528b 100644
--- a/master/_sources/manual.rst.txt
+++ b/master/_sources/manual.rst.txt
@@ -6,232 +6,299 @@
 Manual
 ======
 
-This manual shows how to use several aspects of auto-sklearn. It either
-references the examples where possible or explains certain configurations.
+This manual gives an overview of different aspects of *auto-sklearn*. For each section, we either references examples or
+give short explanations (click the title to expand text), e.g.
 
-Examples
-========
+.. collapse:: <b>Code examples</b>
 
-We provide examples on using *auto-sklearn* for multiple use cases ranging from
-simple classification to advanced uses such as feature importance, parallel runs
-and customization. They can be found in the :ref:`sphx_glr_examples`.
+    We provide examples on using *auto-sklearn* for multiple use cases ranging from
+    simple classification to advanced uses such as feature importance, parallel runs
+    and customization. They can be found in the :ref:`sphx_glr_examples`.
 
-Time and memory limits
-======================
+.. collapse:: <b>Material from talks and presentations</b>
 
-A crucial feature of *auto-sklearn* is limiting the resources (memory and
-time) which the scikit-learn algorithms are allowed to use. Especially for
-large datasets, on which algorithms can take several hours and make the
-machine swap, it is important to stop the evaluations after some time in order
-to make progress in a reasonable amount of time. Setting the resource limits
-is therefore a tradeoff between optimization time and the number of models
-that can be tested.
+    We provide resources for talks, tutorials and presentations on *auto-sklearn* under `auto-sklearn-talks <https://github.com/automl/auto-sklearn-talks>`_
 
-While *auto-sklearn* alleviates manual hyperparameter tuning, the user still
-has to set memory and time limits. For most datasets a memory limit of 3GB or
-6GB as found on most modern computers is sufficient. For the time limits it
-is harder to give clear guidelines. If possible, a good default is a total
-time limit of one day, and a time limit of 30 minutes for a single run.
+.. _limits:
 
-Further guidelines can be found in
-`auto-sklearn/issues/142 <https://github.com/automl/auto-sklearn/issues/142>`_.
+Resource limits
+===============
 
-.. _restricting_the_searchspace:
+A crucial feature of *auto-sklearn* is limiting the resources (memory and time) which the scikit-learn algorithms are
+allowed to use. Especially for large datasets, on which algorithms can take several hours and make the machine swap,
+it is important to stop the evaluations after some time in order to make progress in a reasonable amount of time.
+Setting the resource limits is therefore a tradeoff between optimization time and the number of models that can be
+tested.
 
-Restricting the searchspace
-===========================
+.. collapse:: <b>Time and memory limits</b>
 
-Instead of using all available estimators, it is possible to restrict
-*auto-sklearn*'s searchspace. The following shows an example of how to exclude
-all preprocessing methods and restrict the configuration space to only
-random forests.
+    While *auto-sklearn* alleviates manual hyperparameter tuning, the user still
+    has to set memory and time limits. For most datasets a memory limit of 3GB or
+    6GB as found on most modern computers is sufficient. For the time limits it
+    is harder to give clear guidelines. If possible, a good default is a total
+    time limit of one day, and a time limit of 30 minutes for a single run.
 
-.. code:: python
+    Further guidelines can be found in
+    `auto-sklearn/issues/142 <https://github.com/automl/auto-sklearn/issues/142>`_.
+
+.. collapse:: <b>CPU cores</b>
 
-    import autosklearn.classification
-    automl = autosklearn.classification.AutoSklearnClassifier(
-        include = {
-            'classifier': ["random_forest"],
-            'feature_preprocessor': ["no_preprocessing"]
-        },
-        exclude=None
-    )
-    automl.fit(X_train, y_train)
-    predictions = automl.predict(X_test)
+    By default, *auto-sklearn* uses **one core**. See also :ref:`parallel` on how to configure this.
 
-**Note:** The strings used to identify estimators and preprocessors are the filenames without *.py*.
+.. _space:
 
-For a full list please have a look at the source code (in `autosklearn/pipeline/components/`):
+The search space
+================
 
-  * `Classifiers <https://github.com/automl/auto-sklearn/tree/master/autosklearn/pipeline/components/classification>`_
-  * `Regressors <https://github.com/automl/auto-sklearn/tree/master/autosklearn/pipeline/components/regression>`_
-  * `Preprocessors <https://github.com/automl/auto-sklearn/tree/master/autosklearn/pipeline/components/feature_preprocessing>`_
+*Auto-sklearn* by default searches a large space to find a well performing configuration. However, it is also possible
+to restrict the searchspace:
 
-We do also provide an example on how to restrict the classifiers to search over
-:ref:`sphx_glr_examples_40_advanced_example_interpretable_models.py`.
+.. collapse:: <b>Restricting the searchspace</b>
 
-Data preprocessing
-~~~~~~~~~~~~~~~~~~
-Data preprocessing includes One-Hot encoding of categorical features, imputation
-of missing values and the normalization of features or samples. These ensure that
-the data the gets to the sklearn models is well formed and can be used for 
-training models.
+ The following shows an example of how to exclude all preprocessing methods and restrict the configuration space to
+ only random forests.
 
-While this is necessary in general, if you'd like to disable this step, please
-refer to this :ref:`example <sphx_glr_examples_80_extending_example_extending_data_preprocessor.py>`.
+    .. code:: python
 
-Feature preprocessing
-~~~~~~~~~~~~~~~~~~~~~
-Feature preprocessing is a single transformer which implements for example feature
-selection or transformation of features into a different space (i.e. PCA).
+        import autosklearn.classification
+        automl = autosklearn.classification.AutoSklearnClassifier(
+            include = {
+                'classifier': ["random_forest"],
+                'feature_preprocessor': ["no_preprocessing"]
+            },
+            exclude=None
+        )
+        automl.fit(X_train, y_train)
+        predictions = automl.predict(X_test)
 
-This can be turned off by setting
-``include={'feature_preprocessor'=["no_preprocessing"]}`` as shown in the example above.
+    **Note:** The strings used to identify estimators and preprocessors are the filenames without *.py*.
 
-Resampling strategies
-=====================
+    For a full list please have a look at the source code (in `autosklearn/pipeline/components/`):
 
-Examples for using holdout and cross-validation can be found in :ref:`auto-sklearn/examples/ <examples>`.
+      * `Classifiers <https://github.com/automl/auto-sklearn/tree/master/autosklearn/pipeline/components/classification>`_
+      * `Regressors <https://github.com/automl/auto-sklearn/tree/master/autosklearn/pipeline/components/regression>`_
+      * `Preprocessors <https://github.com/automl/auto-sklearn/tree/master/autosklearn/pipeline/components/feature_preprocessing>`_
 
-Supported Inputs
-================
-*auto-sklearn* can accept targets for the following tasks (more details on `Sklearn algorithms <https://scikit-learn.org/stable/modules/multiclass.html>`_):
+    We do also provide an example on how to restrict the classifiers to search over
+    :ref:`sphx_glr_examples_40_advanced_example_interpretable_models.py`.
+
+.. collapse:: <b>Turn off data preprocessing</b>
+
+    Data preprocessing includes One-Hot encoding of categorical features, imputation
+    of missing values and the normalization of features or samples. These ensure that
+    the data the gets to the sklearn models is well formed and can be used for
+    training models.
+
+    While this is necessary in general, if you'd like to disable this step, please
+    refer to this :ref:`example <sphx_glr_examples_80_extending_example_extending_data_preprocessor.py>`.
+
+.. collapse:: <b>Turn off feature preprocessing</b>
+
+    Feature preprocessing is a single transformer which implements for example feature
+    selection or transformation of features into a different space (i.e. PCA).
+
+    This can be turned off by setting
+    ``include={'feature_preprocessor'=["no_preprocessing"]}`` as shown in the example above.
+
+.. _bestmodel:
 
-* Binary Classification
-* Multiclass Classification
-* Multilabel Classification
-* Regression
-* Multioutput Regression
+Model selection
+===============
 
-You can provide feature and target training pairs (X_train/y_train) to *auto-sklearn* to fit an
-ensemble of pipelines as described in the next section. This X_train/y_train dataset must belong
-to one of the supported formats: np.ndarray, pd.DataFrame, scipy.sparse.csr_matrix and python lists.
-Optionally, you can measure the ability of this fitted model to generalize to unseen data by
-providing an optional testing pair (X_test/Y_test). For further details, please refer to the
-Example :ref:`sphx_glr_examples_40_advanced_example_pandas_train_test.py`.
-Supported formats for these training and testing pairs are: np.ndarray,
-pd.DataFrame, scipy.sparse.csr_matrix and python lists.
+*Auto-sklearn* implements different strategies to identify the best performing model. For some use cases it might be
+necessary to adapt the resampling strategy or define a custom metric:
 
-If your data contains categorical values (in the features or targets), autosklearn will automatically encode your data using a `sklearn.preprocessing.LabelEncoder <https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.LabelEncoder.html>`_ for unidimensional data and a `sklearn.preprocessing.OrdinalEncoder <https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.OrdinalEncoder.html>`_ for multidimensional data.
+.. collapse:: <b>Use different resampling strategies</b>
 
-Regarding the features, there are two methods to guide *auto-sklearn* to properly encode categorical columns:
+    Examples for using holdout and cross-validation can be found in :ref:`example <sphx_glr_examples_40_advanced_example_resampling.py>`
 
-* Providing a X_train/X_test numpy array with the optional flag feat_type. For further details, you
-  can check the Example :ref:`sphx_glr_examples_40_advanced_example_feature_types.py`.
-* You can provide a pandas DataFrame, with properly formatted columns. If a column has numerical
-  dtype, *auto-sklearn* will not encode it and it will be passed directly to scikit-learn. If the
-  column has a categorical/boolean class, it will be encoded. If the column is of any other type
-  (Object or Timeseries), an error will be raised. For further details on how to properly encode
-  your data, you can check the Pandas Example
-  `Working with categorical data <https://pandas.pydata.org/pandas-docs/stable/user_guide/categorical.html>`_).
-  If you are working with time series, it is recommended that you follow this approach
-  `Working with time data <https://stats.stackexchange.com/questions/311494/>`_.
+.. collapse:: <b>Use a custom metric</b>
 
-Regarding the targets (y_train/y_test), if the task involves a classification problem, such features will be automatically encoded. It is recommended to provide both y_train and y_test during fit, so that a common encoding is created between these splits (if only y_train is provided during fit, the categorical encoder will not be able to handle new classes that are exclusive to y_test). If the task is regression, no encoding happens on the targets.
+    Examples for using a custom metric can be found in :ref:`example <sphx_glr_examples_40_advanced_example_metrics.py>`
 
-Ensemble Building Process
-=========================
+.. _ensembles:
 
-*auto-sklearn* uses ensemble selection by `Caruana et al. (2004) <https://dl.acm.org/doi/pdf/10.1145/1015330.1015432>`_
-to build an ensemble based on the models’ prediction for the validation set. The following hyperparameters control how the ensemble is constructed:
+Ensembling
+==========
 
-* ``ensemble_size`` determines the maximal size of the ensemble. If it is set to zero, no ensemble will be constructed.
-* ``ensemble_nbest`` allows the user to directly specify the number of models considered for the ensemble.  This hyperparameter can be an integer *n*, such that only the best *n* models are used in the final ensemble. If a float between 0.0 and 1.0 is provided, ``ensemble_nbest`` would be interpreted as a fraction suggesting the percentage of models to use in the ensemble building process (namely, if ensemble_nbest is a float, library pruning is implemented as described in `Caruana et al. (2006) <https://dl.acm.org/doi/10.1109/ICDM.2006.76>`_).
-* ``max_models_on_disc`` defines the maximum number of models that are kept on the disc, as a mechanism to control the amount of disc space consumed by *auto-sklearn*. Throughout the automl process, different individual models are optimized, and their predictions (and other metadata) is stored on disc. The user can set the upper bound on how many models are acceptable to keep on disc, yet this variable takes priority in the definition of the number of models used by the ensemble builder (that is, the minimum of ``ensemble_size``, ``ensemble_nbest`` and ``max_models_on_disc`` determines the maximal amount of models used in the ensemble). If set to None, this feature is disabled.
+To get the best performance out of the evaluated models, *auto-sklearn* uses ensemble selection by `Caruana et al. (2004) <https://dl.acm.org/doi/pdf/10.1145/1015330.1015432>`_
+to build an ensemble based on the models’ prediction for the validation set.
 
-.. _inspecting_the_results:
+.. collapse:: <b>Configure the ensemble building process</b>
+
+    The following hyperparameters control how the ensemble is constructed:
+
+    * ``ensemble_size`` determines the maximal size of the ensemble. If it is set to zero, no ensemble will be constructed.
+    * ``ensemble_nbest`` allows the user to directly specify the number of models considered for the ensemble.  This hyperparameter can be an integer *n*, such that only the best *n* models are used in the final ensemble. If a float between 0.0 and 1.0 is provided, ``ensemble_nbest`` would be interpreted as a fraction suggesting the percentage of models to use in the ensemble building process (namely, if ensemble_nbest is a float, library pruning is implemented as described in `Caruana et al. (2006) <https://dl.acm.org/doi/10.1109/ICDM.2006.76>`_).
+    * ``max_models_on_disc`` defines the maximum number of models that are kept on the disc, as a mechanism to control the amount of disc space consumed by *auto-sklearn*. Throughout the automl process, different individual models are optimized, and their predictions (and other metadata) is stored on disc. The user can set the upper bound on how many models are acceptable to keep on disc, yet this variable takes priority in the definition of the number of models used by the ensemble builder (that is, the minimum of ``ensemble_size``, ``ensemble_nbest`` and ``max_models_on_disc`` determines the maximal amount of models used in the ensemble). If set to None, this feature is disabled.
+
+.. collapse:: <b>Inspect the final ensemble</b>
+
+    The results obtained from the final ensemble can be printed by calling ``show_models()``.
+    The *auto-sklearn* ensemble is composed of scikit-learn models that can be inspected as exemplified
+    in the Example :ref:`sphx_glr_examples_40_advanced_example_get_pipeline_components.py`.
+
+.. collapse:: <b>Fit ensemble post-hoc</b>
+
+    To use a single core only, it is possible to build ensembles post-hoc. An example on how to do this (first searching
+    for individual models, and then building an ensemble from them) can be seen in
+    :ref:`sphx_glr_examples_60_search_example_sequential.py`.
+
+
+.. _inspect:
 
 Inspecting the results
 ======================
 
-*auto-sklearn* allows users to inspect the training results and statistics. The following example shows how different
-statistics can be printed for the inspection.
+*auto-sklearn* allows users to inspect the training results and statistics. Assume we have a fitted estimator:
 
 .. code:: python
 
-    import autosklearn.classification
-    automl = autosklearn.classification.AutoSklearnClassifier()
-    automl.fit(X_train, y_train)
-    automl.cv_results_
-    automl.performance_over_time_.plot(
-        x='Timestamp',
-        kind='line',
-        legend=True,
-        title='Auto-sklearn accuracy over time',
-        grid=True,
-    )
-    plt.show()
-
-    automl.sprint_statistics()
-    automl.show_models()
-
-``cv_results_`` returns a dict with keys as column headers and values as columns, that can be imported into a pandas DataFrame.
-``performance_over_time_``  returns a DataFrame containing the models performance over time data, which can be used for plotting directly (Here is an example: :ref:`sphx_glr_examples_40_advanced_example_pandas_train_test.py`).
-``sprint_statistics()`` is a method that prints the name of the  dataset, the metric used, and the best validation score
-obtained by running *auto-sklearn*. It additionally prints the number of both successful and unsuccessful
-algorithm runs.
-
-The results obtained from the final ensemble can be printed by calling ``show_models()``.
-*auto-sklearn* ensemble is composed of scikit-learn models that can be inspected as exemplified
-in the Example :ref:`sphx_glr_examples_40_advanced_example_get_pipeline_components.py`.
+        import autosklearn.classification
+        automl = autosklearn.classification.AutoSklearnClassifier()
+        automl.fit(X_train, y_train)
+
+*auto-sklearn* offers the following ways to inspect the results
+
+.. collapse:: <b>Basic statistics</b>
+
+    ``sprint_statistics()`` is a method that prints the name of the  dataset, the metric used, and the best validation score
+    obtained by running *auto-sklearn*. It additionally prints the number of both successful and unsuccessful
+    algorithm runs.
+
+.. collapse:: <b>Performance over Time</b>
+
+    ``performance_over_time_``  returns a DataFrame containing the models performance over time data, which can
+    be used for plotting directly (Here is an example: :ref:`sphx_glr_examples_40_advanced_example_pandas_train_test.py`).
+
+    .. code:: python
+
+        automl.performance_over_time_.plot(
+                x='Timestamp',
+                kind='line',
+                legend=True,
+                title='Auto-sklearn accuracy over time',
+                grid=True,
+            )
+            plt.show()
+
+.. collapse:: <b>Evaluated models</b>
+
+    The results obtained from the final ensemble can be printed by calling ``show_models()``.
+
+.. collapse:: <b>Leaderboard</b>
+
+    ``automl.leaderboard()`` shows the ensemble members, check the :meth:`docs <autosklearn.classification.AutoSklearnClassifier.leaderboard>` for using leaderboard for getting information on *all* runs.
+
+.. collapse:: <b>Other</b>
+
+    ``cv_results_`` returns a dict with keys as column headers and values as columns, that can be imported into a pandas DataFrame.
+
+.. _parallel:
 
 Parallel computation
 ====================
 
-In it's default mode, *auto-sklearn* already uses two cores. The first one is
-used for model building, the second for building an ensemble every time a new
-machine learning model has finished training. An example on how to do this sequentially (first searching for individual models, and then building an ensemble from them) can be seen in
-:ref:`sphx_glr_examples_60_search_example_sequential.py`.
+In it's default mode, *auto-sklearn* uses **one core** and interleaves ensemble building with evaluating new
+configurations.
 
-Nevertheless, *auto-sklearn* also supports parallel Bayesian optimization via the use of
-`Dask.distributed  <https://distributed.dask.org/>`_. By providing the arguments ``n_jobs``
-to the estimator construction, one can control the number of cores available to *auto-sklearn*
-(As shown in the Example :ref:`sphx_glr_examples_60_search_example_parallel_n_jobs.py`).
-Distributed processes are also supported by providing a custom client object to *auto-sklearn* like
-in the Example: :ref:`sphx_glr_examples_60_search_example_parallel_manual_spawning_cli.py`. When
-multiple cores are
-available, *auto-sklearn* will create a worker per core, and use the available workers to both search
-for better machine learning models as well as building an ensemble with them until the time resource
-is exhausted.
+.. collapse:: <b>Parallelization with Dask</b>
 
-**Note:** *auto-sklearn* requires all workers to have access to a shared file system for storing training data and models.
+    Nevertheless, *auto-sklearn* also supports parallel Bayesian optimization via the use of
+    `Dask.distributed  <https://distributed.dask.org/>`_. By providing the arguments ``n_jobs``
+    to the estimator construction, one can control the number of cores available to *auto-sklearn*
+    (As shown in the Example :ref:`sphx_glr_examples_60_search_example_parallel_n_jobs.py`).
+    Distributed processes are also supported by providing a custom client object to *auto-sklearn* like
+    in the Example: :ref:`sphx_glr_examples_60_search_example_parallel_manual_spawning_cli.py`. When
+    multiple cores are
+    available, *auto-sklearn* will create a worker per core, and use the available workers to both search
+    for better machine learning models as well as building an ensemble with them until the time resource
+    is exhausted.
 
-*auto-sklearn* employs `threadpoolctl <https://github.com/joblib/threadpoolctl/>`_ to control the number of threads employed by scientific libraries like numpy or scikit-learn. This is done exclusively during the building procedure of models, not during inference. In particular, *auto-sklearn* allows each pipeline to use at most 1 thread during training. At predicting and scoring time this limitation is not enforced by *auto-sklearn*. You can control the number of resources
-employed by the pipelines by setting the following variables in your environment, prior to running *auto-sklearn*:
+    **Note:** *auto-sklearn* requires all workers to have access to a shared file system for storing training data and models.
 
-.. code-block:: shell-session
+    *auto-sklearn* employs `threadpoolctl <https://github.com/joblib/threadpoolctl/>`_ to control the number of threads employed by scientific libraries like numpy or scikit-learn. This is done exclusively during the building procedure of models, not during inference. In particular, *auto-sklearn* allows each pipeline to use at most 1 thread during training. At predicting and scoring time this limitation is not enforced by *auto-sklearn*. You can control the number of resources
+    employed by the pipelines by setting the following variables in your environment, prior to running *auto-sklearn*:
 
-    $ export OPENBLAS_NUM_THREADS=1
-    $ export MKL_NUM_THREADS=1
-    $ export OMP_NUM_THREADS=1
+    .. code-block:: shell-session
 
+        $ export OPENBLAS_NUM_THREADS=1
+        $ export MKL_NUM_THREADS=1
+        $ export OMP_NUM_THREADS=1
 
-For further information about how scikit-learn handles multiprocessing, please check the `Parallelism, resource management, and configuration <https://scikit-learn.org/stable/computing/parallelism.html>`_ documentation from the library.
 
-Model persistence
-=================
+    For further information about how scikit-learn handles multiprocessing, please check the `Parallelism, resource management, and configuration <https://scikit-learn.org/stable/computing/parallelism.html>`_ documentation from the library.
 
-*auto-sklearn* is mostly a wrapper around scikit-learn. Therefore, it is
-possible to follow the
-`persistence Example <https://scikit-learn.org/stable/modules/model_persistence.html>`_
-from scikit-learn.
+.. _othermanual:
 
-Vanilla auto-sklearn
-====================
+Other
+=====
 
-In order to obtain *vanilla auto-sklearn* as used in `Efficient and Robust Automated Machine Learning
-<https://papers.nips.cc/paper/5872-efficient-and-robust-automated-machine -learning>`_
-set ``ensemble_size=1`` and ``initial_configurations_via_metalearning=0``:
+.. collapse:: <b>Supported input types</b>
 
-.. code:: python
+    *auto-sklearn* can accept targets for the following tasks (more details on `Sklearn algorithms <https://scikit-learn.org/stable/modules/multiclass.html>`_):
+
+    * Binary Classification
+    * Multiclass Classification
+    * Multilabel Classification
+    * Regression
+    * Multioutput Regression
+
+    You can provide feature and target training pairs (X_train/y_train) to *auto-sklearn* to fit an
+    ensemble of pipelines as described in the next section. This X_train/y_train dataset must belong
+    to one of the supported formats: np.ndarray, pd.DataFrame, scipy.sparse.csr_matrix and python lists.
+    Optionally, you can measure the ability of this fitted model to generalize to unseen data by
+    providing an optional testing pair (X_test/Y_test). For further details, please refer to the
+    Example :ref:`sphx_glr_examples_40_advanced_example_pandas_train_test.py`.
+    Supported formats for these training and testing pairs are: np.ndarray,
+    pd.DataFrame, scipy.sparse.csr_matrix and python lists.
+
+    If your data contains categorical values (in the features or targets), autosklearn will automatically encode your
+    data using a `sklearn.preprocessing.LabelEncoder <https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.LabelEncoder.html>`_
+    for unidimensional data and a `sklearn.preprocessing.OrdinalEncoder <https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.OrdinalEncoder.html>`_
+    for multidimensional data.
+
+    Regarding the features, there are two methods to guide *auto-sklearn* to properly encode categorical columns:
+
+    * Providing a X_train/X_test numpy array with the optional flag feat_type. For further details, you
+      can check the Example :ref:`sphx_glr_examples_40_advanced_example_feature_types.py`.
+    * You can provide a pandas DataFrame, with properly formatted columns. If a column has numerical
+      dtype, *auto-sklearn* will not encode it and it will be passed directly to scikit-learn. If the
+      column has a categorical/boolean class, it will be encoded. If the column is of any other type
+      (Object or Timeseries), an error will be raised. For further details on how to properly encode
+      your data, you can check the Pandas Example
+      `Working with categorical data <https://pandas.pydata.org/pandas-docs/stable/user_guide/categorical.html>`_).
+      If you are working with time series, it is recommended that you follow this approach
+      `Working with time data <https://stats.stackexchange.com/questions/311494/>`_.
+
+    Regarding the targets (y_train/y_test), if the task involves a classification problem, such features will be
+    automatically encoded. It is recommended to provide both y_train and y_test during fit, so that a common encoding
+    is created between these splits (if only y_train is provided during fit, the categorical encoder will not be able
+    to handle new classes that are exclusive to y_test). If the task is regression, no encoding happens on the
+    targets.
+
+.. collapse:: <b>Model persistence</b>
+
+    *auto-sklearn* is mostly a wrapper around scikit-learn. Therefore, it is
+    possible to follow the
+    `persistence Example <https://scikit-learn.org/stable/modules/model_persistence.html>`_
+    from scikit-learn.
+
+.. collapse:: <b>Vanilla auto-sklearn</b>
+
+    In order to obtain *vanilla auto-sklearn* as used in `Efficient and Robust Automated Machine Learning
+    <https://papers.nips.cc/paper/5872-efficient-and-robust-automated-machine -learning>`_
+    set ``ensemble_size=1`` and ``initial_configurations_via_metalearning=0``:
+
+    .. code:: python
 
-    import autosklearn.classification
-    automl = autosklearn.classification.AutoSklearnClassifier(
-        ensemble_size=1,
-        initial_configurations_via_metalearning=0
-    )
+        import autosklearn.classification
+        automl = autosklearn.classification.AutoSklearnClassifier(
+            ensemble_size=1,
+            initial_configurations_via_metalearning=0
+        )
 
-An ensemble of size one will result in always choosing the current best model
-according to its performance on the validation set. Setting the initial
-configurations found by meta-learning to zero makes *auto-sklearn* use the
-regular SMAC algorithm for suggesting new hyperparameter configurations.
+    An ensemble of size one will result in always choosing the current best model
+    according to its performance on the validation set. Setting the initial
+    configurations found by meta-learning to zero makes *auto-sklearn* use the
+    regular SMAC algorithm for suggesting new hyperparameter configurations.
diff --git a/master/_sources/releases.rst.txt b/master/_sources/releases.rst.txt
index c89418d851..bc7c33a4a1 100644
--- a/master/_sources/releases.rst.txt
+++ b/master/_sources/releases.rst.txt
@@ -9,6 +9,50 @@
 Releases
 ========
 
+Version 0.14.4
+==============
+
+* Fix #1356: SVR degree hyperparameter now only active with "poly" kernel.
+* Add #1311: Black format checking (non-strict).
+* Maint #1306: Run history is now saved every iteration
+* Doc #1309: Updated the doc faqs to include many use cases and the manual for early introductions
+* Doc #1322: Fix typo in contribution guide
+* Maint #1326: Add isort checker (non-strict)
+* Maint #1238, #1346, #1368, #1370: Update warnings in tests
+* Maint #1325: Test workflow can now be manually triggered
+* Maint #1332: Update docstring and typing of ``include`` and ``exclude`` params
+* Add #1260: Support for Python 3.10
+* Add #1318: First update to use the shared backend in a new submodule `automl_common <https://github.com/automl/automl_common>`_
+* Fix #1339: Resolve dependancy issues with ``sphinx_toolbox``
+* Fix #1335: Fix issue where some regression algorithm gave incorrect output dimensions as raised in #1297
+* Doc #1340: Update example for predefined splits
+* Fix #1329: Fix random state not being passed to the ConfigurationSpace
+* Maint #1348: Stop double triggering of github workflows
+* Doc #1349: Rename OSX to macOS in docs
+* Add #1321: Change ``show_models()`` to produce actual pipeline objects and not a ``str``
+* Maint #1361: Remove ``flaky`` dependency
+* Maint #1366: Make ``SimpleClassificationPipeline`` tests more deterministic
+* Maint #1367: Update test values for ``MLPRegressor`` with newer numpy
+
+Contributors v0.14.4
+********************
+
+* Eddie Bergman
+* Matthias Feurer
+* Katharina Eggensperger
+* UserFindingSelf
+* partev
+
+Version 0.14.3
+==============
+
+* HOTFIX #1356: Updates dask to ``dask.distributed >=2012.12``.
+
+Contributors v0.14.3
+********************
+
+* Eddie Bergman
+
 Version 0.14.2
 ==============
 
@@ -35,7 +79,7 @@ Version 0.14.1
 Contributors v0.14.1
 ********************
 
-* Edward Bergman
+* Eddie Bergman
 * Michael Becker
 * Katharina Eggensperger
 
@@ -621,7 +665,7 @@ Version 0.4.0
   minimization problem.
 * Implements `#271 <https://github.com/automl/auto-sklearn/issues/271>`_:
   XGBoost is available again, even configuring the new dropout functionality.
-* New documentation section :ref:`inspecting_the_results`.
+* New documentation section :ref:`inspect`.
 * Fixes `#444 <https://github.com/automl/auto-sklearn/issues/444>`_:
   Auto-sklearn now only loads models for refit which are actually relevant
   for the ensemble.
diff --git a/master/_static/documentation_options.js b/master/_static/documentation_options.js
index 64463cd050..1664578c28 100644
--- a/master/_static/documentation_options.js
+++ b/master/_static/documentation_options.js
@@ -1,6 +1,6 @@
 var DOCUMENTATION_OPTIONS = {
     URL_ROOT: document.getElementById("documentation_options").getAttribute('data-url_root'),
-    VERSION: '0.14.2',
+    VERSION: '0.14.4',
     LANGUAGE: 'None',
     COLLAPSE_INDEX: false,
     BUILDER: 'html',
diff --git a/master/api.html b/master/api.html
index 254500d493..4276b24f42 100644
--- a/master/api.html
+++ b/master/api.html
@@ -3,9 +3,8 @@
 <html>
   <head>
     <meta charset="utf-8" />
-    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="generator" content="Docutils 0.17.1: http://docutils.sourceforge.net/" />
-
-    <title>APIs &#8212; AutoSklearn 0.14.2 documentation</title>
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>APIs &#8212; AutoSklearn 0.14.4 documentation</title>
     <link rel="stylesheet" type="text/css" href="_static/pygments.css" />
     <link rel="stylesheet" type="text/css" href="_static/bootstrap-sphinx.css" />
     <link rel="stylesheet" type="text/css" href="_static/sg_gallery.css" />
@@ -56,7 +55,7 @@
         </button>
         <a class="navbar-brand" href="index.html">
           auto-sklearn</a>
-        <span class="navbar-text navbar-version pull-left"><b>0.14.2</b></span>
+        <span class="navbar-text navbar-version pull-left"><b>0.14.4</b></span>
       </div>
 
         <div class="collapse navbar-collapse nav-collapse">
@@ -131,15 +130,15 @@
       </div>
     <div class="body col-md-9 content" role="main">
       
-  <section id="apis">
+  <div class="section" id="apis">
 <span id="api"></span><h1>APIs<a class="headerlink" href="#apis" title="Permalink to this headline">¶</a></h1>
-<section id="main-modules">
+<div class="section" id="main-modules">
 <h2>Main modules<a class="headerlink" href="#main-modules" title="Permalink to this headline">¶</a></h2>
-<section id="classification">
+<div class="section" id="classification">
 <h3>Classification<a class="headerlink" href="#classification" title="Permalink to this headline">¶</a></h3>
 <dl class="py class">
 <dt class="sig sig-object py" id="autosklearn.classification.AutoSklearnClassifier">
-<em class="property"><span class="pre">class</span> </em><span class="sig-prename descclassname"><span class="pre">autosklearn.classification.</span></span><span class="sig-name descname"><span class="pre">AutoSklearnClassifier</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">time_left_for_this_task</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">3600</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">per_run_time_limit</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">initial_configurations_via_metalearning</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">25</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">ensemble_size</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">int</span></span> <span class="o"><span class="pre">=</span></span> <span class="default_value"><span class="pre">50</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">ensemble_nbest</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">50</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">max_models_on_disc</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">50</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">seed</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">1</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">memory_limit</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">3072</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">include</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">exclude</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">resampling_strategy</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">'holdout'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">resampling_strategy_arguments</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tmp_folder</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">delete_tmp_folder_after_terminate</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">n_jobs</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">Optional</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span></span> <span class="o"><span class="pre">=</span></span> <span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">dask_client</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">Optional</span><span class="p"><span class="pre">[</span></span><span class="pre">distributed.client.Client</span><span class="p"><span class="pre">]</span></span></span> <span class="o"><span class="pre">=</span></span> <span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">disable_evaluator_output</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">get_smac_object_callback</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">smac_scenario_args</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">logging_config</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">metadata_directory</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">metric</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">scoring_functions</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">Optional</span><span class="p"><span class="pre">[</span></span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">autosklearn.metrics.Scorer</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span> <span class="o"><span class="pre">=</span></span> <span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">load_models</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">bool</span></span> <span class="o"><span class="pre">=</span></span> <span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">get_trials_callback</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/autosklearn/estimators.html#AutoSklearnClassifier"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#autosklearn.classification.AutoSklearnClassifier" title="Permalink to this definition">¶</a></dt>
+<em class="property"><span class="pre">class</span> </em><span class="sig-prename descclassname"><span class="pre">autosklearn.classification.</span></span><span class="sig-name descname"><span class="pre">AutoSklearnClassifier</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">time_left_for_this_task</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">3600</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">per_run_time_limit</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">initial_configurations_via_metalearning</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">25</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">ensemble_size</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">int</span></span> <span class="o"><span class="pre">=</span></span> <span class="default_value"><span class="pre">50</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">ensemble_nbest</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">50</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">max_models_on_disc</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">50</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">seed</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">1</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">memory_limit</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">3072</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">include</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">Optional</span><span class="p"><span class="pre">[</span></span><span class="pre">Dict</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span> <span class="o"><span class="pre">=</span></span> <span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">exclude</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">Optional</span><span class="p"><span class="pre">[</span></span><span class="pre">Dict</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span> <span class="o"><span class="pre">=</span></span> <span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">resampling_strategy</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">'holdout'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">resampling_strategy_arguments</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tmp_folder</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">delete_tmp_folder_after_terminate</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">n_jobs</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">Optional</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span></span> <span class="o"><span class="pre">=</span></span> <span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">dask_client</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">Optional</span><span class="p"><span class="pre">[</span></span><span class="pre">distributed.client.Client</span><span class="p"><span class="pre">]</span></span></span> <span class="o"><span class="pre">=</span></span> <span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">disable_evaluator_output</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">get_smac_object_callback</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">smac_scenario_args</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">logging_config</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">metadata_directory</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">metric</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">scoring_functions</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">Optional</span><span class="p"><span class="pre">[</span></span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">autosklearn.metrics.Scorer</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span> <span class="o"><span class="pre">=</span></span> <span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">load_models</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">bool</span></span> <span class="o"><span class="pre">=</span></span> <span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">get_trials_callback</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/autosklearn/estimators.html#AutoSklearnClassifier"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#autosklearn.classification.AutoSklearnClassifier" title="Permalink to this definition">¶</a></dt>
 <dd><p>This class implements the classification task.</p>
 <dl class="field-list">
 <dt class="field-odd">Parameters</dt>
@@ -161,7 +160,7 @@ <h3>Classification<a class="headerlink" href="#classification" title="Permalink
 </dd>
 <dt><strong>ensemble_size</strong><span class="classifier">int, optional (default=50)</span></dt><dd><p>Number of models added to the ensemble built by <em>Ensemble
 selection from libraries of models</em>. Models are drawn with
-replacement.</p>
+replacement. If set to <code class="docutils literal notranslate"><span class="pre">0</span></code> no ensemble is fit.</p>
 </dd>
 <dt><strong>ensemble_nbest</strong><span class="classifier">int, optional (default=50)</span></dt><dd><p>Only consider the <code class="docutils literal notranslate"><span class="pre">ensemble_nbest</span></code> models when building an
 ensemble.</p>
@@ -177,19 +176,54 @@ <h3>Classification<a class="headerlink" href="#classification" title="Permalink
 </dd>
 <dt><strong>memory_limit</strong><span class="classifier">int, optional (3072)</span></dt><dd><p>Memory limit in MB for the machine learning algorithm.
 <cite>auto-sklearn</cite> will stop fitting the machine learning algorithm if
-it tries to allocate more than <cite>memory_limit</cite> MB.
-If None is provided, no memory limit is set.
-In case of multi-processing, <cite>memory_limit</cite> will be per job.
-This memory limit also applies to the ensemble creation process.</p>
+it tries to allocate more than <code class="docutils literal notranslate"><span class="pre">memory_limit</span></code> MB.</p>
+<p><strong>Important notes:</strong></p>
+<ul class="simple">
+<li><p>If <code class="docutils literal notranslate"><span class="pre">None</span></code> is provided, no memory limit is set.</p></li>
+<li><p>In case of multi-processing, <code class="docutils literal notranslate"><span class="pre">memory_limit</span></code> will be <em>per job</em>, so the total usage is
+<code class="docutils literal notranslate"><span class="pre">n_jobs</span> <span class="pre">x</span> <span class="pre">memory_limit</span></code>.</p></li>
+<li><p>The memory limit also applies to the ensemble creation process.</p></li>
+</ul>
 </dd>
-<dt><strong>include</strong><span class="classifier">dict, optional (None)</span></dt><dd><p>If None, all possible algorithms are used. Otherwise specifies
-set of algorithms for each added component is used. Include and 
-exclude are incompatible if used together on the same component</p>
+<dt><strong>include</strong><span class="classifier">Optional[Dict[str, List[str]]] = None</span></dt><dd><p>If None, all possible algorithms are used.</p>
+<p>Otherwise, specifies a step and the components that are included in search.
+See <code class="docutils literal notranslate"><span class="pre">/pipeline/components/&lt;step&gt;/*</span></code> for available components.</p>
+<p>Incompatible with parameter <code class="docutils literal notranslate"><span class="pre">exclude</span></code>.</p>
+<p><strong>Possible Steps</strong>:</p>
+<ul class="simple">
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;data_preprocessor&quot;</span></code></p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;balancing&quot;</span></code></p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;feature_preprocessor&quot;</span></code></p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;classifier&quot;</span></code> - Only for when when using <code class="docutils literal notranslate"><span class="pre">AutoSklearnClasssifier</span></code></p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;regressor&quot;</span></code> - Only for when when using <code class="docutils literal notranslate"><span class="pre">AutoSklearnRegressor</span></code></p></li>
+</ul>
+<p><strong>Example</strong>:</p>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="n">include</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="s1">&#39;classifier&#39;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;random_forest&quot;</span><span class="p">],</span>
+    <span class="s1">&#39;feature_preprocessor&#39;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;no_preprocessing&quot;</span><span class="p">]</span>
+<span class="p">}</span>
+</pre></div>
+</div>
 </dd>
-<dt><strong>exclude</strong><span class="classifier">dict, optional (None)</span></dt><dd><p>If None, all possible algorithms are used. Otherwise specifies
-set of algorithms for each added component is not used.
-Incompatible with include. Include and exclude are incompatible
-if used together on the same component</p>
+<dt><strong>exclude</strong><span class="classifier">Optional[Dict[str, List[str]]] = None</span></dt><dd><p>If None, all possible algorithms are used.</p>
+<p>Otherwise, specifies a step and the components that are excluded from search.
+See <code class="docutils literal notranslate"><span class="pre">/pipeline/components/&lt;step&gt;/*</span></code> for available components.</p>
+<p>Incompatible with parameter <code class="docutils literal notranslate"><span class="pre">include</span></code>.</p>
+<p><strong>Possible Steps</strong>:</p>
+<ul class="simple">
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;data_preprocessor&quot;</span></code></p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;balancing&quot;</span></code></p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;feature_preprocessor&quot;</span></code></p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;classifier&quot;</span></code> - Only for when when using <code class="docutils literal notranslate"><span class="pre">AutoSklearnClasssifier</span></code></p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;regressor&quot;</span></code> - Only for when when using <code class="docutils literal notranslate"><span class="pre">AutoSklearnRegressor</span></code></p></li>
+</ul>
+<p><strong>Example</strong>:</p>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="n">exclude</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="s1">&#39;classifier&#39;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;random_forest&quot;</span><span class="p">],</span>
+    <span class="s1">&#39;feature_preprocessor&#39;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;no_preprocessing&quot;</span><span class="p">]</span>
+<span class="p">}</span>
+</pre></div>
+</div>
 </dd>
 <dt><strong>resampling_strategy</strong><span class="classifier">string or object, optional (‘holdout’)</span></dt><dd><p>how to to handle overfitting, might need ‘resampling_strategy_arguments’</p>
 <ul class="simple">
@@ -231,14 +265,11 @@ <h3>Classification<a class="headerlink" href="#classification" title="Permalink
 <li><p>‘cv’: {‘folds’: int}</p></li>
 <li><p>‘cv-iterative-fit’: {‘folds’: int}</p></li>
 <li><p>‘partial-cv’: {‘folds’: int, ‘shuffle’: bool}</p></li>
-<li><dl class="simple">
-<dt>BaseCrossValidator or _RepeatedSplits or BaseShuffleSplit object: all arguments</dt><dd><p>required by chosen class as specified in scikit-learn documentation.
+<li><p>BaseCrossValidator or _RepeatedSplits or BaseShuffleSplit object: all arguments
+required by chosen class as specified in scikit-learn documentation.
 If arguments are not provided, scikit-learn defaults are used.
 If no defaults are available, an exception is raised.
-Refer to the ‘n_splits’ argument as ‘folds’.</p>
-</dd>
-</dl>
-</li>
+Refer to the ‘n_splits’ argument as ‘folds’.</p></li>
 </ul>
 </dd>
 <dt><strong>tmp_folder</strong><span class="classifier">string, optional (None)</span></dt><dd><p>folder to store configuration output and log files, if <code class="docutils literal notranslate"><span class="pre">None</span></code>
@@ -248,13 +279,15 @@ <h3>Classification<a class="headerlink" href="#classification" title="Permalink
 tmp_dir will always be deleted</p>
 </dd>
 <dt><strong>n_jobs</strong><span class="classifier">int, optional, experimental</span></dt><dd><p>The number of jobs to run in parallel for <code class="docutils literal notranslate"><span class="pre">fit()</span></code>. <code class="docutils literal notranslate"><span class="pre">-1</span></code> means
-using all processors. By default, Auto-sklearn uses a single core
-for fitting the machine learning model and a single core for fitting
-an ensemble. Ensemble building is not affected by <code class="docutils literal notranslate"><span class="pre">n_jobs</span></code> but
-can be controlled by the number of models in the ensemble. In
-contrast to most scikit-learn models, <code class="docutils literal notranslate"><span class="pre">n_jobs</span></code> given in the
-constructor is not applied to the <code class="docutils literal notranslate"><span class="pre">predict()</span></code> method. If
-<code class="docutils literal notranslate"><span class="pre">dask_client</span></code> is None, a new dask client is created.</p>
+using all processors.</p>
+<p><strong>Important notes</strong>:</p>
+<ul class="simple">
+<li><p>By default, Auto-sklearn uses one core.</p></li>
+<li><p>Ensemble building is not affected by <code class="docutils literal notranslate"><span class="pre">n_jobs</span></code> but can be controlled by the number 
+of models in the ensemble.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">predict()</span></code> is not affected by <code class="docutils literal notranslate"><span class="pre">n_jobs</span></code> (in contrast to most scikit-learn models)</p></li>
+<li><p>If <code class="docutils literal notranslate"><span class="pre">dask_client</span></code> is <code class="docutils literal notranslate"><span class="pre">None</span></code>, a new dask client is created.</p></li>
+</ul>
 </dd>
 <dt><strong>dask_client</strong><span class="classifier">dask.distributed.Client, optional</span></dt><dd><p>User-created dask client, can be used to start a dask cluster and then
 attach auto-sklearn to it.</p>
@@ -268,7 +301,7 @@ <h3>Classification<a class="headerlink" href="#classification" title="Permalink
 <li><p><code class="docutils literal notranslate"><span class="pre">'y_optimization'</span></code> : do not save the predictions for the
 optimization/validation set, which would later on be used to build
 an ensemble.</p></li>
-<li><p><code class="docutils literal notranslate"><span class="pre">'model'</span></code> : do not save any model files</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">model</span></code> : do not save any model files</p></li>
 </ul>
 </dd>
 <dt><strong>smac_scenario_args</strong><span class="classifier">dict, optional (None)</span></dt><dd><p>Additional arguments inserted into the scenario of SMAC. See the
@@ -511,7 +544,7 @@ <h3>Classification<a class="headerlink" href="#classification" title="Permalink
 <dd><p>Returns a pandas table of results for all evaluated models.</p>
 <p>Gives an overview of all models trained during the search process along
 with various statistics about their training.</p>
-<p>The availble statistics are:</p>
+<p>The available statistics are:</p>
 <p><strong>Simple</strong>:</p>
 <ul class="simple">
 <li><p><code class="docutils literal notranslate"><span class="pre">&quot;model_id&quot;</span></code> - The id given to a model by <code class="docutils literal notranslate"><span class="pre">autosklearn</span></code>.</p></li>
@@ -692,11 +725,66 @@ <h3>Classification<a class="headerlink" href="#classification" title="Permalink
 <dl class="py method">
 <dt class="sig sig-object py" id="autosklearn.classification.AutoSklearnClassifier.show_models">
 <span class="sig-name descname"><span class="pre">show_models</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="headerlink" href="#autosklearn.classification.AutoSklearnClassifier.show_models" title="Permalink to this definition">¶</a></dt>
-<dd><p>Return a representation of the final ensemble found by auto-sklearn.</p>
+<dd><p>Returns a dictionary containing dictionaries of ensemble models.</p>
+<p>Each model in the ensemble can be accessed by giving its <code class="docutils literal notranslate"><span class="pre">model_id</span></code> as key.</p>
+<p>A model dictionary contains the following:</p>
+<ul class="simple">
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;model_id&quot;</span></code> - The id given to a model by <code class="docutils literal notranslate"><span class="pre">autosklearn</span></code>.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;rank&quot;</span></code> - The rank of the model based on it’s <code class="docutils literal notranslate"><span class="pre">&quot;cost&quot;</span></code>.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;cost&quot;</span></code> - The loss of the model on the validation set.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;ensemble_weight&quot;</span></code> - The weight given to the model in the ensemble.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;voting_model&quot;</span></code> - The <code class="docutils literal notranslate"><span class="pre">cv_voting_ensemble</span></code> model (for ‘cv’ resampling).</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;estimators&quot;</span></code> - List of models (dicts) in <code class="docutils literal notranslate"><span class="pre">cv_voting_ensemble</span></code> (for ‘cv’ resampling).</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;data_preprocessor&quot;</span></code> - The preprocessor used on the data.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;balancing&quot;</span></code> - The balancing used on the data (for classification).</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;feature_preprocessor&quot;</span></code> - The preprocessor for features types.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;classifier&quot;</span></code> or <code class="docutils literal notranslate"><span class="pre">&quot;regressor&quot;</span></code> - The autosklearn wrapped classifier or regressor.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;sklearn_classifier&quot;</span></code> or <code class="docutils literal notranslate"><span class="pre">&quot;sklearn_regressor&quot;</span></code> - The sklearn classifier or regressor.</p></li>
+</ul>
+<p><strong>Example</strong></p>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">sklearn.datasets</span>
+<span class="kn">import</span> <span class="nn">sklearn.metrics</span>
+<span class="kn">import</span> <span class="nn">autosklearn.regression</span>
+
+<span class="n">X</span><span class="p">,</span> <span class="n">y</span> <span class="o">=</span> <span class="n">sklearn</span><span class="o">.</span><span class="n">datasets</span><span class="o">.</span><span class="n">load_diabetes</span><span class="p">(</span><span class="n">return_X_y</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+
+<span class="n">automl</span> <span class="o">=</span> <span class="n">autosklearn</span><span class="o">.</span><span class="n">regression</span><span class="o">.</span><span class="n">AutoSklearnRegressor</span><span class="p">(</span>
+    <span class="n">time_left_for_this_task</span><span class="o">=</span><span class="mi">120</span>
+    <span class="p">)</span>
+<span class="n">automl</span><span class="o">.</span><span class="n">fit</span><span class="p">(</span><span class="n">X_train</span><span class="p">,</span> <span class="n">y_train</span><span class="p">,</span> <span class="n">dataset_name</span><span class="o">=</span><span class="s1">&#39;diabetes&#39;</span><span class="p">)</span>
+
+<span class="n">ensemble_dict</span> <span class="o">=</span> <span class="n">automl</span><span class="o">.</span><span class="n">show_models</span><span class="p">()</span>
+<span class="nb">print</span><span class="p">(</span><span class="n">ensemble_dict</span><span class="p">)</span>
+</pre></div>
+</div>
+<p>Output:</p>
+<div class="highlight-text notranslate"><div class="highlight"><pre><span></span>{
+    25: {&#39;model_id&#39;: 25.0,
+         &#39;rank&#39;: 1,
+         &#39;cost&#39;: 0.43667876507897496,
+         &#39;ensemble_weight&#39;: 0.38,
+         &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing....&gt;,
+         &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components....&gt;,
+         &#39;regressor&#39;: &lt;autosklearn.pipeline.components.regression....&gt;,
+         &#39;sklearn_regressor&#39;: SGDRegressor(alpha=0.0006517033225329654,...)
+        },
+    6: {&#39;model_id&#39;: 6.0,
+        &#39;rank&#39;: 2,
+        &#39;cost&#39;: 0.4550418898836528,
+        &#39;ensemble_weight&#39;: 0.3,
+        &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing....&gt;,
+        &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components....&gt;,
+        &#39;regressor&#39;: &lt;autosklearn.pipeline.components.regression....&gt;,
+        &#39;sklearn_regressor&#39;: ARDRegression(alpha_1=0.0003701926442639788,...)
+        }...
+}
+</pre></div>
+</div>
 <dl class="field-list simple">
 <dt class="field-odd">Returns</dt>
 <dd class="field-odd"><dl class="simple">
-<dt>str</dt><dd></dd>
+<dt><strong>Dict(int, Any)</strong><span class="classifier">dictionary of length = number of models in the ensemble</span></dt><dd><p>A dictionary of models in the ensemble, where <code class="docutils literal notranslate"><span class="pre">model_id</span></code> is the key.</p>
+</dd>
 </dl>
 </dd>
 </dl>
@@ -732,7 +820,7 @@ <h3>Classification<a class="headerlink" href="#classification" title="Permalink
 <em class="property"><span class="pre">class</span> </em><span class="sig-prename descclassname"><span class="pre">autosklearn.experimental.askl2.</span></span><span class="sig-name descname"><span class="pre">AutoSklearn2Classifier</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">time_left_for_this_task</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">int</span></span> <span class="o"><span class="pre">=</span></span> <span class="default_value"><span class="pre">3600</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">per_run_time_limit</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">ensemble_size</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">int</span></span> <span class="o"><span class="pre">=</span></span> <span class="default_value"><span class="pre">50</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">ensemble_nbest</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">Union</span><span class="p"><span class="pre">[</span></span><span class="pre">float</span><span class="p"><span class="pre">,</span> </span><span class="pre">int</span><span class="p"><span class="pre">]</span></span></span> <span class="o"><span class="pre">=</span></span> <span class="default_value"><span class="pre">50</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">max_models_on_disc</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">int</span></span> <span class="o"><span class="pre">=</span></span> <span class="default_value"><span class="pre">50</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">seed</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">int</span></span> <span class="o"><span class="pre">=</span></span> <span class="default_value"><span class="pre">1</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">memory_limit</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">int</span></span> <span class="o"><span class="pre">=</span></span> <span class="default_value"><span class="pre">3072</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tmp_folder</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">Optional</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">]</span></span></span> <span class="o"><span class="pre">=</span></span> <span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">delete_tmp_folder_after_terminate</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">bool</span></span> <span class="o"><span class="pre">=</span></span> <span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">n_jobs</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">Optional</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span></span> <span class="o"><span class="pre">=</span></span> <span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">dask_client</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">Optional</span><span class="p"><span class="pre">[</span></span><span class="pre">distributed.client.Client</span><span class="p"><span class="pre">]</span></span></span> <span class="o"><span class="pre">=</span></span> <span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">disable_evaluator_output</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">bool</span></span> <span class="o"><span class="pre">=</span></span> <span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">smac_scenario_args</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">Optional</span><span class="p"><span class="pre">[</span></span><span class="pre">Dict</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span> </span><span class="pre">Any</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span> <span class="o"><span class="pre">=</span></span> <span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">logging_config</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">Optional</span><span class="p"><span class="pre">[</span></span><span class="pre">Dict</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span> </span><span class="pre">Any</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span> <span class="o"><span class="pre">=</span></span> <span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">metric</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">Optional</span><span class="p"><span class="pre">[</span></span><span class="pre">autosklearn.metrics.Scorer</span><span class="p"><span class="pre">]</span></span></span> <span class="o"><span class="pre">=</span></span> <span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">scoring_functions</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">Optional</span><span class="p"><span class="pre">[</span></span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">autosklearn.metrics.Scorer</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span> <span class="o"><span class="pre">=</span></span> <span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">load_models</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">bool</span></span> <span class="o"><span class="pre">=</span></span> <span class="default_value"><span class="pre">True</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/autosklearn/experimental/askl2.html#AutoSklearn2Classifier"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#autosklearn.experimental.askl2.AutoSklearn2Classifier" title="Permalink to this definition">¶</a></dt>
 <dd><dl class="field-list">
 <dt class="field-odd">Parameters</dt>
-<dd class="field-odd"><dl class="simple">
+<dd class="field-odd"><dl>
 <dt><strong>time_left_for_this_task</strong><span class="classifier">int, optional (default=3600)</span></dt><dd><p>Time limit in seconds for the search of appropriate
 models. By increasing this value, <em>auto-sklearn</em> has a higher
 chance of finding better models.</p>
@@ -745,7 +833,7 @@ <h3>Classification<a class="headerlink" href="#classification" title="Permalink
 </dd>
 <dt><strong>ensemble_size</strong><span class="classifier">int, optional (default=50)</span></dt><dd><p>Number of models added to the ensemble built by <em>Ensemble
 selection from libraries of models</em>. Models are drawn with
-replacement.</p>
+replacement. If set to <code class="docutils literal notranslate"><span class="pre">0</span></code> no ensemble is fit.</p>
 </dd>
 <dt><strong>ensemble_nbest</strong><span class="classifier">int, optional (default=50)</span></dt><dd><p>Only consider the <code class="docutils literal notranslate"><span class="pre">ensemble_nbest</span></code> models when building an
 ensemble.</p>
@@ -761,10 +849,14 @@ <h3>Classification<a class="headerlink" href="#classification" title="Permalink
 </dd>
 <dt><strong>memory_limit</strong><span class="classifier">int, optional (3072)</span></dt><dd><p>Memory limit in MB for the machine learning algorithm.
 <cite>auto-sklearn</cite> will stop fitting the machine learning algorithm if
-it tries to allocate more than <cite>memory_limit</cite> MB.
-If None is provided, no memory limit is set.
-In case of multi-processing, <cite>memory_limit</cite> will be per job.
-This memory limit also applies to the ensemble creation process.</p>
+it tries to allocate more than <code class="docutils literal notranslate"><span class="pre">memory_limit</span></code> MB.</p>
+<p><strong>Important notes:</strong></p>
+<ul class="simple">
+<li><p>If <code class="docutils literal notranslate"><span class="pre">None</span></code> is provided, no memory limit is set.</p></li>
+<li><p>In case of multi-processing, <code class="docutils literal notranslate"><span class="pre">memory_limit</span></code> will be <em>per job</em>, so the total usage is 
+<code class="docutils literal notranslate"><span class="pre">n_jobs</span> <span class="pre">x</span> <span class="pre">memory_limit</span></code>.</p></li>
+<li><p>The memory limit also applies to the ensemble creation process.</p></li>
+</ul>
 </dd>
 <dt><strong>tmp_folder</strong><span class="classifier">string, optional (None)</span></dt><dd><p>folder to store configuration output and log files, if <code class="docutils literal notranslate"><span class="pre">None</span></code>
 automatically use <code class="docutils literal notranslate"><span class="pre">/tmp/autosklearn_tmp_$pid_$random_number</span></code></p>
@@ -773,13 +865,15 @@ <h3>Classification<a class="headerlink" href="#classification" title="Permalink
 tmp_dir will always be deleted</p>
 </dd>
 <dt><strong>n_jobs</strong><span class="classifier">int, optional, experimental</span></dt><dd><p>The number of jobs to run in parallel for <code class="docutils literal notranslate"><span class="pre">fit()</span></code>. <code class="docutils literal notranslate"><span class="pre">-1</span></code> means
-using all processors. By default, Auto-sklearn uses a single core
-for fitting the machine learning model and a single core for fitting
-an ensemble. Ensemble building is not affected by <code class="docutils literal notranslate"><span class="pre">n_jobs</span></code> but
-can be controlled by the number of models in the ensemble. In
-contrast to most scikit-learn models, <code class="docutils literal notranslate"><span class="pre">n_jobs</span></code> given in the
-constructor is not applied to the <code class="docutils literal notranslate"><span class="pre">predict()</span></code> method. If
-<code class="docutils literal notranslate"><span class="pre">dask_client</span></code> is None, a new dask client is created.</p>
+using all processors.</p>
+<p><strong>Important notes</strong>:</p>
+<ul class="simple">
+<li><p>By default, Auto-sklearn uses one core.</p></li>
+<li><p>Ensemble building is not affected by <code class="docutils literal notranslate"><span class="pre">n_jobs</span></code> but can be controlled by the number 
+of models in the ensemble.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">predict()</span></code> is not affected by <code class="docutils literal notranslate"><span class="pre">n_jobs</span></code> (in contrast to most scikit-learn models)</p></li>
+<li><p>If <code class="docutils literal notranslate"><span class="pre">dask_client</span></code> is <code class="docutils literal notranslate"><span class="pre">None</span></code>, a new dask client is created.</p></li>
+</ul>
 </dd>
 <dt><strong>dask_client</strong><span class="classifier">dask.distributed.Client, optional</span></dt><dd><p>User-created dask client, can be used to start a dask cluster and then
 attach auto-sklearn to it.</p>
@@ -793,7 +887,7 @@ <h3>Classification<a class="headerlink" href="#classification" title="Permalink
 <li><p><code class="docutils literal notranslate"><span class="pre">'y_optimization'</span></code> : do not save the predictions for the
 optimization/validation set, which would later on be used to build
 an ensemble.</p></li>
-<li><p><code class="docutils literal notranslate"><span class="pre">'model'</span></code> : do not save any model files</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">model</span></code> : do not save any model files</p></li>
 </ul>
 </dd>
 <dt><strong>smac_scenario_args</strong><span class="classifier">dict, optional (None)</span></dt><dd><p>Additional arguments inserted into the scenario of SMAC. See the
@@ -1017,7 +1111,7 @@ <h3>Classification<a class="headerlink" href="#classification" title="Permalink
 <dd><p>Returns a pandas table of results for all evaluated models.</p>
 <p>Gives an overview of all models trained during the search process along
 with various statistics about their training.</p>
-<p>The availble statistics are:</p>
+<p>The available statistics are:</p>
 <p><strong>Simple</strong>:</p>
 <ul class="simple">
 <li><p><code class="docutils literal notranslate"><span class="pre">&quot;model_id&quot;</span></code> - The id given to a model by <code class="docutils literal notranslate"><span class="pre">autosklearn</span></code>.</p></li>
@@ -1198,11 +1292,66 @@ <h3>Classification<a class="headerlink" href="#classification" title="Permalink
 <dl class="py method">
 <dt class="sig sig-object py" id="autosklearn.experimental.askl2.AutoSklearn2Classifier.show_models">
 <span class="sig-name descname"><span class="pre">show_models</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="headerlink" href="#autosklearn.experimental.askl2.AutoSklearn2Classifier.show_models" title="Permalink to this definition">¶</a></dt>
-<dd><p>Return a representation of the final ensemble found by auto-sklearn.</p>
+<dd><p>Returns a dictionary containing dictionaries of ensemble models.</p>
+<p>Each model in the ensemble can be accessed by giving its <code class="docutils literal notranslate"><span class="pre">model_id</span></code> as key.</p>
+<p>A model dictionary contains the following:</p>
+<ul class="simple">
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;model_id&quot;</span></code> - The id given to a model by <code class="docutils literal notranslate"><span class="pre">autosklearn</span></code>.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;rank&quot;</span></code> - The rank of the model based on it’s <code class="docutils literal notranslate"><span class="pre">&quot;cost&quot;</span></code>.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;cost&quot;</span></code> - The loss of the model on the validation set.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;ensemble_weight&quot;</span></code> - The weight given to the model in the ensemble.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;voting_model&quot;</span></code> - The <code class="docutils literal notranslate"><span class="pre">cv_voting_ensemble</span></code> model (for ‘cv’ resampling).</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;estimators&quot;</span></code> - List of models (dicts) in <code class="docutils literal notranslate"><span class="pre">cv_voting_ensemble</span></code> (for ‘cv’ resampling).</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;data_preprocessor&quot;</span></code> - The preprocessor used on the data.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;balancing&quot;</span></code> - The balancing used on the data (for classification).</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;feature_preprocessor&quot;</span></code> - The preprocessor for features types.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;classifier&quot;</span></code> or <code class="docutils literal notranslate"><span class="pre">&quot;regressor&quot;</span></code> - The autosklearn wrapped classifier or regressor.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;sklearn_classifier&quot;</span></code> or <code class="docutils literal notranslate"><span class="pre">&quot;sklearn_regressor&quot;</span></code> - The sklearn classifier or regressor.</p></li>
+</ul>
+<p><strong>Example</strong></p>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">sklearn.datasets</span>
+<span class="kn">import</span> <span class="nn">sklearn.metrics</span>
+<span class="kn">import</span> <span class="nn">autosklearn.regression</span>
+
+<span class="n">X</span><span class="p">,</span> <span class="n">y</span> <span class="o">=</span> <span class="n">sklearn</span><span class="o">.</span><span class="n">datasets</span><span class="o">.</span><span class="n">load_diabetes</span><span class="p">(</span><span class="n">return_X_y</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+
+<span class="n">automl</span> <span class="o">=</span> <span class="n">autosklearn</span><span class="o">.</span><span class="n">regression</span><span class="o">.</span><span class="n">AutoSklearnRegressor</span><span class="p">(</span>
+    <span class="n">time_left_for_this_task</span><span class="o">=</span><span class="mi">120</span>
+    <span class="p">)</span>
+<span class="n">automl</span><span class="o">.</span><span class="n">fit</span><span class="p">(</span><span class="n">X_train</span><span class="p">,</span> <span class="n">y_train</span><span class="p">,</span> <span class="n">dataset_name</span><span class="o">=</span><span class="s1">&#39;diabetes&#39;</span><span class="p">)</span>
+
+<span class="n">ensemble_dict</span> <span class="o">=</span> <span class="n">automl</span><span class="o">.</span><span class="n">show_models</span><span class="p">()</span>
+<span class="nb">print</span><span class="p">(</span><span class="n">ensemble_dict</span><span class="p">)</span>
+</pre></div>
+</div>
+<p>Output:</p>
+<div class="highlight-text notranslate"><div class="highlight"><pre><span></span>{
+    25: {&#39;model_id&#39;: 25.0,
+         &#39;rank&#39;: 1,
+         &#39;cost&#39;: 0.43667876507897496,
+         &#39;ensemble_weight&#39;: 0.38,
+         &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing....&gt;,
+         &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components....&gt;,
+         &#39;regressor&#39;: &lt;autosklearn.pipeline.components.regression....&gt;,
+         &#39;sklearn_regressor&#39;: SGDRegressor(alpha=0.0006517033225329654,...)
+        },
+    6: {&#39;model_id&#39;: 6.0,
+        &#39;rank&#39;: 2,
+        &#39;cost&#39;: 0.4550418898836528,
+        &#39;ensemble_weight&#39;: 0.3,
+        &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing....&gt;,
+        &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components....&gt;,
+        &#39;regressor&#39;: &lt;autosklearn.pipeline.components.regression....&gt;,
+        &#39;sklearn_regressor&#39;: ARDRegression(alpha_1=0.0003701926442639788,...)
+        }...
+}
+</pre></div>
+</div>
 <dl class="field-list simple">
 <dt class="field-odd">Returns</dt>
 <dd class="field-odd"><dl class="simple">
-<dt>str</dt><dd></dd>
+<dt><strong>Dict(int, Any)</strong><span class="classifier">dictionary of length = number of models in the ensemble</span></dt><dd><p>A dictionary of models in the ensemble, where <code class="docutils literal notranslate"><span class="pre">model_id</span></code> is the key.</p>
+</dd>
 </dl>
 </dd>
 </dl>
@@ -1233,12 +1382,12 @@ <h3>Classification<a class="headerlink" href="#classification" title="Permalink
 
 </dd></dl>
 
-</section>
-<section id="regression">
+</div>
+<div class="section" id="regression">
 <h3>Regression<a class="headerlink" href="#regression" title="Permalink to this headline">¶</a></h3>
 <dl class="py class">
 <dt class="sig sig-object py" id="autosklearn.regression.AutoSklearnRegressor">
-<em class="property"><span class="pre">class</span> </em><span class="sig-prename descclassname"><span class="pre">autosklearn.regression.</span></span><span class="sig-name descname"><span class="pre">AutoSklearnRegressor</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">time_left_for_this_task</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">3600</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">per_run_time_limit</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">initial_configurations_via_metalearning</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">25</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">ensemble_size</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">int</span></span> <span class="o"><span class="pre">=</span></span> <span class="default_value"><span class="pre">50</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">ensemble_nbest</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">50</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">max_models_on_disc</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">50</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">seed</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">1</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">memory_limit</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">3072</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">include</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">exclude</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">resampling_strategy</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">'holdout'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">resampling_strategy_arguments</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tmp_folder</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">delete_tmp_folder_after_terminate</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">n_jobs</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">Optional</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span></span> <span class="o"><span class="pre">=</span></span> <span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">dask_client</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">Optional</span><span class="p"><span class="pre">[</span></span><span class="pre">distributed.client.Client</span><span class="p"><span class="pre">]</span></span></span> <span class="o"><span class="pre">=</span></span> <span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">disable_evaluator_output</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">get_smac_object_callback</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">smac_scenario_args</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">logging_config</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">metadata_directory</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">metric</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">scoring_functions</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">Optional</span><span class="p"><span class="pre">[</span></span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">autosklearn.metrics.Scorer</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span> <span class="o"><span class="pre">=</span></span> <span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">load_models</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">bool</span></span> <span class="o"><span class="pre">=</span></span> <span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">get_trials_callback</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/autosklearn/estimators.html#AutoSklearnRegressor"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#autosklearn.regression.AutoSklearnRegressor" title="Permalink to this definition">¶</a></dt>
+<em class="property"><span class="pre">class</span> </em><span class="sig-prename descclassname"><span class="pre">autosklearn.regression.</span></span><span class="sig-name descname"><span class="pre">AutoSklearnRegressor</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">time_left_for_this_task</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">3600</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">per_run_time_limit</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">initial_configurations_via_metalearning</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">25</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">ensemble_size</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">int</span></span> <span class="o"><span class="pre">=</span></span> <span class="default_value"><span class="pre">50</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">ensemble_nbest</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">50</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">max_models_on_disc</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">50</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">seed</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">1</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">memory_limit</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">3072</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">include</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">Optional</span><span class="p"><span class="pre">[</span></span><span class="pre">Dict</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span> <span class="o"><span class="pre">=</span></span> <span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">exclude</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">Optional</span><span class="p"><span class="pre">[</span></span><span class="pre">Dict</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">,</span> </span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">str</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span> <span class="o"><span class="pre">=</span></span> <span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">resampling_strategy</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">'holdout'</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">resampling_strategy_arguments</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">tmp_folder</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">delete_tmp_folder_after_terminate</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">n_jobs</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">Optional</span><span class="p"><span class="pre">[</span></span><span class="pre">int</span><span class="p"><span class="pre">]</span></span></span> <span class="o"><span class="pre">=</span></span> <span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">dask_client</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">Optional</span><span class="p"><span class="pre">[</span></span><span class="pre">distributed.client.Client</span><span class="p"><span class="pre">]</span></span></span> <span class="o"><span class="pre">=</span></span> <span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">disable_evaluator_output</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">False</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">get_smac_object_callback</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">smac_scenario_args</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">logging_config</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">metadata_directory</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">metric</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">scoring_functions</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">Optional</span><span class="p"><span class="pre">[</span></span><span class="pre">List</span><span class="p"><span class="pre">[</span></span><span class="pre">autosklearn.metrics.Scorer</span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span> <span class="o"><span class="pre">=</span></span> <span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">load_models</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">bool</span></span> <span class="o"><span class="pre">=</span></span> <span class="default_value"><span class="pre">True</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">get_trials_callback</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span><a class="reference internal" href="_modules/autosklearn/estimators.html#AutoSklearnRegressor"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#autosklearn.regression.AutoSklearnRegressor" title="Permalink to this definition">¶</a></dt>
 <dd><p>This class implements the regression task.</p>
 <dl class="field-list">
 <dt class="field-odd">Parameters</dt>
@@ -1260,7 +1409,7 @@ <h3>Regression<a class="headerlink" href="#regression" title="Permalink to this
 </dd>
 <dt><strong>ensemble_size</strong><span class="classifier">int, optional (default=50)</span></dt><dd><p>Number of models added to the ensemble built by <em>Ensemble
 selection from libraries of models</em>. Models are drawn with
-replacement.</p>
+replacement. If set to <code class="docutils literal notranslate"><span class="pre">0</span></code> no ensemble is fit.</p>
 </dd>
 <dt><strong>ensemble_nbest</strong><span class="classifier">int, optional (default=50)</span></dt><dd><p>Only consider the <code class="docutils literal notranslate"><span class="pre">ensemble_nbest</span></code> models when building an
 ensemble.</p>
@@ -1276,19 +1425,54 @@ <h3>Regression<a class="headerlink" href="#regression" title="Permalink to this
 </dd>
 <dt><strong>memory_limit</strong><span class="classifier">int, optional (3072)</span></dt><dd><p>Memory limit in MB for the machine learning algorithm.
 <cite>auto-sklearn</cite> will stop fitting the machine learning algorithm if
-it tries to allocate more than <cite>memory_limit</cite> MB.
-If None is provided, no memory limit is set.
-In case of multi-processing, <cite>memory_limit</cite> will be per job.
-This memory limit also applies to the ensemble creation process.</p>
+it tries to allocate more than <code class="docutils literal notranslate"><span class="pre">memory_limit</span></code> MB.</p>
+<p><strong>Important notes:</strong></p>
+<ul class="simple">
+<li><p>If <code class="docutils literal notranslate"><span class="pre">None</span></code> is provided, no memory limit is set.</p></li>
+<li><p>In case of multi-processing, <code class="docutils literal notranslate"><span class="pre">memory_limit</span></code> will be <em>per job</em>, so the total usage is
+<code class="docutils literal notranslate"><span class="pre">n_jobs</span> <span class="pre">x</span> <span class="pre">memory_limit</span></code>.</p></li>
+<li><p>The memory limit also applies to the ensemble creation process.</p></li>
+</ul>
 </dd>
-<dt><strong>include</strong><span class="classifier">dict, optional (None)</span></dt><dd><p>If None, all possible algorithms are used. Otherwise specifies
-set of algorithms for each added component is used. Include and 
-exclude are incompatible if used together on the same component</p>
+<dt><strong>include</strong><span class="classifier">Optional[Dict[str, List[str]]] = None</span></dt><dd><p>If None, all possible algorithms are used.</p>
+<p>Otherwise, specifies a step and the components that are included in search.
+See <code class="docutils literal notranslate"><span class="pre">/pipeline/components/&lt;step&gt;/*</span></code> for available components.</p>
+<p>Incompatible with parameter <code class="docutils literal notranslate"><span class="pre">exclude</span></code>.</p>
+<p><strong>Possible Steps</strong>:</p>
+<ul class="simple">
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;data_preprocessor&quot;</span></code></p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;balancing&quot;</span></code></p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;feature_preprocessor&quot;</span></code></p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;classifier&quot;</span></code> - Only for when when using <code class="docutils literal notranslate"><span class="pre">AutoSklearnClasssifier</span></code></p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;regressor&quot;</span></code> - Only for when when using <code class="docutils literal notranslate"><span class="pre">AutoSklearnRegressor</span></code></p></li>
+</ul>
+<p><strong>Example</strong>:</p>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="n">include</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="s1">&#39;classifier&#39;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;random_forest&quot;</span><span class="p">],</span>
+    <span class="s1">&#39;feature_preprocessor&#39;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;no_preprocessing&quot;</span><span class="p">]</span>
+<span class="p">}</span>
+</pre></div>
+</div>
 </dd>
-<dt><strong>exclude</strong><span class="classifier">dict, optional (None)</span></dt><dd><p>If None, all possible algorithms are used. Otherwise specifies
-set of algorithms for each added component is not used.
-Incompatible with include. Include and exclude are incompatible
-if used together on the same component</p>
+<dt><strong>exclude</strong><span class="classifier">Optional[Dict[str, List[str]]] = None</span></dt><dd><p>If None, all possible algorithms are used.</p>
+<p>Otherwise, specifies a step and the components that are excluded from search.
+See <code class="docutils literal notranslate"><span class="pre">/pipeline/components/&lt;step&gt;/*</span></code> for available components.</p>
+<p>Incompatible with parameter <code class="docutils literal notranslate"><span class="pre">include</span></code>.</p>
+<p><strong>Possible Steps</strong>:</p>
+<ul class="simple">
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;data_preprocessor&quot;</span></code></p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;balancing&quot;</span></code></p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;feature_preprocessor&quot;</span></code></p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;classifier&quot;</span></code> - Only for when when using <code class="docutils literal notranslate"><span class="pre">AutoSklearnClasssifier</span></code></p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;regressor&quot;</span></code> - Only for when when using <code class="docutils literal notranslate"><span class="pre">AutoSklearnRegressor</span></code></p></li>
+</ul>
+<p><strong>Example</strong>:</p>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="n">exclude</span> <span class="o">=</span> <span class="p">{</span>
+    <span class="s1">&#39;classifier&#39;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;random_forest&quot;</span><span class="p">],</span>
+    <span class="s1">&#39;feature_preprocessor&#39;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;no_preprocessing&quot;</span><span class="p">]</span>
+<span class="p">}</span>
+</pre></div>
+</div>
 </dd>
 <dt><strong>resampling_strategy</strong><span class="classifier">string or object, optional (‘holdout’)</span></dt><dd><p>how to to handle overfitting, might need ‘resampling_strategy_arguments’</p>
 <ul class="simple">
@@ -1330,14 +1514,11 @@ <h3>Regression<a class="headerlink" href="#regression" title="Permalink to this
 <li><p>‘cv’: {‘folds’: int}</p></li>
 <li><p>‘cv-iterative-fit’: {‘folds’: int}</p></li>
 <li><p>‘partial-cv’: {‘folds’: int, ‘shuffle’: bool}</p></li>
-<li><dl class="simple">
-<dt>BaseCrossValidator or _RepeatedSplits or BaseShuffleSplit object: all arguments</dt><dd><p>required by chosen class as specified in scikit-learn documentation.
+<li><p>BaseCrossValidator or _RepeatedSplits or BaseShuffleSplit object: all arguments
+required by chosen class as specified in scikit-learn documentation.
 If arguments are not provided, scikit-learn defaults are used.
 If no defaults are available, an exception is raised.
-Refer to the ‘n_splits’ argument as ‘folds’.</p>
-</dd>
-</dl>
-</li>
+Refer to the ‘n_splits’ argument as ‘folds’.</p></li>
 </ul>
 </dd>
 <dt><strong>tmp_folder</strong><span class="classifier">string, optional (None)</span></dt><dd><p>folder to store configuration output and log files, if <code class="docutils literal notranslate"><span class="pre">None</span></code>
@@ -1347,13 +1528,15 @@ <h3>Regression<a class="headerlink" href="#regression" title="Permalink to this
 tmp_dir will always be deleted</p>
 </dd>
 <dt><strong>n_jobs</strong><span class="classifier">int, optional, experimental</span></dt><dd><p>The number of jobs to run in parallel for <code class="docutils literal notranslate"><span class="pre">fit()</span></code>. <code class="docutils literal notranslate"><span class="pre">-1</span></code> means
-using all processors. By default, Auto-sklearn uses a single core
-for fitting the machine learning model and a single core for fitting
-an ensemble. Ensemble building is not affected by <code class="docutils literal notranslate"><span class="pre">n_jobs</span></code> but
-can be controlled by the number of models in the ensemble. In
-contrast to most scikit-learn models, <code class="docutils literal notranslate"><span class="pre">n_jobs</span></code> given in the
-constructor is not applied to the <code class="docutils literal notranslate"><span class="pre">predict()</span></code> method. If
-<code class="docutils literal notranslate"><span class="pre">dask_client</span></code> is None, a new dask client is created.</p>
+using all processors.</p>
+<p><strong>Important notes</strong>:</p>
+<ul class="simple">
+<li><p>By default, Auto-sklearn uses one core.</p></li>
+<li><p>Ensemble building is not affected by <code class="docutils literal notranslate"><span class="pre">n_jobs</span></code> but can be controlled by the number 
+of models in the ensemble.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">predict()</span></code> is not affected by <code class="docutils literal notranslate"><span class="pre">n_jobs</span></code> (in contrast to most scikit-learn models)</p></li>
+<li><p>If <code class="docutils literal notranslate"><span class="pre">dask_client</span></code> is <code class="docutils literal notranslate"><span class="pre">None</span></code>, a new dask client is created.</p></li>
+</ul>
 </dd>
 <dt><strong>dask_client</strong><span class="classifier">dask.distributed.Client, optional</span></dt><dd><p>User-created dask client, can be used to start a dask cluster and then
 attach auto-sklearn to it.</p>
@@ -1367,7 +1550,7 @@ <h3>Regression<a class="headerlink" href="#regression" title="Permalink to this
 <li><p><code class="docutils literal notranslate"><span class="pre">'y_optimization'</span></code> : do not save the predictions for the
 optimization/validation set, which would later on be used to build
 an ensemble.</p></li>
-<li><p><code class="docutils literal notranslate"><span class="pre">'model'</span></code> : do not save any model files</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">model</span></code> : do not save any model files</p></li>
 </ul>
 </dd>
 <dt><strong>smac_scenario_args</strong><span class="classifier">dict, optional (None)</span></dt><dd><p>Additional arguments inserted into the scenario of SMAC. See the
@@ -1608,7 +1791,7 @@ <h3>Regression<a class="headerlink" href="#regression" title="Permalink to this
 <dd><p>Returns a pandas table of results for all evaluated models.</p>
 <p>Gives an overview of all models trained during the search process along
 with various statistics about their training.</p>
-<p>The availble statistics are:</p>
+<p>The available statistics are:</p>
 <p><strong>Simple</strong>:</p>
 <ul class="simple">
 <li><p><code class="docutils literal notranslate"><span class="pre">&quot;model_id&quot;</span></code> - The id given to a model by <code class="docutils literal notranslate"><span class="pre">autosklearn</span></code>.</p></li>
@@ -1782,11 +1965,66 @@ <h3>Regression<a class="headerlink" href="#regression" title="Permalink to this
 <dl class="py method">
 <dt class="sig sig-object py" id="autosklearn.regression.AutoSklearnRegressor.show_models">
 <span class="sig-name descname"><span class="pre">show_models</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="headerlink" href="#autosklearn.regression.AutoSklearnRegressor.show_models" title="Permalink to this definition">¶</a></dt>
-<dd><p>Return a representation of the final ensemble found by auto-sklearn.</p>
+<dd><p>Returns a dictionary containing dictionaries of ensemble models.</p>
+<p>Each model in the ensemble can be accessed by giving its <code class="docutils literal notranslate"><span class="pre">model_id</span></code> as key.</p>
+<p>A model dictionary contains the following:</p>
+<ul class="simple">
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;model_id&quot;</span></code> - The id given to a model by <code class="docutils literal notranslate"><span class="pre">autosklearn</span></code>.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;rank&quot;</span></code> - The rank of the model based on it’s <code class="docutils literal notranslate"><span class="pre">&quot;cost&quot;</span></code>.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;cost&quot;</span></code> - The loss of the model on the validation set.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;ensemble_weight&quot;</span></code> - The weight given to the model in the ensemble.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;voting_model&quot;</span></code> - The <code class="docutils literal notranslate"><span class="pre">cv_voting_ensemble</span></code> model (for ‘cv’ resampling).</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;estimators&quot;</span></code> - List of models (dicts) in <code class="docutils literal notranslate"><span class="pre">cv_voting_ensemble</span></code> (for ‘cv’ resampling).</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;data_preprocessor&quot;</span></code> - The preprocessor used on the data.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;balancing&quot;</span></code> - The balancing used on the data (for classification).</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;feature_preprocessor&quot;</span></code> - The preprocessor for features types.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;classifier&quot;</span></code> or <code class="docutils literal notranslate"><span class="pre">&quot;regressor&quot;</span></code> - The autosklearn wrapped classifier or regressor.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">&quot;sklearn_classifier&quot;</span></code> or <code class="docutils literal notranslate"><span class="pre">&quot;sklearn_regressor&quot;</span></code> - The sklearn classifier or regressor.</p></li>
+</ul>
+<p><strong>Example</strong></p>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">sklearn.datasets</span>
+<span class="kn">import</span> <span class="nn">sklearn.metrics</span>
+<span class="kn">import</span> <span class="nn">autosklearn.regression</span>
+
+<span class="n">X</span><span class="p">,</span> <span class="n">y</span> <span class="o">=</span> <span class="n">sklearn</span><span class="o">.</span><span class="n">datasets</span><span class="o">.</span><span class="n">load_diabetes</span><span class="p">(</span><span class="n">return_X_y</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+
+<span class="n">automl</span> <span class="o">=</span> <span class="n">autosklearn</span><span class="o">.</span><span class="n">regression</span><span class="o">.</span><span class="n">AutoSklearnRegressor</span><span class="p">(</span>
+    <span class="n">time_left_for_this_task</span><span class="o">=</span><span class="mi">120</span>
+    <span class="p">)</span>
+<span class="n">automl</span><span class="o">.</span><span class="n">fit</span><span class="p">(</span><span class="n">X_train</span><span class="p">,</span> <span class="n">y_train</span><span class="p">,</span> <span class="n">dataset_name</span><span class="o">=</span><span class="s1">&#39;diabetes&#39;</span><span class="p">)</span>
+
+<span class="n">ensemble_dict</span> <span class="o">=</span> <span class="n">automl</span><span class="o">.</span><span class="n">show_models</span><span class="p">()</span>
+<span class="nb">print</span><span class="p">(</span><span class="n">ensemble_dict</span><span class="p">)</span>
+</pre></div>
+</div>
+<p>Output:</p>
+<div class="highlight-text notranslate"><div class="highlight"><pre><span></span>{
+    25: {&#39;model_id&#39;: 25.0,
+         &#39;rank&#39;: 1,
+         &#39;cost&#39;: 0.43667876507897496,
+         &#39;ensemble_weight&#39;: 0.38,
+         &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing....&gt;,
+         &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components....&gt;,
+         &#39;regressor&#39;: &lt;autosklearn.pipeline.components.regression....&gt;,
+         &#39;sklearn_regressor&#39;: SGDRegressor(alpha=0.0006517033225329654,...)
+        },
+    6: {&#39;model_id&#39;: 6.0,
+        &#39;rank&#39;: 2,
+        &#39;cost&#39;: 0.4550418898836528,
+        &#39;ensemble_weight&#39;: 0.3,
+        &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing....&gt;,
+        &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components....&gt;,
+        &#39;regressor&#39;: &lt;autosklearn.pipeline.components.regression....&gt;,
+        &#39;sklearn_regressor&#39;: ARDRegression(alpha_1=0.0003701926442639788,...)
+        }...
+}
+</pre></div>
+</div>
 <dl class="field-list simple">
 <dt class="field-odd">Returns</dt>
 <dd class="field-odd"><dl class="simple">
-<dt>str</dt><dd></dd>
+<dt><strong>Dict(int, Any)</strong><span class="classifier">dictionary of length = number of models in the ensemble</span></dt><dd><p>A dictionary of models in the ensemble, where <code class="docutils literal notranslate"><span class="pre">model_id</span></code> is the key.</p>
+</dd>
 </dl>
 </dd>
 </dl>
@@ -1817,9 +2055,9 @@ <h3>Regression<a class="headerlink" href="#regression" title="Permalink to this
 
 </dd></dl>
 
-</section>
-</section>
-<section id="metrics">
+</div>
+</div>
+<div class="section" id="metrics">
 <h2>Metrics<a class="headerlink" href="#metrics" title="Permalink to this headline">¶</a></h2>
 <dl class="py function">
 <dt class="sig sig-object py" id="autosklearn.metrics.make_scorer">
@@ -1859,9 +2097,9 @@ <h2>Metrics<a class="headerlink" href="#metrics" title="Permalink to this headli
 </dl>
 </dd></dl>
 
-<section id="built-in-metrics">
+<div class="section" id="built-in-metrics">
 <h3>Built-in Metrics<a class="headerlink" href="#built-in-metrics" title="Permalink to this headline">¶</a></h3>
-<section id="classification-metrics">
+<div class="section" id="classification-metrics">
 <h4>Classification metrics<a class="headerlink" href="#classification-metrics" title="Permalink to this headline">¶</a></h4>
 <p>Note: The default <code class="docutils literal notranslate"><span class="pre">autosklearn.metrics.f1</span></code>, <code class="docutils literal notranslate"><span class="pre">autosklearn.metrics.precision</span></code> and <code class="docutils literal notranslate"><span class="pre">autosklearn.metrics.recall</span></code>
 built-in metrics are applicable only for binary classification. In order to apply them on multilabel and multiclass
@@ -1988,8 +2226,8 @@ <h4>Classification metrics<a class="headerlink" href="#classification-metrics" t
 <dd><p>alias of log_loss</p>
 </dd></dl>
 
-</section>
-<section id="regression-metrics">
+</div>
+<div class="section" id="regression-metrics">
 <h4>Regression metrics<a class="headerlink" href="#regression-metrics" title="Permalink to this headline">¶</a></h4>
 <dl class="py attribute">
 <dt class="sig sig-object py" id="autosklearn.metrics.r2">
@@ -2015,10 +2253,10 @@ <h4>Regression metrics<a class="headerlink" href="#regression-metrics" title="Pe
 <dd><p>alias of median_absolute_error</p>
 </dd></dl>
 
-</section>
-</section>
-</section>
-<section id="extension-interfaces">
+</div>
+</div>
+</div>
+<div class="section" id="extension-interfaces">
 <h2>Extension Interfaces<a class="headerlink" href="#extension-interfaces" title="Permalink to this headline">¶</a></h2>
 <dl class="py class">
 <dt class="sig sig-object py" id="autosklearn.pipeline.components.base.AutoSklearnClassificationAlgorithm">
@@ -2171,8 +2409,8 @@ <h2>Extension Interfaces<a class="headerlink" href="#extension-interfaces" title
 
 </dd></dl>
 
-</section>
-</section>
+</div>
+</div>
 
 
     </div>
@@ -2193,7 +2431,7 @@ <h2>Extension Interfaces<a class="headerlink" href="#extension-interfaces" title
       
     </p>
     <p>
-        &copy; Copyright 2014-2021, Machine Learning Professorship Freiburg.<br/>
+        &copy; Copyright 2014-2022, Machine Learning Professorship Freiburg.<br/>
       Created using <a href="http://sphinx-doc.org/">Sphinx</a> 4.2.0.<br/>
     </p>
   </div>
diff --git a/master/examples/20_basic/example_classification.html b/master/examples/20_basic/example_classification.html
index b809415215..7f68587190 100644
--- a/master/examples/20_basic/example_classification.html
+++ b/master/examples/20_basic/example_classification.html
@@ -3,9 +3,8 @@
 <html>
   <head>
     <meta charset="utf-8" />
-    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="generator" content="Docutils 0.17.1: http://docutils.sourceforge.net/" />
-
-    <title>Classification &#8212; AutoSklearn 0.14.2 documentation</title>
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>Classification &#8212; AutoSklearn 0.14.4 documentation</title>
     <link rel="stylesheet" type="text/css" href="../../_static/pygments.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/bootstrap-sphinx.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/sg_gallery.css" />
@@ -55,7 +54,7 @@
         </button>
         <a class="navbar-brand" href="../../index.html">
           auto-sklearn</a>
-        <span class="navbar-text navbar-version pull-left"><b>0.14.2</b></span>
+        <span class="navbar-text navbar-version pull-left"><b>0.14.4</b></span>
       </div>
 
         <div class="collapse navbar-collapse nav-collapse">
@@ -126,25 +125,27 @@
 <p>Click <a class="reference internal" href="#sphx-glr-download-examples-20-basic-example-classification-py"><span class="std std-ref">here</span></a>
 to download the full example code or to run this example in your browser via Binder</p>
 </div>
-<section class="sphx-glr-example-title" id="classification">
+<div class="sphx-glr-example-title section" id="classification">
 <span id="sphx-glr-examples-20-basic-example-classification-py"></span><h1>Classification<a class="headerlink" href="#classification" title="Permalink to this headline">¶</a></h1>
 <p>The following example shows how to fit a simple classification model with
 <em>auto-sklearn</em>.</p>
-<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">sklearn.datasets</span>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">pprint</span> <span class="kn">import</span> <span class="n">pprint</span>
+
+<span class="kn">import</span> <span class="nn">sklearn.datasets</span>
 <span class="kn">import</span> <span class="nn">sklearn.metrics</span>
 
 <span class="kn">import</span> <span class="nn">autosklearn.classification</span>
 </pre></div>
 </div>
-<section id="data-loading">
+<div class="section" id="data-loading">
 <h2>Data Loading<a class="headerlink" href="#data-loading" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">X</span><span class="p">,</span> <span class="n">y</span> <span class="o">=</span> <span class="n">sklearn</span><span class="o">.</span><span class="n">datasets</span><span class="o">.</span><span class="n">load_breast_cancer</span><span class="p">(</span><span class="n">return_X_y</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
 <span class="n">X_train</span><span class="p">,</span> <span class="n">X_test</span><span class="p">,</span> <span class="n">y_train</span><span class="p">,</span> <span class="n">y_test</span> <span class="o">=</span> \
     <span class="n">sklearn</span><span class="o">.</span><span class="n">model_selection</span><span class="o">.</span><span class="n">train_test_split</span><span class="p">(</span><span class="n">X</span><span class="p">,</span> <span class="n">y</span><span class="p">,</span> <span class="n">random_state</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
 </pre></div>
 </div>
-</section>
-<section id="build-and-fit-a-classifier">
+</div>
+<div class="section" id="build-and-fit-a-classifier">
 <h2>Build and fit a classifier<a class="headerlink" href="#build-and-fit-a-classifier" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">automl</span> <span class="o">=</span> <span class="n">autosklearn</span><span class="o">.</span><span class="n">classification</span><span class="o">.</span><span class="n">AutoSklearnClassifier</span><span class="p">(</span>
     <span class="n">time_left_for_this_task</span><span class="o">=</span><span class="mi">120</span><span class="p">,</span>
@@ -155,12 +156,17 @@ <h2>Build and fit a classifier<a class="headerlink" href="#build-and-fit-a-class
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>AutoSklearnClassifier(per_run_time_limit=30, time_left_for_this_task=120,
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:68: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+  self.metafeatures = self.metafeatures.append(metafeatures)
+/home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:72: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+  self.algorithm_runs[metric].append(runs)
+
+AutoSklearnClassifier(per_run_time_limit=30, time_left_for_this_task=120,
                       tmp_folder=&#39;/tmp/autosklearn_classification_example_tmp&#39;)
 </pre></div>
 </div>
-</section>
-<section id="view-the-models-found-by-auto-sklearn">
+</div>
+<div class="section" id="view-the-models-found-by-auto-sklearn">
 <h2>View the models found by auto-sklearn<a class="headerlink" href="#view-the-models-found-by-auto-sklearn" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="nb">print</span><span class="p">(</span><span class="n">automl</span><span class="o">.</span><span class="n">leaderboard</span><span class="p">())</span>
 </pre></div>
@@ -168,201 +174,397 @@ <h2>View the models found by auto-sklearn<a class="headerlink" href="#view-the-m
 <p class="sphx-glr-script-out">Out:</p>
 <div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>          rank  ensemble_weight                type      cost  duration
 model_id
-34           1             0.08         extra_trees  0.014184  1.673455
-7            2             0.06         extra_trees  0.014184  1.503079
-29           3             0.06         extra_trees  0.021277  1.718799
-16           4             0.02   gradient_boosting  0.021277  1.020501
-3            5             0.08                 mlp  0.028369  1.030776
-22           6             0.04   gradient_boosting  0.028369  1.163809
-26           7             0.02         extra_trees  0.028369  2.200900
-6            8             0.02                 mlp  0.028369  1.111869
-10           9             0.04       random_forest  0.028369  1.950961
-5           10             0.04       random_forest  0.035461  2.045974
-9           11             0.06         extra_trees  0.042553  1.824072
-27          12             0.04         extra_trees  0.042553  2.006197
-30          13             0.02       liblinear_svc  0.042553  0.949611
-32          14             0.12         extra_trees  0.049645  1.795890
-40          15             0.12          libsvm_svc  0.049645  0.821530
-33          16             0.02       random_forest  0.056738  1.952558
-20          17             0.02  passive_aggressive  0.078014  0.691132
-37          18             0.12          libsvm_svc  0.085106  0.968546
-35          19             0.02         extra_trees  0.085106  1.519158
+39           1             0.06         extra_trees  0.014184  1.416798
+7            2             0.04         extra_trees  0.014184  1.423290
+21           3             0.04         extra_trees  0.021277  1.313003
+2            4             0.04       random_forest  0.028369  1.564060
+32           5             0.04       liblinear_svc  0.028369  0.748310
+29           6             0.02         extra_trees  0.028369  1.389265
+27           7             0.04         extra_trees  0.028369  8.149227
+26           8             0.04         extra_trees  0.028369  2.153994
+13           9             0.02   gradient_boosting  0.028369  1.305902
+19          10             0.02         extra_trees  0.028369  2.652099
+11          11             0.04       random_forest  0.028369  1.933282
+6           12             0.08                 mlp  0.028369  0.973582
+10          13             0.08       random_forest  0.028369  1.716530
+5           14             0.04       random_forest  0.035461  1.797078
+31          15             0.02       liblinear_svc  0.035461  0.807887
+12          16             0.04   gradient_boosting  0.035461  1.177767
+8           17             0.02       random_forest  0.035461  1.783555
+18          18             0.04       random_forest  0.035461  1.805679
+17          19             0.02   gradient_boosting  0.035461  1.504512
+9           20             0.02         extra_trees  0.042553  1.689027
+15          21             0.02                 mlp  0.049645  3.394691
+23          22             0.04                 mlp  0.049645  1.976257
+4           23             0.04                 mlp  0.056738  1.186914
+28          24             0.02                 lda  0.063830  0.863830
+34          25             0.02       random_forest  0.063830  1.685210
+24          26             0.02       random_forest  0.070922  1.560245
+41          27             0.02       decision_tree  0.070922  0.741800
+37          28             0.02         extra_trees  0.085106  1.440288
+43          29             0.02         extra_trees  0.092199  1.604374
+42          30             0.02  passive_aggressive  0.120567  0.747741
 </pre></div>
 </div>
-</section>
-<section id="print-the-final-ensemble-constructed-by-auto-sklearn">
+</div>
+<div class="section" id="print-the-final-ensemble-constructed-by-auto-sklearn">
 <h2>Print the final ensemble constructed by auto-sklearn<a class="headerlink" href="#print-the-final-ensemble-constructed-by-auto-sklearn" title="Permalink to this headline">¶</a></h2>
-<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="nb">print</span><span class="p">(</span><span class="n">automl</span><span class="o">.</span><span class="n">show_models</span><span class="p">())</span>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">pprint</span><span class="p">(</span><span class="n">automl</span><span class="o">.</span><span class="n">show_models</span><span class="p">(),</span> <span class="n">indent</span><span class="o">=</span><span class="mi">4</span><span class="p">)</span>
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>[(0.120000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;extra_trees&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;feature_agglomeration&#39;, &#39;classifier:extra_trees:bootstrap&#39;: &#39;True&#39;, &#39;classifier:extra_trees:criterion&#39;: &#39;entropy&#39;, &#39;classifier:extra_trees:max_depth&#39;: &#39;None&#39;, &#39;classifier:extra_trees:max_features&#39;: 0.3871420537981852, &#39;classifier:extra_trees:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:extra_trees:min_impurity_decrease&#39;: 0.0, &#39;classifier:extra_trees:min_samples_leaf&#39;: 5, &#39;classifier:extra_trees:min_samples_split&#39;: 14, &#39;classifier:extra_trees:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;robust_scaler&#39;, &#39;feature_preprocessor:feature_agglomeration:affinity&#39;: &#39;manhattan&#39;, &#39;feature_preprocessor:feature_agglomeration:linkage&#39;: &#39;complete&#39;, &#39;feature_preprocessor:feature_agglomeration:n_clusters&#39;: 355, &#39;feature_preprocessor:feature_agglomeration:pooling_func&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max&#39;: 0.726182961545702, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min&#39;: 0.02260942170772907},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.120000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;libsvm_svc&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;feature_agglomeration&#39;, &#39;classifier:libsvm_svc:C&#39;: 0.07563955664363409, &#39;classifier:libsvm_svc:gamma&#39;: 1.9836355993007913, &#39;classifier:libsvm_svc:kernel&#39;: &#39;rbf&#39;, &#39;classifier:libsvm_svc:max_iter&#39;: -1, &#39;classifier:libsvm_svc:shrinking&#39;: &#39;True&#39;, &#39;classifier:libsvm_svc:tol&#39;: 0.005003173341372008, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;minmax&#39;, &#39;feature_preprocessor:feature_agglomeration:affinity&#39;: &#39;cosine&#39;, &#39;feature_preprocessor:feature_agglomeration:linkage&#39;: &#39;complete&#39;, &#39;feature_preprocessor:feature_agglomeration:n_clusters&#39;: 58, &#39;feature_preprocessor:feature_agglomeration:pooling_func&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.15890408760884822},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.120000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;libsvm_svc&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;select_percentile_classification&#39;, &#39;classifier:libsvm_svc:C&#39;: 5121.989374401993, &#39;classifier:libsvm_svc:gamma&#39;: 0.0024386428371511353, &#39;classifier:libsvm_svc:kernel&#39;: &#39;rbf&#39;, &#39;classifier:libsvm_svc:max_iter&#39;: -1, &#39;classifier:libsvm_svc:shrinking&#39;: &#39;True&#39;, &#39;classifier:libsvm_svc:tol&#39;: 0.0706216214952021, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;median&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;robust_scaler&#39;, &#39;feature_preprocessor:select_percentile_classification:percentile&#39;: 90.1749860017522, &#39;feature_preprocessor:select_percentile_classification:score_func&#39;: &#39;chi2&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.00017888891850981892, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max&#39;: 0.7393440007191528, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min&#39;: 0.04986510953238},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.080000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;extra_trees&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;polynomial&#39;, &#39;classifier:extra_trees:bootstrap&#39;: &#39;False&#39;, &#39;classifier:extra_trees:criterion&#39;: &#39;gini&#39;, &#39;classifier:extra_trees:max_depth&#39;: &#39;None&#39;, &#39;classifier:extra_trees:max_features&#39;: 0.4791448484072812, &#39;classifier:extra_trees:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:extra_trees:min_impurity_decrease&#39;: 0.0, &#39;classifier:extra_trees:min_samples_leaf&#39;: 1, &#39;classifier:extra_trees:min_samples_split&#39;: 6, &#39;classifier:extra_trees:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;none&#39;, &#39;feature_preprocessor:polynomial:degree&#39;: 3, &#39;feature_preprocessor:polynomial:include_bias&#39;: &#39;True&#39;, &#39;feature_preprocessor:polynomial:interaction_only&#39;: &#39;True&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.17184142431776436},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.080000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;mlp&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;feature_agglomeration&#39;, &#39;classifier:mlp:activation&#39;: &#39;tanh&#39;, &#39;classifier:mlp:alpha&#39;: 0.0001363185819149026, &#39;classifier:mlp:batch_size&#39;: &#39;auto&#39;, &#39;classifier:mlp:beta_1&#39;: 0.9, &#39;classifier:mlp:beta_2&#39;: 0.999, &#39;classifier:mlp:early_stopping&#39;: &#39;valid&#39;, &#39;classifier:mlp:epsilon&#39;: 1e-08, &#39;classifier:mlp:hidden_layer_depth&#39;: 3, &#39;classifier:mlp:learning_rate_init&#39;: 0.00018009776276177523, &#39;classifier:mlp:n_iter_no_change&#39;: 32, &#39;classifier:mlp:num_nodes_per_layer&#39;: 115, &#39;classifier:mlp:shuffle&#39;: &#39;True&#39;, &#39;classifier:mlp:solver&#39;: &#39;adam&#39;, &#39;classifier:mlp:tol&#39;: 0.0001, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;median&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;standardize&#39;, &#39;feature_preprocessor:feature_agglomeration:affinity&#39;: &#39;euclidean&#39;, &#39;feature_preprocessor:feature_agglomeration:linkage&#39;: &#39;ward&#39;, &#39;feature_preprocessor:feature_agglomeration:n_clusters&#39;: 182, &#39;feature_preprocessor:feature_agglomeration:pooling_func&#39;: &#39;mean&#39;, &#39;classifier:mlp:validation_fraction&#39;: 0.1},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.060000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;extra_trees&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;polynomial&#39;, &#39;classifier:extra_trees:bootstrap&#39;: &#39;False&#39;, &#39;classifier:extra_trees:criterion&#39;: &#39;gini&#39;, &#39;classifier:extra_trees:max_depth&#39;: &#39;None&#39;, &#39;classifier:extra_trees:max_features&#39;: 0.5707983257382487, &#39;classifier:extra_trees:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:extra_trees:min_impurity_decrease&#39;: 0.0, &#39;classifier:extra_trees:min_samples_leaf&#39;: 3, &#39;classifier:extra_trees:min_samples_split&#39;: 11, &#39;classifier:extra_trees:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;median&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;none&#39;, &#39;feature_preprocessor:polynomial:degree&#39;: 2, &#39;feature_preprocessor:polynomial:include_bias&#39;: &#39;False&#39;, &#39;feature_preprocessor:polynomial:interaction_only&#39;: &#39;False&#39;},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.060000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;extra_trees&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:extra_trees:bootstrap&#39;: &#39;False&#39;, &#39;classifier:extra_trees:criterion&#39;: &#39;gini&#39;, &#39;classifier:extra_trees:max_depth&#39;: &#39;None&#39;, &#39;classifier:extra_trees:max_features&#39;: 0.661377361743419, &#39;classifier:extra_trees:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:extra_trees:min_impurity_decrease&#39;: 0.0, &#39;classifier:extra_trees:min_samples_leaf&#39;: 2, &#39;classifier:extra_trees:min_samples_split&#39;: 13, &#39;classifier:extra_trees:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;none&#39;},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.060000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;extra_trees&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;extra_trees_preproc_for_classification&#39;, &#39;classifier:extra_trees:bootstrap&#39;: &#39;False&#39;, &#39;classifier:extra_trees:criterion&#39;: &#39;gini&#39;, &#39;classifier:extra_trees:max_depth&#39;: &#39;None&#39;, &#39;classifier:extra_trees:max_features&#39;: 0.9292309396985746, &#39;classifier:extra_trees:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:extra_trees:min_impurity_decrease&#39;: 0.0, &#39;classifier:extra_trees:min_samples_leaf&#39;: 1, &#39;classifier:extra_trees:min_samples_split&#39;: 10, &#39;classifier:extra_trees:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;median&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;normalize&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:bootstrap&#39;: &#39;False&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:criterion&#39;: &#39;entropy&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:max_depth&#39;: &#39;None&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:max_features&#39;: 0.9929881254946676, &#39;feature_preprocessor:extra_trees_preproc_for_classification:max_leaf_nodes&#39;: &#39;None&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_impurity_decrease&#39;: 0.0, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_samples_leaf&#39;: 1, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_samples_split&#39;: 2, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_weight_fraction_leaf&#39;: 0.0, &#39;feature_preprocessor:extra_trees_preproc_for_classification:n_estimators&#39;: 100, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.032719158639429445},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.040000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;random_forest&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;select_percentile_classification&#39;, &#39;classifier:random_forest:bootstrap&#39;: &#39;True&#39;, &#39;classifier:random_forest:criterion&#39;: &#39;entropy&#39;, &#39;classifier:random_forest:max_depth&#39;: &#39;None&#39;, &#39;classifier:random_forest:max_features&#39;: 0.49138075723513286, &#39;classifier:random_forest:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:random_forest:min_impurity_decrease&#39;: 0.0, &#39;classifier:random_forest:min_samples_leaf&#39;: 1, &#39;classifier:random_forest:min_samples_split&#39;: 6, &#39;classifier:random_forest:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;minmax&#39;, &#39;feature_preprocessor:select_percentile_classification:percentile&#39;: 56.97947373958566, &#39;feature_preprocessor:select_percentile_classification:score_func&#39;: &#39;mutual_info&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.42693600390988135},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.040000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;gradient_boosting&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;extra_trees_preproc_for_classification&#39;, &#39;classifier:gradient_boosting:early_stop&#39;: &#39;valid&#39;, &#39;classifier:gradient_boosting:l2_regularization&#39;: 8.057778875694463e-05, &#39;classifier:gradient_boosting:learning_rate&#39;: 0.09179220974965213, &#39;classifier:gradient_boosting:loss&#39;: &#39;auto&#39;, &#39;classifier:gradient_boosting:max_bins&#39;: 255, &#39;classifier:gradient_boosting:max_depth&#39;: &#39;None&#39;, &#39;classifier:gradient_boosting:max_leaf_nodes&#39;: 200, &#39;classifier:gradient_boosting:min_samples_leaf&#39;: 20, &#39;classifier:gradient_boosting:scoring&#39;: &#39;loss&#39;, &#39;classifier:gradient_boosting:tol&#39;: 1e-07, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;quantile_transformer&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:bootstrap&#39;: &#39;False&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:criterion&#39;: &#39;gini&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:max_depth&#39;: &#39;None&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:max_features&#39;: 0.9984367650965825, &#39;feature_preprocessor:extra_trees_preproc_for_classification:max_leaf_nodes&#39;: &#39;None&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_impurity_decrease&#39;: 0.0, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_samples_leaf&#39;: 13, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_samples_split&#39;: 18, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_weight_fraction_leaf&#39;: 0.0, &#39;feature_preprocessor:extra_trees_preproc_for_classification:n_estimators&#39;: 100, &#39;classifier:gradient_boosting:n_iter_no_change&#39;: 18, &#39;classifier:gradient_boosting:validation_fraction&#39;: 0.14295295806077554, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.008542188583124829, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles&#39;: 1102, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution&#39;: &#39;uniform&#39;},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.040000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;random_forest&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;extra_trees_preproc_for_classification&#39;, &#39;classifier:random_forest:bootstrap&#39;: &#39;True&#39;, &#39;classifier:random_forest:criterion&#39;: &#39;entropy&#39;, &#39;classifier:random_forest:max_depth&#39;: &#39;None&#39;, &#39;classifier:random_forest:max_features&#39;: 0.4775492074518431, &#39;classifier:random_forest:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:random_forest:min_impurity_decrease&#39;: 0.0, &#39;classifier:random_forest:min_samples_leaf&#39;: 2, &#39;classifier:random_forest:min_samples_split&#39;: 2, &#39;classifier:random_forest:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;robust_scaler&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:bootstrap&#39;: &#39;False&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:criterion&#39;: &#39;entropy&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:max_depth&#39;: &#39;None&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:max_features&#39;: 0.5662900693317384, &#39;feature_preprocessor:extra_trees_preproc_for_classification:max_leaf_nodes&#39;: &#39;None&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_impurity_decrease&#39;: 0.0, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_samples_leaf&#39;: 1, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_samples_split&#39;: 7, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_weight_fraction_leaf&#39;: 0.0, &#39;feature_preprocessor:extra_trees_preproc_for_classification:n_estimators&#39;: 100, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.015996368052062886, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max&#39;: 0.7845396961078424, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min&#39;: 0.25545052141264185},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.040000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;extra_trees&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;polynomial&#39;, &#39;classifier:extra_trees:bootstrap&#39;: &#39;True&#39;, &#39;classifier:extra_trees:criterion&#39;: &#39;entropy&#39;, &#39;classifier:extra_trees:max_depth&#39;: &#39;None&#39;, &#39;classifier:extra_trees:max_features&#39;: 0.3793256905143867, &#39;classifier:extra_trees:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:extra_trees:min_impurity_decrease&#39;: 0.0, &#39;classifier:extra_trees:min_samples_leaf&#39;: 20, &#39;classifier:extra_trees:min_samples_split&#39;: 5, &#39;classifier:extra_trees:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;none&#39;, &#39;feature_preprocessor:polynomial:degree&#39;: 3, &#39;feature_preprocessor:polynomial:include_bias&#39;: &#39;False&#39;, &#39;feature_preprocessor:polynomial:interaction_only&#39;: &#39;True&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.00010083678209087315},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.020000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;gradient_boosting&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;select_percentile_classification&#39;, &#39;classifier:gradient_boosting:early_stop&#39;: &#39;train&#39;, &#39;classifier:gradient_boosting:l2_regularization&#39;: 3.387912939529945e-10, &#39;classifier:gradient_boosting:learning_rate&#39;: 0.30755227194768237, &#39;classifier:gradient_boosting:loss&#39;: &#39;auto&#39;, &#39;classifier:gradient_boosting:max_bins&#39;: 255, &#39;classifier:gradient_boosting:max_depth&#39;: &#39;None&#39;, &#39;classifier:gradient_boosting:max_leaf_nodes&#39;: 60, &#39;classifier:gradient_boosting:min_samples_leaf&#39;: 39, &#39;classifier:gradient_boosting:scoring&#39;: &#39;loss&#39;, &#39;classifier:gradient_boosting:tol&#39;: 1e-07, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;none&#39;, &#39;feature_preprocessor:select_percentile_classification:percentile&#39;: 93.39844669585806, &#39;feature_preprocessor:select_percentile_classification:score_func&#39;: &#39;f_classif&#39;, &#39;classifier:gradient_boosting:n_iter_no_change&#39;: 18, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.010000000000000004},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.020000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;mlp&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;fast_ica&#39;, &#39;classifier:mlp:activation&#39;: &#39;relu&#39;, &#39;classifier:mlp:alpha&#39;: 0.0017940473175767063, &#39;classifier:mlp:batch_size&#39;: &#39;auto&#39;, &#39;classifier:mlp:beta_1&#39;: 0.9, &#39;classifier:mlp:beta_2&#39;: 0.999, &#39;classifier:mlp:early_stopping&#39;: &#39;valid&#39;, &#39;classifier:mlp:epsilon&#39;: 1e-08, &#39;classifier:mlp:hidden_layer_depth&#39;: 2, &#39;classifier:mlp:learning_rate_init&#39;: 0.0004684917334431039, &#39;classifier:mlp:n_iter_no_change&#39;: 32, &#39;classifier:mlp:num_nodes_per_layer&#39;: 101, &#39;classifier:mlp:shuffle&#39;: &#39;True&#39;, &#39;classifier:mlp:solver&#39;: &#39;adam&#39;, &#39;classifier:mlp:tol&#39;: 0.0001, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;standardize&#39;, &#39;feature_preprocessor:fast_ica:algorithm&#39;: &#39;parallel&#39;, &#39;feature_preprocessor:fast_ica:fun&#39;: &#39;exp&#39;, &#39;feature_preprocessor:fast_ica:whiten&#39;: &#39;False&#39;, &#39;classifier:mlp:validation_fraction&#39;: 0.1},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.020000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;extra_trees&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;polynomial&#39;, &#39;classifier:extra_trees:bootstrap&#39;: &#39;False&#39;, &#39;classifier:extra_trees:criterion&#39;: &#39;entropy&#39;, &#39;classifier:extra_trees:max_depth&#39;: &#39;None&#39;, &#39;classifier:extra_trees:max_features&#39;: 0.971348315581384, &#39;classifier:extra_trees:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:extra_trees:min_impurity_decrease&#39;: 0.0, &#39;classifier:extra_trees:min_samples_leaf&#39;: 2, &#39;classifier:extra_trees:min_samples_split&#39;: 19, &#39;classifier:extra_trees:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;normalize&#39;, &#39;feature_preprocessor:polynomial:degree&#39;: 2, &#39;feature_preprocessor:polynomial:include_bias&#39;: &#39;False&#39;, &#39;feature_preprocessor:polynomial:interaction_only&#39;: &#39;False&#39;},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.020000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;liblinear_svc&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;select_percentile_classification&#39;, &#39;classifier:liblinear_svc:C&#39;: 140.23434217313954, &#39;classifier:liblinear_svc:dual&#39;: &#39;False&#39;, &#39;classifier:liblinear_svc:fit_intercept&#39;: &#39;True&#39;, &#39;classifier:liblinear_svc:intercept_scaling&#39;: 1, &#39;classifier:liblinear_svc:loss&#39;: &#39;squared_hinge&#39;, &#39;classifier:liblinear_svc:multi_class&#39;: &#39;ovr&#39;, &#39;classifier:liblinear_svc:penalty&#39;: &#39;l2&#39;, &#39;classifier:liblinear_svc:tol&#39;: 0.0002944604673080368, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;quantile_transformer&#39;, &#39;feature_preprocessor:select_percentile_classification:percentile&#39;: 29.807816607300627, &#39;feature_preprocessor:select_percentile_classification:score_func&#39;: &#39;chi2&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.034330895903523305, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles&#39;: 1450, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution&#39;: &#39;uniform&#39;},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.020000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;random_forest&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;select_rates_classification&#39;, &#39;classifier:random_forest:bootstrap&#39;: &#39;True&#39;, &#39;classifier:random_forest:criterion&#39;: &#39;gini&#39;, &#39;classifier:random_forest:max_depth&#39;: &#39;None&#39;, &#39;classifier:random_forest:max_features&#39;: 0.5561389303694175, &#39;classifier:random_forest:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:random_forest:min_impurity_decrease&#39;: 0.0, &#39;classifier:random_forest:min_samples_leaf&#39;: 16, &#39;classifier:random_forest:min_samples_split&#39;: 4, &#39;classifier:random_forest:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;quantile_transformer&#39;, &#39;feature_preprocessor:select_rates_classification:alpha&#39;: 0.48507733455436836, &#39;feature_preprocessor:select_rates_classification:score_func&#39;: &#39;f_classif&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles&#39;: 1567, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution&#39;: &#39;normal&#39;, &#39;feature_preprocessor:select_rates_classification:mode&#39;: &#39;fpr&#39;},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.020000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;passive_aggressive&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:passive_aggressive:C&#39;: 0.14268277711454813, &#39;classifier:passive_aggressive:average&#39;: &#39;False&#39;, &#39;classifier:passive_aggressive:fit_intercept&#39;: &#39;True&#39;, &#39;classifier:passive_aggressive:loss&#39;: &#39;hinge&#39;, &#39;classifier:passive_aggressive:tol&#39;: 0.0002600768160857831, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;none&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.0070580904199417415},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.020000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;extra_trees&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;pca&#39;, &#39;classifier:extra_trees:bootstrap&#39;: &#39;False&#39;, &#39;classifier:extra_trees:criterion&#39;: &#39;gini&#39;, &#39;classifier:extra_trees:max_depth&#39;: &#39;None&#39;, &#39;classifier:extra_trees:max_features&#39;: 0.5137138356508442, &#39;classifier:extra_trees:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:extra_trees:min_impurity_decrease&#39;: 0.0, &#39;classifier:extra_trees:min_samples_leaf&#39;: 9, &#39;classifier:extra_trees:min_samples_split&#39;: 15, &#39;classifier:extra_trees:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;none&#39;, &#39;feature_preprocessor:pca:keep_variance&#39;: 0.9694955791292086, &#39;feature_preprocessor:pca:whiten&#39;: &#39;True&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.00014078218950356456},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-]
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>{   2: {   &#39;balancing&#39;: Balancing(random_state=1),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc790129ac0&gt;,
+           &#39;cost&#39;: 0.028368794326241176,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7903bd040&gt;,
+           &#39;ensemble_weight&#39;: 0.04,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc79023cd90&gt;,
+           &#39;model_id&#39;: 2,
+           &#39;rank&#39;: 4,
+           &#39;sklearn_classifier&#39;: RandomForestClassifier(max_features=5, n_estimators=512, n_jobs=1,
+                       random_state=1, warm_start=True)},
+    4: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78b6bc040&gt;,
+           &#39;cost&#39;: 0.05673758865248224,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78b85f6d0&gt;,
+           &#39;ensemble_weight&#39;: 0.04,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78b714ca0&gt;,
+           &#39;model_id&#39;: 4,
+           &#39;rank&#39;: 23,
+           &#39;sklearn_classifier&#39;: MLPClassifier(activation=&#39;tanh&#39;, alpha=0.00021148999718383549, beta_1=0.999,
+              beta_2=0.9, hidden_layer_sizes=(113, 113, 113),
+              learning_rate_init=0.0007452270241186694, max_iter=64,
+              n_iter_no_change=32, random_state=1, validation_fraction=0.0,
+              verbose=0, warm_start=True)},
+    5: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78c414be0&gt;,
+           &#39;cost&#39;: 0.03546099290780147,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78c864820&gt;,
+           &#39;ensemble_weight&#39;: 0.04,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c4ed730&gt;,
+           &#39;model_id&#39;: 5,
+           &#39;rank&#39;: 14,
+           &#39;sklearn_classifier&#39;: RandomForestClassifier(criterion=&#39;entropy&#39;, max_features=3, min_samples_leaf=2,
+                       n_estimators=512, n_jobs=1, random_state=1,
+                       warm_start=True)},
+    6: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc790040550&gt;,
+           &#39;cost&#39;: 0.028368794326241176,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc79015a340&gt;,
+           &#39;ensemble_weight&#39;: 0.08,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc7900401f0&gt;,
+           &#39;model_id&#39;: 6,
+           &#39;rank&#39;: 5,
+           &#39;sklearn_classifier&#39;: MLPClassifier(alpha=0.0017940473175767063, beta_1=0.999, beta_2=0.9,
+              early_stopping=True, hidden_layer_sizes=(101, 101),
+              learning_rate_init=0.0004684917334431039, max_iter=32,
+              n_iter_no_change=32, random_state=1, verbose=0, warm_start=True)},
+    7: {   &#39;balancing&#39;: Balancing(random_state=1),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc79068b250&gt;,
+           &#39;cost&#39;: 0.014184397163120588,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc793f99a30&gt;,
+           &#39;ensemble_weight&#39;: 0.04,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc79068b370&gt;,
+           &#39;model_id&#39;: 7,
+           &#39;rank&#39;: 1,
+           &#39;sklearn_classifier&#39;: ExtraTreesClassifier(max_features=34, min_samples_leaf=3, min_samples_split=11,
+                     n_estimators=512, n_jobs=1, random_state=1,
+                     warm_start=True)},
+    8: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78c1f30a0&gt;,
+           &#39;cost&#39;: 0.03546099290780147,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78c70c0a0&gt;,
+           &#39;ensemble_weight&#39;: 0.02,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c3b75e0&gt;,
+           &#39;model_id&#39;: 8,
+           &#39;rank&#39;: 15,
+           &#39;sklearn_classifier&#39;: RandomForestClassifier(max_features=2, min_samples_leaf=2, n_estimators=512,
+                       n_jobs=1, random_state=1, warm_start=True)},
+    9: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78b820eb0&gt;,
+           &#39;cost&#39;: 0.04255319148936165,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78be25d90&gt;,
+           &#39;ensemble_weight&#39;: 0.02,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78b9c7250&gt;,
+           &#39;model_id&#39;: 9,
+           &#39;rank&#39;: 20,
+           &#39;sklearn_classifier&#39;: ExtraTreesClassifier(max_features=9, min_samples_split=10, n_estimators=512,
+                     n_jobs=1, random_state=1, warm_start=True)},
+    10: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78fe73d60&gt;,
+            &#39;cost&#39;: 0.028368794326241176,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc79016ebe0&gt;,
+            &#39;ensemble_weight&#39;: 0.08,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78fe73970&gt;,
+            &#39;model_id&#39;: 10,
+            &#39;rank&#39;: 6,
+            &#39;sklearn_classifier&#39;: RandomForestClassifier(criterion=&#39;entropy&#39;, max_features=4, min_samples_split=6,
+                       n_estimators=512, n_jobs=1, random_state=1,
+                       warm_start=True)},
+    11: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78fd0ae80&gt;,
+            &#39;cost&#39;: 0.028368794326241176,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7901eaf70&gt;,
+            &#39;ensemble_weight&#39;: 0.04,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78fd0ac70&gt;,
+            &#39;model_id&#39;: 11,
+            &#39;rank&#39;: 7,
+            &#39;sklearn_classifier&#39;: RandomForestClassifier(criterion=&#39;entropy&#39;, max_features=23, min_samples_leaf=7,
+                       n_estimators=512, n_jobs=1, random_state=1,
+                       warm_start=True)},
+    12: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78c1f3c70&gt;,
+            &#39;cost&#39;: 0.03546099290780147,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78c5fa5b0&gt;,
+            &#39;ensemble_weight&#39;: 0.04,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c1f37f0&gt;,
+            &#39;model_id&#39;: 12,
+            &#39;rank&#39;: 16,
+            &#39;sklearn_classifier&#39;: HistGradientBoostingClassifier(early_stopping=True,
+                               l2_regularization=0.005326508887463406,
+                               learning_rate=0.060800813211425456, max_iter=512,
+                               max_leaf_nodes=6, min_samples_leaf=5,
+                               n_iter_no_change=5, random_state=1,
+                               validation_fraction=None, warm_start=True)},
+    13: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78fa69a30&gt;,
+            &#39;cost&#39;: 0.028368794326241176,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78feca6a0&gt;,
+            &#39;ensemble_weight&#39;: 0.02,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78fa69730&gt;,
+            &#39;model_id&#39;: 13,
+            &#39;rank&#39;: 8,
+            &#39;sklearn_classifier&#39;: HistGradientBoostingClassifier(early_stopping=False,
+                               l2_regularization=1.0647401999412075e-10,
+                               learning_rate=0.08291320147381159, max_iter=512,
+                               max_leaf_nodes=39, n_iter_no_change=0,
+                               random_state=1, validation_fraction=None,
+                               warm_start=True)},
+    15: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78b836a90&gt;,
+            &#39;cost&#39;: 0.049645390070921946,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78ba74a90&gt;,
+            &#39;ensemble_weight&#39;: 0.02,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78b836370&gt;,
+            &#39;model_id&#39;: 15,
+            &#39;rank&#39;: 21,
+            &#39;sklearn_classifier&#39;: MLPClassifier(alpha=4.2841884333778574e-06, beta_1=0.999, beta_2=0.9,
+              hidden_layer_sizes=(263, 263, 263),
+              learning_rate_init=0.0011804284312897009, max_iter=128,
+              n_iter_no_change=32, random_state=1, validation_fraction=0.0,
+              verbose=0, warm_start=True)},
+    17: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78be46b20&gt;,
+            &#39;cost&#39;: 0.03546099290780147,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78c917c10&gt;,
+            &#39;ensemble_weight&#39;: 0.02,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78bc74550&gt;,
+            &#39;model_id&#39;: 17,
+            &#39;rank&#39;: 17,
+            &#39;sklearn_classifier&#39;: HistGradientBoostingClassifier(early_stopping=True,
+                               l2_regularization=0.4635442279519353,
+                               learning_rate=0.09809681787962342, max_iter=512,
+                               max_leaf_nodes=328, min_samples_leaf=2,
+                               n_iter_no_change=2, random_state=1,
+                               validation_fraction=None, warm_start=True)},
+    18: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78ba1bee0&gt;,
+            &#39;cost&#39;: 0.03546099290780147,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78c345ee0&gt;,
+            &#39;ensemble_weight&#39;: 0.04,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78be46c10&gt;,
+            &#39;model_id&#39;: 18,
+            &#39;rank&#39;: 18,
+            &#39;sklearn_classifier&#39;: RandomForestClassifier(criterion=&#39;entropy&#39;, max_features=3, n_estimators=512,
+                       n_jobs=1, random_state=1, warm_start=True)},
+    19: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78cb3ca30&gt;,
+            &#39;cost&#39;: 0.028368794326241176,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78fe1ea00&gt;,
+            &#39;ensemble_weight&#39;: 0.02,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78cb3c880&gt;,
+            &#39;model_id&#39;: 19,
+            &#39;rank&#39;: 9,
+            &#39;sklearn_classifier&#39;: ExtraTreesClassifier(criterion=&#39;entropy&#39;, max_features=448, min_samples_leaf=2,
+                     min_samples_split=20, n_estimators=512, n_jobs=1,
+                     random_state=1, warm_start=True)},
+    21: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc790288640&gt;,
+            &#39;cost&#39;: 0.021276595744680882,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7902d5a30&gt;,
+            &#39;ensemble_weight&#39;: 0.04,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc790279550&gt;,
+            &#39;model_id&#39;: 21,
+            &#39;rank&#39;: 3,
+            &#39;sklearn_classifier&#39;: ExtraTreesClassifier(criterion=&#39;entropy&#39;, max_features=4, min_samples_leaf=2,
+                     min_samples_split=15, n_estimators=512, n_jobs=1,
+                     random_state=1, warm_start=True)},
+    23: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78b75a580&gt;,
+            &#39;cost&#39;: 0.049645390070921946,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78ba3ddf0&gt;,
+            &#39;ensemble_weight&#39;: 0.04,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78b75a130&gt;,
+            &#39;model_id&#39;: 23,
+            &#39;rank&#39;: 22,
+            &#39;sklearn_classifier&#39;: MLPClassifier(alpha=0.02847755502162456, beta_1=0.999, beta_2=0.9,
+              hidden_layer_sizes=(123, 123),
+              learning_rate_init=0.000421568792103947, max_iter=256,
+              n_iter_no_change=32, random_state=1, validation_fraction=0.0,
+              verbose=0, warm_start=True)},
+    24: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78b44ee50&gt;,
+            &#39;cost&#39;: 0.07092198581560283,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78b58aa90&gt;,
+            &#39;ensemble_weight&#39;: 0.02,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc790288bb0&gt;,
+            &#39;model_id&#39;: 24,
+            &#39;rank&#39;: 26,
+            &#39;sklearn_classifier&#39;: RandomForestClassifier(criterion=&#39;entropy&#39;, max_features=16, n_estimators=512,
+                       n_jobs=1, random_state=1, warm_start=True)},
+    26: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78ca98610&gt;,
+            &#39;cost&#39;: 0.028368794326241176,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78fbb9b50&gt;,
+            &#39;ensemble_weight&#39;: 0.04,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78ca98190&gt;,
+            &#39;model_id&#39;: 26,
+            &#39;rank&#39;: 10,
+            &#39;sklearn_classifier&#39;: ExtraTreesClassifier(criterion=&#39;entropy&#39;, max_features=414, min_samples_leaf=2,
+                     min_samples_split=19, n_estimators=512, n_jobs=1,
+                     random_state=1, warm_start=True)},
+    27: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78c944760&gt;,
+            &#39;cost&#39;: 0.028368794326241176,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78cb34910&gt;,
+            &#39;ensemble_weight&#39;: 0.04,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c93bdc0&gt;,
+            &#39;model_id&#39;: 27,
+            &#39;rank&#39;: 11,
+            &#39;sklearn_classifier&#39;: ExtraTreesClassifier(bootstrap=True, criterion=&#39;entropy&#39;, max_features=4126,
+                     min_samples_leaf=7, min_samples_split=13, n_estimators=512,
+                     n_jobs=1, random_state=1, warm_start=True)},
+    28: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78b545430&gt;,
+            &#39;cost&#39;: 0.06382978723404253,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78b79d940&gt;,
+            &#39;ensemble_weight&#39;: 0.02,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78b5451f0&gt;,
+            &#39;model_id&#39;: 28,
+            &#39;rank&#39;: 24,
+            &#39;sklearn_classifier&#39;: LinearDiscriminantAnalysis(tol=0.032840833359764676)},
+    29: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78c883580&gt;,
+            &#39;cost&#39;: 0.028368794326241176,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78cae7460&gt;,
+            &#39;ensemble_weight&#39;: 0.02,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c883070&gt;,
+            &#39;model_id&#39;: 29,
+            &#39;rank&#39;: 12,
+            &#39;sklearn_classifier&#39;: ExtraTreesClassifier(criterion=&#39;entropy&#39;, max_features=8, min_samples_leaf=11,
+                     min_samples_split=5, n_estimators=512, n_jobs=1,
+                     random_state=1, warm_start=True)},
+    31: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78ba251f0&gt;,
+            &#39;cost&#39;: 0.03546099290780147,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78bd04670&gt;,
+            &#39;ensemble_weight&#39;: 0.02,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78ba1bdc0&gt;,
+            &#39;model_id&#39;: 31,
+            &#39;rank&#39;: 19,
+            &#39;sklearn_classifier&#39;: LinearSVC(class_weight=&#39;balanced&#39;, dual=False, intercept_scaling=1.0,
+          random_state=1, tol=0.0001350687148423338)},
+    32: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78c667580&gt;,
+            &#39;cost&#39;: 0.028368794326241176,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78ca8e9a0&gt;,
+            &#39;ensemble_weight&#39;: 0.04,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c6cef10&gt;,
+            &#39;model_id&#39;: 32,
+            &#39;rank&#39;: 13,
+            &#39;sklearn_classifier&#39;: LinearSVC(C=832.5744883902069, dual=False, intercept_scaling=1.0,
+          random_state=1, tol=0.015340961719445064)},
+    34: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78f7b95e0&gt;,
+            &#39;cost&#39;: 0.06382978723404253,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78b753850&gt;,
+            &#39;ensemble_weight&#39;: 0.02,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c3e6490&gt;,
+            &#39;model_id&#39;: 34,
+            &#39;rank&#39;: 25,
+            &#39;sklearn_classifier&#39;: RandomForestClassifier(bootstrap=False, criterion=&#39;entropy&#39;, max_features=54,
+                       min_samples_leaf=14, min_samples_split=3,
+                       n_estimators=512, n_jobs=1, random_state=1,
+                       warm_start=True)},
+    37: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78acf7d00&gt;,
+            &#39;cost&#39;: 0.08510638297872342,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc790382190&gt;,
+            &#39;ensemble_weight&#39;: 0.02,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78ad54a00&gt;,
+            &#39;model_id&#39;: 37,
+            &#39;rank&#39;: 28,
+            &#39;sklearn_classifier&#39;: ExtraTreesClassifier(criterion=&#39;entropy&#39;, max_features=1, min_samples_leaf=10,
+                     min_samples_split=18, n_estimators=512, n_jobs=1,
+                     random_state=1, warm_start=True)},
+    39: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc790362c40&gt;,
+            &#39;cost&#39;: 0.014184397163120588,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7902d58b0&gt;,
+            &#39;ensemble_weight&#39;: 0.06,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc7901c2250&gt;,
+            &#39;model_id&#39;: 39,
+            &#39;rank&#39;: 2,
+            &#39;sklearn_classifier&#39;: ExtraTreesClassifier(criterion=&#39;entropy&#39;, max_features=32, min_samples_leaf=8,
+                     min_samples_split=8, n_estimators=512, n_jobs=1,
+                     random_state=1, warm_start=True)},
+    41: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78adce8b0&gt;,
+            &#39;cost&#39;: 0.07092198581560283,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78bff5f70&gt;,
+            &#39;ensemble_weight&#39;: 0.02,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78adc2d30&gt;,
+            &#39;model_id&#39;: 41,
+            &#39;rank&#39;: 27,
+            &#39;sklearn_classifier&#39;: DecisionTreeClassifier(max_depth=1, min_samples_leaf=15, min_samples_split=5,
+                       random_state=1)},
+    42: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78a8d7a90&gt;,
+            &#39;cost&#39;: 0.12056737588652477,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78ad91f70&gt;,
+            &#39;ensemble_weight&#39;: 0.02,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78a8d7790&gt;,
+            &#39;model_id&#39;: 42,
+            &#39;rank&#39;: 30,
+            &#39;sklearn_classifier&#39;: PassiveAggressiveClassifier(C=8.738643048253599e-05, average=True,
+                            loss=&#39;squared_hinge&#39;, max_iter=32, random_state=1,
+                            tol=0.0010303157708058924, warm_start=True)},
+    43: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78ab3f340&gt;,
+            &#39;cost&#39;: 0.09219858156028371,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78adfc730&gt;,
+            &#39;ensemble_weight&#39;: 0.02,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78aba4cd0&gt;,
+            &#39;model_id&#39;: 43,
+            &#39;rank&#39;: 29,
+            &#39;sklearn_classifier&#39;: ExtraTreesClassifier(bootstrap=True, criterion=&#39;entropy&#39;, max_features=12,
+                     min_samples_leaf=19, min_samples_split=17,
+                     n_estimators=512, n_jobs=1, random_state=1,
+                     warm_start=True)}}
 </pre></div>
 </div>
-</section>
-<section id="get-the-score-of-the-final-ensemble">
+</div>
+<div class="section" id="get-the-score-of-the-final-ensemble">
 <h2>Get the Score of the final ensemble<a class="headerlink" href="#get-the-score-of-the-final-ensemble" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">predictions</span> <span class="o">=</span> <span class="n">automl</span><span class="o">.</span><span class="n">predict</span><span class="p">(</span><span class="n">X_test</span><span class="p">)</span>
 <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Accuracy score:&quot;</span><span class="p">,</span> <span class="n">sklearn</span><span class="o">.</span><span class="n">metrics</span><span class="o">.</span><span class="n">accuracy_score</span><span class="p">(</span><span class="n">y_test</span><span class="p">,</span> <span class="n">predictions</span><span class="p">))</span>
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Accuracy score: 0.9440559440559441
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Accuracy score: 0.958041958041958
 </pre></div>
 </div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes  0.650 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes  1.747 seconds)</p>
 <div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-examples-20-basic-example-classification-py">
 <div class="binder-badge docutils container">
 <a class="reference external image-reference" href="https://mybinder.org/v2/gh/automl/auto-sklearn/master?urlpath=lab/tree/notebooks/examples/20_basic/example_classification.ipynb"><img alt="Launch binder" src="../../_images/binder_badge_logo.svg" width="150px" /></a>
@@ -375,8 +577,8 @@ <h2>Get the Score of the final ensemble<a class="headerlink" href="#get-the-scor
 </div>
 </div>
 <p class="sphx-glr-signature"><a class="reference external" href="https://sphinx-gallery.github.io">Gallery generated by Sphinx-Gallery</a></p>
-</section>
-</section>
+</div>
+</div>
 
 
     </div>
@@ -397,7 +599,7 @@ <h2>Get the Score of the final ensemble<a class="headerlink" href="#get-the-scor
       
     </p>
     <p>
-        &copy; Copyright 2014-2021, Machine Learning Professorship Freiburg.<br/>
+        &copy; Copyright 2014-2022, Machine Learning Professorship Freiburg.<br/>
       Created using <a href="http://sphinx-doc.org/">Sphinx</a> 4.2.0.<br/>
     </p>
   </div>
diff --git a/master/examples/20_basic/example_multilabel_classification.html b/master/examples/20_basic/example_multilabel_classification.html
index ebf93caaef..00d10309cc 100644
--- a/master/examples/20_basic/example_multilabel_classification.html
+++ b/master/examples/20_basic/example_multilabel_classification.html
@@ -3,9 +3,8 @@
 <html>
   <head>
     <meta charset="utf-8" />
-    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="generator" content="Docutils 0.17.1: http://docutils.sourceforge.net/" />
-
-    <title>Multi-label Classification &#8212; AutoSklearn 0.14.2 documentation</title>
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>Multi-label Classification &#8212; AutoSklearn 0.14.4 documentation</title>
     <link rel="stylesheet" type="text/css" href="../../_static/pygments.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/bootstrap-sphinx.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/sg_gallery.css" />
@@ -55,7 +54,7 @@
         </button>
         <a class="navbar-brand" href="../../index.html">
           auto-sklearn</a>
-        <span class="navbar-text navbar-version pull-left"><b>0.14.2</b></span>
+        <span class="navbar-text navbar-version pull-left"><b>0.14.4</b></span>
       </div>
 
         <div class="collapse navbar-collapse nav-collapse">
@@ -127,12 +126,13 @@
 <p>Click <a class="reference internal" href="#sphx-glr-download-examples-20-basic-example-multilabel-classification-py"><span class="std std-ref">here</span></a>
 to download the full example code or to run this example in your browser via Binder</p>
 </div>
-<section class="sphx-glr-example-title" id="multi-label-classification">
+<div class="sphx-glr-example-title section" id="multi-label-classification">
 <span id="sphx-glr-examples-20-basic-example-multilabel-classification-py"></span><h1>Multi-label Classification<a class="headerlink" href="#multi-label-classification" title="Permalink to this headline">¶</a></h1>
 <p>This examples shows how to format the targets for a multilabel classification
 problem. Details on multilabel classification can be found
 <a class="reference external" href="https://scikit-learn.org/stable/modules/multiclass.html">here</a>.</p>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
+<span class="kn">from</span> <span class="nn">pprint</span> <span class="kn">import</span> <span class="n">pprint</span>
 
 <span class="kn">import</span> <span class="nn">sklearn.datasets</span>
 <span class="kn">import</span> <span class="nn">sklearn.metrics</span>
@@ -141,7 +141,7 @@
 <span class="kn">import</span> <span class="nn">autosklearn.classification</span>
 </pre></div>
 </div>
-<section id="data-loading">
+<div class="section" id="data-loading">
 <h2>Data Loading<a class="headerlink" href="#data-loading" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="c1"># Using reuters multilabel dataset -- https://www.openml.org/d/40594</span>
 <span class="n">X</span><span class="p">,</span> <span class="n">y</span> <span class="o">=</span> <span class="n">sklearn</span><span class="o">.</span><span class="n">datasets</span><span class="o">.</span><span class="n">fetch_openml</span><span class="p">(</span><span class="n">data_id</span><span class="o">=</span><span class="mi">40594</span><span class="p">,</span> <span class="n">return_X_y</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span> <span class="n">as_frame</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span>
@@ -153,7 +153,7 @@ <h2>Data Loading<a class="headerlink" href="#data-loading" title="Permalink to t
 <span class="c1"># More information on: https://scikit-learn.org/stable/modules/multiclass.html</span>
 <span class="n">y</span><span class="p">[</span><span class="n">y</span> <span class="o">==</span> <span class="s1">&#39;TRUE&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="mi">1</span>
 <span class="n">y</span><span class="p">[</span><span class="n">y</span> <span class="o">==</span> <span class="s1">&#39;FALSE&#39;</span><span class="p">]</span> <span class="o">=</span> <span class="mi">0</span>
-<span class="n">y</span> <span class="o">=</span> <span class="n">y</span><span class="o">.</span><span class="n">astype</span><span class="p">(</span><span class="n">np</span><span class="o">.</span><span class="n">int</span><span class="p">)</span>
+<span class="n">y</span> <span class="o">=</span> <span class="n">y</span><span class="o">.</span><span class="n">astype</span><span class="p">(</span><span class="nb">int</span><span class="p">)</span>
 
 <span class="c1"># Using type of target is a good way to make sure your data</span>
 <span class="c1"># is properly formatted</span>
@@ -165,14 +165,11 @@ <h2>Data Loading<a class="headerlink" href="#data-loading" title="Permalink to t
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/home/runner/work/auto-sklearn/auto-sklearn/examples/20_basic/example_multilabel_classification.py:33: DeprecationWarning: `np.int` is a deprecated alias for the builtin `int`. To silence this warning, use `int` by itself. Doing this will not modify any behavior and is safe. When replacing `np.int`, you may wish to use e.g. `np.int64` or `np.int32` to specify the precision. If you wish to review your current use, check the release note link for additional information.
-Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations
-  y = y.astype(np.int)
-type_of_target=multilabel-indicator
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>type_of_target=multilabel-indicator
 </pre></div>
 </div>
-</section>
-<section id="building-the-classifier">
+</div>
+<div class="section" id="building-the-classifier">
 <h2>Building the classifier<a class="headerlink" href="#building-the-classifier" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">automl</span> <span class="o">=</span> <span class="n">autosklearn</span><span class="o">.</span><span class="n">classification</span><span class="o">.</span><span class="n">AutoSklearnClassifier</span><span class="p">(</span>
     <span class="n">time_left_for_this_task</span><span class="o">=</span><span class="mi">60</span><span class="p">,</span>
@@ -192,8 +189,8 @@ <h2>Building the classifier<a class="headerlink" href="#building-the-classifier"
                       time_left_for_this_task=60)
 </pre></div>
 </div>
-</section>
-<section id="view-the-models-found-by-auto-sklearn">
+</div>
+<div class="section" id="view-the-models-found-by-auto-sklearn">
 <h2>View the models found by auto-sklearn<a class="headerlink" href="#view-the-models-found-by-auto-sklearn" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="nb">print</span><span class="p">(</span><span class="n">automl</span><span class="o">.</span><span class="n">leaderboard</span><span class="p">())</span>
 </pre></div>
@@ -201,29 +198,30 @@ <h2>View the models found by auto-sklearn<a class="headerlink" href="#view-the-m
 <p class="sphx-glr-script-out">Out:</p>
 <div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>          rank  ensemble_weight           type      cost  duration
 model_id
-2            1              1.0  random_forest  0.447294  3.636021
+2            1              1.0  random_forest  0.447294  3.264884
 </pre></div>
 </div>
-</section>
-<section id="print-the-final-ensemble-constructed-by-auto-sklearn">
+</div>
+<div class="section" id="print-the-final-ensemble-constructed-by-auto-sklearn">
 <h2>Print the final ensemble constructed by auto-sklearn<a class="headerlink" href="#print-the-final-ensemble-constructed-by-auto-sklearn" title="Permalink to this headline">¶</a></h2>
-<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="nb">print</span><span class="p">(</span><span class="n">automl</span><span class="o">.</span><span class="n">show_models</span><span class="p">())</span>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">pprint</span><span class="p">(</span><span class="n">automl</span><span class="o">.</span><span class="n">show_models</span><span class="p">(),</span> <span class="n">indent</span><span class="o">=</span><span class="mi">4</span><span class="p">)</span>
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>[(1.000000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;random_forest&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:random_forest:bootstrap&#39;: &#39;True&#39;, &#39;classifier:random_forest:criterion&#39;: &#39;gini&#39;, &#39;classifier:random_forest:max_depth&#39;: &#39;None&#39;, &#39;classifier:random_forest:max_features&#39;: 0.5, &#39;classifier:random_forest:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:random_forest:min_impurity_decrease&#39;: 0.0, &#39;classifier:random_forest:min_samples_leaf&#39;: 1, &#39;classifier:random_forest:min_samples_split&#39;: 2, &#39;classifier:random_forest:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;standardize&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.01},
-dataset_properties={
-  &#39;task&#39;: 3,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: True,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-]
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>{   2: {   &#39;balancing&#39;: Balancing(random_state=1),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78ca21f10&gt;,
+           &#39;cost&#39;: 0.4472941828699525,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78a8d7610&gt;,
+           &#39;ensemble_weight&#39;: 1.0,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78ca21670&gt;,
+           &#39;model_id&#39;: 2,
+           &#39;rank&#39;: 1,
+           &#39;sklearn_classifier&#39;: RandomForestClassifier(max_features=15, n_estimators=512, n_jobs=1,
+                       random_state=1, warm_start=True)}}
 </pre></div>
 </div>
-</section>
-<section id="print-statistics-about-the-auto-sklearn-run">
+</div>
+<div class="section" id="print-statistics-about-the-auto-sklearn-run">
 <h2>Print statistics about the auto-sklearn run<a class="headerlink" href="#print-statistics-about-the-auto-sklearn-run" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="c1"># Print statistics about the auto-sklearn run such as number of</span>
 <span class="c1"># iterations, number of models failed with a time out.</span>
@@ -242,8 +240,8 @@ <h2>Print statistics about the auto-sklearn run<a class="headerlink" href="#prin
   Number of target algorithms that exceeded the memory limit: 0
 </pre></div>
 </div>
-</section>
-<section id="get-the-score-of-the-final-ensemble">
+</div>
+<div class="section" id="get-the-score-of-the-final-ensemble">
 <h2>Get the Score of the final ensemble<a class="headerlink" href="#get-the-score-of-the-final-ensemble" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">predictions</span> <span class="o">=</span> <span class="n">automl</span><span class="o">.</span><span class="n">predict</span><span class="p">(</span><span class="n">X_test</span><span class="p">)</span>
 <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Accuracy score&quot;</span><span class="p">,</span> <span class="n">sklearn</span><span class="o">.</span><span class="n">metrics</span><span class="o">.</span><span class="n">accuracy_score</span><span class="p">(</span><span class="n">y_test</span><span class="p">,</span> <span class="n">predictions</span><span class="p">))</span>
@@ -253,7 +251,7 @@ <h2>Get the Score of the final ensemble<a class="headerlink" href="#get-the-scor
 <div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Accuracy score 0.604
 </pre></div>
 </div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 0 minutes  15.973 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 0 minutes  12.175 seconds)</p>
 <div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-examples-20-basic-example-multilabel-classification-py">
 <div class="binder-badge docutils container">
 <a class="reference external image-reference" href="https://mybinder.org/v2/gh/automl/auto-sklearn/master?urlpath=lab/tree/notebooks/examples/20_basic/example_multilabel_classification.ipynb"><img alt="Launch binder" src="../../_images/binder_badge_logo.svg" width="150px" /></a>
@@ -266,8 +264,8 @@ <h2>Get the Score of the final ensemble<a class="headerlink" href="#get-the-scor
 </div>
 </div>
 <p class="sphx-glr-signature"><a class="reference external" href="https://sphinx-gallery.github.io">Gallery generated by Sphinx-Gallery</a></p>
-</section>
-</section>
+</div>
+</div>
 
 
     </div>
@@ -288,7 +286,7 @@ <h2>Get the Score of the final ensemble<a class="headerlink" href="#get-the-scor
       
     </p>
     <p>
-        &copy; Copyright 2014-2021, Machine Learning Professorship Freiburg.<br/>
+        &copy; Copyright 2014-2022, Machine Learning Professorship Freiburg.<br/>
       Created using <a href="http://sphinx-doc.org/">Sphinx</a> 4.2.0.<br/>
     </p>
   </div>
diff --git a/master/examples/20_basic/example_multioutput_regression.html b/master/examples/20_basic/example_multioutput_regression.html
index 3f6aad5f09..76cf4bbe68 100644
--- a/master/examples/20_basic/example_multioutput_regression.html
+++ b/master/examples/20_basic/example_multioutput_regression.html
@@ -3,9 +3,8 @@
 <html>
   <head>
     <meta charset="utf-8" />
-    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="generator" content="Docutils 0.17.1: http://docutils.sourceforge.net/" />
-
-    <title>Multi-output Regression &#8212; AutoSklearn 0.14.2 documentation</title>
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>Multi-output Regression &#8212; AutoSklearn 0.14.4 documentation</title>
     <link rel="stylesheet" type="text/css" href="../../_static/pygments.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/bootstrap-sphinx.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/sg_gallery.css" />
@@ -55,7 +54,7 @@
         </button>
         <a class="navbar-brand" href="../../index.html">
           auto-sklearn</a>
-        <span class="navbar-text navbar-version pull-left"><b>0.14.2</b></span>
+        <span class="navbar-text navbar-version pull-left"><b>0.14.4</b></span>
       </div>
 
         <div class="collapse navbar-collapse nav-collapse">
@@ -127,11 +126,12 @@
 <p>Click <a class="reference internal" href="#sphx-glr-download-examples-20-basic-example-multioutput-regression-py"><span class="std std-ref">here</span></a>
 to download the full example code or to run this example in your browser via Binder</p>
 </div>
-<section class="sphx-glr-example-title" id="multi-output-regression">
+<div class="sphx-glr-example-title section" id="multi-output-regression">
 <span id="sphx-glr-examples-20-basic-example-multioutput-regression-py"></span><h1>Multi-output Regression<a class="headerlink" href="#multi-output-regression" title="Permalink to this headline">¶</a></h1>
 <p>The following example shows how to fit a multioutput regression model with
 <em>auto-sklearn</em>.</p>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">numpy</span>
+<span class="kn">from</span> <span class="nn">pprint</span> <span class="kn">import</span> <span class="n">pprint</span>
 
 <span class="kn">from</span> <span class="nn">sklearn.datasets</span> <span class="kn">import</span> <span class="n">make_regression</span>
 <span class="kn">from</span> <span class="nn">sklearn.metrics</span> <span class="kn">import</span> <span class="n">r2_score</span>
@@ -140,15 +140,15 @@
 <span class="kn">from</span> <span class="nn">autosklearn.regression</span> <span class="kn">import</span> <span class="n">AutoSklearnRegressor</span>
 </pre></div>
 </div>
-<section id="data-loading">
+<div class="section" id="data-loading">
 <h2>Data Loading<a class="headerlink" href="#data-loading" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">X</span><span class="p">,</span> <span class="n">y</span> <span class="o">=</span> <span class="n">make_regression</span><span class="p">(</span><span class="n">n_samples</span><span class="o">=</span><span class="mi">1000</span><span class="p">,</span> <span class="n">n_features</span><span class="o">=</span><span class="mi">10</span><span class="p">,</span> <span class="n">n_informative</span><span class="o">=</span><span class="mi">5</span><span class="p">,</span> <span class="n">n_targets</span><span class="o">=</span><span class="mi">3</span><span class="p">)</span>
 
 <span class="n">X_train</span><span class="p">,</span> <span class="n">X_test</span><span class="p">,</span> <span class="n">y_train</span><span class="p">,</span> <span class="n">y_test</span> <span class="o">=</span> <span class="n">train_test_split</span><span class="p">(</span><span class="n">X</span><span class="p">,</span> <span class="n">y</span><span class="p">,</span> <span class="n">random_state</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
 </pre></div>
 </div>
-</section>
-<section id="build-and-fit-a-regressor">
+</div>
+<div class="section" id="build-and-fit-a-regressor">
 <h2>Build and fit a regressor<a class="headerlink" href="#build-and-fit-a-regressor" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">automl</span> <span class="o">=</span> <span class="n">AutoSklearnRegressor</span><span class="p">(</span>
     <span class="n">time_left_for_this_task</span><span class="o">=</span><span class="mi">120</span><span class="p">,</span>
@@ -163,64 +163,72 @@ <h2>Build and fit a regressor<a class="headerlink" href="#build-and-fit-a-regres
                      tmp_folder=&#39;/tmp/autosklearn_multioutput_regression_example_tmp&#39;)
 </pre></div>
 </div>
-</section>
-<section id="view-the-models-found-by-auto-sklearn">
+</div>
+<div class="section" id="view-the-models-found-by-auto-sklearn">
 <h2>View the models found by auto-sklearn<a class="headerlink" href="#view-the-models-found-by-auto-sklearn" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="nb">print</span><span class="p">(</span><span class="n">automl</span><span class="o">.</span><span class="n">leaderboard</span><span class="p">())</span>
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>          rank  ensemble_weight                 type      cost  duration
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>          rank  ensemble_weight           type      cost  duration
 model_id
-2            1             0.62        random_forest  0.157218  2.282688
-5            2             0.18  k_nearest_neighbors  0.207332  0.459119
-25           3             0.20  k_nearest_neighbors  0.318861  0.454935
+22           1             0.32    extra_trees  0.136510  5.840576
+2            2             0.42  random_forest  0.142814  2.232426
+21           3             0.26  random_forest  0.224804  1.676522
 </pre></div>
 </div>
-</section>
-<section id="print-the-final-ensemble-constructed-by-auto-sklearn">
+</div>
+<div class="section" id="print-the-final-ensemble-constructed-by-auto-sklearn">
 <h2>Print the final ensemble constructed by auto-sklearn<a class="headerlink" href="#print-the-final-ensemble-constructed-by-auto-sklearn" title="Permalink to this headline">¶</a></h2>
-<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="nb">print</span><span class="p">(</span><span class="n">automl</span><span class="o">.</span><span class="n">show_models</span><span class="p">())</span>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">pprint</span><span class="p">(</span><span class="n">automl</span><span class="o">.</span><span class="n">show_models</span><span class="p">(),</span> <span class="n">indent</span><span class="o">=</span><span class="mi">4</span><span class="p">)</span>
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>[(0.620000, SimpleRegressionPipeline({&#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;regressor:__choice__&#39;: &#39;random_forest&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;standardize&#39;, &#39;regressor:random_forest:bootstrap&#39;: &#39;True&#39;, &#39;regressor:random_forest:criterion&#39;: &#39;mse&#39;, &#39;regressor:random_forest:max_depth&#39;: &#39;None&#39;, &#39;regressor:random_forest:max_features&#39;: 1.0, &#39;regressor:random_forest:max_leaf_nodes&#39;: &#39;None&#39;, &#39;regressor:random_forest:min_impurity_decrease&#39;: 0.0, &#39;regressor:random_forest:min_samples_leaf&#39;: 1, &#39;regressor:random_forest:min_samples_split&#39;: 2, &#39;regressor:random_forest:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.01},
-dataset_properties={
-  &#39;task&#39;: 5,
-  &#39;sparse&#39;: False,
-  &#39;multioutput&#39;: True,
-  &#39;target_type&#39;: &#39;regression&#39;,
-  &#39;signed&#39;: False})),
-(0.200000, SimpleRegressionPipeline({&#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;polynomial&#39;, &#39;regressor:__choice__&#39;: &#39;k_nearest_neighbors&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;minmax&#39;, &#39;feature_preprocessor:polynomial:degree&#39;: 3, &#39;feature_preprocessor:polynomial:include_bias&#39;: &#39;False&#39;, &#39;feature_preprocessor:polynomial:interaction_only&#39;: &#39;True&#39;, &#39;regressor:k_nearest_neighbors:n_neighbors&#39;: 1, &#39;regressor:k_nearest_neighbors:p&#39;: 2, &#39;regressor:k_nearest_neighbors:weights&#39;: &#39;distance&#39;},
-dataset_properties={
-  &#39;task&#39;: 5,
-  &#39;sparse&#39;: False,
-  &#39;multioutput&#39;: True,
-  &#39;target_type&#39;: &#39;regression&#39;,
-  &#39;signed&#39;: False})),
-(0.180000, SimpleRegressionPipeline({&#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;pca&#39;, &#39;regressor:__choice__&#39;: &#39;k_nearest_neighbors&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;normalize&#39;, &#39;feature_preprocessor:pca:keep_variance&#39;: 0.8772654126094648, &#39;feature_preprocessor:pca:whiten&#39;: &#39;True&#39;, &#39;regressor:k_nearest_neighbors:n_neighbors&#39;: 14, &#39;regressor:k_nearest_neighbors:p&#39;: 1, &#39;regressor:k_nearest_neighbors:weights&#39;: &#39;uniform&#39;},
-dataset_properties={
-  &#39;task&#39;: 5,
-  &#39;sparse&#39;: False,
-  &#39;multioutput&#39;: True,
-  &#39;target_type&#39;: &#39;regression&#39;,
-  &#39;signed&#39;: False})),
-]
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>{   2: {   &#39;cost&#39;: 0.14281372797350056,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78f7b94c0&gt;,
+           &#39;ensemble_weight&#39;: 0.42,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c920640&gt;,
+           &#39;model_id&#39;: 2,
+           &#39;rank&#39;: 2,
+           &#39;regressor&#39;: &lt;autosklearn.pipeline.components.regression.RegressorChoice object at 0x7fc78c920430&gt;,
+           &#39;sklearn_regressor&#39;: RandomForestRegressor(max_features=1.0, n_estimators=512, n_jobs=1,
+                      random_state=1, warm_start=True)},
+    21: {   &#39;cost&#39;: 0.22480405782181234,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78f8d10a0&gt;,
+            &#39;ensemble_weight&#39;: 0.26,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78ca212e0&gt;,
+            &#39;model_id&#39;: 21,
+            &#39;rank&#39;: 3,
+            &#39;regressor&#39;: &lt;autosklearn.pipeline.components.regression.RegressorChoice object at 0x7fc78ca210d0&gt;,
+            &#39;sklearn_regressor&#39;: RandomForestRegressor(criterion=&#39;friedman_mse&#39;, max_features=0.9689529284463153,
+                      min_samples_leaf=5, min_samples_split=4, n_estimators=512,
+                      n_jobs=1, random_state=1, warm_start=True)},
+    22: {   &#39;cost&#39;: 0.13651009800609504,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78f7b9640&gt;,
+            &#39;ensemble_weight&#39;: 0.32,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78f7b94f0&gt;,
+            &#39;model_id&#39;: 22,
+            &#39;rank&#39;: 1,
+            &#39;regressor&#39;: &lt;autosklearn.pipeline.components.regression.RegressorChoice object at 0x7fc78f7b9f10&gt;,
+            &#39;sklearn_regressor&#39;: ExtraTreesRegressor(bootstrap=True, criterion=&#39;mae&#39;,
+                    max_features=0.9195228400065838, min_samples_leaf=2,
+                    min_samples_split=7, n_estimators=512, n_jobs=1,
+                    random_state=1, warm_start=True)}}
 </pre></div>
 </div>
-</section>
-<section id="get-the-score-of-the-final-ensemble">
+</div>
+<div class="section" id="get-the-score-of-the-final-ensemble">
 <h2>Get the Score of the final ensemble<a class="headerlink" href="#get-the-score-of-the-final-ensemble" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">predictions</span> <span class="o">=</span> <span class="n">automl</span><span class="o">.</span><span class="n">predict</span><span class="p">(</span><span class="n">X_test</span><span class="p">)</span>
 <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;R2 score:&quot;</span><span class="p">,</span> <span class="n">r2_score</span><span class="p">(</span><span class="n">y_test</span><span class="p">,</span> <span class="n">predictions</span><span class="p">))</span>
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>R2 score: 0.8683277328634489
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>R2 score: 0.8858495026618883
 </pre></div>
 </div>
-</section>
-<section id="get-the-configuration-space">
+</div>
+<div class="section" id="get-the-configuration-space">
 <h2>Get the configuration space<a class="headerlink" href="#get-the-configuration-space" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="c1"># The configuration space is reduced, i.e. no SVM.</span>
 <span class="nb">print</span><span class="p">(</span><span class="n">automl</span><span class="o">.</span><span class="n">get_configuration_space</span><span class="p">(</span><span class="n">X_train</span><span class="p">,</span> <span class="n">y_train</span><span class="p">))</span>
@@ -417,7 +425,7 @@ <h2>Get the configuration space<a class="headerlink" href="#get-the-configuratio
     (Forbidden: regressor:__choice__ == &#39;random_forest&#39; &amp;&amp; Forbidden: feature_preprocessor:__choice__ == &#39;nystroem_sampler&#39;)
 </pre></div>
 </div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  55.310 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  56.218 seconds)</p>
 <div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-examples-20-basic-example-multioutput-regression-py">
 <div class="binder-badge docutils container">
 <a class="reference external image-reference" href="https://mybinder.org/v2/gh/automl/auto-sklearn/master?urlpath=lab/tree/notebooks/examples/20_basic/example_multioutput_regression.ipynb"><img alt="Launch binder" src="../../_images/binder_badge_logo.svg" width="150px" /></a>
@@ -430,8 +438,8 @@ <h2>Get the configuration space<a class="headerlink" href="#get-the-configuratio
 </div>
 </div>
 <p class="sphx-glr-signature"><a class="reference external" href="https://sphinx-gallery.github.io">Gallery generated by Sphinx-Gallery</a></p>
-</section>
-</section>
+</div>
+</div>
 
 
     </div>
@@ -452,7 +460,7 @@ <h2>Get the configuration space<a class="headerlink" href="#get-the-configuratio
       
     </p>
     <p>
-        &copy; Copyright 2014-2021, Machine Learning Professorship Freiburg.<br/>
+        &copy; Copyright 2014-2022, Machine Learning Professorship Freiburg.<br/>
       Created using <a href="http://sphinx-doc.org/">Sphinx</a> 4.2.0.<br/>
     </p>
   </div>
diff --git a/master/examples/20_basic/example_regression.html b/master/examples/20_basic/example_regression.html
index 772ac1975b..8b77b1d09b 100644
--- a/master/examples/20_basic/example_regression.html
+++ b/master/examples/20_basic/example_regression.html
@@ -3,9 +3,8 @@
 <html>
   <head>
     <meta charset="utf-8" />
-    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="generator" content="Docutils 0.17.1: http://docutils.sourceforge.net/" />
-
-    <title>Regression &#8212; AutoSklearn 0.14.2 documentation</title>
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>Regression &#8212; AutoSklearn 0.14.4 documentation</title>
     <link rel="stylesheet" type="text/css" href="../../_static/pygments.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/bootstrap-sphinx.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/sg_gallery.css" />
@@ -55,7 +54,7 @@
         </button>
         <a class="navbar-brand" href="../../index.html">
           auto-sklearn</a>
-        <span class="navbar-text navbar-version pull-left"><b>0.14.2</b></span>
+        <span class="navbar-text navbar-version pull-left"><b>0.14.4</b></span>
       </div>
 
         <div class="collapse navbar-collapse nav-collapse">
@@ -127,18 +126,20 @@
 <p>Click <a class="reference internal" href="#sphx-glr-download-examples-20-basic-example-regression-py"><span class="std std-ref">here</span></a>
 to download the full example code or to run this example in your browser via Binder</p>
 </div>
-<section class="sphx-glr-example-title" id="regression">
+<div class="sphx-glr-example-title section" id="regression">
 <span id="sphx-glr-examples-20-basic-example-regression-py"></span><h1>Regression<a class="headerlink" href="#regression" title="Permalink to this headline">¶</a></h1>
 <p>The following example shows how to fit a simple regression model with
 <em>auto-sklearn</em>.</p>
-<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">sklearn.datasets</span>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">pprint</span> <span class="kn">import</span> <span class="n">pprint</span>
+
+<span class="kn">import</span> <span class="nn">sklearn.datasets</span>
 <span class="kn">import</span> <span class="nn">sklearn.metrics</span>
 
 <span class="kn">import</span> <span class="nn">autosklearn.regression</span>
 <span class="kn">import</span> <span class="nn">matplotlib.pyplot</span> <span class="k">as</span> <span class="nn">plt</span>
 </pre></div>
 </div>
-<section id="data-loading">
+<div class="section" id="data-loading">
 <h2>Data Loading<a class="headerlink" href="#data-loading" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">X</span><span class="p">,</span> <span class="n">y</span> <span class="o">=</span> <span class="n">sklearn</span><span class="o">.</span><span class="n">datasets</span><span class="o">.</span><span class="n">load_diabetes</span><span class="p">(</span><span class="n">return_X_y</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
 
@@ -146,8 +147,8 @@ <h2>Data Loading<a class="headerlink" href="#data-loading" title="Permalink to t
     <span class="n">sklearn</span><span class="o">.</span><span class="n">model_selection</span><span class="o">.</span><span class="n">train_test_split</span><span class="p">(</span><span class="n">X</span><span class="p">,</span> <span class="n">y</span><span class="p">,</span> <span class="n">random_state</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
 </pre></div>
 </div>
-</section>
-<section id="build-and-fit-a-regressor">
+</div>
+<div class="section" id="build-and-fit-a-regressor">
 <h2>Build and fit a regressor<a class="headerlink" href="#build-and-fit-a-regressor" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">automl</span> <span class="o">=</span> <span class="n">autosklearn</span><span class="o">.</span><span class="n">regression</span><span class="o">.</span><span class="n">AutoSklearnRegressor</span><span class="p">(</span>
     <span class="n">time_left_for_this_task</span><span class="o">=</span><span class="mi">120</span><span class="p">,</span>
@@ -158,12 +159,17 @@ <h2>Build and fit a regressor<a class="headerlink" href="#build-and-fit-a-regres
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>AutoSklearnRegressor(per_run_time_limit=30, time_left_for_this_task=120,
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:68: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+  self.metafeatures = self.metafeatures.append(metafeatures)
+/home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:72: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+  self.algorithm_runs[metric].append(runs)
+
+AutoSklearnRegressor(per_run_time_limit=30, time_left_for_this_task=120,
                      tmp_folder=&#39;/tmp/autosklearn_regression_example_tmp&#39;)
 </pre></div>
 </div>
-</section>
-<section id="view-the-models-found-by-auto-sklearn">
+</div>
+<div class="section" id="view-the-models-found-by-auto-sklearn">
 <h2>View the models found by auto-sklearn<a class="headerlink" href="#view-the-models-found-by-auto-sklearn" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="nb">print</span><span class="p">(</span><span class="n">automl</span><span class="o">.</span><span class="n">leaderboard</span><span class="p">())</span>
 </pre></div>
@@ -171,60 +177,69 @@ <h2>View the models found by auto-sklearn<a class="headerlink" href="#view-the-m
 <p class="sphx-glr-script-out">Out:</p>
 <div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>          rank  ensemble_weight               type      cost  duration
 model_id
-25           1             0.46                sgd  0.436679  0.628683
-6            2             0.32     ard_regression  0.455042  0.645808
-27           3             0.14     ard_regression  0.462249  0.639516
-11           4             0.02      random_forest  0.507400  8.241306
-7            5             0.06  gradient_boosting  0.518673  1.216493
+25           1             0.44                sgd  0.436679  0.608678
+6            2             0.34     ard_regression  0.455042  0.619489
+39           3             0.18     ard_regression  0.474807  0.617622
+7            4             0.04  gradient_boosting  0.518673  1.104580
 </pre></div>
 </div>
-</section>
-<section id="print-the-final-ensemble-constructed-by-auto-sklearn">
+</div>
+<div class="section" id="print-the-final-ensemble-constructed-by-auto-sklearn">
 <h2>Print the final ensemble constructed by auto-sklearn<a class="headerlink" href="#print-the-final-ensemble-constructed-by-auto-sklearn" title="Permalink to this headline">¶</a></h2>
-<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="nb">print</span><span class="p">(</span><span class="n">automl</span><span class="o">.</span><span class="n">show_models</span><span class="p">())</span>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">pprint</span><span class="p">(</span><span class="n">automl</span><span class="o">.</span><span class="n">show_models</span><span class="p">(),</span> <span class="n">indent</span><span class="o">=</span><span class="mi">4</span><span class="p">)</span>
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>[(0.460000, SimpleRegressionPipeline({&#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;polynomial&#39;, &#39;regressor:__choice__&#39;: &#39;sgd&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;minmax&#39;, &#39;feature_preprocessor:polynomial:degree&#39;: 3, &#39;feature_preprocessor:polynomial:include_bias&#39;: &#39;False&#39;, &#39;feature_preprocessor:polynomial:interaction_only&#39;: &#39;False&#39;, &#39;regressor:sgd:alpha&#39;: 0.0006517033225329654, &#39;regressor:sgd:average&#39;: &#39;False&#39;, &#39;regressor:sgd:fit_intercept&#39;: &#39;True&#39;, &#39;regressor:sgd:learning_rate&#39;: &#39;invscaling&#39;, &#39;regressor:sgd:loss&#39;: &#39;epsilon_insensitive&#39;, &#39;regressor:sgd:penalty&#39;: &#39;elasticnet&#39;, &#39;regressor:sgd:tol&#39;: 0.002431731981071206, &#39;regressor:sgd:epsilon&#39;: 0.012150149892783745, &#39;regressor:sgd:eta0&#39;: 0.016444224834275295, &#39;regressor:sgd:l1_ratio&#39;: 1.7462342366289323e-09, &#39;regressor:sgd:power_t&#39;: 0.21521743568582094},
-dataset_properties={
-  &#39;task&#39;: 4,
-  &#39;sparse&#39;: False,
-  &#39;multioutput&#39;: False,
-  &#39;target_type&#39;: &#39;regression&#39;,
-  &#39;signed&#39;: False})),
-(0.320000, SimpleRegressionPipeline({&#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;polynomial&#39;, &#39;regressor:__choice__&#39;: &#39;ard_regression&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;standardize&#39;, &#39;feature_preprocessor:polynomial:degree&#39;: 2, &#39;feature_preprocessor:polynomial:include_bias&#39;: &#39;True&#39;, &#39;feature_preprocessor:polynomial:interaction_only&#39;: &#39;False&#39;, &#39;regressor:ard_regression:alpha_1&#39;: 0.0003701926442639788, &#39;regressor:ard_regression:alpha_2&#39;: 2.2118001735899097e-07, &#39;regressor:ard_regression:fit_intercept&#39;: &#39;True&#39;, &#39;regressor:ard_regression:lambda_1&#39;: 1.2037591637980971e-06, &#39;regressor:ard_regression:lambda_2&#39;: 4.358378124977852e-09, &#39;regressor:ard_regression:n_iter&#39;: 300, &#39;regressor:ard_regression:threshold_lambda&#39;: 1136.5286041327277, &#39;regressor:ard_regression:tol&#39;: 0.021944240404849075},
-dataset_properties={
-  &#39;task&#39;: 4,
-  &#39;sparse&#39;: False,
-  &#39;multioutput&#39;: False,
-  &#39;target_type&#39;: &#39;regression&#39;,
-  &#39;signed&#39;: False})),
-(0.140000, SimpleRegressionPipeline({&#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;select_rates_regression&#39;, &#39;regressor:__choice__&#39;: &#39;ard_regression&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;robust_scaler&#39;, &#39;feature_preprocessor:select_rates_regression:alpha&#39;: 0.3621762718897781, &#39;feature_preprocessor:select_rates_regression:mode&#39;: &#39;fwe&#39;, &#39;feature_preprocessor:select_rates_regression:score_func&#39;: &#39;f_regression&#39;, &#39;regressor:ard_regression:alpha_1&#39;: 2.7664515192592053e-05, &#39;regressor:ard_regression:alpha_2&#39;: 9.504988116581138e-07, &#39;regressor:ard_regression:fit_intercept&#39;: &#39;True&#39;, &#39;regressor:ard_regression:lambda_1&#39;: 6.50650698230178e-09, &#39;regressor:ard_regression:lambda_2&#39;: 4.238533890074848e-07, &#39;regressor:ard_regression:n_iter&#39;: 300, &#39;regressor:ard_regression:threshold_lambda&#39;: 78251.58542976103, &#39;regressor:ard_regression:tol&#39;: 0.0007301343236220855, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.0001745391328519669, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max&#39;: 0.8057830372269097, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min&#39;: 0.24982831110057324},
-dataset_properties={
-  &#39;task&#39;: 4,
-  &#39;sparse&#39;: False,
-  &#39;multioutput&#39;: False,
-  &#39;target_type&#39;: &#39;regression&#39;,
-  &#39;signed&#39;: False})),
-(0.060000, SimpleRegressionPipeline({&#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;regressor:__choice__&#39;: &#39;gradient_boosting&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;quantile_transformer&#39;, &#39;regressor:gradient_boosting:early_stop&#39;: &#39;off&#39;, &#39;regressor:gradient_boosting:l2_regularization&#39;: 1.8428972335335263e-10, &#39;regressor:gradient_boosting:learning_rate&#39;: 0.012607824914758717, &#39;regressor:gradient_boosting:loss&#39;: &#39;least_squares&#39;, &#39;regressor:gradient_boosting:max_bins&#39;: 255, &#39;regressor:gradient_boosting:max_depth&#39;: &#39;None&#39;, &#39;regressor:gradient_boosting:max_leaf_nodes&#39;: 10, &#39;regressor:gradient_boosting:min_samples_leaf&#39;: 8, &#39;regressor:gradient_boosting:scoring&#39;: &#39;loss&#39;, &#39;regressor:gradient_boosting:tol&#39;: 1e-07, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles&#39;: 952, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution&#39;: &#39;uniform&#39;},
-dataset_properties={
-  &#39;task&#39;: 4,
-  &#39;sparse&#39;: False,
-  &#39;multioutput&#39;: False,
-  &#39;target_type&#39;: &#39;regression&#39;,
-  &#39;signed&#39;: False})),
-(0.020000, SimpleRegressionPipeline({&#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;polynomial&#39;, &#39;regressor:__choice__&#39;: &#39;random_forest&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;minmax&#39;, &#39;feature_preprocessor:polynomial:degree&#39;: 2, &#39;feature_preprocessor:polynomial:include_bias&#39;: &#39;True&#39;, &#39;feature_preprocessor:polynomial:interaction_only&#39;: &#39;True&#39;, &#39;regressor:random_forest:bootstrap&#39;: &#39;False&#39;, &#39;regressor:random_forest:criterion&#39;: &#39;mae&#39;, &#39;regressor:random_forest:max_depth&#39;: &#39;None&#39;, &#39;regressor:random_forest:max_features&#39;: 0.6277363920171745, &#39;regressor:random_forest:max_leaf_nodes&#39;: &#39;None&#39;, &#39;regressor:random_forest:min_impurity_decrease&#39;: 0.0, &#39;regressor:random_forest:min_samples_leaf&#39;: 6, &#39;regressor:random_forest:min_samples_split&#39;: 15, &#39;regressor:random_forest:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.0010413452644415357},
-dataset_properties={
-  &#39;task&#39;: 4,
-  &#39;sparse&#39;: False,
-  &#39;multioutput&#39;: False,
-  &#39;target_type&#39;: &#39;regression&#39;,
-  &#39;signed&#39;: False})),
-]
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>{   6: {   &#39;cost&#39;: 0.4550418898836528,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78a826a00&gt;,
+           &#39;ensemble_weight&#39;: 0.34,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78b503820&gt;,
+           &#39;model_id&#39;: 6,
+           &#39;rank&#39;: 2,
+           &#39;regressor&#39;: &lt;autosklearn.pipeline.components.regression.RegressorChoice object at 0x7fc78b503880&gt;,
+           &#39;sklearn_regressor&#39;: ARDRegression(alpha_1=0.0003701926442639788, alpha_2=2.2118001735899097e-07,
+              copy_X=False, lambda_1=1.2037591637980971e-06,
+              lambda_2=4.358378124977852e-09,
+              threshold_lambda=1136.5286041327277, tol=0.021944240404849075)},
+    7: {   &#39;cost&#39;: 0.5186726734789994,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78f913a00&gt;,
+           &#39;ensemble_weight&#39;: 0.04,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78b48c700&gt;,
+           &#39;model_id&#39;: 7,
+           &#39;rank&#39;: 4,
+           &#39;regressor&#39;: &lt;autosklearn.pipeline.components.regression.RegressorChoice object at 0x7fc78a5d3580&gt;,
+           &#39;sklearn_regressor&#39;: HistGradientBoostingRegressor(l2_regularization=1.8428972335335263e-10,
+                              learning_rate=0.012607824914758717, max_iter=512,
+                              max_leaf_nodes=10, min_samples_leaf=8,
+                              n_iter_no_change=0, random_state=1,
+                              validation_fraction=None, warm_start=True)},
+    25: {   &#39;cost&#39;: 0.43667876507897496,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78c94d400&gt;,
+            &#39;ensemble_weight&#39;: 0.44,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78f913df0&gt;,
+            &#39;model_id&#39;: 25,
+            &#39;rank&#39;: 1,
+            &#39;regressor&#39;: &lt;autosklearn.pipeline.components.regression.RegressorChoice object at 0x7fc78f80b160&gt;,
+            &#39;sklearn_regressor&#39;: SGDRegressor(alpha=0.0006517033225329654, epsilon=0.012150149892783745,
+             eta0=0.016444224834275295, l1_ratio=1.7462342366289323e-09,
+             loss=&#39;epsilon_insensitive&#39;, max_iter=16, penalty=&#39;elasticnet&#39;,
+             power_t=0.21521743568582094, random_state=1,
+             tol=0.002431731981071206, warm_start=True)},
+    39: {   &#39;cost&#39;: 0.4748068089650166,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78f80b610&gt;,
+            &#39;ensemble_weight&#39;: 0.18,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c482280&gt;,
+            &#39;model_id&#39;: 39,
+            &#39;rank&#39;: 3,
+            &#39;regressor&#39;: &lt;autosklearn.pipeline.components.regression.RegressorChoice object at 0x7fc78c4826a0&gt;,
+            &#39;sklearn_regressor&#39;: ARDRegression(alpha_1=0.0005012365297609799, alpha_2=3.025360750168211e-08,
+              copy_X=False, lambda_1=4.9749646614525684e-05,
+              lambda_2=3.2368037115065363e-10,
+              threshold_lambda=18669.665899307194, tol=0.0012624032013298571)}}
 </pre></div>
 </div>
-</section>
-<section id="get-the-score-of-the-final-ensemble">
+</div>
+<div class="section" id="get-the-score-of-the-final-ensemble">
 <h2>Get the Score of the final ensemble<a class="headerlink" href="#get-the-score-of-the-final-ensemble" title="Permalink to this headline">¶</a></h2>
 <p>After training the estimator, we can now quantify the goodness of fit. One possibility for
 is the <a class="reference external" href="https://scikit-learn.org/stable/modules/model_evaluation.html#r2-score">R2 score</a>.
@@ -237,12 +252,12 @@ <h2>Get the Score of the final ensemble<a class="headerlink" href="#get-the-scor
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Train R2 score: 0.5944780427522034
-Test R2 score: 0.3959585042866587
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Train R2 score: 0.5855373845454157
+Test R2 score: 0.39879073225079487
 </pre></div>
 </div>
-</section>
-<section id="plot-the-predictions">
+</div>
+<div class="section" id="plot-the-predictions">
 <h2>Plot the predictions<a class="headerlink" href="#plot-the-predictions" title="Permalink to this headline">¶</a></h2>
 <p>Furthermore, we can now visually inspect the predictions. We plot the true value against the
 predictions and show results on train and test data. Points on the diagonal depict perfect
@@ -261,7 +276,7 @@ <h2>Plot the predictions<a class="headerlink" href="#plot-the-predictions" title
 <span class="n">plt</span><span class="o">.</span><span class="n">show</span><span class="p">()</span>
 </pre></div>
 </div>
-<img src="../../_images/sphx_glr_example_regression_001.png" srcset="../../_images/sphx_glr_example_regression_001.png" alt="example regression" class = "sphx-glr-single-img"/><p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  58.195 seconds)</p>
+<img src="../../_images/sphx_glr_example_regression_001.png" srcset="../../_images/sphx_glr_example_regression_001.png" alt="example regression" class = "sphx-glr-single-img"/><p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  59.382 seconds)</p>
 <div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-examples-20-basic-example-regression-py">
 <div class="binder-badge docutils container">
 <a class="reference external image-reference" href="https://mybinder.org/v2/gh/automl/auto-sklearn/master?urlpath=lab/tree/notebooks/examples/20_basic/example_regression.ipynb"><img alt="Launch binder" src="../../_images/binder_badge_logo.svg" width="150px" /></a>
@@ -274,8 +289,8 @@ <h2>Plot the predictions<a class="headerlink" href="#plot-the-predictions" title
 </div>
 </div>
 <p class="sphx-glr-signature"><a class="reference external" href="https://sphinx-gallery.github.io">Gallery generated by Sphinx-Gallery</a></p>
-</section>
-</section>
+</div>
+</div>
 
 
     </div>
@@ -296,7 +311,7 @@ <h2>Plot the predictions<a class="headerlink" href="#plot-the-predictions" title
       
     </p>
     <p>
-        &copy; Copyright 2014-2021, Machine Learning Professorship Freiburg.<br/>
+        &copy; Copyright 2014-2022, Machine Learning Professorship Freiburg.<br/>
       Created using <a href="http://sphinx-doc.org/">Sphinx</a> 4.2.0.<br/>
     </p>
   </div>
diff --git a/master/examples/20_basic/sg_execution_times.html b/master/examples/20_basic/sg_execution_times.html
index 6d6e8ee070..e76015261d 100644
--- a/master/examples/20_basic/sg_execution_times.html
+++ b/master/examples/20_basic/sg_execution_times.html
@@ -3,9 +3,8 @@
 <html>
   <head>
     <meta charset="utf-8" />
-    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="generator" content="Docutils 0.17.1: http://docutils.sourceforge.net/" />
-
-    <title>Computation times &#8212; AutoSklearn 0.14.2 documentation</title>
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>Computation times &#8212; AutoSklearn 0.14.4 documentation</title>
     <link rel="stylesheet" type="text/css" href="../../_static/pygments.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/bootstrap-sphinx.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/sg_gallery.css" />
@@ -55,7 +54,7 @@
         </button>
         <a class="navbar-brand" href="../../index.html">
           auto-sklearn</a>
-        <span class="navbar-text navbar-version pull-left"><b>0.14.2</b></span>
+        <span class="navbar-text navbar-version pull-left"><b>0.14.4</b></span>
       </div>
 
         <div class="collapse navbar-collapse nav-collapse">
@@ -114,9 +113,9 @@
       </div>
     <div class="body col-md-9 content" role="main">
       
-  <section id="computation-times">
+  <div class="section" id="computation-times">
 <span id="sphx-glr-examples-20-basic-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>06:10.128</strong> total execution time for <strong>examples_20_basic</strong> files:</p>
+<p><strong>06:09.522</strong> total execution time for <strong>examples_20_basic</strong> files:</p>
 <table class="docutils align-default">
 <colgroup>
 <col style="width: 86%" />
@@ -125,24 +124,24 @@
 </colgroup>
 <tbody>
 <tr class="row-odd"><td><p><a class="reference internal" href="example_classification.html#sphx-glr-examples-20-basic-example-classification-py"><span class="std std-ref">Classification</span></a> (<code class="docutils literal notranslate"><span class="pre">example_classification.py</span></code>)</p></td>
-<td><p>02:00.650</p></td>
+<td><p>02:01.747</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="example_regression.html#sphx-glr-examples-20-basic-example-regression-py"><span class="std std-ref">Regression</span></a> (<code class="docutils literal notranslate"><span class="pre">example_regression.py</span></code>)</p></td>
-<td><p>01:58.195</p></td>
+<td><p>01:59.382</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-odd"><td><p><a class="reference internal" href="example_multioutput_regression.html#sphx-glr-examples-20-basic-example-multioutput-regression-py"><span class="std std-ref">Multi-output Regression</span></a> (<code class="docutils literal notranslate"><span class="pre">example_multioutput_regression.py</span></code>)</p></td>
-<td><p>01:55.310</p></td>
+<td><p>01:56.218</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="example_multilabel_classification.html#sphx-glr-examples-20-basic-example-multilabel-classification-py"><span class="std std-ref">Multi-label Classification</span></a> (<code class="docutils literal notranslate"><span class="pre">example_multilabel_classification.py</span></code>)</p></td>
-<td><p>00:15.973</p></td>
+<td><p>00:12.175</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 </tbody>
 </table>
-</section>
+</div>
 
 
     </div>
@@ -163,7 +162,7 @@
       
     </p>
     <p>
-        &copy; Copyright 2014-2021, Machine Learning Professorship Freiburg.<br/>
+        &copy; Copyright 2014-2022, Machine Learning Professorship Freiburg.<br/>
       Created using <a href="http://sphinx-doc.org/">Sphinx</a> 4.2.0.<br/>
     </p>
   </div>
diff --git a/master/examples/40_advanced/example_calc_multiple_metrics.html b/master/examples/40_advanced/example_calc_multiple_metrics.html
index 4adb7cf7d2..f454e578a4 100644
--- a/master/examples/40_advanced/example_calc_multiple_metrics.html
+++ b/master/examples/40_advanced/example_calc_multiple_metrics.html
@@ -3,9 +3,8 @@
 <html>
   <head>
     <meta charset="utf-8" />
-    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="generator" content="Docutils 0.17.1: http://docutils.sourceforge.net/" />
-
-    <title>Metrics &#8212; AutoSklearn 0.14.2 documentation</title>
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>Metrics &#8212; AutoSklearn 0.14.4 documentation</title>
     <link rel="stylesheet" type="text/css" href="../../_static/pygments.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/bootstrap-sphinx.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/sg_gallery.css" />
@@ -55,7 +54,7 @@
         </button>
         <a class="navbar-brand" href="../../index.html">
           auto-sklearn</a>
-        <span class="navbar-text navbar-version pull-left"><b>0.14.2</b></span>
+        <span class="navbar-text navbar-version pull-left"><b>0.14.4</b></span>
       </div>
 
         <div class="collapse navbar-collapse nav-collapse">
@@ -124,7 +123,7 @@
 <p>Click <a class="reference internal" href="#sphx-glr-download-examples-40-advanced-example-calc-multiple-metrics-py"><span class="std std-ref">here</span></a>
 to download the full example code or to run this example in your browser via Binder</p>
 </div>
-<section class="sphx-glr-example-title" id="metrics">
+<div class="sphx-glr-example-title section" id="metrics">
 <span id="sphx-glr-examples-40-advanced-example-calc-multiple-metrics-py"></span><h1>Metrics<a class="headerlink" href="#metrics" title="Permalink to this headline">¶</a></h1>
 <p>In <em>Auto-sklearn</em>, model is optimized over a metric, either built-in or
 custom metric. Moreover, it is also possible to calculate multiple metrics
@@ -151,15 +150,15 @@
     <span class="k">return</span> <span class="n">results</span><span class="p">[</span><span class="n">cols</span><span class="p">]</span>
 </pre></div>
 </div>
-<section id="data-loading">
+<div class="section" id="data-loading">
 <h2>Data Loading<a class="headerlink" href="#data-loading" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">X</span><span class="p">,</span> <span class="n">y</span> <span class="o">=</span> <span class="n">sklearn</span><span class="o">.</span><span class="n">datasets</span><span class="o">.</span><span class="n">load_breast_cancer</span><span class="p">(</span><span class="n">return_X_y</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
 <span class="n">X_train</span><span class="p">,</span> <span class="n">X_test</span><span class="p">,</span> <span class="n">y_train</span><span class="p">,</span> <span class="n">y_test</span> <span class="o">=</span> \
     <span class="n">sklearn</span><span class="o">.</span><span class="n">model_selection</span><span class="o">.</span><span class="n">train_test_split</span><span class="p">(</span><span class="n">X</span><span class="p">,</span> <span class="n">y</span><span class="p">,</span> <span class="n">random_state</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
 </pre></div>
 </div>
-</section>
-<section id="build-and-fit-a-classifier">
+</div>
+<div class="section" id="build-and-fit-a-classifier">
 <h2>Build and fit a classifier<a class="headerlink" href="#build-and-fit-a-classifier" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">error_rate</span> <span class="o">=</span> <span class="n">autosklearn</span><span class="o">.</span><span class="n">metrics</span><span class="o">.</span><span class="n">make_scorer</span><span class="p">(</span>
     <span class="n">name</span><span class="o">=</span><span class="s1">&#39;custom_error&#39;</span><span class="p">,</span>
@@ -178,14 +177,19 @@ <h2>Build and fit a classifier<a class="headerlink" href="#build-and-fit-a-class
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>AutoSklearnClassifier(per_run_time_limit=30,
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:68: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+  self.metafeatures = self.metafeatures.append(metafeatures)
+/home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:72: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+  self.algorithm_runs[metric].append(runs)
+
+AutoSklearnClassifier(per_run_time_limit=30,
                       scoring_functions=[balanced_accuracy, precision, recall,
                                          f1, custom_error],
                       time_left_for_this_task=120)
 </pre></div>
 </div>
-</section>
-<section id="get-the-score-of-the-final-ensemble">
+</div>
+<div class="section" id="get-the-score-of-the-final-ensemble">
 <h2>Get the Score of the final ensemble<a class="headerlink" href="#get-the-score-of-the-final-ensemble" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">predictions</span> <span class="o">=</span> <span class="bp">cls</span><span class="o">.</span><span class="n">predict</span><span class="p">(</span><span class="n">X_test</span><span class="p">)</span>
 <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Accuracy score&quot;</span><span class="p">,</span> <span class="n">sklearn</span><span class="o">.</span><span class="n">metrics</span><span class="o">.</span><span class="n">accuracy_score</span><span class="p">(</span><span class="n">y_test</span><span class="p">,</span> <span class="n">predictions</span><span class="p">))</span>
@@ -196,50 +200,54 @@ <h2>Get the Score of the final ensemble<a class="headerlink" href="#get-the-scor
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Accuracy score 0.9440559440559441
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Accuracy score 0.958041958041958
 ################################################################################
 Metric results
  rank_test_scores param_classifier:__choice__  mean_test_score  metric_balanced_accuracy  metric_precision  metric_recall  metric_f1  metric_custom_error
-                7               random_forest         0.971631                  0.969533          0.977528       0.977528   0.977528             0.028369
-                7                         mlp         0.971631                  0.961538          0.956989       1.000000   0.978022             0.028369
-               29                         mlp         0.943262                  0.935069          0.945055       0.966292   0.955556             0.056738
-               17               random_forest         0.964539                  0.959918          0.966667       0.977528   0.972067             0.035461
-                7                         mlp         0.971631                  0.961538          0.956989       1.000000   0.978022             0.028369
+                5               random_forest         0.971631                  0.969533          0.977528       0.977528   0.977528             0.028369
+                5                         mlp         0.971631                  0.961538          0.956989       1.000000   0.978022             0.028369
+               31                         mlp         0.943262                  0.935069          0.945055       0.966292   0.955556             0.056738
+               19               random_forest         0.964539                  0.959918          0.966667       0.977528   0.972067             0.035461
+                5                         mlp         0.971631                  0.961538          0.956989       1.000000   0.978022             0.028369
                 1                 extra_trees         0.985816                  0.984767          0.988764       0.988764   0.988764             0.014184
-               17               random_forest         0.964539                  0.963915          0.977273       0.966292   0.971751             0.035461
-               22                 extra_trees         0.957447                  0.954300          0.966292       0.966292   0.966292             0.042553
-                7               random_forest         0.971631                  0.969533          0.977528       0.977528   0.977528             0.028369
-                7               random_forest         0.971631                  0.969533          0.977528       0.977528   0.977528             0.028369
-               17           gradient_boosting         0.964539                  0.963915          0.977273       0.966292   0.971751             0.035461
-                7           gradient_boosting         0.971631                  0.965536          0.967033       0.988764   0.977778             0.028369
-                7                         mlp         0.971631                  0.965536          0.967033       0.988764   0.977778             0.028369
-               26                         mlp         0.950355                  0.948682          0.965909       0.955056   0.960452             0.049645
+               19               random_forest         0.964539                  0.963915          0.977273       0.966292   0.971751             0.035461
+               25                 extra_trees         0.957447                  0.954300          0.966292       0.966292   0.966292             0.042553
+                5               random_forest         0.971631                  0.969533          0.977528       0.977528   0.977528             0.028369
+                5               random_forest         0.971631                  0.969533          0.977528       0.977528   0.977528             0.028369
+               19           gradient_boosting         0.964539                  0.963915          0.977273       0.966292   0.971751             0.035461
+                5           gradient_boosting         0.971631                  0.965536          0.967033       0.988764   0.977778             0.028369
+                5                         mlp         0.971631                  0.965536          0.967033       0.988764   0.977778             0.028369
+               28                         mlp         0.950355                  0.948682          0.965909       0.955056   0.960452             0.049645
                 3           gradient_boosting         0.978723                  0.975151          0.977778       0.988764   0.983240             0.021277
-               17           gradient_boosting         0.964539                  0.959918          0.966667       0.977528   0.972067             0.035461
-               17               random_forest         0.964539                  0.959918          0.966667       0.977528   0.972067             0.035461
-                7                 extra_trees         0.971631                  0.969533          0.977528       0.977528   0.977528             0.028369
-               34          passive_aggressive         0.921986                  0.894231          0.890000       1.000000   0.941799             0.078014
+               19           gradient_boosting         0.964539                  0.959918          0.966667       0.977528   0.972067             0.035461
+               19               random_forest         0.964539                  0.959918          0.966667       0.977528   0.972067             0.035461
+                5                 extra_trees         0.971631                  0.969533          0.977528       0.977528   0.977528             0.028369
+               37          passive_aggressive         0.921986                  0.894231          0.890000       1.000000   0.941799             0.078014
                 3                 extra_trees         0.978723                  0.975151          0.977778       0.988764   0.983240             0.021277
-                7           gradient_boosting         0.971631                  0.965536          0.967033       0.988764   0.977778             0.028369
-               26                         mlp         0.950355                  0.940687          0.945652       0.977528   0.961326             0.049645
-               32               random_forest         0.929078                  0.923833          0.943820       0.943820   0.943820             0.070922
-               22                    adaboost         0.957447                  0.950303          0.956044       0.977528   0.966667             0.042553
-                7                 extra_trees         0.971631                  0.965536          0.967033       0.988764   0.977778             0.028369
-               22                 extra_trees         0.957447                  0.954300          0.966292       0.966292   0.966292             0.042553
-               32                bernoulli_nb         0.929078                  0.923833          0.943820       0.943820   0.943820             0.070922
-                3                 extra_trees         0.978723                  0.979149          0.988636       0.977528   0.983051             0.021277
-               22               liblinear_svc         0.957447                  0.954300          0.966292       0.966292   0.966292             0.042553
-               29                 gaussian_nb         0.943262                  0.935069          0.945055       0.966292   0.955556             0.056738
-               26                 extra_trees         0.950355                  0.936690          0.936170       0.988764   0.961749             0.049645
-               29               random_forest         0.943262                  0.943064          0.965517       0.943820   0.954545             0.056738
+                5           gradient_boosting         0.971631                  0.965536          0.967033       0.988764   0.977778             0.028369
+               28                         mlp         0.950355                  0.940687          0.945652       0.977528   0.961326             0.049645
+               35               random_forest         0.929078                  0.923833          0.943820       0.943820   0.943820             0.070922
+               25                    adaboost         0.957447                  0.950303          0.956044       0.977528   0.966667             0.042553
+                5                 extra_trees         0.971631                  0.965536          0.967033       0.988764   0.977778             0.028369
+                5                 extra_trees         0.971631                  0.969533          0.977528       0.977528   0.977528             0.028369
+               33                         lda         0.936170                  0.925454          0.934783       0.966292   0.950276             0.063830
+                5                 extra_trees         0.971631                  0.965536          0.967033       0.988764   0.977778             0.028369
+                5                         lda         0.971631                  0.961538          0.956989       1.000000   0.978022             0.028369
+               19               liblinear_svc         0.964539                  0.959918          0.966667       0.977528   0.972067             0.035461
+                5               liblinear_svc         0.971631                  0.965536          0.967033       0.988764   0.977778             0.028369
+               31                    adaboost         0.943262                  0.943064          0.965517       0.943820   0.954545             0.056738
+               33               random_forest         0.936170                  0.937446          0.965116       0.932584   0.948571             0.063830
+               28                    adaboost         0.950355                  0.952679          0.976744       0.943820   0.960000             0.049645
+               40                 extra_trees         0.631206                  0.500000          0.631206       1.000000   0.773913             0.368794
+               38                 extra_trees         0.914894                  0.884615          0.881188       1.000000   0.936842             0.085106
+               41                         sgd         0.368794                  0.500000          0.000000       0.000000   0.000000             0.631206
                 1                 extra_trees         0.985816                  0.984767          0.988764       0.988764   0.988764             0.014184
-               35                 extra_trees         0.914894                  0.884615          0.881188       1.000000   0.936842             0.085106
-                3                 extra_trees         0.978723                  0.975151          0.977778       0.988764   0.983240             0.021277
-               35                  libsvm_svc         0.914894                  0.924589          0.975309       0.887640   0.929412             0.085106
-               37                         sgd         0.829787                  0.845182          0.933333       0.786517   0.853659             0.170213
+               25                         mlp         0.957447                  0.950303          0.956044       0.977528   0.966667             0.042553
+               35               decision_tree         0.929078                  0.919836          0.934066       0.955056   0.944444             0.070922
+               39          passive_aggressive         0.879433                  0.868518          0.900000       0.910112   0.905028             0.120567
 </pre></div>
 </div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes  8.166 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  59.942 seconds)</p>
 <div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-examples-40-advanced-example-calc-multiple-metrics-py">
 <div class="binder-badge docutils container">
 <a class="reference external image-reference" href="https://mybinder.org/v2/gh/automl/auto-sklearn/master?urlpath=lab/tree/notebooks/examples/40_advanced/example_calc_multiple_metrics.ipynb"><img alt="Launch binder" src="../../_images/binder_badge_logo1.svg" width="150px" /></a>
@@ -252,8 +260,8 @@ <h2>Get the Score of the final ensemble<a class="headerlink" href="#get-the-scor
 </div>
 </div>
 <p class="sphx-glr-signature"><a class="reference external" href="https://sphinx-gallery.github.io">Gallery generated by Sphinx-Gallery</a></p>
-</section>
-</section>
+</div>
+</div>
 
 
     </div>
@@ -274,7 +282,7 @@ <h2>Get the Score of the final ensemble<a class="headerlink" href="#get-the-scor
       
     </p>
     <p>
-        &copy; Copyright 2014-2021, Machine Learning Professorship Freiburg.<br/>
+        &copy; Copyright 2014-2022, Machine Learning Professorship Freiburg.<br/>
       Created using <a href="http://sphinx-doc.org/">Sphinx</a> 4.2.0.<br/>
     </p>
   </div>
diff --git a/master/examples/40_advanced/example_debug_logging.html b/master/examples/40_advanced/example_debug_logging.html
index b9d031fe09..378dc6bf7c 100644
--- a/master/examples/40_advanced/example_debug_logging.html
+++ b/master/examples/40_advanced/example_debug_logging.html
@@ -3,9 +3,8 @@
 <html>
   <head>
     <meta charset="utf-8" />
-    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="generator" content="Docutils 0.17.1: http://docutils.sourceforge.net/" />
-
-    <title>Logging and debugging &#8212; AutoSklearn 0.14.2 documentation</title>
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>Logging and debugging &#8212; AutoSklearn 0.14.4 documentation</title>
     <link rel="stylesheet" type="text/css" href="../../_static/pygments.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/bootstrap-sphinx.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/sg_gallery.css" />
@@ -55,7 +54,7 @@
         </button>
         <a class="navbar-brand" href="../../index.html">
           auto-sklearn</a>
-        <span class="navbar-text navbar-version pull-left"><b>0.14.2</b></span>
+        <span class="navbar-text navbar-version pull-left"><b>0.14.4</b></span>
       </div>
 
         <div class="collapse navbar-collapse nav-collapse">
@@ -124,7 +123,7 @@
 <p>Click <a class="reference internal" href="#sphx-glr-download-examples-40-advanced-example-debug-logging-py"><span class="std std-ref">here</span></a>
 to download the full example code or to run this example in your browser via Binder</p>
 </div>
-<section class="sphx-glr-example-title" id="logging-and-debugging">
+<div class="sphx-glr-example-title section" id="logging-and-debugging">
 <span id="sphx-glr-examples-40-advanced-example-debug-logging-py"></span><h1>Logging and debugging<a class="headerlink" href="#logging-and-debugging" title="Permalink to this headline">¶</a></h1>
 <p>This example shows how to provide a custom logging configuration to <em>auto-sklearn</em>.
 We will be fitting 2 pipelines and showing any INFO-level msg on console.
@@ -142,7 +141,7 @@
 <span class="kn">import</span> <span class="nn">autosklearn.classification</span>
 </pre></div>
 </div>
-<section id="data-loading">
+<div class="section" id="data-loading">
 <h2>Data Loading<a class="headerlink" href="#data-loading" title="Permalink to this headline">¶</a></h2>
 <p>Load kr-vs-kp dataset from <a class="reference external" href="https://www.openml.org/d/3">https://www.openml.org/d/3</a></p>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">X</span><span class="p">,</span> <span class="n">y</span> <span class="o">=</span> <span class="n">data</span> <span class="o">=</span> <span class="n">sklearn</span><span class="o">.</span><span class="n">datasets</span><span class="o">.</span><span class="n">fetch_openml</span><span class="p">(</span><span class="n">data_id</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">return_X_y</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span> <span class="n">as_frame</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
@@ -151,8 +150,8 @@ <h2>Data Loading<a class="headerlink" href="#data-loading" title="Permalink to t
      <span class="n">sklearn</span><span class="o">.</span><span class="n">model_selection</span><span class="o">.</span><span class="n">train_test_split</span><span class="p">(</span><span class="n">X</span><span class="p">,</span> <span class="n">y</span><span class="p">,</span> <span class="n">random_state</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
 </pre></div>
 </div>
-</section>
-<section id="create-a-logging-config">
+</div>
+<div class="section" id="create-a-logging-config">
 <h2>Create a logging config<a class="headerlink" href="#create-a-logging-config" title="Permalink to this headline">¶</a></h2>
 <p><em>auto-sklearn</em> uses a default
 <a class="reference external" href="https://github.com/automl/auto-sklearn/blob/master/autosklearn/util/logging.yaml">logging config</a>
@@ -190,8 +189,8 @@ <h2>Create a logging config<a class="headerlink" href="#create-a-logging-config"
 <span class="p">}</span>
 </pre></div>
 </div>
-</section>
-<section id="build-and-fit-a-classifier">
+</div>
+<div class="section" id="build-and-fit-a-classifier">
 <h2>Build and fit a classifier<a class="headerlink" href="#build-and-fit-a-classifier" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="bp">cls</span> <span class="o">=</span> <span class="n">autosklearn</span><span class="o">.</span><span class="n">classification</span><span class="o">.</span><span class="n">AutoSklearnClassifier</span><span class="p">(</span>
     <span class="n">time_left_for_this_task</span><span class="o">=</span><span class="mi">30</span><span class="p">,</span>
@@ -224,16 +223,16 @@ <h2>Build and fit a classifier<a class="headerlink" href="#build-and-fit-a-class
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>2021-11-09 19:52:24,412 - Client-EnsembleBuilder - INFO - DummyFuture: ([{&#39;Timestamp&#39;: Timestamp(&#39;2021-11-09 19:52:24.397684&#39;), &#39;ensemble_optimization_score&#39;: 0.9886363636363636, &#39;ensemble_test_score&#39;: 0.9899874843554443}], 50, None, None, None)/SingleThreadedClient() Started Ensemble builder job at 2021.11.09-19.52.24 for iteration 0.
-2021-11-09 19:52:28,964 - Client-EnsembleBuilder - INFO - DummyFuture: ([{&#39;Timestamp&#39;: Timestamp(&#39;2021-11-09 19:52:28.945787&#39;), &#39;ensemble_optimization_score&#39;: 0.9911616161616161, &#39;ensemble_test_score&#39;: 0.9899874843554443}], 50, None, None, None)/SingleThreadedClient() Started Ensemble builder job at 2021.11.09-19.52.28 for iteration 1.
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>2022-01-25 16:22:18,915 - Client-EnsembleBuilder - INFO - DummyFuture: ([{&#39;Timestamp&#39;: Timestamp(&#39;2022-01-25 16:22:18.897617&#39;), &#39;ensemble_optimization_score&#39;: 0.9886363636363636, &#39;ensemble_test_score&#39;: 0.9899874843554443}], 50, None, None, None)/SingleThreadedClient() Started Ensemble builder job at 2022.01.25-16.22.18 for iteration 0.
+2022-01-25 16:22:23,231 - Client-EnsembleBuilder - INFO - DummyFuture: ([{&#39;Timestamp&#39;: Timestamp(&#39;2022-01-25 16:22:23.213965&#39;), &#39;ensemble_optimization_score&#39;: 0.9911616161616161, &#39;ensemble_test_score&#39;: 0.9899874843554443}], 50, None, None, None)/SingleThreadedClient() Started Ensemble builder job at 2022.01.25-16.22.23 for iteration 1.
+tmp_folder/distributed.log
 tmp_folder/.auto-sklearn
-tmp_folder/AutoML(1):893634d8-4196-11ec-8a32-1b4c952643c3.log
-tmp_folder/space.json
 tmp_folder/smac3-output
-tmp_folder/distributed.log
+tmp_folder/AutoML(1):f52bd18f-7dfa-11ec-875f-3137a7553b01.log
+tmp_folder/space.json
 </pre></div>
 </div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 0 minutes  22.911 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 0 minutes  14.975 seconds)</p>
 <div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-examples-40-advanced-example-debug-logging-py">
 <div class="binder-badge docutils container">
 <a class="reference external image-reference" href="https://mybinder.org/v2/gh/automl/auto-sklearn/master?urlpath=lab/tree/notebooks/examples/40_advanced/example_debug_logging.ipynb"><img alt="Launch binder" src="../../_images/binder_badge_logo1.svg" width="150px" /></a>
@@ -246,8 +245,8 @@ <h2>Build and fit a classifier<a class="headerlink" href="#build-and-fit-a-class
 </div>
 </div>
 <p class="sphx-glr-signature"><a class="reference external" href="https://sphinx-gallery.github.io">Gallery generated by Sphinx-Gallery</a></p>
-</section>
-</section>
+</div>
+</div>
 
 
     </div>
@@ -268,7 +267,7 @@ <h2>Build and fit a classifier<a class="headerlink" href="#build-and-fit-a-class
       
     </p>
     <p>
-        &copy; Copyright 2014-2021, Machine Learning Professorship Freiburg.<br/>
+        &copy; Copyright 2014-2022, Machine Learning Professorship Freiburg.<br/>
       Created using <a href="http://sphinx-doc.org/">Sphinx</a> 4.2.0.<br/>
     </p>
   </div>
diff --git a/master/examples/40_advanced/example_feature_types.html b/master/examples/40_advanced/example_feature_types.html
index 96117a93f0..b0ba66821c 100644
--- a/master/examples/40_advanced/example_feature_types.html
+++ b/master/examples/40_advanced/example_feature_types.html
@@ -3,9 +3,8 @@
 <html>
   <head>
     <meta charset="utf-8" />
-    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="generator" content="Docutils 0.17.1: http://docutils.sourceforge.net/" />
-
-    <title>Feature Types &#8212; AutoSklearn 0.14.2 documentation</title>
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>Feature Types &#8212; AutoSklearn 0.14.4 documentation</title>
     <link rel="stylesheet" type="text/css" href="../../_static/pygments.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/bootstrap-sphinx.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/sg_gallery.css" />
@@ -55,7 +54,7 @@
         </button>
         <a class="navbar-brand" href="../../index.html">
           auto-sklearn</a>
-        <span class="navbar-text navbar-version pull-left"><b>0.14.2</b></span>
+        <span class="navbar-text navbar-version pull-left"><b>0.14.4</b></span>
       </div>
 
         <div class="collapse navbar-collapse nav-collapse">
@@ -124,7 +123,7 @@
 <p>Click <a class="reference internal" href="#sphx-glr-download-examples-40-advanced-example-feature-types-py"><span class="std std-ref">here</span></a>
 to download the full example code or to run this example in your browser via Binder</p>
 </div>
-<section class="sphx-glr-example-title" id="feature-types">
+<div class="sphx-glr-example-title section" id="feature-types">
 <span id="sphx-glr-examples-40-advanced-example-feature-types-py"></span><h1>Feature Types<a class="headerlink" href="#feature-types" title="Permalink to this headline">¶</a></h1>
 <p>In <em>auto-sklearn</em> it is possible to specify the feature types of a dataset when calling the method
 <a class="reference internal" href="../../api.html#autosklearn.classification.AutoSklearnClassifier.fit" title="autosklearn.classification.AutoSklearnClassifier.fit"><code class="xref py py-meth docutils literal notranslate"><span class="pre">fit()</span></code></a> by specifying the argument
@@ -141,7 +140,7 @@
 <span class="kn">import</span> <span class="nn">autosklearn.classification</span>
 </pre></div>
 </div>
-<section id="data-loading">
+<div class="section" id="data-loading">
 <h2>Data Loading<a class="headerlink" href="#data-loading" title="Permalink to this headline">¶</a></h2>
 <p>Load Australian dataset from <a class="reference external" href="https://www.openml.org/d/40981">https://www.openml.org/d/40981</a></p>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">bunch</span> <span class="o">=</span> <span class="n">data</span> <span class="o">=</span> <span class="n">sklearn</span><span class="o">.</span><span class="n">datasets</span><span class="o">.</span><span class="n">fetch_openml</span><span class="p">(</span><span class="n">data_id</span><span class="o">=</span><span class="mi">40981</span><span class="p">,</span> <span class="n">as_frame</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
@@ -168,8 +167,8 @@ <h2>Data Loading<a class="headerlink" href="#data-loading" title="Permalink to t
   X = bunch[&#39;data&#39;].to_numpy(np.float)
 </pre></div>
 </div>
-</section>
-<section id="build-and-fit-a-classifier">
+</div>
+<div class="section" id="build-and-fit-a-classifier">
 <h2>Build and fit a classifier<a class="headerlink" href="#build-and-fit-a-classifier" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="bp">cls</span> <span class="o">=</span> <span class="n">autosklearn</span><span class="o">.</span><span class="n">classification</span><span class="o">.</span><span class="n">AutoSklearnClassifier</span><span class="p">(</span>
     <span class="n">time_left_for_this_task</span><span class="o">=</span><span class="mi">30</span><span class="p">,</span>
@@ -188,8 +187,8 @@ <h2>Build and fit a classifier<a class="headerlink" href="#build-and-fit-a-class
                       time_left_for_this_task=30)
 </pre></div>
 </div>
-</section>
-<section id="get-the-score-of-the-final-ensemble">
+</div>
+<div class="section" id="get-the-score-of-the-final-ensemble">
 <h2>Get the Score of the final ensemble<a class="headerlink" href="#get-the-score-of-the-final-ensemble" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">predictions</span> <span class="o">=</span> <span class="bp">cls</span><span class="o">.</span><span class="n">predict</span><span class="p">(</span><span class="n">X_test</span><span class="p">)</span>
 <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Accuracy score&quot;</span><span class="p">,</span> <span class="n">sklearn</span><span class="o">.</span><span class="n">metrics</span><span class="o">.</span><span class="n">accuracy_score</span><span class="p">(</span><span class="n">y_test</span><span class="p">,</span> <span class="n">predictions</span><span class="p">))</span>
@@ -199,7 +198,7 @@ <h2>Get the Score of the final ensemble<a class="headerlink" href="#get-the-scor
 <div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Accuracy score 0.8786127167630058
 </pre></div>
 </div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 0 minutes  10.620 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 0 minutes  8.763 seconds)</p>
 <div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-examples-40-advanced-example-feature-types-py">
 <div class="binder-badge docutils container">
 <a class="reference external image-reference" href="https://mybinder.org/v2/gh/automl/auto-sklearn/master?urlpath=lab/tree/notebooks/examples/40_advanced/example_feature_types.ipynb"><img alt="Launch binder" src="../../_images/binder_badge_logo1.svg" width="150px" /></a>
@@ -212,8 +211,8 @@ <h2>Get the Score of the final ensemble<a class="headerlink" href="#get-the-scor
 </div>
 </div>
 <p class="sphx-glr-signature"><a class="reference external" href="https://sphinx-gallery.github.io">Gallery generated by Sphinx-Gallery</a></p>
-</section>
-</section>
+</div>
+</div>
 
 
     </div>
@@ -234,7 +233,7 @@ <h2>Get the Score of the final ensemble<a class="headerlink" href="#get-the-scor
       
     </p>
     <p>
-        &copy; Copyright 2014-2021, Machine Learning Professorship Freiburg.<br/>
+        &copy; Copyright 2014-2022, Machine Learning Professorship Freiburg.<br/>
       Created using <a href="http://sphinx-doc.org/">Sphinx</a> 4.2.0.<br/>
     </p>
   </div>
diff --git a/master/examples/40_advanced/example_get_pipeline_components.html b/master/examples/40_advanced/example_get_pipeline_components.html
index 262c59d75c..c44ab836d7 100644
--- a/master/examples/40_advanced/example_get_pipeline_components.html
+++ b/master/examples/40_advanced/example_get_pipeline_components.html
@@ -3,9 +3,8 @@
 <html>
   <head>
     <meta charset="utf-8" />
-    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="generator" content="Docutils 0.17.1: http://docutils.sourceforge.net/" />
-
-    <title>Obtain run information &#8212; AutoSklearn 0.14.2 documentation</title>
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>Obtain run information &#8212; AutoSklearn 0.14.4 documentation</title>
     <link rel="stylesheet" type="text/css" href="../../_static/pygments.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/bootstrap-sphinx.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/sg_gallery.css" />
@@ -55,7 +54,7 @@
         </button>
         <a class="navbar-brand" href="../../index.html">
           auto-sklearn</a>
-        <span class="navbar-text navbar-version pull-left"><b>0.14.2</b></span>
+        <span class="navbar-text navbar-version pull-left"><b>0.14.4</b></span>
       </div>
 
         <div class="collapse navbar-collapse nav-collapse">
@@ -129,7 +128,7 @@
 <p>Click <a class="reference internal" href="#sphx-glr-download-examples-40-advanced-example-get-pipeline-components-py"><span class="std std-ref">here</span></a>
 to download the full example code or to run this example in your browser via Binder</p>
 </div>
-<section class="sphx-glr-example-title" id="obtain-run-information">
+<div class="sphx-glr-example-title section" id="obtain-run-information">
 <span id="sphx-glr-examples-40-advanced-example-get-pipeline-components-py"></span><h1>Obtain run information<a class="headerlink" href="#obtain-run-information" title="Permalink to this headline">¶</a></h1>
 <p>The following example shows how to obtain information from a finished
 Auto-sklearn run. In particular, it shows:
@@ -139,21 +138,23 @@
 <p>Auto-sklearn is a wrapper on top of
 the sklearn models. This example illustrates how to interact
 with the sklearn components directly, in this case a PCA preprocessor.</p>
-<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">sklearn.datasets</span>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">pprint</span> <span class="kn">import</span> <span class="n">pprint</span>
+
+<span class="kn">import</span> <span class="nn">sklearn.datasets</span>
 <span class="kn">import</span> <span class="nn">sklearn.metrics</span>
 
 <span class="kn">import</span> <span class="nn">autosklearn.classification</span>
 </pre></div>
 </div>
-<section id="data-loading">
+<div class="section" id="data-loading">
 <h2>Data Loading<a class="headerlink" href="#data-loading" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">X</span><span class="p">,</span> <span class="n">y</span> <span class="o">=</span> <span class="n">sklearn</span><span class="o">.</span><span class="n">datasets</span><span class="o">.</span><span class="n">load_breast_cancer</span><span class="p">(</span><span class="n">return_X_y</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
 <span class="n">X_train</span><span class="p">,</span> <span class="n">X_test</span><span class="p">,</span> <span class="n">y_train</span><span class="p">,</span> <span class="n">y_test</span> <span class="o">=</span> \
     <span class="n">sklearn</span><span class="o">.</span><span class="n">model_selection</span><span class="o">.</span><span class="n">train_test_split</span><span class="p">(</span><span class="n">X</span><span class="p">,</span> <span class="n">y</span><span class="p">,</span> <span class="n">random_state</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
 </pre></div>
 </div>
-</section>
-<section id="build-and-fit-the-classifier">
+</div>
+<div class="section" id="build-and-fit-the-classifier">
 <h2>Build and fit the classifier<a class="headerlink" href="#build-and-fit-the-classifier" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">automl</span> <span class="o">=</span> <span class="n">autosklearn</span><span class="o">.</span><span class="n">classification</span><span class="o">.</span><span class="n">AutoSklearnClassifier</span><span class="p">(</span>
     <span class="n">time_left_for_this_task</span><span class="o">=</span><span class="mi">30</span><span class="p">,</span>
@@ -169,12 +170,17 @@ <h2>Build and fit the classifier<a class="headerlink" href="#build-and-fit-the-c
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>AutoSklearnClassifier(include={&#39;feature_preprocessor&#39;: [&#39;pca&#39;]},
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:68: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+  self.metafeatures = self.metafeatures.append(metafeatures)
+/home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:72: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+  self.algorithm_runs[metric].append(runs)
+
+AutoSklearnClassifier(include={&#39;feature_preprocessor&#39;: [&#39;pca&#39;]},
                       per_run_time_limit=10, time_left_for_this_task=30)
 </pre></div>
 </div>
-</section>
-<section id="predict-using-the-model">
+</div>
+<div class="section" id="predict-using-the-model">
 <h2>Predict using the model<a class="headerlink" href="#predict-using-the-model" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">predictions</span> <span class="o">=</span> <span class="n">automl</span><span class="o">.</span><span class="n">predict</span><span class="p">(</span><span class="n">X_test</span><span class="p">)</span>
 <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Accuracy score:</span><span class="si">{}</span><span class="s2">&quot;</span><span class="o">.</span><span class="n">format</span><span class="p">(</span>
@@ -183,91 +189,132 @@ <h2>Predict using the model<a class="headerlink" href="#predict-using-the-model"
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Accuracy score:0.951048951048951
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Accuracy score:0.958041958041958
 </pre></div>
 </div>
-</section>
-<section id="report-the-models-found-by-auto-sklearn">
+</div>
+<div class="section" id="report-the-models-found-by-auto-sklearn">
 <h2>Report the models found by Auto-Sklearn<a class="headerlink" href="#report-the-models-found-by-auto-sklearn" title="Permalink to this headline">¶</a></h2>
 <p>Auto-sklearn uses
 <a class="reference external" href="https://www.cs.cornell.edu/~alexn/papers/shotgun.icml04.revised.rev2.pdf">Ensemble Selection</a>
 to construct ensembles in a post-hoc fashion. The ensemble is a linear
 weighting of all models constructed during the hyperparameter optimization.
-This prints the final ensemble. It is a list of tuples, each tuple being
-the model weight in the ensemble and the model itself.</p>
-<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="nb">print</span><span class="p">(</span><span class="n">automl</span><span class="o">.</span><span class="n">show_models</span><span class="p">())</span>
+This prints the final ensemble. It is a dictionary where <code class="docutils literal notranslate"><span class="pre">model_id</span></code> of
+each model is a key, and value is a dictionary containing information
+of that model. A model’s dict contains its <code class="docutils literal notranslate"><span class="pre">'model_id'</span></code>, <code class="docutils literal notranslate"><span class="pre">'rank'</span></code>,
+<code class="docutils literal notranslate"><span class="pre">'cost'</span></code>, <code class="docutils literal notranslate"><span class="pre">'ensemble_weight'</span></code>, and the model itself. The model is
+given by the <code class="docutils literal notranslate"><span class="pre">'data_preprocessor'</span></code>, <code class="docutils literal notranslate"><span class="pre">'feature_preprocessor'</span></code>,
+<code class="docutils literal notranslate"><span class="pre">'regressor'/'classifier'</span></code> and <code class="docutils literal notranslate"><span class="pre">'sklearn_regressor'/'sklearn_classifier'</span></code>
+entries. But for the <code class="docutils literal notranslate"><span class="pre">'cv'</span></code> resampling strategy, the same for each cv
+model is stored in the <code class="docutils literal notranslate"><span class="pre">'estimators'</span></code> list in the dict, along with the
+<code class="docutils literal notranslate"><span class="pre">'voting_model'</span></code>.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">pprint</span><span class="p">(</span><span class="n">automl</span><span class="o">.</span><span class="n">show_models</span><span class="p">(),</span> <span class="n">indent</span><span class="o">=</span><span class="mi">4</span><span class="p">)</span>
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>[(0.340000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;sgd&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;pca&#39;, &#39;classifier:sgd:alpha&#39;: 1.0333981795670597e-07, &#39;classifier:sgd:average&#39;: &#39;False&#39;, &#39;classifier:sgd:fit_intercept&#39;: &#39;True&#39;, &#39;classifier:sgd:learning_rate&#39;: &#39;constant&#39;, &#39;classifier:sgd:loss&#39;: &#39;modified_huber&#39;, &#39;classifier:sgd:penalty&#39;: &#39;l2&#39;, &#39;classifier:sgd:tol&#39;: 2.5846457091661748e-05, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;minmax&#39;, &#39;feature_preprocessor:pca:keep_variance&#39;: 0.9755563101648113, &#39;feature_preprocessor:pca:whiten&#39;: &#39;True&#39;, &#39;classifier:sgd:epsilon&#39;: 0.03457277874230573, &#39;classifier:sgd:eta0&#39;: 0.004635634934629576, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.00015788887334393744},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.160000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;mlp&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;pca&#39;, &#39;classifier:mlp:activation&#39;: &#39;tanh&#39;, &#39;classifier:mlp:alpha&#39;: 1.103855734598575e-05, &#39;classifier:mlp:batch_size&#39;: &#39;auto&#39;, &#39;classifier:mlp:beta_1&#39;: 0.9, &#39;classifier:mlp:beta_2&#39;: 0.999, &#39;classifier:mlp:early_stopping&#39;: &#39;valid&#39;, &#39;classifier:mlp:epsilon&#39;: 1e-08, &#39;classifier:mlp:hidden_layer_depth&#39;: 3, &#39;classifier:mlp:learning_rate_init&#39;: 0.00014375616988222174, &#39;classifier:mlp:n_iter_no_change&#39;: 32, &#39;classifier:mlp:num_nodes_per_layer&#39;: 229, &#39;classifier:mlp:shuffle&#39;: &#39;True&#39;, &#39;classifier:mlp:solver&#39;: &#39;adam&#39;, &#39;classifier:mlp:tol&#39;: 0.0001, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;quantile_transformer&#39;, &#39;feature_preprocessor:pca:keep_variance&#39;: 0.7895711479212801, &#39;feature_preprocessor:pca:whiten&#39;: &#39;True&#39;, &#39;classifier:mlp:validation_fraction&#39;: 0.1, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.002766772136115771, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles&#39;: 180, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution&#39;: &#39;uniform&#39;},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.140000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;liblinear_svc&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;pca&#39;, &#39;classifier:liblinear_svc:C&#39;: 10.369811497206404, &#39;classifier:liblinear_svc:dual&#39;: &#39;False&#39;, &#39;classifier:liblinear_svc:fit_intercept&#39;: &#39;True&#39;, &#39;classifier:liblinear_svc:intercept_scaling&#39;: 1, &#39;classifier:liblinear_svc:loss&#39;: &#39;squared_hinge&#39;, &#39;classifier:liblinear_svc:multi_class&#39;: &#39;ovr&#39;, &#39;classifier:liblinear_svc:penalty&#39;: &#39;l2&#39;, &#39;classifier:liblinear_svc:tol&#39;: 0.0015130257264171173, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;power_transformer&#39;, &#39;feature_preprocessor:pca:keep_variance&#39;: 0.6661824659281315, &#39;feature_preprocessor:pca:whiten&#39;: &#39;False&#39;},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.120000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;random_forest&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;pca&#39;, &#39;classifier:random_forest:bootstrap&#39;: &#39;True&#39;, &#39;classifier:random_forest:criterion&#39;: &#39;gini&#39;, &#39;classifier:random_forest:max_depth&#39;: &#39;None&#39;, &#39;classifier:random_forest:max_features&#39;: 0.9331254454871041, &#39;classifier:random_forest:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:random_forest:min_impurity_decrease&#39;: 0.0, &#39;classifier:random_forest:min_samples_leaf&#39;: 2, &#39;classifier:random_forest:min_samples_split&#39;: 20, &#39;classifier:random_forest:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;none&#39;, &#39;feature_preprocessor:pca:keep_variance&#39;: 0.9967857433838874, &#39;feature_preprocessor:pca:whiten&#39;: &#39;False&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.005069923784737444},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.120000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;random_forest&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;pca&#39;, &#39;classifier:random_forest:bootstrap&#39;: &#39;True&#39;, &#39;classifier:random_forest:criterion&#39;: &#39;gini&#39;, &#39;classifier:random_forest:max_depth&#39;: &#39;None&#39;, &#39;classifier:random_forest:max_features&#39;: 0.5, &#39;classifier:random_forest:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:random_forest:min_impurity_decrease&#39;: 0.0, &#39;classifier:random_forest:min_samples_leaf&#39;: 1, &#39;classifier:random_forest:min_samples_split&#39;: 2, &#39;classifier:random_forest:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;standardize&#39;, &#39;feature_preprocessor:pca:keep_variance&#39;: 0.9999, &#39;feature_preprocessor:pca:whiten&#39;: &#39;False&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.01},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.040000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;decision_tree&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;pca&#39;, &#39;classifier:decision_tree:criterion&#39;: &#39;gini&#39;, &#39;classifier:decision_tree:max_depth_factor&#39;: 0.14069368736662313, &#39;classifier:decision_tree:max_features&#39;: 1.0, &#39;classifier:decision_tree:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:decision_tree:min_impurity_decrease&#39;: 0.0, &#39;classifier:decision_tree:min_samples_leaf&#39;: 5, &#39;classifier:decision_tree:min_samples_split&#39;: 19, &#39;classifier:decision_tree:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;quantile_transformer&#39;, &#39;feature_preprocessor:pca:keep_variance&#39;: 0.7772788812704434, &#39;feature_preprocessor:pca:whiten&#39;: &#39;True&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles&#39;: 526, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution&#39;: &#39;uniform&#39;},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.040000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;k_nearest_neighbors&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;pca&#39;, &#39;classifier:k_nearest_neighbors:n_neighbors&#39;: 4, &#39;classifier:k_nearest_neighbors:p&#39;: 2, &#39;classifier:k_nearest_neighbors:weights&#39;: &#39;distance&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;normalize&#39;, &#39;feature_preprocessor:pca:keep_variance&#39;: 0.8047274080856589, &#39;feature_preprocessor:pca:whiten&#39;: &#39;False&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.008015420020402715},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.040000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;libsvm_svc&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;pca&#39;, &#39;classifier:libsvm_svc:C&#39;: 100.5905006626969, &#39;classifier:libsvm_svc:gamma&#39;: 0.011333066835975528, &#39;classifier:libsvm_svc:kernel&#39;: &#39;poly&#39;, &#39;classifier:libsvm_svc:max_iter&#39;: -1, &#39;classifier:libsvm_svc:shrinking&#39;: &#39;True&#39;, &#39;classifier:libsvm_svc:tol&#39;: 0.012391313886912093, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;minmax&#39;, &#39;feature_preprocessor:pca:keep_variance&#39;: 0.9290439925152777, &#39;feature_preprocessor:pca:whiten&#39;: &#39;False&#39;, &#39;classifier:libsvm_svc:coef0&#39;: 0.08087614244138486, &#39;classifier:libsvm_svc:degree&#39;: 3, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.010000000000000004},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-]
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>{   2: {   &#39;balancing&#39;: Balancing(random_state=1),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78b929ac0&gt;,
+           &#39;cost&#39;: 0.07801418439716312,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78ca86640&gt;,
+           &#39;ensemble_weight&#39;: 0.04,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc790382760&gt;,
+           &#39;model_id&#39;: 2,
+           &#39;rank&#39;: 7,
+           &#39;sklearn_classifier&#39;: RandomForestClassifier(max_features=5, n_estimators=512, n_jobs=1,
+                       random_state=1, warm_start=True)},
+    3: {   &#39;balancing&#39;: Balancing(random_state=1),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc786b361f0&gt;,
+           &#39;cost&#39;: 0.07092198581560283,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78a826520&gt;,
+           &#39;ensemble_weight&#39;: 0.18,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc786b36910&gt;,
+           &#39;model_id&#39;: 3,
+           &#39;rank&#39;: 6,
+           &#39;sklearn_classifier&#39;: RandomForestClassifier(max_features=1, min_samples_leaf=2, min_samples_split=20,
+                       n_estimators=512, n_jobs=1, random_state=1,
+                       warm_start=True)},
+    4: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc7901c5820&gt;,
+           &#39;cost&#39;: 0.028368794326241176,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7901c53d0&gt;,
+           &#39;ensemble_weight&#39;: 0.08,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc79032af10&gt;,
+           &#39;model_id&#39;: 4,
+           &#39;rank&#39;: 1,
+           &#39;sklearn_classifier&#39;: MLPClassifier(activation=&#39;tanh&#39;, alpha=1.103855734598575e-05, beta_1=0.999,
+              beta_2=0.9, early_stopping=True,
+              hidden_layer_sizes=(229, 229, 229),
+              learning_rate_init=0.00014375616988222174, max_iter=32,
+              n_iter_no_change=32, random_state=1, verbose=0, warm_start=True)},
+    5: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78c9441c0&gt;,
+           &#39;cost&#39;: 0.1063829787234043,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78a5eb790&gt;,
+           &#39;ensemble_weight&#39;: 0.04,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78f8d1970&gt;,
+           &#39;model_id&#39;: 5,
+           &#39;rank&#39;: 8,
+           &#39;sklearn_classifier&#39;: KNeighborsClassifier(n_neighbors=4, weights=&#39;distance&#39;)},
+    6: {   &#39;balancing&#39;: Balancing(random_state=1),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78c90c6d0&gt;,
+           &#39;cost&#39;: 0.11347517730496459,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78b289160&gt;,
+           &#39;ensemble_weight&#39;: 0.16,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c90c400&gt;,
+           &#39;model_id&#39;: 6,
+           &#39;rank&#39;: 9,
+           &#39;sklearn_classifier&#39;: SVC(C=100.5905006626969, cache_size=1799.9375, coef0=0.08087614244138486,
+    gamma=0.011333066835975528, kernel=&#39;poly&#39;, max_iter=-1.0, random_state=1,
+    tol=0.012391313886912093)},
+    7: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78a7e35b0&gt;,
+           &#39;cost&#39;: 0.03546099290780147,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7903465e0&gt;,
+           &#39;ensemble_weight&#39;: 0.16,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc790169d30&gt;,
+           &#39;model_id&#39;: 7,
+           &#39;rank&#39;: 3,
+           &#39;sklearn_classifier&#39;: LinearSVC(C=10.369811497206404, class_weight=&#39;balanced&#39;, dual=False,
+          intercept_scaling=1.0, random_state=1, tol=0.0015130257264171173)},
+    8: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78bcf7dc0&gt;,
+           &#39;cost&#39;: 0.05673758865248224,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78a892370&gt;,
+           &#39;ensemble_weight&#39;: 0.02,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78bcf7910&gt;,
+           &#39;model_id&#39;: 8,
+           &#39;rank&#39;: 4,
+           &#39;sklearn_classifier&#39;: DecisionTreeClassifier(class_weight=&#39;balanced&#39;, max_depth=1, min_samples_leaf=5,
+                       min_samples_split=19, random_state=1)},
+    9: {   &#39;balancing&#39;: Balancing(random_state=1),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78a5d3d60&gt;,
+           &#39;cost&#39;: 0.028368794326241176,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78a593c70&gt;,
+           &#39;ensemble_weight&#39;: 0.26,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc793fbc9d0&gt;,
+           &#39;model_id&#39;: 9,
+           &#39;rank&#39;: 2,
+           &#39;sklearn_classifier&#39;: SGDClassifier(alpha=1.0333981795670635e-07, epsilon=0.03457277874230573,
+              eta0=0.004635634934629583, learning_rate=&#39;constant&#39;,
+              loss=&#39;modified_huber&#39;, max_iter=128, random_state=1,
+              tol=2.5846457091661748e-05, warm_start=True)},
+    10: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc786687c10&gt;,
+            &#39;cost&#39;: 0.05673758865248224,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78c94dc10&gt;,
+            &#39;ensemble_weight&#39;: 0.06,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc7903bdbb0&gt;,
+            &#39;model_id&#39;: 10,
+            &#39;rank&#39;: 5,
+            &#39;sklearn_classifier&#39;: SGDClassifier(alpha=0.0006046872822704108, epsilon=0.0006085470273398056,
+              eta0=0.0003474820560182061, l1_ratio=4.5783995454757295e-09,
+              learning_rate=&#39;invscaling&#39;, loss=&#39;modified_huber&#39;, max_iter=1024,
+              penalty=&#39;elasticnet&#39;, power_t=0.6064541566557635, random_state=1,
+              tol=1.7244397014112777e-05, warm_start=True)}}
 </pre></div>
 </div>
-</section>
-<section id="report-statistics-about-the-search">
+</div>
+<div class="section" id="report-statistics-about-the-search">
 <h2>Report statistics about the search<a class="headerlink" href="#report-statistics-about-the-search" title="Permalink to this headline">¶</a></h2>
 <p>Print statistics about the auto-sklearn run such as number of
 iterations, number of models failed with a time out etc.</p>
@@ -279,15 +326,15 @@ <h2>Report statistics about the search<a class="headerlink" href="#report-statis
   Dataset name: breast_cancer
   Metric: accuracy
   Best validation score: 0.971631
-  Number of target algorithm runs: 8
-  Number of successful target algorithm runs: 8
+  Number of target algorithm runs: 9
+  Number of successful target algorithm runs: 9
   Number of crashed target algorithm runs: 0
   Number of target algorithms that exceeded the time limit: 0
   Number of target algorithms that exceeded the memory limit: 0
 </pre></div>
 </div>
-</section>
-<section id="detailed-statistics-about-the-search-part-1">
+</div>
+<div class="section" id="detailed-statistics-about-the-search-part-1">
 <h2>Detailed statistics about the search - part 1<a class="headerlink" href="#detailed-statistics-about-the-search-part-1" title="Permalink to this headline">¶</a></h2>
 <p>Auto-sklearn also keeps detailed statistics of the hyperparameter
 optimization procedurce, which are stored in a so-called
@@ -296,7 +343,7 @@ <h2>Detailed statistics about the search - part 1<a class="headerlink" href="#de
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>&lt;smac.runhistory.runhistory.RunHistory object at 0x7f9379a40f10&gt;
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>&lt;smac.runhistory.runhistory.RunHistory object at 0x7fc78ab3f670&gt;
 </pre></div>
 </div>
 <p>Runs are stored inside an <code class="docutils literal notranslate"><span class="pre">OrderedDict</span></code> called <code class="docutils literal notranslate"><span class="pre">data</span></code>:</p>
@@ -304,7 +351,7 @@ <h2>Detailed statistics about the search - part 1<a class="headerlink" href="#de
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>9
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>10
 </pre></div>
 </div>
 <p>Let’s iterative over all entries</p>
@@ -317,31 +364,34 @@ <h2>Detailed statistics about the search - part 1<a class="headerlink" href="#de
 <p class="sphx-glr-script-out">Out:</p>
 <div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>#########
 RunKey(config_id=1, instance_id=&#39;{&quot;task_id&quot;: &quot;breast_cancer&quot;}&#39;, seed=0, budget=0.0)
-RunValue(cost=0.07801418439716312, time=1.9545562267303467, status=&lt;StatusType.SUCCESS: 1&gt;, starttime=1636488092.7797232, endtime=1636488094.7563245, additional_info={&#39;duration&#39;: 1.8572731018066406, &#39;num_run&#39;: 2, &#39;train_loss&#39;: 0.0, &#39;configuration_origin&#39;: &#39;Initial design&#39;})
+RunValue(cost=0.07801418439716312, time=1.5860536098480225, status=&lt;StatusType.SUCCESS: 1&gt;, starttime=1643128262.388968, endtime=1643128263.9952888, additional_info={&#39;duration&#39;: 1.4848618507385254, &#39;num_run&#39;: 2, &#39;train_loss&#39;: 0.0, &#39;configuration_origin&#39;: &#39;Initial design&#39;})
 #########
 RunKey(config_id=2, instance_id=&#39;{&quot;task_id&quot;: &quot;breast_cancer&quot;}&#39;, seed=0, budget=0.0)
-RunValue(cost=0.07092198581560283, time=1.6941378116607666, status=&lt;StatusType.SUCCESS: 1&gt;, starttime=1636488094.8219044, endtime=1636488096.537016, additional_info={&#39;duration&#39;: 1.6041510105133057, &#39;num_run&#39;: 3, &#39;train_loss&#39;: 0.06315789473684208, &#39;configuration_origin&#39;: &#39;Initial design&#39;})
+RunValue(cost=0.07092198581560283, time=1.3667099475860596, status=&lt;StatusType.SUCCESS: 1&gt;, starttime=1643128264.0590467, endtime=1643128265.4495397, additional_info={&#39;duration&#39;: 1.2673461437225342, &#39;num_run&#39;: 3, &#39;train_loss&#39;: 0.06315789473684208, &#39;configuration_origin&#39;: &#39;Initial design&#39;})
 #########
 RunKey(config_id=3, instance_id=&#39;{&quot;task_id&quot;: &quot;breast_cancer&quot;}&#39;, seed=0, budget=0.0)
-RunValue(cost=0.028368794326241176, time=1.6140191555023193, status=&lt;StatusType.SUCCESS: 1&gt;, starttime=1636488096.6401973, endtime=1636488098.2758095, additional_info={&#39;duration&#39;: 1.5358099937438965, &#39;num_run&#39;: 4, &#39;train_loss&#39;: 0.04210526315789476, &#39;configuration_origin&#39;: &#39;Initial design&#39;})
+RunValue(cost=0.028368794326241176, time=1.1778380870819092, status=&lt;StatusType.SUCCESS: 1&gt;, starttime=1643128265.5568986, endtime=1643128266.7576518, additional_info={&#39;duration&#39;: 1.1020641326904297, &#39;num_run&#39;: 4, &#39;train_loss&#39;: 0.04210526315789476, &#39;configuration_origin&#39;: &#39;Initial design&#39;})
 #########
 RunKey(config_id=4, instance_id=&#39;{&quot;task_id&quot;: &quot;breast_cancer&quot;}&#39;, seed=0, budget=0.0)
-RunValue(cost=0.1063829787234043, time=0.6702926158905029, status=&lt;StatusType.SUCCESS: 1&gt;, starttime=1636488098.4037113, endtime=1636488099.0938587, additional_info={&#39;duration&#39;: 0.6066544055938721, &#39;num_run&#39;: 5, &#39;train_loss&#39;: 0.0, &#39;configuration_origin&#39;: &#39;Initial design&#39;})
+RunValue(cost=0.1063829787234043, time=0.6564128398895264, status=&lt;StatusType.SUCCESS: 1&gt;, starttime=1643128266.868573, endtime=1643128267.5475628, additional_info={&#39;duration&#39;: 0.5852375030517578, &#39;num_run&#39;: 5, &#39;train_loss&#39;: 0.0, &#39;configuration_origin&#39;: &#39;Initial design&#39;})
 #########
 RunKey(config_id=5, instance_id=&#39;{&quot;task_id&quot;: &quot;breast_cancer&quot;}&#39;, seed=0, budget=0.0)
-RunValue(cost=0.11347517730496459, time=0.5629768371582031, status=&lt;StatusType.SUCCESS: 1&gt;, starttime=1636488099.2329674, endtime=1636488099.8163307, additional_info={&#39;duration&#39;: 0.5027651786804199, &#39;num_run&#39;: 6, &#39;train_loss&#39;: 0.09122807017543855, &#39;configuration_origin&#39;: &#39;Initial design&#39;})
+RunValue(cost=0.11347517730496459, time=0.5501728057861328, status=&lt;StatusType.SUCCESS: 1&gt;, starttime=1643128267.676605, endtime=1643128268.2494493, additional_info={&#39;duration&#39;: 0.4777560234069824, &#39;num_run&#39;: 6, &#39;train_loss&#39;: 0.09122807017543855, &#39;configuration_origin&#39;: &#39;Initial design&#39;})
 #########
 RunKey(config_id=6, instance_id=&#39;{&quot;task_id&quot;: &quot;breast_cancer&quot;}&#39;, seed=0, budget=0.0)
-RunValue(cost=0.03546099290780147, time=0.9639365673065186, status=&lt;StatusType.SUCCESS: 1&gt;, starttime=1636488102.2641013, endtime=1636488103.2531207, additional_info={&#39;duration&#39;: 0.8704240322113037, &#39;num_run&#39;: 7, &#39;train_loss&#39;: 0.04561403508771933, &#39;configuration_origin&#39;: &#39;Random Search (sorted)&#39;})
+RunValue(cost=0.03546099290780147, time=0.7203934192657471, status=&lt;StatusType.SUCCESS: 1&gt;, starttime=1643128270.32824, endtime=1643128271.0703733, additional_info={&#39;duration&#39;: 0.6468265056610107, &#39;num_run&#39;: 7, &#39;train_loss&#39;: 0.04561403508771933, &#39;configuration_origin&#39;: &#39;Random Search (sorted)&#39;})
 #########
 RunKey(config_id=7, instance_id=&#39;{&quot;task_id&quot;: &quot;breast_cancer&quot;}&#39;, seed=0, budget=0.0)
-RunValue(cost=0.05673758865248224, time=0.8888447284698486, status=&lt;StatusType.SUCCESS: 1&gt;, starttime=1636488105.9350605, endtime=1636488106.851598, additional_info={&#39;duration&#39;: 0.7990398406982422, &#39;num_run&#39;: 8, &#39;train_loss&#39;: 0.06315789473684208, &#39;configuration_origin&#39;: &#39;Random Search (sorted)&#39;})
+RunValue(cost=0.05673758865248224, time=0.6808576583862305, status=&lt;StatusType.SUCCESS: 1&gt;, starttime=1643128273.2836225, endtime=1643128273.9862375, additional_info={&#39;duration&#39;: 0.6066226959228516, &#39;num_run&#39;: 8, &#39;train_loss&#39;: 0.06315789473684208, &#39;configuration_origin&#39;: &#39;Random Search (sorted)&#39;})
 #########
 RunKey(config_id=8, instance_id=&#39;{&quot;task_id&quot;: &quot;breast_cancer&quot;}&#39;, seed=0, budget=0.0)
-RunValue(cost=0.028368794326241176, time=0.7385828495025635, status=&lt;StatusType.SUCCESS: 1&gt;, starttime=1636488107.1076975, endtime=1636488107.8694334, additional_info={&#39;duration&#39;: 0.6549544334411621, &#39;num_run&#39;: 9, &#39;train_loss&#39;: 0.007017543859649145, &#39;configuration_origin&#39;: &#39;Random Search&#39;})
+RunValue(cost=0.028368794326241176, time=0.5615131855010986, status=&lt;StatusType.SUCCESS: 1&gt;, starttime=1643128274.1657884, endtime=1643128274.7493172, additional_info={&#39;duration&#39;: 0.48226284980773926, &#39;num_run&#39;: 9, &#39;train_loss&#39;: 0.007017543859649145, &#39;configuration_origin&#39;: &#39;Random Search&#39;})
 #########
 RunKey(config_id=9, instance_id=&#39;{&quot;task_id&quot;: &quot;breast_cancer&quot;}&#39;, seed=0, budget=0.0)
-RunValue(cost=1.0, time=0.0, status=&lt;StatusType.STOP: 8&gt;, starttime=1636488110.8355012, endtime=1636488110.8355012, additional_info={})
+RunValue(cost=0.05673758865248224, time=0.6582047939300537, status=&lt;StatusType.SUCCESS: 1&gt;, starttime=1643128277.5765781, endtime=1643128278.254898, additional_info={&#39;duration&#39;: 0.5808424949645996, &#39;num_run&#39;: 10, &#39;train_loss&#39;: 0.06666666666666665, &#39;configuration_origin&#39;: &#39;Random Search (sorted)&#39;})
+#########
+RunKey(config_id=10, instance_id=&#39;{&quot;task_id&quot;: &quot;breast_cancer&quot;}&#39;, seed=0, budget=0.0)
+RunValue(cost=1.0, time=0.0, status=&lt;StatusType.STOP: 8&gt;, starttime=1643128280.7154517, endtime=1643128280.715452, additional_info={})
 </pre></div>
 </div>
 <p>and have a detailed look at one entry:</p>
@@ -408,11 +458,11 @@ <h2>Detailed statistics about the search - part 1<a class="headerlink" href="#de
 </div>
 <p class="sphx-glr-script-out">Out:</p>
 <div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Cost: 0.07801418439716312
-Time: 1.9545562267303467
+Time: 1.5860536098480225
 Status: StatusType.SUCCESS
-Additional information: {&#39;duration&#39;: 1.8572731018066406, &#39;num_run&#39;: 2, &#39;train_loss&#39;: 0.0, &#39;configuration_origin&#39;: &#39;Initial design&#39;}
-Start time: 1636488092.7797232
-End time 1636488094.7563245
+Additional information: {&#39;duration&#39;: 1.4848618507385254, &#39;num_run&#39;: 2, &#39;train_loss&#39;: 0.0, &#39;configuration_origin&#39;: &#39;Initial design&#39;}
+Start time: 1643128262.388968
+End time 1643128263.9952888
 </pre></div>
 </div>
 <p>Cost is basically the same as a loss. In case the metric to optimize for
@@ -469,8 +519,8 @@ <h2>Detailed statistics about the search - part 1<a class="headerlink" href="#de
   feature_preprocessor:pca:whiten, Value: &#39;True&#39;
 </pre></div>
 </div>
-</section>
-<section id="detailed-statistics-about-the-search-part-2">
+</div>
+<div class="section" id="detailed-statistics-about-the-search-part-2">
 <h2>Detailed statistics about the search - part 2<a class="headerlink" href="#detailed-statistics-about-the-search-part-2" title="Permalink to this headline">¶</a></h2>
 <p>To maintain compatibility with scikit-learn, Auto-sklearn gives the
 same data as
@@ -480,331 +530,449 @@ <h2>Detailed statistics about the search - part 2<a class="headerlink" href="#de
 </div>
 <p class="sphx-glr-script-out">Out:</p>
 <div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>{&#39;mean_test_score&#39;: array([0.92198582, 0.92907801, 0.97163121, 0.89361702, 0.88652482,
-       0.96453901, 0.94326241, 0.97163121]), &#39;mean_fit_time&#39;: array([1.95455623, 1.69413781, 1.61401916, 0.67029262, 0.56297684,
-       0.96393657, 0.88884473, 0.73858285]), &#39;params&#39;: [{&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;random_forest&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;pca&#39;, &#39;classifier:random_forest:bootstrap&#39;: &#39;True&#39;, &#39;classifier:random_forest:criterion&#39;: &#39;gini&#39;, &#39;classifier:random_forest:max_depth&#39;: &#39;None&#39;, &#39;classifier:random_forest:max_features&#39;: 0.5, &#39;classifier:random_forest:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:random_forest:min_impurity_decrease&#39;: 0.0, &#39;classifier:random_forest:min_samples_leaf&#39;: 1, &#39;classifier:random_forest:min_samples_split&#39;: 2, &#39;classifier:random_forest:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;standardize&#39;, &#39;feature_preprocessor:pca:keep_variance&#39;: 0.9999, &#39;feature_preprocessor:pca:whiten&#39;: &#39;False&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.01}, {&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;random_forest&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;pca&#39;, &#39;classifier:random_forest:bootstrap&#39;: &#39;True&#39;, &#39;classifier:random_forest:criterion&#39;: &#39;gini&#39;, &#39;classifier:random_forest:max_depth&#39;: &#39;None&#39;, &#39;classifier:random_forest:max_features&#39;: 0.9331254454871041, &#39;classifier:random_forest:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:random_forest:min_impurity_decrease&#39;: 0.0, &#39;classifier:random_forest:min_samples_leaf&#39;: 2, &#39;classifier:random_forest:min_samples_split&#39;: 20, &#39;classifier:random_forest:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;none&#39;, &#39;feature_preprocessor:pca:keep_variance&#39;: 0.9967857433838874, &#39;feature_preprocessor:pca:whiten&#39;: &#39;False&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.005069923784737444}, {&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;mlp&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;pca&#39;, &#39;classifier:mlp:activation&#39;: &#39;tanh&#39;, &#39;classifier:mlp:alpha&#39;: 1.103855734598575e-05, &#39;classifier:mlp:batch_size&#39;: &#39;auto&#39;, &#39;classifier:mlp:beta_1&#39;: 0.9, &#39;classifier:mlp:beta_2&#39;: 0.999, &#39;classifier:mlp:early_stopping&#39;: &#39;valid&#39;, &#39;classifier:mlp:epsilon&#39;: 1e-08, &#39;classifier:mlp:hidden_layer_depth&#39;: 3, &#39;classifier:mlp:learning_rate_init&#39;: 0.00014375616988222174, &#39;classifier:mlp:n_iter_no_change&#39;: 32, &#39;classifier:mlp:num_nodes_per_layer&#39;: 229, &#39;classifier:mlp:shuffle&#39;: &#39;True&#39;, &#39;classifier:mlp:solver&#39;: &#39;adam&#39;, &#39;classifier:mlp:tol&#39;: 0.0001, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;quantile_transformer&#39;, &#39;feature_preprocessor:pca:keep_variance&#39;: 0.7895711479212801, &#39;feature_preprocessor:pca:whiten&#39;: &#39;True&#39;, &#39;classifier:mlp:validation_fraction&#39;: 0.1, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.002766772136115771, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles&#39;: 180, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution&#39;: &#39;uniform&#39;}, {&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;k_nearest_neighbors&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;pca&#39;, &#39;classifier:k_nearest_neighbors:n_neighbors&#39;: 4, &#39;classifier:k_nearest_neighbors:p&#39;: 2, &#39;classifier:k_nearest_neighbors:weights&#39;: &#39;distance&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;normalize&#39;, &#39;feature_preprocessor:pca:keep_variance&#39;: 0.8047274080856589, &#39;feature_preprocessor:pca:whiten&#39;: &#39;False&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.008015420020402715}, {&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;libsvm_svc&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;pca&#39;, &#39;classifier:libsvm_svc:C&#39;: 100.5905006626969, &#39;classifier:libsvm_svc:gamma&#39;: 0.011333066835975528, &#39;classifier:libsvm_svc:kernel&#39;: &#39;poly&#39;, &#39;classifier:libsvm_svc:max_iter&#39;: -1, &#39;classifier:libsvm_svc:shrinking&#39;: &#39;True&#39;, &#39;classifier:libsvm_svc:tol&#39;: 0.012391313886912093, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;minmax&#39;, &#39;feature_preprocessor:pca:keep_variance&#39;: 0.9290439925152777, &#39;feature_preprocessor:pca:whiten&#39;: &#39;False&#39;, &#39;classifier:libsvm_svc:coef0&#39;: 0.08087614244138486, &#39;classifier:libsvm_svc:degree&#39;: 3, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.010000000000000004}, {&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;liblinear_svc&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;pca&#39;, &#39;classifier:liblinear_svc:C&#39;: 10.369811497206404, &#39;classifier:liblinear_svc:dual&#39;: &#39;False&#39;, &#39;classifier:liblinear_svc:fit_intercept&#39;: &#39;True&#39;, &#39;classifier:liblinear_svc:intercept_scaling&#39;: 1, &#39;classifier:liblinear_svc:loss&#39;: &#39;squared_hinge&#39;, &#39;classifier:liblinear_svc:multi_class&#39;: &#39;ovr&#39;, &#39;classifier:liblinear_svc:penalty&#39;: &#39;l2&#39;, &#39;classifier:liblinear_svc:tol&#39;: 0.0015130257264171173, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;power_transformer&#39;, &#39;feature_preprocessor:pca:keep_variance&#39;: 0.6661824659281315, &#39;feature_preprocessor:pca:whiten&#39;: &#39;False&#39;}, {&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;decision_tree&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;pca&#39;, &#39;classifier:decision_tree:criterion&#39;: &#39;gini&#39;, &#39;classifier:decision_tree:max_depth_factor&#39;: 0.14069368736662313, &#39;classifier:decision_tree:max_features&#39;: 1.0, &#39;classifier:decision_tree:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:decision_tree:min_impurity_decrease&#39;: 0.0, &#39;classifier:decision_tree:min_samples_leaf&#39;: 5, &#39;classifier:decision_tree:min_samples_split&#39;: 19, &#39;classifier:decision_tree:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;quantile_transformer&#39;, &#39;feature_preprocessor:pca:keep_variance&#39;: 0.7772788812704434, &#39;feature_preprocessor:pca:whiten&#39;: &#39;True&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles&#39;: 526, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution&#39;: &#39;uniform&#39;}, {&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;sgd&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;pca&#39;, &#39;classifier:sgd:alpha&#39;: 1.0333981795670597e-07, &#39;classifier:sgd:average&#39;: &#39;False&#39;, &#39;classifier:sgd:fit_intercept&#39;: &#39;True&#39;, &#39;classifier:sgd:learning_rate&#39;: &#39;constant&#39;, &#39;classifier:sgd:loss&#39;: &#39;modified_huber&#39;, &#39;classifier:sgd:penalty&#39;: &#39;l2&#39;, &#39;classifier:sgd:tol&#39;: 2.5846457091661748e-05, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;minmax&#39;, &#39;feature_preprocessor:pca:keep_variance&#39;: 0.9755563101648113, &#39;feature_preprocessor:pca:whiten&#39;: &#39;True&#39;, &#39;classifier:sgd:epsilon&#39;: 0.03457277874230573, &#39;classifier:sgd:eta0&#39;: 0.004635634934629576, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.00015788887334393744}], &#39;rank_test_scores&#39;: array([6, 5, 1, 7, 8, 3, 4, 1]), &#39;status&#39;: [&#39;Success&#39;, &#39;Success&#39;, &#39;Success&#39;, &#39;Success&#39;, &#39;Success&#39;, &#39;Success&#39;, &#39;Success&#39;, &#39;Success&#39;], &#39;budgets&#39;: [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0], &#39;param_balancing:strategy&#39;: masked_array(data=[&#39;none&#39;, &#39;none&#39;, &#39;weighting&#39;, &#39;weighting&#39;, &#39;none&#39;,
-                   &#39;weighting&#39;, &#39;weighting&#39;, &#39;none&#39;],
-             mask=[False, False, False, False, False, False, False, False],
+       0.96453901, 0.94326241, 0.97163121, 0.94326241]), &#39;mean_fit_time&#39;: array([1.58605361, 1.36670995, 1.17783809, 0.65641284, 0.55017281,
+       0.72039342, 0.68085766, 0.56151319, 0.65820479]), &#39;params&#39;: [{&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;random_forest&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;pca&#39;, &#39;classifier:random_forest:bootstrap&#39;: &#39;True&#39;, &#39;classifier:random_forest:criterion&#39;: &#39;gini&#39;, &#39;classifier:random_forest:max_depth&#39;: &#39;None&#39;, &#39;classifier:random_forest:max_features&#39;: 0.5, &#39;classifier:random_forest:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:random_forest:min_impurity_decrease&#39;: 0.0, &#39;classifier:random_forest:min_samples_leaf&#39;: 1, &#39;classifier:random_forest:min_samples_split&#39;: 2, &#39;classifier:random_forest:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;standardize&#39;, &#39;feature_preprocessor:pca:keep_variance&#39;: 0.9999, &#39;feature_preprocessor:pca:whiten&#39;: &#39;False&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.01}, {&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;random_forest&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;pca&#39;, &#39;classifier:random_forest:bootstrap&#39;: &#39;True&#39;, &#39;classifier:random_forest:criterion&#39;: &#39;gini&#39;, &#39;classifier:random_forest:max_depth&#39;: &#39;None&#39;, &#39;classifier:random_forest:max_features&#39;: 0.9331254454871041, &#39;classifier:random_forest:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:random_forest:min_impurity_decrease&#39;: 0.0, &#39;classifier:random_forest:min_samples_leaf&#39;: 2, &#39;classifier:random_forest:min_samples_split&#39;: 20, &#39;classifier:random_forest:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;none&#39;, &#39;feature_preprocessor:pca:keep_variance&#39;: 0.9967857433838874, &#39;feature_preprocessor:pca:whiten&#39;: &#39;False&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.005069923784737444}, {&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;mlp&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;pca&#39;, &#39;classifier:mlp:activation&#39;: &#39;tanh&#39;, &#39;classifier:mlp:alpha&#39;: 1.103855734598575e-05, &#39;classifier:mlp:batch_size&#39;: &#39;auto&#39;, &#39;classifier:mlp:beta_1&#39;: 0.9, &#39;classifier:mlp:beta_2&#39;: 0.999, &#39;classifier:mlp:early_stopping&#39;: &#39;valid&#39;, &#39;classifier:mlp:epsilon&#39;: 1e-08, &#39;classifier:mlp:hidden_layer_depth&#39;: 3, &#39;classifier:mlp:learning_rate_init&#39;: 0.00014375616988222174, &#39;classifier:mlp:n_iter_no_change&#39;: 32, &#39;classifier:mlp:num_nodes_per_layer&#39;: 229, &#39;classifier:mlp:shuffle&#39;: &#39;True&#39;, &#39;classifier:mlp:solver&#39;: &#39;adam&#39;, &#39;classifier:mlp:tol&#39;: 0.0001, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;quantile_transformer&#39;, &#39;feature_preprocessor:pca:keep_variance&#39;: 0.7895711479212801, &#39;feature_preprocessor:pca:whiten&#39;: &#39;True&#39;, &#39;classifier:mlp:validation_fraction&#39;: 0.1, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.002766772136115771, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles&#39;: 180, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution&#39;: &#39;uniform&#39;}, {&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;k_nearest_neighbors&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;pca&#39;, &#39;classifier:k_nearest_neighbors:n_neighbors&#39;: 4, &#39;classifier:k_nearest_neighbors:p&#39;: 2, &#39;classifier:k_nearest_neighbors:weights&#39;: &#39;distance&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;normalize&#39;, &#39;feature_preprocessor:pca:keep_variance&#39;: 0.8047274080856589, &#39;feature_preprocessor:pca:whiten&#39;: &#39;False&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.008015420020402715}, {&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;libsvm_svc&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;pca&#39;, &#39;classifier:libsvm_svc:C&#39;: 100.5905006626969, &#39;classifier:libsvm_svc:gamma&#39;: 0.011333066835975528, &#39;classifier:libsvm_svc:kernel&#39;: &#39;poly&#39;, &#39;classifier:libsvm_svc:max_iter&#39;: -1, &#39;classifier:libsvm_svc:shrinking&#39;: &#39;True&#39;, &#39;classifier:libsvm_svc:tol&#39;: 0.012391313886912093, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;minmax&#39;, &#39;feature_preprocessor:pca:keep_variance&#39;: 0.9290439925152777, &#39;feature_preprocessor:pca:whiten&#39;: &#39;False&#39;, &#39;classifier:libsvm_svc:coef0&#39;: 0.08087614244138486, &#39;classifier:libsvm_svc:degree&#39;: 3, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.010000000000000004}, {&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;liblinear_svc&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;pca&#39;, &#39;classifier:liblinear_svc:C&#39;: 10.369811497206404, &#39;classifier:liblinear_svc:dual&#39;: &#39;False&#39;, &#39;classifier:liblinear_svc:fit_intercept&#39;: &#39;True&#39;, &#39;classifier:liblinear_svc:intercept_scaling&#39;: 1, &#39;classifier:liblinear_svc:loss&#39;: &#39;squared_hinge&#39;, &#39;classifier:liblinear_svc:multi_class&#39;: &#39;ovr&#39;, &#39;classifier:liblinear_svc:penalty&#39;: &#39;l2&#39;, &#39;classifier:liblinear_svc:tol&#39;: 0.0015130257264171173, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;power_transformer&#39;, &#39;feature_preprocessor:pca:keep_variance&#39;: 0.6661824659281315, &#39;feature_preprocessor:pca:whiten&#39;: &#39;False&#39;}, {&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;decision_tree&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;pca&#39;, &#39;classifier:decision_tree:criterion&#39;: &#39;gini&#39;, &#39;classifier:decision_tree:max_depth_factor&#39;: 0.14069368736662313, &#39;classifier:decision_tree:max_features&#39;: 1.0, &#39;classifier:decision_tree:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:decision_tree:min_impurity_decrease&#39;: 0.0, &#39;classifier:decision_tree:min_samples_leaf&#39;: 5, &#39;classifier:decision_tree:min_samples_split&#39;: 19, &#39;classifier:decision_tree:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;quantile_transformer&#39;, &#39;feature_preprocessor:pca:keep_variance&#39;: 0.7772788812704434, &#39;feature_preprocessor:pca:whiten&#39;: &#39;True&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles&#39;: 526, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution&#39;: &#39;uniform&#39;}, {&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;sgd&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;pca&#39;, &#39;classifier:sgd:alpha&#39;: 1.0333981795670635e-07, &#39;classifier:sgd:average&#39;: &#39;False&#39;, &#39;classifier:sgd:fit_intercept&#39;: &#39;True&#39;, &#39;classifier:sgd:learning_rate&#39;: &#39;constant&#39;, &#39;classifier:sgd:loss&#39;: &#39;modified_huber&#39;, &#39;classifier:sgd:penalty&#39;: &#39;l2&#39;, &#39;classifier:sgd:tol&#39;: 2.5846457091661748e-05, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;minmax&#39;, &#39;feature_preprocessor:pca:keep_variance&#39;: 0.9755563101648113, &#39;feature_preprocessor:pca:whiten&#39;: &#39;True&#39;, &#39;classifier:sgd:epsilon&#39;: 0.03457277874230573, &#39;classifier:sgd:eta0&#39;: 0.004635634934629583, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.00015788887334393714}, {&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;sgd&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;pca&#39;, &#39;classifier:sgd:alpha&#39;: 0.0006046872822704108, &#39;classifier:sgd:average&#39;: &#39;False&#39;, &#39;classifier:sgd:fit_intercept&#39;: &#39;True&#39;, &#39;classifier:sgd:learning_rate&#39;: &#39;invscaling&#39;, &#39;classifier:sgd:loss&#39;: &#39;modified_huber&#39;, &#39;classifier:sgd:penalty&#39;: &#39;elasticnet&#39;, &#39;classifier:sgd:tol&#39;: 1.7244397014112777e-05, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;power_transformer&#39;, &#39;feature_preprocessor:pca:keep_variance&#39;: 0.9941379232526211, &#39;feature_preprocessor:pca:whiten&#39;: &#39;False&#39;, &#39;classifier:sgd:epsilon&#39;: 0.0006085470273398056, &#39;classifier:sgd:eta0&#39;: 0.0003474820560182061, &#39;classifier:sgd:l1_ratio&#39;: 4.5783995454757295e-09, &#39;classifier:sgd:power_t&#39;: 0.6064541566557635}], &#39;rank_test_scores&#39;: array([7, 6, 1, 8, 9, 3, 4, 1, 4]), &#39;status&#39;: [&#39;Success&#39;, &#39;Success&#39;, &#39;Success&#39;, &#39;Success&#39;, &#39;Success&#39;, &#39;Success&#39;, &#39;Success&#39;, &#39;Success&#39;, &#39;Success&#39;], &#39;budgets&#39;: [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0], &#39;param_balancing:strategy&#39;: masked_array(data=[&#39;none&#39;, &#39;none&#39;, &#39;weighting&#39;, &#39;weighting&#39;, &#39;none&#39;,
+                   &#39;weighting&#39;, &#39;weighting&#39;, &#39;none&#39;, &#39;none&#39;],
+             mask=[False, False, False, False, False, False, False, False,
+                   False],
        fill_value=&#39;N/A&#39;,
             dtype=&#39;&lt;U9&#39;), &#39;param_classifier:__choice__&#39;: masked_array(data=[&#39;random_forest&#39;, &#39;random_forest&#39;, &#39;mlp&#39;,
                    &#39;k_nearest_neighbors&#39;, &#39;libsvm_svc&#39;, &#39;liblinear_svc&#39;,
-                   &#39;decision_tree&#39;, &#39;sgd&#39;],
-             mask=[False, False, False, False, False, False, False, False],
+                   &#39;decision_tree&#39;, &#39;sgd&#39;, &#39;sgd&#39;],
+             mask=[False, False, False, False, False, False, False, False,
+                   False],
        fill_value=&#39;N/A&#39;,
             dtype=&#39;&lt;U19&#39;), &#39;param_data_preprocessor:__choice__&#39;: masked_array(data=[&#39;feature_type&#39;, &#39;feature_type&#39;, &#39;feature_type&#39;,
                    &#39;feature_type&#39;, &#39;feature_type&#39;, &#39;feature_type&#39;,
-                   &#39;feature_type&#39;, &#39;feature_type&#39;],
-             mask=[False, False, False, False, False, False, False, False],
+                   &#39;feature_type&#39;, &#39;feature_type&#39;, &#39;feature_type&#39;],
+             mask=[False, False, False, False, False, False, False, False,
+                   False],
        fill_value=&#39;N/A&#39;,
-            dtype=&#39;&lt;U12&#39;), &#39;param_feature_preprocessor:__choice__&#39;: masked_array(data=[&#39;pca&#39;, &#39;pca&#39;, &#39;pca&#39;, &#39;pca&#39;, &#39;pca&#39;, &#39;pca&#39;, &#39;pca&#39;, &#39;pca&#39;],
-             mask=[False, False, False, False, False, False, False, False],
+            dtype=&#39;&lt;U12&#39;), &#39;param_feature_preprocessor:__choice__&#39;: masked_array(data=[&#39;pca&#39;, &#39;pca&#39;, &#39;pca&#39;, &#39;pca&#39;, &#39;pca&#39;, &#39;pca&#39;, &#39;pca&#39;, &#39;pca&#39;,
+                   &#39;pca&#39;],
+             mask=[False, False, False, False, False, False, False, False,
+                   False],
        fill_value=&#39;N/A&#39;,
-            dtype=&#39;&lt;U3&#39;), &#39;param_classifier:adaboost:algorithm&#39;: masked_array(data=[--, --, --, --, --, --, --, --],
-             mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+            dtype=&#39;&lt;U3&#39;), &#39;param_classifier:adaboost:algorithm&#39;: masked_array(data=[--, --, --, --, --, --, --, --, --],
+             mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                    True],
        fill_value=1e+20,
-            dtype=float64), &#39;param_classifier:adaboost:learning_rate&#39;: masked_array(data=[--, --, --, --, --, --, --, --],
-             mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+            dtype=float64), &#39;param_classifier:adaboost:learning_rate&#39;: masked_array(data=[--, --, --, --, --, --, --, --, --],
+             mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                    True],
        fill_value=1e+20,
-            dtype=float64), &#39;param_classifier:adaboost:max_depth&#39;: masked_array(data=[--, --, --, --, --, --, --, --],
-             mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+            dtype=float64), &#39;param_classifier:adaboost:max_depth&#39;: masked_array(data=[--, --, --, --, --, --, --, --, --],
+             mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                    True],
        fill_value=1e+20,
-            dtype=float64), &#39;param_classifier:adaboost:n_estimators&#39;: masked_array(data=[--, --, --, --, --, --, --, --],
-             mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+            dtype=float64), &#39;param_classifier:adaboost:n_estimators&#39;: masked_array(data=[--, --, --, --, --, --, --, --, --],
+             mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                    True],
        fill_value=1e+20,
-            dtype=float64), &#39;param_classifier:bernoulli_nb:alpha&#39;: masked_array(data=[--, --, --, --, --, --, --, --],
-             mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+            dtype=float64), &#39;param_classifier:bernoulli_nb:alpha&#39;: masked_array(data=[--, --, --, --, --, --, --, --, --],
+             mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                    True],
        fill_value=1e+20,
-            dtype=float64), &#39;param_classifier:bernoulli_nb:fit_prior&#39;: masked_array(data=[--, --, --, --, --, --, --, --],
-             mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+            dtype=float64), &#39;param_classifier:bernoulli_nb:fit_prior&#39;: masked_array(data=[--, --, --, --, --, --, --, --, --],
+             mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                    True],
        fill_value=1e+20,
-            dtype=float64), &#39;param_classifier:decision_tree:criterion&#39;: masked_array(data=[--, --, --, --, --, --, &#39;gini&#39;, --],
-             mask=[ True,  True,  True,  True,  True,  True, False,  True],
+            dtype=float64), &#39;param_classifier:decision_tree:criterion&#39;: masked_array(data=[--, --, --, --, --, --, &#39;gini&#39;, --, --],
+             mask=[ True,  True,  True,  True,  True,  True, False,  True,
+                    True],
        fill_value=&#39;N/A&#39;,
-            dtype=&#39;&lt;U32&#39;), &#39;param_classifier:decision_tree:max_depth_factor&#39;: masked_array(data=[--, --, --, --, --, --, 0.14069368736662313, --],
-             mask=[ True,  True,  True,  True,  True,  True, False,  True],
-       fill_value=1e+20), &#39;param_classifier:decision_tree:max_features&#39;: masked_array(data=[--, --, --, --, --, --, 1.0, --],
-             mask=[ True,  True,  True,  True,  True,  True, False,  True],
-       fill_value=1e+20), &#39;param_classifier:decision_tree:max_leaf_nodes&#39;: masked_array(data=[--, --, --, --, --, --, &#39;None&#39;, --],
-             mask=[ True,  True,  True,  True,  True,  True, False,  True],
+            dtype=&#39;&lt;U32&#39;), &#39;param_classifier:decision_tree:max_depth_factor&#39;: masked_array(data=[--, --, --, --, --, --, 0.14069368736662313, --, --],
+             mask=[ True,  True,  True,  True,  True,  True, False,  True,
+                    True],
+       fill_value=1e+20), &#39;param_classifier:decision_tree:max_features&#39;: masked_array(data=[--, --, --, --, --, --, 1.0, --, --],
+             mask=[ True,  True,  True,  True,  True,  True, False,  True,
+                    True],
+       fill_value=1e+20), &#39;param_classifier:decision_tree:max_leaf_nodes&#39;: masked_array(data=[--, --, --, --, --, --, &#39;None&#39;, --, --],
+             mask=[ True,  True,  True,  True,  True,  True, False,  True,
+                    True],
        fill_value=&#39;N/A&#39;,
-            dtype=&#39;&lt;U32&#39;), &#39;param_classifier:decision_tree:min_impurity_decrease&#39;: masked_array(data=[--, --, --, --, --, --, 0.0, --],
-             mask=[ True,  True,  True,  True,  True,  True, False,  True],
-       fill_value=1e+20), &#39;param_classifier:decision_tree:min_samples_leaf&#39;: masked_array(data=[--, --, --, --, --, --, 5.0, --],
-             mask=[ True,  True,  True,  True,  True,  True, False,  True],
-       fill_value=1e+20), &#39;param_classifier:decision_tree:min_samples_split&#39;: masked_array(data=[--, --, --, --, --, --, 19.0, --],
-             mask=[ True,  True,  True,  True,  True,  True, False,  True],
-       fill_value=1e+20), &#39;param_classifier:decision_tree:min_weight_fraction_leaf&#39;: masked_array(data=[--, --, --, --, --, --, 0.0, --],
-             mask=[ True,  True,  True,  True,  True,  True, False,  True],
-       fill_value=1e+20), &#39;param_classifier:extra_trees:bootstrap&#39;: masked_array(data=[--, --, --, --, --, --, --, --],
-             mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+            dtype=&#39;&lt;U32&#39;), &#39;param_classifier:decision_tree:min_impurity_decrease&#39;: masked_array(data=[--, --, --, --, --, --, 0.0, --, --],
+             mask=[ True,  True,  True,  True,  True,  True, False,  True,
+                    True],
+       fill_value=1e+20), &#39;param_classifier:decision_tree:min_samples_leaf&#39;: masked_array(data=[--, --, --, --, --, --, 5.0, --, --],
+             mask=[ True,  True,  True,  True,  True,  True, False,  True,
+                    True],
+       fill_value=1e+20), &#39;param_classifier:decision_tree:min_samples_split&#39;: masked_array(data=[--, --, --, --, --, --, 19.0, --, --],
+             mask=[ True,  True,  True,  True,  True,  True, False,  True,
+                    True],
+       fill_value=1e+20), &#39;param_classifier:decision_tree:min_weight_fraction_leaf&#39;: masked_array(data=[--, --, --, --, --, --, 0.0, --, --],
+             mask=[ True,  True,  True,  True,  True,  True, False,  True,
+                    True],
+       fill_value=1e+20), &#39;param_classifier:extra_trees:bootstrap&#39;: masked_array(data=[--, --, --, --, --, --, --, --, --],
+             mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                    True],
+       fill_value=1e+20,
+            dtype=float64), &#39;param_classifier:extra_trees:criterion&#39;: masked_array(data=[--, --, --, --, --, --, --, --, --],
+             mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                    True],
+       fill_value=1e+20,
+            dtype=float64), &#39;param_classifier:extra_trees:max_depth&#39;: masked_array(data=[--, --, --, --, --, --, --, --, --],
+             mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                    True],
        fill_value=1e+20,
-            dtype=float64), &#39;param_classifier:extra_trees:criterion&#39;: masked_array(data=[--, --, --, --, --, --, --, --],
-             mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+            dtype=float64), &#39;param_classifier:extra_trees:max_features&#39;: masked_array(data=[--, --, --, --, --, --, --, --, --],
+             mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                    True],
        fill_value=1e+20,
-            dtype=float64), &#39;param_classifier:extra_trees:max_depth&#39;: masked_array(data=[--, --, --, --, --, --, --, --],
-             mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+            dtype=float64), &#39;param_classifier:extra_trees:max_leaf_nodes&#39;: masked_array(data=[--, --, --, --, --, --, --, --, --],
+             mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                    True],
        fill_value=1e+20,
-            dtype=float64), &#39;param_classifier:extra_trees:max_features&#39;: masked_array(data=[--, --, --, --, --, --, --, --],
-             mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+            dtype=float64), &#39;param_classifier:extra_trees:min_impurity_decrease&#39;: masked_array(data=[--, --, --, --, --, --, --, --, --],
+             mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                    True],
        fill_value=1e+20,
-            dtype=float64), &#39;param_classifier:extra_trees:max_leaf_nodes&#39;: masked_array(data=[--, --, --, --, --, --, --, --],
-             mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+            dtype=float64), &#39;param_classifier:extra_trees:min_samples_leaf&#39;: masked_array(data=[--, --, --, --, --, --, --, --, --],
+             mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                    True],
        fill_value=1e+20,
-            dtype=float64), &#39;param_classifier:extra_trees:min_impurity_decrease&#39;: masked_array(data=[--, --, --, --, --, --, --, --],
-             mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+            dtype=float64), &#39;param_classifier:extra_trees:min_samples_split&#39;: masked_array(data=[--, --, --, --, --, --, --, --, --],
+             mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                    True],
        fill_value=1e+20,
-            dtype=float64), &#39;param_classifier:extra_trees:min_samples_leaf&#39;: masked_array(data=[--, --, --, --, --, --, --, --],
-             mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+            dtype=float64), &#39;param_classifier:extra_trees:min_weight_fraction_leaf&#39;: masked_array(data=[--, --, --, --, --, --, --, --, --],
+             mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                    True],
        fill_value=1e+20,
-            dtype=float64), &#39;param_classifier:extra_trees:min_samples_split&#39;: masked_array(data=[--, --, --, --, --, --, --, --],
-             mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+            dtype=float64), &#39;param_classifier:gradient_boosting:early_stop&#39;: masked_array(data=[--, --, --, --, --, --, --, --, --],
+             mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                    True],
        fill_value=1e+20,
-            dtype=float64), &#39;param_classifier:extra_trees:min_weight_fraction_leaf&#39;: masked_array(data=[--, --, --, --, --, --, --, --],
-             mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+            dtype=float64), &#39;param_classifier:gradient_boosting:l2_regularization&#39;: masked_array(data=[--, --, --, --, --, --, --, --, --],
+             mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                    True],
        fill_value=1e+20,
-            dtype=float64), &#39;param_classifier:gradient_boosting:early_stop&#39;: masked_array(data=[--, --, --, --, --, --, --, --],
-             mask=[ True,  True,  True,  True,  True,  True,  True,  True],
-       fill_value=1e+20,
-            dtype=float64), &#39;param_classifier:gradient_boosting:l2_regularization&#39;: masked_array(data=[--, --, --, --, --, --, --, --],
-             mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+            dtype=float64), &#39;param_classifier:gradient_boosting:learning_rate&#39;: masked_array(data=[--, --, --, --, --, --, --, --, --],
+             mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                    True],
        fill_value=1e+20,
-            dtype=float64), &#39;param_classifier:gradient_boosting:learning_rate&#39;: masked_array(data=[--, --, --, --, --, --, --, --],
-             mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+            dtype=float64), &#39;param_classifier:gradient_boosting:loss&#39;: masked_array(data=[--, --, --, --, --, --, --, --, --],
+             mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                    True],
        fill_value=1e+20,
-            dtype=float64), &#39;param_classifier:gradient_boosting:loss&#39;: masked_array(data=[--, --, --, --, --, --, --, --],
-             mask=[ True,  True,  True,  True,  True,  True,  True,  True],
-       fill_value=1e+20,
-            dtype=float64), &#39;param_classifier:gradient_boosting:max_bins&#39;: masked_array(data=[--, --, --, --, --, --, --, --],
-             mask=[ True,  True,  True,  True,  True,  True,  True,  True],
-       fill_value=1e+20,
-            dtype=float64), &#39;param_classifier:gradient_boosting:max_depth&#39;: masked_array(data=[--, --, --, --, --, --, --, --],
-             mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+            dtype=float64), &#39;param_classifier:gradient_boosting:max_bins&#39;: masked_array(data=[--, --, --, --, --, --, --, --, --],
+             mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                    True],
        fill_value=1e+20,
-            dtype=float64), &#39;param_classifier:gradient_boosting:max_leaf_nodes&#39;: masked_array(data=[--, --, --, --, --, --, --, --],
-             mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+            dtype=float64), &#39;param_classifier:gradient_boosting:max_depth&#39;: masked_array(data=[--, --, --, --, --, --, --, --, --],
+             mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                    True],
        fill_value=1e+20,
-            dtype=float64), &#39;param_classifier:gradient_boosting:min_samples_leaf&#39;: masked_array(data=[--, --, --, --, --, --, --, --],
-             mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+            dtype=float64), &#39;param_classifier:gradient_boosting:max_leaf_nodes&#39;: masked_array(data=[--, --, --, --, --, --, --, --, --],
+             mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                    True],
        fill_value=1e+20,
-            dtype=float64), &#39;param_classifier:gradient_boosting:scoring&#39;: masked_array(data=[--, --, --, --, --, --, --, --],
-             mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+            dtype=float64), &#39;param_classifier:gradient_boosting:min_samples_leaf&#39;: masked_array(data=[--, --, --, --, --, --, --, --, --],
+             mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                    True],
        fill_value=1e+20,
-            dtype=float64), &#39;param_classifier:gradient_boosting:tol&#39;: masked_array(data=[--, --, --, --, --, --, --, --],
-             mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+            dtype=float64), &#39;param_classifier:gradient_boosting:scoring&#39;: masked_array(data=[--, --, --, --, --, --, --, --, --],
+             mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                    True],
        fill_value=1e+20,
-            dtype=float64), &#39;param_classifier:k_nearest_neighbors:n_neighbors&#39;: masked_array(data=[--, --, --, 4.0, --, --, --, --],
-             mask=[ True,  True,  True, False,  True,  True,  True,  True],
-       fill_value=1e+20), &#39;param_classifier:k_nearest_neighbors:p&#39;: masked_array(data=[--, --, --, 2.0, --, --, --, --],
-             mask=[ True,  True,  True, False,  True,  True,  True,  True],
-       fill_value=1e+20), &#39;param_classifier:k_nearest_neighbors:weights&#39;: masked_array(data=[--, --, --, &#39;distance&#39;, --, --, --, --],
-             mask=[ True,  True,  True, False,  True,  True,  True,  True],
+            dtype=float64), &#39;param_classifier:gradient_boosting:tol&#39;: masked_array(data=[--, --, --, --, --, --, --, --, --],
+             mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                    True],
+       fill_value=1e+20,
+            dtype=float64), &#39;param_classifier:k_nearest_neighbors:n_neighbors&#39;: masked_array(data=[--, --, --, 4.0, --, --, --, --, --],
+             mask=[ True,  True,  True, False,  True,  True,  True,  True,
+                    True],
+       fill_value=1e+20), &#39;param_classifier:k_nearest_neighbors:p&#39;: masked_array(data=[--, --, --, 2.0, --, --, --, --, --],
+             mask=[ True,  True,  True, False,  True,  True,  True,  True,
+                    True],
+       fill_value=1e+20), &#39;param_classifier:k_nearest_neighbors:weights&#39;: masked_array(data=[--, --, --, &#39;distance&#39;, --, --, --, --, --],
+             mask=[ True,  True,  True, False,  True,  True,  True,  True,
+                    True],
        fill_value=&#39;N/A&#39;,
-            dtype=&#39;&lt;U32&#39;), &#39;param_classifier:lda:shrinkage&#39;: masked_array(data=[--, --, --, --, --, --, --, --],
-             mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+            dtype=&#39;&lt;U32&#39;), &#39;param_classifier:lda:shrinkage&#39;: masked_array(data=[--, --, --, --, --, --, --, --, --],
+             mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                    True],
        fill_value=1e+20,
-            dtype=float64), &#39;param_classifier:lda:tol&#39;: masked_array(data=[--, --, --, --, --, --, --, --],
-             mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+            dtype=float64), &#39;param_classifier:lda:tol&#39;: masked_array(data=[--, --, --, --, --, --, --, --, --],
+             mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                    True],
        fill_value=1e+20,
-            dtype=float64), &#39;param_classifier:liblinear_svc:C&#39;: masked_array(data=[--, --, --, --, --, 10.369811497206404, --, --],
-             mask=[ True,  True,  True,  True,  True, False,  True,  True],
-       fill_value=1e+20), &#39;param_classifier:liblinear_svc:dual&#39;: masked_array(data=[--, --, --, --, --, &#39;False&#39;, --, --],
-             mask=[ True,  True,  True,  True,  True, False,  True,  True],
-       fill_value=&#39;N/A&#39;,
-            dtype=&#39;&lt;U32&#39;), &#39;param_classifier:liblinear_svc:fit_intercept&#39;: masked_array(data=[--, --, --, --, --, &#39;True&#39;, --, --],
-             mask=[ True,  True,  True,  True,  True, False,  True,  True],
-       fill_value=&#39;N/A&#39;,
-            dtype=&#39;&lt;U32&#39;), &#39;param_classifier:liblinear_svc:intercept_scaling&#39;: masked_array(data=[--, --, --, --, --, 1.0, --, --],
-             mask=[ True,  True,  True,  True,  True, False,  True,  True],
-       fill_value=1e+20), &#39;param_classifier:liblinear_svc:loss&#39;: masked_array(data=[--, --, --, --, --, &#39;squared_hinge&#39;, --, --],
-             mask=[ True,  True,  True,  True,  True, False,  True,  True],
-       fill_value=&#39;N/A&#39;,
-            dtype=&#39;&lt;U32&#39;), &#39;param_classifier:liblinear_svc:multi_class&#39;: masked_array(data=[--, --, --, --, --, &#39;ovr&#39;, --, --],
-             mask=[ True,  True,  True,  True,  True, False,  True,  True],
-       fill_value=&#39;N/A&#39;,
-            dtype=&#39;&lt;U32&#39;), &#39;param_classifier:liblinear_svc:penalty&#39;: masked_array(data=[--, --, --, --, --, &#39;l2&#39;, --, --],
-             mask=[ True,  True,  True,  True,  True, False,  True,  True],
-       fill_value=&#39;N/A&#39;,
-            dtype=&#39;&lt;U32&#39;), &#39;param_classifier:liblinear_svc:tol&#39;: masked_array(data=[--, --, --, --, --, 0.0015130257264171173, --, --],
-             mask=[ True,  True,  True,  True,  True, False,  True,  True],
-       fill_value=1e+20), &#39;param_classifier:libsvm_svc:C&#39;: masked_array(data=[--, --, --, --, 100.5905006626969, --, --, --],
-             mask=[ True,  True,  True,  True, False,  True,  True,  True],
-       fill_value=1e+20), &#39;param_classifier:libsvm_svc:gamma&#39;: masked_array(data=[--, --, --, --, 0.011333066835975528, --, --, --],
-             mask=[ True,  True,  True,  True, False,  True,  True,  True],
-       fill_value=1e+20), &#39;param_classifier:libsvm_svc:kernel&#39;: masked_array(data=[--, --, --, --, &#39;poly&#39;, --, --, --],
-             mask=[ True,  True,  True,  True, False,  True,  True,  True],
-       fill_value=&#39;N/A&#39;,
-            dtype=&#39;&lt;U32&#39;), &#39;param_classifier:libsvm_svc:max_iter&#39;: masked_array(data=[--, --, --, --, -1.0, --, --, --],
-             mask=[ True,  True,  True,  True, False,  True,  True,  True],
-       fill_value=1e+20), &#39;param_classifier:libsvm_svc:shrinking&#39;: masked_array(data=[--, --, --, --, &#39;True&#39;, --, --, --],
-             mask=[ True,  True,  True,  True, False,  True,  True,  True],
+            dtype=float64), &#39;param_classifier:liblinear_svc:C&#39;: masked_array(data=[--, --, --, --, --, 10.369811497206404, --, --, --],
+             mask=[ True,  True,  True,  True,  True, False,  True,  True,
+                    True],
+       fill_value=1e+20), &#39;param_classifier:liblinear_svc:dual&#39;: masked_array(data=[--, --, --, --, --, &#39;False&#39;, --, --, --],
+             mask=[ True,  True,  True,  True,  True, False,  True,  True,
+                    True],
+       fill_value=&#39;N/A&#39;,
+            dtype=&#39;&lt;U32&#39;), &#39;param_classifier:liblinear_svc:fit_intercept&#39;: masked_array(data=[--, --, --, --, --, &#39;True&#39;, --, --, --],
+             mask=[ True,  True,  True,  True,  True, False,  True,  True,
+                    True],
+       fill_value=&#39;N/A&#39;,
+            dtype=&#39;&lt;U32&#39;), &#39;param_classifier:liblinear_svc:intercept_scaling&#39;: masked_array(data=[--, --, --, --, --, 1.0, --, --, --],
+             mask=[ True,  True,  True,  True,  True, False,  True,  True,
+                    True],
+       fill_value=1e+20), &#39;param_classifier:liblinear_svc:loss&#39;: masked_array(data=[--, --, --, --, --, &#39;squared_hinge&#39;, --, --, --],
+             mask=[ True,  True,  True,  True,  True, False,  True,  True,
+                    True],
+       fill_value=&#39;N/A&#39;,
+            dtype=&#39;&lt;U32&#39;), &#39;param_classifier:liblinear_svc:multi_class&#39;: masked_array(data=[--, --, --, --, --, &#39;ovr&#39;, --, --, --],
+             mask=[ True,  True,  True,  True,  True, False,  True,  True,
+                    True],
+       fill_value=&#39;N/A&#39;,
+            dtype=&#39;&lt;U32&#39;), &#39;param_classifier:liblinear_svc:penalty&#39;: masked_array(data=[--, --, --, --, --, &#39;l2&#39;, --, --, --],
+             mask=[ True,  True,  True,  True,  True, False,  True,  True,
+                    True],
+       fill_value=&#39;N/A&#39;,
+            dtype=&#39;&lt;U32&#39;), &#39;param_classifier:liblinear_svc:tol&#39;: masked_array(data=[--, --, --, --, --, 0.0015130257264171173, --, --, --],
+             mask=[ True,  True,  True,  True,  True, False,  True,  True,
+                    True],
+       fill_value=1e+20), &#39;param_classifier:libsvm_svc:C&#39;: masked_array(data=[--, --, --, --, 100.5905006626969, --, --, --, --],
+             mask=[ True,  True,  True,  True, False,  True,  True,  True,
+                    True],
+       fill_value=1e+20), &#39;param_classifier:libsvm_svc:gamma&#39;: masked_array(data=[--, --, --, --, 0.011333066835975528, --, --, --, --],
+             mask=[ True,  True,  True,  True, False,  True,  True,  True,
+                    True],
+       fill_value=1e+20), &#39;param_classifier:libsvm_svc:kernel&#39;: masked_array(data=[--, --, --, --, &#39;poly&#39;, --, --, --, --],
+             mask=[ True,  True,  True,  True, False,  True,  True,  True,
+                    True],
        fill_value=&#39;N/A&#39;,
-            dtype=&#39;&lt;U32&#39;), &#39;param_classifier:libsvm_svc:tol&#39;: masked_array(data=[--, --, --, --, 0.012391313886912093, --, --, --],
-             mask=[ True,  True,  True,  True, False,  True,  True,  True],
-       fill_value=1e+20), &#39;param_classifier:mlp:activation&#39;: masked_array(data=[--, --, &#39;tanh&#39;, --, --, --, --, --],
-             mask=[ True,  True, False,  True,  True,  True,  True,  True],
+            dtype=&#39;&lt;U32&#39;), &#39;param_classifier:libsvm_svc:max_iter&#39;: masked_array(data=[--, --, --, --, -1.0, --, --, --, --],
+             mask=[ True,  True,  True,  True, False,  True,  True,  True,
+                    True],
+       fill_value=1e+20), &#39;param_classifier:libsvm_svc:shrinking&#39;: masked_array(data=[--, --, --, --, &#39;True&#39;, --, --, --, --],
+             mask=[ True,  True,  True,  True, False,  True,  True,  True,
+                    True],
        fill_value=&#39;N/A&#39;,
-            dtype=&#39;&lt;U32&#39;), &#39;param_classifier:mlp:alpha&#39;: masked_array(data=[--, --, 1.103855734598575e-05, --, --, --, --, --],
-             mask=[ True,  True, False,  True,  True,  True,  True,  True],
-       fill_value=1e+20), &#39;param_classifier:mlp:batch_size&#39;: masked_array(data=[--, --, &#39;auto&#39;, --, --, --, --, --],
-             mask=[ True,  True, False,  True,  True,  True,  True,  True],
-       fill_value=&#39;N/A&#39;,
-            dtype=&#39;&lt;U32&#39;), &#39;param_classifier:mlp:beta_1&#39;: masked_array(data=[--, --, 0.9, --, --, --, --, --],
-             mask=[ True,  True, False,  True,  True,  True,  True,  True],
-       fill_value=1e+20), &#39;param_classifier:mlp:beta_2&#39;: masked_array(data=[--, --, 0.999, --, --, --, --, --],
-             mask=[ True,  True, False,  True,  True,  True,  True,  True],
-       fill_value=1e+20), &#39;param_classifier:mlp:early_stopping&#39;: masked_array(data=[--, --, &#39;valid&#39;, --, --, --, --, --],
-             mask=[ True,  True, False,  True,  True,  True,  True,  True],
-       fill_value=&#39;N/A&#39;,
-            dtype=&#39;&lt;U32&#39;), &#39;param_classifier:mlp:epsilon&#39;: masked_array(data=[--, --, 1e-08, --, --, --, --, --],
-             mask=[ True,  True, False,  True,  True,  True,  True,  True],
-       fill_value=1e+20), &#39;param_classifier:mlp:hidden_layer_depth&#39;: masked_array(data=[--, --, 3.0, --, --, --, --, --],
-             mask=[ True,  True, False,  True,  True,  True,  True,  True],
-       fill_value=1e+20), &#39;param_classifier:mlp:learning_rate_init&#39;: masked_array(data=[--, --, 0.00014375616988222174, --, --, --, --, --],
-             mask=[ True,  True, False,  True,  True,  True,  True,  True],
-       fill_value=1e+20), &#39;param_classifier:mlp:n_iter_no_change&#39;: masked_array(data=[--, --, 32.0, --, --, --, --, --],
-             mask=[ True,  True, False,  True,  True,  True,  True,  True],
-       fill_value=1e+20), &#39;param_classifier:mlp:num_nodes_per_layer&#39;: masked_array(data=[--, --, 229.0, --, --, --, --, --],
-             mask=[ True,  True, False,  True,  True,  True,  True,  True],
-       fill_value=1e+20), &#39;param_classifier:mlp:shuffle&#39;: masked_array(data=[--, --, &#39;True&#39;, --, --, --, --, --],
-             mask=[ True,  True, False,  True,  True,  True,  True,  True],
-       fill_value=&#39;N/A&#39;,
-            dtype=&#39;&lt;U32&#39;), &#39;param_classifier:mlp:solver&#39;: masked_array(data=[--, --, &#39;adam&#39;, --, --, --, --, --],
-             mask=[ True,  True, False,  True,  True,  True,  True,  True],
-       fill_value=&#39;N/A&#39;,
-            dtype=&#39;&lt;U32&#39;), &#39;param_classifier:mlp:tol&#39;: masked_array(data=[--, --, 0.0001, --, --, --, --, --],
-             mask=[ True,  True, False,  True,  True,  True,  True,  True],
-       fill_value=1e+20), &#39;param_classifier:passive_aggressive:C&#39;: masked_array(data=[--, --, --, --, --, --, --, --],
-             mask=[ True,  True,  True,  True,  True,  True,  True,  True],
-       fill_value=1e+20,
-            dtype=float64), &#39;param_classifier:passive_aggressive:average&#39;: masked_array(data=[--, --, --, --, --, --, --, --],
-             mask=[ True,  True,  True,  True,  True,  True,  True,  True],
-       fill_value=1e+20,
-            dtype=float64), &#39;param_classifier:passive_aggressive:fit_intercept&#39;: masked_array(data=[--, --, --, --, --, --, --, --],
-             mask=[ True,  True,  True,  True,  True,  True,  True,  True],
-       fill_value=1e+20,
-            dtype=float64), &#39;param_classifier:passive_aggressive:loss&#39;: masked_array(data=[--, --, --, --, --, --, --, --],
-             mask=[ True,  True,  True,  True,  True,  True,  True,  True],
-       fill_value=1e+20,
-            dtype=float64), &#39;param_classifier:passive_aggressive:tol&#39;: masked_array(data=[--, --, --, --, --, --, --, --],
-             mask=[ True,  True,  True,  True,  True,  True,  True,  True],
-       fill_value=1e+20,
-            dtype=float64), &#39;param_classifier:qda:reg_param&#39;: masked_array(data=[--, --, --, --, --, --, --, --],
-             mask=[ True,  True,  True,  True,  True,  True,  True,  True],
-       fill_value=1e+20,
-            dtype=float64), &#39;param_classifier:random_forest:bootstrap&#39;: masked_array(data=[&#39;True&#39;, &#39;True&#39;, --, --, --, --, --, --],
-             mask=[False, False,  True,  True,  True,  True,  True,  True],
-       fill_value=&#39;N/A&#39;,
-            dtype=&#39;&lt;U32&#39;), &#39;param_classifier:random_forest:criterion&#39;: masked_array(data=[&#39;gini&#39;, &#39;gini&#39;, --, --, --, --, --, --],
-             mask=[False, False,  True,  True,  True,  True,  True,  True],
-       fill_value=&#39;N/A&#39;,
-            dtype=&#39;&lt;U32&#39;), &#39;param_classifier:random_forest:max_depth&#39;: masked_array(data=[&#39;None&#39;, &#39;None&#39;, --, --, --, --, --, --],
-             mask=[False, False,  True,  True,  True,  True,  True,  True],
-       fill_value=&#39;N/A&#39;,
-            dtype=&#39;&lt;U32&#39;), &#39;param_classifier:random_forest:max_features&#39;: masked_array(data=[0.5, 0.9331254454871041, --, --, --, --, --, --],
-             mask=[False, False,  True,  True,  True,  True,  True,  True],
-       fill_value=1e+20), &#39;param_classifier:random_forest:max_leaf_nodes&#39;: masked_array(data=[&#39;None&#39;, &#39;None&#39;, --, --, --, --, --, --],
-             mask=[False, False,  True,  True,  True,  True,  True,  True],
-       fill_value=&#39;N/A&#39;,
-            dtype=&#39;&lt;U32&#39;), &#39;param_classifier:random_forest:min_impurity_decrease&#39;: masked_array(data=[0.0, 0.0, --, --, --, --, --, --],
-             mask=[False, False,  True,  True,  True,  True,  True,  True],
-       fill_value=1e+20), &#39;param_classifier:random_forest:min_samples_leaf&#39;: masked_array(data=[1.0, 2.0, --, --, --, --, --, --],
-             mask=[False, False,  True,  True,  True,  True,  True,  True],
-       fill_value=1e+20), &#39;param_classifier:random_forest:min_samples_split&#39;: masked_array(data=[2.0, 20.0, --, --, --, --, --, --],
-             mask=[False, False,  True,  True,  True,  True,  True,  True],
-       fill_value=1e+20), &#39;param_classifier:random_forest:min_weight_fraction_leaf&#39;: masked_array(data=[0.0, 0.0, --, --, --, --, --, --],
-             mask=[False, False,  True,  True,  True,  True,  True,  True],
-       fill_value=1e+20), &#39;param_classifier:sgd:alpha&#39;: masked_array(data=[--, --, --, --, --, --, --, 1.0333981795670597e-07],
-             mask=[ True,  True,  True,  True,  True,  True,  True, False],
-       fill_value=1e+20), &#39;param_classifier:sgd:average&#39;: masked_array(data=[--, --, --, --, --, --, --, &#39;False&#39;],
-             mask=[ True,  True,  True,  True,  True,  True,  True, False],
-       fill_value=&#39;N/A&#39;,
-            dtype=&#39;&lt;U32&#39;), &#39;param_classifier:sgd:fit_intercept&#39;: masked_array(data=[--, --, --, --, --, --, --, &#39;True&#39;],
-             mask=[ True,  True,  True,  True,  True,  True,  True, False],
-       fill_value=&#39;N/A&#39;,
-            dtype=&#39;&lt;U32&#39;), &#39;param_classifier:sgd:learning_rate&#39;: masked_array(data=[--, --, --, --, --, --, --, &#39;constant&#39;],
-             mask=[ True,  True,  True,  True,  True,  True,  True, False],
-       fill_value=&#39;N/A&#39;,
-            dtype=&#39;&lt;U32&#39;), &#39;param_classifier:sgd:loss&#39;: masked_array(data=[--, --, --, --, --, --, --, &#39;modified_huber&#39;],
-             mask=[ True,  True,  True,  True,  True,  True,  True, False],
-       fill_value=&#39;N/A&#39;,
-            dtype=&#39;&lt;U32&#39;), &#39;param_classifier:sgd:penalty&#39;: masked_array(data=[--, --, --, --, --, --, --, &#39;l2&#39;],
-             mask=[ True,  True,  True,  True,  True,  True,  True, False],
-       fill_value=&#39;N/A&#39;,
-            dtype=&#39;&lt;U32&#39;), &#39;param_classifier:sgd:tol&#39;: masked_array(data=[--, --, --, --, --, --, --, 2.5846457091661748e-05],
-             mask=[ True,  True,  True,  True,  True,  True,  True, False],
+            dtype=&#39;&lt;U32&#39;), &#39;param_classifier:libsvm_svc:tol&#39;: masked_array(data=[--, --, --, --, 0.012391313886912093, --, --, --, --],
+             mask=[ True,  True,  True,  True, False,  True,  True,  True,
+                    True],
+       fill_value=1e+20), &#39;param_classifier:mlp:activation&#39;: masked_array(data=[--, --, &#39;tanh&#39;, --, --, --, --, --, --],
+             mask=[ True,  True, False,  True,  True,  True,  True,  True,
+                    True],
+       fill_value=&#39;N/A&#39;,
+            dtype=&#39;&lt;U32&#39;), &#39;param_classifier:mlp:alpha&#39;: masked_array(data=[--, --, 1.103855734598575e-05, --, --, --, --, --, --],
+             mask=[ True,  True, False,  True,  True,  True,  True,  True,
+                    True],
+       fill_value=1e+20), &#39;param_classifier:mlp:batch_size&#39;: masked_array(data=[--, --, &#39;auto&#39;, --, --, --, --, --, --],
+             mask=[ True,  True, False,  True,  True,  True,  True,  True,
+                    True],
+       fill_value=&#39;N/A&#39;,
+            dtype=&#39;&lt;U32&#39;), &#39;param_classifier:mlp:beta_1&#39;: masked_array(data=[--, --, 0.9, --, --, --, --, --, --],
+             mask=[ True,  True, False,  True,  True,  True,  True,  True,
+                    True],
+       fill_value=1e+20), &#39;param_classifier:mlp:beta_2&#39;: masked_array(data=[--, --, 0.999, --, --, --, --, --, --],
+             mask=[ True,  True, False,  True,  True,  True,  True,  True,
+                    True],
+       fill_value=1e+20), &#39;param_classifier:mlp:early_stopping&#39;: masked_array(data=[--, --, &#39;valid&#39;, --, --, --, --, --, --],
+             mask=[ True,  True, False,  True,  True,  True,  True,  True,
+                    True],
+       fill_value=&#39;N/A&#39;,
+            dtype=&#39;&lt;U32&#39;), &#39;param_classifier:mlp:epsilon&#39;: masked_array(data=[--, --, 1e-08, --, --, --, --, --, --],
+             mask=[ True,  True, False,  True,  True,  True,  True,  True,
+                    True],
+       fill_value=1e+20), &#39;param_classifier:mlp:hidden_layer_depth&#39;: masked_array(data=[--, --, 3.0, --, --, --, --, --, --],
+             mask=[ True,  True, False,  True,  True,  True,  True,  True,
+                    True],
+       fill_value=1e+20), &#39;param_classifier:mlp:learning_rate_init&#39;: masked_array(data=[--, --, 0.00014375616988222174, --, --, --, --, --, --],
+             mask=[ True,  True, False,  True,  True,  True,  True,  True,
+                    True],
+       fill_value=1e+20), &#39;param_classifier:mlp:n_iter_no_change&#39;: masked_array(data=[--, --, 32.0, --, --, --, --, --, --],
+             mask=[ True,  True, False,  True,  True,  True,  True,  True,
+                    True],
+       fill_value=1e+20), &#39;param_classifier:mlp:num_nodes_per_layer&#39;: masked_array(data=[--, --, 229.0, --, --, --, --, --, --],
+             mask=[ True,  True, False,  True,  True,  True,  True,  True,
+                    True],
+       fill_value=1e+20), &#39;param_classifier:mlp:shuffle&#39;: masked_array(data=[--, --, &#39;True&#39;, --, --, --, --, --, --],
+             mask=[ True,  True, False,  True,  True,  True,  True,  True,
+                    True],
+       fill_value=&#39;N/A&#39;,
+            dtype=&#39;&lt;U32&#39;), &#39;param_classifier:mlp:solver&#39;: masked_array(data=[--, --, &#39;adam&#39;, --, --, --, --, --, --],
+             mask=[ True,  True, False,  True,  True,  True,  True,  True,
+                    True],
+       fill_value=&#39;N/A&#39;,
+            dtype=&#39;&lt;U32&#39;), &#39;param_classifier:mlp:tol&#39;: masked_array(data=[--, --, 0.0001, --, --, --, --, --, --],
+             mask=[ True,  True, False,  True,  True,  True,  True,  True,
+                    True],
+       fill_value=1e+20), &#39;param_classifier:passive_aggressive:C&#39;: masked_array(data=[--, --, --, --, --, --, --, --, --],
+             mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                    True],
+       fill_value=1e+20,
+            dtype=float64), &#39;param_classifier:passive_aggressive:average&#39;: masked_array(data=[--, --, --, --, --, --, --, --, --],
+             mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                    True],
+       fill_value=1e+20,
+            dtype=float64), &#39;param_classifier:passive_aggressive:fit_intercept&#39;: masked_array(data=[--, --, --, --, --, --, --, --, --],
+             mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                    True],
+       fill_value=1e+20,
+            dtype=float64), &#39;param_classifier:passive_aggressive:loss&#39;: masked_array(data=[--, --, --, --, --, --, --, --, --],
+             mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                    True],
+       fill_value=1e+20,
+            dtype=float64), &#39;param_classifier:passive_aggressive:tol&#39;: masked_array(data=[--, --, --, --, --, --, --, --, --],
+             mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                    True],
+       fill_value=1e+20,
+            dtype=float64), &#39;param_classifier:qda:reg_param&#39;: masked_array(data=[--, --, --, --, --, --, --, --, --],
+             mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                    True],
+       fill_value=1e+20,
+            dtype=float64), &#39;param_classifier:random_forest:bootstrap&#39;: masked_array(data=[&#39;True&#39;, &#39;True&#39;, --, --, --, --, --, --, --],
+             mask=[False, False,  True,  True,  True,  True,  True,  True,
+                    True],
+       fill_value=&#39;N/A&#39;,
+            dtype=&#39;&lt;U32&#39;), &#39;param_classifier:random_forest:criterion&#39;: masked_array(data=[&#39;gini&#39;, &#39;gini&#39;, --, --, --, --, --, --, --],
+             mask=[False, False,  True,  True,  True,  True,  True,  True,
+                    True],
+       fill_value=&#39;N/A&#39;,
+            dtype=&#39;&lt;U32&#39;), &#39;param_classifier:random_forest:max_depth&#39;: masked_array(data=[&#39;None&#39;, &#39;None&#39;, --, --, --, --, --, --, --],
+             mask=[False, False,  True,  True,  True,  True,  True,  True,
+                    True],
+       fill_value=&#39;N/A&#39;,
+            dtype=&#39;&lt;U32&#39;), &#39;param_classifier:random_forest:max_features&#39;: masked_array(data=[0.5, 0.9331254454871041, --, --, --, --, --, --, --],
+             mask=[False, False,  True,  True,  True,  True,  True,  True,
+                    True],
+       fill_value=1e+20), &#39;param_classifier:random_forest:max_leaf_nodes&#39;: masked_array(data=[&#39;None&#39;, &#39;None&#39;, --, --, --, --, --, --, --],
+             mask=[False, False,  True,  True,  True,  True,  True,  True,
+                    True],
+       fill_value=&#39;N/A&#39;,
+            dtype=&#39;&lt;U32&#39;), &#39;param_classifier:random_forest:min_impurity_decrease&#39;: masked_array(data=[0.0, 0.0, --, --, --, --, --, --, --],
+             mask=[False, False,  True,  True,  True,  True,  True,  True,
+                    True],
+       fill_value=1e+20), &#39;param_classifier:random_forest:min_samples_leaf&#39;: masked_array(data=[1.0, 2.0, --, --, --, --, --, --, --],
+             mask=[False, False,  True,  True,  True,  True,  True,  True,
+                    True],
+       fill_value=1e+20), &#39;param_classifier:random_forest:min_samples_split&#39;: masked_array(data=[2.0, 20.0, --, --, --, --, --, --, --],
+             mask=[False, False,  True,  True,  True,  True,  True,  True,
+                    True],
+       fill_value=1e+20), &#39;param_classifier:random_forest:min_weight_fraction_leaf&#39;: masked_array(data=[0.0, 0.0, --, --, --, --, --, --, --],
+             mask=[False, False,  True,  True,  True,  True,  True,  True,
+                    True],
+       fill_value=1e+20), &#39;param_classifier:sgd:alpha&#39;: masked_array(data=[--, --, --, --, --, --, --, 1.0333981795670635e-07,
+                   0.0006046872822704108],
+             mask=[ True,  True,  True,  True,  True,  True,  True, False,
+                   False],
+       fill_value=1e+20), &#39;param_classifier:sgd:average&#39;: masked_array(data=[--, --, --, --, --, --, --, &#39;False&#39;, &#39;False&#39;],
+             mask=[ True,  True,  True,  True,  True,  True,  True, False,
+                   False],
+       fill_value=&#39;N/A&#39;,
+            dtype=&#39;&lt;U32&#39;), &#39;param_classifier:sgd:fit_intercept&#39;: masked_array(data=[--, --, --, --, --, --, --, &#39;True&#39;, &#39;True&#39;],
+             mask=[ True,  True,  True,  True,  True,  True,  True, False,
+                   False],
+       fill_value=&#39;N/A&#39;,
+            dtype=&#39;&lt;U32&#39;), &#39;param_classifier:sgd:learning_rate&#39;: masked_array(data=[--, --, --, --, --, --, --, &#39;constant&#39;, &#39;invscaling&#39;],
+             mask=[ True,  True,  True,  True,  True,  True,  True, False,
+                   False],
+       fill_value=&#39;N/A&#39;,
+            dtype=&#39;&lt;U32&#39;), &#39;param_classifier:sgd:loss&#39;: masked_array(data=[--, --, --, --, --, --, --, &#39;modified_huber&#39;,
+                   &#39;modified_huber&#39;],
+             mask=[ True,  True,  True,  True,  True,  True,  True, False,
+                   False],
+       fill_value=&#39;N/A&#39;,
+            dtype=&#39;&lt;U32&#39;), &#39;param_classifier:sgd:penalty&#39;: masked_array(data=[--, --, --, --, --, --, --, &#39;l2&#39;, &#39;elasticnet&#39;],
+             mask=[ True,  True,  True,  True,  True,  True,  True, False,
+                   False],
+       fill_value=&#39;N/A&#39;,
+            dtype=&#39;&lt;U32&#39;), &#39;param_classifier:sgd:tol&#39;: masked_array(data=[--, --, --, --, --, --, --, 2.5846457091661748e-05,
+                   1.7244397014112777e-05],
+             mask=[ True,  True,  True,  True,  True,  True,  True, False,
+                   False],
        fill_value=1e+20), &#39;param_data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: masked_array(data=[&#39;one_hot_encoding&#39;, &#39;no_encoding&#39;, &#39;no_encoding&#39;,
                    &#39;no_encoding&#39;, &#39;one_hot_encoding&#39;, &#39;no_encoding&#39;,
-                   &#39;no_encoding&#39;, &#39;encoding&#39;],
-             mask=[False, False, False, False, False, False, False, False],
+                   &#39;no_encoding&#39;, &#39;encoding&#39;, &#39;no_encoding&#39;],
+             mask=[False, False, False, False, False, False, False, False,
+                   False],
        fill_value=&#39;N/A&#39;,
             dtype=&#39;&lt;U16&#39;), &#39;param_data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: masked_array(data=[&#39;minority_coalescer&#39;, &#39;minority_coalescer&#39;,
                    &#39;minority_coalescer&#39;, &#39;minority_coalescer&#39;,
                    &#39;minority_coalescer&#39;, &#39;no_coalescense&#39;,
-                   &#39;no_coalescense&#39;, &#39;minority_coalescer&#39;],
-             mask=[False, False, False, False, False, False, False, False],
+                   &#39;no_coalescense&#39;, &#39;minority_coalescer&#39;,
+                   &#39;no_coalescense&#39;],
+             mask=[False, False, False, False, False, False, False, False,
+                   False],
        fill_value=&#39;N/A&#39;,
             dtype=&#39;&lt;U18&#39;), &#39;param_data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: masked_array(data=[&#39;mean&#39;, &#39;mean&#39;, &#39;most_frequent&#39;, &#39;mean&#39;, &#39;mean&#39;,
-                   &#39;mean&#39;, &#39;mean&#39;, &#39;most_frequent&#39;],
-             mask=[False, False, False, False, False, False, False, False],
+                   &#39;mean&#39;, &#39;mean&#39;, &#39;most_frequent&#39;, &#39;mean&#39;],
+             mask=[False, False, False, False, False, False, False, False,
+                   False],
        fill_value=&#39;N/A&#39;,
             dtype=&#39;&lt;U13&#39;), &#39;param_data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: masked_array(data=[&#39;standardize&#39;, &#39;none&#39;, &#39;quantile_transformer&#39;,
                    &#39;normalize&#39;, &#39;minmax&#39;, &#39;power_transformer&#39;,
-                   &#39;quantile_transformer&#39;, &#39;minmax&#39;],
-             mask=[False, False, False, False, False, False, False, False],
+                   &#39;quantile_transformer&#39;, &#39;minmax&#39;, &#39;power_transformer&#39;],
+             mask=[False, False, False, False, False, False, False, False,
+                   False],
        fill_value=&#39;N/A&#39;,
             dtype=&#39;&lt;U20&#39;), &#39;param_feature_preprocessor:pca:keep_variance&#39;: masked_array(data=[0.9999, 0.9967857433838874, 0.7895711479212801,
                    0.8047274080856589, 0.9290439925152777,
                    0.6661824659281315, 0.7772788812704434,
-                   0.9755563101648113],
-             mask=[False, False, False, False, False, False, False, False],
+                   0.9755563101648113, 0.9941379232526211],
+             mask=[False, False, False, False, False, False, False, False,
+                   False],
        fill_value=1e+20), &#39;param_feature_preprocessor:pca:whiten&#39;: masked_array(data=[&#39;False&#39;, &#39;False&#39;, &#39;True&#39;, &#39;False&#39;, &#39;False&#39;, &#39;False&#39;,
-                   &#39;True&#39;, &#39;True&#39;],
-             mask=[False, False, False, False, False, False, False, False],
-       fill_value=&#39;N/A&#39;,
-            dtype=&#39;&lt;U5&#39;), &#39;param_classifier:gradient_boosting:n_iter_no_change&#39;: masked_array(data=[--, --, --, --, --, --, --, --],
-             mask=[ True,  True,  True,  True,  True,  True,  True,  True],
-       fill_value=1e+20,
-            dtype=float64), &#39;param_classifier:gradient_boosting:validation_fraction&#39;: masked_array(data=[--, --, --, --, --, --, --, --],
-             mask=[ True,  True,  True,  True,  True,  True,  True,  True],
-       fill_value=1e+20,
-            dtype=float64), &#39;param_classifier:lda:shrinkage_factor&#39;: masked_array(data=[--, --, --, --, --, --, --, --],
-             mask=[ True,  True,  True,  True,  True,  True,  True,  True],
-       fill_value=1e+20,
-            dtype=float64), &#39;param_classifier:libsvm_svc:coef0&#39;: masked_array(data=[--, --, --, --, 0.08087614244138486, --, --, --],
-             mask=[ True,  True,  True,  True, False,  True,  True,  True],
-       fill_value=1e+20), &#39;param_classifier:libsvm_svc:degree&#39;: masked_array(data=[--, --, --, --, 3.0, --, --, --],
-             mask=[ True,  True,  True,  True, False,  True,  True,  True],
-       fill_value=1e+20), &#39;param_classifier:mlp:validation_fraction&#39;: masked_array(data=[--, --, 0.1, --, --, --, --, --],
-             mask=[ True,  True, False,  True,  True,  True,  True,  True],
-       fill_value=1e+20), &#39;param_classifier:sgd:epsilon&#39;: masked_array(data=[--, --, --, --, --, --, --, 0.03457277874230573],
-             mask=[ True,  True,  True,  True,  True,  True,  True, False],
-       fill_value=1e+20), &#39;param_classifier:sgd:eta0&#39;: masked_array(data=[--, --, --, --, --, --, --, 0.004635634934629576],
-             mask=[ True,  True,  True,  True,  True,  True,  True, False],
-       fill_value=1e+20), &#39;param_classifier:sgd:l1_ratio&#39;: masked_array(data=[--, --, --, --, --, --, --, --],
-             mask=[ True,  True,  True,  True,  True,  True,  True,  True],
-       fill_value=1e+20,
-            dtype=float64), &#39;param_classifier:sgd:power_t&#39;: masked_array(data=[--, --, --, --, --, --, --, --],
-             mask=[ True,  True,  True,  True,  True,  True,  True,  True],
-       fill_value=1e+20,
-            dtype=float64), &#39;param_data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: masked_array(data=[0.01, 0.005069923784737444, 0.002766772136115771,
+                   &#39;True&#39;, &#39;True&#39;, &#39;False&#39;],
+             mask=[False, False, False, False, False, False, False, False,
+                   False],
+       fill_value=&#39;N/A&#39;,
+            dtype=&#39;&lt;U5&#39;), &#39;param_classifier:gradient_boosting:n_iter_no_change&#39;: masked_array(data=[--, --, --, --, --, --, --, --, --],
+             mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                    True],
+       fill_value=1e+20,
+            dtype=float64), &#39;param_classifier:gradient_boosting:validation_fraction&#39;: masked_array(data=[--, --, --, --, --, --, --, --, --],
+             mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                    True],
+       fill_value=1e+20,
+            dtype=float64), &#39;param_classifier:lda:shrinkage_factor&#39;: masked_array(data=[--, --, --, --, --, --, --, --, --],
+             mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                    True],
+       fill_value=1e+20,
+            dtype=float64), &#39;param_classifier:libsvm_svc:coef0&#39;: masked_array(data=[--, --, --, --, 0.08087614244138486, --, --, --, --],
+             mask=[ True,  True,  True,  True, False,  True,  True,  True,
+                    True],
+       fill_value=1e+20), &#39;param_classifier:libsvm_svc:degree&#39;: masked_array(data=[--, --, --, --, 3.0, --, --, --, --],
+             mask=[ True,  True,  True,  True, False,  True,  True,  True,
+                    True],
+       fill_value=1e+20), &#39;param_classifier:mlp:validation_fraction&#39;: masked_array(data=[--, --, 0.1, --, --, --, --, --, --],
+             mask=[ True,  True, False,  True,  True,  True,  True,  True,
+                    True],
+       fill_value=1e+20), &#39;param_classifier:sgd:epsilon&#39;: masked_array(data=[--, --, --, --, --, --, --, 0.03457277874230573,
+                   0.0006085470273398056],
+             mask=[ True,  True,  True,  True,  True,  True,  True, False,
+                   False],
+       fill_value=1e+20), &#39;param_classifier:sgd:eta0&#39;: masked_array(data=[--, --, --, --, --, --, --, 0.004635634934629583,
+                   0.0003474820560182061],
+             mask=[ True,  True,  True,  True,  True,  True,  True, False,
+                   False],
+       fill_value=1e+20), &#39;param_classifier:sgd:l1_ratio&#39;: masked_array(data=[--, --, --, --, --, --, --, --, 4.5783995454757295e-09],
+             mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                   False],
+       fill_value=1e+20), &#39;param_classifier:sgd:power_t&#39;: masked_array(data=[--, --, --, --, --, --, --, --, 0.6064541566557635],
+             mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                   False],
+       fill_value=1e+20), &#39;param_data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: masked_array(data=[0.01, 0.005069923784737444, 0.002766772136115771,
                    0.008015420020402715, 0.010000000000000004, --, --,
-                   0.00015788887334393744],
-             mask=[False, False, False, False, False,  True,  True, False],
-       fill_value=1e+20), &#39;param_data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles&#39;: masked_array(data=[--, --, 180.0, --, --, --, 526.0, --],
-             mask=[ True,  True, False,  True,  True,  True, False,  True],
-       fill_value=1e+20), &#39;param_data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution&#39;: masked_array(data=[--, --, &#39;uniform&#39;, --, --, --, &#39;uniform&#39;, --],
-             mask=[ True,  True, False,  True,  True,  True, False,  True],
+                   0.00015788887334393714, --],
+             mask=[False, False, False, False, False,  True,  True, False,
+                    True],
+       fill_value=1e+20), &#39;param_data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles&#39;: masked_array(data=[--, --, 180.0, --, --, --, 526.0, --, --],
+             mask=[ True,  True, False,  True,  True,  True, False,  True,
+                    True],
+       fill_value=1e+20), &#39;param_data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution&#39;: masked_array(data=[--, --, &#39;uniform&#39;, --, --, --, &#39;uniform&#39;, --, --],
+             mask=[ True,  True, False,  True,  True,  True, False,  True,
+                    True],
        fill_value=&#39;N/A&#39;,
-            dtype=&#39;&lt;U32&#39;), &#39;param_data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max&#39;: masked_array(data=[--, --, --, --, --, --, --, --],
-             mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+            dtype=&#39;&lt;U32&#39;), &#39;param_data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max&#39;: masked_array(data=[--, --, --, --, --, --, --, --, --],
+             mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                    True],
        fill_value=1e+20,
-            dtype=float64), &#39;param_data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min&#39;: masked_array(data=[--, --, --, --, --, --, --, --],
-             mask=[ True,  True,  True,  True,  True,  True,  True,  True],
+            dtype=float64), &#39;param_data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min&#39;: masked_array(data=[--, --, --, --, --, --, --, --, --],
+             mask=[ True,  True,  True,  True,  True,  True,  True,  True,
+                    True],
        fill_value=1e+20,
             dtype=float64)}
 </pre></div>
 </div>
-</section>
-<section id="inspect-the-components-of-the-best-model">
+</div>
+<div class="section" id="inspect-the-components-of-the-best-model">
 <h2>Inspect the components of the best model<a class="headerlink" href="#inspect-the-components-of-the-best-model" title="Permalink to this headline">¶</a></h2>
 <p>Iterate over the components of the model and print
 The explained variance ratio per stage</p>
@@ -827,23 +995,26 @@ <h2>Inspect the components of the best model<a class="headerlink" href="#inspect
 <div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>The 0th pipeline has a explained variance of [0.49503611 0.16649281 0.09111888 0.07213284 0.04865917 0.03208923
  0.01851537 0.01223987 0.0104432  0.01006215 0.00894849 0.0076293
  0.00684163]
-The 1th pipeline has a explained variance of [0.45954164 0.18010277 0.09814953 0.06334655]
+The 1th pipeline has a explained variance of [0.98080571 0.01684553]
 The 2th pipeline has a explained variance of [0.44685232 0.17122959 0.1085501 ]
-The 3th pipeline has a explained variance of [0.98080571 0.01684553]
-The 4th pipeline has a explained variance of [4.32956881e-01 1.79057296e-01 1.11737571e-01 6.80724345e-02
+The 3th pipeline has a explained variance of [0.49503611 0.16649281 0.09111888 0.07213284 0.04865917 0.03208923
+ 0.01851537 0.01223987]
+The 4th pipeline has a explained variance of [0.45954164 0.18010277 0.09814953 0.06334655]
+The 5th pipeline has a explained variance of [0.44685232 0.17122959 0.1085501  0.07063243 0.06437417 0.03986668
+ 0.02462169 0.01589975 0.01257378 0.00948519 0.00858029 0.0067563
+ 0.0059319  0.00431052 0.0030875  0.00172583]
+The 6th pipeline has a explained variance of [4.32956881e-01 1.79057296e-01 1.11737571e-01 6.80724345e-02
  5.94611519e-02 3.70629898e-02 2.38430977e-02 1.49326086e-02
  1.37641366e-02 1.13704890e-02 1.03737258e-02 8.74116751e-03
  7.57629717e-03 4.86528503e-03 3.32225143e-03 2.55773043e-03
  2.20759805e-03 1.88675402e-03 1.36245140e-03 1.03409213e-03
  8.39749085e-04 7.91287172e-04 6.75655689e-04 5.42961621e-04
  5.02641737e-04 2.07827509e-04 1.74597367e-04]
-The 5th pipeline has a explained variance of [0.4595393  0.18012072 0.09809101 0.06332899]
-The 6th pipeline has a explained variance of [0.76699224 0.17152095]
-The 7th pipeline has a explained variance of [0.49503611 0.16649281 0.09111888 0.07213284 0.04865917 0.03208923
- 0.01851537 0.01223987]
+The 7th pipeline has a explained variance of [0.76699224 0.17152095]
+The 8th pipeline has a explained variance of [0.4595393  0.18012072 0.09809101 0.06332899]
 </pre></div>
 </div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 0 minutes  34.468 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 0 minutes  24.928 seconds)</p>
 <div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-examples-40-advanced-example-get-pipeline-components-py">
 <div class="binder-badge docutils container">
 <a class="reference external image-reference" href="https://mybinder.org/v2/gh/automl/auto-sklearn/master?urlpath=lab/tree/notebooks/examples/40_advanced/example_get_pipeline_components.ipynb"><img alt="Launch binder" src="../../_images/binder_badge_logo1.svg" width="150px" /></a>
@@ -856,8 +1027,8 @@ <h2>Inspect the components of the best model<a class="headerlink" href="#inspect
 </div>
 </div>
 <p class="sphx-glr-signature"><a class="reference external" href="https://sphinx-gallery.github.io">Gallery generated by Sphinx-Gallery</a></p>
-</section>
-</section>
+</div>
+</div>
 
 
     </div>
@@ -878,7 +1049,7 @@ <h2>Inspect the components of the best model<a class="headerlink" href="#inspect
       
     </p>
     <p>
-        &copy; Copyright 2014-2021, Machine Learning Professorship Freiburg.<br/>
+        &copy; Copyright 2014-2022, Machine Learning Professorship Freiburg.<br/>
       Created using <a href="http://sphinx-doc.org/">Sphinx</a> 4.2.0.<br/>
     </p>
   </div>
diff --git a/master/examples/40_advanced/example_inspect_predictions.html b/master/examples/40_advanced/example_inspect_predictions.html
index 4b4459b5c3..321143ed27 100644
--- a/master/examples/40_advanced/example_inspect_predictions.html
+++ b/master/examples/40_advanced/example_inspect_predictions.html
@@ -3,9 +3,8 @@
 <html>
   <head>
     <meta charset="utf-8" />
-    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="generator" content="Docutils 0.17.1: http://docutils.sourceforge.net/" />
-
-    <title>Model Explanation &#8212; AutoSklearn 0.14.2 documentation</title>
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>Model Explanation &#8212; AutoSklearn 0.14.4 documentation</title>
     <link rel="stylesheet" type="text/css" href="../../_static/pygments.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/bootstrap-sphinx.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/sg_gallery.css" />
@@ -55,7 +54,7 @@
         </button>
         <a class="navbar-brand" href="../../index.html">
           auto-sklearn</a>
-        <span class="navbar-text navbar-version pull-left"><b>0.14.2</b></span>
+        <span class="navbar-text navbar-version pull-left"><b>0.14.4</b></span>
       </div>
 
         <div class="collapse navbar-collapse nav-collapse">
@@ -125,7 +124,7 @@
 <p>Click <a class="reference internal" href="#sphx-glr-download-examples-40-advanced-example-inspect-predictions-py"><span class="std std-ref">here</span></a>
 to download the full example code or to run this example in your browser via Binder</p>
 </div>
-<section class="sphx-glr-example-title" id="model-explanation">
+<div class="sphx-glr-example-title section" id="model-explanation">
 <span id="sphx-glr-examples-40-advanced-example-inspect-predictions-py"></span><h1>Model Explanation<a class="headerlink" href="#model-explanation" title="Permalink to this headline">¶</a></h1>
 <p>The following example shows how to fit a simple classification model with
 <em>auto-sklearn</em> and use the <a class="reference external" href="https://scikit-learn.org/stable/inspection.html">inspect module</a> from
@@ -136,7 +135,7 @@
 <span class="kn">import</span> <span class="nn">autosklearn.classification</span>
 </pre></div>
 </div>
-<section id="load-data-and-build-a-model">
+<div class="section" id="load-data-and-build-a-model">
 <h2>Load Data and Build a Model<a class="headerlink" href="#load-data-and-build-a-model" title="Permalink to this headline">¶</a></h2>
 <p>We start by loading the “Run or walk” dataset from OpenML and train an auto-sklearn model on it.
 For this dataset, the goal is to predict whether a person is running or walking based on
@@ -166,12 +165,16 @@ <h2>Load Data and Build a Model<a class="headerlink" href="#load-data-and-build-
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Train score 0.9954285714285714
-Test score 0.9826666666666667
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:68: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+  self.metafeatures = self.metafeatures.append(metafeatures)
+/home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:72: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+  self.algorithm_runs[metric].append(runs)
+Train score 0.9957142857142857
+Test score 0.984
 </pre></div>
 </div>
-</section>
-<section id="compute-permutation-importance-part-1">
+</div>
+<div class="section" id="compute-permutation-importance-part-1">
 <h2>Compute permutation importance - part 1<a class="headerlink" href="#compute-permutation-importance-part-1" title="Permalink to this headline">¶</a></h2>
 <p>Since auto-sklearn implements the scikit-learn interface, it can be used with the scikit-learn’s
 inspection module. So, now we first look at the <a class="reference external" href="https://christophm.github.io/interpretable-ml-book/feature-importance.html">permutation importance</a>, which defines the
@@ -195,16 +198,16 @@ <h2>Compute permutation importance - part 1<a class="headerlink" href="#compute-
 </pre></div>
 </div>
 <img src="../../_images/sphx_glr_example_inspect_predictions_001.png" srcset="../../_images/sphx_glr_example_inspect_predictions_001.png" alt="example inspect predictions" class = "sphx-glr-single-img"/><p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>gyro_y    : 0.000 +/- 0.001
-gyro_x    : 0.025 +/- 0.003
-gyro_z    : 0.037 +/- 0.003
-acceleration_x: 0.056 +/- 0.006
-acceleration_z: 0.126 +/- 0.006
-acceleration_y: 0.280 +/- 0.006
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>gyro_y    : 0.001 +/- 0.002
+gyro_x    : 0.023 +/- 0.002
+gyro_z    : 0.036 +/- 0.003
+acceleration_x: 0.051 +/- 0.006
+acceleration_z: 0.144 +/- 0.007
+acceleration_y: 0.272 +/- 0.006
 </pre></div>
 </div>
-</section>
-<section id="create-partial-dependence-pd-and-individual-conditional-expectation-ice-plots-part-2">
+</div>
+<div class="section" id="create-partial-dependence-pd-and-individual-conditional-expectation-ice-plots-part-2">
 <h2>Create partial dependence (PD) and individual conditional expectation (ICE) plots - part 2<a class="headerlink" href="#create-partial-dependence-pd-and-individual-conditional-expectation-ice-plots-part-2" title="Permalink to this headline">¶</a></h2>
 <p><a class="reference external" href="https://christophm.github.io/interpretable-ml-book/ice.html">ICE plots</a> describe the relation
 between feature values and the response value for each sample
@@ -226,8 +229,8 @@ <h2>Create partial dependence (PD) and individual conditional expectation (ICE)
 <span class="n">plt</span><span class="o">.</span><span class="n">show</span><span class="p">()</span>
 </pre></div>
 </div>
-<img src="../../_images/sphx_glr_example_inspect_predictions_002.png" srcset="../../_images/sphx_glr_example_inspect_predictions_002.png" alt="example inspect predictions" class = "sphx-glr-single-img"/></section>
-<section id="create-partial-dependence-pdp-plots-for-more-than-one-feature-part-3">
+<img src="../../_images/sphx_glr_example_inspect_predictions_002.png" srcset="../../_images/sphx_glr_example_inspect_predictions_002.png" alt="example inspect predictions" class = "sphx-glr-single-img"/></div>
+<div class="section" id="create-partial-dependence-pdp-plots-for-more-than-one-feature-part-3">
 <h2>Create partial dependence (PDP) plots for more than one feature - part 3<a class="headerlink" href="#create-partial-dependence-pdp-plots-for-more-than-one-feature-part-3" title="Permalink to this headline">¶</a></h2>
 <p>A PD plot can also be generated for two features and thus allow to inspect the interaction between
 these features. Again, we’ll look at acceleration_y and acceleration_z.</p>
@@ -240,7 +243,7 @@ <h2>Create partial dependence (PDP) plots for more than one feature - part 3<a c
 <span class="n">plt</span><span class="o">.</span><span class="n">show</span><span class="p">()</span>
 </pre></div>
 </div>
-<img src="../../_images/sphx_glr_example_inspect_predictions_003.png" srcset="../../_images/sphx_glr_example_inspect_predictions_003.png" alt="example inspect predictions" class = "sphx-glr-single-img"/><p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 4 minutes  2.065 seconds)</p>
+<img src="../../_images/sphx_glr_example_inspect_predictions_003.png" srcset="../../_images/sphx_glr_example_inspect_predictions_003.png" alt="example inspect predictions" class = "sphx-glr-single-img"/><p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 4 minutes  6.567 seconds)</p>
 <div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-examples-40-advanced-example-inspect-predictions-py">
 <div class="binder-badge docutils container">
 <a class="reference external image-reference" href="https://mybinder.org/v2/gh/automl/auto-sklearn/master?urlpath=lab/tree/notebooks/examples/40_advanced/example_inspect_predictions.ipynb"><img alt="Launch binder" src="../../_images/binder_badge_logo1.svg" width="150px" /></a>
@@ -253,8 +256,8 @@ <h2>Create partial dependence (PDP) plots for more than one feature - part 3<a c
 </div>
 </div>
 <p class="sphx-glr-signature"><a class="reference external" href="https://sphinx-gallery.github.io">Gallery generated by Sphinx-Gallery</a></p>
-</section>
-</section>
+</div>
+</div>
 
 
     </div>
@@ -275,7 +278,7 @@ <h2>Create partial dependence (PDP) plots for more than one feature - part 3<a c
       
     </p>
     <p>
-        &copy; Copyright 2014-2021, Machine Learning Professorship Freiburg.<br/>
+        &copy; Copyright 2014-2022, Machine Learning Professorship Freiburg.<br/>
       Created using <a href="http://sphinx-doc.org/">Sphinx</a> 4.2.0.<br/>
     </p>
   </div>
diff --git a/master/examples/40_advanced/example_interpretable_models.html b/master/examples/40_advanced/example_interpretable_models.html
index 56113c24dc..9b1cc90a41 100644
--- a/master/examples/40_advanced/example_interpretable_models.html
+++ b/master/examples/40_advanced/example_interpretable_models.html
@@ -3,9 +3,8 @@
 <html>
   <head>
     <meta charset="utf-8" />
-    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="generator" content="Docutils 0.17.1: http://docutils.sourceforge.net/" />
-
-    <title>Interpretable models &#8212; AutoSklearn 0.14.2 documentation</title>
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>Interpretable models &#8212; AutoSklearn 0.14.4 documentation</title>
     <link rel="stylesheet" type="text/css" href="../../_static/pygments.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/bootstrap-sphinx.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/sg_gallery.css" />
@@ -55,7 +54,7 @@
         </button>
         <a class="navbar-brand" href="../../index.html">
           auto-sklearn</a>
-        <span class="navbar-text navbar-version pull-left"><b>0.14.2</b></span>
+        <span class="navbar-text navbar-version pull-left"><b>0.14.4</b></span>
       </div>
 
         <div class="collapse navbar-collapse nav-collapse">
@@ -127,16 +126,18 @@
 <p>Click <a class="reference internal" href="#sphx-glr-download-examples-40-advanced-example-interpretable-models-py"><span class="std std-ref">here</span></a>
 to download the full example code or to run this example in your browser via Binder</p>
 </div>
-<section class="sphx-glr-example-title" id="interpretable-models">
+<div class="sphx-glr-example-title section" id="interpretable-models">
 <span id="sphx-glr-examples-40-advanced-example-interpretable-models-py"></span><h1>Interpretable models<a class="headerlink" href="#interpretable-models" title="Permalink to this headline">¶</a></h1>
 <p>The following example shows how to inspect the models which <em>auto-sklearn</em>
 optimizes over and how to restrict them to an interpretable subset.</p>
-<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">autosklearn.classification</span>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">pprint</span> <span class="kn">import</span> <span class="n">pprint</span>
+
+<span class="kn">import</span> <span class="nn">autosklearn.classification</span>
 <span class="kn">import</span> <span class="nn">sklearn.datasets</span>
 <span class="kn">import</span> <span class="nn">sklearn.metrics</span>
 </pre></div>
 </div>
-<section id="show-available-classification-models">
+<div class="section" id="show-available-classification-models">
 <h2>Show available classification models<a class="headerlink" href="#show-available-classification-models" title="Permalink to this headline">¶</a></h2>
 <p>We will first list all classifiers Auto-sklearn chooses from. A similar
 call is available for preprocessors (see below) and regression (not shown)
@@ -166,8 +167,8 @@ <h2>Show available classification models<a class="headerlink" href="#show-availa
 sgd
 </pre></div>
 </div>
-</section>
-<section id="show-available-preprocessors">
+</div>
+<div class="section" id="show-available-preprocessors">
 <h2>Show available preprocessors<a class="headerlink" href="#show-available-preprocessors" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">autosklearn.pipeline.components.feature_preprocessing</span> <span class="kn">import</span> <span class="n">FeaturePreprocessorChoice</span>
 
@@ -196,16 +197,16 @@ <h2>Show available preprocessors<a class="headerlink" href="#show-available-prep
 truncatedSVD
 </pre></div>
 </div>
-</section>
-<section id="data-loading">
+</div>
+<div class="section" id="data-loading">
 <h2>Data Loading<a class="headerlink" href="#data-loading" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">X</span><span class="p">,</span> <span class="n">y</span> <span class="o">=</span> <span class="n">sklearn</span><span class="o">.</span><span class="n">datasets</span><span class="o">.</span><span class="n">load_breast_cancer</span><span class="p">(</span><span class="n">return_X_y</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
 <span class="n">X_train</span><span class="p">,</span> <span class="n">X_test</span><span class="p">,</span> <span class="n">y_train</span><span class="p">,</span> <span class="n">y_test</span> <span class="o">=</span> \
     <span class="n">sklearn</span><span class="o">.</span><span class="n">model_selection</span><span class="o">.</span><span class="n">train_test_split</span><span class="p">(</span><span class="n">X</span><span class="p">,</span> <span class="n">y</span><span class="p">,</span> <span class="n">random_state</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
 </pre></div>
 </div>
-</section>
-<section id="build-and-fit-a-classifier">
+</div>
+<div class="section" id="build-and-fit-a-classifier">
 <h2>Build and fit a classifier<a class="headerlink" href="#build-and-fit-a-classifier" title="Permalink to this headline">¶</a></h2>
 <p>We will now only use a subset of the given classifiers and preprocessors.
 Furthermore, we will restrict the ensemble size to <code class="docutils literal notranslate"><span class="pre">1</span></code> to only use the
@@ -230,7 +231,12 @@ <h2>Build and fit a classifier<a class="headerlink" href="#build-and-fit-a-class
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>AutoSklearnClassifier(ensemble_size=1,
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:68: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+  self.metafeatures = self.metafeatures.append(metafeatures)
+/home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:72: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+  self.algorithm_runs[metric].append(runs)
+
+AutoSklearnClassifier(ensemble_size=1,
                       include={&#39;classifier&#39;: [&#39;decision_tree&#39;, &#39;lda&#39;, &#39;sgd&#39;],
                                &#39;feature_preprocessor&#39;: [&#39;no_preprocessing&#39;,
                                                         &#39;polynomial&#39;,
@@ -239,36 +245,39 @@ <h2>Build and fit a classifier<a class="headerlink" href="#build-and-fit-a-class
                       tmp_folder=&#39;/tmp/autosklearn_interpretable_models_example_tmp&#39;)
 </pre></div>
 </div>
-</section>
-<section id="print-the-final-ensemble-constructed-by-auto-sklearn">
+</div>
+<div class="section" id="print-the-final-ensemble-constructed-by-auto-sklearn">
 <h2>Print the final ensemble constructed by auto-sklearn<a class="headerlink" href="#print-the-final-ensemble-constructed-by-auto-sklearn" title="Permalink to this headline">¶</a></h2>
-<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="nb">print</span><span class="p">(</span><span class="n">automl</span><span class="o">.</span><span class="n">show_models</span><span class="p">())</span>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">pprint</span><span class="p">(</span><span class="n">automl</span><span class="o">.</span><span class="n">show_models</span><span class="p">(),</span> <span class="n">indent</span><span class="o">=</span><span class="mi">4</span><span class="p">)</span>
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>[(1.000000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;sgd&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;polynomial&#39;, &#39;classifier:sgd:alpha&#39;: 0.0003336259810364362, &#39;classifier:sgd:average&#39;: &#39;False&#39;, &#39;classifier:sgd:fit_intercept&#39;: &#39;True&#39;, &#39;classifier:sgd:learning_rate&#39;: &#39;invscaling&#39;, &#39;classifier:sgd:loss&#39;: &#39;log&#39;, &#39;classifier:sgd:penalty&#39;: &#39;l2&#39;, &#39;classifier:sgd:tol&#39;: 0.006915066463246923, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;median&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;standardize&#39;, &#39;feature_preprocessor:polynomial:degree&#39;: 3, &#39;feature_preprocessor:polynomial:include_bias&#39;: &#39;True&#39;, &#39;feature_preprocessor:polynomial:interaction_only&#39;: &#39;False&#39;, &#39;classifier:sgd:eta0&#39;: 0.010000000000000014, &#39;classifier:sgd:power_t&#39;: 0.5, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.012199160249499144},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-]
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>{   63: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78c482160&gt;,
+            &#39;cost&#39;: 0.007092198581560294,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78f80b910&gt;,
+            &#39;ensemble_weight&#39;: 1.0,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c482eb0&gt;,
+            &#39;model_id&#39;: 63,
+            &#39;rank&#39;: 1,
+            &#39;sklearn_classifier&#39;: SGDClassifier(alpha=9.999999999999991e-05, epsilon=1.3700860264543197e-05,
+              eta0=0.010000000000000014, learning_rate=&#39;invscaling&#39;,
+              loss=&#39;modified_huber&#39;, max_iter=64, random_state=1,
+              tol=0.0004668987145850572, warm_start=True)}}
 </pre></div>
 </div>
-</section>
-<section id="get-the-score-of-the-final-ensemble">
+</div>
+<div class="section" id="get-the-score-of-the-final-ensemble">
 <h2>Get the Score of the final ensemble<a class="headerlink" href="#get-the-score-of-the-final-ensemble" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">predictions</span> <span class="o">=</span> <span class="n">automl</span><span class="o">.</span><span class="n">predict</span><span class="p">(</span><span class="n">X_test</span><span class="p">)</span>
 <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Accuracy score:&quot;</span><span class="p">,</span> <span class="n">sklearn</span><span class="o">.</span><span class="n">metrics</span><span class="o">.</span><span class="n">accuracy_score</span><span class="p">(</span><span class="n">y_test</span><span class="p">,</span> <span class="n">predictions</span><span class="p">))</span>
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Accuracy score: 0.958041958041958
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Accuracy score: 0.965034965034965
 </pre></div>
 </div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  55.571 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  54.044 seconds)</p>
 <div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-examples-40-advanced-example-interpretable-models-py">
 <div class="binder-badge docutils container">
 <a class="reference external image-reference" href="https://mybinder.org/v2/gh/automl/auto-sklearn/master?urlpath=lab/tree/notebooks/examples/40_advanced/example_interpretable_models.ipynb"><img alt="Launch binder" src="../../_images/binder_badge_logo1.svg" width="150px" /></a>
@@ -281,8 +290,8 @@ <h2>Get the Score of the final ensemble<a class="headerlink" href="#get-the-scor
 </div>
 </div>
 <p class="sphx-glr-signature"><a class="reference external" href="https://sphinx-gallery.github.io">Gallery generated by Sphinx-Gallery</a></p>
-</section>
-</section>
+</div>
+</div>
 
 
     </div>
@@ -303,7 +312,7 @@ <h2>Get the Score of the final ensemble<a class="headerlink" href="#get-the-scor
       
     </p>
     <p>
-        &copy; Copyright 2014-2021, Machine Learning Professorship Freiburg.<br/>
+        &copy; Copyright 2014-2022, Machine Learning Professorship Freiburg.<br/>
       Created using <a href="http://sphinx-doc.org/">Sphinx</a> 4.2.0.<br/>
     </p>
   </div>
diff --git a/master/examples/40_advanced/example_metrics.html b/master/examples/40_advanced/example_metrics.html
index 72ff5f781c..bf7071332e 100644
--- a/master/examples/40_advanced/example_metrics.html
+++ b/master/examples/40_advanced/example_metrics.html
@@ -3,9 +3,8 @@
 <html>
   <head>
     <meta charset="utf-8" />
-    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="generator" content="Docutils 0.17.1: http://docutils.sourceforge.net/" />
-
-    <title>Metrics &#8212; AutoSklearn 0.14.2 documentation</title>
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>Metrics &#8212; AutoSklearn 0.14.4 documentation</title>
     <link rel="stylesheet" type="text/css" href="../../_static/pygments.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/bootstrap-sphinx.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/sg_gallery.css" />
@@ -55,7 +54,7 @@
         </button>
         <a class="navbar-brand" href="../../index.html">
           auto-sklearn</a>
-        <span class="navbar-text navbar-version pull-left"><b>0.14.2</b></span>
+        <span class="navbar-text navbar-version pull-left"><b>0.14.4</b></span>
       </div>
 
         <div class="collapse navbar-collapse nav-collapse">
@@ -129,7 +128,7 @@
 <p>Click <a class="reference internal" href="#sphx-glr-download-examples-40-advanced-example-metrics-py"><span class="std std-ref">here</span></a>
 to download the full example code or to run this example in your browser via Binder</p>
 </div>
-<section class="sphx-glr-example-title" id="metrics">
+<div class="sphx-glr-example-title section" id="metrics">
 <span id="sphx-glr-examples-40-advanced-example-metrics-py"></span><h1>Metrics<a class="headerlink" href="#metrics" title="Permalink to this headline">¶</a></h1>
 <p><em>Auto-sklearn</em> supports various built-in metrics, which can be found in the
 <a class="reference internal" href="../../api.html#built-in-metrics"><span class="std std-ref">metrics section in the API</span></a>. However, it is also
@@ -146,7 +145,7 @@
 <span class="kn">import</span> <span class="nn">autosklearn.metrics</span>
 </pre></div>
 </div>
-<section id="custom-metrics">
+<div class="section" id="custom-metrics">
 <h2>Custom Metrics<a class="headerlink" href="#custom-metrics" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="k">def</span> <span class="nf">accuracy</span><span class="p">(</span><span class="n">solution</span><span class="p">,</span> <span class="n">prediction</span><span class="p">):</span>
     <span class="c1"># custom function defining accuracy</span>
@@ -170,16 +169,16 @@ <h2>Custom Metrics<a class="headerlink" href="#custom-metrics" title="Permalink
     <span class="k">return</span> <span class="n">np</span><span class="o">.</span><span class="n">mean</span><span class="p">(</span><span class="n">solution</span> <span class="o">!=</span> <span class="n">prediction</span><span class="p">)</span>
 </pre></div>
 </div>
-</section>
-<section id="data-loading">
+</div>
+<div class="section" id="data-loading">
 <h2>Data Loading<a class="headerlink" href="#data-loading" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">X</span><span class="p">,</span> <span class="n">y</span> <span class="o">=</span> <span class="n">sklearn</span><span class="o">.</span><span class="n">datasets</span><span class="o">.</span><span class="n">load_breast_cancer</span><span class="p">(</span><span class="n">return_X_y</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
 <span class="n">X_train</span><span class="p">,</span> <span class="n">X_test</span><span class="p">,</span> <span class="n">y_train</span><span class="p">,</span> <span class="n">y_test</span> <span class="o">=</span> \
     <span class="n">sklearn</span><span class="o">.</span><span class="n">model_selection</span><span class="o">.</span><span class="n">train_test_split</span><span class="p">(</span><span class="n">X</span><span class="p">,</span> <span class="n">y</span><span class="p">,</span> <span class="n">random_state</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
 </pre></div>
 </div>
-</section>
-<section id="print-a-list-of-available-metrics">
+</div>
+<div class="section" id="print-a-list-of-available-metrics">
 <h2>Print a list of available metrics<a class="headerlink" href="#print-a-list-of-available-metrics" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Available CLASSIFICATION metrics autosklearn.metrics.*:&quot;</span><span class="p">)</span>
 <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;</span><span class="se">\t</span><span class="s2">*&quot;</span> <span class="o">+</span> <span class="s2">&quot;</span><span class="se">\n\t</span><span class="s2">*&quot;</span><span class="o">.</span><span class="n">join</span><span class="p">(</span><span class="n">autosklearn</span><span class="o">.</span><span class="n">metrics</span><span class="o">.</span><span class="n">CLASSIFICATION_METRICS</span><span class="p">))</span>
@@ -195,17 +194,14 @@ <h2>Print a list of available metrics<a class="headerlink" href="#print-a-list-o
         *roc_auc
         *average_precision
         *log_loss
-        *precision
         *precision_macro
         *precision_micro
         *precision_samples
         *precision_weighted
-        *recall
         *recall_macro
         *recall_micro
         *recall_samples
         *recall_weighted
-        *f1
         *f1_macro
         *f1_micro
         *f1_samples
@@ -219,8 +215,8 @@ <h2>Print a list of available metrics<a class="headerlink" href="#print-a-list-o
         *r2
 </pre></div>
 </div>
-</section>
-<section id="first-example-use-predefined-accuracy-metric">
+</div>
+<div class="section" id="first-example-use-predefined-accuracy-metric">
 <h2>First example: Use predefined accuracy metric<a class="headerlink" href="#first-example-use-predefined-accuracy-metric" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="nb">print</span><span class="p">(</span><span class="s2">&quot;#&quot;</span><span class="o">*</span><span class="mi">80</span><span class="p">)</span>
 <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Use predefined accuracy metric&quot;</span><span class="p">)</span>
@@ -241,11 +237,15 @@ <h2>First example: Use predefined accuracy metric<a class="headerlink" href="#fi
 <p class="sphx-glr-script-out">Out:</p>
 <div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>################################################################################
 Use predefined accuracy metric
-Accuracy score 0.944 using accuracy
+/home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:68: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+  self.metafeatures = self.metafeatures.append(metafeatures)
+/home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:72: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+  self.algorithm_runs[metric].append(runs)
+Accuracy score 0.958 using accuracy
 </pre></div>
 </div>
-</section>
-<section id="second-example-use-own-accuracy-metric">
+</div>
+<div class="section" id="second-example-use-own-accuracy-metric">
 <h2>Second example: Use own accuracy metric<a class="headerlink" href="#second-example-use-own-accuracy-metric" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="nb">print</span><span class="p">(</span><span class="s2">&quot;#&quot;</span><span class="o">*</span><span class="mi">80</span><span class="p">)</span>
 <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Use self defined accuracy metric&quot;</span><span class="p">)</span>
@@ -277,8 +277,8 @@ <h2>Second example: Use own accuracy metric<a class="headerlink" href="#second-e
 Accuracy score 0.937 using accu
 </pre></div>
 </div>
-</section>
-<section id="third-example-use-own-error-metric">
+</div>
+<div class="section" id="third-example-use-own-error-metric">
 <h2>Third example: Use own error metric<a class="headerlink" href="#third-example-use-own-error-metric" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="nb">print</span><span class="p">(</span><span class="s2">&quot;#&quot;</span><span class="o">*</span><span class="mi">80</span><span class="p">)</span>
 <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Use self defined error metric&quot;</span><span class="p">)</span>
@@ -307,22 +307,24 @@ <h2>Third example: Use own error metric<a class="headerlink" href="#third-exampl
 <p class="sphx-glr-script-out">Out:</p>
 <div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>################################################################################
 Use self defined error metric
-[WARNING] [2021-11-09 20:04:18,731:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
-[WARNING] [2021-11-09 20:04:19,737:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
-[WARNING] [2021-11-09 20:04:23,184:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
-[WARNING] [2021-11-09 20:04:24,140:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
-[WARNING] [2021-11-09 20:04:25,232:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
-[WARNING] [2021-11-09 20:04:29,419:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
-[WARNING] [2021-11-09 20:04:31,278:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
-[WARNING] [2021-11-09 20:04:35,411:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
-[WARNING] [2021-11-09 20:04:40,045:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
-[WARNING] [2021-11-09 20:04:45,068:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
-[WARNING] [2021-11-09 20:04:50,071:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+[WARNING] [2022-01-25 16:33:33,542:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+[WARNING] [2022-01-25 16:33:34,650:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+[WARNING] [2022-01-25 16:33:37,944:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+[WARNING] [2022-01-25 16:33:38,788:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+[WARNING] [2022-01-25 16:33:39,764:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+[WARNING] [2022-01-25 16:33:43,502:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+[WARNING] [2022-01-25 16:33:45,472:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+[WARNING] [2022-01-25 16:33:49,255:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+[WARNING] [2022-01-25 16:33:53,483:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+[WARNING] [2022-01-25 16:33:58,566:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+[WARNING] [2022-01-25 16:34:03,041:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+[WARNING] [2022-01-25 16:34:07,836:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+[WARNING] [2022-01-25 16:34:08,936:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
 Error score -0.063 using error
 </pre></div>
 </div>
-</section>
-<section id="fourth-example-use-own-accuracy-metric-with-additional-argument">
+</div>
+<div class="section" id="fourth-example-use-own-accuracy-metric-with-additional-argument">
 <h2>Fourth example: Use own accuracy metric with additional argument<a class="headerlink" href="#fourth-example-use-own-accuracy-metric-with-additional-argument" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="nb">print</span><span class="p">(</span><span class="s2">&quot;#&quot;</span><span class="o">*</span><span class="mi">80</span><span class="p">)</span>
 <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Use self defined accuracy with additional argument&quot;</span><span class="p">)</span>
@@ -355,8 +357,8 @@ <h2>Fourth example: Use own accuracy metric with additional argument<a class="he
 Accuracy score 0.937 using accu_add
 </pre></div>
 </div>
-</section>
-<section id="fifth-example-use-own-accuracy-metric-with-additional-argument">
+</div>
+<div class="section" id="fifth-example-use-own-accuracy-metric-with-additional-argument">
 <h2>Fifth example: Use own accuracy metric with additional argument<a class="headerlink" href="#fifth-example-use-own-accuracy-metric-with-additional-argument" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="nb">print</span><span class="p">(</span><span class="s2">&quot;#&quot;</span><span class="o">*</span><span class="mi">80</span><span class="p">)</span>
 <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Use self defined error with additional argument&quot;</span><span class="p">)</span>
@@ -386,24 +388,25 @@ <h2>Fifth example: Use own accuracy metric with additional argument<a class="hea
 <p class="sphx-glr-script-out">Out:</p>
 <div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>################################################################################
 Use self defined error with additional argument
-[WARNING] [2021-11-09 20:06:23,413:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
-[WARNING] [2021-11-09 20:06:27,269:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
-[WARNING] [2021-11-09 20:06:34,117:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
-[WARNING] [2021-11-09 20:06:35,067:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
-[WARNING] [2021-11-09 20:06:38,388:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
-[WARNING] [2021-11-09 20:06:39,226:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
-[WARNING] [2021-11-09 20:06:40,191:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
-[WARNING] [2021-11-09 20:06:45,538:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
-[WARNING] [2021-11-09 20:06:49,773:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
-[WARNING] [2021-11-09 20:06:54,348:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
-[WARNING] [2021-11-09 20:06:56,606:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
-[WARNING] [2021-11-09 20:07:01,814:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
-[WARNING] [2021-11-09 20:07:06,326:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
-[WARNING] [2021-11-09 20:07:07,428:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+[WARNING] [2022-01-25 16:35:15,090:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+[WARNING] [2022-01-25 16:35:18,533:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+[WARNING] [2022-01-25 16:35:24,273:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+[WARNING] [2022-01-25 16:35:25,186:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+[WARNING] [2022-01-25 16:35:28,418:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+[WARNING] [2022-01-25 16:35:29,368:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+[WARNING] [2022-01-25 16:35:30,457:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+[WARNING] [2022-01-25 16:35:35,650:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+[WARNING] [2022-01-25 16:35:39,374:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+[WARNING] [2022-01-25 16:35:43,590:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+[WARNING] [2022-01-25 16:35:45,754:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+[WARNING] [2022-01-25 16:35:50,626:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+[WARNING] [2022-01-25 16:35:54,807:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+[WARNING] [2022-01-25 16:35:55,786:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
+[WARNING] [2022-01-25 16:36:00,907:smac.runhistory.runhistory2epm.RunHistory2EPM4LogCost] Got cost of smaller/equal to 0. Replace by 0.000010 since we use log cost.
 Error score 0.615 using error_add
 </pre></div>
 </div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 5 minutes  19.475 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 4 minutes  39.637 seconds)</p>
 <div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-examples-40-advanced-example-metrics-py">
 <div class="binder-badge docutils container">
 <a class="reference external image-reference" href="https://mybinder.org/v2/gh/automl/auto-sklearn/master?urlpath=lab/tree/notebooks/examples/40_advanced/example_metrics.ipynb"><img alt="Launch binder" src="../../_images/binder_badge_logo1.svg" width="150px" /></a>
@@ -416,8 +419,8 @@ <h2>Fifth example: Use own accuracy metric with additional argument<a class="hea
 </div>
 </div>
 <p class="sphx-glr-signature"><a class="reference external" href="https://sphinx-gallery.github.io">Gallery generated by Sphinx-Gallery</a></p>
-</section>
-</section>
+</div>
+</div>
 
 
     </div>
@@ -438,7 +441,7 @@ <h2>Fifth example: Use own accuracy metric with additional argument<a class="hea
       
     </p>
     <p>
-        &copy; Copyright 2014-2021, Machine Learning Professorship Freiburg.<br/>
+        &copy; Copyright 2014-2022, Machine Learning Professorship Freiburg.<br/>
       Created using <a href="http://sphinx-doc.org/">Sphinx</a> 4.2.0.<br/>
     </p>
   </div>
diff --git a/master/examples/40_advanced/example_pandas_train_test.html b/master/examples/40_advanced/example_pandas_train_test.html
index 8eb13e8f68..e99ba12f3f 100644
--- a/master/examples/40_advanced/example_pandas_train_test.html
+++ b/master/examples/40_advanced/example_pandas_train_test.html
@@ -3,9 +3,8 @@
 <html>
   <head>
     <meta charset="utf-8" />
-    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="generator" content="Docutils 0.17.1: http://docutils.sourceforge.net/" />
-
-    <title>Performance-over-time plot &#8212; AutoSklearn 0.14.2 documentation</title>
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>Performance-over-time plot &#8212; AutoSklearn 0.14.4 documentation</title>
     <link rel="stylesheet" type="text/css" href="../../_static/pygments.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/bootstrap-sphinx.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/sg_gallery.css" />
@@ -55,7 +54,7 @@
         </button>
         <a class="navbar-brand" href="../../index.html">
           auto-sklearn</a>
-        <span class="navbar-text navbar-version pull-left"><b>0.14.2</b></span>
+        <span class="navbar-text navbar-version pull-left"><b>0.14.4</b></span>
       </div>
 
         <div class="collapse navbar-collapse nav-collapse">
@@ -125,7 +124,7 @@
 <p>Click <a class="reference internal" href="#sphx-glr-download-examples-40-advanced-example-pandas-train-test-py"><span class="std std-ref">here</span></a>
 to download the full example code or to run this example in your browser via Binder</p>
 </div>
-<section class="sphx-glr-example-title" id="performance-over-time-plot">
+<div class="sphx-glr-example-title section" id="performance-over-time-plot">
 <span id="sphx-glr-examples-40-advanced-example-pandas-train-test-py"></span><h1>Performance-over-time plot<a class="headerlink" href="#performance-over-time-plot" title="Permalink to this headline">¶</a></h1>
 <p>This example shows, how to use the <em>performance_over_time_</em> attribute to plot the performance
 over train time.  <em>performance_over_time_</em> can contain multiple metrics within a pandas dataframe, namely:</p>
@@ -166,7 +165,7 @@
 <span class="kn">import</span> <span class="nn">autosklearn.classification</span>
 </pre></div>
 </div>
-<section id="data-loading">
+<div class="section" id="data-loading">
 <h2>Data Loading<a class="headerlink" href="#data-loading" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="c1"># Using Australian dataset https://www.openml.org/d/40981.</span>
 <span class="c1"># This example will use the command fetch_openml, which will</span>
@@ -220,8 +219,8 @@ <h2>Data Loading<a class="headerlink" href="#data-loading" title="Permalink to t
 dtype: object
 </pre></div>
 </div>
-</section>
-<section id="build-and-fit-a-classifier">
+</div>
+<div class="section" id="build-and-fit-a-classifier">
 <h2>Build and fit a classifier<a class="headerlink" href="#build-and-fit-a-classifier" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="bp">cls</span> <span class="o">=</span> <span class="n">autosklearn</span><span class="o">.</span><span class="n">classification</span><span class="o">.</span><span class="n">AutoSklearnClassifier</span><span class="p">(</span>
     <span class="n">time_left_for_this_task</span><span class="o">=</span><span class="mi">120</span><span class="p">,</span>
@@ -231,22 +230,27 @@ <h2>Build and fit a classifier<a class="headerlink" href="#build-and-fit-a-class
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>AutoSklearnClassifier(per_run_time_limit=30, time_left_for_this_task=120)
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:68: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+  self.metafeatures = self.metafeatures.append(metafeatures)
+/home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:72: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+  self.algorithm_runs[metric].append(runs)
+
+AutoSklearnClassifier(per_run_time_limit=30, time_left_for_this_task=120)
 </pre></div>
 </div>
-</section>
-<section id="get-the-score-of-the-final-ensemble">
+</div>
+<div class="section" id="get-the-score-of-the-final-ensemble">
 <h2>Get the Score of the final ensemble<a class="headerlink" href="#get-the-score-of-the-final-ensemble" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">predictions</span> <span class="o">=</span> <span class="bp">cls</span><span class="o">.</span><span class="n">predict</span><span class="p">(</span><span class="n">X_test</span><span class="p">)</span>
 <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Accuracy score&quot;</span><span class="p">,</span> <span class="n">sklearn</span><span class="o">.</span><span class="n">metrics</span><span class="o">.</span><span class="n">accuracy_score</span><span class="p">(</span><span class="n">y_test</span><span class="p">,</span> <span class="n">predictions</span><span class="p">))</span>
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Accuracy score 0.8579710144927536
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Accuracy score 0.8492753623188406
 </pre></div>
 </div>
-</section>
-<section id="plot-the-ensemble-performance">
+</div>
+<div class="section" id="plot-the-ensemble-performance">
 <h2>Plot the ensemble performance<a class="headerlink" href="#plot-the-ensemble-performance" title="Permalink to this headline">¶</a></h2>
 <p>The <em>performance_over_time_</em> attribute returns a pandas dataframe, which can
 be directly used for plotting</p>
@@ -261,7 +265,7 @@ <h2>Plot the ensemble performance<a class="headerlink" href="#plot-the-ensemble-
 <span class="n">plt</span><span class="o">.</span><span class="n">show</span><span class="p">()</span>
 </pre></div>
 </div>
-<img src="../../_images/sphx_glr_example_pandas_train_test_001.png" srcset="../../_images/sphx_glr_example_pandas_train_test_001.png" alt="Auto-sklearn accuracy over time" class = "sphx-glr-single-img"/><p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes  1.039 seconds)</p>
+<img src="../../_images/sphx_glr_example_pandas_train_test_001.png" srcset="../../_images/sphx_glr_example_pandas_train_test_001.png" alt="Auto-sklearn accuracy over time" class = "sphx-glr-single-img"/><p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  55.852 seconds)</p>
 <div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-examples-40-advanced-example-pandas-train-test-py">
 <div class="binder-badge docutils container">
 <a class="reference external image-reference" href="https://mybinder.org/v2/gh/automl/auto-sklearn/master?urlpath=lab/tree/notebooks/examples/40_advanced/example_pandas_train_test.ipynb"><img alt="Launch binder" src="../../_images/binder_badge_logo1.svg" width="150px" /></a>
@@ -274,8 +278,8 @@ <h2>Plot the ensemble performance<a class="headerlink" href="#plot-the-ensemble-
 </div>
 </div>
 <p class="sphx-glr-signature"><a class="reference external" href="https://sphinx-gallery.github.io">Gallery generated by Sphinx-Gallery</a></p>
-</section>
-</section>
+</div>
+</div>
 
 
     </div>
@@ -296,7 +300,7 @@ <h2>Plot the ensemble performance<a class="headerlink" href="#plot-the-ensemble-
       
     </p>
     <p>
-        &copy; Copyright 2014-2021, Machine Learning Professorship Freiburg.<br/>
+        &copy; Copyright 2014-2022, Machine Learning Professorship Freiburg.<br/>
       Created using <a href="http://sphinx-doc.org/">Sphinx</a> 4.2.0.<br/>
     </p>
   </div>
diff --git a/master/examples/40_advanced/example_resampling.html b/master/examples/40_advanced/example_resampling.html
index ac8550f931..c3d5a76fc3 100644
--- a/master/examples/40_advanced/example_resampling.html
+++ b/master/examples/40_advanced/example_resampling.html
@@ -3,9 +3,8 @@
 <html>
   <head>
     <meta charset="utf-8" />
-    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="generator" content="Docutils 0.17.1: http://docutils.sourceforge.net/" />
-
-    <title>Resampling Strategies &#8212; AutoSklearn 0.14.2 documentation</title>
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>Resampling Strategies &#8212; AutoSklearn 0.14.4 documentation</title>
     <link rel="stylesheet" type="text/css" href="../../_static/pygments.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/bootstrap-sphinx.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/sg_gallery.css" />
@@ -55,7 +54,7 @@
         </button>
         <a class="navbar-brand" href="../../index.html">
           auto-sklearn</a>
-        <span class="navbar-text navbar-version pull-left"><b>0.14.2</b></span>
+        <span class="navbar-text navbar-version pull-left"><b>0.14.4</b></span>
       </div>
 
         <div class="collapse navbar-collapse nav-collapse">
@@ -128,7 +127,7 @@
 <p>Click <a class="reference internal" href="#sphx-glr-download-examples-40-advanced-example-resampling-py"><span class="std std-ref">here</span></a>
 to download the full example code or to run this example in your browser via Binder</p>
 </div>
-<section class="sphx-glr-example-title" id="resampling-strategies">
+<div class="sphx-glr-example-title section" id="resampling-strategies">
 <span id="sphx-glr-examples-40-advanced-example-resampling-py"></span><h1>Resampling Strategies<a class="headerlink" href="#resampling-strategies" title="Permalink to this headline">¶</a></h1>
 <p>In <em>auto-sklearn</em> it is possible to use different resampling strategies
 by specifying the arguments <code class="docutils literal notranslate"><span class="pre">resampling_strategy</span></code> and
@@ -142,15 +141,15 @@
 <span class="kn">import</span> <span class="nn">autosklearn.classification</span>
 </pre></div>
 </div>
-<section id="data-loading">
+<div class="section" id="data-loading">
 <h2>Data Loading<a class="headerlink" href="#data-loading" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">X</span><span class="p">,</span> <span class="n">y</span> <span class="o">=</span> <span class="n">sklearn</span><span class="o">.</span><span class="n">datasets</span><span class="o">.</span><span class="n">load_breast_cancer</span><span class="p">(</span><span class="n">return_X_y</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
 <span class="n">X_train</span><span class="p">,</span> <span class="n">X_test</span><span class="p">,</span> <span class="n">y_train</span><span class="p">,</span> <span class="n">y_test</span> <span class="o">=</span> \
     <span class="n">sklearn</span><span class="o">.</span><span class="n">model_selection</span><span class="o">.</span><span class="n">train_test_split</span><span class="p">(</span><span class="n">X</span><span class="p">,</span> <span class="n">y</span><span class="p">,</span> <span class="n">random_state</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
 </pre></div>
 </div>
-</section>
-<section id="holdout">
+</div>
+<div class="section" id="holdout">
 <h2>Holdout<a class="headerlink" href="#holdout" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">automl</span> <span class="o">=</span> <span class="n">autosklearn</span><span class="o">.</span><span class="n">classification</span><span class="o">.</span><span class="n">AutoSklearnClassifier</span><span class="p">(</span>
     <span class="n">time_left_for_this_task</span><span class="o">=</span><span class="mi">120</span><span class="p">,</span>
@@ -167,25 +166,30 @@ <h2>Holdout<a class="headerlink" href="#holdout" title="Permalink to this headli
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>AutoSklearnClassifier(per_run_time_limit=30,
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:68: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+  self.metafeatures = self.metafeatures.append(metafeatures)
+/home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:72: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+  self.algorithm_runs[metric].append(runs)
+
+AutoSklearnClassifier(per_run_time_limit=30,
                       resampling_strategy_arguments={&#39;train_size&#39;: 0.67},
                       time_left_for_this_task=120,
                       tmp_folder=&#39;/tmp/autosklearn_resampling_example_tmp&#39;)
 </pre></div>
 </div>
-</section>
-<section id="get-the-score-of-the-final-ensemble">
+</div>
+<div class="section" id="get-the-score-of-the-final-ensemble">
 <h2>Get the Score of the final ensemble<a class="headerlink" href="#get-the-score-of-the-final-ensemble" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">predictions</span> <span class="o">=</span> <span class="n">automl</span><span class="o">.</span><span class="n">predict</span><span class="p">(</span><span class="n">X_test</span><span class="p">)</span>
 <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Accuracy score holdout: &quot;</span><span class="p">,</span> <span class="n">sklearn</span><span class="o">.</span><span class="n">metrics</span><span class="o">.</span><span class="n">accuracy_score</span><span class="p">(</span><span class="n">y_test</span><span class="p">,</span> <span class="n">predictions</span><span class="p">))</span>
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Accuracy score holdout:  0.9440559440559441
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Accuracy score holdout:  0.958041958041958
 </pre></div>
 </div>
-</section>
-<section id="cross-validation">
+</div>
+<div class="section" id="cross-validation">
 <h2>Cross-validation<a class="headerlink" href="#cross-validation" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">automl</span> <span class="o">=</span> <span class="n">autosklearn</span><span class="o">.</span><span class="n">classification</span><span class="o">.</span><span class="n">AutoSklearnClassifier</span><span class="p">(</span>
     <span class="n">time_left_for_this_task</span><span class="o">=</span><span class="mi">120</span><span class="p">,</span>
@@ -207,12 +211,16 @@ <h2>Cross-validation<a class="headerlink" href="#cross-validation" title="Permal
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Before re-fit
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:68: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+  self.metafeatures = self.metafeatures.append(metafeatures)
+/home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:72: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+  self.algorithm_runs[metric].append(runs)
+Before re-fit
 Accuracy score CV 0.965034965034965
 </pre></div>
 </div>
-</section>
-<section id="perform-a-refit">
+</div>
+<div class="section" id="perform-a-refit">
 <h2>Perform a refit<a class="headerlink" href="#perform-a-refit" title="Permalink to this headline">¶</a></h2>
 <p>During fit(), models are fit on individual cross-validation folds. To use
 all available data, we call refit() which trains all models in the
@@ -225,11 +233,11 @@ <h2>Perform a refit<a class="headerlink" href="#perform-a-refit" title="Permalin
 </div>
 <p class="sphx-glr-script-out">Out:</p>
 <div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>After re-fit
-Accuracy score CV 0.958041958041958
+Accuracy score CV 0.965034965034965
 </pre></div>
 </div>
-</section>
-<section id="scikit-learn-splitter-objects">
+</div>
+<div class="section" id="scikit-learn-splitter-objects">
 <h2>scikit-learn splitter objects<a class="headerlink" href="#scikit-learn-splitter-objects" title="Permalink to this headline">¶</a></h2>
 <p>It is also possible to use
 <a class="reference external" href="https://scikit-learn.org/stable/modules/classes.html#splitter-classes">scikit-learn’s splitter classes</a> to further customize the outputs. In case one needs to have 100% control over the
@@ -238,8 +246,9 @@ <h2>scikit-learn splitter objects<a class="headerlink" href="#scikit-learn-split
 <p>Below is an example of using a predefined split. We split the training
 data by the first feature. In practice, one would use a splitting according
 to the use case at hand.</p>
-<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">resampling_strategy</span> <span class="o">=</span> <span class="n">sklearn</span><span class="o">.</span><span class="n">model_selection</span><span class="o">.</span><span class="n">PredefinedSplit</span><span class="p">(</span>
-    <span class="n">test_fold</span><span class="o">=</span><span class="n">np</span><span class="o">.</span><span class="n">where</span><span class="p">(</span><span class="n">X_train</span><span class="p">[:,</span> <span class="mi">0</span><span class="p">]</span> <span class="o">&lt;</span> <span class="n">np</span><span class="o">.</span><span class="n">mean</span><span class="p">(</span><span class="n">X_train</span><span class="p">[:,</span> <span class="mi">0</span><span class="p">]))[</span><span class="mi">0</span><span class="p">]</span>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">selected_indices</span> <span class="o">=</span> <span class="p">(</span><span class="n">X_train</span><span class="p">[:,</span> <span class="mi">0</span><span class="p">]</span> <span class="o">&lt;</span> <span class="n">np</span><span class="o">.</span><span class="n">mean</span><span class="p">(</span><span class="n">X_train</span><span class="p">[:,</span> <span class="mi">0</span><span class="p">]))</span><span class="o">.</span><span class="n">astype</span><span class="p">(</span><span class="nb">int</span><span class="p">)</span>
+<span class="n">resampling_strategy</span> <span class="o">=</span> <span class="n">sklearn</span><span class="o">.</span><span class="n">model_selection</span><span class="o">.</span><span class="n">PredefinedSplit</span><span class="p">(</span>
+    <span class="n">test_fold</span><span class="o">=</span><span class="n">selected_indices</span>
 <span class="p">)</span>
 
 <span class="n">automl</span> <span class="o">=</span> <span class="n">autosklearn</span><span class="o">.</span><span class="n">classification</span><span class="o">.</span><span class="n">AutoSklearnClassifier</span><span class="p">(</span>
@@ -250,13 +259,24 @@ <h2>scikit-learn splitter objects<a class="headerlink" href="#scikit-learn-split
     <span class="n">resampling_strategy</span><span class="o">=</span><span class="n">resampling_strategy</span><span class="p">,</span>
 <span class="p">)</span>
 <span class="n">automl</span><span class="o">.</span><span class="n">fit</span><span class="p">(</span><span class="n">X_train</span><span class="p">,</span> <span class="n">y_train</span><span class="p">,</span> <span class="n">dataset_name</span><span class="o">=</span><span class="s1">&#39;breast_cancer&#39;</span><span class="p">)</span>
+
+<span class="nb">print</span><span class="p">(</span><span class="n">automl</span><span class="o">.</span><span class="n">sprint_statistics</span><span class="p">())</span>
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>AutoSklearnClassifier(per_run_time_limit=30,
-                      resampling_strategy=PredefinedSplit(test_fold=array([  3,   4, ..., 424, 425])),
-                      time_left_for_this_task=120,
-                      tmp_folder=&#39;/tmp/autosklearn_resampling_example_tmp&#39;)
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:68: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+  self.metafeatures = self.metafeatures.append(metafeatures)
+/home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:72: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+  self.algorithm_runs[metric].append(runs)
+auto-sklearn results:
+  Dataset name: breast_cancer
+  Metric: accuracy
+  Best validation score: 0.964789
+  Number of target algorithm runs: 31
+  Number of successful target algorithm runs: 31
+  Number of crashed target algorithm runs: 0
+  Number of target algorithms that exceeded the time limit: 0
+  Number of target algorithms that exceeded the memory limit: 0
 </pre></div>
 </div>
 <p>For custom resampling strategies (i.e. resampling strategies that are not
@@ -266,13 +286,13 @@ <h2>scikit-learn splitter objects<a class="headerlink" href="#scikit-learn-split
 </div>
 <p class="sphx-glr-script-out">Out:</p>
 <div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>AutoSklearnClassifier(per_run_time_limit=30,
-                      resampling_strategy=PredefinedSplit(test_fold=array([  3,   4, ..., 424, 425])),
+                      resampling_strategy=PredefinedSplit(test_fold=array([0, 0, ..., 1, 1])),
                       time_left_for_this_task=120,
                       tmp_folder=&#39;/tmp/autosklearn_resampling_example_tmp&#39;)
 </pre></div>
 </div>
-</section>
-<section id="get-the-score-of-the-final-ensemble-again">
+</div>
+<div class="section" id="get-the-score-of-the-final-ensemble-again">
 <h2>Get the Score of the final ensemble (again)<a class="headerlink" href="#get-the-score-of-the-final-ensemble-again" title="Permalink to this headline">¶</a></h2>
 <p>Obviously, this score is pretty bad as we “destroyed” the dataset by
 splitting it on the first feature.</p>
@@ -281,10 +301,10 @@ <h2>Get the Score of the final ensemble (again)<a class="headerlink" href="#get-
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Accuracy score custom split 0.38461538461538464
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Accuracy score custom split 0.958041958041958
 </pre></div>
 </div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 6 minutes  22.500 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 6 minutes  27.008 seconds)</p>
 <div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-examples-40-advanced-example-resampling-py">
 <div class="binder-badge docutils container">
 <a class="reference external image-reference" href="https://mybinder.org/v2/gh/automl/auto-sklearn/master?urlpath=lab/tree/notebooks/examples/40_advanced/example_resampling.ipynb"><img alt="Launch binder" src="../../_images/binder_badge_logo1.svg" width="150px" /></a>
@@ -297,8 +317,8 @@ <h2>Get the Score of the final ensemble (again)<a class="headerlink" href="#get-
 </div>
 </div>
 <p class="sphx-glr-signature"><a class="reference external" href="https://sphinx-gallery.github.io">Gallery generated by Sphinx-Gallery</a></p>
-</section>
-</section>
+</div>
+</div>
 
 
     </div>
@@ -319,7 +339,7 @@ <h2>Get the Score of the final ensemble (again)<a class="headerlink" href="#get-
       
     </p>
     <p>
-        &copy; Copyright 2014-2021, Machine Learning Professorship Freiburg.<br/>
+        &copy; Copyright 2014-2022, Machine Learning Professorship Freiburg.<br/>
       Created using <a href="http://sphinx-doc.org/">Sphinx</a> 4.2.0.<br/>
     </p>
   </div>
diff --git a/master/examples/40_advanced/example_single_configuration.html b/master/examples/40_advanced/example_single_configuration.html
index 85e52edfdf..9ec78492a5 100644
--- a/master/examples/40_advanced/example_single_configuration.html
+++ b/master/examples/40_advanced/example_single_configuration.html
@@ -3,9 +3,8 @@
 <html>
   <head>
     <meta charset="utf-8" />
-    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="generator" content="Docutils 0.17.1: http://docutils.sourceforge.net/" />
-
-    <title>Fit a single configuration &#8212; AutoSklearn 0.14.2 documentation</title>
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>Fit a single configuration &#8212; AutoSklearn 0.14.4 documentation</title>
     <link rel="stylesheet" type="text/css" href="../../_static/pygments.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/bootstrap-sphinx.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/sg_gallery.css" />
@@ -55,7 +54,7 @@
         </button>
         <a class="navbar-brand" href="../../index.html">
           auto-sklearn</a>
-        <span class="navbar-text navbar-version pull-left"><b>0.14.2</b></span>
+        <span class="navbar-text navbar-version pull-left"><b>0.14.4</b></span>
       </div>
 
         <div class="collapse navbar-collapse nav-collapse">
@@ -124,7 +123,7 @@
 <p>Click <a class="reference internal" href="#sphx-glr-download-examples-40-advanced-example-single-configuration-py"><span class="std std-ref">here</span></a>
 to download the full example code or to run this example in your browser via Binder</p>
 </div>
-<section class="sphx-glr-example-title" id="fit-a-single-configuration">
+<div class="sphx-glr-example-title section" id="fit-a-single-configuration">
 <span id="sphx-glr-examples-40-advanced-example-single-configuration-py"></span><h1>Fit a single configuration<a class="headerlink" href="#fit-a-single-configuration" title="Permalink to this headline">¶</a></h1>
 <p><em>Auto-sklearn</em> searches for the best combination of machine learning algorithms
 and their hyper-parameter configuration for a given task, using Scikit-Learn Pipelines.
@@ -144,7 +143,7 @@
 <span class="kn">import</span> <span class="nn">autosklearn.classification</span>
 </pre></div>
 </div>
-<section id="data-loading">
+<div class="section" id="data-loading">
 <h2>Data Loading<a class="headerlink" href="#data-loading" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">X</span><span class="p">,</span> <span class="n">y</span> <span class="o">=</span> <span class="n">sklearn</span><span class="o">.</span><span class="n">datasets</span><span class="o">.</span><span class="n">fetch_openml</span><span class="p">(</span><span class="n">data_id</span><span class="o">=</span><span class="mi">3</span><span class="p">,</span> <span class="n">return_X_y</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span> <span class="n">as_frame</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
 <span class="n">X_train</span><span class="p">,</span> <span class="n">X_test</span><span class="p">,</span> <span class="n">y_train</span><span class="p">,</span> <span class="n">y_test</span> <span class="o">=</span> <span class="n">sklearn</span><span class="o">.</span><span class="n">model_selection</span><span class="o">.</span><span class="n">train_test_split</span><span class="p">(</span>
@@ -152,8 +151,8 @@ <h2>Data Loading<a class="headerlink" href="#data-loading" title="Permalink to t
 <span class="p">)</span>
 </pre></div>
 </div>
-</section>
-<section id="define-an-estimator">
+</div>
+<div class="section" id="define-an-estimator">
 <h2>Define an estimator<a class="headerlink" href="#define-an-estimator" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="bp">cls</span> <span class="o">=</span> <span class="n">autosklearn</span><span class="o">.</span><span class="n">classification</span><span class="o">.</span><span class="n">AutoSklearnClassifier</span><span class="p">(</span>
     <span class="n">time_left_for_this_task</span><span class="o">=</span><span class="mi">120</span><span class="p">,</span>
@@ -169,8 +168,8 @@ <h2>Define an estimator<a class="headerlink" href="#define-an-estimator" title="
 <span class="p">)</span>
 </pre></div>
 </div>
-</section>
-<section id="fit-an-user-provided-configuration">
+</div>
+<div class="section" id="fit-an-user-provided-configuration">
 <h2>Fit an user provided configuration<a class="headerlink" href="#fit-an-user-provided-configuration" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="c1"># We will create a configuration that has a user defined</span>
 <span class="c1"># min_samples_split in the Random Forest. We recommend you to look into</span>
@@ -209,64 +208,76 @@ <h2>Fit an user provided configuration<a class="headerlink" href="#fit-an-user-p
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>{&#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7f937c5a9ca0&gt;, &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;), &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7f937b0d0e50&gt;, &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7f937b0d0a00&gt;}
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>{&#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78ffee460&gt;, &#39;balancing&#39;: Balancing(random_state=1), &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78ab79cd0&gt;, &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78aae48b0&gt;}
 RunInfo(config=Configuration:
-  balancing:strategy, Value: &#39;weighting&#39;
+  balancing:strategy, Value: &#39;none&#39;
   classifier:__choice__, Value: &#39;random_forest&#39;
-  classifier:random_forest:bootstrap, Value: &#39;False&#39;
-  classifier:random_forest:criterion, Value: &#39;gini&#39;
+  classifier:random_forest:bootstrap, Value: &#39;True&#39;
+  classifier:random_forest:criterion, Value: &#39;entropy&#39;
   classifier:random_forest:max_depth, Constant: &#39;None&#39;
-  classifier:random_forest:max_features, Value: 0.40665293991189455
+  classifier:random_forest:max_features, Value: 0.7056146361355262
   classifier:random_forest:max_leaf_nodes, Constant: &#39;None&#39;
   classifier:random_forest:min_impurity_decrease, Constant: 0.0
-  classifier:random_forest:min_samples_leaf, Value: 13
+  classifier:random_forest:min_samples_leaf, Value: 11
   classifier:random_forest:min_samples_split, Value: 11
   classifier:random_forest:min_weight_fraction_leaf, Constant: 0.0
   data_preprocessor:__choice__, Value: &#39;feature_type&#39;
-  data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__, Value: &#39;no_encoding&#39;
-  data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__, Value: &#39;minority_coalescer&#39;
-  data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction, Value: 0.0012559712452161484
-  data_preprocessor:feature_type:numerical_transformer:imputation:strategy, Value: &#39;median&#39;
-  data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__, Value: &#39;quantile_transformer&#39;
-  data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles, Value: 204
-  data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution, Value: &#39;normal&#39;
-  feature_preprocessor:__choice__, Value: &#39;polynomial&#39;
-  feature_preprocessor:polynomial:degree, Value: 2
-  feature_preprocessor:polynomial:include_bias, Value: &#39;False&#39;
-  feature_preprocessor:polynomial:interaction_only, Value: &#39;True&#39;
+  data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__, Value: &#39;one_hot_encoding&#39;
+  data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__, Value: &#39;no_coalescense&#39;
+  data_preprocessor:feature_type:numerical_transformer:imputation:strategy, Value: &#39;mean&#39;
+  data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__, Value: &#39;none&#39;
+  feature_preprocessor:__choice__, Value: &#39;extra_trees_preproc_for_classification&#39;
+  feature_preprocessor:extra_trees_preproc_for_classification:bootstrap, Value: &#39;False&#39;
+  feature_preprocessor:extra_trees_preproc_for_classification:criterion, Value: &#39;gini&#39;
+  feature_preprocessor:extra_trees_preproc_for_classification:max_depth, Constant: &#39;None&#39;
+  feature_preprocessor:extra_trees_preproc_for_classification:max_features, Value: 0.41746352744644133
+  feature_preprocessor:extra_trees_preproc_for_classification:max_leaf_nodes, Constant: &#39;None&#39;
+  feature_preprocessor:extra_trees_preproc_for_classification:min_impurity_decrease, Constant: 0.0
+  feature_preprocessor:extra_trees_preproc_for_classification:min_samples_leaf, Value: 5
+  feature_preprocessor:extra_trees_preproc_for_classification:min_samples_split, Value: 2
+  feature_preprocessor:extra_trees_preproc_for_classification:min_weight_fraction_leaf, Constant: 0.0
+  feature_preprocessor:extra_trees_preproc_for_classification:n_estimators, Constant: 100
 , instance=None, instance_specific=None, seed=1, cutoff=60, capped=False, budget=0.0, source_id=0)
-RunValue(cost=0.012322274881516604, time=2.914306402206421, status=&lt;StatusType.SUCCESS: 1&gt;, starttime=1636487580.5942185, endtime=1636487583.5312755, additional_info={&#39;duration&#39;: 2.8303675651550293, &#39;num_run&#39;: 2, &#39;train_loss&#39;: 0.012143858010275621, &#39;configuration_origin&#39;: None})
+RunValue(cost=0.03033175355450235, time=2.2301056385040283, status=&lt;StatusType.SUCCESS: 1&gt;, starttime=1643127754.4012537, endtime=1643127756.6555753, additional_info={&#39;duration&#39;: 2.1325058937072754, &#39;num_run&#39;: 2, &#39;train_loss&#39;: 0.047641289117234975, &#39;configuration_origin&#39;: None})
 Passed Configuration: Configuration:
-  balancing:strategy, Value: &#39;weighting&#39;
+  balancing:strategy, Value: &#39;none&#39;
   classifier:__choice__, Value: &#39;random_forest&#39;
-  classifier:random_forest:bootstrap, Value: &#39;False&#39;
-  classifier:random_forest:criterion, Value: &#39;gini&#39;
+  classifier:random_forest:bootstrap, Value: &#39;True&#39;
+  classifier:random_forest:criterion, Value: &#39;entropy&#39;
   classifier:random_forest:max_depth, Constant: &#39;None&#39;
-  classifier:random_forest:max_features, Value: 0.40665293991189455
+  classifier:random_forest:max_features, Value: 0.7056146361355262
   classifier:random_forest:max_leaf_nodes, Constant: &#39;None&#39;
   classifier:random_forest:min_impurity_decrease, Constant: 0.0
-  classifier:random_forest:min_samples_leaf, Value: 13
+  classifier:random_forest:min_samples_leaf, Value: 11
   classifier:random_forest:min_samples_split, Value: 11
   classifier:random_forest:min_weight_fraction_leaf, Constant: 0.0
   data_preprocessor:__choice__, Value: &#39;feature_type&#39;
-  data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__, Value: &#39;no_encoding&#39;
-  data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__, Value: &#39;minority_coalescer&#39;
-  data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction, Value: 0.0012559712452161484
-  data_preprocessor:feature_type:numerical_transformer:imputation:strategy, Value: &#39;median&#39;
-  data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__, Value: &#39;quantile_transformer&#39;
-  data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles, Value: 204
-  data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution, Value: &#39;normal&#39;
-  feature_preprocessor:__choice__, Value: &#39;polynomial&#39;
-  feature_preprocessor:polynomial:degree, Value: 2
-  feature_preprocessor:polynomial:include_bias, Value: &#39;False&#39;
-  feature_preprocessor:polynomial:interaction_only, Value: &#39;True&#39;
+  data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__, Value: &#39;one_hot_encoding&#39;
+  data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__, Value: &#39;no_coalescense&#39;
+  data_preprocessor:feature_type:numerical_transformer:imputation:strategy, Value: &#39;mean&#39;
+  data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__, Value: &#39;none&#39;
+  feature_preprocessor:__choice__, Value: &#39;extra_trees_preproc_for_classification&#39;
+  feature_preprocessor:extra_trees_preproc_for_classification:bootstrap, Value: &#39;False&#39;
+  feature_preprocessor:extra_trees_preproc_for_classification:criterion, Value: &#39;gini&#39;
+  feature_preprocessor:extra_trees_preproc_for_classification:max_depth, Constant: &#39;None&#39;
+  feature_preprocessor:extra_trees_preproc_for_classification:max_features, Value: 0.41746352744644133
+  feature_preprocessor:extra_trees_preproc_for_classification:max_leaf_nodes, Constant: &#39;None&#39;
+  feature_preprocessor:extra_trees_preproc_for_classification:min_impurity_decrease, Constant: 0.0
+  feature_preprocessor:extra_trees_preproc_for_classification:min_samples_leaf, Value: 5
+  feature_preprocessor:extra_trees_preproc_for_classification:min_samples_split, Value: 2
+  feature_preprocessor:extra_trees_preproc_for_classification:min_weight_fraction_leaf, Constant: 0.0
+  feature_preprocessor:extra_trees_preproc_for_classification:n_estimators, Constant: 100
 
-Random Forest: RandomForestClassifier(bootstrap=False, max_features=14, min_samples_leaf=13,
+Random Forest: RandomForestClassifier(criterion=&#39;entropy&#39;, max_features=6, min_samples_leaf=11,
                        min_samples_split=11, n_estimators=512, n_jobs=1,
                        random_state=1, warm_start=True)
+/home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:68: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+  self.metafeatures = self.metafeatures.append(metafeatures)
+/home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:72: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+  self.algorithm_runs[metric].append(runs)
 </pre></div>
 </div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes  24.877 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes  5.409 seconds)</p>
 <div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-examples-40-advanced-example-single-configuration-py">
 <div class="binder-badge docutils container">
 <a class="reference external image-reference" href="https://mybinder.org/v2/gh/automl/auto-sklearn/master?urlpath=lab/tree/notebooks/examples/40_advanced/example_single_configuration.ipynb"><img alt="Launch binder" src="../../_images/binder_badge_logo1.svg" width="150px" /></a>
@@ -279,8 +290,8 @@ <h2>Fit an user provided configuration<a class="headerlink" href="#fit-an-user-p
 </div>
 </div>
 <p class="sphx-glr-signature"><a class="reference external" href="https://sphinx-gallery.github.io">Gallery generated by Sphinx-Gallery</a></p>
-</section>
-</section>
+</div>
+</div>
 
 
     </div>
@@ -301,7 +312,7 @@ <h2>Fit an user provided configuration<a class="headerlink" href="#fit-an-user-p
       
     </p>
     <p>
-        &copy; Copyright 2014-2021, Machine Learning Professorship Freiburg.<br/>
+        &copy; Copyright 2014-2022, Machine Learning Professorship Freiburg.<br/>
       Created using <a href="http://sphinx-doc.org/">Sphinx</a> 4.2.0.<br/>
     </p>
   </div>
diff --git a/master/examples/40_advanced/sg_execution_times.html b/master/examples/40_advanced/sg_execution_times.html
index d13123a6d0..af1c0e7504 100644
--- a/master/examples/40_advanced/sg_execution_times.html
+++ b/master/examples/40_advanced/sg_execution_times.html
@@ -3,9 +3,8 @@
 <html>
   <head>
     <meta charset="utf-8" />
-    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="generator" content="Docutils 0.17.1: http://docutils.sourceforge.net/" />
-
-    <title>Computation times &#8212; AutoSklearn 0.14.2 documentation</title>
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>Computation times &#8212; AutoSklearn 0.14.4 documentation</title>
     <link rel="stylesheet" type="text/css" href="../../_static/pygments.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/bootstrap-sphinx.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/sg_gallery.css" />
@@ -55,7 +54,7 @@
         </button>
         <a class="navbar-brand" href="../../index.html">
           auto-sklearn</a>
-        <span class="navbar-text navbar-version pull-left"><b>0.14.2</b></span>
+        <span class="navbar-text navbar-version pull-left"><b>0.14.4</b></span>
       </div>
 
         <div class="collapse navbar-collapse nav-collapse">
@@ -114,9 +113,9 @@
       </div>
     <div class="body col-md-9 content" role="main">
       
-  <section id="computation-times">
+  <div class="section" id="computation-times">
 <span id="sphx-glr-examples-40-advanced-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>25:21.692</strong> total execution time for <strong>examples_40_advanced</strong> files:</p>
+<p><strong>23:57.125</strong> total execution time for <strong>examples_40_advanced</strong> files:</p>
 <table class="docutils align-default">
 <colgroup>
 <col style="width: 86%" />
@@ -125,48 +124,48 @@
 </colgroup>
 <tbody>
 <tr class="row-odd"><td><p><a class="reference internal" href="example_resampling.html#sphx-glr-examples-40-advanced-example-resampling-py"><span class="std std-ref">Resampling Strategies</span></a> (<code class="docutils literal notranslate"><span class="pre">example_resampling.py</span></code>)</p></td>
-<td><p>06:22.500</p></td>
+<td><p>06:27.008</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="example_metrics.html#sphx-glr-examples-40-advanced-example-metrics-py"><span class="std std-ref">Metrics</span></a> (<code class="docutils literal notranslate"><span class="pre">example_metrics.py</span></code>)</p></td>
-<td><p>05:19.475</p></td>
+<td><p>04:39.637</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-odd"><td><p><a class="reference internal" href="example_inspect_predictions.html#sphx-glr-examples-40-advanced-example-inspect-predictions-py"><span class="std std-ref">Model Explanation</span></a> (<code class="docutils literal notranslate"><span class="pre">example_inspect_predictions.py</span></code>)</p></td>
-<td><p>04:02.065</p></td>
+<td><p>04:06.567</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="example_single_configuration.html#sphx-glr-examples-40-advanced-example-single-configuration-py"><span class="std std-ref">Fit a single configuration</span></a> (<code class="docutils literal notranslate"><span class="pre">example_single_configuration.py</span></code>)</p></td>
-<td><p>02:24.877</p></td>
+<td><p>02:05.409</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-odd"><td><p><a class="reference internal" href="example_calc_multiple_metrics.html#sphx-glr-examples-40-advanced-example-calc-multiple-metrics-py"><span class="std std-ref">Metrics</span></a> (<code class="docutils literal notranslate"><span class="pre">example_calc_multiple_metrics.py</span></code>)</p></td>
-<td><p>02:08.166</p></td>
+<td><p>01:59.942</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="example_pandas_train_test.html#sphx-glr-examples-40-advanced-example-pandas-train-test-py"><span class="std std-ref">Performance-over-time plot</span></a> (<code class="docutils literal notranslate"><span class="pre">example_pandas_train_test.py</span></code>)</p></td>
-<td><p>02:01.039</p></td>
+<td><p>01:55.852</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-odd"><td><p><a class="reference internal" href="example_interpretable_models.html#sphx-glr-examples-40-advanced-example-interpretable-models-py"><span class="std std-ref">Interpretable models</span></a> (<code class="docutils literal notranslate"><span class="pre">example_interpretable_models.py</span></code>)</p></td>
-<td><p>01:55.571</p></td>
+<td><p>01:54.044</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="example_get_pipeline_components.html#sphx-glr-examples-40-advanced-example-get-pipeline-components-py"><span class="std std-ref">Obtain run information</span></a> (<code class="docutils literal notranslate"><span class="pre">example_get_pipeline_components.py</span></code>)</p></td>
-<td><p>00:34.468</p></td>
+<td><p>00:24.928</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-odd"><td><p><a class="reference internal" href="example_debug_logging.html#sphx-glr-examples-40-advanced-example-debug-logging-py"><span class="std std-ref">Logging and debugging</span></a> (<code class="docutils literal notranslate"><span class="pre">example_debug_logging.py</span></code>)</p></td>
-<td><p>00:22.911</p></td>
+<td><p>00:14.975</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="example_feature_types.html#sphx-glr-examples-40-advanced-example-feature-types-py"><span class="std std-ref">Feature Types</span></a> (<code class="docutils literal notranslate"><span class="pre">example_feature_types.py</span></code>)</p></td>
-<td><p>00:10.620</p></td>
+<td><p>00:08.763</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 </tbody>
 </table>
-</section>
+</div>
 
 
     </div>
@@ -187,7 +186,7 @@
       
     </p>
     <p>
-        &copy; Copyright 2014-2021, Machine Learning Professorship Freiburg.<br/>
+        &copy; Copyright 2014-2022, Machine Learning Professorship Freiburg.<br/>
       Created using <a href="http://sphinx-doc.org/">Sphinx</a> 4.2.0.<br/>
     </p>
   </div>
diff --git a/master/examples/60_search/example_parallel_manual_spawning_cli.html b/master/examples/60_search/example_parallel_manual_spawning_cli.html
index ed3e5df7b9..526cc958d3 100644
--- a/master/examples/60_search/example_parallel_manual_spawning_cli.html
+++ b/master/examples/60_search/example_parallel_manual_spawning_cli.html
@@ -3,9 +3,8 @@
 <html>
   <head>
     <meta charset="utf-8" />
-    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="generator" content="Docutils 0.17.1: http://docutils.sourceforge.net/" />
-
-    <title>Parallel Usage: Spawning workers from the command line &#8212; AutoSklearn 0.14.2 documentation</title>
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>Parallel Usage: Spawning workers from the command line &#8212; AutoSklearn 0.14.4 documentation</title>
     <link rel="stylesheet" type="text/css" href="../../_static/pygments.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/bootstrap-sphinx.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/sg_gallery.css" />
@@ -55,7 +54,7 @@
         </button>
         <a class="navbar-brand" href="../../index.html">
           auto-sklearn</a>
-        <span class="navbar-text navbar-version pull-left"><b>0.14.2</b></span>
+        <span class="navbar-text navbar-version pull-left"><b>0.14.4</b></span>
       </div>
 
         <div class="collapse navbar-collapse nav-collapse">
@@ -131,7 +130,7 @@
 <p>Click <a class="reference internal" href="#sphx-glr-download-examples-60-search-example-parallel-manual-spawning-cli-py"><span class="std std-ref">here</span></a>
 to download the full example code or to run this example in your browser via Binder</p>
 </div>
-<section class="sphx-glr-example-title" id="parallel-usage-spawning-workers-from-the-command-line">
+<div class="sphx-glr-example-title section" id="parallel-usage-spawning-workers-from-the-command-line">
 <span id="sphx-glr-examples-60-search-example-parallel-manual-spawning-cli-py"></span><h1>Parallel Usage: Spawning workers from the command line<a class="headerlink" href="#parallel-usage-spawning-workers-from-the-command-line" title="Permalink to this headline">¶</a></h1>
 <p><em>Auto-sklearn</em> uses
 <a class="reference external" href="https://distributed.dask.org/en/latest/index.html">dask.distributed</a>
@@ -139,17 +138,19 @@
 <p>This example shows how to start the dask scheduler and spawn
 workers for <em>Auto-sklearn</em> manually from the command line. Use this example
 as a starting point to parallelize <em>Auto-sklearn</em> across multiple
-machines. If you want to start everything manually from within Python
-please see <a class="reference internal" href="example_parallel_manual_spawning_python.html#sphx-glr-examples-60-search-example-parallel-manual-spawning-python-py"><span class="std std-ref">Parallel Usage: Spawning workers from within Python</span></a>.
-To run <em>Auto-sklearn</em> in parallel on a single machine check out the example
+machines.</p>
+<p>To run <em>Auto-sklearn</em> in parallel on a single machine check out the example
 <a class="reference internal" href="example_parallel_n_jobs.html#sphx-glr-examples-60-search-example-parallel-n-jobs-py"><span class="std std-ref">Parallel Usage  on a single machine</span></a>.</p>
+<p>If you want to start everything manually from within Python
+please see <code class="docutils literal notranslate"><span class="pre">:ref:sphx_glr_examples_60_search_example_parallel_manual_spawning_python.py</span></code>.</p>
+<p><strong>NOTE:</strong> Above example is disabled due to issue <a class="reference external" href="https://github.com/dask/distributed/issues/5627">https://github.com/dask/distributed/issues/5627</a></p>
 <p>You can learn more about the dask command line interface from
 <a class="reference external" href="https://docs.dask.org/en/latest/setup/cli.html">https://docs.dask.org/en/latest/setup/cli.html</a>.</p>
 <p>When manually passing a dask client to Auto-sklearn, all logic
 must be guarded by <code class="docutils literal notranslate"><span class="pre">if</span> <span class="pre">__name__</span> <span class="pre">==</span> <span class="pre">&quot;__main__&quot;:</span></code> statements! We use
 multiple such statements to properly render this example as a notebook
 and also allow execution via the command line.</p>
-<section id="background">
+<div class="section" id="background">
 <h2>Background<a class="headerlink" href="#background" title="Permalink to this headline">¶</a></h2>
 <p>To run Auto-sklearn distributed on multiple machines we need to set
 up three components:</p>
@@ -169,8 +170,8 @@ <h2>Background<a class="headerlink" href="#background" title="Permalink to this
 be started from the command line and not from within a Python file via
 the <code class="docutils literal notranslate"><span class="pre">subprocess</span></code> module as done here (for the sake of having a self-contained
 example).</p>
-</section>
-<section id="import-statements">
+</div>
+<div class="section" id="import-statements">
 <h2>Import statements<a class="headerlink" href="#import-statements" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">multiprocessing</span>
 <span class="kn">import</span> <span class="nn">subprocess</span>
@@ -188,8 +189,8 @@ <h2>Import statements<a class="headerlink" href="#import-statements" title="Perm
 <span class="n">worker_processes</span> <span class="o">=</span> <span class="p">[]</span>
 </pre></div>
 </div>
-</section>
-<section id="setup-client-scheduler-communication">
+</div>
+<div class="section" id="setup-client-scheduler-communication">
 <h2>0. Setup client-scheduler communication<a class="headerlink" href="#setup-client-scheduler-communication" title="Permalink to this headline">¶</a></h2>
 <p>In this examples the dask scheduler is started without an explicit
 address and port. Instead, the scheduler takes a free port and stores
@@ -199,8 +200,8 @@ <h2>0. Setup client-scheduler communication<a class="headerlink" href="#setup-cl
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">scheduler_file_name</span> <span class="o">=</span> <span class="s1">&#39;scheduler-file.json&#39;</span>
 </pre></div>
 </div>
-</section>
-<section id="start-scheduler">
+</div>
+<div class="section" id="start-scheduler">
 <h2>1. Start scheduler<a class="headerlink" href="#start-scheduler" title="Permalink to this headline">¶</a></h2>
 <p>Starting the scheduler is done with the following bash command:</p>
 <div class="highlight-bash notranslate"><div class="highlight"><pre><span></span>dask-scheduler --scheduler-file scheduler-file.json --idle-timeout <span class="m">10</span>
@@ -230,8 +231,8 @@ <h2>1. Start scheduler<a class="headerlink" href="#start-scheduler" title="Perma
     <span class="n">time</span><span class="o">.</span><span class="n">sleep</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span>
 </pre></div>
 </div>
-</section>
-<section id="start-two-workers">
+</div>
+<div class="section" id="start-two-workers">
 <h2>2. Start two workers<a class="headerlink" href="#start-two-workers" title="Permalink to this headline">¶</a></h2>
 <p>Starting the scheduler is done with the following bash command:</p>
 <div class="highlight-bash notranslate"><div class="highlight"><pre><span></span><span class="nv">DASK_DISTRIBUTED__WORKER__DAEMON</span><span class="o">=</span>False <span class="se">\</span>
@@ -270,15 +271,15 @@ <h2>2. Start two workers<a class="headerlink" href="#start-two-workers" title="P
     <span class="n">time</span><span class="o">.</span><span class="n">sleep</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span>
 </pre></div>
 </div>
-</section>
-<section id="creating-a-client-in-python">
+</div>
+<div class="section" id="creating-a-client-in-python">
 <h2>3. Creating a client in Python<a class="headerlink" href="#creating-a-client-in-python" title="Permalink to this headline">¶</a></h2>
 <p>Finally we create a dask cluster which also connects to the scheduler via
 the information in the file created by the scheduler.</p>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">client</span> <span class="o">=</span> <span class="n">dask</span><span class="o">.</span><span class="n">distributed</span><span class="o">.</span><span class="n">Client</span><span class="p">(</span><span class="n">scheduler_file</span><span class="o">=</span><span class="n">scheduler_file_name</span><span class="p">)</span>
 </pre></div>
 </div>
-<section id="start-auto-sklearn">
+<div class="section" id="start-auto-sklearn">
 <h3>Start Auto-sklearn<a class="headerlink" href="#start-auto-sklearn" title="Permalink to this headline">¶</a></h3>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="k">if</span> <span class="vm">__name__</span> <span class="o">==</span> <span class="s2">&quot;__main__&quot;</span><span class="p">:</span>
     <span class="n">X</span><span class="p">,</span> <span class="n">y</span> <span class="o">=</span> <span class="n">sklearn</span><span class="o">.</span><span class="n">datasets</span><span class="o">.</span><span class="n">load_breast_cancer</span><span class="p">(</span><span class="n">return_X_y</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
@@ -313,22 +314,26 @@ <h3>Start Auto-sklearn<a class="headerlink" href="#start-auto-sklearn" title="Pe
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>auto-sklearn results:
-  Dataset name: 4433d19e-4199-11ec-8a32-1b4c952643c3
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:68: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+  self.metafeatures = self.metafeatures.append(metafeatures)
+/home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:72: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+  self.algorithm_runs[metric].append(runs)
+auto-sklearn results:
+  Dataset name: 74d6c3db-7dfd-11ec-875f-3137a7553b01
   Metric: accuracy
   Best validation score: 0.992908
-  Number of target algorithm runs: 12
-  Number of successful target algorithm runs: 11
+  Number of target algorithm runs: 14
+  Number of successful target algorithm runs: 13
   Number of crashed target algorithm runs: 0
   Number of target algorithms that exceeded the time limit: 1
   Number of target algorithms that exceeded the memory limit: 0
 
-Accuracy score 0.965034965034965
+Accuracy score 0.958041958041958
 </pre></div>
 </div>
-</section>
-</section>
-<section id="wait-until-all-workers-are-closed">
+</div>
+</div>
+<div class="section" id="wait-until-all-workers-are-closed">
 <h2>Wait until all workers are closed<a class="headerlink" href="#wait-until-all-workers-are-closed" title="Permalink to this headline">¶</a></h2>
 <p>This is only necessary if the workers are started from within this python
 script. In a real application one would start them directly from the command
@@ -339,7 +344,7 @@ <h2>Wait until all workers are closed<a class="headerlink" href="#wait-until-all
         <span class="n">process</span><span class="o">.</span><span class="n">join</span><span class="p">()</span>
 </pre></div>
 </div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 0 minutes  39.363 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 0 minutes  39.167 seconds)</p>
 <div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-examples-60-search-example-parallel-manual-spawning-cli-py">
 <div class="binder-badge docutils container">
 <a class="reference external image-reference" href="https://mybinder.org/v2/gh/automl/auto-sklearn/master?urlpath=lab/tree/notebooks/examples/60_search/example_parallel_manual_spawning_cli.ipynb"><img alt="Launch binder" src="../../_images/binder_badge_logo2.svg" width="150px" /></a>
@@ -352,8 +357,8 @@ <h2>Wait until all workers are closed<a class="headerlink" href="#wait-until-all
 </div>
 </div>
 <p class="sphx-glr-signature"><a class="reference external" href="https://sphinx-gallery.github.io">Gallery generated by Sphinx-Gallery</a></p>
-</section>
-</section>
+</div>
+</div>
 
 
     </div>
@@ -374,7 +379,7 @@ <h2>Wait until all workers are closed<a class="headerlink" href="#wait-until-all
       
     </p>
     <p>
-        &copy; Copyright 2014-2021, Machine Learning Professorship Freiburg.<br/>
+        &copy; Copyright 2014-2022, Machine Learning Professorship Freiburg.<br/>
       Created using <a href="http://sphinx-doc.org/">Sphinx</a> 4.2.0.<br/>
     </p>
   </div>
diff --git a/master/examples/60_search/example_parallel_manual_spawning_python.html b/master/examples/60_search/example_parallel_manual_spawning_python.html
deleted file mode 100644
index f6ae93e719..0000000000
--- a/master/examples/60_search/example_parallel_manual_spawning_python.html
+++ /dev/null
@@ -1,326 +0,0 @@
-<!DOCTYPE html>
-
-<html>
-  <head>
-    <meta charset="utf-8" />
-    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="generator" content="Docutils 0.17.1: http://docutils.sourceforge.net/" />
-
-    <title>Parallel Usage: Spawning workers from within Python &#8212; AutoSklearn 0.14.2 documentation</title>
-    <link rel="stylesheet" type="text/css" href="../../_static/pygments.css" />
-    <link rel="stylesheet" type="text/css" href="../../_static/bootstrap-sphinx.css" />
-    <link rel="stylesheet" type="text/css" href="../../_static/sg_gallery.css" />
-    <link rel="stylesheet" type="text/css" href="../../_static/sg_gallery-binder.css" />
-    <link rel="stylesheet" type="text/css" href="../../_static/sg_gallery-dataframe.css" />
-    <link rel="stylesheet" type="text/css" href="../../_static/sg_gallery-rendered-html.css" />
-    <script data-url_root="../../" id="documentation_options" src="../../_static/documentation_options.js"></script>
-    <script src="../../_static/jquery.js"></script>
-    <script src="../../_static/underscore.js"></script>
-    <script src="../../_static/doctools.js"></script>
-    <link rel="index" title="Index" href="../../genindex.html" />
-    <link rel="search" title="Search" href="../../search.html" />
-<meta charset='utf-8'>
-<meta http-equiv='X-UA-Compatible' content='IE=edge,chrome=1'>
-<meta name='viewport' content='width=device-width, initial-scale=1.0, maximum-scale=1'>
-<meta name="apple-mobile-web-app-capable" content="yes">
-<script type="text/javascript" src="../../_static/js/jquery-1.12.4.min.js "></script>
-<script type="text/javascript" src="../../_static/js/jquery-fix.js "></script>
-<script type="text/javascript" src="../../_static/bootstrap-3.4.1/js/bootstrap.min.js "></script>
-<script type="text/javascript" src="../../_static/bootstrap-sphinx.js "></script>
-
-  </head><body>
-  
-  <a href="https://github.com/automl/auto-sklearn"
-     class="visible-desktop hidden-xs"><img
-    id="gh-banner"
-    style="position: absolute; top: 50px; right: 0; border: 0;"
-    src="https://s3.amazonaws.com/github/ribbons/forkme_right_red_aa0000.png"
-    alt="Fork me on GitHub"></a>
-  <script>
-    // Adjust banner height.
-    $(function () {
-      var navHeight = $(".navbar .container").css("height");
-      $("#gh-banner").css("top", navHeight);
-    });
-  </script>
-
-
-  <div id="navbar" class="navbar navbar-default navbar-fixed-top">
-    <div class="container">
-      <div class="navbar-header">
-        <!-- .btn-navbar is used as the toggle for collapsed navbar content -->
-        <button type="button" class="navbar-toggle" data-toggle="collapse" data-target=".nav-collapse">
-          <span class="icon-bar"></span>
-          <span class="icon-bar"></span>
-          <span class="icon-bar"></span>
-        </button>
-        <a class="navbar-brand" href="../../index.html">
-          auto-sklearn</a>
-        <span class="navbar-text navbar-version pull-left"><b>0.14.2</b></span>
-      </div>
-
-        <div class="collapse navbar-collapse nav-collapse">
-          <ul class="nav navbar-nav">
-            
-                <li><a href="../../index.html">Start</a></li>
-                <li><a href="../../releases.html">Releases</a></li>
-                <li><a href="../../installation.html">Installation</a></li>
-                <li><a href="../../manual.html">Manual</a></li>
-                <li><a href="../index.html">Examples</a></li>
-                <li><a href="../../api.html">API</a></li>
-                <li><a href="../../extending.html">Extending</a></li>
-                <li><a href="../../faq.html">FAQ</a></li>
-            
-            
-              <li class="dropdown globaltoc-container">
-  <a role="button"
-     id="dLabelGlobalToc"
-     data-toggle="dropdown"
-     data-target="#"
-     href="../../index.html">Site <b class="caret"></b></a>
-  <ul class="dropdown-menu globaltoc"
-      role="menu"
-      aria-labelledby="dLabelGlobalToc"></ul>
-</li>
-              
-            
-            
-            
-            
-            
-          </ul>
-
-          
-            
-<form class="navbar-form navbar-right" action="../../search.html" method="get">
- <div class="form-group">
-  <input type="text" name="q" class="form-control" placeholder="Search" />
- </div>
-  <input type="hidden" name="check_keywords" value="yes" />
-  <input type="hidden" name="area" value="default" />
-</form>
-          
-        </div>
-    </div>
-  </div>
-
-<div class="container">
-  <div class="row">
-      <div class="col-md-3">
-        <div id="sidebar" class="bs-sidenav" role="complementary"><ul>
-<li><a class="reference internal" href="#">Parallel Usage: Spawning workers from within Python</a><ul>
-<li><a class="reference internal" href="#background">Background</a></li>
-<li><a class="reference internal" href="#define-function-to-start-worker">Define function to start worker</a></li>
-<li><a class="reference internal" href="#start-auto-sklearn">Start Auto-sklearn</a></li>
-</ul>
-</li>
-</ul>
-
-        </div>
-      </div>
-    <div class="body col-md-9 content" role="main">
-      
-  <div class="sphx-glr-download-link-note admonition note">
-<p class="admonition-title">Note</p>
-<p>Click <a class="reference internal" href="#sphx-glr-download-examples-60-search-example-parallel-manual-spawning-python-py"><span class="std std-ref">here</span></a>
-to download the full example code or to run this example in your browser via Binder</p>
-</div>
-<section class="sphx-glr-example-title" id="parallel-usage-spawning-workers-from-within-python">
-<span id="sphx-glr-examples-60-search-example-parallel-manual-spawning-python-py"></span><h1>Parallel Usage: Spawning workers from within Python<a class="headerlink" href="#parallel-usage-spawning-workers-from-within-python" title="Permalink to this headline">¶</a></h1>
-<p><em>Auto-sklearn</em> uses
-<a class="reference external" href="https://distributed.dask.org/en/latest/index.html">dask.distributed</a>
-for parallel optimization.</p>
-<p>This example shows how to start the dask scheduler and spawn
-workers for <em>Auto-sklearn</em> manually within Python. Use this example
-as a starting point to parallelize <em>Auto-sklearn</em> across multiple
-machines. If you want to start everything manually from the command line
-please see <a class="reference internal" href="example_parallel_manual_spawning_cli.html#sphx-glr-examples-60-search-example-parallel-manual-spawning-cli-py"><span class="std std-ref">Parallel Usage: Spawning workers from the command line</span></a>.
-To run <em>Auto-sklearn</em> in parallel on a single machine check out the example
-<a class="reference internal" href="example_parallel_n_jobs.html#sphx-glr-examples-60-search-example-parallel-n-jobs-py"><span class="std std-ref">Parallel Usage  on a single machine</span></a>.</p>
-<p>When manually passing a dask client to Auto-sklearn, all logic
-must be guarded by <code class="docutils literal notranslate"><span class="pre">if</span> <span class="pre">__name__</span> <span class="pre">==</span> <span class="pre">&quot;__main__&quot;:</span></code> statements! We use
-multiple such statements to properly render this example as a notebook
-and also allow execution via the command line.</p>
-<section id="background">
-<h2>Background<a class="headerlink" href="#background" title="Permalink to this headline">¶</a></h2>
-<p>To run Auto-sklearn distributed on multiple machines we need to set
-up three components:</p>
-<ol class="arabic simple">
-<li><p><strong>Auto-sklearn and a dask client</strong>. This will manage all workload, find new
-configurations to evaluate and submit jobs via a dask client. As this
-runs Bayesian optimization it should be executed on its own CPU.</p></li>
-<li><p><strong>The dask workers</strong>. They will do the actual work of running machine
-learning algorithms and require their own CPU each.</p></li>
-<li><p><strong>The scheduler</strong>. It manages the communication between the dask client
-and the different dask workers. As the client and all workers connect
-to the scheduler it must be started first. This is a light-weight job
-and does not require its own CPU.</p></li>
-</ol>
-<p>We will now start these three components in reverse order: scheduler,
-workers and client. Also, in a real setup, the scheduler and the workers should
-be started from the command line and not from within a Python file via
-the <code class="docutils literal notranslate"><span class="pre">subprocess</span></code> module as done here (for the sake of having a self-contained
-example).</p>
-<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">asyncio</span>
-<span class="kn">import</span> <span class="nn">multiprocessing</span>
-<span class="kn">import</span> <span class="nn">time</span>
-
-<span class="kn">import</span> <span class="nn">dask</span>
-<span class="kn">import</span> <span class="nn">dask.distributed</span>
-<span class="kn">import</span> <span class="nn">sklearn.datasets</span>
-<span class="kn">import</span> <span class="nn">sklearn.metrics</span>
-
-<span class="kn">from</span> <span class="nn">autosklearn.classification</span> <span class="kn">import</span> <span class="n">AutoSklearnClassifier</span>
-<span class="kn">from</span> <span class="nn">autosklearn.constants</span> <span class="kn">import</span> <span class="n">MULTICLASS_CLASSIFICATION</span>
-
-<span class="n">tmp_folder</span> <span class="o">=</span> <span class="s1">&#39;/tmp/autosklearn_parallel_2_example_tmp&#39;</span>
-</pre></div>
-</div>
-</section>
-<section id="define-function-to-start-worker">
-<h2>Define function to start worker<a class="headerlink" href="#define-function-to-start-worker" title="Permalink to this headline">¶</a></h2>
-<p>Define the function to start a dask worker from python. This
-is a bit cumbersome and should ideally be done from the command line.
-We do it here only for illustrational purpose.</p>
-<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="c1"># Check the dask docs at</span>
-<span class="c1"># https://docs.dask.org/en/latest/setup/python-advanced.html for further</span>
-<span class="c1"># information.</span>
-
-<span class="k">def</span> <span class="nf">start_python_worker</span><span class="p">(</span><span class="n">scheduler_address</span><span class="p">):</span>
-    <span class="n">dask</span><span class="o">.</span><span class="n">config</span><span class="o">.</span><span class="n">set</span><span class="p">({</span><span class="s1">&#39;distributed.worker.daemon&#39;</span><span class="p">:</span> <span class="kc">False</span><span class="p">})</span>
-
-    <span class="k">async</span> <span class="k">def</span> <span class="nf">do_work</span><span class="p">():</span>
-        <span class="k">async</span> <span class="k">with</span> <span class="n">dask</span><span class="o">.</span><span class="n">distributed</span><span class="o">.</span><span class="n">Nanny</span><span class="p">(</span>
-            <span class="n">scheduler_ip</span><span class="o">=</span><span class="n">scheduler_address</span><span class="p">,</span>
-            <span class="n">nthreads</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span>
-            <span class="n">lifetime</span><span class="o">=</span><span class="mi">35</span><span class="p">,</span>  <span class="c1"># automatically shut down the worker so this loop ends</span>
-            <span class="n">memory_limit</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span>  <span class="c1"># Disable memory management as it is done by Auto-sklearn itself</span>
-        <span class="p">)</span> <span class="k">as</span> <span class="n">worker</span><span class="p">:</span>
-            <span class="k">await</span> <span class="n">worker</span><span class="o">.</span><span class="n">finished</span><span class="p">()</span>
-
-    <span class="n">asyncio</span><span class="o">.</span><span class="n">get_event_loop</span><span class="p">()</span><span class="o">.</span><span class="n">run_until_complete</span><span class="p">(</span><span class="n">do_work</span><span class="p">())</span>
-</pre></div>
-</div>
-</section>
-<section id="start-auto-sklearn">
-<h2>Start Auto-sklearn<a class="headerlink" href="#start-auto-sklearn" title="Permalink to this headline">¶</a></h2>
-<p>We are now ready to start <em>auto-sklearn</em> and all dask related processes.</p>
-<p>To use auto-sklearn in parallel we must guard the code with
-<code class="docutils literal notranslate"><span class="pre">if</span> <span class="pre">__name__</span> <span class="pre">==</span> <span class="pre">'__main__'</span></code>. We then start a dask cluster as a context,
-which means that it is automatically stopped once all computation is done.</p>
-<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="k">if</span> <span class="vm">__name__</span> <span class="o">==</span> <span class="s1">&#39;__main__&#39;</span><span class="p">:</span>
-    <span class="n">X</span><span class="p">,</span> <span class="n">y</span> <span class="o">=</span> <span class="n">sklearn</span><span class="o">.</span><span class="n">datasets</span><span class="o">.</span><span class="n">load_breast_cancer</span><span class="p">(</span><span class="n">return_X_y</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
-    <span class="n">X_train</span><span class="p">,</span> <span class="n">X_test</span><span class="p">,</span> <span class="n">y_train</span><span class="p">,</span> <span class="n">y_test</span> <span class="o">=</span> \
-        <span class="n">sklearn</span><span class="o">.</span><span class="n">model_selection</span><span class="o">.</span><span class="n">train_test_split</span><span class="p">(</span><span class="n">X</span><span class="p">,</span> <span class="n">y</span><span class="p">,</span> <span class="n">random_state</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
-
-    <span class="c1"># 1. Create a dask scheduler (LocalCluster)</span>
-    <span class="k">with</span> <span class="n">dask</span><span class="o">.</span><span class="n">distributed</span><span class="o">.</span><span class="n">LocalCluster</span><span class="p">(</span>
-        <span class="n">n_workers</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span> <span class="n">processes</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span> <span class="n">threads_per_worker</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span>
-    <span class="p">)</span> <span class="k">as</span> <span class="n">cluster</span><span class="p">:</span>
-
-        <span class="c1"># 2. Start the workers</span>
-        <span class="c1"># now we start the two workers, one from within Python, the other</span>
-        <span class="c1"># via the command line.</span>
-        <span class="n">worker_processes</span> <span class="o">=</span> <span class="p">[]</span>
-        <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">2</span><span class="p">):</span>
-            <span class="n">process_python_worker</span> <span class="o">=</span> <span class="n">multiprocessing</span><span class="o">.</span><span class="n">Process</span><span class="p">(</span>
-                <span class="n">target</span><span class="o">=</span><span class="n">start_python_worker</span><span class="p">,</span>
-                <span class="n">args</span><span class="o">=</span><span class="p">(</span><span class="n">cluster</span><span class="o">.</span><span class="n">scheduler_address</span><span class="p">,</span> <span class="p">),</span>
-            <span class="p">)</span>
-            <span class="n">process_python_worker</span><span class="o">.</span><span class="n">start</span><span class="p">()</span>
-            <span class="n">worker_processes</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">process_python_worker</span><span class="p">)</span>
-
-        <span class="c1"># Wait a second for workers to become available</span>
-        <span class="n">time</span><span class="o">.</span><span class="n">sleep</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span>
-
-        <span class="c1"># 3. Start the client</span>
-        <span class="k">with</span> <span class="n">dask</span><span class="o">.</span><span class="n">distributed</span><span class="o">.</span><span class="n">Client</span><span class="p">(</span><span class="n">address</span><span class="o">=</span><span class="n">cluster</span><span class="o">.</span><span class="n">scheduler_address</span><span class="p">)</span> <span class="k">as</span> <span class="n">client</span><span class="p">:</span>
-            <span class="n">automl</span> <span class="o">=</span> <span class="n">AutoSklearnClassifier</span><span class="p">(</span>
-                <span class="n">delete_tmp_folder_after_terminate</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
-                <span class="n">time_left_for_this_task</span><span class="o">=</span><span class="mi">30</span><span class="p">,</span>
-                <span class="n">per_run_time_limit</span><span class="o">=</span><span class="mi">10</span><span class="p">,</span>
-                <span class="n">memory_limit</span><span class="o">=</span><span class="mi">1024</span><span class="p">,</span>
-                <span class="n">tmp_folder</span><span class="o">=</span><span class="n">tmp_folder</span><span class="p">,</span>
-                <span class="n">seed</span><span class="o">=</span><span class="mi">777</span><span class="p">,</span>
-                <span class="c1"># n_jobs is ignored internally as we pass a dask client.</span>
-                <span class="n">n_jobs</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span>
-                <span class="c1"># Pass a dask client which connects to the previously constructed cluster.</span>
-                <span class="n">dask_client</span><span class="o">=</span><span class="n">client</span><span class="p">,</span>
-            <span class="p">)</span>
-            <span class="n">automl</span><span class="o">.</span><span class="n">fit</span><span class="p">(</span><span class="n">X_train</span><span class="p">,</span> <span class="n">y_train</span><span class="p">)</span>
-
-            <span class="n">automl</span><span class="o">.</span><span class="n">fit_ensemble</span><span class="p">(</span>
-                <span class="n">y_train</span><span class="p">,</span>
-                <span class="n">task</span><span class="o">=</span><span class="n">MULTICLASS_CLASSIFICATION</span><span class="p">,</span>
-                <span class="n">dataset_name</span><span class="o">=</span><span class="s1">&#39;digits&#39;</span><span class="p">,</span>
-                <span class="n">ensemble_size</span><span class="o">=</span><span class="mi">20</span><span class="p">,</span>
-                <span class="n">ensemble_nbest</span><span class="o">=</span><span class="mi">50</span><span class="p">,</span>
-            <span class="p">)</span>
-
-        <span class="n">predictions</span> <span class="o">=</span> <span class="n">automl</span><span class="o">.</span><span class="n">predict</span><span class="p">(</span><span class="n">X_test</span><span class="p">)</span>
-        <span class="nb">print</span><span class="p">(</span><span class="n">automl</span><span class="o">.</span><span class="n">sprint_statistics</span><span class="p">())</span>
-        <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Accuracy score&quot;</span><span class="p">,</span> <span class="n">sklearn</span><span class="o">.</span><span class="n">metrics</span><span class="o">.</span><span class="n">accuracy_score</span><span class="p">(</span><span class="n">y_test</span><span class="p">,</span> <span class="n">predictions</span><span class="p">))</span>
-
-        <span class="c1"># Wait until all workers are closed</span>
-        <span class="k">for</span> <span class="n">process</span> <span class="ow">in</span> <span class="n">worker_processes</span><span class="p">:</span>
-            <span class="n">process_python_worker</span><span class="o">.</span><span class="n">join</span><span class="p">()</span>
-</pre></div>
-</div>
-<p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>[ERROR] [2021-11-09 20:12:34,562:asyncio] _GatheringFuture exception was never retrieved
-future: &lt;_GatheringFuture finished exception=CancelledError()&gt;
-asyncio.exceptions.CancelledError
-auto-sklearn results:
-  Dataset name: 5b2e548d-4199-11ec-8a32-1b4c952643c3
-  Metric: accuracy
-  Best validation score: 0.992908
-  Number of target algorithm runs: 14
-  Number of successful target algorithm runs: 12
-  Number of crashed target algorithm runs: 0
-  Number of target algorithms that exceeded the time limit: 2
-  Number of target algorithms that exceeded the memory limit: 0
-
-Accuracy score 0.965034965034965
-</pre></div>
-</div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 0 minutes  35.796 seconds)</p>
-<div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-examples-60-search-example-parallel-manual-spawning-python-py">
-<div class="binder-badge docutils container">
-<a class="reference external image-reference" href="https://mybinder.org/v2/gh/automl/auto-sklearn/master?urlpath=lab/tree/notebooks/examples/60_search/example_parallel_manual_spawning_python.ipynb"><img alt="Launch binder" src="../../_images/binder_badge_logo2.svg" width="150px" /></a>
-</div>
-<div class="sphx-glr-download sphx-glr-download-python docutils container">
-<p><a class="reference download internal" download="" href="../../_downloads/6507de4916ad481ba20cb60e6dc3309e/example_parallel_manual_spawning_python.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">example_parallel_manual_spawning_python.py</span></code></a></p>
-</div>
-<div class="sphx-glr-download sphx-glr-download-jupyter docutils container">
-<p><a class="reference download internal" download="" href="../../_downloads/986156f70b0b7eee4d669b791b35bd6d/example_parallel_manual_spawning_python.ipynb"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Jupyter</span> <span class="pre">notebook:</span> <span class="pre">example_parallel_manual_spawning_python.ipynb</span></code></a></p>
-</div>
-</div>
-<p class="sphx-glr-signature"><a class="reference external" href="https://sphinx-gallery.github.io">Gallery generated by Sphinx-Gallery</a></p>
-</section>
-</section>
-
-
-    </div>
-      
-  </div>
-</div>
-<footer class="footer">
-  <div class="container">
-    <p class="pull-right">
-      <a href="#">Back to top</a>
-      
-        <br/>
-        
-<div id="sourcelink">
-  <a href="../../_sources/examples/60_search/example_parallel_manual_spawning_python.rst.txt"
-     rel="nofollow">Source</a>
-</div>
-      
-    </p>
-    <p>
-        &copy; Copyright 2014-2021, Machine Learning Professorship Freiburg.<br/>
-      Created using <a href="http://sphinx-doc.org/">Sphinx</a> 4.2.0.<br/>
-    </p>
-  </div>
-</footer>
-  </body>
-</html>
\ No newline at end of file
diff --git a/master/examples/60_search/example_parallel_n_jobs.html b/master/examples/60_search/example_parallel_n_jobs.html
index dda3ac5c4b..a75c4327c5 100644
--- a/master/examples/60_search/example_parallel_n_jobs.html
+++ b/master/examples/60_search/example_parallel_n_jobs.html
@@ -3,9 +3,8 @@
 <html>
   <head>
     <meta charset="utf-8" />
-    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="generator" content="Docutils 0.17.1: http://docutils.sourceforge.net/" />
-
-    <title>Parallel Usage on a single machine &#8212; AutoSklearn 0.14.2 documentation</title>
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>Parallel Usage on a single machine &#8212; AutoSklearn 0.14.4 documentation</title>
     <link rel="stylesheet" type="text/css" href="../../_static/pygments.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/bootstrap-sphinx.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/sg_gallery.css" />
@@ -55,7 +54,7 @@
         </button>
         <a class="navbar-brand" href="../../index.html">
           auto-sklearn</a>
-        <span class="navbar-text navbar-version pull-left"><b>0.14.2</b></span>
+        <span class="navbar-text navbar-version pull-left"><b>0.14.4</b></span>
       </div>
 
         <div class="collapse navbar-collapse nav-collapse">
@@ -123,7 +122,7 @@
 <p>Click <a class="reference internal" href="#sphx-glr-download-examples-60-search-example-parallel-n-jobs-py"><span class="std std-ref">here</span></a>
 to download the full example code or to run this example in your browser via Binder</p>
 </div>
-<section class="sphx-glr-example-title" id="parallel-usage-on-a-single-machine">
+<div class="sphx-glr-example-title section" id="parallel-usage-on-a-single-machine">
 <span id="sphx-glr-examples-60-search-example-parallel-n-jobs-py"></span><h1>Parallel Usage  on a single machine<a class="headerlink" href="#parallel-usage-on-a-single-machine" title="Permalink to this headline">¶</a></h1>
 <p><em>Auto-sklearn</em> uses
 <cite>dask.distributed &lt;https://distributed.dask.org/en/latest/index.html</cite>&gt;_
@@ -141,15 +140,15 @@
 <span class="kn">import</span> <span class="nn">autosklearn.classification</span>
 </pre></div>
 </div>
-<section id="data-loading">
+<div class="section" id="data-loading">
 <h2>Data Loading<a class="headerlink" href="#data-loading" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">X</span><span class="p">,</span> <span class="n">y</span> <span class="o">=</span> <span class="n">sklearn</span><span class="o">.</span><span class="n">datasets</span><span class="o">.</span><span class="n">load_breast_cancer</span><span class="p">(</span><span class="n">return_X_y</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
 <span class="n">X_train</span><span class="p">,</span> <span class="n">X_test</span><span class="p">,</span> <span class="n">y_train</span><span class="p">,</span> <span class="n">y_test</span> <span class="o">=</span> \
     <span class="n">sklearn</span><span class="o">.</span><span class="n">model_selection</span><span class="o">.</span><span class="n">train_test_split</span><span class="p">(</span><span class="n">X</span><span class="p">,</span> <span class="n">y</span><span class="p">,</span> <span class="n">random_state</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
 </pre></div>
 </div>
-</section>
-<section id="build-and-fit-a-classifier">
+</div>
+<div class="section" id="build-and-fit-a-classifier">
 <h2>Build and fit a classifier<a class="headerlink" href="#build-and-fit-a-classifier" title="Permalink to this headline">¶</a></h2>
 <p>To use <code class="docutils literal notranslate"><span class="pre">n_jobs_</span></code> we must guard the code</p>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="k">if</span> <span class="vm">__name__</span> <span class="o">==</span> <span class="s1">&#39;__main__&#39;</span><span class="p">:</span>
@@ -171,18 +170,22 @@ <h2>Build and fit a classifier<a class="headerlink" href="#build-and-fit-a-class
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>auto-sklearn results:
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:68: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+  self.metafeatures = self.metafeatures.append(metafeatures)
+/home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:72: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+  self.algorithm_runs[metric].append(runs)
+auto-sklearn results:
   Dataset name: breast_cancer
   Metric: accuracy
   Best validation score: 0.985816
-  Number of target algorithm runs: 49
-  Number of successful target algorithm runs: 48
-  Number of crashed target algorithm runs: 0
-  Number of target algorithms that exceeded the time limit: 1
+  Number of target algorithm runs: 52
+  Number of successful target algorithm runs: 49
+  Number of crashed target algorithm runs: 1
+  Number of target algorithms that exceeded the time limit: 2
   Number of target algorithms that exceeded the memory limit: 0
 </pre></div>
 </div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes  24.442 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  57.601 seconds)</p>
 <div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-examples-60-search-example-parallel-n-jobs-py">
 <div class="binder-badge docutils container">
 <a class="reference external image-reference" href="https://mybinder.org/v2/gh/automl/auto-sklearn/master?urlpath=lab/tree/notebooks/examples/60_search/example_parallel_n_jobs.ipynb"><img alt="Launch binder" src="../../_images/binder_badge_logo2.svg" width="150px" /></a>
@@ -195,8 +198,8 @@ <h2>Build and fit a classifier<a class="headerlink" href="#build-and-fit-a-class
 </div>
 </div>
 <p class="sphx-glr-signature"><a class="reference external" href="https://sphinx-gallery.github.io">Gallery generated by Sphinx-Gallery</a></p>
-</section>
-</section>
+</div>
+</div>
 
 
     </div>
@@ -217,7 +220,7 @@ <h2>Build and fit a classifier<a class="headerlink" href="#build-and-fit-a-class
       
     </p>
     <p>
-        &copy; Copyright 2014-2021, Machine Learning Professorship Freiburg.<br/>
+        &copy; Copyright 2014-2022, Machine Learning Professorship Freiburg.<br/>
       Created using <a href="http://sphinx-doc.org/">Sphinx</a> 4.2.0.<br/>
     </p>
   </div>
diff --git a/master/examples/60_search/example_random_search.html b/master/examples/60_search/example_random_search.html
index 12f2c49149..191949e60a 100644
--- a/master/examples/60_search/example_random_search.html
+++ b/master/examples/60_search/example_random_search.html
@@ -3,9 +3,8 @@
 <html>
   <head>
     <meta charset="utf-8" />
-    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="generator" content="Docutils 0.17.1: http://docutils.sourceforge.net/" />
-
-    <title>Random Search &#8212; AutoSklearn 0.14.2 documentation</title>
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>Random Search &#8212; AutoSklearn 0.14.4 documentation</title>
     <link rel="stylesheet" type="text/css" href="../../_static/pygments.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/bootstrap-sphinx.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/sg_gallery.css" />
@@ -55,7 +54,7 @@
         </button>
         <a class="navbar-brand" href="../../index.html">
           auto-sklearn</a>
-        <span class="navbar-text navbar-version pull-left"><b>0.14.2</b></span>
+        <span class="navbar-text navbar-version pull-left"><b>0.14.4</b></span>
       </div>
 
         <div class="collapse navbar-collapse nav-collapse">
@@ -124,7 +123,7 @@
 <p>Click <a class="reference internal" href="#sphx-glr-download-examples-60-search-example-random-search-py"><span class="std std-ref">here</span></a>
 to download the full example code or to run this example in your browser via Binder</p>
 </div>
-<section class="sphx-glr-example-title" id="random-search">
+<div class="sphx-glr-example-title section" id="random-search">
 <span id="sphx-glr-examples-60-search-example-random-search-py"></span><h1>Random Search<a class="headerlink" href="#random-search" title="Permalink to this headline">¶</a></h1>
 <p>A crucial feature of <em>auto-sklearn</em> is automatically optimizing the hyperparameters through SMAC,
 introduced <a class="reference external" href="https://ml.informatik.uni-freiburg.de/papers/11-LION5-SMAC.pdf">here</a>.
@@ -134,7 +133,9 @@
 <a class="reference external" href="https://ml.informatik.uni-freiburg.de/papers/11-LION5-SMAC.pdf">Random Online Aggressive Racing (ROAR)</a>
 as yet another alternative optimizatino strategy.
 Both examples are intended to show how the optimization strategy in <em>auto-sklearn</em> can be adapted.</p>
-<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">sklearn.model_selection</span>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">pprint</span> <span class="kn">import</span> <span class="n">pprint</span>
+
+<span class="kn">import</span> <span class="nn">sklearn.model_selection</span>
 <span class="kn">import</span> <span class="nn">sklearn.datasets</span>
 <span class="kn">import</span> <span class="nn">sklearn.metrics</span>
 
@@ -144,15 +145,15 @@
 <span class="kn">import</span> <span class="nn">autosklearn.classification</span>
 </pre></div>
 </div>
-<section id="data-loading">
+<div class="section" id="data-loading">
 <h2>Data Loading<a class="headerlink" href="#data-loading" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">X</span><span class="p">,</span> <span class="n">y</span> <span class="o">=</span> <span class="n">sklearn</span><span class="o">.</span><span class="n">datasets</span><span class="o">.</span><span class="n">load_breast_cancer</span><span class="p">(</span><span class="n">return_X_y</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
 <span class="n">X_train</span><span class="p">,</span> <span class="n">X_test</span><span class="p">,</span> <span class="n">y_train</span><span class="p">,</span> <span class="n">y_test</span> <span class="o">=</span> \
     <span class="n">sklearn</span><span class="o">.</span><span class="n">model_selection</span><span class="o">.</span><span class="n">train_test_split</span><span class="p">(</span><span class="n">X</span><span class="p">,</span> <span class="n">y</span><span class="p">,</span> <span class="n">random_state</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
 </pre></div>
 </div>
-</section>
-<section id="fit-a-classifier-using-roar">
+</div>
+<div class="section" id="fit-a-classifier-using-roar">
 <h2>Fit a classifier using ROAR<a class="headerlink" href="#fit-a-classifier-using-roar" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="k">def</span> <span class="nf">get_roar_object_callback</span><span class="p">(</span>
     <span class="n">scenario_dict</span><span class="p">,</span>
@@ -194,7 +195,7 @@ <h2>Fit a classifier using ROAR<a class="headerlink" href="#fit-a-classifier-usi
 <span class="nb">print</span><span class="p">(</span><span class="s1">&#39;#&#39;</span> <span class="o">*</span> <span class="mi">80</span><span class="p">)</span>
 <span class="nb">print</span><span class="p">(</span><span class="s1">&#39;Results for ROAR.&#39;</span><span class="p">)</span>
 <span class="c1"># Print the final ensemble constructed by auto-sklearn via ROAR.</span>
-<span class="nb">print</span><span class="p">(</span><span class="n">automl</span><span class="o">.</span><span class="n">show_models</span><span class="p">())</span>
+<span class="n">pprint</span><span class="p">(</span><span class="n">automl</span><span class="o">.</span><span class="n">show_models</span><span class="p">(),</span> <span class="n">indent</span><span class="o">=</span><span class="mi">4</span><span class="p">)</span>
 <span class="n">predictions</span> <span class="o">=</span> <span class="n">automl</span><span class="o">.</span><span class="n">predict</span><span class="p">(</span><span class="n">X_test</span><span class="p">)</span>
 <span class="c1"># Print statistics about the auto-sklearn run such as number of</span>
 <span class="c1"># iterations, number of models failed with a time out.</span>
@@ -203,208 +204,287 @@ <h2>Fit a classifier using ROAR<a class="headerlink" href="#fit-a-classifier-usi
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>################################################################################
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>[ERROR] [2022-01-25 16:41:15,309:asyncio] _GatheringFuture exception was never retrieved
+future: &lt;_GatheringFuture finished exception=CancelledError()&gt;
+asyncio.exceptions.CancelledError
+################################################################################
 Results for ROAR.
-[(0.120000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;random_forest&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;select_rates_classification&#39;, &#39;classifier:random_forest:bootstrap&#39;: &#39;False&#39;, &#39;classifier:random_forest:criterion&#39;: &#39;entropy&#39;, &#39;classifier:random_forest:max_depth&#39;: &#39;None&#39;, &#39;classifier:random_forest:max_features&#39;: 0.13366756044911932, &#39;classifier:random_forest:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:random_forest:min_impurity_decrease&#39;: 0.0, &#39;classifier:random_forest:min_samples_leaf&#39;: 10, &#39;classifier:random_forest:min_samples_split&#39;: 18, &#39;classifier:random_forest:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;robust_scaler&#39;, &#39;feature_preprocessor:select_rates_classification:alpha&#39;: 0.06835899987255477, &#39;feature_preprocessor:select_rates_classification:score_func&#39;: &#39;chi2&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max&#39;: 0.791639810538127, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min&#39;: 0.2949722712387337, &#39;feature_preprocessor:select_rates_classification:mode&#39;: &#39;fwe&#39;},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.100000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;qda&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:qda:reg_param&#39;: 0.45597150689539434, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;standardize&#39;},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.100000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;adaboost&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;fast_ica&#39;, &#39;classifier:adaboost:algorithm&#39;: &#39;SAMME&#39;, &#39;classifier:adaboost:learning_rate&#39;: 1.6587261797143242, &#39;classifier:adaboost:max_depth&#39;: 8, &#39;classifier:adaboost:n_estimators&#39;: 368, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;median&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;normalize&#39;, &#39;feature_preprocessor:fast_ica:algorithm&#39;: &#39;parallel&#39;, &#39;feature_preprocessor:fast_ica:fun&#39;: &#39;cube&#39;, &#39;feature_preprocessor:fast_ica:whiten&#39;: &#39;True&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.00031132126574883666, &#39;feature_preprocessor:fast_ica:n_components&#39;: 530},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.080000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;passive_aggressive&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;extra_trees_preproc_for_classification&#39;, &#39;classifier:passive_aggressive:C&#39;: 1.4622203211010016e-05, &#39;classifier:passive_aggressive:average&#39;: &#39;True&#39;, &#39;classifier:passive_aggressive:fit_intercept&#39;: &#39;True&#39;, &#39;classifier:passive_aggressive:loss&#39;: &#39;hinge&#39;, &#39;classifier:passive_aggressive:tol&#39;: 0.0017964371726429177, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;power_transformer&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:bootstrap&#39;: &#39;False&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:criterion&#39;: &#39;entropy&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:max_depth&#39;: &#39;None&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:max_features&#39;: 0.10392737384710127, &#39;feature_preprocessor:extra_trees_preproc_for_classification:max_leaf_nodes&#39;: &#39;None&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_impurity_decrease&#39;: 0.0, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_samples_leaf&#39;: 11, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_samples_split&#39;: 9, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_weight_fraction_leaf&#39;: 0.0, &#39;feature_preprocessor:extra_trees_preproc_for_classification:n_estimators&#39;: 100},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.060000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;lda&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;nystroem_sampler&#39;, &#39;classifier:lda:shrinkage&#39;: &#39;manual&#39;, &#39;classifier:lda:tol&#39;: 0.0001384744065375479, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;standardize&#39;, &#39;feature_preprocessor:nystroem_sampler:kernel&#39;: &#39;poly&#39;, &#39;feature_preprocessor:nystroem_sampler:n_components&#39;: 414, &#39;classifier:lda:shrinkage_factor&#39;: 0.2192999307296778, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.0010364937923091088, &#39;feature_preprocessor:nystroem_sampler:coef0&#39;: 0.9586161969293623, &#39;feature_preprocessor:nystroem_sampler:degree&#39;: 4, &#39;feature_preprocessor:nystroem_sampler:gamma&#39;: 4.85086661900421e-05},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.060000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;lda&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;select_rates_classification&#39;, &#39;classifier:lda:shrinkage&#39;: &#39;None&#39;, &#39;classifier:lda:tol&#39;: 0.018038793247120707, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;power_transformer&#39;, &#39;feature_preprocessor:select_rates_classification:alpha&#39;: 0.3720251865973017, &#39;feature_preprocessor:select_rates_classification:score_func&#39;: &#39;mutual_info_classif&#39;},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.040000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;random_forest&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:random_forest:bootstrap&#39;: &#39;True&#39;, &#39;classifier:random_forest:criterion&#39;: &#39;gini&#39;, &#39;classifier:random_forest:max_depth&#39;: &#39;None&#39;, &#39;classifier:random_forest:max_features&#39;: 0.5, &#39;classifier:random_forest:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:random_forest:min_impurity_decrease&#39;: 0.0, &#39;classifier:random_forest:min_samples_leaf&#39;: 1, &#39;classifier:random_forest:min_samples_split&#39;: 2, &#39;classifier:random_forest:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;standardize&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.01},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.040000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;qda&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:qda:reg_param&#39;: 0.01667385805274091, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;median&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;robust_scaler&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.24966112905354662, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max&#39;: 0.8911872287672372, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min&#39;: 0.27951285357994865},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.040000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;qda&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;select_percentile_classification&#39;, &#39;classifier:qda:reg_param&#39;: 0.23910875621465733, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;robust_scaler&#39;, &#39;feature_preprocessor:select_percentile_classification:percentile&#39;: 87.86214879612422, &#39;feature_preprocessor:select_percentile_classification:score_func&#39;: &#39;mutual_info&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.0015543757334360092, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max&#39;: 0.7530273530198002, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min&#39;: 0.0792482460648931},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.040000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;liblinear_svc&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;fast_ica&#39;, &#39;classifier:liblinear_svc:C&#39;: 64.64437157661465, &#39;classifier:liblinear_svc:dual&#39;: &#39;False&#39;, &#39;classifier:liblinear_svc:fit_intercept&#39;: &#39;True&#39;, &#39;classifier:liblinear_svc:intercept_scaling&#39;: 1, &#39;classifier:liblinear_svc:loss&#39;: &#39;squared_hinge&#39;, &#39;classifier:liblinear_svc:multi_class&#39;: &#39;ovr&#39;, &#39;classifier:liblinear_svc:penalty&#39;: &#39;l2&#39;, &#39;classifier:liblinear_svc:tol&#39;: 0.013399889539803811, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;none&#39;, &#39;feature_preprocessor:fast_ica:algorithm&#39;: &#39;deflation&#39;, &#39;feature_preprocessor:fast_ica:fun&#39;: &#39;exp&#39;, &#39;feature_preprocessor:fast_ica:whiten&#39;: &#39;True&#39;, &#39;feature_preprocessor:fast_ica:n_components&#39;: 43},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.040000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;mlp&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;random_trees_embedding&#39;, &#39;classifier:mlp:activation&#39;: &#39;relu&#39;, &#39;classifier:mlp:alpha&#39;: 1.040063255316952e-07, &#39;classifier:mlp:batch_size&#39;: &#39;auto&#39;, &#39;classifier:mlp:beta_1&#39;: 0.9, &#39;classifier:mlp:beta_2&#39;: 0.999, &#39;classifier:mlp:early_stopping&#39;: &#39;valid&#39;, &#39;classifier:mlp:epsilon&#39;: 1e-08, &#39;classifier:mlp:hidden_layer_depth&#39;: 1, &#39;classifier:mlp:learning_rate_init&#39;: 0.0018475934154961247, &#39;classifier:mlp:n_iter_no_change&#39;: 32, &#39;classifier:mlp:num_nodes_per_layer&#39;: 35, &#39;classifier:mlp:shuffle&#39;: &#39;True&#39;, &#39;classifier:mlp:solver&#39;: &#39;adam&#39;, &#39;classifier:mlp:tol&#39;: 0.0001, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;none&#39;, &#39;feature_preprocessor:random_trees_embedding:bootstrap&#39;: &#39;True&#39;, &#39;feature_preprocessor:random_trees_embedding:max_depth&#39;: 9, &#39;feature_preprocessor:random_trees_embedding:max_leaf_nodes&#39;: &#39;None&#39;, &#39;feature_preprocessor:random_trees_embedding:min_samples_leaf&#39;: 6, &#39;feature_preprocessor:random_trees_embedding:min_samples_split&#39;: 17, &#39;feature_preprocessor:random_trees_embedding:min_weight_fraction_leaf&#39;: 1.0, &#39;feature_preprocessor:random_trees_embedding:n_estimators&#39;: 32, &#39;classifier:mlp:validation_fraction&#39;: 0.1},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.040000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;decision_tree&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;pca&#39;, &#39;classifier:decision_tree:criterion&#39;: &#39;entropy&#39;, &#39;classifier:decision_tree:max_depth_factor&#39;: 1.7056696489196201, &#39;classifier:decision_tree:max_features&#39;: 1.0, &#39;classifier:decision_tree:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:decision_tree:min_impurity_decrease&#39;: 0.0, &#39;classifier:decision_tree:min_samples_leaf&#39;: 18, &#39;classifier:decision_tree:min_samples_split&#39;: 8, &#39;classifier:decision_tree:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;median&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;robust_scaler&#39;, &#39;feature_preprocessor:pca:keep_variance&#39;: 0.927211764000264, &#39;feature_preprocessor:pca:whiten&#39;: &#39;False&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max&#39;: 0.9968675728600899, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min&#39;: 0.29662315421901003},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.040000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;passive_aggressive&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;kernel_pca&#39;, &#39;classifier:passive_aggressive:C&#39;: 0.114391364050916, &#39;classifier:passive_aggressive:average&#39;: &#39;False&#39;, &#39;classifier:passive_aggressive:fit_intercept&#39;: &#39;True&#39;, &#39;classifier:passive_aggressive:loss&#39;: &#39;hinge&#39;, &#39;classifier:passive_aggressive:tol&#39;: 1.3440590734654458e-05, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;normalize&#39;, &#39;feature_preprocessor:kernel_pca:kernel&#39;: &#39;sigmoid&#39;, &#39;feature_preprocessor:kernel_pca:n_components&#39;: 1117, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.023878029144121318, &#39;feature_preprocessor:kernel_pca:coef0&#39;: -0.34481806774236556},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.020000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;gradient_boosting&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:gradient_boosting:early_stop&#39;: &#39;valid&#39;, &#39;classifier:gradient_boosting:l2_regularization&#39;: 0.009193751229756878, &#39;classifier:gradient_boosting:learning_rate&#39;: 0.6836522210129492, &#39;classifier:gradient_boosting:loss&#39;: &#39;auto&#39;, &#39;classifier:gradient_boosting:max_bins&#39;: 255, &#39;classifier:gradient_boosting:max_depth&#39;: &#39;None&#39;, &#39;classifier:gradient_boosting:max_leaf_nodes&#39;: 828, &#39;classifier:gradient_boosting:min_samples_leaf&#39;: 3, &#39;classifier:gradient_boosting:scoring&#39;: &#39;loss&#39;, &#39;classifier:gradient_boosting:tol&#39;: 1e-07, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;normalize&#39;, &#39;classifier:gradient_boosting:n_iter_no_change&#39;: 6, &#39;classifier:gradient_boosting:validation_fraction&#39;: 0.28817393948552533, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.00010868227587711292},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.020000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;liblinear_svc&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;pca&#39;, &#39;classifier:liblinear_svc:C&#39;: 2.285306370007781, &#39;classifier:liblinear_svc:dual&#39;: &#39;False&#39;, &#39;classifier:liblinear_svc:fit_intercept&#39;: &#39;True&#39;, &#39;classifier:liblinear_svc:intercept_scaling&#39;: 1, &#39;classifier:liblinear_svc:loss&#39;: &#39;squared_hinge&#39;, &#39;classifier:liblinear_svc:multi_class&#39;: &#39;ovr&#39;, &#39;classifier:liblinear_svc:penalty&#39;: &#39;l2&#39;, &#39;classifier:liblinear_svc:tol&#39;: 0.005821344021856017, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;quantile_transformer&#39;, &#39;feature_preprocessor:pca:keep_variance&#39;: 0.7978382221839412, &#39;feature_preprocessor:pca:whiten&#39;: &#39;True&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles&#39;: 1644, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution&#39;: &#39;normal&#39;},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.020000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;gradient_boosting&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;select_rates_classification&#39;, &#39;classifier:gradient_boosting:early_stop&#39;: &#39;train&#39;, &#39;classifier:gradient_boosting:l2_regularization&#39;: 4.520111487139148e-06, &#39;classifier:gradient_boosting:learning_rate&#39;: 0.12585871697357337, &#39;classifier:gradient_boosting:loss&#39;: &#39;auto&#39;, &#39;classifier:gradient_boosting:max_bins&#39;: 255, &#39;classifier:gradient_boosting:max_depth&#39;: &#39;None&#39;, &#39;classifier:gradient_boosting:max_leaf_nodes&#39;: 1894, &#39;classifier:gradient_boosting:min_samples_leaf&#39;: 2, &#39;classifier:gradient_boosting:scoring&#39;: &#39;loss&#39;, &#39;classifier:gradient_boosting:tol&#39;: 1e-07, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;median&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;robust_scaler&#39;, &#39;feature_preprocessor:select_rates_classification:alpha&#39;: 0.06707365823758892, &#39;feature_preprocessor:select_rates_classification:score_func&#39;: &#39;chi2&#39;, &#39;classifier:gradient_boosting:n_iter_no_change&#39;: 20, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.0014466154794086487, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max&#39;: 0.7104048919054863, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min&#39;: 0.2345103479872152, &#39;feature_preprocessor:select_rates_classification:mode&#39;: &#39;fpr&#39;},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.020000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;lda&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;kernel_pca&#39;, &#39;classifier:lda:shrinkage&#39;: &#39;auto&#39;, &#39;classifier:lda:tol&#39;: 0.014034416932373166, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;median&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;power_transformer&#39;, &#39;feature_preprocessor:kernel_pca:kernel&#39;: &#39;cosine&#39;, &#39;feature_preprocessor:kernel_pca:n_components&#39;: 285, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.0074025063474275335},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.020000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;lda&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;liblinear_svc_preprocessor&#39;, &#39;classifier:lda:shrinkage&#39;: &#39;None&#39;, &#39;classifier:lda:tol&#39;: 0.035543606049804645, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;none&#39;, &#39;feature_preprocessor:liblinear_svc_preprocessor:C&#39;: 4.279105226196228, &#39;feature_preprocessor:liblinear_svc_preprocessor:dual&#39;: &#39;False&#39;, &#39;feature_preprocessor:liblinear_svc_preprocessor:fit_intercept&#39;: &#39;True&#39;, &#39;feature_preprocessor:liblinear_svc_preprocessor:intercept_scaling&#39;: 1, &#39;feature_preprocessor:liblinear_svc_preprocessor:loss&#39;: &#39;squared_hinge&#39;, &#39;feature_preprocessor:liblinear_svc_preprocessor:multi_class&#39;: &#39;ovr&#39;, &#39;feature_preprocessor:liblinear_svc_preprocessor:penalty&#39;: &#39;l1&#39;, &#39;feature_preprocessor:liblinear_svc_preprocessor:tol&#39;: 4.410651487451908e-05},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.020000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;mlp&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;select_rates_classification&#39;, &#39;classifier:mlp:activation&#39;: &#39;tanh&#39;, &#39;classifier:mlp:alpha&#39;: 4.421745686388309e-05, &#39;classifier:mlp:batch_size&#39;: &#39;auto&#39;, &#39;classifier:mlp:beta_1&#39;: 0.9, &#39;classifier:mlp:beta_2&#39;: 0.999, &#39;classifier:mlp:early_stopping&#39;: &#39;train&#39;, &#39;classifier:mlp:epsilon&#39;: 1e-08, &#39;classifier:mlp:hidden_layer_depth&#39;: 3, &#39;classifier:mlp:learning_rate_init&#39;: 0.0005485701552075972, &#39;classifier:mlp:n_iter_no_change&#39;: 32, &#39;classifier:mlp:num_nodes_per_layer&#39;: 36, &#39;classifier:mlp:shuffle&#39;: &#39;True&#39;, &#39;classifier:mlp:solver&#39;: &#39;adam&#39;, &#39;classifier:mlp:tol&#39;: 0.0001, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;minmax&#39;, &#39;feature_preprocessor:select_rates_classification:alpha&#39;: 0.44455023733729254, &#39;feature_preprocessor:select_rates_classification:score_func&#39;: &#39;mutual_info_classif&#39;},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.020000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;mlp&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;extra_trees_preproc_for_classification&#39;, &#39;classifier:mlp:activation&#39;: &#39;relu&#39;, &#39;classifier:mlp:alpha&#39;: 1.4882838781863879e-06, &#39;classifier:mlp:batch_size&#39;: &#39;auto&#39;, &#39;classifier:mlp:beta_1&#39;: 0.9, &#39;classifier:mlp:beta_2&#39;: 0.999, &#39;classifier:mlp:early_stopping&#39;: &#39;train&#39;, &#39;classifier:mlp:epsilon&#39;: 1e-08, &#39;classifier:mlp:hidden_layer_depth&#39;: 2, &#39;classifier:mlp:learning_rate_init&#39;: 0.06290764761689918, &#39;classifier:mlp:n_iter_no_change&#39;: 32, &#39;classifier:mlp:num_nodes_per_layer&#39;: 23, &#39;classifier:mlp:shuffle&#39;: &#39;True&#39;, &#39;classifier:mlp:solver&#39;: &#39;adam&#39;, &#39;classifier:mlp:tol&#39;: 0.0001, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;median&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;quantile_transformer&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:bootstrap&#39;: &#39;False&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:criterion&#39;: &#39;gini&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:max_depth&#39;: &#39;None&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:max_features&#39;: 0.9565048776795676, &#39;feature_preprocessor:extra_trees_preproc_for_classification:max_leaf_nodes&#39;: &#39;None&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_impurity_decrease&#39;: 0.0, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_samples_leaf&#39;: 7, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_samples_split&#39;: 13, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_weight_fraction_leaf&#39;: 0.0, &#39;feature_preprocessor:extra_trees_preproc_for_classification:n_estimators&#39;: 100, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles&#39;: 273, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution&#39;: &#39;normal&#39;},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.020000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;decision_tree&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;random_trees_embedding&#39;, &#39;classifier:decision_tree:criterion&#39;: &#39;gini&#39;, &#39;classifier:decision_tree:max_depth_factor&#39;: 0.6228726042266908, &#39;classifier:decision_tree:max_features&#39;: 1.0, &#39;classifier:decision_tree:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:decision_tree:min_impurity_decrease&#39;: 0.0, &#39;classifier:decision_tree:min_samples_leaf&#39;: 16, &#39;classifier:decision_tree:min_samples_split&#39;: 19, &#39;classifier:decision_tree:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;quantile_transformer&#39;, &#39;feature_preprocessor:random_trees_embedding:bootstrap&#39;: &#39;False&#39;, &#39;feature_preprocessor:random_trees_embedding:max_depth&#39;: 2, &#39;feature_preprocessor:random_trees_embedding:max_leaf_nodes&#39;: &#39;None&#39;, &#39;feature_preprocessor:random_trees_embedding:min_samples_leaf&#39;: 17, &#39;feature_preprocessor:random_trees_embedding:min_samples_split&#39;: 12, &#39;feature_preprocessor:random_trees_embedding:min_weight_fraction_leaf&#39;: 1.0, &#39;feature_preprocessor:random_trees_embedding:n_estimators&#39;: 38, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles&#39;: 606, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution&#39;: &#39;uniform&#39;},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.020000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;decision_tree&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;pca&#39;, &#39;classifier:decision_tree:criterion&#39;: &#39;entropy&#39;, &#39;classifier:decision_tree:max_depth_factor&#39;: 1.1387664107898294, &#39;classifier:decision_tree:max_features&#39;: 1.0, &#39;classifier:decision_tree:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:decision_tree:min_impurity_decrease&#39;: 0.0, &#39;classifier:decision_tree:min_samples_leaf&#39;: 8, &#39;classifier:decision_tree:min_samples_split&#39;: 9, &#39;classifier:decision_tree:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;none&#39;, &#39;feature_preprocessor:pca:keep_variance&#39;: 0.9503612101678187, &#39;feature_preprocessor:pca:whiten&#39;: &#39;True&#39;},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.020000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;bernoulli_nb&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;select_percentile_classification&#39;, &#39;classifier:bernoulli_nb:alpha&#39;: 8.843515731149308, &#39;classifier:bernoulli_nb:fit_prior&#39;: &#39;True&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;robust_scaler&#39;, &#39;feature_preprocessor:select_percentile_classification:percentile&#39;: 50.58615414999356, &#39;feature_preprocessor:select_percentile_classification:score_func&#39;: &#39;chi2&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max&#39;: 0.7852218490061461, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min&#39;: 0.15162626175424537},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-]
+{   2: {   &#39;balancing&#39;: Balancing(random_state=1),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc787e05a30&gt;,
+           &#39;cost&#39;: 0.028368794326241176,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7861841f0&gt;,
+           &#39;ensemble_weight&#39;: 0.04,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78a86d100&gt;,
+           &#39;model_id&#39;: 2,
+           &#39;rank&#39;: 3,
+           &#39;sklearn_classifier&#39;: RandomForestClassifier(max_features=5, n_estimators=512, n_jobs=1,
+                       random_state=1, warm_start=True)},
+    4: {   &#39;balancing&#39;: Balancing(random_state=1),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc787302940&gt;,
+           &#39;cost&#39;: 0.05673758865248224,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7876620a0&gt;,
+           &#39;ensemble_weight&#39;: 0.02,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc786554580&gt;,
+           &#39;model_id&#39;: 4,
+           &#39;rank&#39;: 15,
+           &#39;sklearn_classifier&#39;: MLPClassifier(alpha=1.0400632553169558e-07, beta_1=0.999, beta_2=0.9,
+              early_stopping=True, hidden_layer_sizes=(35,),
+              learning_rate_init=0.0018475934154961215, max_iter=32,
+              n_iter_no_change=32, random_state=1, verbose=0, warm_start=True)},
+    5: {   &#39;balancing&#39;: Balancing(random_state=1),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc787374fd0&gt;,
+           &#39;cost&#39;: 0.11347517730496459,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78706ef70&gt;,
+           &#39;ensemble_weight&#39;: 0.02,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc787374400&gt;,
+           &#39;model_id&#39;: 5,
+           &#39;rank&#39;: 26,
+           &#39;sklearn_classifier&#39;: BernoulliNB(alpha=0.010232681621243814)},
+    6: {   &#39;balancing&#39;: Balancing(random_state=1),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc787748400&gt;,
+           &#39;cost&#39;: 0.049645390070921946,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78582ee20&gt;,
+           &#39;ensemble_weight&#39;: 0.06,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc787748250&gt;,
+           &#39;model_id&#39;: 6,
+           &#39;rank&#39;: 11,
+           &#39;sklearn_classifier&#39;: LinearDiscriminantAnalysis(tol=1.3202262453017803e-05)},
+    7: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc787aa5220&gt;,
+           &#39;cost&#39;: 0.099290780141844,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78af89d60&gt;,
+           &#39;ensemble_weight&#39;: 0.02,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc787a8deb0&gt;,
+           &#39;model_id&#39;: 7,
+           &#39;rank&#39;: 24,
+           &#39;sklearn_classifier&#39;: BernoulliNB(alpha=8.843515731149308)},
+    9: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc787793460&gt;,
+           &#39;cost&#39;: 0.07801418439716312,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78ff456d0&gt;,
+           &#39;ensemble_weight&#39;: 0.02,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc787788ca0&gt;,
+           &#39;model_id&#39;: 9,
+           &#39;rank&#39;: 20,
+           &#39;sklearn_classifier&#39;: BernoulliNB(alpha=0.04662763358613446, fit_prior=False)},
+    10: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc787e05310&gt;,
+            &#39;cost&#39;: 0.007092198581560294,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78509f9a0&gt;,
+            &#39;ensemble_weight&#39;: 0.08,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78bd13160&gt;,
+            &#39;model_id&#39;: 10,
+            &#39;rank&#39;: 1,
+            &#39;sklearn_classifier&#39;: QuadraticDiscriminantAnalysis(reg_param=0.45597150689539434)},
+    11: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc79068b160&gt;,
+            &#39;cost&#39;: 0.028368794326241176,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7850a8850&gt;,
+            &#39;ensemble_weight&#39;: 0.06,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78a791250&gt;,
+            &#39;model_id&#39;: 11,
+            &#39;rank&#39;: 4,
+            &#39;sklearn_classifier&#39;: QuadraticDiscriminantAnalysis(reg_param=0.0995499108996053)},
+    12: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc787ba29a0&gt;,
+            &#39;cost&#39;: 0.049645390070921946,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78580d130&gt;,
+            &#39;ensemble_weight&#39;: 0.04,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc787ba2040&gt;,
+            &#39;model_id&#39;: 12,
+            &#39;rank&#39;: 12,
+            &#39;sklearn_classifier&#39;: LinearDiscriminantAnalysis(shrinkage=&#39;auto&#39;, solver=&#39;lsqr&#39;,
+                           tol=0.014034416932373166)},
+    13: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78fa980a0&gt;,
+            &#39;cost&#39;: 0.021276595744680882,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc787888340&gt;,
+            &#39;ensemble_weight&#39;: 0.06,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78fa98670&gt;,
+            &#39;model_id&#39;: 13,
+            &#39;rank&#39;: 2,
+            &#39;sklearn_classifier&#39;: LinearDiscriminantAnalysis(shrinkage=0.2192999307296778, solver=&#39;lsqr&#39;,
+                           tol=0.0001384744065375479)},
+    14: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc787e47580&gt;,
+            &#39;cost&#39;: 0.05673758865248224,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc784bbe130&gt;,
+            &#39;ensemble_weight&#39;: 0.04,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc787e47400&gt;,
+            &#39;model_id&#39;: 14,
+            &#39;rank&#39;: 16,
+            &#39;sklearn_classifier&#39;: LinearDiscriminantAnalysis(tol=0.035543606049804645)},
+    16: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78af79460&gt;,
+            &#39;cost&#39;: 0.08510638297872342,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc79068b340&gt;,
+            &#39;ensemble_weight&#39;: 0.04,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78af90700&gt;,
+            &#39;model_id&#39;: 16,
+            &#39;rank&#39;: 21,
+            &#39;sklearn_classifier&#39;: DecisionTreeClassifier(criterion=&#39;entropy&#39;, max_depth=1, min_samples_leaf=8,
+                       min_samples_split=9, random_state=1)},
+    17: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78b9260a0&gt;,
+            &#39;cost&#39;: 0.04255319148936165,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7865df100&gt;,
+            &#39;ensemble_weight&#39;: 0.04,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78b926c40&gt;,
+            &#39;model_id&#39;: 17,
+            &#39;rank&#39;: 10,
+            &#39;sklearn_classifier&#39;: LinearSVC(C=64.64437157661465, dual=False, intercept_scaling=1.0,
+          random_state=1, tol=0.013399889539803811)},
+    18: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc786bff640&gt;,
+            &#39;cost&#39;: 0.099290780141844,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc786aaaf10&gt;,
+            &#39;ensemble_weight&#39;: 0.02,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc786bd1b50&gt;,
+            &#39;model_id&#39;: 18,
+            &#39;rank&#39;: 25,
+            &#39;sklearn_classifier&#39;: PassiveAggressiveClassifier(C=0.114391364050916, max_iter=128, random_state=1,
+                            tol=1.3440590734654458e-05, warm_start=True)},
+    19: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78f986610&gt;,
+            &#39;cost&#39;: 0.03546099290780147,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78f84d1f0&gt;,
+            &#39;ensemble_weight&#39;: 0.02,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc786854070&gt;,
+            &#39;model_id&#39;: 19,
+            &#39;rank&#39;: 8,
+            &#39;sklearn_classifier&#39;: QuadraticDiscriminantAnalysis(reg_param=0.01667385805274091)},
+    21: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78f69fd90&gt;,
+            &#39;cost&#39;: 0.028368794326241176,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78c43e280&gt;,
+            &#39;ensemble_weight&#39;: 0.06,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c94d1f0&gt;,
+            &#39;model_id&#39;: 21,
+            &#39;rank&#39;: 5,
+            &#39;sklearn_classifier&#39;: LinearSVC(C=2.285306370007781, dual=False, intercept_scaling=1.0,
+          random_state=1, tol=0.005821344021856017)},
+    24: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc7903bddf0&gt;,
+            &#39;cost&#39;: 0.06382978723404253,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc786581c40&gt;,
+            &#39;ensemble_weight&#39;: 0.02,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c9bf490&gt;,
+            &#39;model_id&#39;: 24,
+            &#39;rank&#39;: 17,
+            &#39;sklearn_classifier&#39;: SGDClassifier(alpha=5.189696963506304e-05, eta0=0.007608275318572964,
+              learning_rate=&#39;constant&#39;, loss=&#39;perceptron&#39;, max_iter=64,
+              random_state=1, tol=0.0005024616279807072, warm_start=True)},
+    25: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc786149370&gt;,
+            &#39;cost&#39;: 0.08510638297872342,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78770be20&gt;,
+            &#39;ensemble_weight&#39;: 0.02,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc786c845e0&gt;,
+            &#39;model_id&#39;: 25,
+            &#39;rank&#39;: 22,
+            &#39;sklearn_classifier&#39;: PassiveAggressiveClassifier(C=1.4622203211010016e-05, average=True,
+                            max_iter=128, random_state=1,
+                            tol=0.0017964371726429177, warm_start=True)},
+    26: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78ff433d0&gt;,
+            &#39;cost&#39;: 0.06382978723404253,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc787e52070&gt;,
+            &#39;ensemble_weight&#39;: 0.06,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78ff43730&gt;,
+            &#39;model_id&#39;: 26,
+            &#39;rank&#39;: 18,
+            &#39;sklearn_classifier&#39;: AdaBoostClassifier(algorithm=&#39;SAMME&#39;,
+                   base_estimator=DecisionTreeClassifier(max_depth=8),
+                   learning_rate=1.6587261797143242, n_estimators=368,
+                   random_state=1)},
+    27: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc786149910&gt;,
+            &#39;cost&#39;: 0.09219858156028371,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78face370&gt;,
+            &#39;ensemble_weight&#39;: 0.02,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc7861490d0&gt;,
+            &#39;model_id&#39;: 27,
+            &#39;rank&#39;: 23,
+            &#39;sklearn_classifier&#39;: DecisionTreeClassifier(criterion=&#39;entropy&#39;, max_depth=4, min_samples_leaf=18,
+                       min_samples_split=8, random_state=1)},
+    30: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78c9bff70&gt;,
+            &#39;cost&#39;: 0.028368794326241176,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78a25e7c0&gt;,
+            &#39;ensemble_weight&#39;: 0.06,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78a4be100&gt;,
+            &#39;model_id&#39;: 30,
+            &#39;rank&#39;: 6,
+            &#39;sklearn_classifier&#39;: RandomForestClassifier(bootstrap=False, criterion=&#39;entropy&#39;, max_features=1,
+                       min_samples_leaf=10, min_samples_split=18,
+                       n_estimators=512, n_jobs=1, random_state=1,
+                       warm_start=True)},
+    31: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc7877cdaf0&gt;,
+            &#39;cost&#39;: 0.049645390070921946,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78733d670&gt;,
+            &#39;ensemble_weight&#39;: 0.02,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc7877cd460&gt;,
+            &#39;model_id&#39;: 31,
+            &#39;rank&#39;: 13,
+            &#39;sklearn_classifier&#39;: LinearDiscriminantAnalysis(tol=0.018038793247120707)},
+    33: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc786796cd0&gt;,
+            &#39;cost&#39;: 0.03546099290780147,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc785b99af0&gt;,
+            &#39;ensemble_weight&#39;: 0.04,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c5839d0&gt;,
+            &#39;model_id&#39;: 33,
+            &#39;rank&#39;: 9,
+            &#39;sklearn_classifier&#39;: HistGradientBoostingClassifier(early_stopping=True,
+                               l2_regularization=4.5201114871391565e-06,
+                               learning_rate=0.12585871697357337, max_iter=256,
+                               max_leaf_nodes=1894, min_samples_leaf=2,
+                               n_iter_no_change=20, random_state=1,
+                               validation_fraction=None, warm_start=True)},
+    35: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78b257790&gt;,
+            &#39;cost&#39;: 0.049645390070921946,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78776fac0&gt;,
+            &#39;ensemble_weight&#39;: 0.02,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc784bd90d0&gt;,
+            &#39;model_id&#39;: 35,
+            &#39;rank&#39;: 14,
+            &#39;sklearn_classifier&#39;: KNeighborsClassifier(n_neighbors=2)},
+    36: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc786806670&gt;,
+            &#39;cost&#39;: 0.07092198581560283,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78c752cd0&gt;,
+            &#39;ensemble_weight&#39;: 0.06,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78a86d9a0&gt;,
+            &#39;model_id&#39;: 36,
+            &#39;rank&#39;: 19,
+            &#39;sklearn_classifier&#39;: GaussianNB()},
+    37: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78aac0730&gt;,
+            &#39;cost&#39;: 0.028368794326241176,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78f83e5e0&gt;,
+            &#39;ensemble_weight&#39;: 0.04,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78b6561c0&gt;,
+            &#39;model_id&#39;: 37,
+            &#39;rank&#39;: 7,
+            &#39;sklearn_classifier&#39;: LinearDiscriminantAnalysis(shrinkage=0.32361629219618826, solver=&#39;lsqr&#39;,
+                           tol=3.142032801716559e-05)}}
 auto-sklearn results:
   Dataset name: breast_cancer
   Metric: accuracy
   Best validation score: 0.992908
-  Number of target algorithm runs: 33
-  Number of successful target algorithm runs: 31
+  Number of target algorithm runs: 36
+  Number of successful target algorithm runs: 35
   Number of crashed target algorithm runs: 1
-  Number of target algorithms that exceeded the time limit: 1
+  Number of target algorithms that exceeded the time limit: 0
   Number of target algorithms that exceeded the memory limit: 0
 
-Accuracy score 0.9440559440559441
+Accuracy score 0.958041958041958
 </pre></div>
 </div>
-</section>
-<section id="fit-a-classifier-using-random-search">
+</div>
+<div class="section" id="fit-a-classifier-using-random-search">
 <h2>Fit a classifier using Random Search<a class="headerlink" href="#fit-a-classifier-using-random-search" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="k">def</span> <span class="nf">get_random_search_object_callback</span><span class="p">(</span>
     <span class="n">scenario_dict</span><span class="p">,</span>
@@ -449,7 +529,7 @@ <h2>Fit a classifier using Random Search<a class="headerlink" href="#fit-a-class
 <span class="nb">print</span><span class="p">(</span><span class="s1">&#39;Results for random search.&#39;</span><span class="p">)</span>
 
 <span class="c1"># Print the final ensemble constructed by auto-sklearn via random search.</span>
-<span class="nb">print</span><span class="p">(</span><span class="n">automl</span><span class="o">.</span><span class="n">show_models</span><span class="p">())</span>
+<span class="n">pprint</span><span class="p">(</span><span class="n">automl</span><span class="o">.</span><span class="n">show_models</span><span class="p">(),</span> <span class="n">indent</span><span class="o">=</span><span class="mi">4</span><span class="p">)</span>
 
 <span class="c1"># Print statistics about the auto-sklearn run such as number of</span>
 <span class="c1"># iterations, number of models failed with a time out.</span>
@@ -462,205 +542,310 @@ <h2>Fit a classifier using Random Search<a class="headerlink" href="#fit-a-class
 <p class="sphx-glr-script-out">Out:</p>
 <div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>################################################################################
 Results for random search.
-[(0.120000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;random_forest&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;select_rates_classification&#39;, &#39;classifier:random_forest:bootstrap&#39;: &#39;False&#39;, &#39;classifier:random_forest:criterion&#39;: &#39;entropy&#39;, &#39;classifier:random_forest:max_depth&#39;: &#39;None&#39;, &#39;classifier:random_forest:max_features&#39;: 0.13366756044911932, &#39;classifier:random_forest:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:random_forest:min_impurity_decrease&#39;: 0.0, &#39;classifier:random_forest:min_samples_leaf&#39;: 10, &#39;classifier:random_forest:min_samples_split&#39;: 18, &#39;classifier:random_forest:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;robust_scaler&#39;, &#39;feature_preprocessor:select_rates_classification:alpha&#39;: 0.06835899987255477, &#39;feature_preprocessor:select_rates_classification:score_func&#39;: &#39;chi2&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max&#39;: 0.791639810538127, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min&#39;: 0.2949722712387337, &#39;feature_preprocessor:select_rates_classification:mode&#39;: &#39;fwe&#39;},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.100000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;qda&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:qda:reg_param&#39;: 0.45597150689539434, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;standardize&#39;},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.100000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;adaboost&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;fast_ica&#39;, &#39;classifier:adaboost:algorithm&#39;: &#39;SAMME&#39;, &#39;classifier:adaboost:learning_rate&#39;: 1.6587261797143242, &#39;classifier:adaboost:max_depth&#39;: 8, &#39;classifier:adaboost:n_estimators&#39;: 368, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;median&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;normalize&#39;, &#39;feature_preprocessor:fast_ica:algorithm&#39;: &#39;parallel&#39;, &#39;feature_preprocessor:fast_ica:fun&#39;: &#39;cube&#39;, &#39;feature_preprocessor:fast_ica:whiten&#39;: &#39;True&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.00031132126574883666, &#39;feature_preprocessor:fast_ica:n_components&#39;: 530},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.080000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;passive_aggressive&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;extra_trees_preproc_for_classification&#39;, &#39;classifier:passive_aggressive:C&#39;: 1.4622203211010016e-05, &#39;classifier:passive_aggressive:average&#39;: &#39;True&#39;, &#39;classifier:passive_aggressive:fit_intercept&#39;: &#39;True&#39;, &#39;classifier:passive_aggressive:loss&#39;: &#39;hinge&#39;, &#39;classifier:passive_aggressive:tol&#39;: 0.0017964371726429177, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;power_transformer&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:bootstrap&#39;: &#39;False&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:criterion&#39;: &#39;entropy&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:max_depth&#39;: &#39;None&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:max_features&#39;: 0.10392737384710127, &#39;feature_preprocessor:extra_trees_preproc_for_classification:max_leaf_nodes&#39;: &#39;None&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_impurity_decrease&#39;: 0.0, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_samples_leaf&#39;: 11, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_samples_split&#39;: 9, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_weight_fraction_leaf&#39;: 0.0, &#39;feature_preprocessor:extra_trees_preproc_for_classification:n_estimators&#39;: 100},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.060000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;lda&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;nystroem_sampler&#39;, &#39;classifier:lda:shrinkage&#39;: &#39;manual&#39;, &#39;classifier:lda:tol&#39;: 0.0001384744065375479, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;standardize&#39;, &#39;feature_preprocessor:nystroem_sampler:kernel&#39;: &#39;poly&#39;, &#39;feature_preprocessor:nystroem_sampler:n_components&#39;: 414, &#39;classifier:lda:shrinkage_factor&#39;: 0.2192999307296778, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.0010364937923091088, &#39;feature_preprocessor:nystroem_sampler:coef0&#39;: 0.9586161969293623, &#39;feature_preprocessor:nystroem_sampler:degree&#39;: 4, &#39;feature_preprocessor:nystroem_sampler:gamma&#39;: 4.85086661900421e-05},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.060000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;lda&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;select_rates_classification&#39;, &#39;classifier:lda:shrinkage&#39;: &#39;None&#39;, &#39;classifier:lda:tol&#39;: 0.018038793247120707, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;power_transformer&#39;, &#39;feature_preprocessor:select_rates_classification:alpha&#39;: 0.3720251865973017, &#39;feature_preprocessor:select_rates_classification:score_func&#39;: &#39;mutual_info_classif&#39;},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.040000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;random_forest&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:random_forest:bootstrap&#39;: &#39;True&#39;, &#39;classifier:random_forest:criterion&#39;: &#39;gini&#39;, &#39;classifier:random_forest:max_depth&#39;: &#39;None&#39;, &#39;classifier:random_forest:max_features&#39;: 0.5, &#39;classifier:random_forest:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:random_forest:min_impurity_decrease&#39;: 0.0, &#39;classifier:random_forest:min_samples_leaf&#39;: 1, &#39;classifier:random_forest:min_samples_split&#39;: 2, &#39;classifier:random_forest:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;standardize&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.01},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.040000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;qda&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:qda:reg_param&#39;: 0.01667385805274091, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;median&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;robust_scaler&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.24966112905354662, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max&#39;: 0.8911872287672372, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min&#39;: 0.27951285357994865},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.040000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;qda&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;select_percentile_classification&#39;, &#39;classifier:qda:reg_param&#39;: 0.23910875621465733, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;robust_scaler&#39;, &#39;feature_preprocessor:select_percentile_classification:percentile&#39;: 87.86214879612422, &#39;feature_preprocessor:select_percentile_classification:score_func&#39;: &#39;mutual_info&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.0015543757334360092, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max&#39;: 0.7530273530198002, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min&#39;: 0.0792482460648931},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.040000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;liblinear_svc&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;fast_ica&#39;, &#39;classifier:liblinear_svc:C&#39;: 64.64437157661465, &#39;classifier:liblinear_svc:dual&#39;: &#39;False&#39;, &#39;classifier:liblinear_svc:fit_intercept&#39;: &#39;True&#39;, &#39;classifier:liblinear_svc:intercept_scaling&#39;: 1, &#39;classifier:liblinear_svc:loss&#39;: &#39;squared_hinge&#39;, &#39;classifier:liblinear_svc:multi_class&#39;: &#39;ovr&#39;, &#39;classifier:liblinear_svc:penalty&#39;: &#39;l2&#39;, &#39;classifier:liblinear_svc:tol&#39;: 0.013399889539803811, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;none&#39;, &#39;feature_preprocessor:fast_ica:algorithm&#39;: &#39;deflation&#39;, &#39;feature_preprocessor:fast_ica:fun&#39;: &#39;exp&#39;, &#39;feature_preprocessor:fast_ica:whiten&#39;: &#39;True&#39;, &#39;feature_preprocessor:fast_ica:n_components&#39;: 43},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.040000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;mlp&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;random_trees_embedding&#39;, &#39;classifier:mlp:activation&#39;: &#39;relu&#39;, &#39;classifier:mlp:alpha&#39;: 1.040063255316952e-07, &#39;classifier:mlp:batch_size&#39;: &#39;auto&#39;, &#39;classifier:mlp:beta_1&#39;: 0.9, &#39;classifier:mlp:beta_2&#39;: 0.999, &#39;classifier:mlp:early_stopping&#39;: &#39;valid&#39;, &#39;classifier:mlp:epsilon&#39;: 1e-08, &#39;classifier:mlp:hidden_layer_depth&#39;: 1, &#39;classifier:mlp:learning_rate_init&#39;: 0.0018475934154961247, &#39;classifier:mlp:n_iter_no_change&#39;: 32, &#39;classifier:mlp:num_nodes_per_layer&#39;: 35, &#39;classifier:mlp:shuffle&#39;: &#39;True&#39;, &#39;classifier:mlp:solver&#39;: &#39;adam&#39;, &#39;classifier:mlp:tol&#39;: 0.0001, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;none&#39;, &#39;feature_preprocessor:random_trees_embedding:bootstrap&#39;: &#39;True&#39;, &#39;feature_preprocessor:random_trees_embedding:max_depth&#39;: 9, &#39;feature_preprocessor:random_trees_embedding:max_leaf_nodes&#39;: &#39;None&#39;, &#39;feature_preprocessor:random_trees_embedding:min_samples_leaf&#39;: 6, &#39;feature_preprocessor:random_trees_embedding:min_samples_split&#39;: 17, &#39;feature_preprocessor:random_trees_embedding:min_weight_fraction_leaf&#39;: 1.0, &#39;feature_preprocessor:random_trees_embedding:n_estimators&#39;: 32, &#39;classifier:mlp:validation_fraction&#39;: 0.1},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.040000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;decision_tree&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;pca&#39;, &#39;classifier:decision_tree:criterion&#39;: &#39;entropy&#39;, &#39;classifier:decision_tree:max_depth_factor&#39;: 1.7056696489196201, &#39;classifier:decision_tree:max_features&#39;: 1.0, &#39;classifier:decision_tree:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:decision_tree:min_impurity_decrease&#39;: 0.0, &#39;classifier:decision_tree:min_samples_leaf&#39;: 18, &#39;classifier:decision_tree:min_samples_split&#39;: 8, &#39;classifier:decision_tree:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;median&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;robust_scaler&#39;, &#39;feature_preprocessor:pca:keep_variance&#39;: 0.927211764000264, &#39;feature_preprocessor:pca:whiten&#39;: &#39;False&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max&#39;: 0.9968675728600899, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min&#39;: 0.29662315421901003},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.040000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;passive_aggressive&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;kernel_pca&#39;, &#39;classifier:passive_aggressive:C&#39;: 0.114391364050916, &#39;classifier:passive_aggressive:average&#39;: &#39;False&#39;, &#39;classifier:passive_aggressive:fit_intercept&#39;: &#39;True&#39;, &#39;classifier:passive_aggressive:loss&#39;: &#39;hinge&#39;, &#39;classifier:passive_aggressive:tol&#39;: 1.3440590734654458e-05, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;normalize&#39;, &#39;feature_preprocessor:kernel_pca:kernel&#39;: &#39;sigmoid&#39;, &#39;feature_preprocessor:kernel_pca:n_components&#39;: 1117, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.023878029144121318, &#39;feature_preprocessor:kernel_pca:coef0&#39;: -0.34481806774236556},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.020000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;gradient_boosting&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:gradient_boosting:early_stop&#39;: &#39;valid&#39;, &#39;classifier:gradient_boosting:l2_regularization&#39;: 0.009193751229756878, &#39;classifier:gradient_boosting:learning_rate&#39;: 0.6836522210129492, &#39;classifier:gradient_boosting:loss&#39;: &#39;auto&#39;, &#39;classifier:gradient_boosting:max_bins&#39;: 255, &#39;classifier:gradient_boosting:max_depth&#39;: &#39;None&#39;, &#39;classifier:gradient_boosting:max_leaf_nodes&#39;: 828, &#39;classifier:gradient_boosting:min_samples_leaf&#39;: 3, &#39;classifier:gradient_boosting:scoring&#39;: &#39;loss&#39;, &#39;classifier:gradient_boosting:tol&#39;: 1e-07, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;normalize&#39;, &#39;classifier:gradient_boosting:n_iter_no_change&#39;: 6, &#39;classifier:gradient_boosting:validation_fraction&#39;: 0.28817393948552533, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.00010868227587711292},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.020000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;liblinear_svc&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;pca&#39;, &#39;classifier:liblinear_svc:C&#39;: 2.285306370007781, &#39;classifier:liblinear_svc:dual&#39;: &#39;False&#39;, &#39;classifier:liblinear_svc:fit_intercept&#39;: &#39;True&#39;, &#39;classifier:liblinear_svc:intercept_scaling&#39;: 1, &#39;classifier:liblinear_svc:loss&#39;: &#39;squared_hinge&#39;, &#39;classifier:liblinear_svc:multi_class&#39;: &#39;ovr&#39;, &#39;classifier:liblinear_svc:penalty&#39;: &#39;l2&#39;, &#39;classifier:liblinear_svc:tol&#39;: 0.005821344021856017, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;quantile_transformer&#39;, &#39;feature_preprocessor:pca:keep_variance&#39;: 0.7978382221839412, &#39;feature_preprocessor:pca:whiten&#39;: &#39;True&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles&#39;: 1644, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution&#39;: &#39;normal&#39;},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.020000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;gradient_boosting&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;select_rates_classification&#39;, &#39;classifier:gradient_boosting:early_stop&#39;: &#39;train&#39;, &#39;classifier:gradient_boosting:l2_regularization&#39;: 4.520111487139148e-06, &#39;classifier:gradient_boosting:learning_rate&#39;: 0.12585871697357337, &#39;classifier:gradient_boosting:loss&#39;: &#39;auto&#39;, &#39;classifier:gradient_boosting:max_bins&#39;: 255, &#39;classifier:gradient_boosting:max_depth&#39;: &#39;None&#39;, &#39;classifier:gradient_boosting:max_leaf_nodes&#39;: 1894, &#39;classifier:gradient_boosting:min_samples_leaf&#39;: 2, &#39;classifier:gradient_boosting:scoring&#39;: &#39;loss&#39;, &#39;classifier:gradient_boosting:tol&#39;: 1e-07, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;median&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;robust_scaler&#39;, &#39;feature_preprocessor:select_rates_classification:alpha&#39;: 0.06707365823758892, &#39;feature_preprocessor:select_rates_classification:score_func&#39;: &#39;chi2&#39;, &#39;classifier:gradient_boosting:n_iter_no_change&#39;: 20, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.0014466154794086487, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max&#39;: 0.7104048919054863, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min&#39;: 0.2345103479872152, &#39;feature_preprocessor:select_rates_classification:mode&#39;: &#39;fpr&#39;},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.020000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;lda&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;kernel_pca&#39;, &#39;classifier:lda:shrinkage&#39;: &#39;auto&#39;, &#39;classifier:lda:tol&#39;: 0.014034416932373166, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;median&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;power_transformer&#39;, &#39;feature_preprocessor:kernel_pca:kernel&#39;: &#39;cosine&#39;, &#39;feature_preprocessor:kernel_pca:n_components&#39;: 285, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.0074025063474275335},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.020000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;lda&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;liblinear_svc_preprocessor&#39;, &#39;classifier:lda:shrinkage&#39;: &#39;None&#39;, &#39;classifier:lda:tol&#39;: 0.035543606049804645, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;none&#39;, &#39;feature_preprocessor:liblinear_svc_preprocessor:C&#39;: 4.279105226196228, &#39;feature_preprocessor:liblinear_svc_preprocessor:dual&#39;: &#39;False&#39;, &#39;feature_preprocessor:liblinear_svc_preprocessor:fit_intercept&#39;: &#39;True&#39;, &#39;feature_preprocessor:liblinear_svc_preprocessor:intercept_scaling&#39;: 1, &#39;feature_preprocessor:liblinear_svc_preprocessor:loss&#39;: &#39;squared_hinge&#39;, &#39;feature_preprocessor:liblinear_svc_preprocessor:multi_class&#39;: &#39;ovr&#39;, &#39;feature_preprocessor:liblinear_svc_preprocessor:penalty&#39;: &#39;l1&#39;, &#39;feature_preprocessor:liblinear_svc_preprocessor:tol&#39;: 4.410651487451908e-05},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.020000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;mlp&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;select_rates_classification&#39;, &#39;classifier:mlp:activation&#39;: &#39;tanh&#39;, &#39;classifier:mlp:alpha&#39;: 4.421745686388309e-05, &#39;classifier:mlp:batch_size&#39;: &#39;auto&#39;, &#39;classifier:mlp:beta_1&#39;: 0.9, &#39;classifier:mlp:beta_2&#39;: 0.999, &#39;classifier:mlp:early_stopping&#39;: &#39;train&#39;, &#39;classifier:mlp:epsilon&#39;: 1e-08, &#39;classifier:mlp:hidden_layer_depth&#39;: 3, &#39;classifier:mlp:learning_rate_init&#39;: 0.0005485701552075972, &#39;classifier:mlp:n_iter_no_change&#39;: 32, &#39;classifier:mlp:num_nodes_per_layer&#39;: 36, &#39;classifier:mlp:shuffle&#39;: &#39;True&#39;, &#39;classifier:mlp:solver&#39;: &#39;adam&#39;, &#39;classifier:mlp:tol&#39;: 0.0001, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;minmax&#39;, &#39;feature_preprocessor:select_rates_classification:alpha&#39;: 0.44455023733729254, &#39;feature_preprocessor:select_rates_classification:score_func&#39;: &#39;mutual_info_classif&#39;},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.020000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;mlp&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;extra_trees_preproc_for_classification&#39;, &#39;classifier:mlp:activation&#39;: &#39;relu&#39;, &#39;classifier:mlp:alpha&#39;: 1.4882838781863879e-06, &#39;classifier:mlp:batch_size&#39;: &#39;auto&#39;, &#39;classifier:mlp:beta_1&#39;: 0.9, &#39;classifier:mlp:beta_2&#39;: 0.999, &#39;classifier:mlp:early_stopping&#39;: &#39;train&#39;, &#39;classifier:mlp:epsilon&#39;: 1e-08, &#39;classifier:mlp:hidden_layer_depth&#39;: 2, &#39;classifier:mlp:learning_rate_init&#39;: 0.06290764761689918, &#39;classifier:mlp:n_iter_no_change&#39;: 32, &#39;classifier:mlp:num_nodes_per_layer&#39;: 23, &#39;classifier:mlp:shuffle&#39;: &#39;True&#39;, &#39;classifier:mlp:solver&#39;: &#39;adam&#39;, &#39;classifier:mlp:tol&#39;: 0.0001, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;median&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;quantile_transformer&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:bootstrap&#39;: &#39;False&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:criterion&#39;: &#39;gini&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:max_depth&#39;: &#39;None&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:max_features&#39;: 0.9565048776795676, &#39;feature_preprocessor:extra_trees_preproc_for_classification:max_leaf_nodes&#39;: &#39;None&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_impurity_decrease&#39;: 0.0, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_samples_leaf&#39;: 7, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_samples_split&#39;: 13, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_weight_fraction_leaf&#39;: 0.0, &#39;feature_preprocessor:extra_trees_preproc_for_classification:n_estimators&#39;: 100, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles&#39;: 273, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution&#39;: &#39;normal&#39;},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.020000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;decision_tree&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;random_trees_embedding&#39;, &#39;classifier:decision_tree:criterion&#39;: &#39;gini&#39;, &#39;classifier:decision_tree:max_depth_factor&#39;: 0.6228726042266908, &#39;classifier:decision_tree:max_features&#39;: 1.0, &#39;classifier:decision_tree:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:decision_tree:min_impurity_decrease&#39;: 0.0, &#39;classifier:decision_tree:min_samples_leaf&#39;: 16, &#39;classifier:decision_tree:min_samples_split&#39;: 19, &#39;classifier:decision_tree:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;quantile_transformer&#39;, &#39;feature_preprocessor:random_trees_embedding:bootstrap&#39;: &#39;False&#39;, &#39;feature_preprocessor:random_trees_embedding:max_depth&#39;: 2, &#39;feature_preprocessor:random_trees_embedding:max_leaf_nodes&#39;: &#39;None&#39;, &#39;feature_preprocessor:random_trees_embedding:min_samples_leaf&#39;: 17, &#39;feature_preprocessor:random_trees_embedding:min_samples_split&#39;: 12, &#39;feature_preprocessor:random_trees_embedding:min_weight_fraction_leaf&#39;: 1.0, &#39;feature_preprocessor:random_trees_embedding:n_estimators&#39;: 38, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles&#39;: 606, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution&#39;: &#39;uniform&#39;},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.020000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;decision_tree&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;pca&#39;, &#39;classifier:decision_tree:criterion&#39;: &#39;entropy&#39;, &#39;classifier:decision_tree:max_depth_factor&#39;: 1.1387664107898294, &#39;classifier:decision_tree:max_features&#39;: 1.0, &#39;classifier:decision_tree:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:decision_tree:min_impurity_decrease&#39;: 0.0, &#39;classifier:decision_tree:min_samples_leaf&#39;: 8, &#39;classifier:decision_tree:min_samples_split&#39;: 9, &#39;classifier:decision_tree:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;none&#39;, &#39;feature_preprocessor:pca:keep_variance&#39;: 0.9503612101678187, &#39;feature_preprocessor:pca:whiten&#39;: &#39;True&#39;},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.020000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;bernoulli_nb&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;select_percentile_classification&#39;, &#39;classifier:bernoulli_nb:alpha&#39;: 8.843515731149308, &#39;classifier:bernoulli_nb:fit_prior&#39;: &#39;True&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;robust_scaler&#39;, &#39;feature_preprocessor:select_percentile_classification:percentile&#39;: 50.58615414999356, &#39;feature_preprocessor:select_percentile_classification:score_func&#39;: &#39;chi2&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max&#39;: 0.7852218490061461, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min&#39;: 0.15162626175424537},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-]
+{   3: {   &#39;balancing&#39;: Balancing(random_state=1),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc787842790&gt;,
+           &#39;cost&#39;: 0.05673758865248224,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78fa984f0&gt;,
+           &#39;ensemble_weight&#39;: 0.02,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc7876ba3d0&gt;,
+           &#39;model_id&#39;: 3,
+           &#39;rank&#39;: 16,
+           &#39;sklearn_classifier&#39;: MLPClassifier(alpha=1.0400632553169558e-07, beta_1=0.999, beta_2=0.9,
+              early_stopping=True, hidden_layer_sizes=(35,),
+              learning_rate_init=0.0018475934154961215, max_iter=32,
+              n_iter_no_change=32, random_state=1, verbose=0, warm_start=True)},
+    4: {   &#39;balancing&#39;: Balancing(random_state=1),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc785c0eb80&gt;,
+           &#39;cost&#39;: 0.028368794326241176,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7877aefa0&gt;,
+           &#39;ensemble_weight&#39;: 0.04,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc785c0eb50&gt;,
+           &#39;model_id&#39;: 4,
+           &#39;rank&#39;: 3,
+           &#39;sklearn_classifier&#39;: RandomForestClassifier(max_features=5, n_estimators=512, n_jobs=1,
+                       random_state=1, warm_start=True)},
+    5: {   &#39;balancing&#39;: Balancing(random_state=1),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc785b6ac70&gt;,
+           &#39;cost&#39;: 0.11347517730496459,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78679fe50&gt;,
+           &#39;ensemble_weight&#39;: 0.02,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc785b6a0a0&gt;,
+           &#39;model_id&#39;: 5,
+           &#39;rank&#39;: 28,
+           &#39;sklearn_classifier&#39;: BernoulliNB(alpha=0.010232681621243814)},
+    6: {   &#39;balancing&#39;: Balancing(random_state=1),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78c9c7700&gt;,
+           &#39;cost&#39;: 0.049645390070921946,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78a7f2c70&gt;,
+           &#39;ensemble_weight&#39;: 0.08,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78a605820&gt;,
+           &#39;model_id&#39;: 6,
+           &#39;rank&#39;: 12,
+           &#39;sklearn_classifier&#39;: LinearDiscriminantAnalysis(tol=1.3202262453017803e-05)},
+    7: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78700b100&gt;,
+           &#39;cost&#39;: 0.099290780141844,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7877938b0&gt;,
+           &#39;ensemble_weight&#39;: 0.02,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc787772ca0&gt;,
+           &#39;model_id&#39;: 7,
+           &#39;rank&#39;: 26,
+           &#39;sklearn_classifier&#39;: BernoulliNB(alpha=8.843515731149308)},
+    8: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78662d580&gt;,
+           &#39;cost&#39;: 0.08510638297872342,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc784a023a0&gt;,
+           &#39;ensemble_weight&#39;: 0.02,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78b94c490&gt;,
+           &#39;model_id&#39;: 8,
+           &#39;rank&#39;: 22,
+           &#39;sklearn_classifier&#39;: SGDClassifier(alpha=3.816570529036241e-06, eta0=0.043879389796003546,
+              l1_ratio=0.003142456770095119, learning_rate=&#39;constant&#39;,
+              loss=&#39;squared_hinge&#39;, max_iter=32, penalty=&#39;elasticnet&#39;,
+              random_state=1, tol=0.000896391087588558, warm_start=True)},
+    9: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78657f370&gt;,
+           &#39;cost&#39;: 0.07801418439716312,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc787ae89d0&gt;,
+           &#39;ensemble_weight&#39;: 0.02,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78657f1c0&gt;,
+           &#39;model_id&#39;: 9,
+           &#39;rank&#39;: 21,
+           &#39;sklearn_classifier&#39;: BernoulliNB(alpha=0.04662763358613446, fit_prior=False)},
+    10: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78af797f0&gt;,
+            &#39;cost&#39;: 0.007092198581560294,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc787788bb0&gt;,
+            &#39;ensemble_weight&#39;: 0.06,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78af79550&gt;,
+            &#39;model_id&#39;: 10,
+            &#39;rank&#39;: 1,
+            &#39;sklearn_classifier&#39;: QuadraticDiscriminantAnalysis(reg_param=0.45597150689539434)},
+    11: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78c8f8bb0&gt;,
+            &#39;cost&#39;: 0.028368794326241176,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78779a040&gt;,
+            &#39;ensemble_weight&#39;: 0.04,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c8f8a60&gt;,
+            &#39;model_id&#39;: 11,
+            &#39;rank&#39;: 4,
+            &#39;sklearn_classifier&#39;: QuadraticDiscriminantAnalysis(reg_param=0.0995499108996053)},
+    12: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78a25eb20&gt;,
+            &#39;cost&#39;: 0.049645390070921946,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78698af70&gt;,
+            &#39;ensemble_weight&#39;: 0.02,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c43e5b0&gt;,
+            &#39;model_id&#39;: 12,
+            &#39;rank&#39;: 13,
+            &#39;sklearn_classifier&#39;: LinearDiscriminantAnalysis(shrinkage=&#39;auto&#39;, solver=&#39;lsqr&#39;,
+                           tol=0.014034416932373166)},
+    13: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78fad68e0&gt;,
+            &#39;cost&#39;: 0.021276595744680882,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78faf4670&gt;,
+            &#39;ensemble_weight&#39;: 0.06,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78fad64f0&gt;,
+            &#39;model_id&#39;: 13,
+            &#39;rank&#39;: 2,
+            &#39;sklearn_classifier&#39;: LinearDiscriminantAnalysis(shrinkage=0.2192999307296778, solver=&#39;lsqr&#39;,
+                           tol=0.0001384744065375479)},
+    14: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78c5d43d0&gt;,
+            &#39;cost&#39;: 0.05673758865248224,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc786184070&gt;,
+            &#39;ensemble_weight&#39;: 0.08,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c5d4250&gt;,
+            &#39;model_id&#39;: 14,
+            &#39;rank&#39;: 17,
+            &#39;sklearn_classifier&#39;: LinearDiscriminantAnalysis(tol=0.035543606049804645)},
+    16: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78faf4e50&gt;,
+            &#39;cost&#39;: 0.08510638297872342,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc784ec5c40&gt;,
+            &#39;ensemble_weight&#39;: 0.02,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78fae4fd0&gt;,
+            &#39;model_id&#39;: 16,
+            &#39;rank&#39;: 23,
+            &#39;sklearn_classifier&#39;: DecisionTreeClassifier(criterion=&#39;entropy&#39;, max_depth=1, min_samples_leaf=8,
+                       min_samples_split=9, random_state=1)},
+    17: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78c917340&gt;,
+            &#39;cost&#39;: 0.04255319148936165,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78f83e280&gt;,
+            &#39;ensemble_weight&#39;: 0.02,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c9171f0&gt;,
+            &#39;model_id&#39;: 17,
+            &#39;rank&#39;: 11,
+            &#39;sklearn_classifier&#39;: LinearSVC(C=64.64437157661465, dual=False, intercept_scaling=1.0,
+          random_state=1, tol=0.013399889539803811)},
+    18: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc784f7fa00&gt;,
+            &#39;cost&#39;: 0.099290780141844,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78a4c8340&gt;,
+            &#39;ensemble_weight&#39;: 0.04,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc784f7f190&gt;,
+            &#39;model_id&#39;: 18,
+            &#39;rank&#39;: 27,
+            &#39;sklearn_classifier&#39;: PassiveAggressiveClassifier(C=0.114391364050916, max_iter=128, random_state=1,
+                            tol=1.3440590734654458e-05, warm_start=True)},
+    19: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78ca21940&gt;,
+            &#39;cost&#39;: 0.03546099290780147,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7860954f0&gt;,
+            &#39;ensemble_weight&#39;: 0.02,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc785b994f0&gt;,
+            &#39;model_id&#39;: 19,
+            &#39;rank&#39;: 8,
+            &#39;sklearn_classifier&#39;: QuadraticDiscriminantAnalysis(reg_param=0.01667385805274091)},
+    21: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc7871ff3d0&gt;,
+            &#39;cost&#39;: 0.028368794326241176,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78770aca0&gt;,
+            &#39;ensemble_weight&#39;: 0.06,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78fb2b040&gt;,
+            &#39;model_id&#39;: 21,
+            &#39;rank&#39;: 5,
+            &#39;sklearn_classifier&#39;: LinearSVC(C=2.285306370007781, dual=False, intercept_scaling=1.0,
+          random_state=1, tol=0.005821344021856017)},
+    24: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc784b36370&gt;,
+            &#39;cost&#39;: 0.06382978723404253,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc786bb7a90&gt;,
+            &#39;ensemble_weight&#39;: 0.02,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc786625e20&gt;,
+            &#39;model_id&#39;: 24,
+            &#39;rank&#39;: 18,
+            &#39;sklearn_classifier&#39;: SGDClassifier(alpha=5.189696963506304e-05, eta0=0.007608275318572964,
+              learning_rate=&#39;constant&#39;, loss=&#39;perceptron&#39;, max_iter=64,
+              random_state=1, tol=0.0005024616279807072, warm_start=True)},
+    27: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc785e313d0&gt;,
+            &#39;cost&#39;: 0.09219858156028371,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78b498ac0&gt;,
+            &#39;ensemble_weight&#39;: 0.06,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc784bf14f0&gt;,
+            &#39;model_id&#39;: 27,
+            &#39;rank&#39;: 25,
+            &#39;sklearn_classifier&#39;: DecisionTreeClassifier(criterion=&#39;entropy&#39;, max_depth=4, min_samples_leaf=18,
+                       min_samples_split=8, random_state=1)},
+    28: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc784b39ee0&gt;,
+            &#39;cost&#39;: 0.06382978723404253,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78c5a1e80&gt;,
+            &#39;ensemble_weight&#39;: 0.02,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc784b361f0&gt;,
+            &#39;model_id&#39;: 28,
+            &#39;rank&#39;: 19,
+            &#39;sklearn_classifier&#39;: DecisionTreeClassifier(max_depth=1, min_samples_leaf=16, min_samples_split=19,
+                       random_state=1)},
+    29: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc7880f9eb0&gt;,
+            &#39;cost&#39;: 0.03546099290780147,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78687a1f0&gt;,
+            &#39;ensemble_weight&#39;: 0.04,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78b503dc0&gt;,
+            &#39;model_id&#39;: 29,
+            &#39;rank&#39;: 9,
+            &#39;sklearn_classifier&#39;: QuadraticDiscriminantAnalysis(reg_param=0.23910875621465733)},
+    30: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78662deb0&gt;,
+            &#39;cost&#39;: 0.028368794326241176,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc785e49970&gt;,
+            &#39;ensemble_weight&#39;: 0.04,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78582e1f0&gt;,
+            &#39;model_id&#39;: 30,
+            &#39;rank&#39;: 6,
+            &#39;sklearn_classifier&#39;: RandomForestClassifier(bootstrap=False, criterion=&#39;entropy&#39;, max_features=1,
+                       min_samples_leaf=10, min_samples_split=18,
+                       n_estimators=512, n_jobs=1, random_state=1,
+                       warm_start=True)},
+    31: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc787888d00&gt;,
+            &#39;cost&#39;: 0.049645390070921946,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78a791d60&gt;,
+            &#39;ensemble_weight&#39;: 0.02,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc787888340&gt;,
+            &#39;model_id&#39;: 31,
+            &#39;rank&#39;: 14,
+            &#39;sklearn_classifier&#39;: LinearDiscriminantAnalysis(tol=0.018038793247120707)},
+    33: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc790171700&gt;,
+            &#39;cost&#39;: 0.03546099290780147,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78fad6d60&gt;,
+            &#39;ensemble_weight&#39;: 0.02,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc790179040&gt;,
+            &#39;model_id&#39;: 33,
+            &#39;rank&#39;: 10,
+            &#39;sklearn_classifier&#39;: HistGradientBoostingClassifier(early_stopping=True,
+                               l2_regularization=4.5201114871391565e-06,
+                               learning_rate=0.12585871697357337, max_iter=256,
+                               max_leaf_nodes=1894, min_samples_leaf=2,
+                               n_iter_no_change=20, random_state=1,
+                               validation_fraction=None, warm_start=True)},
+    34: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78b3ce220&gt;,
+            &#39;cost&#39;: 0.24822695035460995,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc787011e50&gt;,
+            &#39;ensemble_weight&#39;: 0.02,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78b3d35e0&gt;,
+            &#39;model_id&#39;: 34,
+            &#39;rank&#39;: 29,
+            &#39;sklearn_classifier&#39;: ExtraTreesClassifier(criterion=&#39;entropy&#39;, max_features=8, min_samples_leaf=15,
+                     min_samples_split=17, n_estimators=512, n_jobs=1,
+                     random_state=1, warm_start=True)},
+    35: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc787e35f70&gt;,
+            &#39;cost&#39;: 0.049645390070921946,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc790367be0&gt;,
+            &#39;ensemble_weight&#39;: 0.04,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78509f550&gt;,
+            &#39;model_id&#39;: 35,
+            &#39;rank&#39;: 15,
+            &#39;sklearn_classifier&#39;: KNeighborsClassifier(n_neighbors=2)},
+    36: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78fa97970&gt;,
+            &#39;cost&#39;: 0.07092198581560283,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc787282b20&gt;,
+            &#39;ensemble_weight&#39;: 0.02,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78fa97670&gt;,
+            &#39;model_id&#39;: 36,
+            &#39;rank&#39;: 20,
+            &#39;sklearn_classifier&#39;: GaussianNB()},
+    37: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc786887dc0&gt;,
+            &#39;cost&#39;: 0.028368794326241176,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78580d250&gt;,
+            &#39;ensemble_weight&#39;: 0.04,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78b2743a0&gt;,
+            &#39;model_id&#39;: 37,
+            &#39;rank&#39;: 7,
+            &#39;sklearn_classifier&#39;: LinearDiscriminantAnalysis(shrinkage=0.32361629219618826, solver=&#39;lsqr&#39;,
+                           tol=3.142032801716559e-05)},
+    38: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78770b160&gt;,
+            &#39;cost&#39;: 0.08510638297872342,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc786567e80&gt;,
+            &#39;ensemble_weight&#39;: 0.02,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78770ba60&gt;,
+            &#39;model_id&#39;: 38,
+            &#39;rank&#39;: 24,
+            &#39;sklearn_classifier&#39;: DecisionTreeClassifier(criterion=&#39;entropy&#39;, max_depth=1, min_samples_leaf=13,
+                       min_samples_split=5, random_state=1)}}
 auto-sklearn results:
   Dataset name: breast_cancer
   Metric: accuracy
   Best validation score: 0.992908
-  Number of target algorithm runs: 33
-  Number of successful target algorithm runs: 31
+  Number of target algorithm runs: 37
+  Number of successful target algorithm runs: 36
   Number of crashed target algorithm runs: 1
-  Number of target algorithms that exceeded the time limit: 1
+  Number of target algorithms that exceeded the time limit: 0
   Number of target algorithms that exceeded the memory limit: 0
 
 Accuracy score 0.9440559440559441
 </pre></div>
 </div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  52.038 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  53.718 seconds)</p>
 <div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-examples-60-search-example-random-search-py">
 <div class="binder-badge docutils container">
 <a class="reference external image-reference" href="https://mybinder.org/v2/gh/automl/auto-sklearn/master?urlpath=lab/tree/notebooks/examples/60_search/example_random_search.ipynb"><img alt="Launch binder" src="../../_images/binder_badge_logo2.svg" width="150px" /></a>
@@ -673,8 +858,8 @@ <h2>Fit a classifier using Random Search<a class="headerlink" href="#fit-a-class
 </div>
 </div>
 <p class="sphx-glr-signature"><a class="reference external" href="https://sphinx-gallery.github.io">Gallery generated by Sphinx-Gallery</a></p>
-</section>
-</section>
+</div>
+</div>
 
 
     </div>
@@ -695,7 +880,7 @@ <h2>Fit a classifier using Random Search<a class="headerlink" href="#fit-a-class
       
     </p>
     <p>
-        &copy; Copyright 2014-2021, Machine Learning Professorship Freiburg.<br/>
+        &copy; Copyright 2014-2022, Machine Learning Professorship Freiburg.<br/>
       Created using <a href="http://sphinx-doc.org/">Sphinx</a> 4.2.0.<br/>
     </p>
   </div>
diff --git a/master/examples/60_search/example_sequential.html b/master/examples/60_search/example_sequential.html
index 953ab49774..79b0662916 100644
--- a/master/examples/60_search/example_sequential.html
+++ b/master/examples/60_search/example_sequential.html
@@ -3,9 +3,8 @@
 <html>
   <head>
     <meta charset="utf-8" />
-    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="generator" content="Docutils 0.17.1: http://docutils.sourceforge.net/" />
-
-    <title>Sequential Usage &#8212; AutoSklearn 0.14.2 documentation</title>
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>Sequential Usage &#8212; AutoSklearn 0.14.4 documentation</title>
     <link rel="stylesheet" type="text/css" href="../../_static/pygments.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/bootstrap-sphinx.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/sg_gallery.css" />
@@ -55,7 +54,7 @@
         </button>
         <a class="navbar-brand" href="../../index.html">
           auto-sklearn</a>
-        <span class="navbar-text navbar-version pull-left"><b>0.14.2</b></span>
+        <span class="navbar-text navbar-version pull-left"><b>0.14.4</b></span>
       </div>
 
         <div class="collapse navbar-collapse nav-collapse">
@@ -125,28 +124,30 @@
 <p>Click <a class="reference internal" href="#sphx-glr-download-examples-60-search-example-sequential-py"><span class="std std-ref">here</span></a>
 to download the full example code or to run this example in your browser via Binder</p>
 </div>
-<section class="sphx-glr-example-title" id="sequential-usage">
+<div class="sphx-glr-example-title section" id="sequential-usage">
 <span id="sphx-glr-examples-60-search-example-sequential-py"></span><h1>Sequential Usage<a class="headerlink" href="#sequential-usage" title="Permalink to this headline">¶</a></h1>
 <p>By default, <em>auto-sklearn</em> fits the machine learning models and build their
 ensembles in parallel. However, it is also possible to run the two processes
 sequentially. The example below shows how to first fit the models and build the
 ensembles afterwards.</p>
-<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">sklearn.model_selection</span>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">pprint</span> <span class="kn">import</span> <span class="n">pprint</span>
+
+<span class="kn">import</span> <span class="nn">sklearn.model_selection</span>
 <span class="kn">import</span> <span class="nn">sklearn.datasets</span>
 <span class="kn">import</span> <span class="nn">sklearn.metrics</span>
 
 <span class="kn">import</span> <span class="nn">autosklearn.classification</span>
 </pre></div>
 </div>
-<section id="data-loading">
+<div class="section" id="data-loading">
 <h2>Data Loading<a class="headerlink" href="#data-loading" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">X</span><span class="p">,</span> <span class="n">y</span> <span class="o">=</span> <span class="n">sklearn</span><span class="o">.</span><span class="n">datasets</span><span class="o">.</span><span class="n">load_breast_cancer</span><span class="p">(</span><span class="n">return_X_y</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
 <span class="n">X_train</span><span class="p">,</span> <span class="n">X_test</span><span class="p">,</span> <span class="n">y_train</span><span class="p">,</span> <span class="n">y_test</span> <span class="o">=</span> \
     <span class="n">sklearn</span><span class="o">.</span><span class="n">model_selection</span><span class="o">.</span><span class="n">train_test_split</span><span class="p">(</span><span class="n">X</span><span class="p">,</span> <span class="n">y</span><span class="p">,</span> <span class="n">random_state</span><span class="o">=</span><span class="mi">1</span><span class="p">)</span>
 </pre></div>
 </div>
-</section>
-<section id="build-and-fit-the-classifier">
+</div>
+<div class="section" id="build-and-fit-the-classifier">
 <h2>Build and fit the classifier<a class="headerlink" href="#build-and-fit-the-classifier" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">automl</span> <span class="o">=</span> <span class="n">autosklearn</span><span class="o">.</span><span class="n">classification</span><span class="o">.</span><span class="n">AutoSklearnClassifier</span><span class="p">(</span>
     <span class="n">time_left_for_this_task</span><span class="o">=</span><span class="mi">120</span><span class="p">,</span>
@@ -166,199 +167,317 @@ <h2>Build and fit the classifier<a class="headerlink" href="#build-and-fit-the-c
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>AutoSklearnClassifier(delete_tmp_folder_after_terminate=False, ensemble_size=0,
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:68: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+  self.metafeatures = self.metafeatures.append(metafeatures)
+/home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:72: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+  self.algorithm_runs[metric].append(runs)
+
+AutoSklearnClassifier(delete_tmp_folder_after_terminate=False, ensemble_size=0,
                       per_run_time_limit=30, time_left_for_this_task=120,
                       tmp_folder=&#39;/tmp/autosklearn_sequential_example_tmp&#39;)
 </pre></div>
 </div>
-</section>
-<section id="print-the-final-ensemble-constructed-by-auto-sklearn">
+</div>
+<div class="section" id="print-the-final-ensemble-constructed-by-auto-sklearn">
 <h2>Print the final ensemble constructed by auto-sklearn<a class="headerlink" href="#print-the-final-ensemble-constructed-by-auto-sklearn" title="Permalink to this headline">¶</a></h2>
-<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="nb">print</span><span class="p">(</span><span class="n">automl</span><span class="o">.</span><span class="n">show_models</span><span class="p">())</span>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">pprint</span><span class="p">(</span><span class="n">automl</span><span class="o">.</span><span class="n">show_models</span><span class="p">(),</span> <span class="n">indent</span><span class="o">=</span><span class="mi">4</span><span class="p">)</span>
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>[(0.100000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;libsvm_svc&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;feature_agglomeration&#39;, &#39;classifier:libsvm_svc:C&#39;: 0.07563955664363409, &#39;classifier:libsvm_svc:gamma&#39;: 1.9836355993007913, &#39;classifier:libsvm_svc:kernel&#39;: &#39;rbf&#39;, &#39;classifier:libsvm_svc:max_iter&#39;: -1, &#39;classifier:libsvm_svc:shrinking&#39;: &#39;True&#39;, &#39;classifier:libsvm_svc:tol&#39;: 0.005003173341372008, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;minmax&#39;, &#39;feature_preprocessor:feature_agglomeration:affinity&#39;: &#39;cosine&#39;, &#39;feature_preprocessor:feature_agglomeration:linkage&#39;: &#39;complete&#39;, &#39;feature_preprocessor:feature_agglomeration:n_clusters&#39;: 58, &#39;feature_preprocessor:feature_agglomeration:pooling_func&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.15890408760884822},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.080000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;mlp&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;feature_agglomeration&#39;, &#39;classifier:mlp:activation&#39;: &#39;tanh&#39;, &#39;classifier:mlp:alpha&#39;: 0.0001363185819149026, &#39;classifier:mlp:batch_size&#39;: &#39;auto&#39;, &#39;classifier:mlp:beta_1&#39;: 0.9, &#39;classifier:mlp:beta_2&#39;: 0.999, &#39;classifier:mlp:early_stopping&#39;: &#39;valid&#39;, &#39;classifier:mlp:epsilon&#39;: 1e-08, &#39;classifier:mlp:hidden_layer_depth&#39;: 3, &#39;classifier:mlp:learning_rate_init&#39;: 0.00018009776276177523, &#39;classifier:mlp:n_iter_no_change&#39;: 32, &#39;classifier:mlp:num_nodes_per_layer&#39;: 115, &#39;classifier:mlp:shuffle&#39;: &#39;True&#39;, &#39;classifier:mlp:solver&#39;: &#39;adam&#39;, &#39;classifier:mlp:tol&#39;: 0.0001, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;median&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;standardize&#39;, &#39;feature_preprocessor:feature_agglomeration:affinity&#39;: &#39;euclidean&#39;, &#39;feature_preprocessor:feature_agglomeration:linkage&#39;: &#39;ward&#39;, &#39;feature_preprocessor:feature_agglomeration:n_clusters&#39;: 182, &#39;feature_preprocessor:feature_agglomeration:pooling_func&#39;: &#39;mean&#39;, &#39;classifier:mlp:validation_fraction&#39;: 0.1},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.080000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;gradient_boosting&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;extra_trees_preproc_for_classification&#39;, &#39;classifier:gradient_boosting:early_stop&#39;: &#39;valid&#39;, &#39;classifier:gradient_boosting:l2_regularization&#39;: 8.057778875694463e-05, &#39;classifier:gradient_boosting:learning_rate&#39;: 0.09179220974965213, &#39;classifier:gradient_boosting:loss&#39;: &#39;auto&#39;, &#39;classifier:gradient_boosting:max_bins&#39;: 255, &#39;classifier:gradient_boosting:max_depth&#39;: &#39;None&#39;, &#39;classifier:gradient_boosting:max_leaf_nodes&#39;: 200, &#39;classifier:gradient_boosting:min_samples_leaf&#39;: 20, &#39;classifier:gradient_boosting:scoring&#39;: &#39;loss&#39;, &#39;classifier:gradient_boosting:tol&#39;: 1e-07, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;quantile_transformer&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:bootstrap&#39;: &#39;False&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:criterion&#39;: &#39;gini&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:max_depth&#39;: &#39;None&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:max_features&#39;: 0.9984367650965825, &#39;feature_preprocessor:extra_trees_preproc_for_classification:max_leaf_nodes&#39;: &#39;None&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_impurity_decrease&#39;: 0.0, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_samples_leaf&#39;: 13, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_samples_split&#39;: 18, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_weight_fraction_leaf&#39;: 0.0, &#39;feature_preprocessor:extra_trees_preproc_for_classification:n_estimators&#39;: 100, &#39;classifier:gradient_boosting:n_iter_no_change&#39;: 18, &#39;classifier:gradient_boosting:validation_fraction&#39;: 0.14295295806077554, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.008542188583124829, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles&#39;: 1102, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution&#39;: &#39;uniform&#39;},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.080000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;extra_trees&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;feature_agglomeration&#39;, &#39;classifier:extra_trees:bootstrap&#39;: &#39;True&#39;, &#39;classifier:extra_trees:criterion&#39;: &#39;entropy&#39;, &#39;classifier:extra_trees:max_depth&#39;: &#39;None&#39;, &#39;classifier:extra_trees:max_features&#39;: 0.3871420537981852, &#39;classifier:extra_trees:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:extra_trees:min_impurity_decrease&#39;: 0.0, &#39;classifier:extra_trees:min_samples_leaf&#39;: 5, &#39;classifier:extra_trees:min_samples_split&#39;: 14, &#39;classifier:extra_trees:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;robust_scaler&#39;, &#39;feature_preprocessor:feature_agglomeration:affinity&#39;: &#39;manhattan&#39;, &#39;feature_preprocessor:feature_agglomeration:linkage&#39;: &#39;complete&#39;, &#39;feature_preprocessor:feature_agglomeration:n_clusters&#39;: 355, &#39;feature_preprocessor:feature_agglomeration:pooling_func&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max&#39;: 0.726182961545702, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min&#39;: 0.02260942170772907},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.060000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;extra_trees&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;polynomial&#39;, &#39;classifier:extra_trees:bootstrap&#39;: &#39;False&#39;, &#39;classifier:extra_trees:criterion&#39;: &#39;gini&#39;, &#39;classifier:extra_trees:max_depth&#39;: &#39;None&#39;, &#39;classifier:extra_trees:max_features&#39;: 0.5707983257382487, &#39;classifier:extra_trees:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:extra_trees:min_impurity_decrease&#39;: 0.0, &#39;classifier:extra_trees:min_samples_leaf&#39;: 3, &#39;classifier:extra_trees:min_samples_split&#39;: 11, &#39;classifier:extra_trees:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;median&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;none&#39;, &#39;feature_preprocessor:polynomial:degree&#39;: 2, &#39;feature_preprocessor:polynomial:include_bias&#39;: &#39;False&#39;, &#39;feature_preprocessor:polynomial:interaction_only&#39;: &#39;False&#39;},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.060000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;extra_trees&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;polynomial&#39;, &#39;classifier:extra_trees:bootstrap&#39;: &#39;False&#39;, &#39;classifier:extra_trees:criterion&#39;: &#39;gini&#39;, &#39;classifier:extra_trees:max_depth&#39;: &#39;None&#39;, &#39;classifier:extra_trees:max_features&#39;: 0.4791448484072812, &#39;classifier:extra_trees:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:extra_trees:min_impurity_decrease&#39;: 0.0, &#39;classifier:extra_trees:min_samples_leaf&#39;: 1, &#39;classifier:extra_trees:min_samples_split&#39;: 6, &#39;classifier:extra_trees:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;none&#39;, &#39;feature_preprocessor:polynomial:degree&#39;: 3, &#39;feature_preprocessor:polynomial:include_bias&#39;: &#39;True&#39;, &#39;feature_preprocessor:polynomial:interaction_only&#39;: &#39;True&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.17184142431776436},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.060000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;gradient_boosting&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;liblinear_svc_preprocessor&#39;, &#39;classifier:gradient_boosting:early_stop&#39;: &#39;train&#39;, &#39;classifier:gradient_boosting:l2_regularization&#39;: 0.005326508887463406, &#39;classifier:gradient_boosting:learning_rate&#39;: 0.060800813211425456, &#39;classifier:gradient_boosting:loss&#39;: &#39;auto&#39;, &#39;classifier:gradient_boosting:max_bins&#39;: 255, &#39;classifier:gradient_boosting:max_depth&#39;: &#39;None&#39;, &#39;classifier:gradient_boosting:max_leaf_nodes&#39;: 6, &#39;classifier:gradient_boosting:min_samples_leaf&#39;: 5, &#39;classifier:gradient_boosting:scoring&#39;: &#39;loss&#39;, &#39;classifier:gradient_boosting:tol&#39;: 1e-07, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;median&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;minmax&#39;, &#39;feature_preprocessor:liblinear_svc_preprocessor:C&#39;: 13.550960330919455, &#39;feature_preprocessor:liblinear_svc_preprocessor:dual&#39;: &#39;False&#39;, &#39;feature_preprocessor:liblinear_svc_preprocessor:fit_intercept&#39;: &#39;True&#39;, &#39;feature_preprocessor:liblinear_svc_preprocessor:intercept_scaling&#39;: 1, &#39;feature_preprocessor:liblinear_svc_preprocessor:loss&#39;: &#39;squared_hinge&#39;, &#39;feature_preprocessor:liblinear_svc_preprocessor:multi_class&#39;: &#39;ovr&#39;, &#39;feature_preprocessor:liblinear_svc_preprocessor:penalty&#39;: &#39;l1&#39;, &#39;feature_preprocessor:liblinear_svc_preprocessor:tol&#39;: 1.2958033930435781e-05, &#39;classifier:gradient_boosting:n_iter_no_change&#39;: 5},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.060000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;extra_trees&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;polynomial&#39;, &#39;classifier:extra_trees:bootstrap&#39;: &#39;True&#39;, &#39;classifier:extra_trees:criterion&#39;: &#39;entropy&#39;, &#39;classifier:extra_trees:max_depth&#39;: &#39;None&#39;, &#39;classifier:extra_trees:max_features&#39;: 0.3793256905143867, &#39;classifier:extra_trees:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:extra_trees:min_impurity_decrease&#39;: 0.0, &#39;classifier:extra_trees:min_samples_leaf&#39;: 20, &#39;classifier:extra_trees:min_samples_split&#39;: 5, &#39;classifier:extra_trees:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;none&#39;, &#39;feature_preprocessor:polynomial:degree&#39;: 3, &#39;feature_preprocessor:polynomial:include_bias&#39;: &#39;False&#39;, &#39;feature_preprocessor:polynomial:interaction_only&#39;: &#39;True&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.00010083678209087315},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.040000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;gradient_boosting&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;select_percentile_classification&#39;, &#39;classifier:gradient_boosting:early_stop&#39;: &#39;train&#39;, &#39;classifier:gradient_boosting:l2_regularization&#39;: 3.387912939529945e-10, &#39;classifier:gradient_boosting:learning_rate&#39;: 0.30755227194768237, &#39;classifier:gradient_boosting:loss&#39;: &#39;auto&#39;, &#39;classifier:gradient_boosting:max_bins&#39;: 255, &#39;classifier:gradient_boosting:max_depth&#39;: &#39;None&#39;, &#39;classifier:gradient_boosting:max_leaf_nodes&#39;: 60, &#39;classifier:gradient_boosting:min_samples_leaf&#39;: 39, &#39;classifier:gradient_boosting:scoring&#39;: &#39;loss&#39;, &#39;classifier:gradient_boosting:tol&#39;: 1e-07, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;none&#39;, &#39;feature_preprocessor:select_percentile_classification:percentile&#39;: 93.39844669585806, &#39;feature_preprocessor:select_percentile_classification:score_func&#39;: &#39;f_classif&#39;, &#39;classifier:gradient_boosting:n_iter_no_change&#39;: 18, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.010000000000000004},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.040000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;extra_trees&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:extra_trees:bootstrap&#39;: &#39;False&#39;, &#39;classifier:extra_trees:criterion&#39;: &#39;gini&#39;, &#39;classifier:extra_trees:max_depth&#39;: &#39;None&#39;, &#39;classifier:extra_trees:max_features&#39;: 0.661377361743419, &#39;classifier:extra_trees:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:extra_trees:min_impurity_decrease&#39;: 0.0, &#39;classifier:extra_trees:min_samples_leaf&#39;: 2, &#39;classifier:extra_trees:min_samples_split&#39;: 13, &#39;classifier:extra_trees:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;none&#39;},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.040000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;extra_trees&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;polynomial&#39;, &#39;classifier:extra_trees:bootstrap&#39;: &#39;False&#39;, &#39;classifier:extra_trees:criterion&#39;: &#39;entropy&#39;, &#39;classifier:extra_trees:max_depth&#39;: &#39;None&#39;, &#39;classifier:extra_trees:max_features&#39;: 0.993803313878608, &#39;classifier:extra_trees:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:extra_trees:min_impurity_decrease&#39;: 0.0, &#39;classifier:extra_trees:min_samples_leaf&#39;: 2, &#39;classifier:extra_trees:min_samples_split&#39;: 20, &#39;classifier:extra_trees:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;median&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;robust_scaler&#39;, &#39;feature_preprocessor:polynomial:degree&#39;: 2, &#39;feature_preprocessor:polynomial:include_bias&#39;: &#39;True&#39;, &#39;feature_preprocessor:polynomial:interaction_only&#39;: &#39;True&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.41826215858914706, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max&#39;: 0.7305615609807856, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min&#39;: 0.25595970768123566},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.040000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;gradient_boosting&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;extra_trees_preproc_for_classification&#39;, &#39;classifier:gradient_boosting:early_stop&#39;: &#39;train&#39;, &#39;classifier:gradient_boosting:l2_regularization&#39;: 0.4635442279519353, &#39;classifier:gradient_boosting:learning_rate&#39;: 0.09809681787962342, &#39;classifier:gradient_boosting:loss&#39;: &#39;auto&#39;, &#39;classifier:gradient_boosting:max_bins&#39;: 255, &#39;classifier:gradient_boosting:max_depth&#39;: &#39;None&#39;, &#39;classifier:gradient_boosting:max_leaf_nodes&#39;: 328, &#39;classifier:gradient_boosting:min_samples_leaf&#39;: 2, &#39;classifier:gradient_boosting:scoring&#39;: &#39;loss&#39;, &#39;classifier:gradient_boosting:tol&#39;: 1e-07, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;median&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;standardize&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:bootstrap&#39;: &#39;False&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:criterion&#39;: &#39;entropy&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:max_depth&#39;: &#39;None&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:max_features&#39;: 0.989729761503726, &#39;feature_preprocessor:extra_trees_preproc_for_classification:max_leaf_nodes&#39;: &#39;None&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_impurity_decrease&#39;: 0.0, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_samples_leaf&#39;: 19, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_samples_split&#39;: 5, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_weight_fraction_leaf&#39;: 0.0, &#39;feature_preprocessor:extra_trees_preproc_for_classification:n_estimators&#39;: 100, &#39;classifier:gradient_boosting:n_iter_no_change&#39;: 2, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.0208475757765491},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.040000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;extra_trees&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;extra_trees_preproc_for_classification&#39;, &#39;classifier:extra_trees:bootstrap&#39;: &#39;False&#39;, &#39;classifier:extra_trees:criterion&#39;: &#39;gini&#39;, &#39;classifier:extra_trees:max_depth&#39;: &#39;None&#39;, &#39;classifier:extra_trees:max_features&#39;: 0.9292309396985746, &#39;classifier:extra_trees:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:extra_trees:min_impurity_decrease&#39;: 0.0, &#39;classifier:extra_trees:min_samples_leaf&#39;: 1, &#39;classifier:extra_trees:min_samples_split&#39;: 10, &#39;classifier:extra_trees:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;median&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;normalize&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:bootstrap&#39;: &#39;False&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:criterion&#39;: &#39;entropy&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:max_depth&#39;: &#39;None&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:max_features&#39;: 0.9929881254946676, &#39;feature_preprocessor:extra_trees_preproc_for_classification:max_leaf_nodes&#39;: &#39;None&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_impurity_decrease&#39;: 0.0, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_samples_leaf&#39;: 1, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_samples_split&#39;: 2, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_weight_fraction_leaf&#39;: 0.0, &#39;feature_preprocessor:extra_trees_preproc_for_classification:n_estimators&#39;: 100, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.032719158639429445},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.040000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;random_forest&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;select_rates_classification&#39;, &#39;classifier:random_forest:bootstrap&#39;: &#39;True&#39;, &#39;classifier:random_forest:criterion&#39;: &#39;gini&#39;, &#39;classifier:random_forest:max_depth&#39;: &#39;None&#39;, &#39;classifier:random_forest:max_features&#39;: 0.5561389303694175, &#39;classifier:random_forest:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:random_forest:min_impurity_decrease&#39;: 0.0, &#39;classifier:random_forest:min_samples_leaf&#39;: 16, &#39;classifier:random_forest:min_samples_split&#39;: 4, &#39;classifier:random_forest:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;quantile_transformer&#39;, &#39;feature_preprocessor:select_rates_classification:alpha&#39;: 0.48507733455436836, &#39;feature_preprocessor:select_rates_classification:score_func&#39;: &#39;f_classif&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles&#39;: 1567, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution&#39;: &#39;normal&#39;, &#39;feature_preprocessor:select_rates_classification:mode&#39;: &#39;fpr&#39;},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.040000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;libsvm_svc&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;select_percentile_classification&#39;, &#39;classifier:libsvm_svc:C&#39;: 5121.989374401993, &#39;classifier:libsvm_svc:gamma&#39;: 0.0024386428371511353, &#39;classifier:libsvm_svc:kernel&#39;: &#39;rbf&#39;, &#39;classifier:libsvm_svc:max_iter&#39;: -1, &#39;classifier:libsvm_svc:shrinking&#39;: &#39;True&#39;, &#39;classifier:libsvm_svc:tol&#39;: 0.0706216214952021, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;median&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;robust_scaler&#39;, &#39;feature_preprocessor:select_percentile_classification:percentile&#39;: 90.1749860017522, &#39;feature_preprocessor:select_percentile_classification:score_func&#39;: &#39;chi2&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.00017888891850981892, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max&#39;: 0.7393440007191528, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min&#39;: 0.04986510953238},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.020000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;mlp&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;fast_ica&#39;, &#39;classifier:mlp:activation&#39;: &#39;relu&#39;, &#39;classifier:mlp:alpha&#39;: 0.0017940473175767063, &#39;classifier:mlp:batch_size&#39;: &#39;auto&#39;, &#39;classifier:mlp:beta_1&#39;: 0.9, &#39;classifier:mlp:beta_2&#39;: 0.999, &#39;classifier:mlp:early_stopping&#39;: &#39;valid&#39;, &#39;classifier:mlp:epsilon&#39;: 1e-08, &#39;classifier:mlp:hidden_layer_depth&#39;: 2, &#39;classifier:mlp:learning_rate_init&#39;: 0.0004684917334431039, &#39;classifier:mlp:n_iter_no_change&#39;: 32, &#39;classifier:mlp:num_nodes_per_layer&#39;: 101, &#39;classifier:mlp:shuffle&#39;: &#39;True&#39;, &#39;classifier:mlp:solver&#39;: &#39;adam&#39;, &#39;classifier:mlp:tol&#39;: 0.0001, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;standardize&#39;, &#39;feature_preprocessor:fast_ica:algorithm&#39;: &#39;parallel&#39;, &#39;feature_preprocessor:fast_ica:fun&#39;: &#39;exp&#39;, &#39;feature_preprocessor:fast_ica:whiten&#39;: &#39;False&#39;, &#39;classifier:mlp:validation_fraction&#39;: 0.1},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.020000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;mlp&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;extra_trees_preproc_for_classification&#39;, &#39;classifier:mlp:activation&#39;: &#39;tanh&#39;, &#39;classifier:mlp:alpha&#39;: 2.5550223982458062e-06, &#39;classifier:mlp:batch_size&#39;: &#39;auto&#39;, &#39;classifier:mlp:beta_1&#39;: 0.9, &#39;classifier:mlp:beta_2&#39;: 0.999, &#39;classifier:mlp:early_stopping&#39;: &#39;train&#39;, &#39;classifier:mlp:epsilon&#39;: 1e-08, &#39;classifier:mlp:hidden_layer_depth&#39;: 3, &#39;classifier:mlp:learning_rate_init&#39;: 0.00027271287919467994, &#39;classifier:mlp:n_iter_no_change&#39;: 32, &#39;classifier:mlp:num_nodes_per_layer&#39;: 54, &#39;classifier:mlp:shuffle&#39;: &#39;True&#39;, &#39;classifier:mlp:solver&#39;: &#39;adam&#39;, &#39;classifier:mlp:tol&#39;: 0.0001, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;standardize&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:bootstrap&#39;: &#39;False&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:criterion&#39;: &#39;entropy&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:max_depth&#39;: &#39;None&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:max_features&#39;: 0.7171678618990129, &#39;feature_preprocessor:extra_trees_preproc_for_classification:max_leaf_nodes&#39;: &#39;None&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_impurity_decrease&#39;: 0.0, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_samples_leaf&#39;: 4, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_samples_split&#39;: 17, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_weight_fraction_leaf&#39;: 0.0, &#39;feature_preprocessor:extra_trees_preproc_for_classification:n_estimators&#39;: 100, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.017116661677715188},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.020000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;liblinear_svc&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;select_percentile_classification&#39;, &#39;classifier:liblinear_svc:C&#39;: 140.23434217313954, &#39;classifier:liblinear_svc:dual&#39;: &#39;False&#39;, &#39;classifier:liblinear_svc:fit_intercept&#39;: &#39;True&#39;, &#39;classifier:liblinear_svc:intercept_scaling&#39;: 1, &#39;classifier:liblinear_svc:loss&#39;: &#39;squared_hinge&#39;, &#39;classifier:liblinear_svc:multi_class&#39;: &#39;ovr&#39;, &#39;classifier:liblinear_svc:penalty&#39;: &#39;l2&#39;, &#39;classifier:liblinear_svc:tol&#39;: 0.0002944604673080368, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;quantile_transformer&#39;, &#39;feature_preprocessor:select_percentile_classification:percentile&#39;: 29.807816607300627, &#39;feature_preprocessor:select_percentile_classification:score_func&#39;: &#39;chi2&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.034330895903523305, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles&#39;: 1450, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution&#39;: &#39;uniform&#39;},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.020000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;mlp&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;feature_agglomeration&#39;, &#39;classifier:mlp:activation&#39;: &#39;relu&#39;, &#39;classifier:mlp:alpha&#39;: 4.2841884333778574e-06, &#39;classifier:mlp:batch_size&#39;: &#39;auto&#39;, &#39;classifier:mlp:beta_1&#39;: 0.9, &#39;classifier:mlp:beta_2&#39;: 0.999, &#39;classifier:mlp:early_stopping&#39;: &#39;train&#39;, &#39;classifier:mlp:epsilon&#39;: 1e-08, &#39;classifier:mlp:hidden_layer_depth&#39;: 3, &#39;classifier:mlp:learning_rate_init&#39;: 0.0011804284312897009, &#39;classifier:mlp:n_iter_no_change&#39;: 32, &#39;classifier:mlp:num_nodes_per_layer&#39;: 263, &#39;classifier:mlp:shuffle&#39;: &#39;True&#39;, &#39;classifier:mlp:solver&#39;: &#39;adam&#39;, &#39;classifier:mlp:tol&#39;: 0.0001, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;minmax&#39;, &#39;feature_preprocessor:feature_agglomeration:affinity&#39;: &#39;manhattan&#39;, &#39;feature_preprocessor:feature_agglomeration:linkage&#39;: &#39;average&#39;, &#39;feature_preprocessor:feature_agglomeration:n_clusters&#39;: 338, &#39;feature_preprocessor:feature_agglomeration:pooling_func&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.004071801722749603},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.020000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;mlp&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;fast_ica&#39;, &#39;classifier:mlp:activation&#39;: &#39;relu&#39;, &#39;classifier:mlp:alpha&#39;: 0.02847755502162456, &#39;classifier:mlp:batch_size&#39;: &#39;auto&#39;, &#39;classifier:mlp:beta_1&#39;: 0.9, &#39;classifier:mlp:beta_2&#39;: 0.999, &#39;classifier:mlp:early_stopping&#39;: &#39;train&#39;, &#39;classifier:mlp:epsilon&#39;: 1e-08, &#39;classifier:mlp:hidden_layer_depth&#39;: 2, &#39;classifier:mlp:learning_rate_init&#39;: 0.000421568792103947, &#39;classifier:mlp:n_iter_no_change&#39;: 32, &#39;classifier:mlp:num_nodes_per_layer&#39;: 123, &#39;classifier:mlp:shuffle&#39;: &#39;True&#39;, &#39;classifier:mlp:solver&#39;: &#39;adam&#39;, &#39;classifier:mlp:tol&#39;: 0.0001, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;median&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;minmax&#39;, &#39;feature_preprocessor:fast_ica:algorithm&#39;: &#39;parallel&#39;, &#39;feature_preprocessor:fast_ica:fun&#39;: &#39;exp&#39;, &#39;feature_preprocessor:fast_ica:whiten&#39;: &#39;False&#39;},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.020000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;bernoulli_nb&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;select_rates_classification&#39;, &#39;classifier:bernoulli_nb:alpha&#39;: 39.87397441278958, &#39;classifier:bernoulli_nb:fit_prior&#39;: &#39;False&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;median&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;standardize&#39;, &#39;feature_preprocessor:select_rates_classification:alpha&#39;: 0.3094962987325228, &#39;feature_preprocessor:select_rates_classification:score_func&#39;: &#39;mutual_info_classif&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.09580337973953734},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.020000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;passive_aggressive&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:passive_aggressive:C&#39;: 0.14268277711454813, &#39;classifier:passive_aggressive:average&#39;: &#39;False&#39;, &#39;classifier:passive_aggressive:fit_intercept&#39;: &#39;True&#39;, &#39;classifier:passive_aggressive:loss&#39;: &#39;hinge&#39;, &#39;classifier:passive_aggressive:tol&#39;: 0.0002600768160857831, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;none&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.0070580904199417415},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-]
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>{   2: {   &#39;balancing&#39;: Balancing(random_state=1),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc7901c5dc0&gt;,
+           &#39;cost&#39;: 0.028368794326241176,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78a5f4f70&gt;,
+           &#39;ensemble_weight&#39;: 0.020000000000000004,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc790049a90&gt;,
+           &#39;model_id&#39;: 2,
+           &#39;rank&#39;: 4,
+           &#39;sklearn_classifier&#39;: RandomForestClassifier(max_features=5, n_estimators=512, n_jobs=1,
+                       random_state=1, warm_start=True)},
+    4: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc786184580&gt;,
+           &#39;cost&#39;: 0.05673758865248224,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7879e2c10&gt;,
+           &#39;ensemble_weight&#39;: 0.04000000000000001,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78fb05610&gt;,
+           &#39;model_id&#39;: 4,
+           &#39;rank&#39;: 21,
+           &#39;sklearn_classifier&#39;: MLPClassifier(activation=&#39;tanh&#39;, alpha=0.00021148999718383549, beta_1=0.999,
+              beta_2=0.9, hidden_layer_sizes=(113, 113, 113),
+              learning_rate_init=0.0007452270241186694, max_iter=64,
+              n_iter_no_change=32, random_state=1, validation_fraction=0.0,
+              verbose=0, warm_start=True)},
+    5: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc786918190&gt;,
+           &#39;cost&#39;: 0.03546099290780147,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc785e4f4f0&gt;,
+           &#39;ensemble_weight&#39;: 0.020000000000000004,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc787a9e310&gt;,
+           &#39;model_id&#39;: 5,
+           &#39;rank&#39;: 14,
+           &#39;sklearn_classifier&#39;: RandomForestClassifier(criterion=&#39;entropy&#39;, max_features=3, min_samples_leaf=2,
+                       n_estimators=512, n_jobs=1, random_state=1,
+                       warm_start=True)},
+    6: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78c944ca0&gt;,
+           &#39;cost&#39;: 0.028368794326241176,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7902bd4f0&gt;,
+           &#39;ensemble_weight&#39;: 0.10000000000000002,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c944160&gt;,
+           &#39;model_id&#39;: 6,
+           &#39;rank&#39;: 5,
+           &#39;sklearn_classifier&#39;: MLPClassifier(alpha=0.0017940473175767063, beta_1=0.999, beta_2=0.9,
+              early_stopping=True, hidden_layer_sizes=(101, 101),
+              learning_rate_init=0.0004684917334431039, max_iter=32,
+              n_iter_no_change=32, random_state=1, verbose=0, warm_start=True)},
+    7: {   &#39;balancing&#39;: Balancing(random_state=1),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78a7d9bb0&gt;,
+           &#39;cost&#39;: 0.014184397163120588,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78ca46460&gt;,
+           &#39;ensemble_weight&#39;: 0.04000000000000001,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78ff934c0&gt;,
+           &#39;model_id&#39;: 7,
+           &#39;rank&#39;: 1,
+           &#39;sklearn_classifier&#39;: ExtraTreesClassifier(max_features=34, min_samples_leaf=3, min_samples_split=11,
+                     n_estimators=512, n_jobs=1, random_state=1,
+                     warm_start=True)},
+    8: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc787005400&gt;,
+           &#39;cost&#39;: 0.03546099290780147,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc787e71580&gt;,
+           &#39;ensemble_weight&#39;: 0.060000000000000005,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc786918580&gt;,
+           &#39;model_id&#39;: 8,
+           &#39;rank&#39;: 15,
+           &#39;sklearn_classifier&#39;: RandomForestClassifier(max_features=2, min_samples_leaf=2, n_estimators=512,
+                       n_jobs=1, random_state=1, warm_start=True)},
+    9: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78fc1f100&gt;,
+           &#39;cost&#39;: 0.04255319148936165,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7857b5a30&gt;,
+           &#39;ensemble_weight&#39;: 0.060000000000000005,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78bae9040&gt;,
+           &#39;model_id&#39;: 9,
+           &#39;rank&#39;: 19,
+           &#39;sklearn_classifier&#39;: ExtraTreesClassifier(max_features=9, min_samples_split=10, n_estimators=512,
+                     n_jobs=1, random_state=1, warm_start=True)},
+    10: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78f986940&gt;,
+            &#39;cost&#39;: 0.028368794326241176,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78c482100&gt;,
+            &#39;ensemble_weight&#39;: 0.060000000000000005,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78bcec7c0&gt;,
+            &#39;model_id&#39;: 10,
+            &#39;rank&#39;: 6,
+            &#39;sklearn_classifier&#39;: RandomForestClassifier(criterion=&#39;entropy&#39;, max_features=4, min_samples_split=6,
+                       n_estimators=512, n_jobs=1, random_state=1,
+                       warm_start=True)},
+    11: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78ab3f070&gt;,
+            &#39;cost&#39;: 0.028368794326241176,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78be43c70&gt;,
+            &#39;ensemble_weight&#39;: 0.04000000000000001,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78ab3f550&gt;,
+            &#39;model_id&#39;: 11,
+            &#39;rank&#39;: 7,
+            &#39;sklearn_classifier&#39;: RandomForestClassifier(criterion=&#39;entropy&#39;, max_features=23, min_samples_leaf=7,
+                       n_estimators=512, n_jobs=1, random_state=1,
+                       warm_start=True)},
+    13: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78b2344c0&gt;,
+            &#39;cost&#39;: 0.028368794326241176,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc786095eb0&gt;,
+            &#39;ensemble_weight&#39;: 0.020000000000000004,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78b2451f0&gt;,
+            &#39;model_id&#39;: 13,
+            &#39;rank&#39;: 8,
+            &#39;sklearn_classifier&#39;: HistGradientBoostingClassifier(early_stopping=False,
+                               l2_regularization=1.0647401999412075e-10,
+                               learning_rate=0.08291320147381159, max_iter=512,
+                               max_leaf_nodes=39, n_iter_no_change=0,
+                               random_state=1, validation_fraction=None,
+                               warm_start=True)},
+    15: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78c482310&gt;,
+            &#39;cost&#39;: 0.049645390070921946,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc785872910&gt;,
+            &#39;ensemble_weight&#39;: 0.020000000000000004,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78fc1fc40&gt;,
+            &#39;model_id&#39;: 15,
+            &#39;rank&#39;: 20,
+            &#39;sklearn_classifier&#39;: MLPClassifier(alpha=4.2841884333778574e-06, beta_1=0.999, beta_2=0.9,
+              hidden_layer_sizes=(263, 263, 263),
+              learning_rate_init=0.0011804284312897009, max_iter=128,
+              n_iter_no_change=32, random_state=1, validation_fraction=0.0,
+              verbose=0, warm_start=True)},
+    17: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc785b7d6a0&gt;,
+            &#39;cost&#39;: 0.03546099290780147,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7870329a0&gt;,
+            &#39;ensemble_weight&#39;: 0.020000000000000004,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc7870056d0&gt;,
+            &#39;model_id&#39;: 17,
+            &#39;rank&#39;: 16,
+            &#39;sklearn_classifier&#39;: HistGradientBoostingClassifier(early_stopping=True,
+                               l2_regularization=0.4635442279519353,
+                               learning_rate=0.09809681787962342, max_iter=512,
+                               max_leaf_nodes=328, min_samples_leaf=2,
+                               n_iter_no_change=2, random_state=1,
+                               validation_fraction=None, warm_start=True)},
+    18: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78a38fd30&gt;,
+            &#39;cost&#39;: 0.03546099290780147,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78790c850&gt;,
+            &#39;ensemble_weight&#39;: 0.04000000000000001,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc785b7d5e0&gt;,
+            &#39;model_id&#39;: 18,
+            &#39;rank&#39;: 17,
+            &#39;sklearn_classifier&#39;: RandomForestClassifier(criterion=&#39;entropy&#39;, max_features=3, n_estimators=512,
+                       n_jobs=1, random_state=1, warm_start=True)},
+    19: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78438b220&gt;,
+            &#39;cost&#39;: 0.028368794326241176,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc786c78730&gt;,
+            &#39;ensemble_weight&#39;: 0.020000000000000004,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78438b070&gt;,
+            &#39;model_id&#39;: 19,
+            &#39;rank&#39;: 9,
+            &#39;sklearn_classifier&#39;: ExtraTreesClassifier(criterion=&#39;entropy&#39;, max_features=448, min_samples_leaf=2,
+                     min_samples_split=20, n_estimators=512, n_jobs=1,
+                     random_state=1, warm_start=True)},
+    21: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78a791d30&gt;,
+            &#39;cost&#39;: 0.021276595744680882,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78a593e50&gt;,
+            &#39;ensemble_weight&#39;: 0.04000000000000001,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc79068ba00&gt;,
+            &#39;model_id&#39;: 21,
+            &#39;rank&#39;: 3,
+            &#39;sklearn_classifier&#39;: ExtraTreesClassifier(criterion=&#39;entropy&#39;, max_features=4, min_samples_leaf=2,
+                     min_samples_split=15, n_estimators=512, n_jobs=1,
+                     random_state=1, warm_start=True)},
+    26: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc7857f8c40&gt;,
+            &#39;cost&#39;: 0.028368794326241176,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78ab77d60&gt;,
+            &#39;ensemble_weight&#39;: 0.04000000000000001,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc7857f88e0&gt;,
+            &#39;model_id&#39;: 26,
+            &#39;rank&#39;: 10,
+            &#39;sklearn_classifier&#39;: ExtraTreesClassifier(criterion=&#39;entropy&#39;, max_features=414, min_samples_leaf=2,
+                     min_samples_split=19, n_estimators=512, n_jobs=1,
+                     random_state=1, warm_start=True)},
+    27: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc784e47520&gt;,
+            &#39;cost&#39;: 0.028368794326241176,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc786b74100&gt;,
+            &#39;ensemble_weight&#39;: 0.020000000000000004,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc784e47160&gt;,
+            &#39;model_id&#39;: 27,
+            &#39;rank&#39;: 11,
+            &#39;sklearn_classifier&#39;: ExtraTreesClassifier(bootstrap=True, criterion=&#39;entropy&#39;, max_features=4126,
+                     min_samples_leaf=7, min_samples_split=13, n_estimators=512,
+                     n_jobs=1, random_state=1, warm_start=True)},
+    28: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78b4f2310&gt;,
+            &#39;cost&#39;: 0.06382978723404253,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78666a6d0&gt;,
+            &#39;ensemble_weight&#39;: 0.020000000000000004,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc790366ac0&gt;,
+            &#39;model_id&#39;: 28,
+            &#39;rank&#39;: 22,
+            &#39;sklearn_classifier&#39;: LinearDiscriminantAnalysis(tol=0.032840833359764676)},
+    29: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78c53a9d0&gt;,
+            &#39;cost&#39;: 0.028368794326241176,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc787368bb0&gt;,
+            &#39;ensemble_weight&#39;: 0.020000000000000004,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c53a5e0&gt;,
+            &#39;model_id&#39;: 29,
+            &#39;rank&#39;: 12,
+            &#39;sklearn_classifier&#39;: ExtraTreesClassifier(criterion=&#39;entropy&#39;, max_features=8, min_samples_leaf=11,
+                     min_samples_split=5, n_estimators=512, n_jobs=1,
+                     random_state=1, warm_start=True)},
+    31: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78a378070&gt;,
+            &#39;cost&#39;: 0.03546099290780147,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc785fb2160&gt;,
+            &#39;ensemble_weight&#39;: 0.04000000000000001,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78a38fc40&gt;,
+            &#39;model_id&#39;: 31,
+            &#39;rank&#39;: 18,
+            &#39;sklearn_classifier&#39;: LinearSVC(class_weight=&#39;balanced&#39;, dual=False, intercept_scaling=1.0,
+          random_state=1, tol=0.0001350687148423338)},
+    32: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc7850a2c70&gt;,
+            &#39;cost&#39;: 0.028368794326241176,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc784f51070&gt;,
+            &#39;ensemble_weight&#39;: 0.04000000000000001,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc7850a2760&gt;,
+            &#39;model_id&#39;: 32,
+            &#39;rank&#39;: 13,
+            &#39;sklearn_classifier&#39;: LinearSVC(C=832.5744883902069, dual=False, intercept_scaling=1.0,
+          random_state=1, tol=0.015340961719445064)},
+    34: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc784c102e0&gt;,
+            &#39;cost&#39;: 0.06382978723404253,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc786abb970&gt;,
+            &#39;ensemble_weight&#39;: 0.020000000000000004,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc784b3b400&gt;,
+            &#39;model_id&#39;: 34,
+            &#39;rank&#39;: 23,
+            &#39;sklearn_classifier&#39;: RandomForestClassifier(bootstrap=False, criterion=&#39;entropy&#39;, max_features=54,
+                       min_samples_leaf=14, min_samples_split=3,
+                       n_estimators=512, n_jobs=1, random_state=1,
+                       warm_start=True)},
+    36: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc786a9d310&gt;,
+            &#39;cost&#39;: 0.36879432624113473,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc787e1a9a0&gt;,
+            &#39;ensemble_weight&#39;: 0.08000000000000002,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc786a909a0&gt;,
+            &#39;model_id&#39;: 36,
+            &#39;rank&#39;: 26,
+            &#39;sklearn_classifier&#39;: ExtraTreesClassifier(bootstrap=True, max_features=4, min_samples_leaf=13,
+                     min_samples_split=11, n_estimators=512, n_jobs=1,
+                     random_state=1, warm_start=True)},
+    39: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78f69f220&gt;,
+            &#39;cost&#39;: 0.014184397163120588,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78c384be0&gt;,
+            &#39;ensemble_weight&#39;: 0.060000000000000005,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc790366ee0&gt;,
+            &#39;model_id&#39;: 39,
+            &#39;rank&#39;: 2,
+            &#39;sklearn_classifier&#39;: ExtraTreesClassifier(criterion=&#39;entropy&#39;, max_features=32, min_samples_leaf=8,
+                     min_samples_split=8, n_estimators=512, n_jobs=1,
+                     random_state=1, warm_start=True)},
+    41: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78612c5b0&gt;,
+            &#39;cost&#39;: 0.07092198581560283,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78febd7c0&gt;,
+            &#39;ensemble_weight&#39;: 0.020000000000000004,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78612c280&gt;,
+            &#39;model_id&#39;: 41,
+            &#39;rank&#39;: 24,
+            &#39;sklearn_classifier&#39;: DecisionTreeClassifier(max_depth=1, min_samples_leaf=15, min_samples_split=5,
+                       random_state=1)},
+    43: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78bc70c10&gt;,
+            &#39;cost&#39;: 0.09219858156028371,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7873bf130&gt;,
+            &#39;ensemble_weight&#39;: 0.04000000000000001,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78b800b20&gt;,
+            &#39;model_id&#39;: 43,
+            &#39;rank&#39;: 25,
+            &#39;sklearn_classifier&#39;: ExtraTreesClassifier(bootstrap=True, criterion=&#39;entropy&#39;, max_features=12,
+                     min_samples_leaf=19, min_samples_split=17,
+                     n_estimators=512, n_jobs=1, random_state=1,
+                     warm_start=True)}}
 </pre></div>
 </div>
-</section>
-<section id="get-the-score-of-the-final-ensemble">
+</div>
+<div class="section" id="get-the-score-of-the-final-ensemble">
 <h2>Get the Score of the final ensemble<a class="headerlink" href="#get-the-score-of-the-final-ensemble" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">predictions</span> <span class="o">=</span> <span class="n">automl</span><span class="o">.</span><span class="n">predict</span><span class="p">(</span><span class="n">X_test</span><span class="p">)</span>
 <span class="nb">print</span><span class="p">(</span><span class="n">automl</span><span class="o">.</span><span class="n">sprint_statistics</span><span class="p">())</span>
@@ -370,16 +489,16 @@ <h2>Get the Score of the final ensemble<a class="headerlink" href="#get-the-scor
   Dataset name: breast_cancer
   Metric: accuracy
   Best validation score: 0.985816
-  Number of target algorithm runs: 41
-  Number of successful target algorithm runs: 41
+  Number of target algorithm runs: 44
+  Number of successful target algorithm runs: 43
   Number of crashed target algorithm runs: 0
-  Number of target algorithms that exceeded the time limit: 0
+  Number of target algorithms that exceeded the time limit: 1
   Number of target algorithms that exceeded the memory limit: 0
 
-Accuracy score 0.951048951048951
+Accuracy score 0.958041958041958
 </pre></div>
 </div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes  4.152 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes  1.878 seconds)</p>
 <div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-examples-60-search-example-sequential-py">
 <div class="binder-badge docutils container">
 <a class="reference external image-reference" href="https://mybinder.org/v2/gh/automl/auto-sklearn/master?urlpath=lab/tree/notebooks/examples/60_search/example_sequential.ipynb"><img alt="Launch binder" src="../../_images/binder_badge_logo2.svg" width="150px" /></a>
@@ -392,8 +511,8 @@ <h2>Get the Score of the final ensemble<a class="headerlink" href="#get-the-scor
 </div>
 </div>
 <p class="sphx-glr-signature"><a class="reference external" href="https://sphinx-gallery.github.io">Gallery generated by Sphinx-Gallery</a></p>
-</section>
-</section>
+</div>
+</div>
 
 
     </div>
@@ -414,7 +533,7 @@ <h2>Get the Score of the final ensemble<a class="headerlink" href="#get-the-scor
       
     </p>
     <p>
-        &copy; Copyright 2014-2021, Machine Learning Professorship Freiburg.<br/>
+        &copy; Copyright 2014-2022, Machine Learning Professorship Freiburg.<br/>
       Created using <a href="http://sphinx-doc.org/">Sphinx</a> 4.2.0.<br/>
     </p>
   </div>
diff --git a/master/examples/60_search/example_successive_halving.html b/master/examples/60_search/example_successive_halving.html
index 1774733d32..6a4fb0a402 100644
--- a/master/examples/60_search/example_successive_halving.html
+++ b/master/examples/60_search/example_successive_halving.html
@@ -3,9 +3,8 @@
 <html>
   <head>
     <meta charset="utf-8" />
-    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="generator" content="Docutils 0.17.1: http://docutils.sourceforge.net/" />
-
-    <title>Successive Halving &#8212; AutoSklearn 0.14.2 documentation</title>
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>Successive Halving &#8212; AutoSklearn 0.14.4 documentation</title>
     <link rel="stylesheet" type="text/css" href="../../_static/pygments.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/bootstrap-sphinx.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/sg_gallery.css" />
@@ -55,7 +54,7 @@
         </button>
         <a class="navbar-brand" href="../../index.html">
           auto-sklearn</a>
-        <span class="navbar-text navbar-version pull-left"><b>0.14.2</b></span>
+        <span class="navbar-text navbar-version pull-left"><b>0.14.4</b></span>
       </div>
 
         <div class="collapse navbar-collapse nav-collapse">
@@ -128,7 +127,7 @@
 <p>Click <a class="reference internal" href="#sphx-glr-download-examples-60-search-example-successive-halving-py"><span class="std std-ref">here</span></a>
 to download the full example code or to run this example in your browser via Binder</p>
 </div>
-<section class="sphx-glr-example-title" id="successive-halving">
+<div class="sphx-glr-example-title section" id="successive-halving">
 <span id="sphx-glr-examples-60-search-example-successive-halving-py"></span><h1>Successive Halving<a class="headerlink" href="#successive-halving" title="Permalink to this headline">¶</a></h1>
 <p>This advanced  example illustrates how to interact with
 the SMAC callback and get relevant information from the run, like
@@ -138,14 +137,16 @@
 <p>This results in an adaptation of the <a class="reference external" href="http://proceedings.mlr.press/v80/falkner18a/falkner18a.pdf">BOHB algorithm</a>.
 It uses Successive Halving instead of <a class="reference external" href="https://jmlr.org/papers/volume18/16-558/16-558.pdf">Hyperband</a>, and could be abbreviated as BOSH.
 To get the BOHB algorithm, simply import Hyperband and use it as the intensification strategy.</p>
-<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">sklearn.model_selection</span>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">pprint</span> <span class="kn">import</span> <span class="n">pprint</span>
+
+<span class="kn">import</span> <span class="nn">sklearn.model_selection</span>
 <span class="kn">import</span> <span class="nn">sklearn.datasets</span>
 <span class="kn">import</span> <span class="nn">sklearn.metrics</span>
 
 <span class="kn">import</span> <span class="nn">autosklearn.classification</span>
 </pre></div>
 </div>
-<section id="define-a-callback-that-instantiates-successivehalving">
+<div class="section" id="define-a-callback-that-instantiates-successivehalving">
 <h2>Define a callback that instantiates SuccessiveHalving<a class="headerlink" href="#define-a-callback-that-instantiates-successivehalving" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="k">def</span> <span class="nf">get_smac_object_callback</span><span class="p">(</span><span class="n">budget_type</span><span class="p">):</span>
     <span class="k">def</span> <span class="nf">get_smac_object</span><span class="p">(</span>
@@ -197,16 +198,16 @@ <h2>Define a callback that instantiates SuccessiveHalving<a class="headerlink" h
     <span class="k">return</span> <span class="n">get_smac_object</span>
 </pre></div>
 </div>
-</section>
-<section id="data-loading">
+</div>
+<div class="section" id="data-loading">
 <h2>Data Loading<a class="headerlink" href="#data-loading" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">X</span><span class="p">,</span> <span class="n">y</span> <span class="o">=</span> <span class="n">sklearn</span><span class="o">.</span><span class="n">datasets</span><span class="o">.</span><span class="n">load_breast_cancer</span><span class="p">(</span><span class="n">return_X_y</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
 <span class="n">X_train</span><span class="p">,</span> <span class="n">X_test</span><span class="p">,</span> <span class="n">y_train</span><span class="p">,</span> <span class="n">y_test</span> <span class="o">=</span> \
     <span class="n">sklearn</span><span class="o">.</span><span class="n">model_selection</span><span class="o">.</span><span class="n">train_test_split</span><span class="p">(</span><span class="n">X</span><span class="p">,</span> <span class="n">y</span><span class="p">,</span> <span class="n">random_state</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> <span class="n">shuffle</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
 </pre></div>
 </div>
-</section>
-<section id="build-and-fit-a-classifier">
+</div>
+<div class="section" id="build-and-fit-a-classifier">
 <h2>Build and fit a classifier<a class="headerlink" href="#build-and-fit-a-classifier" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">automl</span> <span class="o">=</span> <span class="n">autosklearn</span><span class="o">.</span><span class="n">classification</span><span class="o">.</span><span class="n">AutoSklearnClassifier</span><span class="p">(</span>
     <span class="n">time_left_for_this_task</span><span class="o">=</span><span class="mi">40</span><span class="p">,</span>
@@ -229,7 +230,7 @@ <h2>Build and fit a classifier<a class="headerlink" href="#build-and-fit-a-class
 <span class="p">)</span>
 <span class="n">automl</span><span class="o">.</span><span class="n">fit</span><span class="p">(</span><span class="n">X_train</span><span class="p">,</span> <span class="n">y_train</span><span class="p">,</span> <span class="n">dataset_name</span><span class="o">=</span><span class="s1">&#39;breast_cancer&#39;</span><span class="p">)</span>
 
-<span class="nb">print</span><span class="p">(</span><span class="n">automl</span><span class="o">.</span><span class="n">show_models</span><span class="p">())</span>
+<span class="n">pprint</span><span class="p">(</span><span class="n">automl</span><span class="o">.</span><span class="n">show_models</span><span class="p">(),</span> <span class="n">indent</span><span class="o">=</span><span class="mi">4</span><span class="p">)</span>
 <span class="n">predictions</span> <span class="o">=</span> <span class="n">automl</span><span class="o">.</span><span class="n">predict</span><span class="p">(</span><span class="n">X_test</span><span class="p">)</span>
 <span class="c1"># Print statistics about the auto-sklearn run such as number of</span>
 <span class="c1"># iterations, number of models failed with a time out.</span>
@@ -238,176 +239,243 @@ <h2>Build and fit a classifier<a class="headerlink" href="#build-and-fit-a-class
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/opt/hostedtoolcache/Python/3.8.12/x64/lib/python3.8/site-packages/smac/intensification/parallel_scheduling.py:152: UserWarning: SuccessiveHalving is intended to be used with more than 1 worker but num_workers=1
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:68: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+  self.metafeatures = self.metafeatures.append(metafeatures)
+/home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:72: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+  self.algorithm_runs[metric].append(runs)
+/opt/hostedtoolcache/Python/3.8.12/x64/lib/python3.8/site-packages/smac/intensification/parallel_scheduling.py:152: UserWarning: SuccessiveHalving is intended to be used with more than 1 worker but num_workers=1
+  warnings.warn(&quot;{} is intended to be used &quot;
+/opt/hostedtoolcache/Python/3.8.12/x64/lib/python3.8/site-packages/smac/intensification/parallel_scheduling.py:152: UserWarning: SuccessiveHalving is intended to be used with more than 1 worker but num_workers=1
   warnings.warn(&quot;{} is intended to be used &quot;
-[(0.120000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;sgd&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:sgd:alpha&#39;: 9.410144741041167e-05, &#39;classifier:sgd:average&#39;: &#39;True&#39;, &#39;classifier:sgd:fit_intercept&#39;: &#39;True&#39;, &#39;classifier:sgd:learning_rate&#39;: &#39;constant&#39;, &#39;classifier:sgd:loss&#39;: &#39;log&#39;, &#39;classifier:sgd:penalty&#39;: &#39;l1&#39;, &#39;classifier:sgd:tol&#39;: 0.05082904256838701, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;normalize&#39;, &#39;classifier:sgd:eta0&#39;: 0.0018055343233337954},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.100000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;sgd&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:sgd:alpha&#39;: 0.0002346515712987664, &#39;classifier:sgd:average&#39;: &#39;True&#39;, &#39;classifier:sgd:fit_intercept&#39;: &#39;True&#39;, &#39;classifier:sgd:learning_rate&#39;: &#39;optimal&#39;, &#39;classifier:sgd:loss&#39;: &#39;log&#39;, &#39;classifier:sgd:penalty&#39;: &#39;l1&#39;, &#39;classifier:sgd:tol&#39;: 1.3716748930467322e-05, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;standardize&#39;},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.080000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;gradient_boosting&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:gradient_boosting:early_stop&#39;: &#39;off&#39;, &#39;classifier:gradient_boosting:l2_regularization&#39;: 1.0945814167023392e-10, &#39;classifier:gradient_boosting:learning_rate&#39;: 0.11042628136263043, &#39;classifier:gradient_boosting:loss&#39;: &#39;auto&#39;, &#39;classifier:gradient_boosting:max_bins&#39;: 255, &#39;classifier:gradient_boosting:max_depth&#39;: &#39;None&#39;, &#39;classifier:gradient_boosting:max_leaf_nodes&#39;: 30, &#39;classifier:gradient_boosting:min_samples_leaf&#39;: 22, &#39;classifier:gradient_boosting:scoring&#39;: &#39;loss&#39;, &#39;classifier:gradient_boosting:tol&#39;: 1e-07, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;standardize&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.05141281638752715},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.080000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;random_forest&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:random_forest:bootstrap&#39;: &#39;True&#39;, &#39;classifier:random_forest:criterion&#39;: &#39;entropy&#39;, &#39;classifier:random_forest:max_depth&#39;: &#39;None&#39;, &#39;classifier:random_forest:max_features&#39;: 0.926283631486858, &#39;classifier:random_forest:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:random_forest:min_impurity_decrease&#39;: 0.0, &#39;classifier:random_forest:min_samples_leaf&#39;: 7, &#39;classifier:random_forest:min_samples_split&#39;: 2, &#39;classifier:random_forest:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;standardize&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.16265262021972576},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.060000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;random_forest&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:random_forest:bootstrap&#39;: &#39;False&#39;, &#39;classifier:random_forest:criterion&#39;: &#39;entropy&#39;, &#39;classifier:random_forest:max_depth&#39;: &#39;None&#39;, &#39;classifier:random_forest:max_features&#39;: 0.41808321658160696, &#39;classifier:random_forest:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:random_forest:min_impurity_decrease&#39;: 0.0, &#39;classifier:random_forest:min_samples_leaf&#39;: 1, &#39;classifier:random_forest:min_samples_split&#39;: 4, &#39;classifier:random_forest:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;median&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;standardize&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.011283688651384545},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.060000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;extra_trees&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:extra_trees:bootstrap&#39;: &#39;False&#39;, &#39;classifier:extra_trees:criterion&#39;: &#39;entropy&#39;, &#39;classifier:extra_trees:max_depth&#39;: &#39;None&#39;, &#39;classifier:extra_trees:max_features&#39;: 0.6128603428070196, &#39;classifier:extra_trees:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:extra_trees:min_impurity_decrease&#39;: 0.0, &#39;classifier:extra_trees:min_samples_leaf&#39;: 1, &#39;classifier:extra_trees:min_samples_split&#39;: 3, &#39;classifier:extra_trees:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;robust_scaler&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.08125391652261632, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max&#39;: 0.8601586365248128, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min&#39;: 0.052862156055921525},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.060000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;random_forest&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:random_forest:bootstrap&#39;: &#39;False&#39;, &#39;classifier:random_forest:criterion&#39;: &#39;entropy&#39;, &#39;classifier:random_forest:max_depth&#39;: &#39;None&#39;, &#39;classifier:random_forest:max_features&#39;: 0.41808321658160696, &#39;classifier:random_forest:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:random_forest:min_impurity_decrease&#39;: 0.0, &#39;classifier:random_forest:min_samples_leaf&#39;: 1, &#39;classifier:random_forest:min_samples_split&#39;: 4, &#39;classifier:random_forest:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;median&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;standardize&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.011283688651384545},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.060000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;gradient_boosting&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:gradient_boosting:early_stop&#39;: &#39;off&#39;, &#39;classifier:gradient_boosting:l2_regularization&#39;: 1e-10, &#39;classifier:gradient_boosting:learning_rate&#39;: 0.16262682406125173, &#39;classifier:gradient_boosting:loss&#39;: &#39;auto&#39;, &#39;classifier:gradient_boosting:max_bins&#39;: 255, &#39;classifier:gradient_boosting:max_depth&#39;: &#39;None&#39;, &#39;classifier:gradient_boosting:max_leaf_nodes&#39;: 66, &#39;classifier:gradient_boosting:min_samples_leaf&#39;: 20, &#39;classifier:gradient_boosting:scoring&#39;: &#39;loss&#39;, &#39;classifier:gradient_boosting:tol&#39;: 1e-07, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;robust_scaler&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.005428587241449129, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max&#39;: 0.75, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min&#39;: 0.23746960178084334},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.060000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;extra_trees&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:extra_trees:bootstrap&#39;: &#39;True&#39;, &#39;classifier:extra_trees:criterion&#39;: &#39;gini&#39;, &#39;classifier:extra_trees:max_depth&#39;: &#39;None&#39;, &#39;classifier:extra_trees:max_features&#39;: 0.37705188916038523, &#39;classifier:extra_trees:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:extra_trees:min_impurity_decrease&#39;: 0.0, &#39;classifier:extra_trees:min_samples_leaf&#39;: 2, &#39;classifier:extra_trees:min_samples_split&#39;: 3, &#39;classifier:extra_trees:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;quantile_transformer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles&#39;: 521, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution&#39;: &#39;uniform&#39;},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.060000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;random_forest&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:random_forest:bootstrap&#39;: &#39;True&#39;, &#39;classifier:random_forest:criterion&#39;: &#39;entropy&#39;, &#39;classifier:random_forest:max_depth&#39;: &#39;None&#39;, &#39;classifier:random_forest:max_features&#39;: 0.09884140378258977, &#39;classifier:random_forest:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:random_forest:min_impurity_decrease&#39;: 0.0, &#39;classifier:random_forest:min_samples_leaf&#39;: 6, &#39;classifier:random_forest:min_samples_split&#39;: 13, &#39;classifier:random_forest:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;standardize&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.022834561782884507},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.040000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;gradient_boosting&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:gradient_boosting:early_stop&#39;: &#39;off&#39;, &#39;classifier:gradient_boosting:l2_regularization&#39;: 1.0945814167023392e-10, &#39;classifier:gradient_boosting:learning_rate&#39;: 0.11042628136263043, &#39;classifier:gradient_boosting:loss&#39;: &#39;auto&#39;, &#39;classifier:gradient_boosting:max_bins&#39;: 255, &#39;classifier:gradient_boosting:max_depth&#39;: &#39;None&#39;, &#39;classifier:gradient_boosting:max_leaf_nodes&#39;: 30, &#39;classifier:gradient_boosting:min_samples_leaf&#39;: 22, &#39;classifier:gradient_boosting:scoring&#39;: &#39;loss&#39;, &#39;classifier:gradient_boosting:tol&#39;: 1e-07, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;standardize&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.05141281638752715},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.040000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;random_forest&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:random_forest:bootstrap&#39;: &#39;False&#39;, &#39;classifier:random_forest:criterion&#39;: &#39;entropy&#39;, &#39;classifier:random_forest:max_depth&#39;: &#39;None&#39;, &#39;classifier:random_forest:max_features&#39;: 0.7708651296941559, &#39;classifier:random_forest:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:random_forest:min_impurity_decrease&#39;: 0.0, &#39;classifier:random_forest:min_samples_leaf&#39;: 13, &#39;classifier:random_forest:min_samples_split&#39;: 2, &#39;classifier:random_forest:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;quantile_transformer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles&#39;: 931, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution&#39;: &#39;normal&#39;},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.040000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;random_forest&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:random_forest:bootstrap&#39;: &#39;False&#39;, &#39;classifier:random_forest:criterion&#39;: &#39;entropy&#39;, &#39;classifier:random_forest:max_depth&#39;: &#39;None&#39;, &#39;classifier:random_forest:max_features&#39;: 0.7323115919225983, &#39;classifier:random_forest:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:random_forest:min_impurity_decrease&#39;: 0.0, &#39;classifier:random_forest:min_samples_leaf&#39;: 15, &#39;classifier:random_forest:min_samples_split&#39;: 6, &#39;classifier:random_forest:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;robust_scaler&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.011901034843417571, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max&#39;: 0.7818500358383581, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min&#39;: 0.20068746139723115},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.040000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;random_forest&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:random_forest:bootstrap&#39;: &#39;True&#39;, &#39;classifier:random_forest:criterion&#39;: &#39;entropy&#39;, &#39;classifier:random_forest:max_depth&#39;: &#39;None&#39;, &#39;classifier:random_forest:max_features&#39;: 0.6779841015398226, &#39;classifier:random_forest:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:random_forest:min_impurity_decrease&#39;: 0.0, &#39;classifier:random_forest:min_samples_leaf&#39;: 14, &#39;classifier:random_forest:min_samples_split&#39;: 14, &#39;classifier:random_forest:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;robust_scaler&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.03961232028373377, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max&#39;: 0.9772109830437746, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min&#39;: 0.13300503334706695},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.020000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;random_forest&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:random_forest:bootstrap&#39;: &#39;True&#39;, &#39;classifier:random_forest:criterion&#39;: &#39;gini&#39;, &#39;classifier:random_forest:max_depth&#39;: &#39;None&#39;, &#39;classifier:random_forest:max_features&#39;: 0.5, &#39;classifier:random_forest:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:random_forest:min_impurity_decrease&#39;: 0.0, &#39;classifier:random_forest:min_samples_leaf&#39;: 1, &#39;classifier:random_forest:min_samples_split&#39;: 2, &#39;classifier:random_forest:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;standardize&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.01},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.020000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;random_forest&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:random_forest:bootstrap&#39;: &#39;False&#39;, &#39;classifier:random_forest:criterion&#39;: &#39;entropy&#39;, &#39;classifier:random_forest:max_depth&#39;: &#39;None&#39;, &#39;classifier:random_forest:max_features&#39;: 0.41808321658160696, &#39;classifier:random_forest:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:random_forest:min_impurity_decrease&#39;: 0.0, &#39;classifier:random_forest:min_samples_leaf&#39;: 1, &#39;classifier:random_forest:min_samples_split&#39;: 4, &#39;classifier:random_forest:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;median&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;standardize&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.011283688651384545},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.020000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;random_forest&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:random_forest:bootstrap&#39;: &#39;False&#39;, &#39;classifier:random_forest:criterion&#39;: &#39;entropy&#39;, &#39;classifier:random_forest:max_depth&#39;: &#39;None&#39;, &#39;classifier:random_forest:max_features&#39;: 0.41808321658160696, &#39;classifier:random_forest:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:random_forest:min_impurity_decrease&#39;: 0.0, &#39;classifier:random_forest:min_samples_leaf&#39;: 1, &#39;classifier:random_forest:min_samples_split&#39;: 4, &#39;classifier:random_forest:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;median&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;standardize&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.011283688651384545},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.020000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;gradient_boosting&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:gradient_boosting:early_stop&#39;: &#39;train&#39;, &#39;classifier:gradient_boosting:l2_regularization&#39;: 3.609412172481434e-10, &#39;classifier:gradient_boosting:learning_rate&#39;: 0.05972079854295879, &#39;classifier:gradient_boosting:loss&#39;: &#39;auto&#39;, &#39;classifier:gradient_boosting:max_bins&#39;: 255, &#39;classifier:gradient_boosting:max_depth&#39;: &#39;None&#39;, &#39;classifier:gradient_boosting:max_leaf_nodes&#39;: 4, &#39;classifier:gradient_boosting:min_samples_leaf&#39;: 2, &#39;classifier:gradient_boosting:scoring&#39;: &#39;loss&#39;, &#39;classifier:gradient_boosting:tol&#39;: 1e-07, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;minmax&#39;, &#39;classifier:gradient_boosting:n_iter_no_change&#39;: 14},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.020000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;gradient_boosting&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:gradient_boosting:early_stop&#39;: &#39;off&#39;, &#39;classifier:gradient_boosting:l2_regularization&#39;: 1.0945814167023392e-10, &#39;classifier:gradient_boosting:learning_rate&#39;: 0.11042628136263043, &#39;classifier:gradient_boosting:loss&#39;: &#39;auto&#39;, &#39;classifier:gradient_boosting:max_bins&#39;: 255, &#39;classifier:gradient_boosting:max_depth&#39;: &#39;None&#39;, &#39;classifier:gradient_boosting:max_leaf_nodes&#39;: 30, &#39;classifier:gradient_boosting:min_samples_leaf&#39;: 22, &#39;classifier:gradient_boosting:scoring&#39;: &#39;loss&#39;, &#39;classifier:gradient_boosting:tol&#39;: 1e-07, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;standardize&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.05141281638752715},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-]
+{   2: {   &#39;balancing&#39;: Balancing(random_state=1),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc784bf1d00&gt;,
+           &#39;cost&#39;: 0.021276595744680882,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78febd310&gt;,
+           &#39;ensemble_weight&#39;: 0.02,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc784bf19d0&gt;,
+           &#39;model_id&#39;: 2,
+           &#39;rank&#39;: 5,
+           &#39;sklearn_classifier&#39;: RandomForestClassifier(max_features=5, n_estimators=64, n_jobs=1,
+                       random_state=1, warm_start=True)},
+    3: {   &#39;balancing&#39;: Balancing(random_state=1),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78c43e580&gt;,
+           &#39;cost&#39;: 0.028368794326241176,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc786095a90&gt;,
+           &#39;ensemble_weight&#39;: 0.04,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78a9ea4f0&gt;,
+           &#39;model_id&#39;: 3,
+           &#39;rank&#39;: 9,
+           &#39;sklearn_classifier&#39;: RandomForestClassifier(criterion=&#39;entropy&#39;, max_features=23, min_samples_leaf=7,
+                       n_estimators=64, n_jobs=1, random_state=1,
+                       warm_start=True)},
+    4: {   &#39;balancing&#39;: Balancing(random_state=1),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78c057850&gt;,
+           &#39;cost&#39;: 0.07801418439716312,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7869195e0&gt;,
+           &#39;ensemble_weight&#39;: 0.08,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c057550&gt;,
+           &#39;model_id&#39;: 4,
+           &#39;rank&#39;: 21,
+           &#39;sklearn_classifier&#39;: PassiveAggressiveClassifier(C=0.14268277711454813, max_iter=128, random_state=1,
+                            tol=0.0002600768160857831, warm_start=True)},
+    5: {   &#39;balancing&#39;: Balancing(random_state=1),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc7871ffbe0&gt;,
+           &#39;cost&#39;: 0.028368794326241176,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78b915d00&gt;,
+           &#39;ensemble_weight&#39;: 0.04,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc7871ff9a0&gt;,
+           &#39;model_id&#39;: 5,
+           &#39;rank&#39;: 10,
+           &#39;sklearn_classifier&#39;: HistGradientBoostingClassifier(early_stopping=False, l2_regularization=1e-10,
+                               learning_rate=0.16262682406125173, max_iter=64,
+                               max_leaf_nodes=66, n_iter_no_change=0,
+                               random_state=1, validation_fraction=None,
+                               warm_start=True)},
+    7: {   &#39;balancing&#39;: Balancing(random_state=1),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc784d70430&gt;,
+           &#39;cost&#39;: 0.03546099290780147,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7867e4640&gt;,
+           &#39;ensemble_weight&#39;: 0.04,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc784d702b0&gt;,
+           &#39;model_id&#39;: 7,
+           &#39;rank&#39;: 13,
+           &#39;sklearn_classifier&#39;: SGDClassifier(alpha=0.0002346515712987664, average=True, eta0=0.01, loss=&#39;log&#39;,
+              max_iter=128, penalty=&#39;l1&#39;, random_state=1,
+              tol=1.3716748930467322e-05, warm_start=True)},
+    8: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78509fd30&gt;,
+           &#39;cost&#39;: 0.014184397163120588,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc790682160&gt;,
+           &#39;ensemble_weight&#39;: 0.06,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78509f910&gt;,
+           &#39;model_id&#39;: 8,
+           &#39;rank&#39;: 1,
+           &#39;sklearn_classifier&#39;: RandomForestClassifier(bootstrap=False, criterion=&#39;entropy&#39;, max_features=4,
+                       min_samples_split=4, n_estimators=64, n_jobs=1,
+                       random_state=1, warm_start=True)},
+    9: {   &#39;balancing&#39;: Balancing(random_state=1),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc786abba00&gt;,
+           &#39;cost&#39;: 0.021276595744680882,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78a86deb0&gt;,
+           &#39;ensemble_weight&#39;: 0.04,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc786abb100&gt;,
+           &#39;model_id&#39;: 9,
+           &#39;rank&#39;: 7,
+           &#39;sklearn_classifier&#39;: ExtraTreesClassifier(criterion=&#39;entropy&#39;, max_features=8, min_samples_split=3,
+                     n_estimators=512, n_jobs=1, random_state=1,
+                     warm_start=True)},
+    10: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78582e940&gt;,
+            &#39;cost&#39;: 0.028368794326241176,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78b656a90&gt;,
+            &#39;ensemble_weight&#39;: 0.04,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78582e3d0&gt;,
+            &#39;model_id&#39;: 10,
+            &#39;rank&#39;: 11,
+            &#39;sklearn_classifier&#39;: HistGradientBoostingClassifier(early_stopping=True,
+                               l2_regularization=5.027708640006448e-08,
+                               learning_rate=0.09750328007832798, max_iter=64,
+                               max_leaf_nodes=1234, min_samples_leaf=25,
+                               n_iter_no_change=1, random_state=1,
+                               validation_fraction=0.08300813783286698,
+                               warm_start=True)},
+    11: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78fc1c340&gt;,
+            &#39;cost&#39;: 0.014184397163120588,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc787888970&gt;,
+            &#39;ensemble_weight&#39;: 0.0,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78fc1c760&gt;,
+            &#39;model_id&#39;: 11,
+            &#39;rank&#39;: 8,
+            &#39;sklearn_classifier&#39;: HistGradientBoostingClassifier(early_stopping=False,
+                               l2_regularization=1.0945814167023392e-10,
+                               learning_rate=0.11042628136263043, max_iter=64,
+                               max_leaf_nodes=30, min_samples_leaf=22,
+                               n_iter_no_change=0, random_state=1,
+                               validation_fraction=None, warm_start=True)},
+    12: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc7878b19d0&gt;,
+            &#39;cost&#39;: 0.04255319148936165,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78a7d53a0&gt;,
+            &#39;ensemble_weight&#39;: 0.1,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc7878b1850&gt;,
+            &#39;model_id&#39;: 12,
+            &#39;rank&#39;: 16,
+            &#39;sklearn_classifier&#39;: RandomForestClassifier(criterion=&#39;entropy&#39;, max_features=1, min_samples_leaf=6,
+                       min_samples_split=13, n_estimators=64, n_jobs=1,
+                       random_state=1, warm_start=True)},
+    13: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc787a63dc0&gt;,
+            &#39;cost&#39;: 0.03546099290780147,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc785869040&gt;,
+            &#39;ensemble_weight&#39;: 0.02,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc787a63bb0&gt;,
+            &#39;model_id&#39;: 13,
+            &#39;rank&#39;: 14,
+            &#39;sklearn_classifier&#39;: HistGradientBoostingClassifier(early_stopping=True,
+                               l2_regularization=2.506856350040198e-06,
+                               learning_rate=0.04634380160611007, max_iter=64,
+                               max_leaf_nodes=11, min_samples_leaf=41,
+                               n_iter_no_change=17, random_state=1,
+                               validation_fraction=None, warm_start=True)},
+    14: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc7871e5400&gt;,
+            &#39;cost&#39;: 0.03546099290780147,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78b418430&gt;,
+            &#39;ensemble_weight&#39;: 0.04,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc7871e5280&gt;,
+            &#39;model_id&#39;: 14,
+            &#39;rank&#39;: 15,
+            &#39;sklearn_classifier&#39;: ExtraTreesClassifier(bootstrap=True, max_features=3, min_samples_leaf=2,
+                     min_samples_split=3, n_estimators=64, n_jobs=1,
+                     random_state=1, warm_start=True)},
+    16: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78a9cc160&gt;,
+            &#39;cost&#39;: 0.049645390070921946,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7871ccb50&gt;,
+            &#39;ensemble_weight&#39;: 0.06,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc7869b8e80&gt;,
+            &#39;model_id&#39;: 16,
+            &#39;rank&#39;: 18,
+            &#39;sklearn_classifier&#39;: RandomForestClassifier(bootstrap=False, criterion=&#39;entropy&#39;, max_features=12,
+                       min_samples_leaf=15, min_samples_split=6,
+                       n_estimators=64, n_jobs=1, random_state=1,
+                       warm_start=True)},
+    17: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78a370be0&gt;,
+            &#39;cost&#39;: 0.099290780141844,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78c0e9b50&gt;,
+            &#39;ensemble_weight&#39;: 0.06,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78a3702e0&gt;,
+            &#39;model_id&#39;: 17,
+            &#39;rank&#39;: 22,
+            &#39;sklearn_classifier&#39;: SGDClassifier(alpha=9.410144741041167e-05, average=True,
+              eta0=0.0018055343233337954, learning_rate=&#39;constant&#39;, loss=&#39;log&#39;,
+              max_iter=128, penalty=&#39;l1&#39;, random_state=1,
+              tol=0.05082904256838701, warm_start=True)},
+    18: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc7845473a0&gt;,
+            &#39;cost&#39;: 0.05673758865248224,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7864a4700&gt;,
+            &#39;ensemble_weight&#39;: 0.06,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc784547220&gt;,
+            &#39;model_id&#39;: 18,
+            &#39;rank&#39;: 20,
+            &#39;sklearn_classifier&#39;: RandomForestClassifier(criterion=&#39;entropy&#39;, max_features=10,
+                       min_samples_leaf=14, min_samples_split=14,
+                       n_estimators=64, n_jobs=1, random_state=1,
+                       warm_start=True)},
+    19: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc7848717c0&gt;,
+            &#39;cost&#39;: 0.028368794326241176,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78b498c70&gt;,
+            &#39;ensemble_weight&#39;: 0.04,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78af79730&gt;,
+            &#39;model_id&#39;: 19,
+            &#39;rank&#39;: 12,
+            &#39;sklearn_classifier&#39;: RandomForestClassifier(bootstrap=False, criterion=&#39;entropy&#39;, max_features=13,
+                       min_samples_leaf=13, n_estimators=64, n_jobs=1,
+                       random_state=1, warm_start=True)},
+    20: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc7868a13a0&gt;,
+            &#39;cost&#39;: 0.049645390070921946,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78758e4c0&gt;,
+            &#39;ensemble_weight&#39;: 0.02,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc7868a1220&gt;,
+            &#39;model_id&#39;: 20,
+            &#39;rank&#39;: 19,
+            &#39;sklearn_classifier&#39;: HistGradientBoostingClassifier(early_stopping=False,
+                               l2_regularization=5.2497721880674565e-06,
+                               learning_rate=0.03162215674470446, max_iter=64,
+                               max_leaf_nodes=74, min_samples_leaf=1,
+                               n_iter_no_change=0, random_state=1,
+                               validation_fraction=None, warm_start=True)},
+    21: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc7864b81f0&gt;,
+            &#39;cost&#39;: 0.04255319148936165,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78bad4130&gt;,
+            &#39;ensemble_weight&#39;: 0.02,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc7864b8130&gt;,
+            &#39;model_id&#39;: 21,
+            &#39;rank&#39;: 17,
+            &#39;sklearn_classifier&#39;: PassiveAggressiveClassifier(C=0.05491437530740365, average=True,
+                            loss=&#39;squared_hinge&#39;, max_iter=128, random_state=1,
+                            tol=0.003640536588885903, warm_start=True)}}
 auto-sklearn results:
   Dataset name: breast_cancer
   Metric: accuracy
   Best validation score: 0.985816
-  Number of target algorithm runs: 24
-  Number of successful target algorithm runs: 24
+  Number of target algorithm runs: 26
+  Number of successful target algorithm runs: 26
   Number of crashed target algorithm runs: 0
   Number of target algorithms that exceeded the time limit: 0
   Number of target algorithms that exceeded the memory limit: 0
 
-Accuracy score 0.9440559440559441
+Accuracy score 0.951048951048951
 </pre></div>
 </div>
-</section>
-<section id="we-can-also-use-cross-validation-with-successive-halving">
+</div>
+<div class="section" id="we-can-also-use-cross-validation-with-successive-halving">
 <h2>We can also use cross-validation with successive halving<a class="headerlink" href="#we-can-also-use-cross-validation-with-successive-halving" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">X</span><span class="p">,</span> <span class="n">y</span> <span class="o">=</span> <span class="n">sklearn</span><span class="o">.</span><span class="n">datasets</span><span class="o">.</span><span class="n">load_breast_cancer</span><span class="p">(</span><span class="n">return_X_y</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
 <span class="n">X_train</span><span class="p">,</span> <span class="n">X_test</span><span class="p">,</span> <span class="n">y_train</span><span class="p">,</span> <span class="n">y_test</span> <span class="o">=</span> \
@@ -431,7 +499,7 @@ <h2>We can also use cross-validation with successive halving<a class="headerlink
 <span class="n">automl</span><span class="o">.</span><span class="n">fit</span><span class="p">(</span><span class="n">X_train</span><span class="p">,</span> <span class="n">y_train</span><span class="p">,</span> <span class="n">dataset_name</span><span class="o">=</span><span class="s1">&#39;breast_cancer&#39;</span><span class="p">)</span>
 
 <span class="c1"># Print the final ensemble constructed by auto-sklearn.</span>
-<span class="nb">print</span><span class="p">(</span><span class="n">automl</span><span class="o">.</span><span class="n">show_models</span><span class="p">())</span>
+<span class="n">pprint</span><span class="p">(</span><span class="n">automl</span><span class="o">.</span><span class="n">show_models</span><span class="p">(),</span> <span class="n">indent</span><span class="o">=</span><span class="mi">4</span><span class="p">)</span>
 <span class="n">automl</span><span class="o">.</span><span class="n">refit</span><span class="p">(</span><span class="n">X_train</span><span class="p">,</span> <span class="n">y_train</span><span class="p">)</span>
 <span class="n">predictions</span> <span class="o">=</span> <span class="n">automl</span><span class="o">.</span><span class="n">predict</span><span class="p">(</span><span class="n">X_test</span><span class="p">)</span>
 <span class="c1"># Print statistics about the auto-sklearn run such as number of</span>
@@ -441,80 +509,362 @@ <h2>We can also use cross-validation with successive halving<a class="headerlink
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/opt/hostedtoolcache/Python/3.8.12/x64/lib/python3.8/site-packages/smac/intensification/parallel_scheduling.py:152: UserWarning: SuccessiveHalving is intended to be used with more than 1 worker but num_workers=1
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:68: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+  self.metafeatures = self.metafeatures.append(metafeatures)
+/home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:72: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+  self.algorithm_runs[metric].append(runs)
+/opt/hostedtoolcache/Python/3.8.12/x64/lib/python3.8/site-packages/smac/intensification/parallel_scheduling.py:152: UserWarning: SuccessiveHalving is intended to be used with more than 1 worker but num_workers=1
   warnings.warn(&quot;{} is intended to be used &quot;
-[(0.280000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;passive_aggressive&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:passive_aggressive:C&#39;: 0.14268277711454813, &#39;classifier:passive_aggressive:average&#39;: &#39;False&#39;, &#39;classifier:passive_aggressive:fit_intercept&#39;: &#39;True&#39;, &#39;classifier:passive_aggressive:loss&#39;: &#39;hinge&#39;, &#39;classifier:passive_aggressive:tol&#39;: 0.0002600768160857831, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;none&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.0070580904199417415},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.180000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;extra_trees&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:extra_trees:bootstrap&#39;: &#39;False&#39;, &#39;classifier:extra_trees:criterion&#39;: &#39;entropy&#39;, &#39;classifier:extra_trees:max_depth&#39;: &#39;None&#39;, &#39;classifier:extra_trees:max_features&#39;: 0.6128603428070196, &#39;classifier:extra_trees:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:extra_trees:min_impurity_decrease&#39;: 0.0, &#39;classifier:extra_trees:min_samples_leaf&#39;: 1, &#39;classifier:extra_trees:min_samples_split&#39;: 3, &#39;classifier:extra_trees:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;robust_scaler&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.08125391652261632, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max&#39;: 0.8601586365248128, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min&#39;: 0.052862156055921525},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.180000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;extra_trees&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:extra_trees:bootstrap&#39;: &#39;False&#39;, &#39;classifier:extra_trees:criterion&#39;: &#39;entropy&#39;, &#39;classifier:extra_trees:max_depth&#39;: &#39;None&#39;, &#39;classifier:extra_trees:max_features&#39;: 0.6128603428070196, &#39;classifier:extra_trees:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:extra_trees:min_impurity_decrease&#39;: 0.0, &#39;classifier:extra_trees:min_samples_leaf&#39;: 1, &#39;classifier:extra_trees:min_samples_split&#39;: 3, &#39;classifier:extra_trees:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;robust_scaler&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.08125391652261632, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max&#39;: 0.8601586365248128, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min&#39;: 0.052862156055921525},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.160000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;extra_trees&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:extra_trees:bootstrap&#39;: &#39;False&#39;, &#39;classifier:extra_trees:criterion&#39;: &#39;entropy&#39;, &#39;classifier:extra_trees:max_depth&#39;: &#39;None&#39;, &#39;classifier:extra_trees:max_features&#39;: 0.6128603428070196, &#39;classifier:extra_trees:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:extra_trees:min_impurity_decrease&#39;: 0.0, &#39;classifier:extra_trees:min_samples_leaf&#39;: 1, &#39;classifier:extra_trees:min_samples_split&#39;: 3, &#39;classifier:extra_trees:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;robust_scaler&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.08125391652261632, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max&#39;: 0.8601586365248128, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min&#39;: 0.052862156055921525},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.100000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;sgd&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:sgd:alpha&#39;: 0.0002346515712987664, &#39;classifier:sgd:average&#39;: &#39;True&#39;, &#39;classifier:sgd:fit_intercept&#39;: &#39;True&#39;, &#39;classifier:sgd:learning_rate&#39;: &#39;optimal&#39;, &#39;classifier:sgd:loss&#39;: &#39;log&#39;, &#39;classifier:sgd:penalty&#39;: &#39;l1&#39;, &#39;classifier:sgd:tol&#39;: 1.3716748930467322e-05, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;standardize&#39;},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.080000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;random_forest&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:random_forest:bootstrap&#39;: &#39;True&#39;, &#39;classifier:random_forest:criterion&#39;: &#39;gini&#39;, &#39;classifier:random_forest:max_depth&#39;: &#39;None&#39;, &#39;classifier:random_forest:max_features&#39;: 0.5, &#39;classifier:random_forest:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:random_forest:min_impurity_decrease&#39;: 0.0, &#39;classifier:random_forest:min_samples_leaf&#39;: 1, &#39;classifier:random_forest:min_samples_split&#39;: 2, &#39;classifier:random_forest:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;standardize&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.01},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.020000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;random_forest&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:random_forest:bootstrap&#39;: &#39;True&#39;, &#39;classifier:random_forest:criterion&#39;: &#39;entropy&#39;, &#39;classifier:random_forest:max_depth&#39;: &#39;None&#39;, &#39;classifier:random_forest:max_features&#39;: 0.926283631486858, &#39;classifier:random_forest:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:random_forest:min_impurity_decrease&#39;: 0.0, &#39;classifier:random_forest:min_samples_leaf&#39;: 7, &#39;classifier:random_forest:min_samples_split&#39;: 2, &#39;classifier:random_forest:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;standardize&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.16265262021972576},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-]
+{   2: {   &#39;cost&#39;: 0.046948356807511755,
+           &#39;ensemble_weight&#39;: 0.14,
+           &#39;estimators&#39;: [   {   &#39;balancing&#39;: Balancing(random_state=1),
+                                 &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78baa93d0&gt;,
+                                 &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc786564c10&gt;,
+                                 &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc7878b0160&gt;,
+                                 &#39;sklearn_classifier&#39;: RandomForestClassifier(max_features=5, n_estimators=64, n_jobs=1,
+                       random_state=1, warm_start=True)},
+                             {   &#39;balancing&#39;: Balancing(random_state=1),
+                                 &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc787c7a3d0&gt;,
+                                 &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc786edd8b0&gt;,
+                                 &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc787c7a280&gt;,
+                                 &#39;sklearn_classifier&#39;: RandomForestClassifier(max_features=5, n_estimators=64, n_jobs=1,
+                       random_state=1, warm_start=True)},
+                             {   &#39;balancing&#39;: Balancing(random_state=1),
+                                 &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78753bcd0&gt;,
+                                 &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78715d9d0&gt;,
+                                 &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78753bb20&gt;,
+                                 &#39;sklearn_classifier&#39;: RandomForestClassifier(max_features=5, n_estimators=64, n_jobs=1,
+                       random_state=1, warm_start=True)},
+                             {   &#39;balancing&#39;: Balancing(random_state=1),
+                                 &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78beb7c40&gt;,
+                                 &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc785684b50&gt;,
+                                 &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78beb7b50&gt;,
+                                 &#39;sklearn_classifier&#39;: RandomForestClassifier(max_features=5, n_estimators=64, n_jobs=1,
+                       random_state=1, warm_start=True)},
+                             {   &#39;balancing&#39;: Balancing(random_state=1),
+                                 &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc7872bce20&gt;,
+                                 &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78728de20&gt;,
+                                 &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc7872bcd30&gt;,
+                                 &#39;sklearn_classifier&#39;: RandomForestClassifier(max_features=5, n_estimators=64, n_jobs=1,
+                       random_state=1, warm_start=True)}],
+           &#39;model_id&#39;: 2,
+           &#39;rank&#39;: 8,
+           &#39;voting_model&#39;: VotingClassifier(estimators=None, voting=&#39;soft&#39;)},
+    3: {   &#39;cost&#39;: 0.05164319248826292,
+           &#39;ensemble_weight&#39;: 0.04,
+           &#39;estimators&#39;: [   {   &#39;balancing&#39;: Balancing(random_state=1),
+                                 &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc787c79550&gt;,
+                                 &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7878b5580&gt;,
+                                 &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc786ed8610&gt;,
+                                 &#39;sklearn_classifier&#39;: RandomForestClassifier(criterion=&#39;entropy&#39;, max_features=23, min_samples_leaf=7,
+                       n_estimators=64, n_jobs=1, random_state=1,
+                       warm_start=True)},
+                             {   &#39;balancing&#39;: Balancing(random_state=1),
+                                 &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78c5a0e80&gt;,
+                                 &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78c0a7940&gt;,
+                                 &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c5a0cd0&gt;,
+                                 &#39;sklearn_classifier&#39;: RandomForestClassifier(criterion=&#39;entropy&#39;, max_features=23, min_samples_leaf=7,
+                       n_estimators=64, n_jobs=1, random_state=1,
+                       warm_start=True)},
+                             {   &#39;balancing&#39;: Balancing(random_state=1),
+                                 &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc7876f1ac0&gt;,
+                                 &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7876d8b80&gt;,
+                                 &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc7876f1be0&gt;,
+                                 &#39;sklearn_classifier&#39;: RandomForestClassifier(criterion=&#39;entropy&#39;, max_features=23, min_samples_leaf=7,
+                       n_estimators=64, n_jobs=1, random_state=1,
+                       warm_start=True)},
+                             {   &#39;balancing&#39;: Balancing(random_state=1),
+                                 &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc787711040&gt;,
+                                 &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc787742160&gt;,
+                                 &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc7877111c0&gt;,
+                                 &#39;sklearn_classifier&#39;: RandomForestClassifier(criterion=&#39;entropy&#39;, max_features=23, min_samples_leaf=7,
+                       n_estimators=64, n_jobs=1, random_state=1,
+                       warm_start=True)},
+                             {   &#39;balancing&#39;: Balancing(random_state=1),
+                                 &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78662fb20&gt;,
+                                 &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc786633bb0&gt;,
+                                 &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78662fc40&gt;,
+                                 &#39;sklearn_classifier&#39;: RandomForestClassifier(criterion=&#39;entropy&#39;, max_features=23, min_samples_leaf=7,
+                       n_estimators=64, n_jobs=1, random_state=1,
+                       warm_start=True)}],
+           &#39;model_id&#39;: 3,
+           &#39;rank&#39;: 10,
+           &#39;voting_model&#39;: VotingClassifier(estimators=None, voting=&#39;soft&#39;)},
+    4: {   &#39;cost&#39;: 0.08215962441314555,
+           &#39;ensemble_weight&#39;: 0.2,
+           &#39;estimators&#39;: [   {   &#39;balancing&#39;: Balancing(random_state=1),
+                                 &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc787157910&gt;,
+                                 &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7877de790&gt;,
+                                 &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78714dfd0&gt;,
+                                 &#39;sklearn_classifier&#39;: PassiveAggressiveClassifier(C=0.14268277711454813, max_iter=128, random_state=1,
+                            tol=0.0002600768160857831, warm_start=True)},
+                             {   &#39;balancing&#39;: Balancing(random_state=1),
+                                 &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc7876d02b0&gt;,
+                                 &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78c4f2e80&gt;,
+                                 &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc7876d0310&gt;,
+                                 &#39;sklearn_classifier&#39;: PassiveAggressiveClassifier(C=0.14268277711454813, max_iter=128, random_state=1,
+                            tol=0.0002600768160857831, warm_start=True)},
+                             {   &#39;balancing&#39;: Balancing(random_state=1),
+                                 &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78c53cb80&gt;,
+                                 &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc787711670&gt;,
+                                 &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c53c5e0&gt;,
+                                 &#39;sklearn_classifier&#39;: PassiveAggressiveClassifier(C=0.14268277711454813, max_iter=128, random_state=1,
+                            tol=0.0002600768160857831, warm_start=True)},
+                             {   &#39;balancing&#39;: Balancing(random_state=1),
+                                 &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc7843c0100&gt;,
+                                 &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7843a12e0&gt;,
+                                 &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc7843b7d60&gt;,
+                                 &#39;sklearn_classifier&#39;: PassiveAggressiveClassifier(C=0.14268277711454813, max_iter=128, random_state=1,
+                            tol=0.0002600768160857831, warm_start=True)},
+                             {   &#39;balancing&#39;: Balancing(random_state=1),
+                                 &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc785b7e4c0&gt;,
+                                 &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc787ec5b20&gt;,
+                                 &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc785b7e400&gt;,
+                                 &#39;sklearn_classifier&#39;: PassiveAggressiveClassifier(C=0.14268277711454813, max_iter=128, random_state=1,
+                            tol=0.0002600768160857831, warm_start=True)}],
+           &#39;model_id&#39;: 4,
+           &#39;rank&#39;: 11,
+           &#39;voting_model&#39;: VotingClassifier(estimators=None, voting=&#39;soft&#39;)},
+    5: {   &#39;cost&#39;: 0.035211267605633804,
+           &#39;ensemble_weight&#39;: 0.02,
+           &#39;estimators&#39;: [   {   &#39;balancing&#39;: Balancing(random_state=1),
+                                 &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc7903592e0&gt;,
+                                 &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc787e052e0&gt;,
+                                 &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78ba37610&gt;,
+                                 &#39;sklearn_classifier&#39;: HistGradientBoostingClassifier(early_stopping=False, l2_regularization=1e-10,
+                               learning_rate=0.16262682406125173, max_iter=64,
+                               max_leaf_nodes=66, n_iter_no_change=0,
+                               random_state=1, validation_fraction=None,
+                               warm_start=True)},
+                             {   &#39;balancing&#39;: Balancing(random_state=1),
+                                 &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc786184130&gt;,
+                                 &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78a956c40&gt;,
+                                 &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78fdee8b0&gt;,
+                                 &#39;sklearn_classifier&#39;: HistGradientBoostingClassifier(early_stopping=False, l2_regularization=1e-10,
+                               learning_rate=0.16262682406125173, max_iter=64,
+                               max_leaf_nodes=66, n_iter_no_change=0,
+                               random_state=1, validation_fraction=None,
+                               warm_start=True)},
+                             {   &#39;balancing&#39;: Balancing(random_state=1),
+                                 &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78656e760&gt;,
+                                 &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7878b0700&gt;,
+                                 &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78656e670&gt;,
+                                 &#39;sklearn_classifier&#39;: HistGradientBoostingClassifier(early_stopping=False, l2_regularization=1e-10,
+                               learning_rate=0.16262682406125173, max_iter=64,
+                               max_leaf_nodes=66, n_iter_no_change=0,
+                               random_state=1, validation_fraction=None,
+                               warm_start=True)},
+                             {   &#39;balancing&#39;: Balancing(random_state=1),
+                                 &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc7850b5e50&gt;,
+                                 &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc785087190&gt;,
+                                 &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc7850b5d30&gt;,
+                                 &#39;sklearn_classifier&#39;: HistGradientBoostingClassifier(early_stopping=False, l2_regularization=1e-10,
+                               learning_rate=0.16262682406125173, max_iter=64,
+                               max_leaf_nodes=66, n_iter_no_change=0,
+                               random_state=1, validation_fraction=None,
+                               warm_start=True)},
+                             {   &#39;balancing&#39;: Balancing(random_state=1),
+                                 &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78690f400&gt;,
+                                 &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7875b5640&gt;,
+                                 &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78690ffd0&gt;,
+                                 &#39;sklearn_classifier&#39;: HistGradientBoostingClassifier(early_stopping=False, l2_regularization=1e-10,
+                               learning_rate=0.16262682406125173, max_iter=64,
+                               max_leaf_nodes=66, n_iter_no_change=0,
+                               random_state=1, validation_fraction=None,
+                               warm_start=True)}],
+           &#39;model_id&#39;: 5,
+           &#39;rank&#39;: 5,
+           &#39;voting_model&#39;: VotingClassifier(estimators=None, voting=&#39;soft&#39;)},
+    6: {   &#39;cost&#39;: 0.04694835680751174,
+           &#39;ensemble_weight&#39;: 0.02,
+           &#39;estimators&#39;: [   {   &#39;balancing&#39;: Balancing(random_state=1),
+                                 &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc786fe3af0&gt;,
+                                 &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78c1023d0&gt;,
+                                 &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc786fe30a0&gt;,
+                                 &#39;sklearn_classifier&#39;: HistGradientBoostingClassifier(early_stopping=True,
+                               l2_regularization=3.609412172481434e-10,
+                               learning_rate=0.05972079854295879, max_iter=64,
+                               max_leaf_nodes=4, min_samples_leaf=2,
+                               n_iter_no_change=14, random_state=1,
+                               validation_fraction=None, warm_start=True)},
+                             {   &#39;balancing&#39;: Balancing(random_state=1),
+                                 &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc7876a55b0&gt;,
+                                 &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78753b430&gt;,
+                                 &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc7876a5520&gt;,
+                                 &#39;sklearn_classifier&#39;: HistGradientBoostingClassifier(early_stopping=True,
+                               l2_regularization=3.609412172481434e-10,
+                               learning_rate=0.05972079854295879, max_iter=64,
+                               max_leaf_nodes=4, min_samples_leaf=2,
+                               n_iter_no_change=14, random_state=1,
+                               validation_fraction=None, warm_start=True)},
+                             {   &#39;balancing&#39;: Balancing(random_state=1),
+                                 &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc785820a90&gt;,
+                                 &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78728db80&gt;,
+                                 &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc785820970&gt;,
+                                 &#39;sklearn_classifier&#39;: HistGradientBoostingClassifier(early_stopping=True,
+                               l2_regularization=3.609412172481434e-10,
+                               learning_rate=0.05972079854295879, max_iter=64,
+                               max_leaf_nodes=4, min_samples_leaf=2,
+                               n_iter_no_change=14, random_state=1,
+                               validation_fraction=None, warm_start=True)},
+                             {   &#39;balancing&#39;: Balancing(random_state=1),
+                                 &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78c5ab9a0&gt;,
+                                 &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78c099940&gt;,
+                                 &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c5ab8b0&gt;,
+                                 &#39;sklearn_classifier&#39;: HistGradientBoostingClassifier(early_stopping=True,
+                               l2_regularization=3.609412172481434e-10,
+                               learning_rate=0.05972079854295879, max_iter=64,
+                               max_leaf_nodes=4, min_samples_leaf=2,
+                               n_iter_no_change=14, random_state=1,
+                               validation_fraction=None, warm_start=True)},
+                             {   &#39;balancing&#39;: Balancing(random_state=1),
+                                 &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc785eda430&gt;,
+                                 &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc784850e80&gt;,
+                                 &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc785eda250&gt;,
+                                 &#39;sklearn_classifier&#39;: HistGradientBoostingClassifier(early_stopping=True,
+                               l2_regularization=3.609412172481434e-10,
+                               learning_rate=0.05972079854295879, max_iter=64,
+                               max_leaf_nodes=4, min_samples_leaf=2,
+                               n_iter_no_change=14, random_state=1,
+                               validation_fraction=None, warm_start=True)}],
+           &#39;model_id&#39;: 6,
+           &#39;rank&#39;: 9,
+           &#39;voting_model&#39;: VotingClassifier(estimators=None, voting=&#39;soft&#39;)},
+    7: {   &#39;cost&#39;: 0.035211267605633784,
+           &#39;ensemble_weight&#39;: 0.16,
+           &#39;estimators&#39;: [   {   &#39;balancing&#39;: Balancing(random_state=1),
+                                 &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc7868546d0&gt;,
+                                 &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78a593d90&gt;,
+                                 &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c9447c0&gt;,
+                                 &#39;sklearn_classifier&#39;: SGDClassifier(alpha=0.0002346515712987664, average=True, eta0=0.01, loss=&#39;log&#39;,
+              max_iter=128, penalty=&#39;l1&#39;, random_state=1,
+              tol=1.3716748930467322e-05, warm_start=True)},
+                             {   &#39;balancing&#39;: Balancing(random_state=1),
+                                 &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc786899d90&gt;,
+                                 &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc784bd3a00&gt;,
+                                 &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc786899b50&gt;,
+                                 &#39;sklearn_classifier&#39;: SGDClassifier(alpha=0.0002346515712987664, average=True, eta0=0.01, loss=&#39;log&#39;,
+              max_iter=128, penalty=&#39;l1&#39;, random_state=1,
+              tol=1.3716748930467322e-05, warm_start=True)},
+                             {   &#39;balancing&#39;: Balancing(random_state=1),
+                                 &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc786fe3550&gt;,
+                                 &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7875c2c40&gt;,
+                                 &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc786fe33d0&gt;,
+                                 &#39;sklearn_classifier&#39;: SGDClassifier(alpha=0.0002346515712987664, average=True, eta0=0.01, loss=&#39;log&#39;,
+              max_iter=128, penalty=&#39;l1&#39;, random_state=1,
+              tol=1.3716748930467322e-05, warm_start=True)},
+                             {   &#39;balancing&#39;: Balancing(random_state=1),
+                                 &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc787a7a910&gt;,
+                                 &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc787a7e670&gt;,
+                                 &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc787a7a7f0&gt;,
+                                 &#39;sklearn_classifier&#39;: SGDClassifier(alpha=0.0002346515712987664, average=True, eta0=0.01, loss=&#39;log&#39;,
+              max_iter=128, penalty=&#39;l1&#39;, random_state=1,
+              tol=1.3716748930467322e-05, warm_start=True)},
+                             {   &#39;balancing&#39;: Balancing(random_state=1),
+                                 &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc786b94460&gt;,
+                                 &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc786b860d0&gt;,
+                                 &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc786b94340&gt;,
+                                 &#39;sklearn_classifier&#39;: SGDClassifier(alpha=0.0002346515712987664, average=True, eta0=0.01, loss=&#39;log&#39;,
+              max_iter=128, penalty=&#39;l1&#39;, random_state=1,
+              tol=1.3716748930467322e-05, warm_start=True)}],
+           &#39;model_id&#39;: 7,
+           &#39;rank&#39;: 6,
+           &#39;voting_model&#39;: VotingClassifier(estimators=None, voting=&#39;soft&#39;)},
+    8: {   &#39;cost&#39;: 0.039906103286385,
+           &#39;ensemble_weight&#39;: 0.08,
+           &#39;estimators&#39;: [   {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+                                 &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78509fe80&gt;,
+                                 &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78b4c3280&gt;,
+                                 &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78ff43b20&gt;,
+                                 &#39;sklearn_classifier&#39;: RandomForestClassifier(bootstrap=False, criterion=&#39;entropy&#39;, max_features=4,
+                       min_samples_split=4, n_estimators=64, n_jobs=1,
+                       random_state=1, warm_start=True)},
+                             {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+                                 &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc787a7a5b0&gt;,
+                                 &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78b73b580&gt;,
+                                 &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc787a7afa0&gt;,
+                                 &#39;sklearn_classifier&#39;: RandomForestClassifier(bootstrap=False, criterion=&#39;entropy&#39;, max_features=4,
+                       min_samples_split=4, n_estimators=64, n_jobs=1,
+                       random_state=1, warm_start=True)},
+                             {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+                                 &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc786ed1c10&gt;,
+                                 &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78b7d9340&gt;,
+                                 &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc786ed1af0&gt;,
+                                 &#39;sklearn_classifier&#39;: RandomForestClassifier(bootstrap=False, criterion=&#39;entropy&#39;, max_features=4,
+                       min_samples_split=4, n_estimators=64, n_jobs=1,
+                       random_state=1, warm_start=True)},
+                             {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+                                 &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78764e8b0&gt;,
+                                 &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc787ba92b0&gt;,
+                                 &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78764e7c0&gt;,
+                                 &#39;sklearn_classifier&#39;: RandomForestClassifier(bootstrap=False, criterion=&#39;entropy&#39;, max_features=4,
+                       min_samples_split=4, n_estimators=64, n_jobs=1,
+                       random_state=1, warm_start=True)},
+                             {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+                                 &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc784c08610&gt;,
+                                 &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78715dfd0&gt;,
+                                 &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc784c08520&gt;,
+                                 &#39;sklearn_classifier&#39;: RandomForestClassifier(bootstrap=False, criterion=&#39;entropy&#39;, max_features=4,
+                       min_samples_split=4, n_estimators=64, n_jobs=1,
+                       random_state=1, warm_start=True)}],
+           &#39;model_id&#39;: 8,
+           &#39;rank&#39;: 7,
+           &#39;voting_model&#39;: VotingClassifier(estimators=None, voting=&#39;soft&#39;)},
+    9: {   &#39;cost&#39;: 0.032863849765258205,
+           &#39;ensemble_weight&#39;: 0.04,
+           &#39;estimators&#39;: [   {   &#39;balancing&#39;: Balancing(random_state=1),
+                                 &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78ca86700&gt;,
+                                 &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7873f3460&gt;,
+                                 &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78ca86b50&gt;,
+                                 &#39;sklearn_classifier&#39;: ExtraTreesClassifier(criterion=&#39;entropy&#39;, max_features=8, min_samples_split=3,
+                     n_estimators=512, n_jobs=1, random_state=1,
+                     warm_start=True)},
+                             {   &#39;balancing&#39;: Balancing(random_state=1),
+                                 &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78c93b3a0&gt;,
+                                 &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78c8b7970&gt;,
+                                 &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c93bf10&gt;,
+                                 &#39;sklearn_classifier&#39;: ExtraTreesClassifier(criterion=&#39;entropy&#39;, max_features=8, min_samples_split=3,
+                     n_estimators=512, n_jobs=1, random_state=1,
+                     warm_start=True)},
+                             {   &#39;balancing&#39;: Balancing(random_state=1),
+                                 &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc787c50580&gt;,
+                                 &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78b7d9970&gt;,
+                                 &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc787c50460&gt;,
+                                 &#39;sklearn_classifier&#39;: ExtraTreesClassifier(criterion=&#39;entropy&#39;, max_features=8, min_samples_split=3,
+                     n_estimators=512, n_jobs=1, random_state=1,
+                     warm_start=True)},
+                             {   &#39;balancing&#39;: Balancing(random_state=1),
+                                 &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78b21c880&gt;,
+                                 &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78beb3ac0&gt;,
+                                 &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78b21c7c0&gt;,
+                                 &#39;sklearn_classifier&#39;: ExtraTreesClassifier(criterion=&#39;entropy&#39;, max_features=8, min_samples_split=3,
+                     n_estimators=512, n_jobs=1, random_state=1,
+                     warm_start=True)},
+                             {   &#39;balancing&#39;: Balancing(random_state=1),
+                                 &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc787e609a0&gt;,
+                                 &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc785edabe0&gt;,
+                                 &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc787e608e0&gt;,
+                                 &#39;sklearn_classifier&#39;: ExtraTreesClassifier(criterion=&#39;entropy&#39;, max_features=8, min_samples_split=3,
+                     n_estimators=512, n_jobs=1, random_state=1,
+                     warm_start=True)}],
+           &#39;model_id&#39;: 9,
+           &#39;rank&#39;: 4,
+           &#39;voting_model&#39;: VotingClassifier(estimators=None, voting=&#39;soft&#39;)}}
 auto-sklearn results:
   Dataset name: breast_cancer
   Metric: accuracy
   Best validation score: 0.971831
   Number of target algorithm runs: 11
-  Number of successful target algorithm runs: 10
+  Number of successful target algorithm runs: 11
   Number of crashed target algorithm runs: 0
-  Number of target algorithms that exceeded the time limit: 1
+  Number of target algorithms that exceeded the time limit: 0
   Number of target algorithms that exceeded the memory limit: 0
 
 Accuracy score 0.965034965034965
 </pre></div>
 </div>
-</section>
-<section id="use-an-iterative-fit-cross-validation-with-successive-halving">
+</div>
+<div class="section" id="use-an-iterative-fit-cross-validation-with-successive-halving">
 <h2>Use an iterative fit cross-validation with successive halving<a class="headerlink" href="#use-an-iterative-fit-cross-validation-with-successive-halving" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">X</span><span class="p">,</span> <span class="n">y</span> <span class="o">=</span> <span class="n">sklearn</span><span class="o">.</span><span class="n">datasets</span><span class="o">.</span><span class="n">load_breast_cancer</span><span class="p">(</span><span class="n">return_X_y</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
 <span class="n">X_train</span><span class="p">,</span> <span class="n">X_test</span><span class="p">,</span> <span class="n">y_train</span><span class="p">,</span> <span class="n">y_test</span> <span class="o">=</span> \
@@ -538,7 +888,7 @@ <h2>Use an iterative fit cross-validation with successive halving<a class="heade
 <span class="n">automl</span><span class="o">.</span><span class="n">fit</span><span class="p">(</span><span class="n">X_train</span><span class="p">,</span> <span class="n">y_train</span><span class="p">,</span> <span class="n">dataset_name</span><span class="o">=</span><span class="s1">&#39;breast_cancer&#39;</span><span class="p">)</span>
 
 <span class="c1"># Print the final ensemble constructed by auto-sklearn.</span>
-<span class="nb">print</span><span class="p">(</span><span class="n">automl</span><span class="o">.</span><span class="n">show_models</span><span class="p">())</span>
+<span class="n">pprint</span><span class="p">(</span><span class="n">automl</span><span class="o">.</span><span class="n">show_models</span><span class="p">(),</span> <span class="n">indent</span><span class="o">=</span><span class="mi">4</span><span class="p">)</span>
 <span class="n">automl</span><span class="o">.</span><span class="n">refit</span><span class="p">(</span><span class="n">X_train</span><span class="p">,</span> <span class="n">y_train</span><span class="p">)</span>
 <span class="n">predictions</span> <span class="o">=</span> <span class="n">automl</span><span class="o">.</span><span class="n">predict</span><span class="p">(</span><span class="n">X_test</span><span class="p">)</span>
 <span class="c1"># Print statistics about the auto-sklearn run such as number of</span>
@@ -548,65 +898,75 @@ <h2>Use an iterative fit cross-validation with successive halving<a class="heade
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/opt/hostedtoolcache/Python/3.8.12/x64/lib/python3.8/site-packages/smac/intensification/parallel_scheduling.py:152: UserWarning: SuccessiveHalving is intended to be used with more than 1 worker but num_workers=1
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:68: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+  self.metafeatures = self.metafeatures.append(metafeatures)
+/home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:72: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+  self.algorithm_runs[metric].append(runs)
+/opt/hostedtoolcache/Python/3.8.12/x64/lib/python3.8/site-packages/smac/intensification/parallel_scheduling.py:152: UserWarning: SuccessiveHalving is intended to be used with more than 1 worker but num_workers=1
   warnings.warn(&quot;{} is intended to be used &quot;
-[(0.240000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;sgd&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:sgd:alpha&#39;: 0.0002346515712987664, &#39;classifier:sgd:average&#39;: &#39;True&#39;, &#39;classifier:sgd:fit_intercept&#39;: &#39;True&#39;, &#39;classifier:sgd:learning_rate&#39;: &#39;optimal&#39;, &#39;classifier:sgd:loss&#39;: &#39;log&#39;, &#39;classifier:sgd:penalty&#39;: &#39;l1&#39;, &#39;classifier:sgd:tol&#39;: 1.3716748930467322e-05, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;standardize&#39;},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.180000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;random_forest&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:random_forest:bootstrap&#39;: &#39;False&#39;, &#39;classifier:random_forest:criterion&#39;: &#39;entropy&#39;, &#39;classifier:random_forest:max_depth&#39;: &#39;None&#39;, &#39;classifier:random_forest:max_features&#39;: 0.41808321658160696, &#39;classifier:random_forest:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:random_forest:min_impurity_decrease&#39;: 0.0, &#39;classifier:random_forest:min_samples_leaf&#39;: 1, &#39;classifier:random_forest:min_samples_split&#39;: 4, &#39;classifier:random_forest:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;median&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;standardize&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.011283688651384545},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.160000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;passive_aggressive&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:passive_aggressive:C&#39;: 0.14268277711454813, &#39;classifier:passive_aggressive:average&#39;: &#39;False&#39;, &#39;classifier:passive_aggressive:fit_intercept&#39;: &#39;True&#39;, &#39;classifier:passive_aggressive:loss&#39;: &#39;hinge&#39;, &#39;classifier:passive_aggressive:tol&#39;: 0.0002600768160857831, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;none&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.0070580904199417415},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.140000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;gradient_boosting&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:gradient_boosting:early_stop&#39;: &#39;train&#39;, &#39;classifier:gradient_boosting:l2_regularization&#39;: 3.609412172481434e-10, &#39;classifier:gradient_boosting:learning_rate&#39;: 0.05972079854295879, &#39;classifier:gradient_boosting:loss&#39;: &#39;auto&#39;, &#39;classifier:gradient_boosting:max_bins&#39;: 255, &#39;classifier:gradient_boosting:max_depth&#39;: &#39;None&#39;, &#39;classifier:gradient_boosting:max_leaf_nodes&#39;: 4, &#39;classifier:gradient_boosting:min_samples_leaf&#39;: 2, &#39;classifier:gradient_boosting:scoring&#39;: &#39;loss&#39;, &#39;classifier:gradient_boosting:tol&#39;: 1e-07, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;minmax&#39;, &#39;classifier:gradient_boosting:n_iter_no_change&#39;: 14},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.120000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;random_forest&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:random_forest:bootstrap&#39;: &#39;True&#39;, &#39;classifier:random_forest:criterion&#39;: &#39;gini&#39;, &#39;classifier:random_forest:max_depth&#39;: &#39;None&#39;, &#39;classifier:random_forest:max_features&#39;: 0.5, &#39;classifier:random_forest:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:random_forest:min_impurity_decrease&#39;: 0.0, &#39;classifier:random_forest:min_samples_leaf&#39;: 1, &#39;classifier:random_forest:min_samples_split&#39;: 2, &#39;classifier:random_forest:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;standardize&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.01},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.100000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;random_forest&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:random_forest:bootstrap&#39;: &#39;True&#39;, &#39;classifier:random_forest:criterion&#39;: &#39;entropy&#39;, &#39;classifier:random_forest:max_depth&#39;: &#39;None&#39;, &#39;classifier:random_forest:max_features&#39;: 0.926283631486858, &#39;classifier:random_forest:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:random_forest:min_impurity_decrease&#39;: 0.0, &#39;classifier:random_forest:min_samples_leaf&#39;: 7, &#39;classifier:random_forest:min_samples_split&#39;: 2, &#39;classifier:random_forest:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;standardize&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.16265262021972576},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.060000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;gradient_boosting&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:gradient_boosting:early_stop&#39;: &#39;off&#39;, &#39;classifier:gradient_boosting:l2_regularization&#39;: 1e-10, &#39;classifier:gradient_boosting:learning_rate&#39;: 0.16262682406125173, &#39;classifier:gradient_boosting:loss&#39;: &#39;auto&#39;, &#39;classifier:gradient_boosting:max_bins&#39;: 255, &#39;classifier:gradient_boosting:max_depth&#39;: &#39;None&#39;, &#39;classifier:gradient_boosting:max_leaf_nodes&#39;: 66, &#39;classifier:gradient_boosting:min_samples_leaf&#39;: 20, &#39;classifier:gradient_boosting:scoring&#39;: &#39;loss&#39;, &#39;classifier:gradient_boosting:tol&#39;: 1e-07, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;robust_scaler&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.005428587241449129, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max&#39;: 0.75, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min&#39;: 0.23746960178084334},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-]
+{   2: {   &#39;balancing&#39;: Balancing(random_state=1),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78648be80&gt;,
+           &#39;cost&#39;: 0.046948356807511755,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78753bfa0&gt;,
+           &#39;ensemble_weight&#39;: 0.12,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78648b5b0&gt;,
+           &#39;model_id&#39;: 2,
+           &#39;rank&#39;: 4,
+           &#39;sklearn_classifier&#39;: None},
+    3: {   &#39;balancing&#39;: Balancing(random_state=1),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc7871573a0&gt;,
+           &#39;cost&#39;: 0.05164319248826292,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc785814430&gt;,
+           &#39;ensemble_weight&#39;: 0.1,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc7871571c0&gt;,
+           &#39;model_id&#39;: 3,
+           &#39;rank&#39;: 6,
+           &#39;sklearn_classifier&#39;: None},
+    4: {   &#39;balancing&#39;: Balancing(random_state=1),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc786ff0ca0&gt;,
+           &#39;cost&#39;: 0.11267605633802817,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7875b5970&gt;,
+           &#39;ensemble_weight&#39;: 0.16,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc786ff0970&gt;,
+           &#39;model_id&#39;: 4,
+           &#39;rank&#39;: 7,
+           &#39;sklearn_classifier&#39;: None},
+    5: {   &#39;balancing&#39;: Balancing(random_state=1),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc786912b20&gt;,
+           &#39;cost&#39;: 0.035211267605633804,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc787bb7550&gt;,
+           &#39;ensemble_weight&#39;: 0.06,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc786912460&gt;,
+           &#39;model_id&#39;: 5,
+           &#39;rank&#39;: 2,
+           &#39;sklearn_classifier&#39;: None},
+    6: {   &#39;balancing&#39;: Balancing(random_state=1),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc785087220&gt;,
+           &#39;cost&#39;: 0.04694835680751174,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78715d850&gt;,
+           &#39;ensemble_weight&#39;: 0.14,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc785087df0&gt;,
+           &#39;model_id&#39;: 6,
+           &#39;rank&#39;: 5,
+           &#39;sklearn_classifier&#39;: None},
+    7: {   &#39;balancing&#39;: Balancing(random_state=1),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc784c0adf0&gt;,
+           &#39;cost&#39;: 0.03286384976525822,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78715d9a0&gt;,
+           &#39;ensemble_weight&#39;: 0.24,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc787bb7430&gt;,
+           &#39;model_id&#39;: 7,
+           &#39;rank&#39;: 1,
+           &#39;sklearn_classifier&#39;: None},
+    8: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78715d310&gt;,
+           &#39;cost&#39;: 0.039906103286385,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc784c0a520&gt;,
+           &#39;ensemble_weight&#39;: 0.18,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78715d3a0&gt;,
+           &#39;model_id&#39;: 8,
+           &#39;rank&#39;: 3,
+           &#39;sklearn_classifier&#39;: None}}
 auto-sklearn results:
   Dataset name: breast_cancer
   Metric: accuracy
@@ -620,8 +980,8 @@ <h2>Use an iterative fit cross-validation with successive halving<a class="heade
 Accuracy score 0.965034965034965
 </pre></div>
 </div>
-</section>
-<section id="next-we-see-the-use-of-subsampling-as-a-budget-in-auto-sklearn">
+</div>
+<div class="section" id="next-we-see-the-use-of-subsampling-as-a-budget-in-auto-sklearn">
 <h2>Next, we see the use of subsampling as a budget in Auto-sklearn<a class="headerlink" href="#next-we-see-the-use-of-subsampling-as-a-budget-in-auto-sklearn" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">X</span><span class="p">,</span> <span class="n">y</span> <span class="o">=</span> <span class="n">sklearn</span><span class="o">.</span><span class="n">datasets</span><span class="o">.</span><span class="n">load_breast_cancer</span><span class="p">(</span><span class="n">return_X_y</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
 <span class="n">X_train</span><span class="p">,</span> <span class="n">X_test</span><span class="p">,</span> <span class="n">y_train</span><span class="p">,</span> <span class="n">y_test</span> <span class="o">=</span> \
@@ -642,7 +1002,7 @@ <h2>Next, we see the use of subsampling as a budget in Auto-sklearn<a class="hea
 <span class="n">automl</span><span class="o">.</span><span class="n">fit</span><span class="p">(</span><span class="n">X_train</span><span class="p">,</span> <span class="n">y_train</span><span class="p">,</span> <span class="n">dataset_name</span><span class="o">=</span><span class="s1">&#39;breast_cancer&#39;</span><span class="p">)</span>
 
 <span class="c1"># Print the final ensemble constructed by auto-sklearn.</span>
-<span class="nb">print</span><span class="p">(</span><span class="n">automl</span><span class="o">.</span><span class="n">show_models</span><span class="p">())</span>
+<span class="n">pprint</span><span class="p">(</span><span class="n">automl</span><span class="o">.</span><span class="n">show_models</span><span class="p">(),</span> <span class="n">indent</span><span class="o">=</span><span class="mi">4</span><span class="p">)</span>
 <span class="n">predictions</span> <span class="o">=</span> <span class="n">automl</span><span class="o">.</span><span class="n">predict</span><span class="p">(</span><span class="n">X_test</span><span class="p">)</span>
 <span class="c1"># Print statistics about the auto-sklearn run such as number of</span>
 <span class="c1"># iterations, number of models failed with a time out.</span>
@@ -651,104 +1011,51 @@ <h2>Next, we see the use of subsampling as a budget in Auto-sklearn<a class="hea
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/opt/hostedtoolcache/Python/3.8.12/x64/lib/python3.8/site-packages/smac/intensification/parallel_scheduling.py:152: UserWarning: SuccessiveHalving is intended to be used with more than 1 worker but num_workers=1
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:68: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+  self.metafeatures = self.metafeatures.append(metafeatures)
+/home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:72: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+  self.algorithm_runs[metric].append(runs)
+/opt/hostedtoolcache/Python/3.8.12/x64/lib/python3.8/site-packages/smac/intensification/parallel_scheduling.py:152: UserWarning: SuccessiveHalving is intended to be used with more than 1 worker but num_workers=1
   warnings.warn(&quot;{} is intended to be used &quot;
-[(0.340000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;mlp&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;fast_ica&#39;, &#39;classifier:mlp:activation&#39;: &#39;relu&#39;, &#39;classifier:mlp:alpha&#39;: 0.0017940473175767063, &#39;classifier:mlp:batch_size&#39;: &#39;auto&#39;, &#39;classifier:mlp:beta_1&#39;: 0.9, &#39;classifier:mlp:beta_2&#39;: 0.999, &#39;classifier:mlp:early_stopping&#39;: &#39;valid&#39;, &#39;classifier:mlp:epsilon&#39;: 1e-08, &#39;classifier:mlp:hidden_layer_depth&#39;: 2, &#39;classifier:mlp:learning_rate_init&#39;: 0.0004684917334431039, &#39;classifier:mlp:n_iter_no_change&#39;: 32, &#39;classifier:mlp:num_nodes_per_layer&#39;: 101, &#39;classifier:mlp:shuffle&#39;: &#39;True&#39;, &#39;classifier:mlp:solver&#39;: &#39;adam&#39;, &#39;classifier:mlp:tol&#39;: 0.0001, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;standardize&#39;, &#39;feature_preprocessor:fast_ica:algorithm&#39;: &#39;parallel&#39;, &#39;feature_preprocessor:fast_ica:fun&#39;: &#39;exp&#39;, &#39;feature_preprocessor:fast_ica:whiten&#39;: &#39;False&#39;, &#39;classifier:mlp:validation_fraction&#39;: 0.1},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.260000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;mlp&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;feature_agglomeration&#39;, &#39;classifier:mlp:activation&#39;: &#39;tanh&#39;, &#39;classifier:mlp:alpha&#39;: 0.0001363185819149026, &#39;classifier:mlp:batch_size&#39;: &#39;auto&#39;, &#39;classifier:mlp:beta_1&#39;: 0.9, &#39;classifier:mlp:beta_2&#39;: 0.999, &#39;classifier:mlp:early_stopping&#39;: &#39;valid&#39;, &#39;classifier:mlp:epsilon&#39;: 1e-08, &#39;classifier:mlp:hidden_layer_depth&#39;: 3, &#39;classifier:mlp:learning_rate_init&#39;: 0.00018009776276177523, &#39;classifier:mlp:n_iter_no_change&#39;: 32, &#39;classifier:mlp:num_nodes_per_layer&#39;: 115, &#39;classifier:mlp:shuffle&#39;: &#39;True&#39;, &#39;classifier:mlp:solver&#39;: &#39;adam&#39;, &#39;classifier:mlp:tol&#39;: 0.0001, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;median&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;standardize&#39;, &#39;feature_preprocessor:feature_agglomeration:affinity&#39;: &#39;euclidean&#39;, &#39;feature_preprocessor:feature_agglomeration:linkage&#39;: &#39;ward&#39;, &#39;feature_preprocessor:feature_agglomeration:n_clusters&#39;: 182, &#39;feature_preprocessor:feature_agglomeration:pooling_func&#39;: &#39;mean&#39;, &#39;classifier:mlp:validation_fraction&#39;: 0.1},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.120000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;random_forest&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;select_percentile_classification&#39;, &#39;classifier:random_forest:bootstrap&#39;: &#39;True&#39;, &#39;classifier:random_forest:criterion&#39;: &#39;entropy&#39;, &#39;classifier:random_forest:max_depth&#39;: &#39;None&#39;, &#39;classifier:random_forest:max_features&#39;: 0.49138075723513286, &#39;classifier:random_forest:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:random_forest:min_impurity_decrease&#39;: 0.0, &#39;classifier:random_forest:min_samples_leaf&#39;: 1, &#39;classifier:random_forest:min_samples_split&#39;: 6, &#39;classifier:random_forest:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;minmax&#39;, &#39;feature_preprocessor:select_percentile_classification:percentile&#39;: 56.97947373958566, &#39;feature_preprocessor:select_percentile_classification:score_func&#39;: &#39;mutual_info&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.42693600390988135},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.060000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;random_forest&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:random_forest:bootstrap&#39;: &#39;True&#39;, &#39;classifier:random_forest:criterion&#39;: &#39;gini&#39;, &#39;classifier:random_forest:max_depth&#39;: &#39;None&#39;, &#39;classifier:random_forest:max_features&#39;: 0.5, &#39;classifier:random_forest:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:random_forest:min_impurity_decrease&#39;: 0.0, &#39;classifier:random_forest:min_samples_leaf&#39;: 1, &#39;classifier:random_forest:min_samples_split&#39;: 2, &#39;classifier:random_forest:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;standardize&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.01},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.040000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;random_forest&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:random_forest:bootstrap&#39;: &#39;True&#39;, &#39;classifier:random_forest:criterion&#39;: &#39;gini&#39;, &#39;classifier:random_forest:max_depth&#39;: &#39;None&#39;, &#39;classifier:random_forest:max_features&#39;: 0.5, &#39;classifier:random_forest:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:random_forest:min_impurity_decrease&#39;: 0.0, &#39;classifier:random_forest:min_samples_leaf&#39;: 1, &#39;classifier:random_forest:min_samples_split&#39;: 2, &#39;classifier:random_forest:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;standardize&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.01},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.040000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;random_forest&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:random_forest:bootstrap&#39;: &#39;True&#39;, &#39;classifier:random_forest:criterion&#39;: &#39;gini&#39;, &#39;classifier:random_forest:max_depth&#39;: &#39;None&#39;, &#39;classifier:random_forest:max_features&#39;: 0.5, &#39;classifier:random_forest:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:random_forest:min_impurity_decrease&#39;: 0.0, &#39;classifier:random_forest:min_samples_leaf&#39;: 1, &#39;classifier:random_forest:min_samples_split&#39;: 2, &#39;classifier:random_forest:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;standardize&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.01},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.040000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;extra_trees&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;polynomial&#39;, &#39;classifier:extra_trees:bootstrap&#39;: &#39;False&#39;, &#39;classifier:extra_trees:criterion&#39;: &#39;gini&#39;, &#39;classifier:extra_trees:max_depth&#39;: &#39;None&#39;, &#39;classifier:extra_trees:max_features&#39;: 0.5707983257382487, &#39;classifier:extra_trees:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:extra_trees:min_impurity_decrease&#39;: 0.0, &#39;classifier:extra_trees:min_samples_leaf&#39;: 3, &#39;classifier:extra_trees:min_samples_split&#39;: 11, &#39;classifier:extra_trees:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;median&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;none&#39;, &#39;feature_preprocessor:polynomial:degree&#39;: 2, &#39;feature_preprocessor:polynomial:include_bias&#39;: &#39;False&#39;, &#39;feature_preprocessor:polynomial:interaction_only&#39;: &#39;False&#39;},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.040000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;random_forest&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;extra_trees_preproc_for_classification&#39;, &#39;classifier:random_forest:bootstrap&#39;: &#39;True&#39;, &#39;classifier:random_forest:criterion&#39;: &#39;gini&#39;, &#39;classifier:random_forest:max_depth&#39;: &#39;None&#39;, &#39;classifier:random_forest:max_features&#39;: 0.43999367631975456, &#39;classifier:random_forest:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:random_forest:min_impurity_decrease&#39;: 0.0, &#39;classifier:random_forest:min_samples_leaf&#39;: 2, &#39;classifier:random_forest:min_samples_split&#39;: 2, &#39;classifier:random_forest:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;standardize&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:bootstrap&#39;: &#39;True&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:criterion&#39;: &#39;entropy&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:max_depth&#39;: &#39;None&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:max_features&#39;: 0.8134515743047006, &#39;feature_preprocessor:extra_trees_preproc_for_classification:max_leaf_nodes&#39;: &#39;None&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_impurity_decrease&#39;: 0.0, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_samples_leaf&#39;: 9, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_samples_split&#39;: 20, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_weight_fraction_leaf&#39;: 0.0, &#39;feature_preprocessor:extra_trees_preproc_for_classification:n_estimators&#39;: 100},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.040000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;extra_trees&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;extra_trees_preproc_for_classification&#39;, &#39;classifier:extra_trees:bootstrap&#39;: &#39;False&#39;, &#39;classifier:extra_trees:criterion&#39;: &#39;gini&#39;, &#39;classifier:extra_trees:max_depth&#39;: &#39;None&#39;, &#39;classifier:extra_trees:max_features&#39;: 0.9292309396985746, &#39;classifier:extra_trees:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:extra_trees:min_impurity_decrease&#39;: 0.0, &#39;classifier:extra_trees:min_samples_leaf&#39;: 1, &#39;classifier:extra_trees:min_samples_split&#39;: 10, &#39;classifier:extra_trees:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;median&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;normalize&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:bootstrap&#39;: &#39;False&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:criterion&#39;: &#39;entropy&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:max_depth&#39;: &#39;None&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:max_features&#39;: 0.9929881254946676, &#39;feature_preprocessor:extra_trees_preproc_for_classification:max_leaf_nodes&#39;: &#39;None&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_impurity_decrease&#39;: 0.0, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_samples_leaf&#39;: 1, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_samples_split&#39;: 2, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_weight_fraction_leaf&#39;: 0.0, &#39;feature_preprocessor:extra_trees_preproc_for_classification:n_estimators&#39;: 100, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.032719158639429445},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.020000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;random_forest&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:random_forest:bootstrap&#39;: &#39;True&#39;, &#39;classifier:random_forest:criterion&#39;: &#39;gini&#39;, &#39;classifier:random_forest:max_depth&#39;: &#39;None&#39;, &#39;classifier:random_forest:max_features&#39;: 0.5, &#39;classifier:random_forest:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:random_forest:min_impurity_decrease&#39;: 0.0, &#39;classifier:random_forest:min_samples_leaf&#39;: 1, &#39;classifier:random_forest:min_samples_split&#39;: 2, &#39;classifier:random_forest:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;standardize&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.01},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-]
+{   3: {   &#39;balancing&#39;: Balancing(random_state=1),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78b503fd0&gt;,
+           &#39;cost&#39;: 0.021276595744680882,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc790367be0&gt;,
+           &#39;ensemble_weight&#39;: 0.4,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78a6db2b0&gt;,
+           &#39;model_id&#39;: 3,
+           &#39;rank&#39;: 2,
+           &#39;sklearn_classifier&#39;: MLPClassifier(activation=&#39;tanh&#39;, alpha=0.0001363185819149026, beta_1=0.999,
+              beta_2=0.9, early_stopping=True,
+              hidden_layer_sizes=(115, 115, 115),
+              learning_rate_init=0.00018009776276177523, max_iter=32,
+              n_iter_no_change=32, random_state=1, verbose=0, warm_start=True)},
+    7: {   &#39;balancing&#39;: Balancing(random_state=1),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78ca21b20&gt;,
+           &#39;cost&#39;: 0.014184397163120588,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78c5ada90&gt;,
+           &#39;ensemble_weight&#39;: 0.04,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78a956520&gt;,
+           &#39;model_id&#39;: 7,
+           &#39;rank&#39;: 3,
+           &#39;sklearn_classifier&#39;: ExtraTreesClassifier(max_features=34, min_samples_leaf=3, min_samples_split=11,
+                     n_estimators=512, n_jobs=1, random_state=1,
+                     warm_start=True)}}
 auto-sklearn results:
   Dataset name: breast_cancer
   Metric: accuracy
-  Best validation score: 0.978723
-  Number of target algorithm runs: 14
-  Number of successful target algorithm runs: 14
+  Best validation score: 0.985816
+  Number of target algorithm runs: 17
+  Number of successful target algorithm runs: 17
   Number of crashed target algorithm runs: 0
   Number of target algorithms that exceeded the time limit: 0
   Number of target algorithms that exceeded the memory limit: 0
 
-Accuracy score 0.951048951048951
+Accuracy score 0.958041958041958
 </pre></div>
 </div>
-</section>
-<section id="mixed-budget-approach">
+</div>
+<div class="section" id="mixed-budget-approach">
 <h2>Mixed budget approach<a class="headerlink" href="#mixed-budget-approach" title="Permalink to this headline">¶</a></h2>
 <p>Finally, there’s a mixed budget type which uses iterations where possible and
 subsamples otherwise</p>
@@ -776,7 +1083,7 @@ <h2>Mixed budget approach<a class="headerlink" href="#mixed-budget-approach" tit
 <span class="n">automl</span><span class="o">.</span><span class="n">fit</span><span class="p">(</span><span class="n">X_train</span><span class="p">,</span> <span class="n">y_train</span><span class="p">,</span> <span class="n">dataset_name</span><span class="o">=</span><span class="s1">&#39;breast_cancer&#39;</span><span class="p">)</span>
 
 <span class="c1"># Print the final ensemble constructed by auto-sklearn.</span>
-<span class="nb">print</span><span class="p">(</span><span class="n">automl</span><span class="o">.</span><span class="n">show_models</span><span class="p">())</span>
+<span class="n">pprint</span><span class="p">(</span><span class="n">automl</span><span class="o">.</span><span class="n">show_models</span><span class="p">(),</span> <span class="n">indent</span><span class="o">=</span><span class="mi">4</span><span class="p">)</span>
 <span class="n">predictions</span> <span class="o">=</span> <span class="n">automl</span><span class="o">.</span><span class="n">predict</span><span class="p">(</span><span class="n">X_test</span><span class="p">)</span>
 <span class="c1"># Print statistics about the auto-sklearn run such as number of</span>
 <span class="c1"># iterations, number of models failed with a time out.</span>
@@ -785,111 +1092,112 @@ <h2>Mixed budget approach<a class="headerlink" href="#mixed-budget-approach" tit
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/opt/hostedtoolcache/Python/3.8.12/x64/lib/python3.8/site-packages/smac/intensification/parallel_scheduling.py:152: UserWarning: SuccessiveHalving is intended to be used with more than 1 worker but num_workers=1
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:68: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+  self.metafeatures = self.metafeatures.append(metafeatures)
+/home/runner/work/auto-sklearn/auto-sklearn/autosklearn/metalearning/metalearning/meta_base.py:72: FutureWarning: The frame.append method is deprecated and will be removed from pandas in a future version. Use pandas.concat instead.
+  self.algorithm_runs[metric].append(runs)
+/opt/hostedtoolcache/Python/3.8.12/x64/lib/python3.8/site-packages/smac/intensification/parallel_scheduling.py:152: UserWarning: SuccessiveHalving is intended to be used with more than 1 worker but num_workers=1
   warnings.warn(&quot;{} is intended to be used &quot;
-[(0.280000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;gradient_boosting&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;select_percentile_classification&#39;, &#39;classifier:gradient_boosting:early_stop&#39;: &#39;train&#39;, &#39;classifier:gradient_boosting:l2_regularization&#39;: 3.387912939529945e-10, &#39;classifier:gradient_boosting:learning_rate&#39;: 0.30755227194768237, &#39;classifier:gradient_boosting:loss&#39;: &#39;auto&#39;, &#39;classifier:gradient_boosting:max_bins&#39;: 255, &#39;classifier:gradient_boosting:max_depth&#39;: &#39;None&#39;, &#39;classifier:gradient_boosting:max_leaf_nodes&#39;: 60, &#39;classifier:gradient_boosting:min_samples_leaf&#39;: 39, &#39;classifier:gradient_boosting:scoring&#39;: &#39;loss&#39;, &#39;classifier:gradient_boosting:tol&#39;: 1e-07, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;none&#39;, &#39;feature_preprocessor:select_percentile_classification:percentile&#39;: 93.39844669585806, &#39;feature_preprocessor:select_percentile_classification:score_func&#39;: &#39;f_classif&#39;, &#39;classifier:gradient_boosting:n_iter_no_change&#39;: 18, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.010000000000000004},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.140000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;extra_trees&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;polynomial&#39;, &#39;classifier:extra_trees:bootstrap&#39;: &#39;False&#39;, &#39;classifier:extra_trees:criterion&#39;: &#39;gini&#39;, &#39;classifier:extra_trees:max_depth&#39;: &#39;None&#39;, &#39;classifier:extra_trees:max_features&#39;: 0.5707983257382487, &#39;classifier:extra_trees:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:extra_trees:min_impurity_decrease&#39;: 0.0, &#39;classifier:extra_trees:min_samples_leaf&#39;: 3, &#39;classifier:extra_trees:min_samples_split&#39;: 11, &#39;classifier:extra_trees:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;median&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;none&#39;, &#39;feature_preprocessor:polynomial:degree&#39;: 2, &#39;feature_preprocessor:polynomial:include_bias&#39;: &#39;False&#39;, &#39;feature_preprocessor:polynomial:interaction_only&#39;: &#39;False&#39;},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.120000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;extra_trees&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;polynomial&#39;, &#39;classifier:extra_trees:bootstrap&#39;: &#39;False&#39;, &#39;classifier:extra_trees:criterion&#39;: &#39;gini&#39;, &#39;classifier:extra_trees:max_depth&#39;: &#39;None&#39;, &#39;classifier:extra_trees:max_features&#39;: 0.5707983257382487, &#39;classifier:extra_trees:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:extra_trees:min_impurity_decrease&#39;: 0.0, &#39;classifier:extra_trees:min_samples_leaf&#39;: 3, &#39;classifier:extra_trees:min_samples_split&#39;: 11, &#39;classifier:extra_trees:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;median&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;none&#39;, &#39;feature_preprocessor:polynomial:degree&#39;: 2, &#39;feature_preprocessor:polynomial:include_bias&#39;: &#39;False&#39;, &#39;feature_preprocessor:polynomial:interaction_only&#39;: &#39;False&#39;},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.100000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;extra_trees&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;polynomial&#39;, &#39;classifier:extra_trees:bootstrap&#39;: &#39;False&#39;, &#39;classifier:extra_trees:criterion&#39;: &#39;gini&#39;, &#39;classifier:extra_trees:max_depth&#39;: &#39;None&#39;, &#39;classifier:extra_trees:max_features&#39;: 0.5707983257382487, &#39;classifier:extra_trees:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:extra_trees:min_impurity_decrease&#39;: 0.0, &#39;classifier:extra_trees:min_samples_leaf&#39;: 3, &#39;classifier:extra_trees:min_samples_split&#39;: 11, &#39;classifier:extra_trees:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;median&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;none&#39;, &#39;feature_preprocessor:polynomial:degree&#39;: 2, &#39;feature_preprocessor:polynomial:include_bias&#39;: &#39;False&#39;, &#39;feature_preprocessor:polynomial:interaction_only&#39;: &#39;False&#39;},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.100000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;random_forest&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:random_forest:bootstrap&#39;: &#39;True&#39;, &#39;classifier:random_forest:criterion&#39;: &#39;gini&#39;, &#39;classifier:random_forest:max_depth&#39;: &#39;None&#39;, &#39;classifier:random_forest:max_features&#39;: 0.5, &#39;classifier:random_forest:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:random_forest:min_impurity_decrease&#39;: 0.0, &#39;classifier:random_forest:min_samples_leaf&#39;: 1, &#39;classifier:random_forest:min_samples_split&#39;: 2, &#39;classifier:random_forest:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;standardize&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.01},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.080000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;extra_trees&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;polynomial&#39;, &#39;classifier:extra_trees:bootstrap&#39;: &#39;False&#39;, &#39;classifier:extra_trees:criterion&#39;: &#39;gini&#39;, &#39;classifier:extra_trees:max_depth&#39;: &#39;None&#39;, &#39;classifier:extra_trees:max_features&#39;: 0.5707983257382487, &#39;classifier:extra_trees:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:extra_trees:min_impurity_decrease&#39;: 0.0, &#39;classifier:extra_trees:min_samples_leaf&#39;: 3, &#39;classifier:extra_trees:min_samples_split&#39;: 11, &#39;classifier:extra_trees:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;median&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;none&#39;, &#39;feature_preprocessor:polynomial:degree&#39;: 2, &#39;feature_preprocessor:polynomial:include_bias&#39;: &#39;False&#39;, &#39;feature_preprocessor:polynomial:interaction_only&#39;: &#39;False&#39;},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.060000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;extra_trees&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;extra_trees_preproc_for_classification&#39;, &#39;classifier:extra_trees:bootstrap&#39;: &#39;False&#39;, &#39;classifier:extra_trees:criterion&#39;: &#39;gini&#39;, &#39;classifier:extra_trees:max_depth&#39;: &#39;None&#39;, &#39;classifier:extra_trees:max_features&#39;: 0.9292309396985746, &#39;classifier:extra_trees:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:extra_trees:min_impurity_decrease&#39;: 0.0, &#39;classifier:extra_trees:min_samples_leaf&#39;: 1, &#39;classifier:extra_trees:min_samples_split&#39;: 10, &#39;classifier:extra_trees:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;median&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;normalize&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:bootstrap&#39;: &#39;False&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:criterion&#39;: &#39;entropy&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:max_depth&#39;: &#39;None&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:max_features&#39;: 0.9929881254946676, &#39;feature_preprocessor:extra_trees_preproc_for_classification:max_leaf_nodes&#39;: &#39;None&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_impurity_decrease&#39;: 0.0, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_samples_leaf&#39;: 1, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_samples_split&#39;: 2, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_weight_fraction_leaf&#39;: 0.0, &#39;feature_preprocessor:extra_trees_preproc_for_classification:n_estimators&#39;: 100, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.032719158639429445},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.040000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;gradient_boosting&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;liblinear_svc_preprocessor&#39;, &#39;classifier:gradient_boosting:early_stop&#39;: &#39;train&#39;, &#39;classifier:gradient_boosting:l2_regularization&#39;: 0.005326508887463406, &#39;classifier:gradient_boosting:learning_rate&#39;: 0.060800813211425456, &#39;classifier:gradient_boosting:loss&#39;: &#39;auto&#39;, &#39;classifier:gradient_boosting:max_bins&#39;: 255, &#39;classifier:gradient_boosting:max_depth&#39;: &#39;None&#39;, &#39;classifier:gradient_boosting:max_leaf_nodes&#39;: 6, &#39;classifier:gradient_boosting:min_samples_leaf&#39;: 5, &#39;classifier:gradient_boosting:scoring&#39;: &#39;loss&#39;, &#39;classifier:gradient_boosting:tol&#39;: 1e-07, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;median&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;minmax&#39;, &#39;feature_preprocessor:liblinear_svc_preprocessor:C&#39;: 13.550960330919455, &#39;feature_preprocessor:liblinear_svc_preprocessor:dual&#39;: &#39;False&#39;, &#39;feature_preprocessor:liblinear_svc_preprocessor:fit_intercept&#39;: &#39;True&#39;, &#39;feature_preprocessor:liblinear_svc_preprocessor:intercept_scaling&#39;: 1, &#39;feature_preprocessor:liblinear_svc_preprocessor:loss&#39;: &#39;squared_hinge&#39;, &#39;feature_preprocessor:liblinear_svc_preprocessor:multi_class&#39;: &#39;ovr&#39;, &#39;feature_preprocessor:liblinear_svc_preprocessor:penalty&#39;: &#39;l1&#39;, &#39;feature_preprocessor:liblinear_svc_preprocessor:tol&#39;: 1.2958033930435781e-05, &#39;classifier:gradient_boosting:n_iter_no_change&#39;: 5},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.020000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;extra_trees&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;select_rates_classification&#39;, &#39;classifier:extra_trees:bootstrap&#39;: &#39;False&#39;, &#39;classifier:extra_trees:criterion&#39;: &#39;entropy&#39;, &#39;classifier:extra_trees:max_depth&#39;: &#39;None&#39;, &#39;classifier:extra_trees:max_features&#39;: 0.562561668029056, &#39;classifier:extra_trees:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:extra_trees:min_impurity_decrease&#39;: 0.0, &#39;classifier:extra_trees:min_samples_leaf&#39;: 2, &#39;classifier:extra_trees:min_samples_split&#39;: 15, &#39;classifier:extra_trees:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;median&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;none&#39;, &#39;feature_preprocessor:select_rates_classification:alpha&#39;: 0.32106218519214325, &#39;feature_preprocessor:select_rates_classification:score_func&#39;: &#39;chi2&#39;, &#39;feature_preprocessor:select_rates_classification:mode&#39;: &#39;fwe&#39;},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.020000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;extra_trees&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;polynomial&#39;, &#39;classifier:extra_trees:bootstrap&#39;: &#39;False&#39;, &#39;classifier:extra_trees:criterion&#39;: &#39;entropy&#39;, &#39;classifier:extra_trees:max_depth&#39;: &#39;None&#39;, &#39;classifier:extra_trees:max_features&#39;: 0.993803313878608, &#39;classifier:extra_trees:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:extra_trees:min_impurity_decrease&#39;: 0.0, &#39;classifier:extra_trees:min_samples_leaf&#39;: 2, &#39;classifier:extra_trees:min_samples_split&#39;: 20, &#39;classifier:extra_trees:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;median&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;robust_scaler&#39;, &#39;feature_preprocessor:polynomial:degree&#39;: 2, &#39;feature_preprocessor:polynomial:include_bias&#39;: &#39;True&#39;, &#39;feature_preprocessor:polynomial:interaction_only&#39;: &#39;True&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.41826215858914706, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max&#39;: 0.7305615609807856, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min&#39;: 0.25595970768123566},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.020000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;gradient_boosting&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;extra_trees_preproc_for_classification&#39;, &#39;classifier:gradient_boosting:early_stop&#39;: &#39;valid&#39;, &#39;classifier:gradient_boosting:l2_regularization&#39;: 8.057778875694463e-05, &#39;classifier:gradient_boosting:learning_rate&#39;: 0.09179220974965213, &#39;classifier:gradient_boosting:loss&#39;: &#39;auto&#39;, &#39;classifier:gradient_boosting:max_bins&#39;: 255, &#39;classifier:gradient_boosting:max_depth&#39;: &#39;None&#39;, &#39;classifier:gradient_boosting:max_leaf_nodes&#39;: 200, &#39;classifier:gradient_boosting:min_samples_leaf&#39;: 20, &#39;classifier:gradient_boosting:scoring&#39;: &#39;loss&#39;, &#39;classifier:gradient_boosting:tol&#39;: 1e-07, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;most_frequent&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;quantile_transformer&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:bootstrap&#39;: &#39;False&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:criterion&#39;: &#39;gini&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:max_depth&#39;: &#39;None&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:max_features&#39;: 0.9984367650965825, &#39;feature_preprocessor:extra_trees_preproc_for_classification:max_leaf_nodes&#39;: &#39;None&#39;, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_impurity_decrease&#39;: 0.0, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_samples_leaf&#39;: 13, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_samples_split&#39;: 18, &#39;feature_preprocessor:extra_trees_preproc_for_classification:min_weight_fraction_leaf&#39;: 0.0, &#39;feature_preprocessor:extra_trees_preproc_for_classification:n_estimators&#39;: 100, &#39;classifier:gradient_boosting:n_iter_no_change&#39;: 18, &#39;classifier:gradient_boosting:validation_fraction&#39;: 0.14295295806077554, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.008542188583124829, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles&#39;: 1102, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution&#39;: &#39;uniform&#39;},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.020000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;random_forest&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;random_trees_embedding&#39;, &#39;classifier:random_forest:bootstrap&#39;: &#39;True&#39;, &#39;classifier:random_forest:criterion&#39;: &#39;entropy&#39;, &#39;classifier:random_forest:max_depth&#39;: &#39;None&#39;, &#39;classifier:random_forest:max_features&#39;: 0.5, &#39;classifier:random_forest:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:random_forest:min_impurity_decrease&#39;: 0.0, &#39;classifier:random_forest:min_samples_leaf&#39;: 1, &#39;classifier:random_forest:min_samples_split&#39;: 2, &#39;classifier:random_forest:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;no_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;no_coalescense&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;median&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;quantile_transformer&#39;, &#39;feature_preprocessor:random_trees_embedding:bootstrap&#39;: &#39;True&#39;, &#39;feature_preprocessor:random_trees_embedding:max_depth&#39;: 5, &#39;feature_preprocessor:random_trees_embedding:max_leaf_nodes&#39;: &#39;None&#39;, &#39;feature_preprocessor:random_trees_embedding:min_samples_leaf&#39;: 1, &#39;feature_preprocessor:random_trees_embedding:min_samples_split&#39;: 2, &#39;feature_preprocessor:random_trees_embedding:min_weight_fraction_leaf&#39;: 1.0, &#39;feature_preprocessor:random_trees_embedding:n_estimators&#39;: 10, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:n_quantiles&#39;: 937, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:quantile_transformer:output_distribution&#39;: &#39;uniform&#39;},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-]
+{   2: {   &#39;balancing&#39;: Balancing(random_state=1),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78fc1f700&gt;,
+           &#39;cost&#39;: 0.021276595744680882,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78a6050d0&gt;,
+           &#39;ensemble_weight&#39;: 0.06,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c944df0&gt;,
+           &#39;model_id&#39;: 2,
+           &#39;rank&#39;: 5,
+           &#39;sklearn_classifier&#39;: RandomForestClassifier(max_features=5, n_estimators=64, n_jobs=1,
+                       random_state=1, warm_start=True)},
+    4: {   &#39;balancing&#39;: Balancing(random_state=1),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc784bf13a0&gt;,
+           &#39;cost&#39;: 0.014184397163120588,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc79068bb80&gt;,
+           &#39;ensemble_weight&#39;: 0.14,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc784bf10a0&gt;,
+           &#39;model_id&#39;: 4,
+           &#39;rank&#39;: 4,
+           &#39;sklearn_classifier&#39;: ExtraTreesClassifier(max_features=34, min_samples_leaf=3, min_samples_split=11,
+                     n_estimators=512, n_jobs=1, random_state=1,
+                     warm_start=True)},
+    6: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc784871e50&gt;,
+           &#39;cost&#39;: 0.04255319148936165,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78580dac0&gt;,
+           &#39;ensemble_weight&#39;: 0.12,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c0ad400&gt;,
+           &#39;model_id&#39;: 6,
+           &#39;rank&#39;: 10,
+           &#39;sklearn_classifier&#39;: ExtraTreesClassifier(max_features=9, min_samples_split=10, n_estimators=64,
+                     n_jobs=1, random_state=1, warm_start=True)},
+    11: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc787e35400&gt;,
+            &#39;cost&#39;: 0.021276595744680882,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78af3f1c0&gt;,
+            &#39;ensemble_weight&#39;: 0.08,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc787e35280&gt;,
+            &#39;model_id&#39;: 11,
+            &#39;rank&#39;: 7,
+            &#39;sklearn_classifier&#39;: HistGradientBoostingClassifier(early_stopping=True,
+                               l2_regularization=3.387912939529945e-10,
+                               learning_rate=0.30755227194768237, max_iter=128,
+                               max_leaf_nodes=60, min_samples_leaf=39,
+                               n_iter_no_change=18, random_state=1,
+                               validation_fraction=None, warm_start=True)},
+    13: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc79039a790&gt;,
+            &#39;cost&#39;: 0.04255319148936165,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78a4bed00&gt;,
+            &#39;ensemble_weight&#39;: 0.02,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78f8a2b80&gt;,
+            &#39;model_id&#39;: 13,
+            &#39;rank&#39;: 11,
+            &#39;sklearn_classifier&#39;: RandomForestClassifier(criterion=&#39;entropy&#39;, max_features=3, n_estimators=64,
+                       n_jobs=1, random_state=1, warm_start=True)},
+    16: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78582ed30&gt;,
+            &#39;cost&#39;: 0.028368794326241176,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78a7e35b0&gt;,
+            &#39;ensemble_weight&#39;: 0.06,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c482ca0&gt;,
+            &#39;model_id&#39;: 16,
+            &#39;rank&#39;: 8,
+            &#39;sklearn_classifier&#39;: HistGradientBoostingClassifier(early_stopping=True,
+                               l2_regularization=8.057778875694463e-05,
+                               learning_rate=0.09179220974965213, max_iter=64,
+                               max_leaf_nodes=200, n_iter_no_change=18,
+                               random_state=1,
+                               validation_fraction=0.14295295806077554,
+                               warm_start=True)},
+    18: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc79039a070&gt;,
+            &#39;cost&#39;: 0.04255319148936165,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78bc2d7c0&gt;,
+            &#39;ensemble_weight&#39;: 0.02,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc79039a820&gt;,
+            &#39;model_id&#39;: 18,
+            &#39;rank&#39;: 12,
+            &#39;sklearn_classifier&#39;: ExtraTreesClassifier(criterion=&#39;entropy&#39;, max_features=414, min_samples_leaf=2,
+                     min_samples_split=19, n_estimators=64, n_jobs=1,
+                     random_state=1, warm_start=True)},
+    19: {   &#39;balancing&#39;: Balancing(random_state=1),
+            &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78ba373a0&gt;,
+            &#39;cost&#39;: 0.028368794326241176,
+            &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7850a8be0&gt;,
+            &#39;ensemble_weight&#39;: 0.02,
+            &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78582efa0&gt;,
+            &#39;model_id&#39;: 19,
+            &#39;rank&#39;: 9,
+            &#39;sklearn_classifier&#39;: HistGradientBoostingClassifier(early_stopping=False, l2_regularization=1e-10,
+                               learning_rate=0.16262682406125173, max_iter=64,
+                               max_leaf_nodes=66, n_iter_no_change=0,
+                               random_state=1, validation_fraction=None,
+                               warm_start=True)}}
 auto-sklearn results:
   Dataset name: breast_cancer
   Metric: accuracy
   Best validation score: 0.985816
-  Number of target algorithm runs: 19
-  Number of successful target algorithm runs: 19
+  Number of target algorithm runs: 24
+  Number of successful target algorithm runs: 24
   Number of crashed target algorithm runs: 0
   Number of target algorithms that exceeded the time limit: 0
   Number of target algorithms that exceeded the memory limit: 0
@@ -897,7 +1205,7 @@ <h2>Mixed budget approach<a class="headerlink" href="#mixed-budget-approach" tit
 Accuracy score 0.9440559440559441
 </pre></div>
 </div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 3 minutes  12.940 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes  59.204 seconds)</p>
 <div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-examples-60-search-example-successive-halving-py">
 <div class="binder-badge docutils container">
 <a class="reference external image-reference" href="https://mybinder.org/v2/gh/automl/auto-sklearn/master?urlpath=lab/tree/notebooks/examples/60_search/example_successive_halving.ipynb"><img alt="Launch binder" src="../../_images/binder_badge_logo2.svg" width="150px" /></a>
@@ -910,8 +1218,8 @@ <h2>Mixed budget approach<a class="headerlink" href="#mixed-budget-approach" tit
 </div>
 </div>
 <p class="sphx-glr-signature"><a class="reference external" href="https://sphinx-gallery.github.io">Gallery generated by Sphinx-Gallery</a></p>
-</section>
-</section>
+</div>
+</div>
 
 
     </div>
@@ -932,7 +1240,7 @@ <h2>Mixed budget approach<a class="headerlink" href="#mixed-budget-approach" tit
       
     </p>
     <p>
-        &copy; Copyright 2014-2021, Machine Learning Professorship Freiburg.<br/>
+        &copy; Copyright 2014-2022, Machine Learning Professorship Freiburg.<br/>
       Created using <a href="http://sphinx-doc.org/">Sphinx</a> 4.2.0.<br/>
     </p>
   </div>
diff --git a/master/examples/60_search/sg_execution_times.html b/master/examples/60_search/sg_execution_times.html
index 33b1d0ae10..cb18ebb0ba 100644
--- a/master/examples/60_search/sg_execution_times.html
+++ b/master/examples/60_search/sg_execution_times.html
@@ -3,9 +3,8 @@
 <html>
   <head>
     <meta charset="utf-8" />
-    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="generator" content="Docutils 0.17.1: http://docutils.sourceforge.net/" />
-
-    <title>Computation times &#8212; AutoSklearn 0.14.2 documentation</title>
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>Computation times &#8212; AutoSklearn 0.14.4 documentation</title>
     <link rel="stylesheet" type="text/css" href="../../_static/pygments.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/bootstrap-sphinx.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/sg_gallery.css" />
@@ -55,7 +54,7 @@
         </button>
         <a class="navbar-brand" href="../../index.html">
           auto-sklearn</a>
-        <span class="navbar-text navbar-version pull-left"><b>0.14.2</b></span>
+        <span class="navbar-text navbar-version pull-left"><b>0.14.4</b></span>
       </div>
 
         <div class="collapse navbar-collapse nav-collapse">
@@ -114,43 +113,39 @@
       </div>
     <div class="body col-md-9 content" role="main">
       
-  <section id="computation-times">
+  <div class="section" id="computation-times">
 <span id="sphx-glr-examples-60-search-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>10:48.731</strong> total execution time for <strong>examples_60_search</strong> files:</p>
+<p><strong>09:31.567</strong> total execution time for <strong>examples_60_search</strong> files:</p>
 <table class="docutils align-default">
 <colgroup>
 <col style="width: 87%" />
-<col style="width: 7%" />
-<col style="width: 5%" />
+<col style="width: 8%" />
+<col style="width: 6%" />
 </colgroup>
 <tbody>
 <tr class="row-odd"><td><p><a class="reference internal" href="example_successive_halving.html#sphx-glr-examples-60-search-example-successive-halving-py"><span class="std std-ref">Successive Halving</span></a> (<code class="docutils literal notranslate"><span class="pre">example_successive_halving.py</span></code>)</p></td>
-<td><p>03:12.940</p></td>
+<td><p>02:59.204</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
-<tr class="row-even"><td><p><a class="reference internal" href="example_parallel_n_jobs.html#sphx-glr-examples-60-search-example-parallel-n-jobs-py"><span class="std std-ref">Parallel Usage  on a single machine</span></a> (<code class="docutils literal notranslate"><span class="pre">example_parallel_n_jobs.py</span></code>)</p></td>
-<td><p>02:24.442</p></td>
+<tr class="row-even"><td><p><a class="reference internal" href="example_sequential.html#sphx-glr-examples-60-search-example-sequential-py"><span class="std std-ref">Sequential Usage</span></a> (<code class="docutils literal notranslate"><span class="pre">example_sequential.py</span></code>)</p></td>
+<td><p>02:01.878</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
-<tr class="row-odd"><td><p><a class="reference internal" href="example_sequential.html#sphx-glr-examples-60-search-example-sequential-py"><span class="std std-ref">Sequential Usage</span></a> (<code class="docutils literal notranslate"><span class="pre">example_sequential.py</span></code>)</p></td>
-<td><p>02:04.152</p></td>
+<tr class="row-odd"><td><p><a class="reference internal" href="example_parallel_n_jobs.html#sphx-glr-examples-60-search-example-parallel-n-jobs-py"><span class="std std-ref">Parallel Usage  on a single machine</span></a> (<code class="docutils literal notranslate"><span class="pre">example_parallel_n_jobs.py</span></code>)</p></td>
+<td><p>01:57.601</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="example_random_search.html#sphx-glr-examples-60-search-example-random-search-py"><span class="std std-ref">Random Search</span></a> (<code class="docutils literal notranslate"><span class="pre">example_random_search.py</span></code>)</p></td>
-<td><p>01:52.038</p></td>
+<td><p>01:53.718</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-odd"><td><p><a class="reference internal" href="example_parallel_manual_spawning_cli.html#sphx-glr-examples-60-search-example-parallel-manual-spawning-cli-py"><span class="std std-ref">Parallel Usage: Spawning workers from the command line</span></a> (<code class="docutils literal notranslate"><span class="pre">example_parallel_manual_spawning_cli.py</span></code>)</p></td>
-<td><p>00:39.363</p></td>
-<td><p>0.0 MB</p></td>
-</tr>
-<tr class="row-even"><td><p><a class="reference internal" href="example_parallel_manual_spawning_python.html#sphx-glr-examples-60-search-example-parallel-manual-spawning-python-py"><span class="std std-ref">Parallel Usage: Spawning workers from within Python</span></a> (<code class="docutils literal notranslate"><span class="pre">example_parallel_manual_spawning_python.py</span></code>)</p></td>
-<td><p>00:35.796</p></td>
+<td><p>00:39.167</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 </tbody>
 </table>
-</section>
+</div>
 
 
     </div>
@@ -171,7 +166,7 @@
       
     </p>
     <p>
-        &copy; Copyright 2014-2021, Machine Learning Professorship Freiburg.<br/>
+        &copy; Copyright 2014-2022, Machine Learning Professorship Freiburg.<br/>
       Created using <a href="http://sphinx-doc.org/">Sphinx</a> 4.2.0.<br/>
     </p>
   </div>
diff --git a/master/examples/80_extending/example_extending_classification.html b/master/examples/80_extending/example_extending_classification.html
index cad357c9a6..f92646dbb6 100644
--- a/master/examples/80_extending/example_extending_classification.html
+++ b/master/examples/80_extending/example_extending_classification.html
@@ -3,9 +3,8 @@
 <html>
   <head>
     <meta charset="utf-8" />
-    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="generator" content="Docutils 0.17.1: http://docutils.sourceforge.net/" />
-
-    <title>Extending Auto-Sklearn with Classification Component &#8212; AutoSklearn 0.14.2 documentation</title>
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>Extending Auto-Sklearn with Classification Component &#8212; AutoSklearn 0.14.4 documentation</title>
     <link rel="stylesheet" type="text/css" href="../../_static/pygments.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/bootstrap-sphinx.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/sg_gallery.css" />
@@ -55,7 +54,7 @@
         </button>
         <a class="navbar-brand" href="../../index.html">
           auto-sklearn</a>
-        <span class="navbar-text navbar-version pull-left"><b>0.14.2</b></span>
+        <span class="navbar-text navbar-version pull-left"><b>0.14.4</b></span>
       </div>
 
         <div class="collapse navbar-collapse nav-collapse">
@@ -125,11 +124,13 @@
 <p>Click <a class="reference internal" href="#sphx-glr-download-examples-80-extending-example-extending-classification-py"><span class="std std-ref">here</span></a>
 to download the full example code or to run this example in your browser via Binder</p>
 </div>
-<section class="sphx-glr-example-title" id="extending-auto-sklearn-with-classification-component">
+<div class="sphx-glr-example-title section" id="extending-auto-sklearn-with-classification-component">
 <span id="sphx-glr-examples-80-extending-example-extending-classification-py"></span><h1>Extending Auto-Sklearn with Classification Component<a class="headerlink" href="#extending-auto-sklearn-with-classification-component" title="Permalink to this headline">¶</a></h1>
 <p>The following example demonstrates how to create a new classification
 component for using in auto-sklearn.</p>
-<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">ConfigSpace.configuration_space</span> <span class="kn">import</span> <span class="n">ConfigurationSpace</span>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">pprint</span> <span class="kn">import</span> <span class="n">pprint</span>
+
+<span class="kn">from</span> <span class="nn">ConfigSpace.configuration_space</span> <span class="kn">import</span> <span class="n">ConfigurationSpace</span>
 <span class="kn">from</span> <span class="nn">ConfigSpace.hyperparameters</span> <span class="kn">import</span> <span class="n">CategoricalHyperparameter</span><span class="p">,</span> \
     <span class="n">UniformIntegerHyperparameter</span><span class="p">,</span> <span class="n">UniformFloatHyperparameter</span>
 
@@ -145,7 +146,7 @@
 <span class="kn">from</span> <span class="nn">sklearn.model_selection</span> <span class="kn">import</span> <span class="n">train_test_split</span>
 </pre></div>
 </div>
-<section id="create-mlp-classifier-component-for-auto-sklearn">
+<div class="section" id="create-mlp-classifier-component-for-auto-sklearn">
 <h2>Create MLP classifier component for auto-sklearn<a class="headerlink" href="#create-mlp-classifier-component-for-auto-sklearn" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="k">class</span> <span class="nc">MLPClassifier</span><span class="p">(</span><span class="n">AutoSklearnClassificationAlgorithm</span><span class="p">):</span>
 
@@ -250,15 +251,15 @@ <h2>Create MLP classifier component for auto-sklearn<a class="headerlink" href="
     solver, Type: Categorical, Choices: {lbfgs, sgd, adam}, Default: adam
 </pre></div>
 </div>
-</section>
-<section id="data-loading">
+</div>
+<div class="section" id="data-loading">
 <h2>Data Loading<a class="headerlink" href="#data-loading" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">X</span><span class="p">,</span> <span class="n">y</span> <span class="o">=</span> <span class="n">load_breast_cancer</span><span class="p">(</span><span class="n">return_X_y</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
 <span class="n">X_train</span><span class="p">,</span> <span class="n">X_test</span><span class="p">,</span> <span class="n">y_train</span><span class="p">,</span> <span class="n">y_test</span> <span class="o">=</span> <span class="n">train_test_split</span><span class="p">(</span><span class="n">X</span><span class="p">,</span> <span class="n">y</span><span class="p">)</span>
 </pre></div>
 </div>
-</section>
-<section id="fit-mlp-classifier-to-the-data">
+</div>
+<div class="section" id="fit-mlp-classifier-to-the-data">
 <h2>Fit MLP classifier to the data<a class="headerlink" href="#fit-mlp-classifier-to-the-data" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">clf</span> <span class="o">=</span> <span class="n">autosklearn</span><span class="o">.</span><span class="n">classification</span><span class="o">.</span><span class="n">AutoSklearnClassifier</span><span class="p">(</span>
     <span class="n">time_left_for_this_task</span><span class="o">=</span><span class="mi">30</span><span class="p">,</span>
@@ -282,36 +283,38 @@ <h2>Fit MLP classifier to the data<a class="headerlink" href="#fit-mlp-classifie
                       time_left_for_this_task=30)
 </pre></div>
 </div>
-</section>
-<section id="print-test-accuracy-and-statistics">
+</div>
+<div class="section" id="print-test-accuracy-and-statistics">
 <h2>Print test accuracy and statistics<a class="headerlink" href="#print-test-accuracy-and-statistics" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">y_pred</span> <span class="o">=</span> <span class="n">clf</span><span class="o">.</span><span class="n">predict</span><span class="p">(</span><span class="n">X_test</span><span class="p">)</span>
 <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;accuracy: &quot;</span><span class="p">,</span> <span class="n">sklearn</span><span class="o">.</span><span class="n">metrics</span><span class="o">.</span><span class="n">accuracy_score</span><span class="p">(</span><span class="n">y_pred</span><span class="p">,</span> <span class="n">y_test</span><span class="p">))</span>
-<span class="nb">print</span><span class="p">(</span><span class="n">clf</span><span class="o">.</span><span class="n">show_models</span><span class="p">())</span>
+<span class="n">pprint</span><span class="p">(</span><span class="n">clf</span><span class="o">.</span><span class="n">show_models</span><span class="p">(),</span> <span class="n">indent</span><span class="o">=</span><span class="mi">4</span><span class="p">)</span>
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>accuracy:  0.951048951048951
-[(0.580000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;MLPClassifier&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:MLPClassifier:activation&#39;: &#39;relu&#39;, &#39;classifier:MLPClassifier:alpha&#39;: 0.0001, &#39;classifier:MLPClassifier:hidden_layer_depth&#39;: 1, &#39;classifier:MLPClassifier:num_nodes_per_layer&#39;: 32, &#39;classifier:MLPClassifier:solver&#39;: &#39;adam&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;standardize&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.01},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.420000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;MLPClassifier&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;select_rates_classification&#39;, &#39;classifier:MLPClassifier:activation&#39;: &#39;relu&#39;, &#39;classifier:MLPClassifier:alpha&#39;: 0.054648061621061846, &#39;classifier:MLPClassifier:hidden_layer_depth&#39;: 1, &#39;classifier:MLPClassifier:num_nodes_per_layer&#39;: 187, &#39;classifier:MLPClassifier:solver&#39;: &#39;adam&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;robust_scaler&#39;, &#39;feature_preprocessor:select_rates_classification:alpha&#39;: 0.17677801820619737, &#39;feature_preprocessor:select_rates_classification:score_func&#39;: &#39;mutual_info_classif&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.06846183238648647, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_max&#39;: 0.8465036896999549, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:robust_scaler:q_min&#39;: 0.14555017892884228},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-]
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>accuracy:  0.986013986013986
+{   2: {   &#39;balancing&#39;: Balancing(random_state=1),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78f92cc70&gt;,
+           &#39;cost&#39;: 0.049645390070921946,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc786854e20&gt;,
+           &#39;ensemble_weight&#39;: 0.52,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78a9eab80&gt;,
+           &#39;model_id&#39;: 2,
+           &#39;rank&#39;: 1,
+           &#39;sklearn_classifier&#39;: MLPClassifier(hidden_layer_sizes=(32,), random_state=1)},
+    5: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78b2896a0&gt;,
+           &#39;cost&#39;: 0.1063829787234043,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc786854430&gt;,
+           &#39;ensemble_weight&#39;: 0.48,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c93bf10&gt;,
+           &#39;model_id&#39;: 5,
+           &#39;rank&#39;: 2,
+           &#39;sklearn_classifier&#39;: MLPClassifier(alpha=0.054648061621061846, hidden_layer_sizes=(187,),
+              random_state=1)}}
 </pre></div>
 </div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 0 minutes  18.369 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 0 minutes  12.500 seconds)</p>
 <div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-examples-80-extending-example-extending-classification-py">
 <div class="binder-badge docutils container">
 <a class="reference external image-reference" href="https://mybinder.org/v2/gh/automl/auto-sklearn/master?urlpath=lab/tree/notebooks/examples/80_extending/example_extending_classification.ipynb"><img alt="Launch binder" src="../../_images/binder_badge_logo3.svg" width="150px" /></a>
@@ -324,8 +327,8 @@ <h2>Print test accuracy and statistics<a class="headerlink" href="#print-test-ac
 </div>
 </div>
 <p class="sphx-glr-signature"><a class="reference external" href="https://sphinx-gallery.github.io">Gallery generated by Sphinx-Gallery</a></p>
-</section>
-</section>
+</div>
+</div>
 
 
     </div>
@@ -346,7 +349,7 @@ <h2>Print test accuracy and statistics<a class="headerlink" href="#print-test-ac
       
     </p>
     <p>
-        &copy; Copyright 2014-2021, Machine Learning Professorship Freiburg.<br/>
+        &copy; Copyright 2014-2022, Machine Learning Professorship Freiburg.<br/>
       Created using <a href="http://sphinx-doc.org/">Sphinx</a> 4.2.0.<br/>
     </p>
   </div>
diff --git a/master/examples/80_extending/example_extending_data_preprocessor.html b/master/examples/80_extending/example_extending_data_preprocessor.html
index b9d44c3ef9..a5556c6d6a 100644
--- a/master/examples/80_extending/example_extending_data_preprocessor.html
+++ b/master/examples/80_extending/example_extending_data_preprocessor.html
@@ -3,9 +3,8 @@
 <html>
   <head>
     <meta charset="utf-8" />
-    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="generator" content="Docutils 0.17.1: http://docutils.sourceforge.net/" />
-
-    <title>Extending Auto-Sklearn with Data Preprocessor Component &#8212; AutoSklearn 0.14.2 documentation</title>
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>Extending Auto-Sklearn with Data Preprocessor Component &#8212; AutoSklearn 0.14.4 documentation</title>
     <link rel="stylesheet" type="text/css" href="../../_static/pygments.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/bootstrap-sphinx.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/sg_gallery.css" />
@@ -55,7 +54,7 @@
         </button>
         <a class="navbar-brand" href="../../index.html">
           auto-sklearn</a>
-        <span class="navbar-text navbar-version pull-left"><b>0.14.2</b></span>
+        <span class="navbar-text navbar-version pull-left"><b>0.14.4</b></span>
       </div>
 
         <div class="collapse navbar-collapse nav-collapse">
@@ -125,10 +124,12 @@
 <p>Click <a class="reference internal" href="#sphx-glr-download-examples-80-extending-example-extending-data-preprocessor-py"><span class="std std-ref">here</span></a>
 to download the full example code or to run this example in your browser via Binder</p>
 </div>
-<section class="sphx-glr-example-title" id="extending-auto-sklearn-with-data-preprocessor-component">
+<div class="sphx-glr-example-title section" id="extending-auto-sklearn-with-data-preprocessor-component">
 <span id="sphx-glr-examples-80-extending-example-extending-data-preprocessor-py"></span><h1>Extending Auto-Sklearn with Data Preprocessor Component<a class="headerlink" href="#extending-auto-sklearn-with-data-preprocessor-component" title="Permalink to this headline">¶</a></h1>
 <p>The following example demonstrates how to turn off data preprocessing step in auto-skearn.</p>
-<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">autosklearn.classification</span>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">pprint</span> <span class="kn">import</span> <span class="n">pprint</span>
+
+<span class="kn">import</span> <span class="nn">autosklearn.classification</span>
 <span class="kn">import</span> <span class="nn">autosklearn.pipeline.components.data_preprocessing</span>
 <span class="kn">import</span> <span class="nn">sklearn.metrics</span>
 <span class="kn">from</span> <span class="nn">ConfigSpace.configuration_space</span> <span class="kn">import</span> <span class="n">ConfigurationSpace</span>
@@ -138,7 +139,7 @@
 <span class="kn">from</span> <span class="nn">sklearn.model_selection</span> <span class="kn">import</span> <span class="n">train_test_split</span>
 </pre></div>
 </div>
-<section id="create-nopreprocessing-component-for-auto-sklearn">
+<div class="section" id="create-nopreprocessing-component-for-auto-sklearn">
 <h2>Create NoPreprocessing component for auto-sklearn<a class="headerlink" href="#create-nopreprocessing-component-for-auto-sklearn" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="k">class</span> <span class="nc">NoPreprocessing</span><span class="p">(</span><span class="n">AutoSklearnPreprocessingAlgorithm</span><span class="p">):</span>
 
@@ -178,15 +179,15 @@ <h2>Create NoPreprocessing component for auto-sklearn<a class="headerlink" href=
 <span class="n">autosklearn</span><span class="o">.</span><span class="n">pipeline</span><span class="o">.</span><span class="n">components</span><span class="o">.</span><span class="n">data_preprocessing</span><span class="o">.</span><span class="n">add_preprocessor</span><span class="p">(</span><span class="n">NoPreprocessing</span><span class="p">)</span>
 </pre></div>
 </div>
-</section>
-<section id="create-dataset">
+</div>
+<div class="section" id="create-dataset">
 <h2>Create dataset<a class="headerlink" href="#create-dataset" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">X</span><span class="p">,</span> <span class="n">y</span> <span class="o">=</span> <span class="n">load_breast_cancer</span><span class="p">(</span><span class="n">return_X_y</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
 <span class="n">X_train</span><span class="p">,</span> <span class="n">X_test</span><span class="p">,</span> <span class="n">y_train</span><span class="p">,</span> <span class="n">y_test</span> <span class="o">=</span> <span class="n">train_test_split</span><span class="p">(</span><span class="n">X</span><span class="p">,</span> <span class="n">y</span><span class="p">)</span>
 </pre></div>
 </div>
-</section>
-<section id="fit-the-model-without-performing-data-preprocessing">
+</div>
+<div class="section" id="fit-the-model-without-performing-data-preprocessing">
 <h2>Fit the model without performing data preprocessing<a class="headerlink" href="#fit-the-model-without-performing-data-preprocessing" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">clf</span> <span class="o">=</span> <span class="n">autosklearn</span><span class="o">.</span><span class="n">classification</span><span class="o">.</span><span class="n">AutoSklearnClassifier</span><span class="p">(</span>
     <span class="n">time_left_for_this_task</span><span class="o">=</span><span class="mi">120</span><span class="p">,</span>
@@ -207,54 +208,60 @@ <h2>Fit the model without performing data preprocessing<a class="headerlink" hre
 </div>
 <p class="sphx-glr-script-out">Out:</p>
 <div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>auto-sklearn results:
-  Dataset name: 2613be54-419a-11ec-8a32-1b4c952643c3
+  Dataset name: 3a1d2162-7dfe-11ec-875f-3137a7553b01
   Metric: accuracy
-  Best validation score: 0.936170
+  Best validation score: 0.964539
   Number of target algorithm runs: 5
-  Number of successful target algorithm runs: 5
-  Number of crashed target algorithm runs: 0
+  Number of successful target algorithm runs: 4
+  Number of crashed target algorithm runs: 1
   Number of target algorithms that exceeded the time limit: 0
   Number of target algorithms that exceeded the memory limit: 0
 </pre></div>
 </div>
-</section>
-<section id="print-prediction-score-and-statistics">
+</div>
+<div class="section" id="print-prediction-score-and-statistics">
 <h2>Print prediction score and statistics<a class="headerlink" href="#print-prediction-score-and-statistics" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">y_pred</span> <span class="o">=</span> <span class="n">clf</span><span class="o">.</span><span class="n">predict</span><span class="p">(</span><span class="n">X_test</span><span class="p">)</span>
 <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;accuracy: &quot;</span><span class="p">,</span> <span class="n">sklearn</span><span class="o">.</span><span class="n">metrics</span><span class="o">.</span><span class="n">accuracy_score</span><span class="p">(</span><span class="n">y_pred</span><span class="p">,</span> <span class="n">y_test</span><span class="p">))</span>
-<span class="nb">print</span><span class="p">(</span><span class="n">clf</span><span class="o">.</span><span class="n">show_models</span><span class="p">())</span>
+<span class="n">pprint</span><span class="p">(</span><span class="n">clf</span><span class="o">.</span><span class="n">show_models</span><span class="p">(),</span> <span class="n">indent</span><span class="o">=</span><span class="mi">4</span><span class="p">)</span>
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
 <div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>accuracy:  0.9370629370629371
-[(0.540000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;weighting&#39;, &#39;classifier:__choice__&#39;: &#39;passive_aggressive&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;NoPreprocessing&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;kernel_pca&#39;, &#39;classifier:passive_aggressive:C&#39;: 7.328568999403367e-05, &#39;classifier:passive_aggressive:average&#39;: &#39;False&#39;, &#39;classifier:passive_aggressive:fit_intercept&#39;: &#39;True&#39;, &#39;classifier:passive_aggressive:loss&#39;: &#39;hinge&#39;, &#39;classifier:passive_aggressive:tol&#39;: 0.02258105767523384, &#39;feature_preprocessor:kernel_pca:kernel&#39;: &#39;cosine&#39;, &#39;feature_preprocessor:kernel_pca:n_components&#39;: 503},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.280000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;random_forest&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;NoPreprocessing&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;classifier:random_forest:bootstrap&#39;: &#39;True&#39;, &#39;classifier:random_forest:criterion&#39;: &#39;gini&#39;, &#39;classifier:random_forest:max_depth&#39;: &#39;None&#39;, &#39;classifier:random_forest:max_features&#39;: 0.5, &#39;classifier:random_forest:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:random_forest:min_impurity_decrease&#39;: 0.0, &#39;classifier:random_forest:min_samples_leaf&#39;: 1, &#39;classifier:random_forest:min_samples_split&#39;: 2, &#39;classifier:random_forest:min_weight_fraction_leaf&#39;: 0.0},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-(0.180000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;lda&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;NoPreprocessing&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;select_rates_classification&#39;, &#39;classifier:lda:shrinkage&#39;: &#39;auto&#39;, &#39;classifier:lda:tol&#39;: 0.00010495766342828794, &#39;feature_preprocessor:select_rates_classification:alpha&#39;: 0.4886183647550176, &#39;feature_preprocessor:select_rates_classification:score_func&#39;: &#39;mutual_info_classif&#39;},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-]
+{   2: {   &#39;balancing&#39;: Balancing(random_state=1),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78f84daf0&gt;,
+           &#39;cost&#39;: 0.03546099290780147,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78fdee5e0&gt;,
+           &#39;ensemble_weight&#39;: 0.44,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78fdeec10&gt;,
+           &#39;model_id&#39;: 2,
+           &#39;rank&#39;: 1,
+           &#39;sklearn_classifier&#39;: RandomForestClassifier(max_features=5, n_estimators=512, n_jobs=1,
+                       random_state=1, warm_start=True)},
+    4: {   &#39;balancing&#39;: Balancing(random_state=1, strategy=&#39;weighting&#39;),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78c4f9700&gt;,
+           &#39;cost&#39;: 0.36170212765957444,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7867d8340&gt;,
+           &#39;ensemble_weight&#39;: 0.4,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c59d910&gt;,
+           &#39;model_id&#39;: 4,
+           &#39;rank&#39;: 3,
+           &#39;sklearn_classifier&#39;: PassiveAggressiveClassifier(C=7.328568999403367e-05, max_iter=16,
+                            random_state=1, tol=0.02258105767523384,
+                            warm_start=True)},
+    6: {   &#39;balancing&#39;: Balancing(random_state=1),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc79015a340&gt;,
+           &#39;cost&#39;: 0.11347517730496459,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78f84d070&gt;,
+           &#39;ensemble_weight&#39;: 0.16,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c98fe20&gt;,
+           &#39;model_id&#39;: 6,
+           &#39;rank&#39;: 2,
+           &#39;sklearn_classifier&#39;: LinearDiscriminantAnalysis(shrinkage=&#39;auto&#39;, solver=&#39;lsqr&#39;,
+                           tol=0.00010495766342828794)}}
 </pre></div>
 </div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 0 minutes  23.312 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 0 minutes  15.803 seconds)</p>
 <div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-examples-80-extending-example-extending-data-preprocessor-py">
 <div class="binder-badge docutils container">
 <a class="reference external image-reference" href="https://mybinder.org/v2/gh/automl/auto-sklearn/master?urlpath=lab/tree/notebooks/examples/80_extending/example_extending_data_preprocessor.ipynb"><img alt="Launch binder" src="../../_images/binder_badge_logo3.svg" width="150px" /></a>
@@ -267,8 +274,8 @@ <h2>Print prediction score and statistics<a class="headerlink" href="#print-pred
 </div>
 </div>
 <p class="sphx-glr-signature"><a class="reference external" href="https://sphinx-gallery.github.io">Gallery generated by Sphinx-Gallery</a></p>
-</section>
-</section>
+</div>
+</div>
 
 
     </div>
@@ -289,7 +296,7 @@ <h2>Print prediction score and statistics<a class="headerlink" href="#print-pred
       
     </p>
     <p>
-        &copy; Copyright 2014-2021, Machine Learning Professorship Freiburg.<br/>
+        &copy; Copyright 2014-2022, Machine Learning Professorship Freiburg.<br/>
       Created using <a href="http://sphinx-doc.org/">Sphinx</a> 4.2.0.<br/>
     </p>
   </div>
diff --git a/master/examples/80_extending/example_extending_preprocessor.html b/master/examples/80_extending/example_extending_preprocessor.html
index 3c0abdd074..a7867bdbaa 100644
--- a/master/examples/80_extending/example_extending_preprocessor.html
+++ b/master/examples/80_extending/example_extending_preprocessor.html
@@ -3,9 +3,8 @@
 <html>
   <head>
     <meta charset="utf-8" />
-    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="generator" content="Docutils 0.17.1: http://docutils.sourceforge.net/" />
-
-    <title>Extending Auto-Sklearn with Preprocessor Component &#8212; AutoSklearn 0.14.2 documentation</title>
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>Extending Auto-Sklearn with Preprocessor Component &#8212; AutoSklearn 0.14.4 documentation</title>
     <link rel="stylesheet" type="text/css" href="../../_static/pygments.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/bootstrap-sphinx.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/sg_gallery.css" />
@@ -55,7 +54,7 @@
         </button>
         <a class="navbar-brand" href="../../index.html">
           auto-sklearn</a>
-        <span class="navbar-text navbar-version pull-left"><b>0.14.2</b></span>
+        <span class="navbar-text navbar-version pull-left"><b>0.14.4</b></span>
       </div>
 
         <div class="collapse navbar-collapse nav-collapse">
@@ -126,12 +125,14 @@
 <p>Click <a class="reference internal" href="#sphx-glr-download-examples-80-extending-example-extending-preprocessor-py"><span class="std std-ref">here</span></a>
 to download the full example code or to run this example in your browser via Binder</p>
 </div>
-<section class="sphx-glr-example-title" id="extending-auto-sklearn-with-preprocessor-component">
+<div class="sphx-glr-example-title section" id="extending-auto-sklearn-with-preprocessor-component">
 <span id="sphx-glr-examples-80-extending-example-extending-preprocessor-py"></span><h1>Extending Auto-Sklearn with Preprocessor Component<a class="headerlink" href="#extending-auto-sklearn-with-preprocessor-component" title="Permalink to this headline">¶</a></h1>
 <p>The following example demonstrates how to create a wrapper around the linear
 discriminant analysis (LDA) algorithm from sklearn and use it as a preprocessor
 in auto-sklearn.</p>
-<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">ConfigSpace.configuration_space</span> <span class="kn">import</span> <span class="n">ConfigurationSpace</span>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">pprint</span> <span class="kn">import</span> <span class="n">pprint</span>
+
+<span class="kn">from</span> <span class="nn">ConfigSpace.configuration_space</span> <span class="kn">import</span> <span class="n">ConfigurationSpace</span>
 <span class="kn">from</span> <span class="nn">ConfigSpace.hyperparameters</span> <span class="kn">import</span> <span class="n">UniformFloatHyperparameter</span><span class="p">,</span> <span class="n">CategoricalHyperparameter</span>
 <span class="kn">from</span> <span class="nn">ConfigSpace.conditions</span> <span class="kn">import</span> <span class="n">InCondition</span>
 
@@ -148,7 +149,7 @@
 <span class="kn">from</span> <span class="nn">sklearn.model_selection</span> <span class="kn">import</span> <span class="n">train_test_split</span>
 </pre></div>
 </div>
-<section id="create-lda-component-for-auto-sklearn">
+<div class="section" id="create-lda-component-for-auto-sklearn">
 <h2>Create LDA component for auto-sklearn<a class="headerlink" href="#create-lda-component-for-auto-sklearn" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="k">class</span> <span class="nc">LDA</span><span class="p">(</span><span class="n">AutoSklearnPreprocessingAlgorithm</span><span class="p">):</span>
 
@@ -217,15 +218,15 @@ <h2>Create LDA component for auto-sklearn<a class="headerlink" href="#create-lda
 <span class="n">autosklearn</span><span class="o">.</span><span class="n">pipeline</span><span class="o">.</span><span class="n">components</span><span class="o">.</span><span class="n">feature_preprocessing</span><span class="o">.</span><span class="n">add_preprocessor</span><span class="p">(</span><span class="n">LDA</span><span class="p">)</span>
 </pre></div>
 </div>
-</section>
-<section id="create-dataset">
+</div>
+<div class="section" id="create-dataset">
 <h2>Create dataset<a class="headerlink" href="#create-dataset" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">X</span><span class="p">,</span> <span class="n">y</span> <span class="o">=</span> <span class="n">load_breast_cancer</span><span class="p">(</span><span class="n">return_X_y</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
 <span class="n">X_train</span><span class="p">,</span> <span class="n">X_test</span><span class="p">,</span> <span class="n">y_train</span><span class="p">,</span> <span class="n">y_test</span> <span class="o">=</span> <span class="n">train_test_split</span><span class="p">(</span><span class="n">X</span><span class="p">,</span> <span class="n">y</span><span class="p">)</span>
 </pre></div>
 </div>
-</section>
-<section id="configuration-space">
+</div>
+<div class="section" id="configuration-space">
 <h2>Configuration space<a class="headerlink" href="#configuration-space" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">cs</span> <span class="o">=</span> <span class="n">LDA</span><span class="o">.</span><span class="n">get_hyperparameter_search_space</span><span class="p">()</span>
 <span class="nb">print</span><span class="p">(</span><span class="n">cs</span><span class="p">)</span>
@@ -241,8 +242,8 @@ <h2>Configuration space<a class="headerlink" href="#configuration-space" title="
     shrinkage | solver in {&#39;lsqr&#39;, &#39;eigen&#39;}
 </pre></div>
 </div>
-</section>
-<section id="fit-the-model-using-lda-as-preprocessor">
+</div>
+<div class="section" id="fit-the-model-using-lda-as-preprocessor">
 <h2>Fit the model using LDA as preprocessor<a class="headerlink" href="#fit-the-model-using-lda-as-preprocessor" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">clf</span> <span class="o">=</span> <span class="n">autosklearn</span><span class="o">.</span><span class="n">classification</span><span class="o">.</span><span class="n">AutoSklearnClassifier</span><span class="p">(</span>
     <span class="n">time_left_for_this_task</span><span class="o">=</span><span class="mi">30</span><span class="p">,</span>
@@ -265,28 +266,29 @@ <h2>Fit the model using LDA as preprocessor<a class="headerlink" href="#fit-the-
                       time_left_for_this_task=30)
 </pre></div>
 </div>
-</section>
-<section id="print-prediction-score-and-statistics">
+</div>
+<div class="section" id="print-prediction-score-and-statistics">
 <h2>Print prediction score and statistics<a class="headerlink" href="#print-prediction-score-and-statistics" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">y_pred</span> <span class="o">=</span> <span class="n">clf</span><span class="o">.</span><span class="n">predict</span><span class="p">(</span><span class="n">X_test</span><span class="p">)</span>
 <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;accuracy: &quot;</span><span class="p">,</span> <span class="n">sklearn</span><span class="o">.</span><span class="n">metrics</span><span class="o">.</span><span class="n">accuracy_score</span><span class="p">(</span><span class="n">y_pred</span><span class="p">,</span> <span class="n">y_test</span><span class="p">))</span>
-<span class="nb">print</span><span class="p">(</span><span class="n">clf</span><span class="o">.</span><span class="n">show_models</span><span class="p">())</span>
+<span class="n">pprint</span><span class="p">(</span><span class="n">clf</span><span class="o">.</span><span class="n">show_models</span><span class="p">(),</span> <span class="n">indent</span><span class="o">=</span><span class="mi">4</span><span class="p">)</span>
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>accuracy:  0.9370629370629371
-[(1.000000, SimpleClassificationPipeline({&#39;balancing:strategy&#39;: &#39;none&#39;, &#39;classifier:__choice__&#39;: &#39;random_forest&#39;, &#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;LDA&#39;, &#39;classifier:random_forest:bootstrap&#39;: &#39;True&#39;, &#39;classifier:random_forest:criterion&#39;: &#39;gini&#39;, &#39;classifier:random_forest:max_depth&#39;: &#39;None&#39;, &#39;classifier:random_forest:max_features&#39;: 0.5, &#39;classifier:random_forest:max_leaf_nodes&#39;: &#39;None&#39;, &#39;classifier:random_forest:min_impurity_decrease&#39;: 0.0, &#39;classifier:random_forest:min_samples_leaf&#39;: 1, &#39;classifier:random_forest:min_samples_split&#39;: 2, &#39;classifier:random_forest:min_weight_fraction_leaf&#39;: 0.0, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;standardize&#39;, &#39;feature_preprocessor:LDA:solver&#39;: &#39;svd&#39;, &#39;feature_preprocessor:LDA:tol&#39;: 0.0001, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.01},
-dataset_properties={
-  &#39;task&#39;: 1,
-  &#39;sparse&#39;: False,
-  &#39;multilabel&#39;: False,
-  &#39;multiclass&#39;: False,
-  &#39;target_type&#39;: &#39;classification&#39;,
-  &#39;signed&#39;: False})),
-]
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>accuracy:  0.958041958041958
+{   2: {   &#39;balancing&#39;: Balancing(random_state=1),
+           &#39;classifier&#39;: &lt;autosklearn.pipeline.components.classification.ClassifierChoice object at 0x7fc78c384d90&gt;,
+           &#39;cost&#39;: 0.021276595744680882,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc78af3f910&gt;,
+           &#39;ensemble_weight&#39;: 1.0,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78c8a0250&gt;,
+           &#39;model_id&#39;: 2,
+           &#39;rank&#39;: 1,
+           &#39;sklearn_classifier&#39;: RandomForestClassifier(max_features=1, n_estimators=512, n_jobs=1,
+                       random_state=1, warm_start=True)}}
 </pre></div>
 </div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 0 minutes  18.761 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 0 minutes  17.499 seconds)</p>
 <div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-examples-80-extending-example-extending-preprocessor-py">
 <div class="binder-badge docutils container">
 <a class="reference external image-reference" href="https://mybinder.org/v2/gh/automl/auto-sklearn/master?urlpath=lab/tree/notebooks/examples/80_extending/example_extending_preprocessor.ipynb"><img alt="Launch binder" src="../../_images/binder_badge_logo3.svg" width="150px" /></a>
@@ -299,8 +301,8 @@ <h2>Print prediction score and statistics<a class="headerlink" href="#print-pred
 </div>
 </div>
 <p class="sphx-glr-signature"><a class="reference external" href="https://sphinx-gallery.github.io">Gallery generated by Sphinx-Gallery</a></p>
-</section>
-</section>
+</div>
+</div>
 
 
     </div>
@@ -321,7 +323,7 @@ <h2>Print prediction score and statistics<a class="headerlink" href="#print-pred
       
     </p>
     <p>
-        &copy; Copyright 2014-2021, Machine Learning Professorship Freiburg.<br/>
+        &copy; Copyright 2014-2022, Machine Learning Professorship Freiburg.<br/>
       Created using <a href="http://sphinx-doc.org/">Sphinx</a> 4.2.0.<br/>
     </p>
   </div>
diff --git a/master/examples/80_extending/example_extending_regression.html b/master/examples/80_extending/example_extending_regression.html
index edf4f03938..93c6dde0cd 100644
--- a/master/examples/80_extending/example_extending_regression.html
+++ b/master/examples/80_extending/example_extending_regression.html
@@ -3,9 +3,8 @@
 <html>
   <head>
     <meta charset="utf-8" />
-    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="generator" content="Docutils 0.17.1: http://docutils.sourceforge.net/" />
-
-    <title>Extending Auto-Sklearn with Regression Component &#8212; AutoSklearn 0.14.2 documentation</title>
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>Extending Auto-Sklearn with Regression Component &#8212; AutoSklearn 0.14.4 documentation</title>
     <link rel="stylesheet" type="text/css" href="../../_static/pygments.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/bootstrap-sphinx.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/sg_gallery.css" />
@@ -55,7 +54,7 @@
         </button>
         <a class="navbar-brand" href="../../index.html">
           auto-sklearn</a>
-        <span class="navbar-text navbar-version pull-left"><b>0.14.2</b></span>
+        <span class="navbar-text navbar-version pull-left"><b>0.14.4</b></span>
       </div>
 
         <div class="collapse navbar-collapse nav-collapse">
@@ -125,11 +124,13 @@
 <p>Click <a class="reference internal" href="#sphx-glr-download-examples-80-extending-example-extending-regression-py"><span class="std std-ref">here</span></a>
 to download the full example code or to run this example in your browser via Binder</p>
 </div>
-<section class="sphx-glr-example-title" id="extending-auto-sklearn-with-regression-component">
+<div class="sphx-glr-example-title section" id="extending-auto-sklearn-with-regression-component">
 <span id="sphx-glr-examples-80-extending-example-extending-regression-py"></span><h1>Extending Auto-Sklearn with Regression Component<a class="headerlink" href="#extending-auto-sklearn-with-regression-component" title="Permalink to this headline">¶</a></h1>
 <p>The following example demonstrates how to create a new regression
 component for using in auto-sklearn.</p>
-<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">ConfigSpace.configuration_space</span> <span class="kn">import</span> <span class="n">ConfigurationSpace</span>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">pprint</span> <span class="kn">import</span> <span class="n">pprint</span>
+
+<span class="kn">from</span> <span class="nn">ConfigSpace.configuration_space</span> <span class="kn">import</span> <span class="n">ConfigurationSpace</span>
 <span class="kn">from</span> <span class="nn">ConfigSpace.hyperparameters</span> <span class="kn">import</span> <span class="n">UniformFloatHyperparameter</span><span class="p">,</span> \
     <span class="n">UniformIntegerHyperparameter</span><span class="p">,</span> <span class="n">CategoricalHyperparameter</span>
 <span class="kn">from</span> <span class="nn">ConfigSpace.conditions</span> <span class="kn">import</span> <span class="n">EqualsCondition</span>
@@ -145,7 +146,7 @@
 <span class="kn">from</span> <span class="nn">sklearn.model_selection</span> <span class="kn">import</span> <span class="n">train_test_split</span>
 </pre></div>
 </div>
-<section id="implement-kernel-ridge-regression-component-for-auto-sklearn">
+<div class="section" id="implement-kernel-ridge-regression-component-for-auto-sklearn">
 <h2>Implement kernel ridge regression component for auto-sklearn<a class="headerlink" href="#implement-kernel-ridge-regression-component-for-auto-sklearn" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="k">class</span> <span class="nc">KernelRidgeRegression</span><span class="p">(</span><span class="n">AutoSklearnRegressionAlgorithm</span><span class="p">):</span>
 
@@ -242,15 +243,15 @@ <h2>Implement kernel ridge regression component for auto-sklearn<a class="header
     degree | kernel == &#39;polynomial&#39;
 </pre></div>
 </div>
-</section>
-<section id="generate-data">
+</div>
+<div class="section" id="generate-data">
 <h2>Generate data<a class="headerlink" href="#generate-data" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">X</span><span class="p">,</span> <span class="n">y</span> <span class="o">=</span> <span class="n">load_diabetes</span><span class="p">(</span><span class="n">return_X_y</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
 <span class="n">X_train</span><span class="p">,</span> <span class="n">X_test</span><span class="p">,</span> <span class="n">y_train</span><span class="p">,</span> <span class="n">y_test</span> <span class="o">=</span> <span class="n">train_test_split</span><span class="p">(</span><span class="n">X</span><span class="p">,</span> <span class="n">y</span><span class="p">)</span>
 </pre></div>
 </div>
-</section>
-<section id="fit-the-model-using-krr">
+</div>
+<div class="section" id="fit-the-model-using-krr">
 <h2>Fit the model using KRR<a class="headerlink" href="#fit-the-model-using-krr" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">reg</span> <span class="o">=</span> <span class="n">autosklearn</span><span class="o">.</span><span class="n">regression</span><span class="o">.</span><span class="n">AutoSklearnRegressor</span><span class="p">(</span>
     <span class="n">time_left_for_this_task</span><span class="o">=</span><span class="mi">30</span><span class="p">,</span>
@@ -274,27 +275,27 @@ <h2>Fit the model using KRR<a class="headerlink" href="#fit-the-model-using-krr"
                      time_left_for_this_task=30)
 </pre></div>
 </div>
-</section>
-<section id="print-prediction-score-and-statistics">
+</div>
+<div class="section" id="print-prediction-score-and-statistics">
 <h2>Print prediction score and statistics<a class="headerlink" href="#print-prediction-score-and-statistics" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">y_pred</span> <span class="o">=</span> <span class="n">reg</span><span class="o">.</span><span class="n">predict</span><span class="p">(</span><span class="n">X_test</span><span class="p">)</span>
 <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;r2 score: &quot;</span><span class="p">,</span> <span class="n">sklearn</span><span class="o">.</span><span class="n">metrics</span><span class="o">.</span><span class="n">r2_score</span><span class="p">(</span><span class="n">y_pred</span><span class="p">,</span> <span class="n">y_test</span><span class="p">))</span>
-<span class="nb">print</span><span class="p">(</span><span class="n">reg</span><span class="o">.</span><span class="n">show_models</span><span class="p">())</span>
+<span class="n">pprint</span><span class="p">(</span><span class="n">reg</span><span class="o">.</span><span class="n">show_models</span><span class="p">(),</span> <span class="n">indent</span><span class="o">=</span><span class="mi">4</span><span class="p">)</span>
 </pre></div>
 </div>
 <p class="sphx-glr-script-out">Out:</p>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>r2 score:  0.1280839244790386
-[(1.000000, SimpleRegressionPipeline({&#39;data_preprocessor:__choice__&#39;: &#39;feature_type&#39;, &#39;feature_preprocessor:__choice__&#39;: &#39;no_preprocessing&#39;, &#39;regressor:__choice__&#39;: &#39;KernelRidgeRegression&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:categorical_encoding:__choice__&#39;: &#39;one_hot_encoding&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:__choice__&#39;: &#39;minority_coalescer&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:imputation:strategy&#39;: &#39;mean&#39;, &#39;data_preprocessor:feature_type:numerical_transformer:rescaling:__choice__&#39;: &#39;standardize&#39;, &#39;regressor:KernelRidgeRegression:alpha&#39;: 1.0, &#39;regressor:KernelRidgeRegression:gamma&#39;: 0.1, &#39;regressor:KernelRidgeRegression:kernel&#39;: &#39;polynomial&#39;, &#39;data_preprocessor:feature_type:categorical_transformer:category_coalescence:minority_coalescer:minimum_fraction&#39;: 0.01, &#39;regressor:KernelRidgeRegression:coef0&#39;: 1.0, &#39;regressor:KernelRidgeRegression:degree&#39;: 3},
-dataset_properties={
-  &#39;task&#39;: 4,
-  &#39;sparse&#39;: False,
-  &#39;multioutput&#39;: False,
-  &#39;target_type&#39;: &#39;regression&#39;,
-  &#39;signed&#39;: False})),
-]
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>r2 score:  0.11103398852619628
+{   2: {   &#39;cost&#39;: 0.65199313632313,
+           &#39;data_preprocessor&#39;: &lt;autosklearn.pipeline.components.data_preprocessing.DataPreprocessorChoice object at 0x7fc7863feee0&gt;,
+           &#39;ensemble_weight&#39;: 1.0,
+           &#39;feature_preprocessor&#39;: &lt;autosklearn.pipeline.components.feature_preprocessing.FeaturePreprocessorChoice object at 0x7fc78f6db9a0&gt;,
+           &#39;model_id&#39;: 2,
+           &#39;rank&#39;: 1,
+           &#39;regressor&#39;: &lt;autosklearn.pipeline.components.regression.RegressorChoice object at 0x7fc78a826580&gt;,
+           &#39;sklearn_regressor&#39;: KernelRidge(alpha=1.0, coef0=1.0, gamma=0.1, kernel=&#39;polynomial&#39;)}}
 </pre></div>
 </div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 0 minutes  11.876 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 0 minutes  11.558 seconds)</p>
 <div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-examples-80-extending-example-extending-regression-py">
 <div class="binder-badge docutils container">
 <a class="reference external image-reference" href="https://mybinder.org/v2/gh/automl/auto-sklearn/master?urlpath=lab/tree/notebooks/examples/80_extending/example_extending_regression.ipynb"><img alt="Launch binder" src="../../_images/binder_badge_logo3.svg" width="150px" /></a>
@@ -307,8 +308,8 @@ <h2>Print prediction score and statistics<a class="headerlink" href="#print-pred
 </div>
 </div>
 <p class="sphx-glr-signature"><a class="reference external" href="https://sphinx-gallery.github.io">Gallery generated by Sphinx-Gallery</a></p>
-</section>
-</section>
+</div>
+</div>
 
 
     </div>
@@ -329,7 +330,7 @@ <h2>Print prediction score and statistics<a class="headerlink" href="#print-pred
       
     </p>
     <p>
-        &copy; Copyright 2014-2021, Machine Learning Professorship Freiburg.<br/>
+        &copy; Copyright 2014-2022, Machine Learning Professorship Freiburg.<br/>
       Created using <a href="http://sphinx-doc.org/">Sphinx</a> 4.2.0.<br/>
     </p>
   </div>
diff --git a/master/examples/80_extending/example_restrict_number_of_hyperparameters.html b/master/examples/80_extending/example_restrict_number_of_hyperparameters.html
index 0ed4b7d95b..bdc8c6f752 100644
--- a/master/examples/80_extending/example_restrict_number_of_hyperparameters.html
+++ b/master/examples/80_extending/example_restrict_number_of_hyperparameters.html
@@ -3,9 +3,8 @@
 <html>
   <head>
     <meta charset="utf-8" />
-    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="generator" content="Docutils 0.17.1: http://docutils.sourceforge.net/" />
-
-    <title>Restricting the number of hyperparameters for an existing component &#8212; AutoSklearn 0.14.2 documentation</title>
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>Restricting the number of hyperparameters for an existing component &#8212; AutoSklearn 0.14.4 documentation</title>
     <link rel="stylesheet" type="text/css" href="../../_static/pygments.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/bootstrap-sphinx.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/sg_gallery.css" />
@@ -55,7 +54,7 @@
         </button>
         <a class="navbar-brand" href="../../index.html">
           auto-sklearn</a>
-        <span class="navbar-text navbar-version pull-left"><b>0.14.2</b></span>
+        <span class="navbar-text navbar-version pull-left"><b>0.14.4</b></span>
       </div>
 
         <div class="collapse navbar-collapse nav-collapse">
@@ -125,7 +124,7 @@
 <p>Click <a class="reference internal" href="#sphx-glr-download-examples-80-extending-example-restrict-number-of-hyperparameters-py"><span class="std std-ref">here</span></a>
 to download the full example code or to run this example in your browser via Binder</p>
 </div>
-<section class="sphx-glr-example-title" id="restricting-the-number-of-hyperparameters-for-an-existing-component">
+<div class="sphx-glr-example-title section" id="restricting-the-number-of-hyperparameters-for-an-existing-component">
 <span id="sphx-glr-examples-80-extending-example-restrict-number-of-hyperparameters-py"></span><h1>Restricting the number of hyperparameters for an existing component<a class="headerlink" href="#restricting-the-number-of-hyperparameters-for-an-existing-component" title="Permalink to this headline">¶</a></h1>
 <p>The following example demonstrates how to replace an existing
 component with a new component, implementing the same classifier,
@@ -143,7 +142,7 @@
 <span class="kn">from</span> <span class="nn">autosklearn.pipeline.constants</span> <span class="kn">import</span> <span class="n">DENSE</span><span class="p">,</span> <span class="n">UNSIGNED_DATA</span><span class="p">,</span> <span class="n">PREDICTIONS</span><span class="p">,</span> <span class="n">SPARSE</span>
 </pre></div>
 </div>
-<section id="subclass-auto-sklearn-s-random-forest-classifier">
+<div class="section" id="subclass-auto-sklearn-s-random-forest-classifier">
 <h2>Subclass auto-sklearn’s random forest classifier<a class="headerlink" href="#subclass-auto-sklearn-s-random-forest-classifier" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="c1"># This classifier only has one of the hyperparameter&#39;s of auto-sklearn&#39;s</span>
 <span class="c1"># default parametrization (``max_features``). Instead, it also</span>
@@ -227,15 +226,15 @@ <h2>Subclass auto-sklearn’s random forest classifier<a class="headerlink" href
     n_estimators, Type: UniformInteger, Range: [10, 1000], Default: 100
 </pre></div>
 </div>
-</section>
-<section id="data-loading">
+</div>
+<div class="section" id="data-loading">
 <h2>Data Loading<a class="headerlink" href="#data-loading" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">X</span><span class="p">,</span> <span class="n">y</span> <span class="o">=</span> <span class="n">load_breast_cancer</span><span class="p">(</span><span class="n">return_X_y</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
 <span class="n">X_train</span><span class="p">,</span> <span class="n">X_test</span><span class="p">,</span> <span class="n">y_train</span><span class="p">,</span> <span class="n">y_test</span> <span class="o">=</span> <span class="n">train_test_split</span><span class="p">(</span><span class="n">X</span><span class="p">,</span> <span class="n">y</span><span class="p">)</span>
 </pre></div>
 </div>
-</section>
-<section id="fit-random-forest-classifier-to-the-data">
+</div>
+<div class="section" id="fit-random-forest-classifier-to-the-data">
 <h2>Fit Random forest classifier to the data<a class="headerlink" href="#fit-random-forest-classifier-to-the-data" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="n">clf</span> <span class="o">=</span> <span class="n">autosklearn</span><span class="o">.</span><span class="n">classification</span><span class="o">.</span><span class="n">AutoSklearnClassifier</span><span class="p">(</span>
     <span class="n">time_left_for_this_task</span><span class="o">=</span><span class="mi">30</span><span class="p">,</span>
@@ -260,8 +259,8 @@ <h2>Fit Random forest classifier to the data<a class="headerlink" href="#fit-ran
                       time_left_for_this_task=30)
 </pre></div>
 </div>
-</section>
-<section id="print-the-configuration-space">
+</div>
+<div class="section" id="print-the-configuration-space">
 <h2>Print the configuration space<a class="headerlink" href="#print-the-configuration-space" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="c1"># Observe that this configuration space only contains our custom random</span>
 <span class="c1"># forest, but not auto-sklearn&#39;s ``random_forest``</span>
@@ -651,7 +650,7 @@ <h2>Print the configuration space<a class="headerlink" href="#print-the-configur
     (Forbidden: feature_preprocessor:__choice__ == &#39;nystroem_sampler&#39; &amp;&amp; Forbidden: classifier:__choice__ == &#39;multinomial_nb&#39;)
 </pre></div>
 </div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 0 minutes  7.799 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 0 minutes  6.778 seconds)</p>
 <div class="sphx-glr-footer class sphx-glr-footer-example docutils container" id="sphx-glr-download-examples-80-extending-example-restrict-number-of-hyperparameters-py">
 <div class="binder-badge docutils container">
 <a class="reference external image-reference" href="https://mybinder.org/v2/gh/automl/auto-sklearn/master?urlpath=lab/tree/notebooks/examples/80_extending/example_restrict_number_of_hyperparameters.ipynb"><img alt="Launch binder" src="../../_images/binder_badge_logo3.svg" width="150px" /></a>
@@ -664,8 +663,8 @@ <h2>Print the configuration space<a class="headerlink" href="#print-the-configur
 </div>
 </div>
 <p class="sphx-glr-signature"><a class="reference external" href="https://sphinx-gallery.github.io">Gallery generated by Sphinx-Gallery</a></p>
-</section>
-</section>
+</div>
+</div>
 
 
     </div>
@@ -686,7 +685,7 @@ <h2>Print the configuration space<a class="headerlink" href="#print-the-configur
       
     </p>
     <p>
-        &copy; Copyright 2014-2021, Machine Learning Professorship Freiburg.<br/>
+        &copy; Copyright 2014-2022, Machine Learning Professorship Freiburg.<br/>
       Created using <a href="http://sphinx-doc.org/">Sphinx</a> 4.2.0.<br/>
     </p>
   </div>
diff --git a/master/examples/80_extending/sg_execution_times.html b/master/examples/80_extending/sg_execution_times.html
index 7d316028a3..e15f5fed30 100644
--- a/master/examples/80_extending/sg_execution_times.html
+++ b/master/examples/80_extending/sg_execution_times.html
@@ -3,9 +3,8 @@
 <html>
   <head>
     <meta charset="utf-8" />
-    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="generator" content="Docutils 0.17.1: http://docutils.sourceforge.net/" />
-
-    <title>Computation times &#8212; AutoSklearn 0.14.2 documentation</title>
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>Computation times &#8212; AutoSklearn 0.14.4 documentation</title>
     <link rel="stylesheet" type="text/css" href="../../_static/pygments.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/bootstrap-sphinx.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/sg_gallery.css" />
@@ -55,7 +54,7 @@
         </button>
         <a class="navbar-brand" href="../../index.html">
           auto-sklearn</a>
-        <span class="navbar-text navbar-version pull-left"><b>0.14.2</b></span>
+        <span class="navbar-text navbar-version pull-left"><b>0.14.4</b></span>
       </div>
 
         <div class="collapse navbar-collapse nav-collapse">
@@ -114,9 +113,9 @@
       </div>
     <div class="body col-md-9 content" role="main">
       
-  <section id="computation-times">
+  <div class="section" id="computation-times">
 <span id="sphx-glr-examples-80-extending-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>01:20.118</strong> total execution time for <strong>examples_80_extending</strong> files:</p>
+<p><strong>01:04.137</strong> total execution time for <strong>examples_80_extending</strong> files:</p>
 <table class="docutils align-default">
 <colgroup>
 <col style="width: 88%" />
@@ -124,29 +123,29 @@
 <col style="width: 5%" />
 </colgroup>
 <tbody>
-<tr class="row-odd"><td><p><a class="reference internal" href="example_extending_data_preprocessor.html#sphx-glr-examples-80-extending-example-extending-data-preprocessor-py"><span class="std std-ref">Extending Auto-Sklearn with Data Preprocessor Component</span></a> (<code class="docutils literal notranslate"><span class="pre">example_extending_data_preprocessor.py</span></code>)</p></td>
-<td><p>00:23.312</p></td>
+<tr class="row-odd"><td><p><a class="reference internal" href="example_extending_preprocessor.html#sphx-glr-examples-80-extending-example-extending-preprocessor-py"><span class="std std-ref">Extending Auto-Sklearn with Preprocessor Component</span></a> (<code class="docutils literal notranslate"><span class="pre">example_extending_preprocessor.py</span></code>)</p></td>
+<td><p>00:17.499</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
-<tr class="row-even"><td><p><a class="reference internal" href="example_extending_preprocessor.html#sphx-glr-examples-80-extending-example-extending-preprocessor-py"><span class="std std-ref">Extending Auto-Sklearn with Preprocessor Component</span></a> (<code class="docutils literal notranslate"><span class="pre">example_extending_preprocessor.py</span></code>)</p></td>
-<td><p>00:18.761</p></td>
+<tr class="row-even"><td><p><a class="reference internal" href="example_extending_data_preprocessor.html#sphx-glr-examples-80-extending-example-extending-data-preprocessor-py"><span class="std std-ref">Extending Auto-Sklearn with Data Preprocessor Component</span></a> (<code class="docutils literal notranslate"><span class="pre">example_extending_data_preprocessor.py</span></code>)</p></td>
+<td><p>00:15.803</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-odd"><td><p><a class="reference internal" href="example_extending_classification.html#sphx-glr-examples-80-extending-example-extending-classification-py"><span class="std std-ref">Extending Auto-Sklearn with Classification Component</span></a> (<code class="docutils literal notranslate"><span class="pre">example_extending_classification.py</span></code>)</p></td>
-<td><p>00:18.369</p></td>
+<td><p>00:12.500</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="example_extending_regression.html#sphx-glr-examples-80-extending-example-extending-regression-py"><span class="std std-ref">Extending Auto-Sklearn with Regression Component</span></a> (<code class="docutils literal notranslate"><span class="pre">example_extending_regression.py</span></code>)</p></td>
-<td><p>00:11.876</p></td>
+<td><p>00:11.558</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-odd"><td><p><a class="reference internal" href="example_restrict_number_of_hyperparameters.html#sphx-glr-examples-80-extending-example-restrict-number-of-hyperparameters-py"><span class="std std-ref">Restricting the number of hyperparameters for an existing component</span></a> (<code class="docutils literal notranslate"><span class="pre">example_restrict_number_of_hyperparameters.py</span></code>)</p></td>
-<td><p>00:07.799</p></td>
+<td><p>00:06.778</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 </tbody>
 </table>
-</section>
+</div>
 
 
     </div>
@@ -167,7 +166,7 @@
       
     </p>
     <p>
-        &copy; Copyright 2014-2021, Machine Learning Professorship Freiburg.<br/>
+        &copy; Copyright 2014-2022, Machine Learning Professorship Freiburg.<br/>
       Created using <a href="http://sphinx-doc.org/">Sphinx</a> 4.2.0.<br/>
     </p>
   </div>
diff --git a/master/examples/index.html b/master/examples/index.html
index 3201930e9c..16f2b73e73 100644
--- a/master/examples/index.html
+++ b/master/examples/index.html
@@ -3,9 +3,8 @@
 <html>
   <head>
     <meta charset="utf-8" />
-    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="generator" content="Docutils 0.17.1: http://docutils.sourceforge.net/" />
-
-    <title>Examples &#8212; AutoSklearn 0.14.2 documentation</title>
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>Examples &#8212; AutoSklearn 0.14.4 documentation</title>
     <link rel="stylesheet" type="text/css" href="../_static/pygments.css" />
     <link rel="stylesheet" type="text/css" href="../_static/bootstrap-sphinx.css" />
     <link rel="stylesheet" type="text/css" href="../_static/sg_gallery.css" />
@@ -55,7 +54,7 @@
         </button>
         <a class="navbar-brand" href="../index.html">
           auto-sklearn</a>
-        <span class="navbar-text navbar-version pull-left"><b>0.14.2</b></span>
+        <span class="navbar-text navbar-version pull-left"><b>0.14.4</b></span>
       </div>
 
         <div class="collapse navbar-collapse nav-collapse">
@@ -126,226 +125,170 @@
       </div>
     <div class="body col-md-9 content" role="main">
       
-  <section id="examples">
+  <div class="section" id="examples">
 <span id="sphx-glr-examples"></span><span id="id1"></span><h1>Examples<a class="headerlink" href="#examples" title="Permalink to this headline">¶</a></h1>
 <p>Practical examples for using <em>auto-sklearn</em>.</p>
-<div class="sphx-glr-clear"></div></section>
-<section id="basic-examples">
+<div class="sphx-glr-clear"></div></div>
+<div class="section" id="basic-examples">
 <span id="sphx-glr-examples-20-basic"></span><span id="id2"></span><h1>Basic Examples<a class="headerlink" href="#basic-examples" title="Permalink to this headline">¶</a></h1>
 <p>Examples for basic classification, regression, multi-output regression, and multi-label classification datasets.</p>
-<div class="sphx-glr-thumbcontainer" tooltip="The following example shows how to fit a simple classification model with *auto-sklearn*."><figure class="align-default" id="id6">
+<div class="sphx-glr-thumbcontainer" tooltip="The following example shows how to fit a simple classification model with *auto-sklearn*."><div class="figure align-default" id="id6">
 <img alt="Classification" src="../_images/sphx_glr_example_classification_thumb.png" />
-<figcaption>
-<p><span class="caption-text"><a class="reference internal" href="20_basic/example_classification.html#sphx-glr-examples-20-basic-example-classification-py"><span class="std std-ref">Classification</span></a></span><a class="headerlink" href="#id6" title="Permalink to this image">¶</a></p>
-</figcaption>
-</figure>
+<p class="caption"><span class="caption-text"><a class="reference internal" href="20_basic/example_classification.html#sphx-glr-examples-20-basic-example-classification-py"><span class="std std-ref">Classification</span></a></span><a class="headerlink" href="#id6" title="Permalink to this image">¶</a></p>
+</div>
 </div><div class="toctree-wrapper compound">
 </div>
-<div class="sphx-glr-thumbcontainer" tooltip="The following example shows how to fit a multioutput regression model with *auto-sklearn*."><figure class="align-default" id="id7">
+<div class="sphx-glr-thumbcontainer" tooltip="The following example shows how to fit a multioutput regression model with *auto-sklearn*."><div class="figure align-default" id="id7">
 <img alt="Multi-output Regression" src="../_images/sphx_glr_example_multioutput_regression_thumb.png" />
-<figcaption>
-<p><span class="caption-text"><a class="reference internal" href="20_basic/example_multioutput_regression.html#sphx-glr-examples-20-basic-example-multioutput-regression-py"><span class="std std-ref">Multi-output Regression</span></a></span><a class="headerlink" href="#id7" title="Permalink to this image">¶</a></p>
-</figcaption>
-</figure>
+<p class="caption"><span class="caption-text"><a class="reference internal" href="20_basic/example_multioutput_regression.html#sphx-glr-examples-20-basic-example-multioutput-regression-py"><span class="std std-ref">Multi-output Regression</span></a></span><a class="headerlink" href="#id7" title="Permalink to this image">¶</a></p>
+</div>
 </div><div class="toctree-wrapper compound">
 </div>
-<div class="sphx-glr-thumbcontainer" tooltip="The following example shows how to fit a simple regression model with *auto-sklearn*."><figure class="align-default" id="id8">
+<div class="sphx-glr-thumbcontainer" tooltip="The following example shows how to fit a simple regression model with *auto-sklearn*."><div class="figure align-default" id="id8">
 <img alt="Regression" src="../_images/sphx_glr_example_regression_thumb.png" />
-<figcaption>
-<p><span class="caption-text"><a class="reference internal" href="20_basic/example_regression.html#sphx-glr-examples-20-basic-example-regression-py"><span class="std std-ref">Regression</span></a></span><a class="headerlink" href="#id8" title="Permalink to this image">¶</a></p>
-</figcaption>
-</figure>
+<p class="caption"><span class="caption-text"><a class="reference internal" href="20_basic/example_regression.html#sphx-glr-examples-20-basic-example-regression-py"><span class="std std-ref">Regression</span></a></span><a class="headerlink" href="#id8" title="Permalink to this image">¶</a></p>
+</div>
 </div><div class="toctree-wrapper compound">
 </div>
-<div class="sphx-glr-thumbcontainer" tooltip="This examples shows how to format the targets for a multilabel classification problem. Details ..."><figure class="align-default" id="id9">
+<div class="sphx-glr-thumbcontainer" tooltip="This examples shows how to format the targets for a multilabel classification problem. Details ..."><div class="figure align-default" id="id9">
 <img alt="Multi-label Classification" src="../_images/sphx_glr_example_multilabel_classification_thumb.png" />
-<figcaption>
-<p><span class="caption-text"><a class="reference internal" href="20_basic/example_multilabel_classification.html#sphx-glr-examples-20-basic-example-multilabel-classification-py"><span class="std std-ref">Multi-label Classification</span></a></span><a class="headerlink" href="#id9" title="Permalink to this image">¶</a></p>
-</figcaption>
-</figure>
+<p class="caption"><span class="caption-text"><a class="reference internal" href="20_basic/example_multilabel_classification.html#sphx-glr-examples-20-basic-example-multilabel-classification-py"><span class="std std-ref">Multi-label Classification</span></a></span><a class="headerlink" href="#id9" title="Permalink to this image">¶</a></p>
+</div>
 </div><div class="toctree-wrapper compound">
 </div>
-<div class="sphx-glr-clear"></div></section>
-<section id="advanced-examples">
+<div class="sphx-glr-clear"></div></div>
+<div class="section" id="advanced-examples">
 <span id="sphx-glr-examples-40-advanced"></span><span id="id3"></span><h1>Advanced Examples<a class="headerlink" href="#advanced-examples" title="Permalink to this headline">¶</a></h1>
 <p>Examples on customizing Auto-sklearn to ones use case by changing the
 metric to optimize, the train-validation split, giving feature types,
 using pandas dataframes as input and inspecting the results of the search
 procedure.</p>
-<div class="sphx-glr-thumbcontainer" tooltip="The following example shows how to inspect the models which *auto-sklearn* optimizes over and h..."><figure class="align-default" id="id10">
+<div class="sphx-glr-thumbcontainer" tooltip="The following example shows how to inspect the models which *auto-sklearn* optimizes over and h..."><div class="figure align-default" id="id10">
 <img alt="Interpretable models" src="../_images/sphx_glr_example_interpretable_models_thumb.png" />
-<figcaption>
-<p><span class="caption-text"><a class="reference internal" href="40_advanced/example_interpretable_models.html#sphx-glr-examples-40-advanced-example-interpretable-models-py"><span class="std std-ref">Interpretable models</span></a></span><a class="headerlink" href="#id10" title="Permalink to this image">¶</a></p>
-</figcaption>
-</figure>
+<p class="caption"><span class="caption-text"><a class="reference internal" href="40_advanced/example_interpretable_models.html#sphx-glr-examples-40-advanced-example-interpretable-models-py"><span class="std std-ref">Interpretable models</span></a></span><a class="headerlink" href="#id10" title="Permalink to this image">¶</a></p>
+</div>
 </div><div class="toctree-wrapper compound">
 </div>
-<div class="sphx-glr-thumbcontainer" tooltip="In *auto-sklearn* it is possible to specify the feature types of a dataset when calling the met..."><figure class="align-default" id="id11">
+<div class="sphx-glr-thumbcontainer" tooltip="In *auto-sklearn* it is possible to specify the feature types of a dataset when calling the met..."><div class="figure align-default" id="id11">
 <img alt="Feature Types" src="../_images/sphx_glr_example_feature_types_thumb.png" />
-<figcaption>
-<p><span class="caption-text"><a class="reference internal" href="40_advanced/example_feature_types.html#sphx-glr-examples-40-advanced-example-feature-types-py"><span class="std std-ref">Feature Types</span></a></span><a class="headerlink" href="#id11" title="Permalink to this image">¶</a></p>
-</figcaption>
-</figure>
+<p class="caption"><span class="caption-text"><a class="reference internal" href="40_advanced/example_feature_types.html#sphx-glr-examples-40-advanced-example-feature-types-py"><span class="std std-ref">Feature Types</span></a></span><a class="headerlink" href="#id11" title="Permalink to this image">¶</a></p>
+</div>
 </div><div class="toctree-wrapper compound">
 </div>
-<div class="sphx-glr-thumbcontainer" tooltip="In *Auto-sklearn*, model is optimized over a metric, either built-in or custom metric. Moreover..."><figure class="align-default" id="id12">
+<div class="sphx-glr-thumbcontainer" tooltip="In *Auto-sklearn*, model is optimized over a metric, either built-in or custom metric. Moreover..."><div class="figure align-default" id="id12">
 <img alt="Metrics" src="../_images/sphx_glr_example_calc_multiple_metrics_thumb.png" />
-<figcaption>
-<p><span class="caption-text"><a class="reference internal" href="40_advanced/example_calc_multiple_metrics.html#sphx-glr-examples-40-advanced-example-calc-multiple-metrics-py"><span class="std std-ref">Metrics</span></a></span><a class="headerlink" href="#id12" title="Permalink to this image">¶</a></p>
-</figcaption>
-</figure>
+<p class="caption"><span class="caption-text"><a class="reference internal" href="40_advanced/example_calc_multiple_metrics.html#sphx-glr-examples-40-advanced-example-calc-multiple-metrics-py"><span class="std std-ref">Metrics</span></a></span><a class="headerlink" href="#id12" title="Permalink to this image">¶</a></p>
+</div>
 </div><div class="toctree-wrapper compound">
 </div>
-<div class="sphx-glr-thumbcontainer" tooltip="The following example shows how to fit a simple classification model with *auto-sklearn* and us..."><figure class="align-default" id="id13">
+<div class="sphx-glr-thumbcontainer" tooltip="The following example shows how to fit a simple classification model with *auto-sklearn* and us..."><div class="figure align-default" id="id13">
 <img alt="Model Explanation" src="../_images/sphx_glr_example_inspect_predictions_thumb.png" />
-<figcaption>
-<p><span class="caption-text"><a class="reference internal" href="40_advanced/example_inspect_predictions.html#sphx-glr-examples-40-advanced-example-inspect-predictions-py"><span class="std std-ref">Model Explanation</span></a></span><a class="headerlink" href="#id13" title="Permalink to this image">¶</a></p>
-</figcaption>
-</figure>
+<p class="caption"><span class="caption-text"><a class="reference internal" href="40_advanced/example_inspect_predictions.html#sphx-glr-examples-40-advanced-example-inspect-predictions-py"><span class="std std-ref">Model Explanation</span></a></span><a class="headerlink" href="#id13" title="Permalink to this image">¶</a></p>
+</div>
 </div><div class="toctree-wrapper compound">
 </div>
-<div class="sphx-glr-thumbcontainer" tooltip="    - ensemble_optimization_score     - ensemble_test_score     - single_best_optimization_scor..."><figure class="align-default" id="id14">
+<div class="sphx-glr-thumbcontainer" tooltip="    - ensemble_optimization_score     - ensemble_test_score     - single_best_optimization_scor..."><div class="figure align-default" id="id14">
 <img alt="Performance-over-time plot" src="../_images/sphx_glr_example_pandas_train_test_thumb.png" />
-<figcaption>
-<p><span class="caption-text"><a class="reference internal" href="40_advanced/example_pandas_train_test.html#sphx-glr-examples-40-advanced-example-pandas-train-test-py"><span class="std std-ref">Performance-over-time plot</span></a></span><a class="headerlink" href="#id14" title="Permalink to this image">¶</a></p>
-</figcaption>
-</figure>
+<p class="caption"><span class="caption-text"><a class="reference internal" href="40_advanced/example_pandas_train_test.html#sphx-glr-examples-40-advanced-example-pandas-train-test-py"><span class="std std-ref">Performance-over-time plot</span></a></span><a class="headerlink" href="#id14" title="Permalink to this image">¶</a></p>
+</div>
 </div><div class="toctree-wrapper compound">
 </div>
-<div class="sphx-glr-thumbcontainer" tooltip="This example shows how to provide a custom logging configuration to *auto-sklearn*. We will be ..."><figure class="align-default" id="id15">
+<div class="sphx-glr-thumbcontainer" tooltip="This example shows how to provide a custom logging configuration to *auto-sklearn*. We will be ..."><div class="figure align-default" id="id15">
 <img alt="Logging and debugging" src="../_images/sphx_glr_example_debug_logging_thumb.png" />
-<figcaption>
-<p><span class="caption-text"><a class="reference internal" href="40_advanced/example_debug_logging.html#sphx-glr-examples-40-advanced-example-debug-logging-py"><span class="std std-ref">Logging and debugging</span></a></span><a class="headerlink" href="#id15" title="Permalink to this image">¶</a></p>
-</figcaption>
-</figure>
+<p class="caption"><span class="caption-text"><a class="reference internal" href="40_advanced/example_debug_logging.html#sphx-glr-examples-40-advanced-example-debug-logging-py"><span class="std std-ref">Logging and debugging</span></a></span><a class="headerlink" href="#id15" title="Permalink to this image">¶</a></p>
+</div>
 </div><div class="toctree-wrapper compound">
 </div>
-<div class="sphx-glr-thumbcontainer" tooltip="*Auto-sklearn* searches for the best combination of machine learning algorithms and their hyper..."><figure class="align-default" id="id16">
+<div class="sphx-glr-thumbcontainer" tooltip="*Auto-sklearn* searches for the best combination of machine learning algorithms and their hyper..."><div class="figure align-default" id="id16">
 <img alt="Fit a single configuration" src="../_images/sphx_glr_example_single_configuration_thumb.png" />
-<figcaption>
-<p><span class="caption-text"><a class="reference internal" href="40_advanced/example_single_configuration.html#sphx-glr-examples-40-advanced-example-single-configuration-py"><span class="std std-ref">Fit a single configuration</span></a></span><a class="headerlink" href="#id16" title="Permalink to this image">¶</a></p>
-</figcaption>
-</figure>
+<p class="caption"><span class="caption-text"><a class="reference internal" href="40_advanced/example_single_configuration.html#sphx-glr-examples-40-advanced-example-single-configuration-py"><span class="std std-ref">Fit a single configuration</span></a></span><a class="headerlink" href="#id16" title="Permalink to this image">¶</a></p>
+</div>
 </div><div class="toctree-wrapper compound">
 </div>
-<div class="sphx-glr-thumbcontainer" tooltip="In *auto-sklearn* it is possible to use different resampling strategies by specifying the argum..."><figure class="align-default" id="id17">
+<div class="sphx-glr-thumbcontainer" tooltip="In *auto-sklearn* it is possible to use different resampling strategies by specifying the argum..."><div class="figure align-default" id="id17">
 <img alt="Resampling Strategies" src="../_images/sphx_glr_example_resampling_thumb.png" />
-<figcaption>
-<p><span class="caption-text"><a class="reference internal" href="40_advanced/example_resampling.html#sphx-glr-examples-40-advanced-example-resampling-py"><span class="std std-ref">Resampling Strategies</span></a></span><a class="headerlink" href="#id17" title="Permalink to this image">¶</a></p>
-</figcaption>
-</figure>
+<p class="caption"><span class="caption-text"><a class="reference internal" href="40_advanced/example_resampling.html#sphx-glr-examples-40-advanced-example-resampling-py"><span class="std std-ref">Resampling Strategies</span></a></span><a class="headerlink" href="#id17" title="Permalink to this image">¶</a></p>
+</div>
 </div><div class="toctree-wrapper compound">
 </div>
-<div class="sphx-glr-thumbcontainer" tooltip="The following example shows how to obtain information from a finished Auto-sklearn run. In part..."><figure class="align-default" id="id18">
+<div class="sphx-glr-thumbcontainer" tooltip="The following example shows how to obtain information from a finished Auto-sklearn run. In part..."><div class="figure align-default" id="id18">
 <img alt="Obtain run information" src="../_images/sphx_glr_example_get_pipeline_components_thumb.png" />
-<figcaption>
-<p><span class="caption-text"><a class="reference internal" href="40_advanced/example_get_pipeline_components.html#sphx-glr-examples-40-advanced-example-get-pipeline-components-py"><span class="std std-ref">Obtain run information</span></a></span><a class="headerlink" href="#id18" title="Permalink to this image">¶</a></p>
-</figcaption>
-</figure>
+<p class="caption"><span class="caption-text"><a class="reference internal" href="40_advanced/example_get_pipeline_components.html#sphx-glr-examples-40-advanced-example-get-pipeline-components-py"><span class="std std-ref">Obtain run information</span></a></span><a class="headerlink" href="#id18" title="Permalink to this image">¶</a></p>
+</div>
 </div><div class="toctree-wrapper compound">
 </div>
-<div class="sphx-glr-thumbcontainer" tooltip="*Auto-sklearn* supports various built-in metrics, which can be found in the metrics section in ..."><figure class="align-default" id="id19">
+<div class="sphx-glr-thumbcontainer" tooltip="*Auto-sklearn* supports various built-in metrics, which can be found in the metrics section in ..."><div class="figure align-default" id="id19">
 <img alt="Metrics" src="../_images/sphx_glr_example_metrics_thumb.png" />
-<figcaption>
-<p><span class="caption-text"><a class="reference internal" href="40_advanced/example_metrics.html#sphx-glr-examples-40-advanced-example-metrics-py"><span class="std std-ref">Metrics</span></a></span><a class="headerlink" href="#id19" title="Permalink to this image">¶</a></p>
-</figcaption>
-</figure>
+<p class="caption"><span class="caption-text"><a class="reference internal" href="40_advanced/example_metrics.html#sphx-glr-examples-40-advanced-example-metrics-py"><span class="std std-ref">Metrics</span></a></span><a class="headerlink" href="#id19" title="Permalink to this image">¶</a></p>
+</div>
 </div><div class="toctree-wrapper compound">
 </div>
-<div class="sphx-glr-clear"></div></section>
-<section id="search-examples">
+<div class="sphx-glr-clear"></div></div>
+<div class="section" id="search-examples">
 <span id="sphx-glr-examples-60-search"></span><span id="id4"></span><h1>Search Examples<a class="headerlink" href="#search-examples" title="Permalink to this headline">¶</a></h1>
 <p>Examples of changing the search procedure of Auto-sklearn.</p>
-<div class="sphx-glr-thumbcontainer" tooltip="*Auto-sklearn* uses dask.distributed &lt;https://distributed.dask.org/en/latest/index.html&gt;_ for p..."><figure class="align-default" id="id20">
+<div class="sphx-glr-thumbcontainer" tooltip="*Auto-sklearn* uses dask.distributed &lt;https://distributed.dask.org/en/latest/index.html&gt;_ for p..."><div class="figure align-default" id="id20">
 <img alt="Parallel Usage  on a single machine" src="../_images/sphx_glr_example_parallel_n_jobs_thumb.png" />
-<figcaption>
-<p><span class="caption-text"><a class="reference internal" href="60_search/example_parallel_n_jobs.html#sphx-glr-examples-60-search-example-parallel-n-jobs-py"><span class="std std-ref">Parallel Usage  on a single machine</span></a></span><a class="headerlink" href="#id20" title="Permalink to this image">¶</a></p>
-</figcaption>
-</figure>
+<p class="caption"><span class="caption-text"><a class="reference internal" href="60_search/example_parallel_n_jobs.html#sphx-glr-examples-60-search-example-parallel-n-jobs-py"><span class="std std-ref">Parallel Usage  on a single machine</span></a></span><a class="headerlink" href="#id20" title="Permalink to this image">¶</a></p>
+</div>
 </div><div class="toctree-wrapper compound">
 </div>
-<div class="sphx-glr-thumbcontainer" tooltip="By default, *auto-sklearn* fits the machine learning models and build their ensembles in parall..."><figure class="align-default" id="id21">
+<div class="sphx-glr-thumbcontainer" tooltip="By default, *auto-sklearn* fits the machine learning models and build their ensembles in parall..."><div class="figure align-default" id="id21">
 <img alt="Sequential Usage" src="../_images/sphx_glr_example_sequential_thumb.png" />
-<figcaption>
-<p><span class="caption-text"><a class="reference internal" href="60_search/example_sequential.html#sphx-glr-examples-60-search-example-sequential-py"><span class="std std-ref">Sequential Usage</span></a></span><a class="headerlink" href="#id21" title="Permalink to this image">¶</a></p>
-</figcaption>
-</figure>
+<p class="caption"><span class="caption-text"><a class="reference internal" href="60_search/example_sequential.html#sphx-glr-examples-60-search-example-sequential-py"><span class="std std-ref">Sequential Usage</span></a></span><a class="headerlink" href="#id21" title="Permalink to this image">¶</a></p>
+</div>
 </div><div class="toctree-wrapper compound">
 </div>
-<div class="sphx-glr-thumbcontainer" tooltip="*Auto-sklearn* uses `dask.distributed &lt;https://distributed.dask.org/en/latest/index.html&gt;`_ for..."><figure class="align-default" id="id22">
+<div class="sphx-glr-thumbcontainer" tooltip="*Auto-sklearn* uses `dask.distributed &lt;https://distributed.dask.org/en/latest/index.html&gt;`_ for..."><div class="figure align-default" id="id22">
 <img alt="Parallel Usage: Spawning workers from the command line" src="../_images/sphx_glr_example_parallel_manual_spawning_cli_thumb.png" />
-<figcaption>
-<p><span class="caption-text"><a class="reference internal" href="60_search/example_parallel_manual_spawning_cli.html#sphx-glr-examples-60-search-example-parallel-manual-spawning-cli-py"><span class="std std-ref">Parallel Usage: Spawning workers from the command line</span></a></span><a class="headerlink" href="#id22" title="Permalink to this image">¶</a></p>
-</figcaption>
-</figure>
-</div><div class="toctree-wrapper compound">
+<p class="caption"><span class="caption-text"><a class="reference internal" href="60_search/example_parallel_manual_spawning_cli.html#sphx-glr-examples-60-search-example-parallel-manual-spawning-cli-py"><span class="std std-ref">Parallel Usage: Spawning workers from the command line</span></a></span><a class="headerlink" href="#id22" title="Permalink to this image">¶</a></p>
 </div>
-<div class="sphx-glr-thumbcontainer" tooltip="*Auto-sklearn* uses `dask.distributed &lt;https://distributed.dask.org/en/latest/index.html&gt;`_ for..."><figure class="align-default" id="id23">
-<img alt="Parallel Usage: Spawning workers from within Python" src="../_images/sphx_glr_example_parallel_manual_spawning_python_thumb.png" />
-<figcaption>
-<p><span class="caption-text"><a class="reference internal" href="60_search/example_parallel_manual_spawning_python.html#sphx-glr-examples-60-search-example-parallel-manual-spawning-python-py"><span class="std std-ref">Parallel Usage: Spawning workers from within Python</span></a></span><a class="headerlink" href="#id23" title="Permalink to this image">¶</a></p>
-</figcaption>
-</figure>
 </div><div class="toctree-wrapper compound">
 </div>
-<div class="sphx-glr-thumbcontainer" tooltip="A crucial feature of *auto-sklearn* is automatically optimizing the hyperparameters through SMA..."><figure class="align-default" id="id24">
+<div class="sphx-glr-thumbcontainer" tooltip="A crucial feature of *auto-sklearn* is automatically optimizing the hyperparameters through SMA..."><div class="figure align-default" id="id23">
 <img alt="Random Search" src="../_images/sphx_glr_example_random_search_thumb.png" />
-<figcaption>
-<p><span class="caption-text"><a class="reference internal" href="60_search/example_random_search.html#sphx-glr-examples-60-search-example-random-search-py"><span class="std std-ref">Random Search</span></a></span><a class="headerlink" href="#id24" title="Permalink to this image">¶</a></p>
-</figcaption>
-</figure>
+<p class="caption"><span class="caption-text"><a class="reference internal" href="60_search/example_random_search.html#sphx-glr-examples-60-search-example-random-search-py"><span class="std std-ref">Random Search</span></a></span><a class="headerlink" href="#id23" title="Permalink to this image">¶</a></p>
+</div>
 </div><div class="toctree-wrapper compound">
 </div>
-<div class="sphx-glr-thumbcontainer" tooltip="This advanced  example illustrates how to interact with the SMAC callback and get relevant info..."><figure class="align-default" id="id25">
+<div class="sphx-glr-thumbcontainer" tooltip="This advanced  example illustrates how to interact with the SMAC callback and get relevant info..."><div class="figure align-default" id="id24">
 <img alt="Successive Halving" src="../_images/sphx_glr_example_successive_halving_thumb.png" />
-<figcaption>
-<p><span class="caption-text"><a class="reference internal" href="60_search/example_successive_halving.html#sphx-glr-examples-60-search-example-successive-halving-py"><span class="std std-ref">Successive Halving</span></a></span><a class="headerlink" href="#id25" title="Permalink to this image">¶</a></p>
-</figcaption>
-</figure>
+<p class="caption"><span class="caption-text"><a class="reference internal" href="60_search/example_successive_halving.html#sphx-glr-examples-60-search-example-successive-halving-py"><span class="std std-ref">Successive Halving</span></a></span><a class="headerlink" href="#id24" title="Permalink to this image">¶</a></p>
+</div>
 </div><div class="toctree-wrapper compound">
 </div>
-<div class="sphx-glr-clear"></div></section>
-<section id="extension-examples">
+<div class="sphx-glr-clear"></div></div>
+<div class="section" id="extension-examples">
 <span id="sphx-glr-examples-80-extending"></span><span id="id5"></span><h1>Extension Examples<a class="headerlink" href="#extension-examples" title="Permalink to this headline">¶</a></h1>
 <p>Examples of extending Auto-sklearn with custom components.</p>
-<div class="sphx-glr-thumbcontainer" tooltip="The following example demonstrates how to turn off data preprocessing step in auto-skearn."><figure class="align-default" id="id26">
+<div class="sphx-glr-thumbcontainer" tooltip="The following example demonstrates how to turn off data preprocessing step in auto-skearn."><div class="figure align-default" id="id25">
 <img alt="Extending Auto-Sklearn with Data Preprocessor Component" src="../_images/sphx_glr_example_extending_data_preprocessor_thumb.png" />
-<figcaption>
-<p><span class="caption-text"><a class="reference internal" href="80_extending/example_extending_data_preprocessor.html#sphx-glr-examples-80-extending-example-extending-data-preprocessor-py"><span class="std std-ref">Extending Auto-Sklearn with Data Preprocessor Component</span></a></span><a class="headerlink" href="#id26" title="Permalink to this image">¶</a></p>
-</figcaption>
-</figure>
+<p class="caption"><span class="caption-text"><a class="reference internal" href="80_extending/example_extending_data_preprocessor.html#sphx-glr-examples-80-extending-example-extending-data-preprocessor-py"><span class="std std-ref">Extending Auto-Sklearn with Data Preprocessor Component</span></a></span><a class="headerlink" href="#id25" title="Permalink to this image">¶</a></p>
+</div>
 </div><div class="toctree-wrapper compound">
 </div>
-<div class="sphx-glr-thumbcontainer" tooltip="The following example demonstrates how to create a wrapper around the linear discriminant analy..."><figure class="align-default" id="id27">
+<div class="sphx-glr-thumbcontainer" tooltip="The following example demonstrates how to create a wrapper around the linear discriminant analy..."><div class="figure align-default" id="id26">
 <img alt="Extending Auto-Sklearn with Preprocessor Component" src="../_images/sphx_glr_example_extending_preprocessor_thumb.png" />
-<figcaption>
-<p><span class="caption-text"><a class="reference internal" href="80_extending/example_extending_preprocessor.html#sphx-glr-examples-80-extending-example-extending-preprocessor-py"><span class="std std-ref">Extending Auto-Sklearn with Preprocessor Component</span></a></span><a class="headerlink" href="#id27" title="Permalink to this image">¶</a></p>
-</figcaption>
-</figure>
+<p class="caption"><span class="caption-text"><a class="reference internal" href="80_extending/example_extending_preprocessor.html#sphx-glr-examples-80-extending-example-extending-preprocessor-py"><span class="std std-ref">Extending Auto-Sklearn with Preprocessor Component</span></a></span><a class="headerlink" href="#id26" title="Permalink to this image">¶</a></p>
+</div>
 </div><div class="toctree-wrapper compound">
 </div>
-<div class="sphx-glr-thumbcontainer" tooltip="The following example demonstrates how to replace an existing component with a new component, i..."><figure class="align-default" id="id28">
+<div class="sphx-glr-thumbcontainer" tooltip="The following example demonstrates how to replace an existing component with a new component, i..."><div class="figure align-default" id="id27">
 <img alt="Restricting the number of hyperparameters for an existing component" src="../_images/sphx_glr_example_restrict_number_of_hyperparameters_thumb.png" />
-<figcaption>
-<p><span class="caption-text"><a class="reference internal" href="80_extending/example_restrict_number_of_hyperparameters.html#sphx-glr-examples-80-extending-example-restrict-number-of-hyperparameters-py"><span class="std std-ref">Restricting the number of hyperparameters for an existing component</span></a></span><a class="headerlink" href="#id28" title="Permalink to this image">¶</a></p>
-</figcaption>
-</figure>
+<p class="caption"><span class="caption-text"><a class="reference internal" href="80_extending/example_restrict_number_of_hyperparameters.html#sphx-glr-examples-80-extending-example-restrict-number-of-hyperparameters-py"><span class="std std-ref">Restricting the number of hyperparameters for an existing component</span></a></span><a class="headerlink" href="#id27" title="Permalink to this image">¶</a></p>
+</div>
 </div><div class="toctree-wrapper compound">
 </div>
-<div class="sphx-glr-thumbcontainer" tooltip="The following example demonstrates how to create a new regression component for using in auto-s..."><figure class="align-default" id="id29">
+<div class="sphx-glr-thumbcontainer" tooltip="The following example demonstrates how to create a new regression component for using in auto-s..."><div class="figure align-default" id="id28">
 <img alt="Extending Auto-Sklearn with Regression Component" src="../_images/sphx_glr_example_extending_regression_thumb.png" />
-<figcaption>
-<p><span class="caption-text"><a class="reference internal" href="80_extending/example_extending_regression.html#sphx-glr-examples-80-extending-example-extending-regression-py"><span class="std std-ref">Extending Auto-Sklearn with Regression Component</span></a></span><a class="headerlink" href="#id29" title="Permalink to this image">¶</a></p>
-</figcaption>
-</figure>
+<p class="caption"><span class="caption-text"><a class="reference internal" href="80_extending/example_extending_regression.html#sphx-glr-examples-80-extending-example-extending-regression-py"><span class="std std-ref">Extending Auto-Sklearn with Regression Component</span></a></span><a class="headerlink" href="#id28" title="Permalink to this image">¶</a></p>
+</div>
 </div><div class="toctree-wrapper compound">
 </div>
-<div class="sphx-glr-thumbcontainer" tooltip="The following example demonstrates how to create a new classification component for using in au..."><figure class="align-default" id="id30">
+<div class="sphx-glr-thumbcontainer" tooltip="The following example demonstrates how to create a new classification component for using in au..."><div class="figure align-default" id="id29">
 <img alt="Extending Auto-Sklearn with Classification Component" src="../_images/sphx_glr_example_extending_classification_thumb.png" />
-<figcaption>
-<p><span class="caption-text"><a class="reference internal" href="80_extending/example_extending_classification.html#sphx-glr-examples-80-extending-example-extending-classification-py"><span class="std std-ref">Extending Auto-Sklearn with Classification Component</span></a></span><a class="headerlink" href="#id30" title="Permalink to this image">¶</a></p>
-</figcaption>
-</figure>
+<p class="caption"><span class="caption-text"><a class="reference internal" href="80_extending/example_extending_classification.html#sphx-glr-examples-80-extending-example-extending-classification-py"><span class="std std-ref">Extending Auto-Sklearn with Classification Component</span></a></span><a class="headerlink" href="#id29" title="Permalink to this image">¶</a></p>
+</div>
 </div><div class="toctree-wrapper compound">
 </div>
 <div class="sphx-glr-clear"></div><div class="sphx-glr-footer class sphx-glr-footer-gallery docutils container">
@@ -357,7 +300,7 @@
 </div>
 </div>
 <p class="sphx-glr-signature"><a class="reference external" href="https://sphinx-gallery.github.io">Gallery generated by Sphinx-Gallery</a></p>
-</section>
+</div>
 
 
     </div>
@@ -378,7 +321,7 @@
       
     </p>
     <p>
-        &copy; Copyright 2014-2021, Machine Learning Professorship Freiburg.<br/>
+        &copy; Copyright 2014-2022, Machine Learning Professorship Freiburg.<br/>
       Created using <a href="http://sphinx-doc.org/">Sphinx</a> 4.2.0.<br/>
     </p>
   </div>
diff --git a/master/extending.html b/master/extending.html
index 05733abc9c..5e62801345 100644
--- a/master/extending.html
+++ b/master/extending.html
@@ -3,9 +3,8 @@
 <html>
   <head>
     <meta charset="utf-8" />
-    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="generator" content="Docutils 0.17.1: http://docutils.sourceforge.net/" />
-
-    <title>Extending auto-sklearn &#8212; AutoSklearn 0.14.2 documentation</title>
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>Extending auto-sklearn &#8212; AutoSklearn 0.14.4 documentation</title>
     <link rel="stylesheet" type="text/css" href="_static/pygments.css" />
     <link rel="stylesheet" type="text/css" href="_static/bootstrap-sphinx.css" />
     <link rel="stylesheet" type="text/css" href="_static/sg_gallery.css" />
@@ -55,7 +54,7 @@
         </button>
         <a class="navbar-brand" href="index.html">
           auto-sklearn</a>
-        <span class="navbar-text navbar-version pull-left"><b>0.14.2</b></span>
+        <span class="navbar-text navbar-version pull-left"><b>0.14.4</b></span>
       </div>
 
         <div class="collapse navbar-collapse nav-collapse">
@@ -124,13 +123,13 @@
       </div>
     <div class="body col-md-9 content" role="main">
       
-  <section id="extending-auto-sklearn">
+  <div class="section" id="extending-auto-sklearn">
 <span id="extending"></span><h1>Extending auto-sklearn<a class="headerlink" href="#extending-auto-sklearn" title="Permalink to this headline">¶</a></h1>
 <p>auto-sklearn can be easily extended with new classification, regression and
 feature preprocessing methods. In order to do so, a user has to implement a
 wrapper class and register it to auto-sklearn. This manual will walk you
 through the process.</p>
-<section id="writing-a-component">
+<div class="section" id="writing-a-component">
 <h2>Writing a component<a class="headerlink" href="#writing-a-component" title="Permalink to this headline">¶</a></h2>
 <p>Depending on the purpose, the component has to be a subclass of one of the
 following base classes:</p>
@@ -165,7 +164,7 @@ <h2>Writing a component<a class="headerlink" href="#writing-a-component" title="
 <span class="sig-prename descclassname"><span class="pre">autosklearn.pipeline.components.feature_preprocessing.</span></span><span class="sig-name descname"><span class="pre">add_preprocessor</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">preprocessor</span></span><span class="p"><span class="pre">:</span></span> <span class="n"><span class="pre">Type</span><span class="p"><span class="pre">[</span></span><a class="reference internal" href="api.html#autosklearn.pipeline.components.base.AutoSklearnPreprocessingAlgorithm" title="autosklearn.pipeline.components.base.AutoSklearnPreprocessingAlgorithm"><span class="pre">autosklearn.pipeline.components.base.AutoSklearnPreprocessingAlgorithm</span></a><span class="p"><span class="pre">]</span></span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">None</span></span></span><a class="reference internal" href="_modules/autosklearn/pipeline/components/feature_preprocessing.html#add_preprocessor"><span class="viewcode-link"><span class="pre">[source]</span></span></a><a class="headerlink" href="#autosklearn.pipeline.components.feature_preprocessing.add_preprocessor" title="Permalink to this definition">¶</a></dt>
 <dd></dd></dl>
 
-<section id="get-hyperparameter-search-space">
+<div class="section" id="get-hyperparameter-search-space">
 <span id="id1"></span><h3>get_hyperparameter_search_space()<a class="headerlink" href="#get-hyperparameter-search-space" title="Permalink to this headline">¶</a></h3>
 <p>Return an instance of <code class="docutils literal notranslate"><span class="pre">ConfigSpace.configuration_space.ConfigurationSpace</span></code>.</p>
 <p>See also the abstract definitions:
@@ -174,8 +173,8 @@ <h2>Writing a component<a class="headerlink" href="#writing-a-component" title="
 <code class="xref py py-meth docutils literal notranslate"><span class="pre">AutoSklearnPreprocessingAlgorithm.get_hyperparameter_search_space()</span></code></p>
 <p>To find out about how to create a <code class="docutils literal notranslate"><span class="pre">ConfigurationSpace</span></code>-object, please look
 at the source code on <a class="reference external" href="https://github.com/automl/auto-sklearn/tree/master/autosklearn/pipeline/components/classification">github.com</a>.</p>
-</section>
-<section id="get-properties">
+</div>
+<div class="section" id="get-properties">
 <span id="id2"></span><h3>get_properties()<a class="headerlink" href="#get-properties" title="Permalink to this headline">¶</a></h3>
 <p>Return a dictionary which defines how the component can be used when
 constructing a machine learning pipeline. The following fields must be
@@ -291,9 +290,9 @@ <h2>Writing a component<a class="headerlink" href="#writing-a-component" title="
 </dl>
 </li>
 </ul>
-</section>
-</section>
-<section id="classification">
+</div>
+</div>
+<div class="section" id="classification">
 <h2>Classification<a class="headerlink" href="#classification" title="Permalink to this headline">¶</a></h2>
 <p>In addition two <cite>get_properties()</cite> and <cite>get_hyperparameter_search_space()</cite>
 you have to implement
@@ -301,8 +300,8 @@ <h2>Classification<a class="headerlink" href="#classification" title="Permalink
 and
 <a class="reference internal" href="api.html#autosklearn.pipeline.components.base.AutoSklearnClassificationAlgorithm.predict" title="autosklearn.pipeline.components.base.AutoSklearnClassificationAlgorithm.predict"><code class="xref py py-meth docutils literal notranslate"><span class="pre">AutoSklearnClassificationAlgorithm.predict()</span></code></a>
 . These are an implementation of the <a class="reference external" href="https://scikit-learn.org/stable/developers/develop.html#apis-of-scikit-learn-objects">scikit-learn predictor API</a>.</p>
-</section>
-<section id="regression">
+</div>
+<div class="section" id="regression">
 <h2>Regression<a class="headerlink" href="#regression" title="Permalink to this headline">¶</a></h2>
 <p>In addition two <cite>get_properties()</cite> and <cite>get_hyperparameter_search_space()</cite>
 you have to implement
@@ -310,8 +309,8 @@ <h2>Regression<a class="headerlink" href="#regression" title="Permalink to this
 and
 <a class="reference internal" href="api.html#autosklearn.pipeline.components.base.AutoSklearnRegressionAlgorithm.predict" title="autosklearn.pipeline.components.base.AutoSklearnRegressionAlgorithm.predict"><code class="xref py py-meth docutils literal notranslate"><span class="pre">AutoSklearnRegressionAlgorithm.predict()</span></code></a>
 . These are an implementation of the <a class="reference external" href="https://scikit-learn.org/stable/developers/develop.html#apis-of-scikit-learn-objects">scikit-learn predictor API</a>.</p>
-</section>
-<section id="feature-preprocessing">
+</div>
+<div class="section" id="feature-preprocessing">
 <h2>Feature Preprocessing<a class="headerlink" href="#feature-preprocessing" title="Permalink to this headline">¶</a></h2>
 <p>In addition two <cite>get_properties()</cite> and <cite>get_hyperparameter_search_space()</cite>
 you have to implement
@@ -319,8 +318,8 @@ <h2>Feature Preprocessing<a class="headerlink" href="#feature-preprocessing" tit
 and
 <a class="reference internal" href="api.html#autosklearn.pipeline.components.base.AutoSklearnPreprocessingAlgorithm.transform" title="autosklearn.pipeline.components.base.AutoSklearnPreprocessingAlgorithm.transform"><code class="xref py py-meth docutils literal notranslate"><span class="pre">AutoSklearnPreprocessingAlgorithm.transform()</span></code></a>
 . These are an implementation of the <a class="reference external" href="https://scikit-learn.org/stable/developers/develop.html#apis-of-scikit-learn-objects">scikit-learn predictor API</a>.</p>
-</section>
-</section>
+</div>
+</div>
 
 
     </div>
@@ -341,7 +340,7 @@ <h2>Feature Preprocessing<a class="headerlink" href="#feature-preprocessing" tit
       
     </p>
     <p>
-        &copy; Copyright 2014-2021, Machine Learning Professorship Freiburg.<br/>
+        &copy; Copyright 2014-2022, Machine Learning Professorship Freiburg.<br/>
       Created using <a href="http://sphinx-doc.org/">Sphinx</a> 4.2.0.<br/>
     </p>
   </div>
diff --git a/master/faq.html b/master/faq.html
index 6b5364b39f..f31f701e2a 100644
--- a/master/faq.html
+++ b/master/faq.html
@@ -3,9 +3,8 @@
 <html>
   <head>
     <meta charset="utf-8" />
-    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="generator" content="Docutils 0.17.1: http://docutils.sourceforge.net/" />
-
-    <title>FAQ &#8212; AutoSklearn 0.14.2 documentation</title>
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>FAQ &#8212; AutoSklearn 0.14.4 documentation</title>
     <link rel="stylesheet" type="text/css" href="_static/pygments.css" />
     <link rel="stylesheet" type="text/css" href="_static/bootstrap-sphinx.css" />
     <link rel="stylesheet" type="text/css" href="_static/sg_gallery.css" />
@@ -55,7 +54,7 @@
         </button>
         <a class="navbar-brand" href="index.html">
           auto-sklearn</a>
-        <span class="navbar-text navbar-version pull-left"><b>0.14.2</b></span>
+        <span class="navbar-text navbar-version pull-left"><b>0.14.4</b></span>
       </div>
 
         <div class="collapse navbar-collapse nav-collapse">
@@ -108,40 +107,15 @@
       <div class="col-md-3">
         <div id="sidebar" class="bs-sidenav" role="complementary"><ul>
 <li><a class="reference internal" href="#">FAQ</a><ul>
-<li><a class="reference internal" href="#issues">Issues</a><ul>
-<li><a class="reference internal" href="#auto-sklearn-is-extremely-memory-hungry-in-a-sequential-setting">Auto-sklearn is extremely memory hungry in a sequential setting</a></li>
-<li><a class="reference internal" href="#auto-sklearn-is-extremely-memory-hungry-in-a-parallel-setting">Auto-sklearn is extremely memory hungry in a parallel setting</a></li>
-<li><a class="reference internal" href="#auto-sklearn-crashes-with-a-segmentation-fault">Auto-sklearn crashes with a segmentation fault</a></li>
-</ul>
-</li>
-<li><a class="reference internal" href="#log-files-and-output">Log files and output</a><ul>
-<li><a class="reference internal" href="#where-does-auto-sklearn-output-files-by-default">Where does Auto-sklearn output files by default?</a></li>
-<li><a class="reference internal" href="#auto-sklearn-eats-up-all-my-disk-space">Auto-sklearn eats up all my disk space</a></li>
-</ul>
-</li>
-<li><a class="reference internal" href="#available-machine-learning-models">Available machine learning models</a><ul>
-<li><a class="reference internal" href="#will-non-scikit-learn-models-be-added-to-auto-sklearn">Will non-scikit-learn models be added to Auto-sklearn?</a></li>
-<li><a class="reference internal" href="#can-the-preprocessing-be-disabled">Can the preprocessing be disabled</a></li>
-</ul>
-</li>
-<li><a class="reference internal" href="#usage">Usage</a><ul>
-<li><a class="reference internal" href="#only-use-interpretable-models">Only use interpretable models</a></li>
-<li><a class="reference internal" href="#limiting-the-number-of-model-evaluations">Limiting the number of model evaluations</a></li>
-<li><a class="reference internal" href="#ensemble-contains-only-a-dummy-model">Ensemble contains only a dummy model</a></li>
-<li><a class="reference internal" href="#parallel-processing-and-oversubscription">Parallel processing and oversubscription</a></li>
-</ul>
-</li>
-<li><a class="reference internal" href="#meta-learning">Meta-Learning</a><ul>
-<li><a class="reference internal" href="#which-datasets-are-used-for-meta-learning">Which datasets are used for meta-learning?</a></li>
-<li><a class="reference internal" href="#how-can-datasets-from-the-meta-data-be-excluded">How can datasets from the meta-data be excluded?</a></li>
-<li><a class="reference internal" href="#which-meta-features-are-used-for-meta-learning">Which meta-features are used for meta-learning?</a></li>
-<li><a class="reference internal" href="#how-is-the-meta-data-generated">How is the meta-data generated?</a><ul>
-<li><a class="reference internal" href="#auto-sklearn-1-0">Auto-sklearn 1.0</a></li>
-<li><a class="reference internal" href="#auto-sklearn-2-0">Auto-sklearn 2.0</a></li>
-</ul>
-</li>
-</ul>
-</li>
+<li><a class="reference internal" href="#general">General</a></li>
+<li><a class="reference internal" href="#resource-management">Resource Management</a></li>
+<li><a class="reference internal" href="#results-log-files-and-output">Results, Log Files and Output</a></li>
+<li><a class="reference internal" href="#the-search-space">The Search Space</a></li>
+<li><a class="reference internal" href="#ensembling">Ensembling</a></li>
+<li><a class="reference internal" href="#configuring-the-search-procedure">Configuring the Search Procedure</a></li>
+<li><a class="reference internal" href="#meta-learning">Meta-Learning</a></li>
+<li><a class="reference internal" href="#issues-and-debugging">Issues and Debugging</a></li>
+<li><a class="reference internal" href="#other">Other</a></li>
 </ul>
 </li>
 </ul>
@@ -150,13 +124,115 @@
       </div>
     <div class="body col-md-9 content" role="main">
       
-  <section id="faq">
+  <div class="section" id="faq">
 <span id="id1"></span><h1>FAQ<a class="headerlink" href="#faq" title="Permalink to this headline">¶</a></h1>
-<section id="issues">
-<h2>Issues<a class="headerlink" href="#issues" title="Permalink to this headline">¶</a></h2>
-<section id="auto-sklearn-is-extremely-memory-hungry-in-a-sequential-setting">
-<h3>Auto-sklearn is extremely memory hungry in a sequential setting<a class="headerlink" href="#auto-sklearn-is-extremely-memory-hungry-in-a-sequential-setting" title="Permalink to this headline">¶</a></h3>
-<p>Auto-sklearn can appear very memory hungry (i.e. requiring a lot of memory for small datasets) due
+<div class="section" id="general">
+<h2>General<a class="headerlink" href="#general" title="Permalink to this headline">¶</a></h2>
+<details class="summary-b-where-can-i-find-examples-on-how-to-use-auto-sklearn-b">
+<summary><b>Where can I find examples on how to use auto-sklearn?</b></summary><p>We provide examples on using <em>auto-sklearn</em> for multiple use cases ranging from
+simple classification to advanced uses such as feature importance, parallel runs
+and customization. They can be found in the <a class="reference internal" href="examples/index.html#sphx-glr-examples"><span class="std std-ref">Examples</span></a>.</p>
+</details><details class="summary-b-what-type-of-tasks-can-auto-sklearn-tackle-b">
+<summary><b>What type of tasks can auto-sklearn tackle?</b></summary><p><em>auto-sklearn</em> can accept targets for the following tasks (more details on <a class="reference external" href="https://scikit-learn.org/stable/modules/multiclass.html">Sklearn algorithms</a>):</p>
+<ul class="simple">
+<li><p>Binary Classification</p></li>
+<li><p>Multiclass Classification</p></li>
+<li><p>Multilabel Classification</p></li>
+<li><p>Regression</p></li>
+<li><p>Multioutput Regression</p></li>
+</ul>
+<p>You can provide feature and target training pairs (X_train/y_train) to <em>auto-sklearn</em> to fit an
+ensemble of pipelines as described in the next section. This X_train/y_train dataset must belong
+to one of the supported formats: np.ndarray, pd.DataFrame, scipy.sparse.csr_matrix and python lists.
+Optionally, you can measure the ability of this fitted model to generalize to unseen data by
+providing an optional testing pair (X_test/Y_test). For further details, please refer to the
+Example <a class="reference internal" href="examples/40_advanced/example_pandas_train_test.html#sphx-glr-examples-40-advanced-example-pandas-train-test-py"><span class="std std-ref">Performance-over-time plot</span></a>.
+Supported formats for these training and testing pairs are: np.ndarray,
+pd.DataFrame, scipy.sparse.csr_matrix and python lists.</p>
+<p>If your data contains categorical values (in the features or targets), autosklearn will automatically encode your
+data using a <a class="reference external" href="https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.LabelEncoder.html">sklearn.preprocessing.LabelEncoder</a>
+for unidimensional data and a <a class="reference external" href="https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.OrdinalEncoder.html">sklearn.preprocessing.OrdinalEncoder</a>
+for multidimensional data.</p>
+<p>Regarding the features, there are two methods to guide <em>auto-sklearn</em> to properly encode categorical columns:</p>
+<ul class="simple">
+<li><p>Providing a X_train/X_test numpy array with the optional flag feat_type. For further details, you
+can check the Example <a class="reference internal" href="examples/40_advanced/example_feature_types.html#sphx-glr-examples-40-advanced-example-feature-types-py"><span class="std std-ref">Feature Types</span></a>.</p></li>
+<li><p>You can provide a pandas DataFrame, with properly formatted columns. If a column has numerical
+dtype, <em>auto-sklearn</em> will not encode it and it will be passed directly to scikit-learn. If the
+column has a categorical/boolean class, it will be encoded. If the column is of any other type
+(Object or Timeseries), an error will be raised. For further details on how to properly encode
+your data, you can check the Pandas Example
+<a class="reference external" href="https://pandas.pydata.org/pandas-docs/stable/user_guide/categorical.html">Working with categorical data</a>).
+If you are working with time series, it is recommended that you follow this approach
+<a class="reference external" href="https://stats.stackexchange.com/questions/311494/">Working with time data</a>.</p></li>
+</ul>
+<p>Regarding the targets (y_train/y_test), if the task involves a classification problem, such features will be
+automatically encoded. It is recommended to provide both y_train and y_test during fit, so that a common encoding
+is created between these splits (if only y_train is provided during fit, the categorical encoder will not be able
+to handle new classes that are exclusive to y_test). If the task is regression, no encoding happens on the
+targets.</p>
+</details><details class="summary-b-where-can-i-find-slides-and-notebooks-from-talks-and-tutorials-b">
+<summary><b>Where can I find slides and notebooks from talks and tutorials?</b></summary><p>We provide resources for talks, tutorials and presentations on <em>auto-sklearn</em> under <a class="reference external" href="https://github.com/automl/auto-sklearn-talks">auto-sklearn-talks</a></p>
+</details><details class="summary-b-how-should-i-cite-auto-sklearn-in-a-scientific-publication-b">
+<summary><b>How should I cite auto-sklearn in a scientific publication?</b></summary><p>If you’ve used auto-sklearn in scientific publications, we would appreciate citations.</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="nd">@inproceedings</span><span class="p">{</span><span class="n">feurer</span><span class="o">-</span><span class="n">neurips15a</span><span class="p">,</span>
+    <span class="n">title</span>     <span class="o">=</span> <span class="p">{</span><span class="n">Efficient</span> <span class="ow">and</span> <span class="n">Robust</span> <span class="n">Automated</span> <span class="n">Machine</span> <span class="n">Learning</span><span class="p">},</span>
+    <span class="n">author</span>    <span class="o">=</span> <span class="p">{</span><span class="n">Feurer</span><span class="p">,</span> <span class="n">Matthias</span> <span class="ow">and</span> <span class="n">Klein</span><span class="p">,</span> <span class="n">Aaron</span> <span class="ow">and</span> <span class="n">Eggensperger</span><span class="p">,</span> <span class="n">Katharina</span>  <span class="n">Springenberg</span><span class="p">,</span> <span class="n">Jost</span> <span class="ow">and</span> <span class="n">Blum</span><span class="p">,</span> <span class="n">Manuel</span> <span class="ow">and</span> <span class="n">Hutter</span><span class="p">,</span> <span class="n">Frank</span><span class="p">},</span>
+    <span class="n">booktitle</span> <span class="o">=</span> <span class="p">{</span><span class="n">Advances</span> <span class="ow">in</span> <span class="n">Neural</span> <span class="n">Information</span> <span class="n">Processing</span> <span class="n">Systems</span> <span class="mi">28</span> <span class="p">(</span><span class="mi">2015</span><span class="p">)},</span>
+    <span class="n">pages</span>     <span class="o">=</span> <span class="p">{</span><span class="mi">2962</span><span class="o">--</span><span class="mi">2970</span><span class="p">},</span>
+    <span class="n">year</span>      <span class="o">=</span> <span class="p">{</span><span class="mi">2015</span><span class="p">}</span>
+<span class="p">}</span>
+</pre></div>
+</div>
+<p>Or this, if you’ve used auto-sklearn 2.0 in your work:</p>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="nd">@article</span><span class="p">{</span><span class="n">feurer</span><span class="o">-</span><span class="n">arxiv20a</span><span class="p">,</span>
+    <span class="n">title</span>     <span class="o">=</span> <span class="p">{</span><span class="n">Auto</span><span class="o">-</span><span class="n">Sklearn</span> <span class="mf">2.0</span><span class="p">:</span> <span class="n">Hands</span><span class="o">-</span><span class="n">free</span> <span class="n">AutoML</span> <span class="n">via</span> <span class="n">Meta</span><span class="o">-</span><span class="n">Learning</span><span class="p">},</span>
+    <span class="n">author</span>    <span class="o">=</span> <span class="p">{</span><span class="n">Feurer</span><span class="p">,</span> <span class="n">Matthias</span> <span class="ow">and</span> <span class="n">Eggensperger</span><span class="p">,</span> <span class="n">Katharina</span> <span class="ow">and</span> <span class="n">Falkner</span><span class="p">,</span> <span class="n">Stefan</span> <span class="ow">and</span> <span class="n">Lindauer</span><span class="p">,</span> <span class="n">Marius</span> <span class="ow">and</span> <span class="n">Hutter</span><span class="p">,</span> <span class="n">Frank</span><span class="p">},</span>
+    <span class="n">booktitle</span> <span class="o">=</span> <span class="p">{</span><span class="n">arXiv</span><span class="p">:</span><span class="mf">2007.04074</span> <span class="p">[</span><span class="n">cs</span><span class="o">.</span><span class="n">LG</span><span class="p">]},</span>
+    <span class="n">year</span>      <span class="o">=</span> <span class="p">{</span><span class="mi">2020</span><span class="p">}</span>
+<span class="p">}</span>
+</pre></div>
+</div>
+</details><details class="summary-b-i-want-to-contribute-what-can-i-do-b">
+<summary><b>I want to contribute. What can I do?</b></summary><p>This sounds great. Please have a look at our <a class="reference external" href="https://github.com/automl/auto-sklearn/blob/master/CONTRIBUTING.md">contribution guide</a></p>
+</details><details class="summary-b-i-have-a-question-which-is-not-answered-here-what-should-i-do-b">
+<summary><b>I have a question which is not answered here. What should I do?</b></summary><p>Thanks a lot. We regularly update this section with questions from our issue tracker. So please use the
+<a class="reference external" href="https://github.com/automl/auto-sklearn/issues">issue tracker</a></p>
+</details></div>
+<div class="section" id="resource-management">
+<h2>Resource Management<a class="headerlink" href="#resource-management" title="Permalink to this headline">¶</a></h2>
+<details class="summary-b-how-should-i-set-the-time-and-memory-limits-b">
+<summary><b>How should I set the time and memory limits?</b></summary><p>While <em>auto-sklearn</em> alleviates manual hyperparameter tuning, the user still
+has to set memory and time limits. For most datasets a memory limit of 3GB or
+6GB as found on most modern computers is sufficient. For the time limits it
+is harder to give clear guidelines. If possible, a good default is a total
+time limit of one day, and a time limit of 30 minutes for a single run.</p>
+<p>Further guidelines can be found in
+<a class="reference external" href="https://github.com/automl/auto-sklearn/issues/142">auto-sklearn/issues/142</a>.</p>
+</details><details class="summary-b-how-many-cpu-cores-does-auto-sklearn-use-by-default-b">
+<summary><b>How many CPU cores does auto-sklearn use by default?</b></summary><p>By default, <em>auto-sklearn</em> uses <strong>one core</strong>. See also <a class="reference internal" href="manual.html#parallel"><span class="std std-ref">Parallel computation</span></a> on how to configure this.</p>
+</details><details class="summary-b-how-can-i-run-auto-sklearn-in-parallel-b">
+<summary><b>How can I run auto-sklearn in parallel?</b></summary><p>Nevertheless, <em>auto-sklearn</em> also supports parallel Bayesian optimization via the use of
+<a class="reference external" href="https://distributed.dask.org/">Dask.distributed</a>. By providing the arguments <code class="docutils literal notranslate"><span class="pre">n_jobs</span></code>
+to the estimator construction, one can control the number of cores available to <em>auto-sklearn</em>
+(As shown in the Example <a class="reference internal" href="examples/60_search/example_parallel_n_jobs.html#sphx-glr-examples-60-search-example-parallel-n-jobs-py"><span class="std std-ref">Parallel Usage  on a single machine</span></a>).
+Distributed processes are also supported by providing a custom client object to <em>auto-sklearn</em> like
+in the Example: <a class="reference internal" href="examples/60_search/example_parallel_manual_spawning_cli.html#sphx-glr-examples-60-search-example-parallel-manual-spawning-cli-py"><span class="std std-ref">Parallel Usage: Spawning workers from the command line</span></a>. When
+multiple cores are
+available, <em>auto-sklearn</em> will create a worker per core, and use the available workers to both search
+for better machine learning models as well as building an ensemble with them until the time resource
+is exhausted.</p>
+<p><strong>Note:</strong> <em>auto-sklearn</em> requires all workers to have access to a shared file system for storing training data and models.</p>
+<p><em>auto-sklearn</em> employs <a class="reference external" href="https://github.com/joblib/threadpoolctl/">threadpoolctl</a> to control the number of threads employed by scientific libraries like numpy or scikit-learn. This is done exclusively during the building procedure of models, not during inference. In particular, <em>auto-sklearn</em> allows each pipeline to use at most 1 thread during training. At predicting and scoring time this limitation is not enforced by <em>auto-sklearn</em>. You can control the number of resources
+employed by the pipelines by setting the following variables in your environment, prior to running <em>auto-sklearn</em>:</p>
+<div class="highlight-shell-session notranslate"><div class="highlight"><pre><span></span><span class="gp">$ </span><span class="nb">export</span> <span class="nv">OPENBLAS_NUM_THREADS</span><span class="o">=</span><span class="m">1</span>
+<span class="gp">$ </span><span class="nb">export</span> <span class="nv">MKL_NUM_THREADS</span><span class="o">=</span><span class="m">1</span>
+<span class="gp">$ </span><span class="nb">export</span> <span class="nv">OMP_NUM_THREADS</span><span class="o">=</span><span class="m">1</span>
+</pre></div>
+</div>
+<p>For further information about how scikit-learn handles multiprocessing, please check the <a class="reference external" href="https://scikit-learn.org/stable/computing/parallelism.html">Parallelism, resource management, and configuration</a> documentation from the library.</p>
+</details><details class="summary-b-auto-sklearn-is-extremely-memory-hungry-in-a-sequential-setting-b">
+<summary><b>Auto-sklearn is extremely memory hungry in a sequential setting</b></summary><p>Auto-sklearn can appear very memory hungry (i.e. requiring a lot of memory for small datasets) due
 to the use of <code class="docutils literal notranslate"><span class="pre">fork</span></code> for creating new processes when running in sequential manner (if this
 happens in a parallel setting or if you pass your own dask client this is due to a different
 issue, see the other issues below).</p>
@@ -197,19 +273,15 @@ <h3>Auto-sklearn is extremely memory hungry in a sequential setting<a class="hea
 </li>
 </ol>
 <p>We therefore suggest using one of the above settings by default.</p>
-</section>
-<section id="auto-sklearn-is-extremely-memory-hungry-in-a-parallel-setting">
-<h3>Auto-sklearn is extremely memory hungry in a parallel setting<a class="headerlink" href="#auto-sklearn-is-extremely-memory-hungry-in-a-parallel-setting" title="Permalink to this headline">¶</a></h3>
-<p>When running Auto-sklearn in a parallel setting it starts new processes for evaluating machine
+</details><details class="summary-b-auto-sklearn-is-extremely-memory-hungry-in-a-parallel-setting-b">
+<summary><b>Auto-sklearn is extremely memory hungry in a parallel setting</b></summary><p>When running Auto-sklearn in a parallel setting it starts new processes for evaluating machine
 learning models using the <code class="docutils literal notranslate"><span class="pre">forkserver</span></code> mechanism. Code that is in the main script and that is
 not guarded by <code class="docutils literal notranslate"><span class="pre">if</span> <span class="pre">__name__</span> <span class="pre">==</span> <span class="pre">&quot;__main__&quot;</span></code> will be executed for each subprocess. If, for example,
 you are loading your dataset outside of the guarded code, your dataset will be loaded for each
 evaluation of a machine learning algorithm and thus blocking your RAM.</p>
 <p>We therefore suggest moving all code inside functions or the main block.</p>
-</section>
-<section id="auto-sklearn-crashes-with-a-segmentation-fault">
-<h3>Auto-sklearn crashes with a segmentation fault<a class="headerlink" href="#auto-sklearn-crashes-with-a-segmentation-fault" title="Permalink to this headline">¶</a></h3>
-<p>Please make sure that you have read and followed the <a class="reference internal" href="installation.html#installation"><span class="std std-ref">Installation</span></a> section! In case
+</details><details class="summary-b-auto-sklearn-crashes-with-a-segmentation-fault-b">
+<summary><b>Auto-sklearn crashes with a segmentation fault</b></summary><p>Please make sure that you have read and followed the <a class="reference internal" href="installation.html#installation"><span class="std std-ref">Installation</span></a> section! In case
 everything is set up correctly, this is most likely due to the dependency
 <a class="reference external" href="https://github.com/automl/random_forest_run">pyrfr</a> not being compiled correctly. If this is the
 case please execute:</p>
@@ -230,27 +302,47 @@ <h3>Auto-sklearn crashes with a segmentation fault<a class="headerlink" href="#a
 github issues for suggestions: <a class="reference external" href="https://github.com/automl/auto-sklearn/issues/1025">1025</a>,
 <a class="reference external" href="https://github.com/automl/auto-sklearn/issues/856">856</a></p></li>
 </ol>
-</section>
-</section>
-<section id="log-files-and-output">
-<h2>Log files and output<a class="headerlink" href="#log-files-and-output" title="Permalink to this headline">¶</a></h2>
-<section id="where-does-auto-sklearn-output-files-by-default">
-<h3>Where does Auto-sklearn output files by default?<a class="headerlink" href="#where-does-auto-sklearn-output-files-by-default" title="Permalink to this headline">¶</a></h3>
-<p><em>Auto-sklearn</em> heavily uses the hard drive to store temporary data, models and log files which can
+</details></div>
+<div class="section" id="results-log-files-and-output">
+<h2>Results, Log Files and Output<a class="headerlink" href="#results-log-files-and-output" title="Permalink to this headline">¶</a></h2>
+<details class="summary-b-how-can-i-get-an-overview-of-the-run-statistics-b">
+<summary><b>How can I get an overview of the run statistics?</b></summary><p><code class="docutils literal notranslate"><span class="pre">sprint_statistics()</span></code> is a method that prints the name of the  dataset, the metric used, and the best validation score
+obtained by running <em>auto-sklearn</em>. It additionally prints the number of both successful and unsuccessful
+algorithm runs.</p>
+</details><details class="summary-b-what-was-the-performance-over-time-b">
+<summary><b>What was the performance over time?</b></summary><p><code class="docutils literal notranslate"><span class="pre">performance_over_time_</span></code>  returns a DataFrame containing the models performance over time data, which can
+be used for plotting directly (Here is an example: <a class="reference internal" href="examples/40_advanced/example_pandas_train_test.html#sphx-glr-examples-40-advanced-example-pandas-train-test-py"><span class="std std-ref">Performance-over-time plot</span></a>).</p>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="n">automl</span><span class="o">.</span><span class="n">performance_over_time_</span><span class="o">.</span><span class="n">plot</span><span class="p">(</span>
+        <span class="n">x</span><span class="o">=</span><span class="s1">&#39;Timestamp&#39;</span><span class="p">,</span>
+        <span class="n">kind</span><span class="o">=</span><span class="s1">&#39;line&#39;</span><span class="p">,</span>
+        <span class="n">legend</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+        <span class="n">title</span><span class="o">=</span><span class="s1">&#39;Auto-sklearn accuracy over time&#39;</span><span class="p">,</span>
+        <span class="n">grid</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+    <span class="p">)</span>
+    <span class="n">plt</span><span class="o">.</span><span class="n">show</span><span class="p">()</span>
+</pre></div>
+</div>
+</details><details class="summary-b-which-models-were-evaluated-b">
+<summary><b>Which models were evaluated?</b></summary><p>You can see all models evaluated using <a class="reference internal" href="api.html#autosklearn.classification.AutoSklearnClassifier.leaderboard" title="autosklearn.classification.AutoSklearnClassifier.leaderboard"><code class="xref py py-meth docutils literal notranslate"><span class="pre">automl.leaderboard(ensemble_only=False)</span></code></a>.</p>
+</details><details class="summary-b-which-models-are-in-the-final-ensemble-b">
+<summary><b>Which models are in the final ensemble?</b></summary><p>Use either <a class="reference internal" href="api.html#autosklearn.classification.AutoSklearnClassifier.leaderboard" title="autosklearn.classification.AutoSklearnClassifier.leaderboard"><code class="xref py py-meth docutils literal notranslate"><span class="pre">automl.leaderboard(ensemble_only=True)</span></code></a> or <code class="docutils literal notranslate"><span class="pre">automl.show_models()</span></code></p>
+</details><details class="summary-b-is-there-more-data-i-can-look-at-b">
+<summary><b>Is there more data I can look at?</b></summary><p><code class="docutils literal notranslate"><span class="pre">cv_results_</span></code> returns a dict with keys as column headers and values as columns, that can be imported into
+a pandas DataFrame, e.g. <code class="docutils literal notranslate"><span class="pre">df</span> <span class="pre">=</span> <span class="pre">pd.DataFrame(automl.cv_results_)</span></code></p>
+</details><details class="summary-b-where-does-auto-sklearn-output-files-by-default-b">
+<summary><b>Where does Auto-sklearn output files by default?</b></summary><p><em>Auto-sklearn</em> heavily uses the hard drive to store temporary data, models and log files which can
 be used to inspect the behavior of Auto-sklearn. Each run of Auto-sklearn requires
 its own directory. If not provided by the user, <em>Auto-sklearn</em> requests a temporary directory from
 Python, which by default is located under <code class="docutils literal notranslate"><span class="pre">/tmp</span></code> and starts with <code class="docutils literal notranslate"><span class="pre">autosklearn_tmp_</span></code> followed
 by a random string. By default, this directory is deleted when the <em>Auto-sklearn</em> object is
-destroyed. If you want to keep these files you can pass the argument
+finished fitting. If you want to keep these files you can pass the argument
 <code class="docutils literal notranslate"><span class="pre">delete_tmp_folder_after_terminate=True</span></code> to the <em>Auto-sklearn</em> object.</p>
 <p>The <a class="reference internal" href="api.html#autosklearn.classification.AutoSklearnClassifier" title="autosklearn.classification.AutoSklearnClassifier"><code class="xref py py-class docutils literal notranslate"><span class="pre">autosklearn.classification.AutoSklearnClassifier</span></code></a> and all other <em>auto-sklearn</em>
-estimators accept the argument <code class="docutils literal notranslate"><span class="pre">tmp_directory</span></code> which change where such output is written to.</p>
+estimators accept the argument <code class="docutils literal notranslate"><span class="pre">tmp_folder</span></code> which change where such output is written to.</p>
 <p>There’s an additional argument <code class="docutils literal notranslate"><span class="pre">output_directory</span></code> which can be passed to <em>Auto-sklearn</em> and it
 controls where test predictions of the ensemble are stored if the test set is passed to <code class="docutils literal notranslate"><span class="pre">fit()</span></code>.</p>
-</section>
-<section id="auto-sklearn-eats-up-all-my-disk-space">
-<h3>Auto-sklearn eats up all my disk space<a class="headerlink" href="#auto-sklearn-eats-up-all-my-disk-space" title="Permalink to this headline">¶</a></h3>
-<p><em>Auto-sklearn</em> heavily uses the hard drive to store temporary data, models and log files which can
+</details><details class="summary-b-auto-sklearn-s-logfiles-eat-up-all-my-disk-space-what-can-i-do-b">
+<summary><b>Auto-sklearn's logfiles eat up all my disk space. What can I do?</b></summary><p><em>Auto-sklearn</em> heavily uses the hard drive to store temporary data, models and log files which can
 be used to inspect the behavior of Auto-sklearn. By default, <em>Auto-sklearn</em> stores 50
 models and their predictions on the validation data (which is a subset of the training data in
 case of holdout and the full training data in case of cross-validation) on the hard drive.
@@ -274,13 +366,51 @@ <h3>Auto-sklearn eats up all my disk space<a class="headerlink" href="#auto-skle
 <p>These limits do only apply to models and their predictions, but not to other files stored in
 the temporary directory such as the log files.</p>
 </div>
-</section>
-</section>
-<section id="available-machine-learning-models">
-<h2>Available machine learning models<a class="headerlink" href="#available-machine-learning-models" title="Permalink to this headline">¶</a></h2>
-<section id="will-non-scikit-learn-models-be-added-to-auto-sklearn">
-<h3>Will non-scikit-learn models be added to Auto-sklearn?<a class="headerlink" href="#will-non-scikit-learn-models-be-added-to-auto-sklearn" title="Permalink to this headline">¶</a></h3>
-<p>The short answer: no.</p>
+</details></div>
+<div class="section" id="the-search-space">
+<h2>The Search Space<a class="headerlink" href="#the-search-space" title="Permalink to this headline">¶</a></h2>
+<details class="summary-b-how-can-i-restrict-the-searchspace-b">
+<summary><b>How can I restrict the searchspace?</b></summary><p>The following shows an example of how to exclude all preprocessing methods and restrict the configuration space to
+only random forests.</p>
+<blockquote>
+<div><div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">autosklearn.classification</span>
+<span class="n">automl</span> <span class="o">=</span> <span class="n">autosklearn</span><span class="o">.</span><span class="n">classification</span><span class="o">.</span><span class="n">AutoSklearnClassifier</span><span class="p">(</span>
+    <span class="n">include</span> <span class="o">=</span> <span class="p">{</span>
+        <span class="s1">&#39;classifier&#39;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;random_forest&quot;</span><span class="p">],</span>
+        <span class="s1">&#39;feature_preprocessor&#39;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;no_preprocessing&quot;</span><span class="p">]</span>
+    <span class="p">},</span>
+    <span class="n">exclude</span><span class="o">=</span><span class="kc">None</span>
+<span class="p">)</span>
+<span class="n">automl</span><span class="o">.</span><span class="n">fit</span><span class="p">(</span><span class="n">X_train</span><span class="p">,</span> <span class="n">y_train</span><span class="p">)</span>
+<span class="n">predictions</span> <span class="o">=</span> <span class="n">automl</span><span class="o">.</span><span class="n">predict</span><span class="p">(</span><span class="n">X_test</span><span class="p">)</span>
+</pre></div>
+</div>
+<p><strong>Note:</strong> The strings used to identify estimators and preprocessors are the filenames without <em>.py</em>.</p>
+<p>For a full list please have a look at the source code (in <cite>autosklearn/pipeline/components/</cite>):</p>
+<blockquote>
+<div><ul class="simple">
+<li><p><a class="reference external" href="https://github.com/automl/auto-sklearn/tree/master/autosklearn/pipeline/components/classification">Classifiers</a></p></li>
+<li><p><a class="reference external" href="https://github.com/automl/auto-sklearn/tree/master/autosklearn/pipeline/components/regression">Regressors</a></p></li>
+<li><p><a class="reference external" href="https://github.com/automl/auto-sklearn/tree/master/autosklearn/pipeline/components/feature_preprocessing">Preprocessors</a></p></li>
+</ul>
+</div></blockquote>
+<p>We do also provide an example on how to restrict the classifiers to search over
+<a class="reference internal" href="examples/40_advanced/example_interpretable_models.html#sphx-glr-examples-40-advanced-example-interpretable-models-py"><span class="std std-ref">Interpretable models</span></a>.</p>
+</div></blockquote>
+</details><details class="summary-b-how-can-i-turn-off-data-preprocessing-b">
+<summary><b>How can I turn off data preprocessing?</b></summary><p>Data preprocessing includes One-Hot encoding of categorical features, imputation
+of missing values and the normalization of features or samples. These ensure that
+the data the gets to the sklearn models is well formed and can be used for
+training models.</p>
+<p>While this is necessary in general, if you’d like to disable this step, please
+refer to this <a class="reference internal" href="examples/80_extending/example_extending_data_preprocessor.html#sphx-glr-examples-80-extending-example-extending-data-preprocessor-py"><span class="std std-ref">example</span></a>.</p>
+</details><details class="summary-b-how-can-i-turn-off-feature-preprocessing-b">
+<summary><b>How can I turn off feature preprocessing?</b></summary><p>Feature preprocessing is a single transformer which implements for example feature
+selection or transformation of features into a different space (i.e. PCA).</p>
+<p>This can be turned off by setting
+<code class="docutils literal notranslate"><span class="pre">include={'feature_preprocessor'=[&quot;no_preprocessing&quot;]}</span></code> as shown in the example above.</p>
+</details><details class="summary-b-will-non-scikit-learn-models-be-added-to-auto-sklearn-b">
+<summary><b>Will non-scikit-learn models be added to Auto-sklearn?</b></summary><p>The short answer: no.</p>
 <p>The long answer answer is a bit more nuanced: maintaining Auto-sklearn requires a lot of time and
 effort, which would grow even larger when depending on more libraries. Also, adding more
 libraries would require us to generate meta-data more often. Lastly, having more choices does not
@@ -290,47 +420,83 @@ <h3>Will non-scikit-learn models be added to Auto-sklearn?<a class="headerlink"
 following the <a class="reference external" href="https://automl.github.io/auto-sklearn/master/examples/index.html#extension-examples">examples on how to extend Auto-sklearn</a>.</p>
 <p>If there is interest in creating a Auto-sklearn-contrib repository with 3rd-party models please
 open an issue for that.</p>
-</section>
-<section id="can-the-preprocessing-be-disabled">
-<h3>Can the preprocessing be disabled<a class="headerlink" href="#can-the-preprocessing-be-disabled" title="Permalink to this headline">¶</a></h3>
-<p>Feature preprocessing can be disabled as discussed in the example
-<a class="reference internal" href="manual.html#restricting-the-searchspace"><span class="std std-ref">Restricting the searchspace</span></a>. Other preprocessing steps such as one hot encoding, missing
-feature imputation and normalization cannot yet be disabled, but we’re working on that.</p>
-</section>
-</section>
-<section id="usage">
-<h2>Usage<a class="headerlink" href="#usage" title="Permalink to this headline">¶</a></h2>
-<section id="only-use-interpretable-models">
-<h3>Only use interpretable models<a class="headerlink" href="#only-use-interpretable-models" title="Permalink to this headline">¶</a></h3>
-<p>Auto-sklearn can be restricted to only use interpretable models and preprocessing algorithms.
-Please see the Section <a class="reference internal" href="manual.html#restricting-the-searchspace"><span class="std std-ref">Restricting the searchspace</span></a> to learn how to restrict the models
+</details><details class="summary-b-how-can-i-only-search-for-interpretable-models-b">
+<summary><b>How can I only search for interpretable models</b></summary><p>Auto-sklearn can be restricted to only use interpretable models and preprocessing algorithms.
+Please see the Section <a class="reference internal" href="manual.html#space"><span class="std std-ref">The search space</span></a> to learn how to restrict the models
 which are searched over or see the Example
 <a class="reference internal" href="examples/40_advanced/example_interpretable_models.html#sphx-glr-examples-40-advanced-example-interpretable-models-py"><span class="std std-ref">Interpretable models</span></a>.</p>
 <p>We don’t provide a judgement which of the models are interpretable as this is very much up to the
 specific use case, but would like to note that decision trees and linear models usually most
 interpretable.</p>
-</section>
-<section id="limiting-the-number-of-model-evaluations">
-<h3>Limiting the number of model evaluations<a class="headerlink" href="#limiting-the-number-of-model-evaluations" title="Permalink to this headline">¶</a></h3>
-<p>In certain cases, for example for debugging, it can be helpful to limit the number of
+</details></div>
+<div class="section" id="ensembling">
+<h2>Ensembling<a class="headerlink" href="#ensembling" title="Permalink to this headline">¶</a></h2>
+<details class="summary-b-what-can-i-configure-wrt-the-ensemble-building-process-b">
+<summary><b>What can I configure wrt the ensemble building process?</b></summary><p>The following hyperparameters control how the ensemble is constructed:</p>
+<ul class="simple">
+<li><p><code class="docutils literal notranslate"><span class="pre">ensemble_size</span></code> determines the maximal size of the ensemble. If it is set to zero, no ensemble will be constructed.</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">ensemble_nbest</span></code> allows the user to directly specify the number of models considered for the ensemble.  This hyperparameter can be an integer <em>n</em>, such that only the best <em>n</em> models are used in the final ensemble. If a float between 0.0 and 1.0 is provided, <code class="docutils literal notranslate"><span class="pre">ensemble_nbest</span></code> would be interpreted as a fraction suggesting the percentage of models to use in the ensemble building process (namely, if ensemble_nbest is a float, library pruning is implemented as described in <a class="reference external" href="https://dl.acm.org/doi/10.1109/ICDM.2006.76">Caruana et al. (2006)</a>).</p></li>
+<li><p><code class="docutils literal notranslate"><span class="pre">max_models_on_disc</span></code> defines the maximum number of models that are kept on the disc, as a mechanism to control the amount of disc space consumed by <em>auto-sklearn</em>. Throughout the automl process, different individual models are optimized, and their predictions (and other metadata) is stored on disc. The user can set the upper bound on how many models are acceptable to keep on disc, yet this variable takes priority in the definition of the number of models used by the ensemble builder (that is, the minimum of <code class="docutils literal notranslate"><span class="pre">ensemble_size</span></code>, <code class="docutils literal notranslate"><span class="pre">ensemble_nbest</span></code> and <code class="docutils literal notranslate"><span class="pre">max_models_on_disc</span></code> determines the maximal amount of models used in the ensemble). If set to None, this feature is disabled.</p></li>
+</ul>
+</details><details class="summary-b-which-models-are-in-the-final-ensemble-b">
+<summary><b>Which models are in the final ensemble?</b></summary><p>The results obtained from the final ensemble can be printed by calling <code class="docutils literal notranslate"><span class="pre">show_models()</span></code> or  <code class="docutils literal notranslate"><span class="pre">leaderboard()</span></code>.
+The <em>auto-sklearn</em> ensemble is composed of scikit-learn models that can be inspected as exemplified
+in the Example <a class="reference internal" href="examples/40_advanced/example_get_pipeline_components.html#sphx-glr-examples-40-advanced-example-get-pipeline-components-py"><span class="std std-ref">Obtain run information</span></a>.</p>
+</details><details class="summary-b-can-i-fit-an-ensemble-also-only-post-hoc-b">
+<summary><b>Can I fit an ensemble also only post-hoc?</b></summary><p>It is possible to build ensembles post-hoc. An example on how to do this (first searching for individual models, and then building an ensemble from them) can be seen in <a class="reference internal" href="examples/60_search/example_sequential.html#sphx-glr-examples-60-search-example-sequential-py"><span class="std std-ref">Sequential Usage</span></a>.</p>
+</details></div>
+<div class="section" id="configuring-the-search-procedure">
+<h2>Configuring the Search Procedure<a class="headerlink" href="#configuring-the-search-procedure" title="Permalink to this headline">¶</a></h2>
+<details class="summary-b-can-i-change-the-resampling-strategy-b">
+<summary><b>Can I change the resampling strategy?</b></summary><p>Examples for using holdout and cross-validation can be found in <a class="reference internal" href="examples/40_advanced/example_resampling.html#sphx-glr-examples-40-advanced-example-resampling-py"><span class="std std-ref">example</span></a></p>
+</details><details class="summary-b-can-i-use-a-custom-metric-b">
+<summary><b>Can I use a custom metric</b></summary><p>Examples for using a custom metric can be found in <a class="reference internal" href="examples/40_advanced/example_metrics.html#sphx-glr-examples-40-advanced-example-metrics-py"><span class="std std-ref">example</span></a></p>
+</details></div>
+<div class="section" id="meta-learning">
+<h2>Meta-Learning<a class="headerlink" href="#meta-learning" title="Permalink to this headline">¶</a></h2>
+<details class="summary-b-which-datasets-are-used-for-meta-learning-b">
+<summary><b>Which datasets are used for meta-learning?</b></summary><p>We updated the list of datasets used for meta-learning several times and this list now differs
+significantly from the original 140 datasets we used in 2015 when the paper and the package were
+released. An up-to-date list of <a class="reference external" href="https://docs.openml.org/#tasks">OpenML task IDs</a> can be found
+on <a class="reference external" href="https://github.com/automl/auto-sklearn/blob/master/scripts/update_metadata_util.py">github</a>.</p>
+</details><details class="summary-b-how-can-datasets-from-the-meta-data-be-excluded-b">
+<summary><b>How can datasets from the meta-data be excluded?</b></summary><p>For <em>Auto-sklearn 1.0</em> one can pass the dataset name via the <code class="docutils literal notranslate"><span class="pre">fit()</span></code> function. If a dataset
+with the same name is within the meta-data, that datasets will not be used.</p>
+<p>For <em>Auto-sklearn 2.0</em> it is not possible to do so because of the method used to construct the
+meta-data.</p>
+</details><details class="summary-b-which-meta-features-are-used-for-meta-learning-b">
+<summary><b>Which meta-features are used for meta-learning?</b></summary><p>We do not have a user guide on meta-features but they are all pretty simple and can be found
+<a class="reference external" href="https://github.com/automl/auto-sklearn/blob/master/autosklearn/metalearning/metafeatures/metafeatures.py">in the source code</a>.</p>
+</details><details class="summary-b-how-is-the-meta-data-generated-for-auto-sklearn-1-0-b">
+<summary><b>How is the meta-data generated for Auto-sklearn 1.0?</b></summary><p>We currently generate meta-data the following way. First, for each of the datasets mentioned
+above, we run Auto-sklearn without meta-learning for a total of two days on multiple metrics (for
+classification these are accuracy, balanced accuracy, log loss and the area under the curce).
+Second, for each run we then have a look at each models that improved the score, i.e. the
+trajectory of the best known model at a time, and refit it on the whole training data. Third, for
+each of these models we then compute all scores we’re interested in, these also include other
+ones such F1 and precision. Finally, for each combination of dataset and metric we store the best
+model we know of.</p>
+</details><details class="summary-b-how-is-the-meta-data-generated-for-auto-sklearn-2-0-b">
+<summary><b>How is the meta-data generated for Auto-sklearn 2.0?</b></summary><p>Please check <a class="reference external" href="https://arxiv.org/abs/2007.04074">our paper</a> for details.</p>
+</details></div>
+<div class="section" id="issues-and-debugging">
+<h2>Issues and Debugging<a class="headerlink" href="#issues-and-debugging" title="Permalink to this headline">¶</a></h2>
+<details class="summary-b-how-can-i-limit-the-number-of-model-evaluations-for-debugging-b">
+<summary><b>How can I limit the number of model evaluations for debugging?</b></summary><p>In certain cases, for example for debugging, it can be helpful to limit the number of
 model evaluations. We do not provide this as an argument in the API as we believe that it
 should NOT be used in practice, but that the user should rather provide time limits.
 An example on how to add the number of models to try as an additional stopping condition
 can be found <a class="reference external" href="https://github.com/automl/auto-sklearn/issues/451#issuecomment-376445607">in this github issue</a>.
 Please note that Auto-sklearn will stop when either the time limit or the number of
 models termination condition is reached.</p>
-</section>
-<section id="ensemble-contains-only-a-dummy-model">
-<h3>Ensemble contains only a dummy model<a class="headerlink" href="#ensemble-contains-only-a-dummy-model" title="Permalink to this headline">¶</a></h3>
-<p>This is a symptom of the problem that all runs started by Auto-sklearn failed. Usually, the issue
+</details><details class="summary-b-why-does-the-final-ensemble-contains-only-a-dummy-model-b">
+<summary><b>Why does the final ensemble contains only a dummy model?</b></summary><p>This is a symptom of the problem that all runs started by Auto-sklearn failed. Usually, the issue
 is that the runtime or memory limit were too tight. Please check the output of
-<code class="docutils literal notranslate"><span class="pre">sprint_statistics</span></code> to see the distribution of why runs failed. If there are mostly crashed
+<code class="docutils literal notranslate"><span class="pre">sprint_statistics()</span></code> to see the distribution of why runs failed. If there are mostly crashed
 runs, please check the log file for further details. If there are mostly runs that exceed the
 memory or time limit, please increase the respective limit and rerun the optimization.</p>
-</section>
-<section id="parallel-processing-and-oversubscription">
-<h3>Parallel processing and oversubscription<a class="headerlink" href="#parallel-processing-and-oversubscription" title="Permalink to this headline">¶</a></h3>
-<p>Auto-sklearn wraps scikit-learn and therefore inherits its parallelism implementation. In short,
+</details><details class="summary-b-auto-sklearn-does-not-use-the-specified-amount-of-resources-b">
+<summary><b>Auto-sklearn does not use the specified amount of resources?</b></summary><p>Auto-sklearn wraps scikit-learn and therefore inherits its parallelism implementation. In short,
 scikit-learn uses two modes of parallelizing computations:</p>
 <ol class="arabic simple">
 <li><p>By using joblib to distribute independent function calls on multiple cores.</p></li>
@@ -346,49 +512,30 @@ <h3>Parallel processing and oversubscription<a class="headerlink" href="#paralle
 variables: <code class="docutils literal notranslate"><span class="pre">MKL_NUM_THREADS</span></code>, <code class="docutils literal notranslate"><span class="pre">OPENBLAS_NUM_THREADS</span></code>, <code class="docutils literal notranslate"><span class="pre">BLIS_NUM_THREADS</span></code> and
 <code class="docutils literal notranslate"><span class="pre">OMP_NUM_THREADS</span></code>.</p>
 <p>More details can be found in the <a class="reference external" href="https://scikit-learn.org/stable/computing/parallelism.html?highlight=joblib#parallelism">scikit-learn docs</a>.</p>
-</section>
-</section>
-<section id="meta-learning">
-<h2>Meta-Learning<a class="headerlink" href="#meta-learning" title="Permalink to this headline">¶</a></h2>
-<section id="which-datasets-are-used-for-meta-learning">
-<h3>Which datasets are used for meta-learning?<a class="headerlink" href="#which-datasets-are-used-for-meta-learning" title="Permalink to this headline">¶</a></h3>
-<p>We updated the list of datasets used for meta-learning several times and this list now differs
-significantly from the original 140 datasets we used in 2015 when the paper and the package were
-released. An up-to-date list of <a class="reference external" href="https://docs.openml.org/#tasks">OpenML task IDs</a> can be found
-on <a class="reference external" href="https://github.com/automl/auto-sklearn/blob/master/scripts/update_metadata_util.py">github</a>.</p>
-</section>
-<section id="how-can-datasets-from-the-meta-data-be-excluded">
-<h3>How can datasets from the meta-data be excluded?<a class="headerlink" href="#how-can-datasets-from-the-meta-data-be-excluded" title="Permalink to this headline">¶</a></h3>
-<p>For <em>Auto-sklearn 1.0</em> one can pass the dataset name via the <code class="docutils literal notranslate"><span class="pre">fit()</span></code> function. If a dataset
-with the same name is within the meta-data, that datasets will not be used.</p>
-<p>For <em>Auto-sklearn 2.0</em> it is not possible to do so because of the method used to construct the
-meta-data.</p>
-</section>
-<section id="which-meta-features-are-used-for-meta-learning">
-<h3>Which meta-features are used for meta-learning?<a class="headerlink" href="#which-meta-features-are-used-for-meta-learning" title="Permalink to this headline">¶</a></h3>
-<p>We do not have a user guide on meta-features but they are all pretty simple and can be found
-<a class="reference external" href="https://github.com/automl/auto-sklearn/blob/master/autosklearn/metalearning/metafeatures/metafeatures.py">in the source code</a>.</p>
-</section>
-<section id="how-is-the-meta-data-generated">
-<h3>How is the meta-data generated?<a class="headerlink" href="#how-is-the-meta-data-generated" title="Permalink to this headline">¶</a></h3>
-<section id="auto-sklearn-1-0">
-<h4>Auto-sklearn 1.0<a class="headerlink" href="#auto-sklearn-1-0" title="Permalink to this headline">¶</a></h4>
-<p>We currently generate meta-data the following way. First, for each of the datasets mentioned
-above, we run Auto-sklearn without meta-learning for a total of two days on multiple metrics (for
-classification these are accuracy, balanced accuracy, log loss and the area under the curce).
-Second, for each run we then have a look at each models that improved the score, i.e. the
-trajectory of the best known model at a time, and refit it on the whole training data. Third, for
-each of these models we then compute all scores we’re interested in, these also include other
-ones such F1 and precision. Finally, for each combination of dataset and metric we store the best
-model we know of.</p>
-</section>
-<section id="auto-sklearn-2-0">
-<h4>Auto-sklearn 2.0<a class="headerlink" href="#auto-sklearn-2-0" title="Permalink to this headline">¶</a></h4>
-<p>Please check <a class="reference external" href="https://arxiv.org/abs/2007.04074">our paper</a> for details.</p>
-</section>
-</section>
-</section>
-</section>
+</details></div>
+<div class="section" id="other">
+<h2>Other<a class="headerlink" href="#other" title="Permalink to this headline">¶</a></h2>
+<details class="summary-b-model-persistence-b">
+<summary><b>Model persistence</b></summary><p><em>auto-sklearn</em> is mostly a wrapper around scikit-learn. Therefore, it is
+possible to follow the
+<a class="reference external" href="https://scikit-learn.org/stable/modules/model_persistence.html">persistence Example</a>
+from scikit-learn.</p>
+</details><details class="summary-b-vanilla-auto-sklearn-b">
+<summary><b>Vanilla auto-sklearn</b></summary><p>In order to obtain <em>vanilla auto-sklearn</em> as used in <a class="reference external" href="https://papers.neurips.cc/paper/5872-efficient-and-robust-automated-machine-learning">Efficient and Robust Automated Machine Learning</a>
+set <code class="docutils literal notranslate"><span class="pre">ensemble_size=1</span></code> and <code class="docutils literal notranslate"><span class="pre">initial_configurations_via_metalearning=0</span></code>:</p>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">autosklearn.classification</span>
+<span class="n">automl</span> <span class="o">=</span> <span class="n">autosklearn</span><span class="o">.</span><span class="n">classification</span><span class="o">.</span><span class="n">AutoSklearnClassifier</span><span class="p">(</span>
+    <span class="n">ensemble_size</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span>
+    <span class="n">initial_configurations_via_metalearning</span><span class="o">=</span><span class="mi">0</span>
+<span class="p">)</span>
+</pre></div>
+</div>
+<p>An ensemble of size one will result in always choosing the current best model
+according to its performance on the validation set. Setting the initial
+configurations found by meta-learning to zero makes <em>auto-sklearn</em> use the
+regular SMAC algorithm for suggesting new hyperparameter configurations.</p>
+</details></div>
+</div>
 
 
     </div>
@@ -409,7 +556,7 @@ <h4>Auto-sklearn 2.0<a class="headerlink" href="#auto-sklearn-2-0" title="Permal
       
     </p>
     <p>
-        &copy; Copyright 2014-2021, Machine Learning Professorship Freiburg.<br/>
+        &copy; Copyright 2014-2022, Machine Learning Professorship Freiburg.<br/>
       Created using <a href="http://sphinx-doc.org/">Sphinx</a> 4.2.0.<br/>
     </p>
   </div>
diff --git a/master/genindex.html b/master/genindex.html
index addca1eaa2..4ce23cd317 100644
--- a/master/genindex.html
+++ b/master/genindex.html
@@ -4,7 +4,7 @@
   <head>
     <meta charset="utf-8" />
     <meta name="viewport" content="width=device-width, initial-scale=1.0" />
-    <title>Index &#8212; AutoSklearn 0.14.2 documentation</title>
+    <title>Index &#8212; AutoSklearn 0.14.4 documentation</title>
     <link rel="stylesheet" type="text/css" href="_static/pygments.css" />
     <link rel="stylesheet" type="text/css" href="_static/bootstrap-sphinx.css" />
     <link rel="stylesheet" type="text/css" href="_static/sg_gallery.css" />
@@ -54,7 +54,7 @@
         </button>
         <a class="navbar-brand" href="index.html">
           auto-sklearn</a>
-        <span class="navbar-text navbar-version pull-left"><b>0.14.2</b></span>
+        <span class="navbar-text navbar-version pull-left"><b>0.14.4</b></span>
       </div>
 
         <div class="collapse navbar-collapse nav-collapse">
@@ -412,7 +412,7 @@ <h2 id="T">T</h2>
       
     </p>
     <p>
-        &copy; Copyright 2014-2021, Machine Learning Professorship Freiburg.<br/>
+        &copy; Copyright 2014-2022, Machine Learning Professorship Freiburg.<br/>
       Created using <a href="http://sphinx-doc.org/">Sphinx</a> 4.2.0.<br/>
     </p>
   </div>
diff --git a/master/index.html b/master/index.html
index d48e27f003..7467bc637e 100644
--- a/master/index.html
+++ b/master/index.html
@@ -3,9 +3,8 @@
 <html>
   <head>
     <meta charset="utf-8" />
-    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="generator" content="Docutils 0.17.1: http://docutils.sourceforge.net/" />
-
-    <title>auto-sklearn &#8212; AutoSklearn 0.14.2 documentation</title>
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>auto-sklearn &#8212; AutoSklearn 0.14.4 documentation</title>
     <link rel="stylesheet" type="text/css" href="_static/pygments.css" />
     <link rel="stylesheet" type="text/css" href="_static/bootstrap-sphinx.css" />
     <link rel="stylesheet" type="text/css" href="_static/sg_gallery.css" />
@@ -55,7 +54,7 @@
         </button>
         <a class="navbar-brand" href="#">
           auto-sklearn</a>
-        <span class="navbar-text navbar-version pull-left"><b>0.14.2</b></span>
+        <span class="navbar-text navbar-version pull-left"><b>0.14.4</b></span>
       </div>
 
         <div class="collapse navbar-collapse nav-collapse">
@@ -121,7 +120,7 @@
       </div>
     <div class="body col-md-9 content" role="main">
       
-  <section id="auto-sklearn">
+  <div class="section" id="auto-sklearn">
 <h1>auto-sklearn<a class="headerlink" href="#auto-sklearn" title="Permalink to this headline">¶</a></h1>
 <p><em>auto-sklearn</em> is an automated machine learning toolkit and a drop-in
 replacement for a scikit-learn estimator:</p>
@@ -135,7 +134,7 @@ <h1>auto-sklearn<a class="headerlink" href="#auto-sklearn" title="Permalink to t
 hyperparameter tuning. It leverages recent advantages in <em>Bayesian
 optimization</em>, <em>meta-learning</em> and <em>ensemble construction</em>. Learn more about
 the technology behind <em>auto-sklearn</em> by reading our paper published at
-<a class="reference external" href="https://papers.neurips.cc/paper/5872-efficient-and-robust-automated-machine-learning.pdf">NIPS 2015</a>
+<a class="reference external" href="https://papers.neurips.cc/paper/5872-efficient-and-robust-automated-machine-learning.pdf">NeurIPS 2015</a>
 .</p>
 <div class="topic">
 <p class="topic-title">NEW: Auto-sklearn 2.0</p>
@@ -147,7 +146,11 @@ <h1>auto-sklearn<a class="headerlink" href="#auto-sklearn" title="Permalink to t
 </pre></div>
 </div>
 <p>A paper describing our advances is available on <a class="reference external" href="https://arxiv.org/abs/2007.04074">arXiv</a>.</p>
-<section id="example">
+<div class="topic">
+<p class="topic-title">NEW: Material from tutorials and presentations</p>
+<p>We provide slides and notebooks from talks and tutorials here: <a class="reference external" href="https://github.com/automl/auto-sklearn-talks">auto-sklearn-talks</a></p>
+</div>
+<div class="section" id="example">
 <h2>Example<a class="headerlink" href="#example" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">autosklearn.classification</span>
 <span class="kn">import</span> <span class="nn">sklearn.model_selection</span>
@@ -165,8 +168,8 @@ <h2>Example<a class="headerlink" href="#example" title="Permalink to this headli
 </pre></div>
 </div>
 <p>This will run for one hour and should result in an accuracy above 0.98.</p>
-</section>
-<section id="manual">
+</div>
+<div class="section" id="manual">
 <h2>Manual<a class="headerlink" href="#manual" title="Permalink to this headline">¶</a></h2>
 <ul class="simple">
 <li><p><a class="reference internal" href="installation.html#installation"><span class="std std-ref">Installation</span></a></p></li>
@@ -175,13 +178,13 @@ <h2>Manual<a class="headerlink" href="#manual" title="Permalink to this headline
 <li><p><a class="reference internal" href="extending.html#extending"><span class="std std-ref">Extending auto-sklearn</span></a></p></li>
 <li><p><a class="reference internal" href="faq.html#faq"><span class="std std-ref">FAQ</span></a></p></li>
 </ul>
-</section>
-<section id="license">
+</div>
+<div class="section" id="license">
 <h2>License<a class="headerlink" href="#license" title="Permalink to this headline">¶</a></h2>
 <p><em>auto-sklearn</em> is licensed the same way as <em>scikit-learn</em>,
 namely the 3-clause BSD license.</p>
-</section>
-<section id="citing-auto-sklearn">
+</div>
+<div class="section" id="citing-auto-sklearn">
 <h2>Citing auto-sklearn<a class="headerlink" href="#citing-auto-sklearn" title="Permalink to this headline">¶</a></h2>
 <p>If you use auto-sklearn in a scientific publication, we would appreciate a
 reference to the following paper:</p>
@@ -212,8 +215,8 @@ <h2>Citing auto-sklearn<a class="headerlink" href="#citing-auto-sklearn" title="
 </pre></div>
 </div>
 </div></blockquote>
-</section>
-<section id="contributing">
+</div>
+<div class="section" id="contributing">
 <h2>Contributing<a class="headerlink" href="#contributing" title="Permalink to this headline">¶</a></h2>
 <p>We appreciate all contribution to auto-sklearn, from bug reports and
 documentation to new features. If you want to contribute to the code, you can
@@ -226,8 +229,8 @@ <h2>Contributing<a class="headerlink" href="#contributing" title="Permalink to t
 get merged, it is highly advised that you contact the developers
 by opening a <a class="reference external" href="https://github.com/automl/auto-sklearn/issues">github issue</a> before starting to work.</p>
 </div>
-</section>
-</section>
+</div>
+</div>
 
 
     </div>
@@ -248,7 +251,7 @@ <h2>Contributing<a class="headerlink" href="#contributing" title="Permalink to t
       
     </p>
     <p>
-        &copy; Copyright 2014-2021, Machine Learning Professorship Freiburg.<br/>
+        &copy; Copyright 2014-2022, Machine Learning Professorship Freiburg.<br/>
       Created using <a href="http://sphinx-doc.org/">Sphinx</a> 4.2.0.<br/>
     </p>
   </div>
diff --git a/master/installation.html b/master/installation.html
index 16340d55cb..1f1605d378 100644
--- a/master/installation.html
+++ b/master/installation.html
@@ -3,9 +3,8 @@
 <html>
   <head>
     <meta charset="utf-8" />
-    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="generator" content="Docutils 0.17.1: http://docutils.sourceforge.net/" />
-
-    <title>Installation &#8212; AutoSklearn 0.14.2 documentation</title>
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>Installation &#8212; AutoSklearn 0.14.4 documentation</title>
     <link rel="stylesheet" type="text/css" href="_static/pygments.css" />
     <link rel="stylesheet" type="text/css" href="_static/bootstrap-sphinx.css" />
     <link rel="stylesheet" type="text/css" href="_static/sg_gallery.css" />
@@ -55,7 +54,7 @@
         </button>
         <a class="navbar-brand" href="index.html">
           auto-sklearn</a>
-        <span class="navbar-text navbar-version pull-left"><b>0.14.2</b></span>
+        <span class="navbar-text navbar-version pull-left"><b>0.14.4</b></span>
       </div>
 
         <div class="collapse navbar-collapse nav-collapse">
@@ -115,9 +114,10 @@
 <li><a class="reference internal" href="#conda-forge">Conda-forge</a></li>
 </ul>
 </li>
-<li><a class="reference internal" href="#windows-osx-compatibility">Windows/OSX compatibility</a><ul>
+<li><a class="reference internal" href="#source-installation">Source Installation</a></li>
+<li><a class="reference internal" href="#windows-macos-compatibility">Windows/macOS compatibility</a><ul>
 <li><a class="reference internal" href="#windows">Windows</a></li>
-<li><a class="reference internal" href="#mac-osx">Mac OSX</a></li>
+<li><a class="reference internal" href="#macos">macOS</a></li>
 </ul>
 </li>
 <li><a class="reference internal" href="#docker-image">Docker Image</a></li>
@@ -129,9 +129,9 @@
       </div>
     <div class="body col-md-9 content" role="main">
       
-  <section id="installation">
+  <div class="section" id="installation">
 <span id="id1"></span><h1>Installation<a class="headerlink" href="#installation" title="Permalink to this headline">¶</a></h1>
-<section id="system-requirements">
+<div class="section" id="system-requirements">
 <h2>System requirements<a class="headerlink" href="#system-requirements" title="Permalink to this headline">¶</a></h2>
 <p>auto-sklearn has the following system requirements:</p>
 <ul class="simple">
@@ -145,10 +145,10 @@ <h2>System requirements<a class="headerlink" href="#system-requirements" title="
 <ul class="simple">
 <li><p>SWIG (<a class="reference external" href="http://www.swig.org/survey.html">get SWIG here</a>).</p></li>
 </ul>
-<p>For an explanation of missing Microsoft Windows and MAC OSX support please
-check the Section <a class="reference internal" href="#windows-osx-compatibility">Windows/OSX compatibility</a>.</p>
-</section>
-<section id="installing-auto-sklearn">
+<p>For an explanation of missing Microsoft Windows and macOS support please
+check the Section <a class="reference internal" href="#windows-macos-compatibility">Windows/macOS compatibility</a>.</p>
+</div>
+<div class="section" id="installing-auto-sklearn">
 <h2>Installing auto-sklearn<a class="headerlink" href="#installing-auto-sklearn" title="Permalink to this headline">¶</a></h2>
 <p>You can install <em>auto-sklearn</em> with <cite>pip</cite> in the usual manner:</p>
 <div class="highlight-bash notranslate"><div class="highlight"><pre><span></span>pip3 install auto-sklearn
@@ -159,16 +159,16 @@ <h2>Installing auto-sklearn<a class="headerlink" href="#installing-auto-sklearn"
 or an
 <a class="reference external" href="https://conda.io/projects/conda/en/latest/user-guide/tasks/manage-environments.html">Anaconda environment</a>.</p>
 <p>If the <code class="docutils literal notranslate"><span class="pre">pip3</span></code> installation command fails, make sure you have the <a class="reference internal" href="#system-requirements">System requirements</a> installed correctly.</p>
-</section>
-<section id="ubuntu-installation">
+</div>
+<div class="section" id="ubuntu-installation">
 <h2>Ubuntu installation<a class="headerlink" href="#ubuntu-installation" title="Permalink to this headline">¶</a></h2>
 <p>To provide Python 3, a C++11 building environment and the latest SWIG version on Ubuntu,
 run:</p>
 <div class="highlight-bash notranslate"><div class="highlight"><pre><span></span>sudo apt-get install build-essential swig python3-dev
 </pre></div>
 </div>
-</section>
-<section id="anaconda-installation">
+</div>
+<div class="section" id="anaconda-installation">
 <h2>Anaconda installation<a class="headerlink" href="#anaconda-installation" title="Permalink to this headline">¶</a></h2>
 <p>You need to enable conda-forge to install <em>auto-sklearn</em> via anaconda. This section explains how to enable conda-forge so
 installation can be done with the command <cite>conda install auto-sklearn</cite>.
@@ -181,7 +181,7 @@ <h2>Anaconda installation<a class="headerlink" href="#anaconda-installation" tit
 <div class="highlight-bash notranslate"><div class="highlight"><pre><span></span>conda install gxx_linux-64 gcc_linux-64 swig
 </pre></div>
 </div>
-<section id="conda-forge">
+<div class="section" id="conda-forge">
 <h3>Conda-forge<a class="headerlink" href="#conda-forge" title="Permalink to this headline">¶</a></h3>
 <p>Installing <cite>auto-sklearn</cite> from the <cite>conda-forge</cite> channel can be achieved by adding <cite>conda-forge</cite> to your channels with:</p>
 <div class="highlight-bash notranslate"><div class="highlight"><pre><span></span>conda config --add channels conda-forge
@@ -201,11 +201,31 @@ <h3>Conda-forge<a class="headerlink" href="#conda-forge" title="Permalink to thi
 <a class="reference external" href="https://github.com/conda-forge/auto-sklearn-feedstock">auto sklearn feedstock</a>.</p>
 <p>for more information about Conda forge check
 <a class="reference external" href="https://conda-forge.org/docs/">conda-forge documentations</a>.</p>
-</section>
-</section>
-<section id="windows-osx-compatibility">
-<h2>Windows/OSX compatibility<a class="headerlink" href="#windows-osx-compatibility" title="Permalink to this headline">¶</a></h2>
-<section id="windows">
+</div>
+</div>
+<div class="section" id="source-installation">
+<h2>Source Installation<a class="headerlink" href="#source-installation" title="Permalink to this headline">¶</a></h2>
+<p>You can install auto-sklearn directly form source by following the below:</p>
+<div class="highlight-bash notranslate"><div class="highlight"><pre><span></span>git clone --recurse-submodules git@github.com:automl/auto-sklearn.git
+<span class="nb">cd</span> auto-sklearn
+
+<span class="c1"># Install it in editable mode with all optional dependencies</span>
+pip install -e <span class="s2">&quot;.[test,doc,examples]&quot;</span>
+</pre></div>
+</div>
+<p>We use submodules so you will have to make sure the submodule is initialized if you
+missed the <cite>–recurse-submodules</cite> option.</p>
+<div class="highlight-bash notranslate"><div class="highlight"><pre><span></span>git clone git@github.com:automl/auto-sklearn.git
+<span class="nb">cd</span> auto-sklearn
+git submodule update --init --recursive
+
+pip install -e <span class="s2">&quot;.[test,doc,examples]&quot;</span>
+</pre></div>
+</div>
+</div>
+<div class="section" id="windows-macos-compatibility">
+<h2>Windows/macOS compatibility<a class="headerlink" href="#windows-macos-compatibility" title="Permalink to this headline">¶</a></h2>
+<div class="section" id="windows">
 <h3>Windows<a class="headerlink" href="#windows" title="Permalink to this headline">¶</a></h3>
 <p><em>auto-sklearn</em> relies heavily on the Python module <code class="docutils literal notranslate"><span class="pre">resource</span></code>. <code class="docutils literal notranslate"><span class="pre">resource</span></code>
 is part of Python’s <a class="reference external" href="https://docs.python.org/3/library/unix.html">Unix Specific Services</a>
@@ -218,15 +238,15 @@ <h3>Windows<a class="headerlink" href="#windows" title="Permalink to this headli
 <li><p>virtual machine</p></li>
 <li><p>docker image</p></li>
 </ul>
-</section>
-<section id="mac-osx">
-<h3>Mac OSX<a class="headerlink" href="#mac-osx" title="Permalink to this headline">¶</a></h3>
-<p>We currently do not know if <em>auto-sklearn</em> works on OSX. There are at least two
-issues holding us back from actively supporting OSX:</p>
+</div>
+<div class="section" id="macos">
+<h3>macOS<a class="headerlink" href="#macos" title="Permalink to this headline">¶</a></h3>
+<p>We currently do not know if <em>auto-sklearn</em> works on macOS. There are at least two
+issues holding us back from actively supporting macOS:</p>
 <ul class="simple">
 <li><p>The <code class="docutils literal notranslate"><span class="pre">resource</span></code> module cannot enforce a memory limit on a Python process
 (see <a class="reference external" href="https://github.com/automl/SMAC3/issues/115">SMAC3/issues/115</a>).</p></li>
-<li><p>Not all dependencies we are using are set up to work on OSX.</p></li>
+<li><p>Not all dependencies we are using are set up to work on macOS.</p></li>
 </ul>
 <p>In case you’re having issues installing the <a class="reference external" href="https://github.com/automl/random_forest_run">pyrfr package</a>, check out
 <a class="reference external" href="https://github.com/automl/auto-sklearn/issues/360#issuecomment-335150470">this installation suggestion on github</a>.</p>
@@ -235,9 +255,9 @@ <h3>Mac OSX<a class="headerlink" href="#mac-osx" title="Permalink to this headli
 <li><p>virtual machine</p></li>
 <li><p>docker image</p></li>
 </ul>
-</section>
-</section>
-<section id="docker-image">
+</div>
+</div>
+<div class="section" id="docker-image">
 <h2>Docker Image<a class="headerlink" href="#docker-image" title="Permalink to this headline">¶</a></h2>
 <p>A Docker image is also provided on dockerhub. To download from dockerhub,
 use:</p>
@@ -258,8 +278,8 @@ <h2>Docker Image<a class="headerlink" href="#docker-image" title="Permalink to t
 </div>
 <p>Alternatively, it is possible to use the development version of auto-sklearn by replacing all
 occurences of <code class="docutils literal notranslate"><span class="pre">master</span></code> by <code class="docutils literal notranslate"><span class="pre">development</span></code>.</p>
-</section>
-</section>
+</div>
+</div>
 
 
     </div>
@@ -280,7 +300,7 @@ <h2>Docker Image<a class="headerlink" href="#docker-image" title="Permalink to t
       
     </p>
     <p>
-        &copy; Copyright 2014-2021, Machine Learning Professorship Freiburg.<br/>
+        &copy; Copyright 2014-2022, Machine Learning Professorship Freiburg.<br/>
       Created using <a href="http://sphinx-doc.org/">Sphinx</a> 4.2.0.<br/>
     </p>
   </div>
diff --git a/master/manual.html b/master/manual.html
index e09ed588ba..7cb83fdfbf 100644
--- a/master/manual.html
+++ b/master/manual.html
@@ -3,9 +3,8 @@
 <html>
   <head>
     <meta charset="utf-8" />
-    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="generator" content="Docutils 0.17.1: http://docutils.sourceforge.net/" />
-
-    <title>Manual &#8212; AutoSklearn 0.14.2 documentation</title>
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>Manual &#8212; AutoSklearn 0.14.4 documentation</title>
     <link rel="stylesheet" type="text/css" href="_static/pygments.css" />
     <link rel="stylesheet" type="text/css" href="_static/bootstrap-sphinx.css" />
     <link rel="stylesheet" type="text/css" href="_static/sg_gallery.css" />
@@ -55,7 +54,7 @@
         </button>
         <a class="navbar-brand" href="index.html">
           auto-sklearn</a>
-        <span class="navbar-text navbar-version pull-left"><b>0.14.2</b></span>
+        <span class="navbar-text navbar-version pull-left"><b>0.14.4</b></span>
       </div>
 
         <div class="collapse navbar-collapse nav-collapse">
@@ -108,20 +107,13 @@
       <div class="col-md-3">
         <div id="sidebar" class="bs-sidenav" role="complementary"><ul>
 <li><a class="reference internal" href="#">Manual</a><ul>
-<li><a class="reference internal" href="#examples">Examples</a></li>
-<li><a class="reference internal" href="#time-and-memory-limits">Time and memory limits</a></li>
-<li><a class="reference internal" href="#restricting-the-searchspace">Restricting the searchspace</a><ul>
-<li><a class="reference internal" href="#data-preprocessing">Data preprocessing</a></li>
-<li><a class="reference internal" href="#feature-preprocessing">Feature preprocessing</a></li>
-</ul>
-</li>
-<li><a class="reference internal" href="#resampling-strategies">Resampling strategies</a></li>
-<li><a class="reference internal" href="#supported-inputs">Supported Inputs</a></li>
-<li><a class="reference internal" href="#ensemble-building-process">Ensemble Building Process</a></li>
+<li><a class="reference internal" href="#resource-limits">Resource limits</a></li>
+<li><a class="reference internal" href="#the-search-space">The search space</a></li>
+<li><a class="reference internal" href="#model-selection">Model selection</a></li>
+<li><a class="reference internal" href="#ensembling">Ensembling</a></li>
 <li><a class="reference internal" href="#inspecting-the-results">Inspecting the results</a></li>
 <li><a class="reference internal" href="#parallel-computation">Parallel computation</a></li>
-<li><a class="reference internal" href="#model-persistence">Model persistence</a></li>
-<li><a class="reference internal" href="#vanilla-auto-sklearn">Vanilla auto-sklearn</a></li>
+<li><a class="reference internal" href="#other">Other</a></li>
 </ul>
 </li>
 </ul>
@@ -130,40 +122,43 @@
       </div>
     <div class="body col-md-9 content" role="main">
       
-  <section id="manual">
+  <div class="section" id="manual">
 <span id="id1"></span><h1>Manual<a class="headerlink" href="#manual" title="Permalink to this headline">¶</a></h1>
-<p>This manual shows how to use several aspects of auto-sklearn. It either
-references the examples where possible or explains certain configurations.</p>
-<section id="examples">
-<h2>Examples<a class="headerlink" href="#examples" title="Permalink to this headline">¶</a></h2>
-<p>We provide examples on using <em>auto-sklearn</em> for multiple use cases ranging from
+<p>This manual gives an overview of different aspects of <em>auto-sklearn</em>. For each section, we either references examples or
+give short explanations (click the title to expand text), e.g.</p>
+<details class="summary-b-code-examples-b">
+<summary><b>Code examples</b></summary><p>We provide examples on using <em>auto-sklearn</em> for multiple use cases ranging from
 simple classification to advanced uses such as feature importance, parallel runs
 and customization. They can be found in the <a class="reference internal" href="examples/index.html#sphx-glr-examples"><span class="std std-ref">Examples</span></a>.</p>
-</section>
-<section id="time-and-memory-limits">
-<h2>Time and memory limits<a class="headerlink" href="#time-and-memory-limits" title="Permalink to this headline">¶</a></h2>
-<p>A crucial feature of <em>auto-sklearn</em> is limiting the resources (memory and
-time) which the scikit-learn algorithms are allowed to use. Especially for
-large datasets, on which algorithms can take several hours and make the
-machine swap, it is important to stop the evaluations after some time in order
-to make progress in a reasonable amount of time. Setting the resource limits
-is therefore a tradeoff between optimization time and the number of models
-that can be tested.</p>
-<p>While <em>auto-sklearn</em> alleviates manual hyperparameter tuning, the user still
+</details><details class="summary-b-material-from-talks-and-presentations-b">
+<summary><b>Material from talks and presentations</b></summary><p>We provide resources for talks, tutorials and presentations on <em>auto-sklearn</em> under <a class="reference external" href="https://github.com/automl/auto-sklearn-talks">auto-sklearn-talks</a></p>
+</details><div class="section" id="resource-limits">
+<span id="limits"></span><h2>Resource limits<a class="headerlink" href="#resource-limits" title="Permalink to this headline">¶</a></h2>
+<p>A crucial feature of <em>auto-sklearn</em> is limiting the resources (memory and time) which the scikit-learn algorithms are
+allowed to use. Especially for large datasets, on which algorithms can take several hours and make the machine swap,
+it is important to stop the evaluations after some time in order to make progress in a reasonable amount of time.
+Setting the resource limits is therefore a tradeoff between optimization time and the number of models that can be
+tested.</p>
+<details class="summary-b-time-and-memory-limits-b">
+<summary><b>Time and memory limits</b></summary><p>While <em>auto-sklearn</em> alleviates manual hyperparameter tuning, the user still
 has to set memory and time limits. For most datasets a memory limit of 3GB or
 6GB as found on most modern computers is sufficient. For the time limits it
 is harder to give clear guidelines. If possible, a good default is a total
 time limit of one day, and a time limit of 30 minutes for a single run.</p>
 <p>Further guidelines can be found in
 <a class="reference external" href="https://github.com/automl/auto-sklearn/issues/142">auto-sklearn/issues/142</a>.</p>
-</section>
-<section id="restricting-the-searchspace">
-<span id="id2"></span><h2>Restricting the searchspace<a class="headerlink" href="#restricting-the-searchspace" title="Permalink to this headline">¶</a></h2>
-<p>Instead of using all available estimators, it is possible to restrict
-<em>auto-sklearn</em>’s searchspace. The following shows an example of how to exclude
-all preprocessing methods and restrict the configuration space to only
-random forests.</p>
-<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">autosklearn.classification</span>
+</details><details class="summary-b-cpu-cores-b">
+<summary><b>CPU cores</b></summary><p>By default, <em>auto-sklearn</em> uses <strong>one core</strong>. See also <a class="reference internal" href="#parallel"><span class="std std-ref">Parallel computation</span></a> on how to configure this.</p>
+</details></div>
+<div class="section" id="the-search-space">
+<span id="space"></span><h2>The search space<a class="headerlink" href="#the-search-space" title="Permalink to this headline">¶</a></h2>
+<p><em>Auto-sklearn</em> by default searches a large space to find a well performing configuration. However, it is also possible
+to restrict the searchspace:</p>
+<details class="summary-b-restricting-the-searchspace-b">
+<summary><b>Restricting the searchspace</b></summary><p>The following shows an example of how to exclude all preprocessing methods and restrict the configuration space to
+only random forests.</p>
+<blockquote>
+<div><div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">autosklearn.classification</span>
 <span class="n">automl</span> <span class="o">=</span> <span class="n">autosklearn</span><span class="o">.</span><span class="n">classification</span><span class="o">.</span><span class="n">AutoSklearnClassifier</span><span class="p">(</span>
     <span class="n">include</span> <span class="o">=</span> <span class="p">{</span>
         <span class="s1">&#39;classifier&#39;</span><span class="p">:</span> <span class="p">[</span><span class="s2">&quot;random_forest&quot;</span><span class="p">],</span>
@@ -186,108 +181,88 @@ <h2>Time and memory limits<a class="headerlink" href="#time-and-memory-limits" t
 </div></blockquote>
 <p>We do also provide an example on how to restrict the classifiers to search over
 <a class="reference internal" href="examples/40_advanced/example_interpretable_models.html#sphx-glr-examples-40-advanced-example-interpretable-models-py"><span class="std std-ref">Interpretable models</span></a>.</p>
-<section id="data-preprocessing">
-<h3>Data preprocessing<a class="headerlink" href="#data-preprocessing" title="Permalink to this headline">¶</a></h3>
-<p>Data preprocessing includes One-Hot encoding of categorical features, imputation
+</div></blockquote>
+</details><details class="summary-b-turn-off-data-preprocessing-b">
+<summary><b>Turn off data preprocessing</b></summary><p>Data preprocessing includes One-Hot encoding of categorical features, imputation
 of missing values and the normalization of features or samples. These ensure that
 the data the gets to the sklearn models is well formed and can be used for
 training models.</p>
 <p>While this is necessary in general, if you’d like to disable this step, please
 refer to this <a class="reference internal" href="examples/80_extending/example_extending_data_preprocessor.html#sphx-glr-examples-80-extending-example-extending-data-preprocessor-py"><span class="std std-ref">example</span></a>.</p>
-</section>
-<section id="feature-preprocessing">
-<h3>Feature preprocessing<a class="headerlink" href="#feature-preprocessing" title="Permalink to this headline">¶</a></h3>
-<p>Feature preprocessing is a single transformer which implements for example feature
+</details><details class="summary-b-turn-off-feature-preprocessing-b">
+<summary><b>Turn off feature preprocessing</b></summary><p>Feature preprocessing is a single transformer which implements for example feature
 selection or transformation of features into a different space (i.e. PCA).</p>
 <p>This can be turned off by setting
 <code class="docutils literal notranslate"><span class="pre">include={'feature_preprocessor'=[&quot;no_preprocessing&quot;]}</span></code> as shown in the example above.</p>
-</section>
-</section>
-<section id="resampling-strategies">
-<h2>Resampling strategies<a class="headerlink" href="#resampling-strategies" title="Permalink to this headline">¶</a></h2>
-<p>Examples for using holdout and cross-validation can be found in <a class="reference internal" href="examples/index.html#examples"><span class="std std-ref">auto-sklearn/examples/</span></a>.</p>
-</section>
-<section id="supported-inputs">
-<h2>Supported Inputs<a class="headerlink" href="#supported-inputs" title="Permalink to this headline">¶</a></h2>
-<p><em>auto-sklearn</em> can accept targets for the following tasks (more details on <a class="reference external" href="https://scikit-learn.org/stable/modules/multiclass.html">Sklearn algorithms</a>):</p>
-<ul class="simple">
-<li><p>Binary Classification</p></li>
-<li><p>Multiclass Classification</p></li>
-<li><p>Multilabel Classification</p></li>
-<li><p>Regression</p></li>
-<li><p>Multioutput Regression</p></li>
-</ul>
-<p>You can provide feature and target training pairs (X_train/y_train) to <em>auto-sklearn</em> to fit an
-ensemble of pipelines as described in the next section. This X_train/y_train dataset must belong
-to one of the supported formats: np.ndarray, pd.DataFrame, scipy.sparse.csr_matrix and python lists.
-Optionally, you can measure the ability of this fitted model to generalize to unseen data by
-providing an optional testing pair (X_test/Y_test). For further details, please refer to the
-Example <a class="reference internal" href="examples/40_advanced/example_pandas_train_test.html#sphx-glr-examples-40-advanced-example-pandas-train-test-py"><span class="std std-ref">Performance-over-time plot</span></a>.
-Supported formats for these training and testing pairs are: np.ndarray,
-pd.DataFrame, scipy.sparse.csr_matrix and python lists.</p>
-<p>If your data contains categorical values (in the features or targets), autosklearn will automatically encode your data using a <a class="reference external" href="https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.LabelEncoder.html">sklearn.preprocessing.LabelEncoder</a> for unidimensional data and a <a class="reference external" href="https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.OrdinalEncoder.html">sklearn.preprocessing.OrdinalEncoder</a> for multidimensional data.</p>
-<p>Regarding the features, there are two methods to guide <em>auto-sklearn</em> to properly encode categorical columns:</p>
-<ul class="simple">
-<li><p>Providing a X_train/X_test numpy array with the optional flag feat_type. For further details, you
-can check the Example <a class="reference internal" href="examples/40_advanced/example_feature_types.html#sphx-glr-examples-40-advanced-example-feature-types-py"><span class="std std-ref">Feature Types</span></a>.</p></li>
-<li><p>You can provide a pandas DataFrame, with properly formatted columns. If a column has numerical
-dtype, <em>auto-sklearn</em> will not encode it and it will be passed directly to scikit-learn. If the
-column has a categorical/boolean class, it will be encoded. If the column is of any other type
-(Object or Timeseries), an error will be raised. For further details on how to properly encode
-your data, you can check the Pandas Example
-<a class="reference external" href="https://pandas.pydata.org/pandas-docs/stable/user_guide/categorical.html">Working with categorical data</a>).
-If you are working with time series, it is recommended that you follow this approach
-<a class="reference external" href="https://stats.stackexchange.com/questions/311494/">Working with time data</a>.</p></li>
-</ul>
-<p>Regarding the targets (y_train/y_test), if the task involves a classification problem, such features will be automatically encoded. It is recommended to provide both y_train and y_test during fit, so that a common encoding is created between these splits (if only y_train is provided during fit, the categorical encoder will not be able to handle new classes that are exclusive to y_test). If the task is regression, no encoding happens on the targets.</p>
-</section>
-<section id="ensemble-building-process">
-<h2>Ensemble Building Process<a class="headerlink" href="#ensemble-building-process" title="Permalink to this headline">¶</a></h2>
-<p><em>auto-sklearn</em> uses ensemble selection by <a class="reference external" href="https://dl.acm.org/doi/pdf/10.1145/1015330.1015432">Caruana et al. (2004)</a>
-to build an ensemble based on the models’ prediction for the validation set. The following hyperparameters control how the ensemble is constructed:</p>
+</details></div>
+<div class="section" id="model-selection">
+<span id="bestmodel"></span><h2>Model selection<a class="headerlink" href="#model-selection" title="Permalink to this headline">¶</a></h2>
+<p><em>Auto-sklearn</em> implements different strategies to identify the best performing model. For some use cases it might be
+necessary to adapt the resampling strategy or define a custom metric:</p>
+<details class="summary-b-use-different-resampling-strategies-b">
+<summary><b>Use different resampling strategies</b></summary><p>Examples for using holdout and cross-validation can be found in <a class="reference internal" href="examples/40_advanced/example_resampling.html#sphx-glr-examples-40-advanced-example-resampling-py"><span class="std std-ref">example</span></a></p>
+</details><details class="summary-b-use-a-custom-metric-b">
+<summary><b>Use a custom metric</b></summary><p>Examples for using a custom metric can be found in <a class="reference internal" href="examples/40_advanced/example_metrics.html#sphx-glr-examples-40-advanced-example-metrics-py"><span class="std std-ref">example</span></a></p>
+</details></div>
+<div class="section" id="ensembling">
+<span id="ensembles"></span><h2>Ensembling<a class="headerlink" href="#ensembling" title="Permalink to this headline">¶</a></h2>
+<p>To get the best performance out of the evaluated models, <em>auto-sklearn</em> uses ensemble selection by <a class="reference external" href="https://dl.acm.org/doi/pdf/10.1145/1015330.1015432">Caruana et al. (2004)</a>
+to build an ensemble based on the models’ prediction for the validation set.</p>
+<details class="summary-b-configure-the-ensemble-building-process-b">
+<summary><b>Configure the ensemble building process</b></summary><p>The following hyperparameters control how the ensemble is constructed:</p>
 <ul class="simple">
 <li><p><code class="docutils literal notranslate"><span class="pre">ensemble_size</span></code> determines the maximal size of the ensemble. If it is set to zero, no ensemble will be constructed.</p></li>
 <li><p><code class="docutils literal notranslate"><span class="pre">ensemble_nbest</span></code> allows the user to directly specify the number of models considered for the ensemble.  This hyperparameter can be an integer <em>n</em>, such that only the best <em>n</em> models are used in the final ensemble. If a float between 0.0 and 1.0 is provided, <code class="docutils literal notranslate"><span class="pre">ensemble_nbest</span></code> would be interpreted as a fraction suggesting the percentage of models to use in the ensemble building process (namely, if ensemble_nbest is a float, library pruning is implemented as described in <a class="reference external" href="https://dl.acm.org/doi/10.1109/ICDM.2006.76">Caruana et al. (2006)</a>).</p></li>
 <li><p><code class="docutils literal notranslate"><span class="pre">max_models_on_disc</span></code> defines the maximum number of models that are kept on the disc, as a mechanism to control the amount of disc space consumed by <em>auto-sklearn</em>. Throughout the automl process, different individual models are optimized, and their predictions (and other metadata) is stored on disc. The user can set the upper bound on how many models are acceptable to keep on disc, yet this variable takes priority in the definition of the number of models used by the ensemble builder (that is, the minimum of <code class="docutils literal notranslate"><span class="pre">ensemble_size</span></code>, <code class="docutils literal notranslate"><span class="pre">ensemble_nbest</span></code> and <code class="docutils literal notranslate"><span class="pre">max_models_on_disc</span></code> determines the maximal amount of models used in the ensemble). If set to None, this feature is disabled.</p></li>
 </ul>
-</section>
-<section id="inspecting-the-results">
-<span id="id3"></span><h2>Inspecting the results<a class="headerlink" href="#inspecting-the-results" title="Permalink to this headline">¶</a></h2>
-<p><em>auto-sklearn</em> allows users to inspect the training results and statistics. The following example shows how different
-statistics can be printed for the inspection.</p>
+</details><details class="summary-b-inspect-the-final-ensemble-b">
+<summary><b>Inspect the final ensemble</b></summary><p>The results obtained from the final ensemble can be printed by calling <code class="docutils literal notranslate"><span class="pre">show_models()</span></code>.
+The <em>auto-sklearn</em> ensemble is composed of scikit-learn models that can be inspected as exemplified
+in the Example <a class="reference internal" href="examples/40_advanced/example_get_pipeline_components.html#sphx-glr-examples-40-advanced-example-get-pipeline-components-py"><span class="std std-ref">Obtain run information</span></a>.</p>
+</details><details class="summary-b-fit-ensemble-post-hoc-b">
+<summary><b>Fit ensemble post-hoc</b></summary><p>To use a single core only, it is possible to build ensembles post-hoc. An example on how to do this (first searching
+for individual models, and then building an ensemble from them) can be seen in
+<a class="reference internal" href="examples/60_search/example_sequential.html#sphx-glr-examples-60-search-example-sequential-py"><span class="std std-ref">Sequential Usage</span></a>.</p>
+</details></div>
+<div class="section" id="inspecting-the-results">
+<span id="inspect"></span><h2>Inspecting the results<a class="headerlink" href="#inspecting-the-results" title="Permalink to this headline">¶</a></h2>
+<p><em>auto-sklearn</em> allows users to inspect the training results and statistics. Assume we have a fitted estimator:</p>
 <div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">autosklearn.classification</span>
 <span class="n">automl</span> <span class="o">=</span> <span class="n">autosklearn</span><span class="o">.</span><span class="n">classification</span><span class="o">.</span><span class="n">AutoSklearnClassifier</span><span class="p">()</span>
 <span class="n">automl</span><span class="o">.</span><span class="n">fit</span><span class="p">(</span><span class="n">X_train</span><span class="p">,</span> <span class="n">y_train</span><span class="p">)</span>
-<span class="n">automl</span><span class="o">.</span><span class="n">cv_results_</span>
-<span class="n">automl</span><span class="o">.</span><span class="n">performance_over_time_</span><span class="o">.</span><span class="n">plot</span><span class="p">(</span>
-    <span class="n">x</span><span class="o">=</span><span class="s1">&#39;Timestamp&#39;</span><span class="p">,</span>
-    <span class="n">kind</span><span class="o">=</span><span class="s1">&#39;line&#39;</span><span class="p">,</span>
-    <span class="n">legend</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
-    <span class="n">title</span><span class="o">=</span><span class="s1">&#39;Auto-sklearn accuracy over time&#39;</span><span class="p">,</span>
-    <span class="n">grid</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
-<span class="p">)</span>
-<span class="n">plt</span><span class="o">.</span><span class="n">show</span><span class="p">()</span>
-
-<span class="n">automl</span><span class="o">.</span><span class="n">sprint_statistics</span><span class="p">()</span>
-<span class="n">automl</span><span class="o">.</span><span class="n">show_models</span><span class="p">()</span>
 </pre></div>
 </div>
-<p><code class="docutils literal notranslate"><span class="pre">cv_results_</span></code> returns a dict with keys as column headers and values as columns, that can be imported into a pandas DataFrame.
-<code class="docutils literal notranslate"><span class="pre">performance_over_time_</span></code>  returns a DataFrame containing the models performance over time data, which can be used for plotting directly (Here is an example: <a class="reference internal" href="examples/40_advanced/example_pandas_train_test.html#sphx-glr-examples-40-advanced-example-pandas-train-test-py"><span class="std std-ref">Performance-over-time plot</span></a>).
-<code class="docutils literal notranslate"><span class="pre">sprint_statistics()</span></code> is a method that prints the name of the  dataset, the metric used, and the best validation score
+<p><em>auto-sklearn</em> offers the following ways to inspect the results</p>
+<details class="summary-b-basic-statistics-b">
+<summary><b>Basic statistics</b></summary><p><code class="docutils literal notranslate"><span class="pre">sprint_statistics()</span></code> is a method that prints the name of the  dataset, the metric used, and the best validation score
 obtained by running <em>auto-sklearn</em>. It additionally prints the number of both successful and unsuccessful
 algorithm runs.</p>
-<p>The results obtained from the final ensemble can be printed by calling <code class="docutils literal notranslate"><span class="pre">show_models()</span></code>.
-<em>auto-sklearn</em> ensemble is composed of scikit-learn models that can be inspected as exemplified
-in the Example <a class="reference internal" href="examples/40_advanced/example_get_pipeline_components.html#sphx-glr-examples-40-advanced-example-get-pipeline-components-py"><span class="std std-ref">Obtain run information</span></a>.</p>
-</section>
-<section id="parallel-computation">
-<h2>Parallel computation<a class="headerlink" href="#parallel-computation" title="Permalink to this headline">¶</a></h2>
-<p>In it’s default mode, <em>auto-sklearn</em> already uses two cores. The first one is
-used for model building, the second for building an ensemble every time a new
-machine learning model has finished training. An example on how to do this sequentially (first searching for individual models, and then building an ensemble from them) can be seen in
-<a class="reference internal" href="examples/60_search/example_sequential.html#sphx-glr-examples-60-search-example-sequential-py"><span class="std std-ref">Sequential Usage</span></a>.</p>
-<p>Nevertheless, <em>auto-sklearn</em> also supports parallel Bayesian optimization via the use of
+</details><details class="summary-b-performance-over-time-b">
+<summary><b>Performance over Time</b></summary><p><code class="docutils literal notranslate"><span class="pre">performance_over_time_</span></code>  returns a DataFrame containing the models performance over time data, which can
+be used for plotting directly (Here is an example: <a class="reference internal" href="examples/40_advanced/example_pandas_train_test.html#sphx-glr-examples-40-advanced-example-pandas-train-test-py"><span class="std std-ref">Performance-over-time plot</span></a>).</p>
+<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="n">automl</span><span class="o">.</span><span class="n">performance_over_time_</span><span class="o">.</span><span class="n">plot</span><span class="p">(</span>
+        <span class="n">x</span><span class="o">=</span><span class="s1">&#39;Timestamp&#39;</span><span class="p">,</span>
+        <span class="n">kind</span><span class="o">=</span><span class="s1">&#39;line&#39;</span><span class="p">,</span>
+        <span class="n">legend</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+        <span class="n">title</span><span class="o">=</span><span class="s1">&#39;Auto-sklearn accuracy over time&#39;</span><span class="p">,</span>
+        <span class="n">grid</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span>
+    <span class="p">)</span>
+    <span class="n">plt</span><span class="o">.</span><span class="n">show</span><span class="p">()</span>
+</pre></div>
+</div>
+</details><details class="summary-b-evaluated-models-b">
+<summary><b>Evaluated models</b></summary><p>The results obtained from the final ensemble can be printed by calling <code class="docutils literal notranslate"><span class="pre">show_models()</span></code>.</p>
+</details><details class="summary-b-leaderboard-b">
+<summary><b>Leaderboard</b></summary><p><code class="docutils literal notranslate"><span class="pre">automl.leaderboard()</span></code> shows the ensemble members, check the <a class="reference internal" href="api.html#autosklearn.classification.AutoSklearnClassifier.leaderboard" title="autosklearn.classification.AutoSklearnClassifier.leaderboard"><code class="xref py py-meth docutils literal notranslate"><span class="pre">docs</span></code></a> for using leaderboard for getting information on <em>all</em> runs.</p>
+</details><details class="summary-b-other-b">
+<summary><b>Other</b></summary><p><code class="docutils literal notranslate"><span class="pre">cv_results_</span></code> returns a dict with keys as column headers and values as columns, that can be imported into a pandas DataFrame.</p>
+</details></div>
+<div class="section" id="parallel-computation">
+<span id="parallel"></span><h2>Parallel computation<a class="headerlink" href="#parallel-computation" title="Permalink to this headline">¶</a></h2>
+<p>In it’s default mode, <em>auto-sklearn</em> uses <strong>one core</strong> and interleaves ensemble building with evaluating new
+configurations.</p>
+<details class="summary-b-parallelization-with-dask-b">
+<summary><b>Parallelization with Dask</b></summary><p>Nevertheless, <em>auto-sklearn</em> also supports parallel Bayesian optimization via the use of
 <a class="reference external" href="https://distributed.dask.org/">Dask.distributed</a>. By providing the arguments <code class="docutils literal notranslate"><span class="pre">n_jobs</span></code>
 to the estimator construction, one can control the number of cores available to <em>auto-sklearn</em>
 (As shown in the Example <a class="reference internal" href="examples/60_search/example_parallel_n_jobs.html#sphx-glr-examples-60-search-example-parallel-n-jobs-py"><span class="std std-ref">Parallel Usage  on a single machine</span></a>).
@@ -306,17 +281,55 @@ <h2>Parallel computation<a class="headerlink" href="#parallel-computation" title
 </pre></div>
 </div>
 <p>For further information about how scikit-learn handles multiprocessing, please check the <a class="reference external" href="https://scikit-learn.org/stable/computing/parallelism.html">Parallelism, resource management, and configuration</a> documentation from the library.</p>
-</section>
-<section id="model-persistence">
-<h2>Model persistence<a class="headerlink" href="#model-persistence" title="Permalink to this headline">¶</a></h2>
-<p><em>auto-sklearn</em> is mostly a wrapper around scikit-learn. Therefore, it is
+</details></div>
+<div class="section" id="other">
+<span id="othermanual"></span><h2>Other<a class="headerlink" href="#other" title="Permalink to this headline">¶</a></h2>
+<details class="summary-b-supported-input-types-b">
+<summary><b>Supported input types</b></summary><p><em>auto-sklearn</em> can accept targets for the following tasks (more details on <a class="reference external" href="https://scikit-learn.org/stable/modules/multiclass.html">Sklearn algorithms</a>):</p>
+<ul class="simple">
+<li><p>Binary Classification</p></li>
+<li><p>Multiclass Classification</p></li>
+<li><p>Multilabel Classification</p></li>
+<li><p>Regression</p></li>
+<li><p>Multioutput Regression</p></li>
+</ul>
+<p>You can provide feature and target training pairs (X_train/y_train) to <em>auto-sklearn</em> to fit an
+ensemble of pipelines as described in the next section. This X_train/y_train dataset must belong
+to one of the supported formats: np.ndarray, pd.DataFrame, scipy.sparse.csr_matrix and python lists.
+Optionally, you can measure the ability of this fitted model to generalize to unseen data by
+providing an optional testing pair (X_test/Y_test). For further details, please refer to the
+Example <a class="reference internal" href="examples/40_advanced/example_pandas_train_test.html#sphx-glr-examples-40-advanced-example-pandas-train-test-py"><span class="std std-ref">Performance-over-time plot</span></a>.
+Supported formats for these training and testing pairs are: np.ndarray,
+pd.DataFrame, scipy.sparse.csr_matrix and python lists.</p>
+<p>If your data contains categorical values (in the features or targets), autosklearn will automatically encode your
+data using a <a class="reference external" href="https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.LabelEncoder.html">sklearn.preprocessing.LabelEncoder</a>
+for unidimensional data and a <a class="reference external" href="https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.OrdinalEncoder.html">sklearn.preprocessing.OrdinalEncoder</a>
+for multidimensional data.</p>
+<p>Regarding the features, there are two methods to guide <em>auto-sklearn</em> to properly encode categorical columns:</p>
+<ul class="simple">
+<li><p>Providing a X_train/X_test numpy array with the optional flag feat_type. For further details, you
+can check the Example <a class="reference internal" href="examples/40_advanced/example_feature_types.html#sphx-glr-examples-40-advanced-example-feature-types-py"><span class="std std-ref">Feature Types</span></a>.</p></li>
+<li><p>You can provide a pandas DataFrame, with properly formatted columns. If a column has numerical
+dtype, <em>auto-sklearn</em> will not encode it and it will be passed directly to scikit-learn. If the
+column has a categorical/boolean class, it will be encoded. If the column is of any other type
+(Object or Timeseries), an error will be raised. For further details on how to properly encode
+your data, you can check the Pandas Example
+<a class="reference external" href="https://pandas.pydata.org/pandas-docs/stable/user_guide/categorical.html">Working with categorical data</a>).
+If you are working with time series, it is recommended that you follow this approach
+<a class="reference external" href="https://stats.stackexchange.com/questions/311494/">Working with time data</a>.</p></li>
+</ul>
+<p>Regarding the targets (y_train/y_test), if the task involves a classification problem, such features will be
+automatically encoded. It is recommended to provide both y_train and y_test during fit, so that a common encoding
+is created between these splits (if only y_train is provided during fit, the categorical encoder will not be able
+to handle new classes that are exclusive to y_test). If the task is regression, no encoding happens on the
+targets.</p>
+</details><details class="summary-b-model-persistence-b">
+<summary><b>Model persistence</b></summary><p><em>auto-sklearn</em> is mostly a wrapper around scikit-learn. Therefore, it is
 possible to follow the
 <a class="reference external" href="https://scikit-learn.org/stable/modules/model_persistence.html">persistence Example</a>
 from scikit-learn.</p>
-</section>
-<section id="vanilla-auto-sklearn">
-<h2>Vanilla auto-sklearn<a class="headerlink" href="#vanilla-auto-sklearn" title="Permalink to this headline">¶</a></h2>
-<p>In order to obtain <em>vanilla auto-sklearn</em> as used in <a class="reference external" href="https://papers.nips.cc/paper/5872-efficient-and-robust-automated-machine-learning">Efficient and Robust Automated Machine Learning</a>
+</details><details class="summary-b-vanilla-auto-sklearn-b">
+<summary><b>Vanilla auto-sklearn</b></summary><p>In order to obtain <em>vanilla auto-sklearn</em> as used in <a class="reference external" href="https://papers.nips.cc/paper/5872-efficient-and-robust-automated-machine-learning">Efficient and Robust Automated Machine Learning</a>
 set <code class="docutils literal notranslate"><span class="pre">ensemble_size=1</span></code> and <code class="docutils literal notranslate"><span class="pre">initial_configurations_via_metalearning=0</span></code>:</p>
 <div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">autosklearn.classification</span>
 <span class="n">automl</span> <span class="o">=</span> <span class="n">autosklearn</span><span class="o">.</span><span class="n">classification</span><span class="o">.</span><span class="n">AutoSklearnClassifier</span><span class="p">(</span>
@@ -329,8 +342,8 @@ <h2>Vanilla auto-sklearn<a class="headerlink" href="#vanilla-auto-sklearn" title
 according to its performance on the validation set. Setting the initial
 configurations found by meta-learning to zero makes <em>auto-sklearn</em> use the
 regular SMAC algorithm for suggesting new hyperparameter configurations.</p>
-</section>
-</section>
+</details></div>
+</div>
 
 
     </div>
@@ -351,7 +364,7 @@ <h2>Vanilla auto-sklearn<a class="headerlink" href="#vanilla-auto-sklearn" title
       
     </p>
     <p>
-        &copy; Copyright 2014-2021, Machine Learning Professorship Freiburg.<br/>
+        &copy; Copyright 2014-2022, Machine Learning Professorship Freiburg.<br/>
       Created using <a href="http://sphinx-doc.org/">Sphinx</a> 4.2.0.<br/>
     </p>
   </div>
diff --git a/master/notebooks/examples/20_basic/example_classification.ipynb b/master/notebooks/examples/20_basic/example_classification.ipynb
index 7cc4b7cd39..934f27b5c7 100644
--- a/master/notebooks/examples/20_basic/example_classification.ipynb
+++ b/master/notebooks/examples/20_basic/example_classification.ipynb
@@ -26,7 +26,7 @@
       },
       "outputs": [],
       "source": [
-        "import sklearn.datasets\nimport sklearn.metrics\n\nimport autosklearn.classification"
+        "from pprint import pprint\n\nimport sklearn.datasets\nimport sklearn.metrics\n\nimport autosklearn.classification"
       ]
     },
     {
@@ -98,7 +98,7 @@
       },
       "outputs": [],
       "source": [
-        "print(automl.show_models())"
+        "pprint(automl.show_models(), indent=4)"
       ]
     },
     {
diff --git a/master/notebooks/examples/20_basic/example_multilabel_classification.ipynb b/master/notebooks/examples/20_basic/example_multilabel_classification.ipynb
index 341f5b333e..bf999ddaad 100644
--- a/master/notebooks/examples/20_basic/example_multilabel_classification.ipynb
+++ b/master/notebooks/examples/20_basic/example_multilabel_classification.ipynb
@@ -26,7 +26,7 @@
       },
       "outputs": [],
       "source": [
-        "import numpy as np\n\nimport sklearn.datasets\nimport sklearn.metrics\nfrom sklearn.utils.multiclass import type_of_target\n\nimport autosklearn.classification"
+        "import numpy as np\nfrom pprint import pprint\n\nimport sklearn.datasets\nimport sklearn.metrics\nfrom sklearn.utils.multiclass import type_of_target\n\nimport autosklearn.classification"
       ]
     },
     {
@@ -44,7 +44,7 @@
       },
       "outputs": [],
       "source": [
-        "# Using reuters multilabel dataset -- https://www.openml.org/d/40594\nX, y = sklearn.datasets.fetch_openml(data_id=40594, return_X_y=True, as_frame=False)\n\n# fetch openml downloads a numpy array with TRUE/FALSE strings. Re-map it to\n# integer dtype with ones and zeros\n# This is to comply with Scikit-learn requirement:\n# \"Positive classes are indicated with 1 and negative classes with 0 or -1.\"\n# More information on: https://scikit-learn.org/stable/modules/multiclass.html\ny[y == 'TRUE'] = 1\ny[y == 'FALSE'] = 0\ny = y.astype(np.int)\n\n# Using type of target is a good way to make sure your data\n# is properly formatted\nprint(f\"type_of_target={type_of_target(y)}\")\n\nX_train, X_test, y_train, y_test = sklearn.model_selection.train_test_split(\n    X, y, random_state=1\n)"
+        "# Using reuters multilabel dataset -- https://www.openml.org/d/40594\nX, y = sklearn.datasets.fetch_openml(data_id=40594, return_X_y=True, as_frame=False)\n\n# fetch openml downloads a numpy array with TRUE/FALSE strings. Re-map it to\n# integer dtype with ones and zeros\n# This is to comply with Scikit-learn requirement:\n# \"Positive classes are indicated with 1 and negative classes with 0 or -1.\"\n# More information on: https://scikit-learn.org/stable/modules/multiclass.html\ny[y == 'TRUE'] = 1\ny[y == 'FALSE'] = 0\ny = y.astype(int)\n\n# Using type of target is a good way to make sure your data\n# is properly formatted\nprint(f\"type_of_target={type_of_target(y)}\")\n\nX_train, X_test, y_train, y_test = sklearn.model_selection.train_test_split(\n    X, y, random_state=1\n)"
       ]
     },
     {
@@ -98,7 +98,7 @@
       },
       "outputs": [],
       "source": [
-        "print(automl.show_models())"
+        "pprint(automl.show_models(), indent=4)"
       ]
     },
     {
diff --git a/master/notebooks/examples/20_basic/example_multioutput_regression.ipynb b/master/notebooks/examples/20_basic/example_multioutput_regression.ipynb
index f3112d78a6..7998b22d6f 100644
--- a/master/notebooks/examples/20_basic/example_multioutput_regression.ipynb
+++ b/master/notebooks/examples/20_basic/example_multioutput_regression.ipynb
@@ -26,7 +26,7 @@
       },
       "outputs": [],
       "source": [
-        "import numpy as numpy\n\nfrom sklearn.datasets import make_regression\nfrom sklearn.metrics import r2_score\nfrom sklearn.model_selection import train_test_split\n\nfrom autosklearn.regression import AutoSklearnRegressor"
+        "import numpy as numpy\nfrom pprint import pprint\n\nfrom sklearn.datasets import make_regression\nfrom sklearn.metrics import r2_score\nfrom sklearn.model_selection import train_test_split\n\nfrom autosklearn.regression import AutoSklearnRegressor"
       ]
     },
     {
@@ -98,7 +98,7 @@
       },
       "outputs": [],
       "source": [
-        "print(automl.show_models())"
+        "pprint(automl.show_models(), indent=4)"
       ]
     },
     {
diff --git a/master/notebooks/examples/20_basic/example_regression.ipynb b/master/notebooks/examples/20_basic/example_regression.ipynb
index 3bfacf545e..acb7673ace 100644
--- a/master/notebooks/examples/20_basic/example_regression.ipynb
+++ b/master/notebooks/examples/20_basic/example_regression.ipynb
@@ -26,7 +26,7 @@
       },
       "outputs": [],
       "source": [
-        "import sklearn.datasets\nimport sklearn.metrics\n\nimport autosklearn.regression\nimport matplotlib.pyplot as plt"
+        "from pprint import pprint\n\nimport sklearn.datasets\nimport sklearn.metrics\n\nimport autosklearn.regression\nimport matplotlib.pyplot as plt"
       ]
     },
     {
@@ -98,7 +98,7 @@
       },
       "outputs": [],
       "source": [
-        "print(automl.show_models())"
+        "pprint(automl.show_models(), indent=4)"
       ]
     },
     {
diff --git a/master/notebooks/examples/40_advanced/example_get_pipeline_components.ipynb b/master/notebooks/examples/40_advanced/example_get_pipeline_components.ipynb
index b26c9ab89c..33f2d673d0 100644
--- a/master/notebooks/examples/40_advanced/example_get_pipeline_components.ipynb
+++ b/master/notebooks/examples/40_advanced/example_get_pipeline_components.ipynb
@@ -26,7 +26,7 @@
       },
       "outputs": [],
       "source": [
-        "import sklearn.datasets\nimport sklearn.metrics\n\nimport autosklearn.classification"
+        "from pprint import pprint\n\nimport sklearn.datasets\nimport sklearn.metrics\n\nimport autosklearn.classification"
       ]
     },
     {
@@ -87,7 +87,7 @@
       "cell_type": "markdown",
       "metadata": {},
       "source": [
-        "## Report the models found by Auto-Sklearn\n\nAuto-sklearn uses\n`Ensemble Selection <https://www.cs.cornell.edu/~alexn/papers/shotgun.icml04.revised.rev2.pdf>`_\nto construct ensembles in a post-hoc fashion. The ensemble is a linear\nweighting of all models constructed during the hyperparameter optimization.\nThis prints the final ensemble. It is a list of tuples, each tuple being\nthe model weight in the ensemble and the model itself.\n\n"
+        "## Report the models found by Auto-Sklearn\n\nAuto-sklearn uses\n`Ensemble Selection <https://www.cs.cornell.edu/~alexn/papers/shotgun.icml04.revised.rev2.pdf>`_\nto construct ensembles in a post-hoc fashion. The ensemble is a linear\nweighting of all models constructed during the hyperparameter optimization.\nThis prints the final ensemble. It is a dictionary where ``model_id`` of\neach model is a key, and value is a dictionary containing information\nof that model. A model's dict contains its ``'model_id'``, ``'rank'``,\n``'cost'``, ``'ensemble_weight'``, and the model itself. The model is\ngiven by the ``'data_preprocessor'``, ``'feature_preprocessor'``,\n``'regressor'/'classifier'`` and ``'sklearn_regressor'/'sklearn_classifier'``\nentries. But for the ``'cv'`` resampling strategy, the same for each cv\nmodel is stored in the ``'estimators'`` list in the dict, along with the\n``'voting_model'``.\n\n"
       ]
     },
     {
@@ -98,7 +98,7 @@
       },
       "outputs": [],
       "source": [
-        "print(automl.show_models())"
+        "pprint(automl.show_models(), indent=4)"
       ]
     },
     {
diff --git a/master/notebooks/examples/40_advanced/example_interpretable_models.ipynb b/master/notebooks/examples/40_advanced/example_interpretable_models.ipynb
index b4b7eb5b6c..04a8cc6aba 100644
--- a/master/notebooks/examples/40_advanced/example_interpretable_models.ipynb
+++ b/master/notebooks/examples/40_advanced/example_interpretable_models.ipynb
@@ -26,7 +26,7 @@
       },
       "outputs": [],
       "source": [
-        "import autosklearn.classification\nimport sklearn.datasets\nimport sklearn.metrics"
+        "from pprint import pprint\n\nimport autosklearn.classification\nimport sklearn.datasets\nimport sklearn.metrics"
       ]
     },
     {
@@ -116,7 +116,7 @@
       },
       "outputs": [],
       "source": [
-        "print(automl.show_models())"
+        "pprint(automl.show_models(), indent=4)"
       ]
     },
     {
diff --git a/master/notebooks/examples/40_advanced/example_resampling.ipynb b/master/notebooks/examples/40_advanced/example_resampling.ipynb
index dbe33aaa65..55c4cb4f14 100644
--- a/master/notebooks/examples/40_advanced/example_resampling.ipynb
+++ b/master/notebooks/examples/40_advanced/example_resampling.ipynb
@@ -141,7 +141,7 @@
       },
       "outputs": [],
       "source": [
-        "resampling_strategy = sklearn.model_selection.PredefinedSplit(\n    test_fold=np.where(X_train[:, 0] < np.mean(X_train[:, 0]))[0]\n)\n\nautoml = autosklearn.classification.AutoSklearnClassifier(\n    time_left_for_this_task=120,\n    per_run_time_limit=30,\n    tmp_folder='/tmp/autosklearn_resampling_example_tmp',\n    disable_evaluator_output=False,\n    resampling_strategy=resampling_strategy,\n)\nautoml.fit(X_train, y_train, dataset_name='breast_cancer')"
+        "selected_indices = (X_train[:, 0] < np.mean(X_train[:, 0])).astype(int)\nresampling_strategy = sklearn.model_selection.PredefinedSplit(\n    test_fold=selected_indices\n)\n\nautoml = autosklearn.classification.AutoSklearnClassifier(\n    time_left_for_this_task=120,\n    per_run_time_limit=30,\n    tmp_folder='/tmp/autosklearn_resampling_example_tmp',\n    disable_evaluator_output=False,\n    resampling_strategy=resampling_strategy,\n)\nautoml.fit(X_train, y_train, dataset_name='breast_cancer')\n\nprint(automl.sprint_statistics())"
       ]
     },
     {
diff --git a/master/notebooks/examples/60_search/example_parallel_manual_spawning_cli.ipynb b/master/notebooks/examples/60_search/example_parallel_manual_spawning_cli.ipynb
index 0da3e8120c..0c7b01d234 100644
--- a/master/notebooks/examples/60_search/example_parallel_manual_spawning_cli.ipynb
+++ b/master/notebooks/examples/60_search/example_parallel_manual_spawning_cli.ipynb
@@ -15,7 +15,7 @@
       "cell_type": "markdown",
       "metadata": {},
       "source": [
-        "\n# Parallel Usage: Spawning workers from the command line\n\n*Auto-sklearn* uses\n`dask.distributed <https://distributed.dask.org/en/latest/index.html>`_\nfor parallel optimization.\n\nThis example shows how to start the dask scheduler and spawn\nworkers for *Auto-sklearn* manually from the command line. Use this example\nas a starting point to parallelize *Auto-sklearn* across multiple\nmachines. If you want to start everything manually from within Python\nplease see `sphx_glr_examples_60_search_example_parallel_manual_spawning_python.py`.\nTo run *Auto-sklearn* in parallel on a single machine check out the example\n`sphx_glr_examples_60_search_example_parallel_n_jobs.py`.\n\nYou can learn more about the dask command line interface from\nhttps://docs.dask.org/en/latest/setup/cli.html.\n\nWhen manually passing a dask client to Auto-sklearn, all logic\nmust be guarded by ``if __name__ == \"__main__\":`` statements! We use\nmultiple such statements to properly render this example as a notebook\nand also allow execution via the command line.\n\n## Background\n\nTo run Auto-sklearn distributed on multiple machines we need to set\nup three components:\n\n1. **Auto-sklearn and a dask client**. This will manage all workload, find new\n   configurations to evaluate and submit jobs via a dask client. As this\n   runs Bayesian optimization it should be executed on its own CPU.\n2. **The dask workers**. They will do the actual work of running machine\n   learning algorithms and require their own CPU each.\n3. **The scheduler**. It manages the communication between the dask client\n   and the different dask workers. As the client and all workers connect\n   to the scheduler it must be started first. This is a light-weight job\n   and does not require its own CPU.\n\nWe will now start these three components in reverse order: scheduler,\nworkers and client. Also, in a real setup, the scheduler and the workers should\nbe started from the command line and not from within a Python file via\nthe ``subprocess`` module as done here (for the sake of having a self-contained\nexample).\n"
+        "\n# Parallel Usage: Spawning workers from the command line\n\n*Auto-sklearn* uses\n`dask.distributed <https://distributed.dask.org/en/latest/index.html>`_\nfor parallel optimization.\n\nThis example shows how to start the dask scheduler and spawn\nworkers for *Auto-sklearn* manually from the command line. Use this example\nas a starting point to parallelize *Auto-sklearn* across multiple\nmachines.\n\nTo run *Auto-sklearn* in parallel on a single machine check out the example\n`sphx_glr_examples_60_search_example_parallel_n_jobs.py`.\n\nIf you want to start everything manually from within Python\nplease see ``:ref:sphx_glr_examples_60_search_example_parallel_manual_spawning_python.py``.\n\n**NOTE:** Above example is disabled due to issue https://github.com/dask/distributed/issues/5627\n\n\nYou can learn more about the dask command line interface from\nhttps://docs.dask.org/en/latest/setup/cli.html.\n\nWhen manually passing a dask client to Auto-sklearn, all logic\nmust be guarded by ``if __name__ == \"__main__\":`` statements! We use\nmultiple such statements to properly render this example as a notebook\nand also allow execution via the command line.\n\n## Background\n\nTo run Auto-sklearn distributed on multiple machines we need to set\nup three components:\n\n1. **Auto-sklearn and a dask client**. This will manage all workload, find new\n   configurations to evaluate and submit jobs via a dask client. As this\n   runs Bayesian optimization it should be executed on its own CPU.\n2. **The dask workers**. They will do the actual work of running machine\n   learning algorithms and require their own CPU each.\n3. **The scheduler**. It manages the communication between the dask client\n   and the different dask workers. As the client and all workers connect\n   to the scheduler it must be started first. This is a light-weight job\n   and does not require its own CPU.\n\nWe will now start these three components in reverse order: scheduler,\nworkers and client. Also, in a real setup, the scheduler and the workers should\nbe started from the command line and not from within a Python file via\nthe ``subprocess`` module as done here (for the sake of having a self-contained\nexample).\n"
       ]
     },
     {
diff --git a/master/notebooks/examples/60_search/example_parallel_manual_spawning_python.ipynb b/master/notebooks/examples/60_search/example_parallel_manual_spawning_python.ipynb
deleted file mode 100644
index 7851c9efc7..0000000000
--- a/master/notebooks/examples/60_search/example_parallel_manual_spawning_python.ipynb
+++ /dev/null
@@ -1,90 +0,0 @@
-{
-  "cells": [
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "collapsed": false
-      },
-      "outputs": [],
-      "source": [
-        "%matplotlib inline"
-      ]
-    },
-    {
-      "cell_type": "markdown",
-      "metadata": {},
-      "source": [
-        "\n# Parallel Usage: Spawning workers from within Python\n\n*Auto-sklearn* uses\n`dask.distributed <https://distributed.dask.org/en/latest/index.html>`_\nfor parallel optimization.\n\nThis example shows how to start the dask scheduler and spawn\nworkers for *Auto-sklearn* manually within Python. Use this example\nas a starting point to parallelize *Auto-sklearn* across multiple\nmachines. If you want to start everything manually from the command line\nplease see `sphx_glr_examples_60_search_example_parallel_manual_spawning_cli.py`.\nTo run *Auto-sklearn* in parallel on a single machine check out the example\n`sphx_glr_examples_60_search_example_parallel_n_jobs.py`.\n\nWhen manually passing a dask client to Auto-sklearn, all logic\nmust be guarded by ``if __name__ == \"__main__\":`` statements! We use\nmultiple such statements to properly render this example as a notebook\nand also allow execution via the command line.\n\n## Background\n\nTo run Auto-sklearn distributed on multiple machines we need to set\nup three components:\n\n1. **Auto-sklearn and a dask client**. This will manage all workload, find new\n   configurations to evaluate and submit jobs via a dask client. As this\n   runs Bayesian optimization it should be executed on its own CPU.\n2. **The dask workers**. They will do the actual work of running machine\n   learning algorithms and require their own CPU each.\n3. **The scheduler**. It manages the communication between the dask client\n   and the different dask workers. As the client and all workers connect\n   to the scheduler it must be started first. This is a light-weight job\n   and does not require its own CPU.\n\nWe will now start these three components in reverse order: scheduler,\nworkers and client. Also, in a real setup, the scheduler and the workers should\nbe started from the command line and not from within a Python file via\nthe ``subprocess`` module as done here (for the sake of having a self-contained\nexample).\n"
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "collapsed": false
-      },
-      "outputs": [],
-      "source": [
-        "import asyncio\nimport multiprocessing\nimport time\n\nimport dask\nimport dask.distributed\nimport sklearn.datasets\nimport sklearn.metrics\n\nfrom autosklearn.classification import AutoSklearnClassifier\nfrom autosklearn.constants import MULTICLASS_CLASSIFICATION\n\ntmp_folder = '/tmp/autosklearn_parallel_2_example_tmp'"
-      ]
-    },
-    {
-      "cell_type": "markdown",
-      "metadata": {},
-      "source": [
-        "## Define function to start worker\n\nDefine the function to start a dask worker from python. This\nis a bit cumbersome and should ideally be done from the command line.\nWe do it here only for illustrational purpose.\n\n"
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "collapsed": false
-      },
-      "outputs": [],
-      "source": [
-        "# Check the dask docs at\n# https://docs.dask.org/en/latest/setup/python-advanced.html for further\n# information.\n\ndef start_python_worker(scheduler_address):\n    dask.config.set({'distributed.worker.daemon': False})\n\n    async def do_work():\n        async with dask.distributed.Nanny(\n            scheduler_ip=scheduler_address,\n            nthreads=1,\n            lifetime=35,  # automatically shut down the worker so this loop ends\n            memory_limit=0,  # Disable memory management as it is done by Auto-sklearn itself\n        ) as worker:\n            await worker.finished()\n\n    asyncio.get_event_loop().run_until_complete(do_work())"
-      ]
-    },
-    {
-      "cell_type": "markdown",
-      "metadata": {},
-      "source": [
-        "## Start Auto-sklearn\n\nWe are now ready to start *auto-sklearn* and all dask related processes.\n\nTo use auto-sklearn in parallel we must guard the code with\n``if __name__ == '__main__'``. We then start a dask cluster as a context,\nwhich means that it is automatically stopped once all computation is done.\n\n"
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "collapsed": false
-      },
-      "outputs": [],
-      "source": [
-        "if __name__ == '__main__':\n    X, y = sklearn.datasets.load_breast_cancer(return_X_y=True)\n    X_train, X_test, y_train, y_test = \\\n        sklearn.model_selection.train_test_split(X, y, random_state=1)\n\n    # 1. Create a dask scheduler (LocalCluster)\n    with dask.distributed.LocalCluster(\n        n_workers=0, processes=True, threads_per_worker=1,\n    ) as cluster:\n\n        # 2. Start the workers\n        # now we start the two workers, one from within Python, the other\n        # via the command line.\n        worker_processes = []\n        for _ in range(2):\n            process_python_worker = multiprocessing.Process(\n                target=start_python_worker,\n                args=(cluster.scheduler_address, ),\n            )\n            process_python_worker.start()\n            worker_processes.append(process_python_worker)\n\n        # Wait a second for workers to become available\n        time.sleep(1)\n\n        # 3. Start the client\n        with dask.distributed.Client(address=cluster.scheduler_address) as client:\n            automl = AutoSklearnClassifier(\n                delete_tmp_folder_after_terminate=False,\n                time_left_for_this_task=30,\n                per_run_time_limit=10,\n                memory_limit=1024,\n                tmp_folder=tmp_folder,\n                seed=777,\n                # n_jobs is ignored internally as we pass a dask client.\n                n_jobs=1,\n                # Pass a dask client which connects to the previously constructed cluster.\n                dask_client=client,\n            )\n            automl.fit(X_train, y_train)\n\n            automl.fit_ensemble(\n                y_train,\n                task=MULTICLASS_CLASSIFICATION,\n                dataset_name='digits',\n                ensemble_size=20,\n                ensemble_nbest=50,\n            )\n\n        predictions = automl.predict(X_test)\n        print(automl.sprint_statistics())\n        print(\"Accuracy score\", sklearn.metrics.accuracy_score(y_test, predictions))\n\n        # Wait until all workers are closed\n        for process in worker_processes:\n            process_python_worker.join()"
-      ]
-    }
-  ],
-  "metadata": {
-    "kernelspec": {
-      "display_name": "Python 3",
-      "language": "python",
-      "name": "python3"
-    },
-    "language_info": {
-      "codemirror_mode": {
-        "name": "ipython",
-        "version": 3
-      },
-      "file_extension": ".py",
-      "mimetype": "text/x-python",
-      "name": "python",
-      "nbconvert_exporter": "python",
-      "pygments_lexer": "ipython3",
-      "version": "3.8.12"
-    }
-  },
-  "nbformat": 4,
-  "nbformat_minor": 0
-}
\ No newline at end of file
diff --git a/master/notebooks/examples/60_search/example_random_search.ipynb b/master/notebooks/examples/60_search/example_random_search.ipynb
index 753e1b54fa..1748e1bda2 100644
--- a/master/notebooks/examples/60_search/example_random_search.ipynb
+++ b/master/notebooks/examples/60_search/example_random_search.ipynb
@@ -26,7 +26,7 @@
       },
       "outputs": [],
       "source": [
-        "import sklearn.model_selection\nimport sklearn.datasets\nimport sklearn.metrics\n\nfrom smac.facade.roar_facade import ROAR\nfrom smac.scenario.scenario import Scenario\n\nimport autosklearn.classification"
+        "from pprint import pprint\n\nimport sklearn.model_selection\nimport sklearn.datasets\nimport sklearn.metrics\n\nfrom smac.facade.roar_facade import ROAR\nfrom smac.scenario.scenario import Scenario\n\nimport autosklearn.classification"
       ]
     },
     {
@@ -62,7 +62,7 @@
       },
       "outputs": [],
       "source": [
-        "def get_roar_object_callback(\n    scenario_dict,\n    seed,\n    ta,\n    ta_kwargs,\n    metalearning_configurations,\n    n_jobs,\n    dask_client,\n):\n    \"\"\"Random online adaptive racing.\"\"\"\n\n    if n_jobs > 1 or (dask_client and len(dask_client.nthreads()) > 1):\n        raise ValueError(\"Please make sure to guard the code invoking Auto-sklearn by \"\n                         \"`if __name__ == '__main__'` and remove this exception.\")\n\n    scenario = Scenario(scenario_dict)\n    return ROAR(\n        scenario=scenario,\n        rng=seed,\n        tae_runner=ta,\n        tae_runner_kwargs=ta_kwargs,\n        run_id=seed,\n        dask_client=dask_client,\n        n_jobs=n_jobs,\n    )\n\n\nautoml = autosklearn.classification.AutoSklearnClassifier(\n    time_left_for_this_task=60,\n    per_run_time_limit=15,\n    tmp_folder='/tmp/autosklearn_random_search_example_tmp',\n    initial_configurations_via_metalearning=0,\n    # The callback to get the SMAC object\n    get_smac_object_callback=get_roar_object_callback,\n)\nautoml.fit(X_train, y_train, dataset_name='breast_cancer')\n\nprint('#' * 80)\nprint('Results for ROAR.')\n# Print the final ensemble constructed by auto-sklearn via ROAR.\nprint(automl.show_models())\npredictions = automl.predict(X_test)\n# Print statistics about the auto-sklearn run such as number of\n# iterations, number of models failed with a time out.\nprint(automl.sprint_statistics())\nprint(\"Accuracy score\", sklearn.metrics.accuracy_score(y_test, predictions))"
+        "def get_roar_object_callback(\n    scenario_dict,\n    seed,\n    ta,\n    ta_kwargs,\n    metalearning_configurations,\n    n_jobs,\n    dask_client,\n):\n    \"\"\"Random online adaptive racing.\"\"\"\n\n    if n_jobs > 1 or (dask_client and len(dask_client.nthreads()) > 1):\n        raise ValueError(\"Please make sure to guard the code invoking Auto-sklearn by \"\n                         \"`if __name__ == '__main__'` and remove this exception.\")\n\n    scenario = Scenario(scenario_dict)\n    return ROAR(\n        scenario=scenario,\n        rng=seed,\n        tae_runner=ta,\n        tae_runner_kwargs=ta_kwargs,\n        run_id=seed,\n        dask_client=dask_client,\n        n_jobs=n_jobs,\n    )\n\n\nautoml = autosklearn.classification.AutoSklearnClassifier(\n    time_left_for_this_task=60,\n    per_run_time_limit=15,\n    tmp_folder='/tmp/autosklearn_random_search_example_tmp',\n    initial_configurations_via_metalearning=0,\n    # The callback to get the SMAC object\n    get_smac_object_callback=get_roar_object_callback,\n)\nautoml.fit(X_train, y_train, dataset_name='breast_cancer')\n\nprint('#' * 80)\nprint('Results for ROAR.')\n# Print the final ensemble constructed by auto-sklearn via ROAR.\npprint(automl.show_models(), indent=4)\npredictions = automl.predict(X_test)\n# Print statistics about the auto-sklearn run such as number of\n# iterations, number of models failed with a time out.\nprint(automl.sprint_statistics())\nprint(\"Accuracy score\", sklearn.metrics.accuracy_score(y_test, predictions))"
       ]
     },
     {
@@ -80,7 +80,7 @@
       },
       "outputs": [],
       "source": [
-        "def get_random_search_object_callback(\n    scenario_dict,\n    seed,\n    ta,\n    ta_kwargs,\n    metalearning_configurations,\n    n_jobs,\n    dask_client\n):\n    \"\"\" Random search \"\"\"\n\n    if n_jobs > 1 or (dask_client and len(dask_client.nthreads()) > 1):\n        raise ValueError(\"Please make sure to guard the code invoking Auto-sklearn by \"\n                         \"`if __name__ == '__main__'` and remove this exception.\")\n\n    scenario_dict['minR'] = len(scenario_dict['instances'])\n    scenario_dict['initial_incumbent'] = 'RANDOM'\n    scenario = Scenario(scenario_dict)\n    return ROAR(\n        scenario=scenario,\n        rng=seed,\n        tae_runner=ta,\n        tae_runner_kwargs=ta_kwargs,\n        run_id=seed,\n        dask_client=dask_client,\n        n_jobs=n_jobs,\n    )\n\n\nautoml = autosklearn.classification.AutoSklearnClassifier(\n    time_left_for_this_task=60,\n    per_run_time_limit=15,\n    tmp_folder='/tmp/autosklearn_random_search_example_tmp',\n    initial_configurations_via_metalearning=0,\n    # Passing the callback to get the SMAC object\n    get_smac_object_callback=get_random_search_object_callback,\n)\nautoml.fit(X_train, y_train, dataset_name='breast_cancer')\n\nprint('#' * 80)\nprint('Results for random search.')\n\n# Print the final ensemble constructed by auto-sklearn via random search.\nprint(automl.show_models())\n\n# Print statistics about the auto-sklearn run such as number of\n# iterations, number of models failed with a time out.\nprint(automl.sprint_statistics())\n\npredictions = automl.predict(X_test)\nprint(\"Accuracy score\", sklearn.metrics.accuracy_score(y_test, predictions))"
+        "def get_random_search_object_callback(\n    scenario_dict,\n    seed,\n    ta,\n    ta_kwargs,\n    metalearning_configurations,\n    n_jobs,\n    dask_client\n):\n    \"\"\" Random search \"\"\"\n\n    if n_jobs > 1 or (dask_client and len(dask_client.nthreads()) > 1):\n        raise ValueError(\"Please make sure to guard the code invoking Auto-sklearn by \"\n                         \"`if __name__ == '__main__'` and remove this exception.\")\n\n    scenario_dict['minR'] = len(scenario_dict['instances'])\n    scenario_dict['initial_incumbent'] = 'RANDOM'\n    scenario = Scenario(scenario_dict)\n    return ROAR(\n        scenario=scenario,\n        rng=seed,\n        tae_runner=ta,\n        tae_runner_kwargs=ta_kwargs,\n        run_id=seed,\n        dask_client=dask_client,\n        n_jobs=n_jobs,\n    )\n\n\nautoml = autosklearn.classification.AutoSklearnClassifier(\n    time_left_for_this_task=60,\n    per_run_time_limit=15,\n    tmp_folder='/tmp/autosklearn_random_search_example_tmp',\n    initial_configurations_via_metalearning=0,\n    # Passing the callback to get the SMAC object\n    get_smac_object_callback=get_random_search_object_callback,\n)\nautoml.fit(X_train, y_train, dataset_name='breast_cancer')\n\nprint('#' * 80)\nprint('Results for random search.')\n\n# Print the final ensemble constructed by auto-sklearn via random search.\npprint(automl.show_models(), indent=4)\n\n# Print statistics about the auto-sklearn run such as number of\n# iterations, number of models failed with a time out.\nprint(automl.sprint_statistics())\n\npredictions = automl.predict(X_test)\nprint(\"Accuracy score\", sklearn.metrics.accuracy_score(y_test, predictions))"
       ]
     }
   ],
diff --git a/master/notebooks/examples/60_search/example_sequential.ipynb b/master/notebooks/examples/60_search/example_sequential.ipynb
index 4b401c5b90..89111256c1 100644
--- a/master/notebooks/examples/60_search/example_sequential.ipynb
+++ b/master/notebooks/examples/60_search/example_sequential.ipynb
@@ -26,7 +26,7 @@
       },
       "outputs": [],
       "source": [
-        "import sklearn.model_selection\nimport sklearn.datasets\nimport sklearn.metrics\n\nimport autosklearn.classification"
+        "from pprint import pprint\n\nimport sklearn.model_selection\nimport sklearn.datasets\nimport sklearn.metrics\n\nimport autosklearn.classification"
       ]
     },
     {
@@ -80,7 +80,7 @@
       },
       "outputs": [],
       "source": [
-        "print(automl.show_models())"
+        "pprint(automl.show_models(), indent=4)"
       ]
     },
     {
diff --git a/master/notebooks/examples/60_search/example_successive_halving.ipynb b/master/notebooks/examples/60_search/example_successive_halving.ipynb
index f285520243..e62b893576 100644
--- a/master/notebooks/examples/60_search/example_successive_halving.ipynb
+++ b/master/notebooks/examples/60_search/example_successive_halving.ipynb
@@ -26,7 +26,7 @@
       },
       "outputs": [],
       "source": [
-        "import sklearn.model_selection\nimport sklearn.datasets\nimport sklearn.metrics\n\nimport autosklearn.classification"
+        "from pprint import pprint\n\nimport sklearn.model_selection\nimport sklearn.datasets\nimport sklearn.metrics\n\nimport autosklearn.classification"
       ]
     },
     {
@@ -80,7 +80,7 @@
       },
       "outputs": [],
       "source": [
-        "automl = autosklearn.classification.AutoSklearnClassifier(\n    time_left_for_this_task=40,\n    per_run_time_limit=10,\n    tmp_folder='/tmp/autosklearn_sh_example_tmp',\n    disable_evaluator_output=False,\n    # 'holdout' with 'train_size'=0.67 is the default argument setting\n    # for AutoSklearnClassifier. It is explicitly specified in this example\n    # for demonstrational purpose.\n    resampling_strategy='holdout',\n    resampling_strategy_arguments={'train_size': 0.67},\n    include={\n        'classifier': [\n            'extra_trees', 'gradient_boosting', 'random_forest',\n            'sgd', 'passive_aggressive'\n        ],\n        'feature_preprocessor': ['no_preprocessing']\n    },\n    get_smac_object_callback=get_smac_object_callback('iterations'),\n)\nautoml.fit(X_train, y_train, dataset_name='breast_cancer')\n\nprint(automl.show_models())\npredictions = automl.predict(X_test)\n# Print statistics about the auto-sklearn run such as number of\n# iterations, number of models failed with a time out.\nprint(automl.sprint_statistics())\nprint(\"Accuracy score\", sklearn.metrics.accuracy_score(y_test, predictions))"
+        "automl = autosklearn.classification.AutoSklearnClassifier(\n    time_left_for_this_task=40,\n    per_run_time_limit=10,\n    tmp_folder='/tmp/autosklearn_sh_example_tmp',\n    disable_evaluator_output=False,\n    # 'holdout' with 'train_size'=0.67 is the default argument setting\n    # for AutoSklearnClassifier. It is explicitly specified in this example\n    # for demonstrational purpose.\n    resampling_strategy='holdout',\n    resampling_strategy_arguments={'train_size': 0.67},\n    include={\n        'classifier': [\n            'extra_trees', 'gradient_boosting', 'random_forest',\n            'sgd', 'passive_aggressive'\n        ],\n        'feature_preprocessor': ['no_preprocessing']\n    },\n    get_smac_object_callback=get_smac_object_callback('iterations'),\n)\nautoml.fit(X_train, y_train, dataset_name='breast_cancer')\n\npprint(automl.show_models(), indent=4)\npredictions = automl.predict(X_test)\n# Print statistics about the auto-sklearn run such as number of\n# iterations, number of models failed with a time out.\nprint(automl.sprint_statistics())\nprint(\"Accuracy score\", sklearn.metrics.accuracy_score(y_test, predictions))"
       ]
     },
     {
@@ -98,7 +98,7 @@
       },
       "outputs": [],
       "source": [
-        "X, y = sklearn.datasets.load_breast_cancer(return_X_y=True)\nX_train, X_test, y_train, y_test = \\\n    sklearn.model_selection.train_test_split(X, y, random_state=1, shuffle=True)\n\nautoml = autosklearn.classification.AutoSklearnClassifier(\n    time_left_for_this_task=40,\n    per_run_time_limit=10,\n    tmp_folder='/tmp/autosklearn_sh_example_tmp_01',\n    disable_evaluator_output=False,\n    resampling_strategy='cv',\n    include={\n        'classifier': [\n            'extra_trees', 'gradient_boosting', 'random_forest', \n            'sgd', 'passive_aggressive'\n        ],\n        'feature_preprocessor': ['no_preprocessing']\n    },\n    get_smac_object_callback=get_smac_object_callback('iterations'),\n)\nautoml.fit(X_train, y_train, dataset_name='breast_cancer')\n\n# Print the final ensemble constructed by auto-sklearn.\nprint(automl.show_models())\nautoml.refit(X_train, y_train)\npredictions = automl.predict(X_test)\n# Print statistics about the auto-sklearn run such as number of\n# iterations, number of models failed with a time out.\nprint(automl.sprint_statistics())\nprint(\"Accuracy score\", sklearn.metrics.accuracy_score(y_test, predictions))"
+        "X, y = sklearn.datasets.load_breast_cancer(return_X_y=True)\nX_train, X_test, y_train, y_test = \\\n    sklearn.model_selection.train_test_split(X, y, random_state=1, shuffle=True)\n\nautoml = autosklearn.classification.AutoSklearnClassifier(\n    time_left_for_this_task=40,\n    per_run_time_limit=10,\n    tmp_folder='/tmp/autosklearn_sh_example_tmp_01',\n    disable_evaluator_output=False,\n    resampling_strategy='cv',\n    include={\n        'classifier': [\n            'extra_trees', 'gradient_boosting', 'random_forest', \n            'sgd', 'passive_aggressive'\n        ],\n        'feature_preprocessor': ['no_preprocessing']\n    },\n    get_smac_object_callback=get_smac_object_callback('iterations'),\n)\nautoml.fit(X_train, y_train, dataset_name='breast_cancer')\n\n# Print the final ensemble constructed by auto-sklearn.\npprint(automl.show_models(), indent=4)\nautoml.refit(X_train, y_train)\npredictions = automl.predict(X_test)\n# Print statistics about the auto-sklearn run such as number of\n# iterations, number of models failed with a time out.\nprint(automl.sprint_statistics())\nprint(\"Accuracy score\", sklearn.metrics.accuracy_score(y_test, predictions))"
       ]
     },
     {
@@ -116,7 +116,7 @@
       },
       "outputs": [],
       "source": [
-        "X, y = sklearn.datasets.load_breast_cancer(return_X_y=True)\nX_train, X_test, y_train, y_test = \\\n    sklearn.model_selection.train_test_split(X, y, random_state=1, shuffle=True)\n\nautoml = autosklearn.classification.AutoSklearnClassifier(\n    time_left_for_this_task=40,\n    per_run_time_limit=10,\n    tmp_folder='/tmp/autosklearn_sh_example_tmp_cv_02',\n    disable_evaluator_output=False,\n    resampling_strategy='cv-iterative-fit',\n    include={\n        'classifier': [\n            'extra_trees', 'gradient_boosting', 'random_forest',\n            'sgd', 'passive_aggressive'\n        ],\n        'feature_preprocessor': ['no_preprocessing']\n    },\n    get_smac_object_callback=get_smac_object_callback('iterations'),\n)\nautoml.fit(X_train, y_train, dataset_name='breast_cancer')\n\n# Print the final ensemble constructed by auto-sklearn.\nprint(automl.show_models())\nautoml.refit(X_train, y_train)\npredictions = automl.predict(X_test)\n# Print statistics about the auto-sklearn run such as number of\n# iterations, number of models failed with a time out.\nprint(automl.sprint_statistics())\nprint(\"Accuracy score\", sklearn.metrics.accuracy_score(y_test, predictions))"
+        "X, y = sklearn.datasets.load_breast_cancer(return_X_y=True)\nX_train, X_test, y_train, y_test = \\\n    sklearn.model_selection.train_test_split(X, y, random_state=1, shuffle=True)\n\nautoml = autosklearn.classification.AutoSklearnClassifier(\n    time_left_for_this_task=40,\n    per_run_time_limit=10,\n    tmp_folder='/tmp/autosklearn_sh_example_tmp_cv_02',\n    disable_evaluator_output=False,\n    resampling_strategy='cv-iterative-fit',\n    include={\n        'classifier': [\n            'extra_trees', 'gradient_boosting', 'random_forest',\n            'sgd', 'passive_aggressive'\n        ],\n        'feature_preprocessor': ['no_preprocessing']\n    },\n    get_smac_object_callback=get_smac_object_callback('iterations'),\n)\nautoml.fit(X_train, y_train, dataset_name='breast_cancer')\n\n# Print the final ensemble constructed by auto-sklearn.\npprint(automl.show_models(), indent=4)\nautoml.refit(X_train, y_train)\npredictions = automl.predict(X_test)\n# Print statistics about the auto-sklearn run such as number of\n# iterations, number of models failed with a time out.\nprint(automl.sprint_statistics())\nprint(\"Accuracy score\", sklearn.metrics.accuracy_score(y_test, predictions))"
       ]
     },
     {
@@ -134,7 +134,7 @@
       },
       "outputs": [],
       "source": [
-        "X, y = sklearn.datasets.load_breast_cancer(return_X_y=True)\nX_train, X_test, y_train, y_test = \\\n    sklearn.model_selection.train_test_split(X, y, random_state=1, shuffle=True)\n\nautoml = autosklearn.classification.AutoSklearnClassifier(\n    time_left_for_this_task=40,\n    per_run_time_limit=10,\n    tmp_folder='/tmp/autosklearn_sh_example_tmp_03',\n    disable_evaluator_output=False,\n    # 'holdout' with 'train_size'=0.67 is the default argument setting\n    # for AutoSklearnClassifier. It is explicitly specified in this example\n    # for demonstrational purpose.\n    resampling_strategy='holdout',\n    resampling_strategy_arguments={'train_size': 0.67},\n    get_smac_object_callback=get_smac_object_callback('subsample'),\n)\nautoml.fit(X_train, y_train, dataset_name='breast_cancer')\n\n# Print the final ensemble constructed by auto-sklearn.\nprint(automl.show_models())\npredictions = automl.predict(X_test)\n# Print statistics about the auto-sklearn run such as number of\n# iterations, number of models failed with a time out.\nprint(automl.sprint_statistics())\nprint(\"Accuracy score\", sklearn.metrics.accuracy_score(y_test, predictions))"
+        "X, y = sklearn.datasets.load_breast_cancer(return_X_y=True)\nX_train, X_test, y_train, y_test = \\\n    sklearn.model_selection.train_test_split(X, y, random_state=1, shuffle=True)\n\nautoml = autosklearn.classification.AutoSklearnClassifier(\n    time_left_for_this_task=40,\n    per_run_time_limit=10,\n    tmp_folder='/tmp/autosklearn_sh_example_tmp_03',\n    disable_evaluator_output=False,\n    # 'holdout' with 'train_size'=0.67 is the default argument setting\n    # for AutoSklearnClassifier. It is explicitly specified in this example\n    # for demonstrational purpose.\n    resampling_strategy='holdout',\n    resampling_strategy_arguments={'train_size': 0.67},\n    get_smac_object_callback=get_smac_object_callback('subsample'),\n)\nautoml.fit(X_train, y_train, dataset_name='breast_cancer')\n\n# Print the final ensemble constructed by auto-sklearn.\npprint(automl.show_models(), indent=4)\npredictions = automl.predict(X_test)\n# Print statistics about the auto-sklearn run such as number of\n# iterations, number of models failed with a time out.\nprint(automl.sprint_statistics())\nprint(\"Accuracy score\", sklearn.metrics.accuracy_score(y_test, predictions))"
       ]
     },
     {
@@ -152,7 +152,7 @@
       },
       "outputs": [],
       "source": [
-        "X, y = sklearn.datasets.load_breast_cancer(return_X_y=True)\nX_train, X_test, y_train, y_test = \\\n    sklearn.model_selection.train_test_split(X, y, random_state=1, shuffle=True)\n\nautoml = autosklearn.classification.AutoSklearnClassifier(\n    time_left_for_this_task=40,\n    per_run_time_limit=10,\n    tmp_folder='/tmp/autosklearn_sh_example_tmp_04',\n    disable_evaluator_output=False,\n    # 'holdout' with 'train_size'=0.67 is the default argument setting\n    # for AutoSklearnClassifier. It is explicitly specified in this example\n    # for demonstrational purpose.\n    resampling_strategy='holdout',\n    resampling_strategy_arguments={'train_size': 0.67},\n    include={\n        'classifier': [\n            'extra_trees', 'gradient_boosting', 'random_forest', 'sgd'\n        ]\n    },\n    get_smac_object_callback=get_smac_object_callback('mixed'),\n)\nautoml.fit(X_train, y_train, dataset_name='breast_cancer')\n\n# Print the final ensemble constructed by auto-sklearn.\nprint(automl.show_models())\npredictions = automl.predict(X_test)\n# Print statistics about the auto-sklearn run such as number of\n# iterations, number of models failed with a time out.\nprint(automl.sprint_statistics())\nprint(\"Accuracy score\", sklearn.metrics.accuracy_score(y_test, predictions))"
+        "X, y = sklearn.datasets.load_breast_cancer(return_X_y=True)\nX_train, X_test, y_train, y_test = \\\n    sklearn.model_selection.train_test_split(X, y, random_state=1, shuffle=True)\n\nautoml = autosklearn.classification.AutoSklearnClassifier(\n    time_left_for_this_task=40,\n    per_run_time_limit=10,\n    tmp_folder='/tmp/autosklearn_sh_example_tmp_04',\n    disable_evaluator_output=False,\n    # 'holdout' with 'train_size'=0.67 is the default argument setting\n    # for AutoSklearnClassifier. It is explicitly specified in this example\n    # for demonstrational purpose.\n    resampling_strategy='holdout',\n    resampling_strategy_arguments={'train_size': 0.67},\n    include={\n        'classifier': [\n            'extra_trees', 'gradient_boosting', 'random_forest', 'sgd'\n        ]\n    },\n    get_smac_object_callback=get_smac_object_callback('mixed'),\n)\nautoml.fit(X_train, y_train, dataset_name='breast_cancer')\n\n# Print the final ensemble constructed by auto-sklearn.\npprint(automl.show_models(), indent=4)\npredictions = automl.predict(X_test)\n# Print statistics about the auto-sklearn run such as number of\n# iterations, number of models failed with a time out.\nprint(automl.sprint_statistics())\nprint(\"Accuracy score\", sklearn.metrics.accuracy_score(y_test, predictions))"
       ]
     }
   ],
diff --git a/master/notebooks/examples/80_extending/example_extending_classification.ipynb b/master/notebooks/examples/80_extending/example_extending_classification.ipynb
index 08d5d78bf2..2a97dc7eac 100644
--- a/master/notebooks/examples/80_extending/example_extending_classification.ipynb
+++ b/master/notebooks/examples/80_extending/example_extending_classification.ipynb
@@ -26,7 +26,7 @@
       },
       "outputs": [],
       "source": [
-        "from ConfigSpace.configuration_space import ConfigurationSpace\nfrom ConfigSpace.hyperparameters import CategoricalHyperparameter, \\\n    UniformIntegerHyperparameter, UniformFloatHyperparameter\n\nimport sklearn.metrics\nimport autosklearn.classification\nimport autosklearn.pipeline.components.classification\nfrom autosklearn.pipeline.components.base \\\n    import AutoSklearnClassificationAlgorithm\nfrom autosklearn.pipeline.constants import DENSE, SIGNED_DATA, UNSIGNED_DATA, \\\n    PREDICTIONS\n\nfrom sklearn.datasets import load_breast_cancer\nfrom sklearn.model_selection import train_test_split"
+        "from pprint import pprint\n\nfrom ConfigSpace.configuration_space import ConfigurationSpace\nfrom ConfigSpace.hyperparameters import CategoricalHyperparameter, \\\n    UniformIntegerHyperparameter, UniformFloatHyperparameter\n\nimport sklearn.metrics\nimport autosklearn.classification\nimport autosklearn.pipeline.components.classification\nfrom autosklearn.pipeline.components.base \\\n    import AutoSklearnClassificationAlgorithm\nfrom autosklearn.pipeline.constants import DENSE, SIGNED_DATA, UNSIGNED_DATA, \\\n    PREDICTIONS\n\nfrom sklearn.datasets import load_breast_cancer\nfrom sklearn.model_selection import train_test_split"
       ]
     },
     {
@@ -98,7 +98,7 @@
       },
       "outputs": [],
       "source": [
-        "y_pred = clf.predict(X_test)\nprint(\"accuracy: \", sklearn.metrics.accuracy_score(y_pred, y_test))\nprint(clf.show_models())"
+        "y_pred = clf.predict(X_test)\nprint(\"accuracy: \", sklearn.metrics.accuracy_score(y_pred, y_test))\npprint(clf.show_models(), indent=4)"
       ]
     }
   ],
diff --git a/master/notebooks/examples/80_extending/example_extending_data_preprocessor.ipynb b/master/notebooks/examples/80_extending/example_extending_data_preprocessor.ipynb
index 490787f633..bb3899d5a3 100644
--- a/master/notebooks/examples/80_extending/example_extending_data_preprocessor.ipynb
+++ b/master/notebooks/examples/80_extending/example_extending_data_preprocessor.ipynb
@@ -26,7 +26,7 @@
       },
       "outputs": [],
       "source": [
-        "import autosklearn.classification\nimport autosklearn.pipeline.components.data_preprocessing\nimport sklearn.metrics\nfrom ConfigSpace.configuration_space import ConfigurationSpace\nfrom autosklearn.pipeline.components.base import AutoSklearnPreprocessingAlgorithm\nfrom autosklearn.pipeline.constants import SPARSE, DENSE, UNSIGNED_DATA, INPUT\nfrom sklearn.datasets import load_breast_cancer\nfrom sklearn.model_selection import train_test_split"
+        "from pprint import pprint\n\nimport autosklearn.classification\nimport autosklearn.pipeline.components.data_preprocessing\nimport sklearn.metrics\nfrom ConfigSpace.configuration_space import ConfigurationSpace\nfrom autosklearn.pipeline.components.base import AutoSklearnPreprocessingAlgorithm\nfrom autosklearn.pipeline.constants import SPARSE, DENSE, UNSIGNED_DATA, INPUT\nfrom sklearn.datasets import load_breast_cancer\nfrom sklearn.model_selection import train_test_split"
       ]
     },
     {
@@ -98,7 +98,7 @@
       },
       "outputs": [],
       "source": [
-        "y_pred = clf.predict(X_test)\nprint(\"accuracy: \", sklearn.metrics.accuracy_score(y_pred, y_test))\nprint(clf.show_models())"
+        "y_pred = clf.predict(X_test)\nprint(\"accuracy: \", sklearn.metrics.accuracy_score(y_pred, y_test))\npprint(clf.show_models(), indent=4)"
       ]
     }
   ],
diff --git a/master/notebooks/examples/80_extending/example_extending_preprocessor.ipynb b/master/notebooks/examples/80_extending/example_extending_preprocessor.ipynb
index a44db2d221..fa090d2f9a 100644
--- a/master/notebooks/examples/80_extending/example_extending_preprocessor.ipynb
+++ b/master/notebooks/examples/80_extending/example_extending_preprocessor.ipynb
@@ -26,7 +26,7 @@
       },
       "outputs": [],
       "source": [
-        "from ConfigSpace.configuration_space import ConfigurationSpace\nfrom ConfigSpace.hyperparameters import UniformFloatHyperparameter, CategoricalHyperparameter\nfrom ConfigSpace.conditions import InCondition\n\nimport sklearn.metrics\nimport autosklearn.classification\nimport autosklearn.pipeline.components.feature_preprocessing\nfrom autosklearn.pipeline.components.base \\\n    import AutoSklearnPreprocessingAlgorithm\nfrom autosklearn.pipeline.constants import DENSE, SIGNED_DATA, \\\n    UNSIGNED_DATA\nfrom autosklearn.util.common import check_none\n\nfrom sklearn.datasets import load_breast_cancer\nfrom sklearn.model_selection import train_test_split"
+        "from pprint import pprint\n\nfrom ConfigSpace.configuration_space import ConfigurationSpace\nfrom ConfigSpace.hyperparameters import UniformFloatHyperparameter, CategoricalHyperparameter\nfrom ConfigSpace.conditions import InCondition\n\nimport sklearn.metrics\nimport autosklearn.classification\nimport autosklearn.pipeline.components.feature_preprocessing\nfrom autosklearn.pipeline.components.base \\\n    import AutoSklearnPreprocessingAlgorithm\nfrom autosklearn.pipeline.constants import DENSE, SIGNED_DATA, \\\n    UNSIGNED_DATA\nfrom autosklearn.util.common import check_none\n\nfrom sklearn.datasets import load_breast_cancer\nfrom sklearn.model_selection import train_test_split"
       ]
     },
     {
@@ -116,7 +116,7 @@
       },
       "outputs": [],
       "source": [
-        "y_pred = clf.predict(X_test)\nprint(\"accuracy: \", sklearn.metrics.accuracy_score(y_pred, y_test))\nprint(clf.show_models())"
+        "y_pred = clf.predict(X_test)\nprint(\"accuracy: \", sklearn.metrics.accuracy_score(y_pred, y_test))\npprint(clf.show_models(), indent=4)"
       ]
     }
   ],
diff --git a/master/notebooks/examples/80_extending/example_extending_regression.ipynb b/master/notebooks/examples/80_extending/example_extending_regression.ipynb
index 5783cf90e0..7e7efbf899 100644
--- a/master/notebooks/examples/80_extending/example_extending_regression.ipynb
+++ b/master/notebooks/examples/80_extending/example_extending_regression.ipynb
@@ -26,7 +26,7 @@
       },
       "outputs": [],
       "source": [
-        "from ConfigSpace.configuration_space import ConfigurationSpace\nfrom ConfigSpace.hyperparameters import UniformFloatHyperparameter, \\\n    UniformIntegerHyperparameter, CategoricalHyperparameter\nfrom ConfigSpace.conditions import EqualsCondition\n\nimport sklearn.metrics\nimport autosklearn.regression\nimport autosklearn.pipeline.components.regression\nfrom autosklearn.pipeline.components.base import AutoSklearnRegressionAlgorithm\nfrom autosklearn.pipeline.constants import SPARSE, DENSE, \\\n    SIGNED_DATA, UNSIGNED_DATA, PREDICTIONS\n\nfrom sklearn.datasets import load_diabetes\nfrom sklearn.model_selection import train_test_split"
+        "from pprint import pprint\n\nfrom ConfigSpace.configuration_space import ConfigurationSpace\nfrom ConfigSpace.hyperparameters import UniformFloatHyperparameter, \\\n    UniformIntegerHyperparameter, CategoricalHyperparameter\nfrom ConfigSpace.conditions import EqualsCondition\n\nimport sklearn.metrics\nimport autosklearn.regression\nimport autosklearn.pipeline.components.regression\nfrom autosklearn.pipeline.components.base import AutoSklearnRegressionAlgorithm\nfrom autosklearn.pipeline.constants import SPARSE, DENSE, \\\n    SIGNED_DATA, UNSIGNED_DATA, PREDICTIONS\n\nfrom sklearn.datasets import load_diabetes\nfrom sklearn.model_selection import train_test_split"
       ]
     },
     {
@@ -98,7 +98,7 @@
       },
       "outputs": [],
       "source": [
-        "y_pred = reg.predict(X_test)\nprint(\"r2 score: \", sklearn.metrics.r2_score(y_pred, y_test))\nprint(reg.show_models())"
+        "y_pred = reg.predict(X_test)\nprint(\"r2 score: \", sklearn.metrics.r2_score(y_pred, y_test))\npprint(reg.show_models(), indent=4)"
       ]
     }
   ],
diff --git a/master/objects.inv b/master/objects.inv
index d5a7ed44583e3cc6450e57e9f6be2062ff009d7a..5177b72da30112d48155973997a91048d2283547 100644
GIT binary patch
delta 7268
zcmV-q9Gl~nKA|~~L;*CBMLd7qa@@9(zSmQ*%Bk9<>Oj$qG?L>xM|N^t$2!(p`MG3q
zIM9g49CAW(B>8S%V_$EdWbsP^BtSMmjGT+a8WR2W*C5aU8h^>5DeEsePO9QG&6B#$
z?z1#$va&dROJ3daSJ&bVuA=?dH!NMWh?0GF`q1q1SKmeGByj{fz0ZFd3bw<|L%9io
z7UQDEyLFBO!|t;^&a)yU>=rk1S{C=&_E52OalKE{&>D8-2IqDBlr<0W6VA2|O&wUn
zKB<yjNKo>7gR6C!RGWaP`wDNev<Z(G@2hg11jMay*lJOyWfihm)yMeg^+Wk23TsHz
zeU%kWTsI_9-DD{c8<T&uyQC?rdddMJ1;v#eNa<(1t#DnJ0G$<2+SAto^wd4?ah2_G
z(Iok4LY^*FsimX02cyAVz7wRQ-*>}ybN+DbSuKE@8t-PnO^u6NfFnCM2ykepR>2w8
zrD=eJJG2jQP=`js0rY4^!nwI{fc9%MNQ#OiK26e;)aLXnl}dja@@#!*Fi7wbSIHKW
zGK15MdU0ojb&@AViZ`)0hx>(xNW4qZs`L`fJO%4yx6g6yBg(xW-Z;zV<u=aC+Fj``
z`GR9szgSy(e<)Jw>_MV9DdJ>ZmpRcD<ErBQ05|m%qyFoV5TE1?vB~BW@h{k|q-Rl6
z<g<D5z6GyC@GgIU)zV!Ar%9fBih1U>6wYP{`{uF`_s`-ojgvzPmXjUB)3n^}%Yrm6
z)Ct$v)ZgbLfw%d#tVkzp2kKSNz-QgDc1Qth!se)R>PQ7p<kGc*sJpE&K}_lRsn~}9
zN@jT8XKr7G{7x2yO_db&eOc|oY5Lp_JHuz20sB}5+oXSLcA@Tt0g}xoPPKux(iy<d
z4K6n1Yr)~~anc+rtP~+CPPOz4)jT~)WPyTLYD@x-JQbgS6q*SPcys{(<QP+cAi6~o
zrVwYv4<O*)Q3D9L%fN67&pARc1?W`sr@$Nr>{Cz<!}Lk0z~8`mNInVZFc9B7@{uoo
z7G1}>*}Q+rlQqt1xKuv<F)KFsSwv2S2x>*%iYL)~@#KW)U_hHPRim!a+uIM_%T|lR
zNG5ng{;Msi*sC_-`jF)f%8E$zMJ+-y8Is9<mU1`z0Rz*U9Sc-1k>1}j!HhH06Sw@X
zjTj7`H5Y6=NIo;J?uqNNfwxJKlJtEm|Dk=zfOmfh&Bt<c$gO}h36e=}l=K{6)Kd9Y
z&GD`*TZEJZT~DJgI<B`%*yir%f$RzC*JP<fJz(j(O7JuN+0}qG>1nY#tpn|n;*jLB
zwBPUqILE9fa3b5ydQ9j&PeghPCvdXgb<En=%d?o-@S}JV8%n>Zz)%@p%X>8fD$J_#
zCQ5&bO+?*(5-B}Cvw;vctmbbJK}7@=B}hU2*{Tpuddy|G5gqp{U_zUuNus<=Xn%4J
z9g2e_&N~YFk^g{GA-$ETdB72Ge9L?G<31~rJZig$GZ+$hL=X{!N98^H!QfE~|Er*I
zXQu$ShqbFFBS>emZmL5{V<+pc5pArIXsv(e7|8S)tU`)DwEl`TeYYMY;aTvQ;U@+z
zIxP3)p$KX-RlrQ<3J_B;F!B!wjS;JMhrG!cLocsq73U7{hQ3BT1HjeTipl*Vt%cNJ
zcc(Z`_LW3uncjDfL#<vG`1qs=+CJ{dPB5USn&5qWr)9Ttr<6;-c5A{oMFHw52wZ<B
zxGPqMILW#^@LbazMRf>e8+XQ?M#MC-&!=icRO2n2-3;rG(-48O;oG)J_KXSIJWb@p
z3$|3G@azvw+#4*xnyG=T+D7r0>d49!ZP07)f}AedY-Ty=ym#O~+tj*S;N3b|NP9O;
zLu#^J(xR0Sk5&d?DjOpW9&z|z?W%wM?FiMrPS|tg#&u}lPOVM0n1-KCDl}4{gH_<8
zHzpRt{x&ViFiXNWBbGdPnKs0=-RQR!pwbnxNjPpZFw^yiaQDY?jW6#GRX38C97#+8
zk!Q6paf-#rs#v!M1wd*?^l*Piucnyp-WK%@G6&#6%7ePmmvx=>Hpb65J+Ob-;U?RO
zAiLFd4%3TRc)25fpZA^EW00;lX<uwAmn`~UDL{w=jWD`6i{<QI7o$t^ls93lEmJ01
z`gsAC76TA5@8T}+K2hLCZSIr-+z>$X<47IQ7{TfJH%*SLgEv%g8;w(kzIOCyI7e|d
zczvKNqqkf6^&vG%lj1e$qkw;)rINkgPJV9<B}{!95<5mc(ut=PP8uB1I>fVp4_4=~
z=AejHBlc8`j-zX{G8ev{_1!?yv!3g8a#O*X$4llJ{pcN}M+ucn^qQ}G3~z5VDPB8%
z;g6!<=pzT6VQR&Y&@pO-PTaGU8K`=eGM$e26?I?In~!*eIF$F74cmXN8F;=(d&brR
zigl53x9AjHS1j_p)_htwba4-z%*YD5?zuNr;{e#JNT5kfSH8VUWtz&g5b!g}vx<$H
z;91$M6fyaNIT<$^r=e;x?nxga6z!9$i59_Z5K7JSWLzntDv6eNE(DLC2g{ewgQZg*
zEdBBzE9yN?Wh}q-62gDs&~<8&VWFD8`9hzVSCH?d6Quh1h&}oMB<5#!X>?$rjO6Ha
zt{183A-$$tNqwm0B8;G)Nx&lp-^Ed)TeDg}8!Egnt7f!t2Qr6i8mn^A=SecNjP#6i
z`C2*|`TA_9MOTfZyMFA;#6(o)xF6Gn^e+suBJ#ru+WY5yo)mw83Sz(!@i83jakZ1d
ztZYXOrm{ZGnvJc9pkbt_LY*qvm|epNd05Ivom4y{UBeCD<6?s;=J)%}53Ey@6`Sla
z+Z@PiG9=w(RFL)}pD7rp@_m-#A0ir7uYKA(d*<-5vS5cgc4*IgV;wi7a6<|=l+o0b
znQRoV?Iz#8gPVW+;re~w<iOd#m-xihlS)R%I1fn?k=h!`L66XUe(cQNi!3Be5M_cW
z6GWIGIc8#+dp@4Iw{U~pvyyggBi3=IUr-DRoEtT3?ofEd`p~r8iSdAF)M!VVcd4mr
zF&yB2$lVSOi97cdf;|I8DyXxUY;Z+jn`BL+lHB9NUG;xZnEUDv>KoCJxDyEzOR=&%
zMo7hIhUc))PTX_MXCxOks$u8+BEb!#g_)r+1~Ex{Bf3?gihjV!XE&q1!AvwbQ|O+@
zD0xii^2!|d7b6G!vQ!aj6`}rSJFe&sz-GRY>elk37^f?^8__;)jO!$?`?Ri{d`k~?
zQ|==7IXHi9){UcdJ104)6A@3x9%AZMQn=6Vn+K`x4H<tEzxh-|?U7VGT}1bkN6Dt`
z@+Q^xKu_oLWzE#eZH$O$MMNu-T1BQJsuguwQLFVAo%QdItf5&`)s3Dtv;&2Siz42r
zugg0lh!QO$IvxG|AZBZfejjB7G_pKY-=qLAy)%DGK#%uvioi@4;WRNXb@G;*_=Svo
zYto15oasz9yk#cNf&kMFROiAa@a>dpm<2VaqvPo^<!J|ppDH@hU5|yOguqN{9<s_{
z5kFaNo;eJ9U=2f-ru?RB|J|T{+>AXLi6nL0R7qCE4UKU_TiJmmZ*OmL#RAPV3{XDe
zimiY0VxxFRB4Ng8W|qeopbZW1Y{V^;J)83(t+3-V3)>eNlnrg$TT9=3lZ@Cb%oVWE
zx2u-BGmARx05-^l%<bSyg})tBK5j{c*;#)sE5|w!xvIZ~xFft9yv?STQFIV*(yFX$
z^qAz?#x)4mBl2nhb=`XaCgte?`h+~IzCwQ_Y@P4(dy&fSXwCP@mK6OTj%yWSN+7mL
zK42Rk%6vm-1O53+gn#mX!J0*#XPZ}O6X?%CAh8WH6zNt+1~bx}i}VaLcy+&#f;Z3q
z)M@sGbR|T%j67=^(wQjN|G{Z9hpUQ<Bra-k$Fz=fr}OD_^{~F366m{4!eV(N>okAy
z_xfH<cRIfQ)5^}*lbXq~(KYEQM%^ivK{zs8(m6#P?z^Eb561V_Zxe(#?@IIc`WD`>
zrGlAJ(4kfp?W^)J+u+Ty%qi%&t3pzZ$*fT(X7hl%dB#O1H}s5)gZK0RSh#KLxjc*c
z`WB_PS|xdo=~316Xfb;H<x@dEOq_q_nK5VFAjHU2q#&c@WJ?CicOpgfR90VbRY&(#
zxs$8Ec60zp*Sq-Aesgvj)wntA$pA9L^rVjT0n?M|Dh7Wj#!ou&Fz^*qhx+P}u8z>U
z1JLp|dvQuvG|_cUEpR{!DY8PDD7({@dGlUM-py0lmWa^I2{w}0OUYeP=*@rr{4`>#
zG1#UC@oA%s9?;)B1TAAy3yYbs0+ba<Xdxu7O(bsFYj4}uN&015v9)61>`*{|ZvXVA
z$X8aerF?kTE_x5clw%_W`$J!wH*fipv|!97)*zP{&^Va_F#<>?kd*{c{c`0l)sL9|
z7xv`bF97Dl{T+h_&fUUW<BWf{BlV|*^iU2(ljRY;KtkqEQqfuEvc@5+6xD^8E<{qc
zpOEN6Lf4;DY(RDv@qfy7?b<EiQl_3j6cOLmc20Lm`T#8~9Me5e4Wl_XNIgv)FuVd%
z1&phv>FD0b^i}4O>^U3KRc%Kr5`SjNsK^&7cd@dRe7b~cG9nVn9+7{3n!X5pLe}qd
z4%Pd8Or6T0BJkz=Jkj`4dl1Ebg%H~YcuKjz{ueU|&)l_{R3_CiETOPA13ta|4Cfsz
z&-Kg&=VFcjI*?8ZUEB++5S;+84R;*?tCcWiKD+!TxrEr(H%i*83inBR)nA2#|75{5
zx8PF+qm~3*>o{%NH&}n)IBL97@pyykYgqT2x14S@@{qFoY#wp^kmQfH2||dVa0l)4
z=%4(#oB9ylH4C@!JvIY<KNn#k?B^;Bq(po=qW8AZu{GVOK?nb2ctZ<HL!*nWm|75K
zu@{kYB(>6ACNv_^X+=Y=?*Lr8LgrV-cG)wnw}%b&rOAFznsa}L#@A*qLq!}FwWv2H
zU5<PU{M`H_dZHT-G4n=j9n(GeX6trREFmK`sU|BteWt}SsvkYkt&x~{P`23V9x3Wk
zLWYvk_N>so=)c|RLx<Of%<3LRp-OW197)L8evalrQbbvU=~TocW{zjDwt+2!Jc32^
z3nm3>wKI>{*Uo=eaT<*OeGYfWd?LjrCAyNlE?K#Gu`D9xohU|BF)Q%k-1*a!u_<GK
zQ&D3AlM_*6!r*DBNd|6{P_NJ8_L{r)OSMlRw=g&%@H~1i-)dW)`RD3guyifAS$k5J
zJ4JW-UhUzuySLevN`Mb{w^Hs4C7RW7ZL68>3Dq=<J)nO#A8%n)eX|76b@DJ5(1&uP
z*}l-f4=W|Ta@plkFmu|s`ZWT~7xlI)b=qIDVAH0cSE`$mwfgM)2FnQAVD-u{F^%?4
zm>wFO>6KW8hOcpA#7Y-px69F!LOYKon1N!zN>4y(m&ntxi=yjxk`CBVnk)KsrXEgp
z05a`YS&e_ZRrR`;z=2xnUFnM&d)7X)G17fX=xa^We9A?s-7*`zMHsiLrybBL{JXpb
z@|}@4V=kA~cCFyf^tzTD8gsm?jZsB_zV|qVUxqjN$pcgoU1us*dr~63Ec0$|EjcA)
zul$e#p&xbn``JbH|KiShMh<Z|-kqu-VyPiKxg~#3xHIoGiU_OGpHJ`t+dbEtyE*ci
z#aQ=+;QLElO?YLxyX1Q8&Jriy0h*m-wKyiC`6db8z*zYcWq|hN<GZ&Z+7NK_5whiQ
zYcxFCi;#6pZzcHDB`pt09Ss~ufSk}91wM;POGHv@8PRJ9I4xQxdbIp(KBC|Md$|0G
zt6P7-($nr<xf0pEAS#zWjMYuWnwd4Te;dSDy;SahhSkbYSz)S?U#cpM#isg4!A<q;
zU!Onw@ZaKs#)Ib!^oLi{a7aF;|5a7s!!V<-*f9iOa8(d$Rkqn;b2$ss8ukrp86eI8
zF$TsmH*6w5%%Ja+T}O16=6i5n!dVivbl-n{6Bmcw8dq_7A3uC00}Z<TY=;}Ut{mh#
zr?J|-vn-uFBb6toQ(%47?6Va_Ewx1*PS3W=JG=@s(mT)!piVubSSUp@h=#l7M;D=H
zQO7CPR&m{95y#H}lJm2pTQ@Uvpaji>ZtaX|_;fo)XTG)7{K|8TJW*(|oZ-pow0M6n
zIrrlQlC+dw#oxj&)Q2@Is9JOJU$wo$9U(I%2faI1TPt34XCnQ7|MS0g+Lnw0POwhM
zOwO@R$b+X@Aw1k>TC-w<pN%%NH*zGvpY_Dv$eo9+kcD;!^-7AMR^+V&v1ws<mN}?k
z={x+NnYaxc4%?=*Ly<4T`e81hLI!^+@Dj$RNG$S&YZ2E7v;urxu!=8-H(Egu5BsCR
zUcXWmRamIPufCfUe=~l?yoT0GI=CQSLRtowW6<W2;WT@B;PGf!UDMQ_w5$ih1<q|R
z)GhVU!`^ilD%M@ZthxREhf$q$b(StIZ<yds`6POe386?5HyN%0io3*7UsivCwO>K0
z0ov}FlrT?KCU-j8dI1`qNRN~lW%TC|<D{*xap*dJpZulPKJrr>+3@zi+jp%R`Ib~H
z@xBe3%g%&w608($Xpu4lC}P@;BdyqV4_Ek(TbY0<7Ym~YWxiFTBzWKc$wck8(KpgM
z9tp@TDdk%pA;_6TR@VoT(<6Vkm9^qLY3~-Y51stG_7XA*OzK6#&-Ab(>S=E1uaDvv
z!zf8kRu)<Q?i2k#OG^rOd8ksl#i>ZPczAkaD=AYVCFN%+ebYXZsE53%|3UAki|(^r
zm?~}vx8hS7det3bA}q_=OT`7?$CA$HNx)C+EmNu9<1rknv|5x>$diBl)|I+DHl`q-
zyX&^J0P?9kik$p;GU3pU>@0ORH_Jf|x@n53T9YI?X4dvj?TV?13?Qj+e-NxxxKC|L
zXDBa`%Yuvx)`w<LgH%X0(Y5m9MC-X(<>h<fVTDwlrPLQ$rR4jM{9hr(x7AvO#OoXO
z&xyQ4NlVR>NT)4sGm?KkRaC4Kw#q%v2df&@=dyTrPljR&$DSb9Z5Dk)JOn*^Gl^UP
zr4Z7D-kP9vZCr{B)V6p^hnn=d`&-8&dG&=j&TN-V1M;+pr+vo-KhPgKq)R?%Y3+F|
zeoR#?3;LjreK5o6I(Ot-)KeOGBJ%%Fce=D_KV0%cG((!m9fyCh=S=;zZt#xOg})A2
zWjd(ApofVdP2_F}qw6`W4@GluvWJW4_WVAC(e*qL2T?vv+8~qt$^SKC{K!8J?S_))
z>nzWjuRvVHM3FA~X9xS~PKQ0ovt8EDdEI%nT?(G`L~)M7Un%LOQ#5%{9ac+d?NO>S
z=1!#m$c!%54%UCAXZ>KA-r}i+k}0e7hCvGPOKfnC(_!J$q}gO$m+(dxlz4{Dq>7^U
zCxt}2mtT!LSpnLe`D*H01B8UH^a5f>Zab_Zp}AyDbR1PcW8YB~Jt28&=sWJeBKe}d
zb?lI1(Tij`I9-5w7PomjlG2!Y3Wo`D{Ky`iV?|0uQtN-=OZiOBLh)a1(ykjuZ-96_
zJ6)XiG;08_)a2;#3@y-e*DvcJM)1#0&zuyTp|ha|0y3gv;izI^Qn8p&v2aweFsWFK
zs8~8$u{2q+9JONUXvNZG#d5-mrK5_aNyTzh#kr%3bCZhmQ56@CDlSYaE=E<X9967L
zDpsQ^E**bWT$)r|j;gqFRB>feaW$&qwWEsHCKazoRGd3jy>nC5J0Goj7mg|}Oe!uW
zR4g1-EKDjEqbin;DwZY{%TX2Qjw;SgD$Yk$TsW$@FsT@+-HyK8<jO}qxuYXD`SDRV
z?pPsA^<lI+IMxPJWw;o}U%25f4ET#N{G}WI(tv+IQVUjY`Li<Q&uT1xR&M#TGUU%{
zEPqyR`Li<Q&uT1xF5U9y(vUxwWBGIChJR(izZ%D1xZy7h_#^rM+KvBf1OL}!{C{%8
z|H**=lZvm6WGtX<n&F`45mz-oN_S#7BN8G$koiLuGonolUKBVwZ23bIGy*QbfD0XP
zVFiC&fB_de;KB&F1T$Rf43`#$OEANw&TwgExC8?(b-<+s@Ei<yt^=N1058CR7dqgD
z1#krhT<L%-3*aRf@KOi7v;ba#0k3qxD+}OjFyLz)@U;=}99*!^^@4qFDcBcazzZGl
z!V0(m11@yHg#~a423+cZOAFvR81P&NJhy)UUVs5FbU<TG0edgqd0BiH*mdci%i_4e
z1yC=4mf{C4eR|=$u%RuWXbTN(VMbd*(UuyTvDB@gmaa5QS7u9BP)k>ur7N?gE2yO_
z&C-?G(o3kNmzt%QW=pT2XjdBAl?`nHMO$cS#(V?%cGa;Pz4}j38-LPlq@7RG`Ky2G
znZI(;6P+Ozi#)=uzMPoT_-;6fQSTp~<85BWoARlkdoEfw@$xKY>tE%Q{@Eh?uNJ~m
z@JEXvH-#uhL@|<zTdUCW3~38*`^V7`#mI^H?vOVb<LPgM9mCn24l0ageA$$`6Ux!7
zA%LGv97o-Q=4_Cc5p#$;kX0_o>FR$(lBYZq)oWd?)J-=NnIWbOxi~}p1@QEE%oIr^
zd8e<JS#0rQoR@5^qMJFosE-*80SN>z`eH?N+<e_*$FHG_`p}olu^ECFeYR#V+hcm0
zMcl3!b~MBW;$sq9Qz)>yWzBxZiPaHtxB1f1FcW}}Q36NVE9z@dgDV2k(0hNfW46-5
z*&+jipGgW3-=Xbilb0i#<GSWJtKl8EPP8po=sFs}Q4mAnXRY~|qTCAUWW(S#NjLJ!
zMVUUVjd$xg@wea3=V*!PR>Db}+9O9}&@DJt%e?LdvfFoghV%|UYS&M3xb7Y$0yoD>
zaeIo7Vd}*zezp{dB6!f(@2-DvAaA(z!il<eLxu-&<=}k>)`R2vT_`JO`+M*m9&GHw
z+jbVV6aT^K$dh)cJ6z+*lQOLac^tSoV%h66)NKru#&u5|$Ac3l4v&!YwbF7Joxh>O
zgAQlf8pO<y7KStqGAedFC&k||hY_>UEWlQNHzUDk12~#2aYs&M&@+F)A&@7j47Y+0
z(Apr02W%Vf=<i*(YdLbb&n3#&+l~$3{F0GA&j8;IeTg^^PXmu0@biiDal6_rob66(
zfKg$Ce$ZGSjHBgPj}d~l!!$++*3{_;5$i!4Lz2o!%Evb-<0r$>8Kdf6I{xVw_PpVL
z0!=wm%+oE@`H}yrZ#I7i8gQhjr(3X79y{?HeE=l>g7Hq@Fz2UrlVn9y9f~L`=uy%(
zkNu58o<5OIz6tF8^FB`sH*;Pf;OQ0T{GdMPCs9JnV8DncG|=hho{k*lbPsOc@-x+G
zN>o15Gwet+747q~0a_G1*y#}C{M`PE?izX&=r#kps}~7(dPRRZ`9-qg9%Jq8glXVj
zj%p3Oqodm3+1F{~N<KZEHlAcyF=18oErY=`V!-;<)^F%RC+vND4f}mn?!+OByB+Ok
z(B{r9BSOcE-NM9>%ex`w93_725;oTFR)$ax961IDj0ZQtg2)~>Z0sB@Hr%v0W(L?N
zPTyEJ#@W*!s*zYQAD5^`yiA9Pjcfcy@pS(%y4S^J&Ux^10ry#PcHf(Xob%(@L3!U;
yyvNlT&bA*jX73lOcfNW6X!ExZ=!b~ohd4rcBE!Hf3g~d&3F|@a%Krn-enLMeJ3BD|

delta 7901
zcmV<39wOnPIh8(;L;*69MLd7oa@@9-efL+e%Bh;9YM`iF7dv`6CCg5tI`)aiO5%Cz
zVzZ$ZkJ)607hT@wH|F=|O9mSk5+DH*0MT+DTym4xYp)Fg8^FdbZt9}?n&G(2kCH5|
zs`M^R;yNwzqf7GYmc6>-Z*UpypD$>-Xc5Kx^yt3cWv{=B(5u)H=;(hgtpRL@>-%C8
z0&Ru!3h&k#4h*|b_c%-Qkg!`^uaY9aOSer))2*s~oP^e}D>gW*R*z|Yzk0;!_P(wH
zYuLwSybB3Res6HOF5+?%5OrVTO`6o<5m)=NSjPczOB}XZR7p{WELPPaK6-UuJo3UC
z5_MmudA+J?lBlZF#1elSleD|IF3M`k0U`;-m7SH+&v;wnswymWmMm#cUkA`r_q4}l
zy2E)LXGbx4x|F4sj^5rH&EDlZK|1<<H*7cOZ;w5z1#na2-3+*?ad8Xw$j%LdJ+xD+
z;0){1G}wbXv=8>64vmBZ=)sDFb93PU?bl|I6eUS~6ekI(&B=c=NX0dIwr*-{CHR2L
zc#BDy!AS~U+!<jVXK|k3&B~j@-NHj;wTqLo@DfZt1*>?s&v4}<%Df-mILl_mc9j*C
zyV70!6|ZRhqHXD2lPBQptweF0ui|x8WJK2rmnG{5xTyz>>aQjyKFJ$mlg=mNU$9$A
z&!Q&Dr}N}}3toSR;9dU8rMn1D;w<wN^UP}@oX!yT&1Ej`pT$M8ikrk*PIL^9l47?n
za?-ef6RxnXzt2Vjm)W)`NhfS))vKO?PrGC7kixDBo1@ODBNafAOV`Rq-A#!JVglo*
zd>j5N>ET(QxqTJ#I~f$#Wt>-cMY#*7>2o{m44-Wp>_dMQY@Mpvg}N68h&P*6q71C1
z&H#R@alRp6%O3s?$8}R;sR)5MXz3TKd3vPC0)UrtOu`;{Dn4OTXeKc1qYD;5jxhxr
zM7K!76ymJ-fepBK)W8PZWnegk=NuuJ0(7eRQ(z7Q_9-ZbVfrLg;BVkOB%cIy7>I8k
z*vJ<@@veVkRc|h`c#Sg%mx`x9ruhax@yIC`L9NIoe-eG*PhJrn6lhZ<a?~}tyt(dP
zwpt`cBEbduR~b^#S8c*|lV&wa^N9CF4MHLr63N~Pxlil?1yh?H15_`O+TYT_j5AXc
zxBRZO7!;m17j!&GJ~J-wi0d+ew@Hzd^g}Cu-9CS$z`GdovDh@35fG9fk>rz<o&k(n
zD&ERD-j!v8kd&b7Y4k<M^>zu{-2FU|eMR~;Y2r{1X!@=ayn~-z4Y0@gZYcEQ@_0qz
zf8<YAnu5vemaPn{fW5+MD@<$ICW`Y-1Rgn#q|TV$KpQr+Qf?4IMFbVaNJ4#QRA^6n
ztTunU5go7c)`T{39Y<LaLsxGO9l${X=RJUaWIybwkd8!D-{Xk2i^U!NahK+C7PY;%
z84L+LB8Z5>qvDSKpzx@L|AkYyHB+#+hc<aSBS=55s>>#Uu*Uj1f_4<~hIEdBRG-Ew
zB<VxzXQb%6@gQlR1rI5Hq~N?SaaS~XP@8{20X3N^KvcoN$Ukjpj99g6vO1*<y}X`L
zoH@Wx@EWlUu&%~NOy(D%7J|X<PH~*<Yk^KPz3&`{TD>gr@ktZ3ecY3oU_ehb!Taz|
z%WUONDwlrk)`W4A0@PCwxK1!vER8<mb<wa~QyfLvgtCn}<4z-@8tLbgH6qY>W6ytX
zhV_SOh^(^V+qOyOj0xI2P2|K2wp66>?3;Sk8^_o+Qv+GKjbbn5L60lipx55TA6>Fp
z&vMXt@4(-g)Vf>X-8vaadpAu(YNB0I0*Mg|BnDtA8Y2ZBG5BB1s{QQ<)xJ*HbKu5x
zXx~n*O{Q>zpG`6}lAwcC;Da|N2E%{;HZ938OTsrJhCFzgHpI2v=(iPCr7L8UaM)&G
zrt1&k?hoS{U)~+cZX_!h5}%zR&njQyfJMp5ST`C4medaD;r@<ZO)}lRE$SPjkBwU?
z59&r=)^%3f6@J1=Lnm(QbjO42M%Ot^&tu`mj`)4nccPE2bX_EUv8h}#=zo7L0U-u7
z!szsPB_`gw7+stttO=uSnKTE|&kJj5J^<lUBJT3;69sP6=1v*d8)DJ?I8wK0jO^+8
zH%*SLgEv(6Hfm?}eC_B@K91sS@VbFT%iFE^`ZhI6kzzIJBZq)eNndX#zqf)Craldc
z9ikqo#FG-oHIASTu`FPN)p>uc87P8kM4$4}adf3u=EB#rzH3N&)^n9krYkt}c!@lN
zkKRFglu)sHu72KQ`}RhWVzuLE_9*%t9y#a?Q!9pq4pA#q;+~~UL)EjCsdV_SsQZ%M
zeE1{8puE3q*mh0B^F`V-x=fERSzLANL9OeGL7vr`-_|v(yMf7!G>3mB$-Su>2f$uM
zvYNzn<=d-Nx~WV90Y8&GtJtUso|Vl=5tA>Nk#VDT8mc1Wp5#74(LOHgXc5c?u2ery
z#*`vZNwmCoA$a&aSiXE7ES>UT>6Zs-UhQ!rV)?C?;0|YvtD23CLz&O@@p%RDPB=l(
z$42bYwIwk>vkR>Q3uS*KLx;IuB%_D)nqrZ2Q;9_wK|d3RM-;w`qeM4)wSG2~cwdzD
zXyFcI4%0MN<)Y7%WM&!Z8Rzn~bTabw*-Q%-bfepT>`TXlSLRherfbu`(8%)04=rdP
zp7vRsTT~DY4v&xFXphUC2xg@_Fc@Tgnl%$!9znxMQiVEIqA`EFvLobSDI0ZycuKm4
z8@$K)1_S2z`^^uuQ<LVK^da3e<TV+RZc-40UgQ&iF)H7u3H~91uzKaw-kCFpjg>h&
z)UiW*)*I`%A&DE3xS<qM12fSmUYSk4d2es>_0@;I$$_(fFYt-0CzTAwIQMZLk=h!G
zK@Zn_a_G$7^DKWPOb{?ZzyuK{NQ{|i=AMja?v1@c?pX<4+lY3Y;S2IXfpeo~%pHJ7
zv=0sCj*ka;qXr#m)};p3d^o`Tkh>iW5_j$^1bYgKWKd@>S>uwx*6|vmlFZ}VyXv7Z
z^VQe#8{Uw(6A2SbF|s^FNJVRg=djOA+;hyQB^NfTVdj7QBEb!<g{h`61~Ex_!@E_X
zihjV!XE&q1!A!JuroeW@D1L}xd1Z$Ci;)9<Sx|&pMX0|Ijw!meU^CxHdDr(rj8hfd
zj%Xh@#&wd{eOgycz9mgn7rV%P4o;bMV<_FGMh5Cc#M7~dn0gfy?$W#ZUZ}eu<8R_O
zAM>a^EsB4oi(tEV6mQxtZ(MF0I5^9fHB~Ei7$Tw-5v@pQ<(Y~=E9$hOR_o6y>pvV>
z+h$E!)oRv22MQAxMZ8g8mv=@GB}ztgI{NuR%vKovG0Mo&$f7B~NdcgGXOw^*@52;<
znJ&U<VqWNEEjRHCDf!l<57Rl*nQU0gOq>M)h7NyJ=fVZ>&6H|83n~nw<LNTxX$OO!
zDmu|^kA<Ox*qKz{r=`XscIeqWb13w{8j37T`CZliherFj8GAAkiK|sz#%aE)A;xXn
z$_^x1dwYXR8fb<vK=FV}y2^`=;vI>E8Kdc09%6ttG{CbFH&FI$&WE(Zj>`;eUt~}=
zv~7QHEPeA$(qhvvSHMEwu3GMn6Y8)7*dP`%w}USk{$@=1xFsc~$MCtV9P32Hs{R(@
zmhi6eHl12V(Lr1!Wl>e=A<oi`YY?nQ<n;jRs`tQ}l&1&i6Y{M35|OZVzR&MPD!ZdK
z-^W{0^nW<4RfsCFu}$!?w(-8mHZU9L&tHEa{L21XYvy&HZeD>V(4PTEq8nrY=|)Ai
zW~4Y5;TdM|>V6{$Z=V0DlJqO-O7L(Qc~;k?Gf}Mnjgxu~R~Z)xT-4%@X&vVd^Xag9
zSlv!(>AQ8pe0d}7G_m*UUQKs8zW&q5&exOb$<fg@=_y9t0h2*EQe4tGMIG+@p)P+9
z#&^bV6NEVHO0)Os7T&O>f|`-TP^*mgW$}=1@a9nF<S_0kkx-*EYm|xJJj>lY<06wA
zdd9`A_w-n>aNE{%e!ODqTcqA<8D|-WQ>61I&fsLq$DDkaRg$IJoN<HTBU7G&jFRIm
z87$xO6wzZ*e#K=K-Ic{otp3`;01$uHyZF++I6jIhTsM0%fJ`x*)REj{I5e)L@J&8`
z(us$Guc<oJSBF$}gx0MEEpM|IN3fy^)-|=j0WAb%i84`kt19#6y%fCb$D%C}p_vnG
zB(Rr)yQI*Y{mD^8S7Xpk4gAwaDV&{O-v=#YLJN(V&;pd^h-)Dvu5=`B=xcv(+tzXN
zbz9Q4V&UwNK;O0h^`^*HR?($=c-Jm^55tsUBMJMuug#mccu86?>Jlp}ml)7EkpfWy
zNF)%I1c83Bau@U?s{e&OIr9q^^WpxEK>=rO;Zi%J?MVGGCOwoUuhT4oyB$RS#3jrs
z7Zna!r9c;=x)4d(enO%Pab163rr3DdDfze=-2{f>(o=v5q<P+r=?v=NBr^i~I8Kz0
zc-lkU>w>qLEMLO1=ffbO3_)$6b%iT?KTc+jopx@KfB#TTtZciLTV(o0<NLy8_7FZ%
z&Vg`lF<<?=SXZv@fl2AyK@^ePZpX5_IJviNTNtMMyp|o!xpP|6#4UfKF_u&<!Z6cx
z)LN?hD)UJ8Dj{(}ZE!>2&ny`w`69(`CG9+!E}@DHpE#k9$gob`GcqCT_c@2^{XV8j
zWl#~=ZXcFtd}odo#eRk0yNp;$u^;DWJqgd;HJMbz<smEqSd#(2U4FLb9W2k)%(c$N
z3jfuRF%9emu&WTA0IPouw;cejm3GQ}HU&>|39$=!l(bhB?vwPYzY+=mxy5O2!Dkjn
zEeW_ZdfK#aupD~Sc&XyC2G!TF?l)^W-D+f$(3`0q@ajI!9!%>AYy`PGXrD*FvgdB<
zLv+^++`{+RH1z#kgn_W1tI&|*(^(PRzzD-s*w+QILelMrf>M9OWJD__7lc{t`NSF$
zT47Tf1TZ_TXsGqQ1=p^S`IWI<`ULg1*?=#N_j}TuJ2bu~dnqbnsHjE#q|@cdH^)!)
ze@2h6PZ?8hM3+6@k#Dwc=g{Md<QZ2(hMayr(_k6sM~|?R8dDES*GAqUNj-`y@}ary
zWR$F^|Cc*`+hKpFCbPOnQK*vKJx2m^wx5GMi1R3|G0a>(VCr}VOQGnh=>u2<Uog&5
ztDSnpzIL|gTVwnmbGSR^6Dc+c(G_P^LCejHWf3XwL@@%zw7^^E&KFO{x{R@$xgHak
zoVOkmww|?~WZ*Vu{q}g(URc?FsrHG@r7cbfERQ~jx5|GB3G8$Au1dQSr!sg_78}cV
z*<SA9w1+*=9oQB=+}%pCjhtwf$F;3yx`kZP%(skx`gmhU)i+Blx=tSETJ)jZXtpoZ
z??X#TuUvL{6wI9Vt$OPS^+mnijh*(FEZEQ#^h$MI&{m&*-(VR54OXuV6Vsq~LiIp!
zrdMJm8oqzV@ewQR+h}*M#JO^OQ!oShfE5l*YIoGY*hSKHGfBtVP{<YaVqy=cIslpW
ztF%Vm$a>vN;6SbLuJEG5p0!VJjBuX<`byC>pK@L(_eBS95yq|RX}4$<{#D#q@|}@4
zV=kB0cBSCX^ty%|8gsm~jgdtxeeYokzX)%#gGzru5iCpRi^G9PFU!1}TSHEX*eg5J
zgzHD0{(g3W{-51B&&VO>#=ApUcq}!9CpY8?bLO2!9%0q`^9f$ocF*<ZZjO9rF;;z{
z^>s(CCcM^Nb#yg$#gP;50L{*!S{xnGY!e4>V2u2UGK==)<NM1HZHRUA5whiQYcxFC
zi;#bHOl>9D)FqUMxQYgjBS225jRKoRg%S~0N=Eb=0!9nUM30uA%?I@Re-D>GanqBv
z^t8K|E;n~Ch|(P~V|7!qW@gRoUmi17FQuE@?P_JHtkBiSuVop=VqN_Ma9w@-=jYEp
z{I@uVc<`Ko{_qAX2Fb?szsL%F7^d)wo?3tM6_+`oR;HUR)|ay|tzqAwmH}c65M`h(
zbHgU`dIo);>^h*k)L)zR63&vSg=0GERo?8@xLg%?tNUj%(17J<J6wxxC|0g>8mrxE
zPt%Dbm05C_0_&@$pDib9!4`QqJ=-ep@T#Se-hoyN>fjmqLMf6#5bl~EU9>d|94CKY
zTg7ybMI1j{ker_#-MX2X10~x$=+@5YhEF$Rbmm)I&aXU&$P<Yc%^8*)rp0^7nIF%Q
zpe6Mx{uX|rYSy%%D$T`zROSkIfXoye)b3bita#C_j`aWi&;Ocf8#2mrf^|Y>a*lOE
z-g=tVhKJisYnpHHlh$VXMvMg5vz~w0h1jUv3K?j3P_KmuYDF#uh)D~*;?O|_P2b_a
zqv9rTd)PLm8H#)v+7EL9<uXWu7ceG8e37rc7JlnUE5OzT%lKk=qZ9=Gus<5?^($pj
ziG?iu`nyT-3(cIq39Xk@ox+qL4TH-uX#L1=n!Pw5eKf4DXlf5;R|DZ3=ca!b>K1z7
zZV1(d@^u$0+T4Es!>CSJoux{P3u#zWK8fCALMW2Nb&4ws#a-g4FDt>CZ>&`SP1n*2
zn8z{|8>3CV00obyM^cP3`ZC0L)z;Tl=sJEE|0UNx@{^zNa`|8FyH<^OODYzBf1#Au
z#nG`rKxttY=vY~52XaHOr5%6RvJbY@fhA>J-IE+(=Yd8Q-ol8MRph(4$!Z%_9E6`r
ziewPPKrWqh4r!8VSVhgAZk*jcN1L?zs_S3t>47k6_H<wF?iqnFuV9x}I)-%83JPpV
zM*kU{m$@s-=jgu4x8x5j_Ke!iSrxA9f#=b)Oe@r4B8G_=CSAda+6{kM0`qbXBX^tR
zttsKGl9tPR*)?RvEbC5a<(t(g9Kh?2Gw--}XTLHzB;>^D<d;*(R6ndZ<d>sMt;rfg
zsRH4RK|d^U<TtHL^Zsj$TDhw!WU61@83C7IRZ)?X42+{qvm<3&bInIfLEec$MG=US
zPen!f6&@cf@F@|qtAu|y!db%M=#XW67EL<BfV6rs8_*LI@l%pD8@&040TK>iha40z
z-2n<M4CJMNiGFod+j1qPQcWcyqodSFT~N8SQk2OEC{>(RRfA0hm%qHGe`xY7-KCnL
zEWa=1A<mk(yFqFQD8zfCX?`<`B>vG-4<l7<qq`KET!<1DO}&3_Of3vaifB?AcH~Jv
zYUC*}&9$~B>Jim6pWxR={)=MJA`DA}r*F1sF=^d^^~l;wD}Eb8q&i!_ELCP|+5;Jl
zH|r`X)BUgm-l}gK1<6~`?COi(Sn8pqG&`VLjqDxU$pmR1?=k5gZwjmq{>8UtA8SpT
zQ%sw7Q~jVY2w;EIpaX;91z}AZZw3>6>C|4@sO4PK<i4xzaPme2Cv)7f<D$F5VP;7H
zI`;yHdjJ5t=SeADU#wM0%;*a~5b>v{piFbLR?3K`q<56l83F0+MPA&AuZ%a~SnEP5
z3`V{L4E4dpJ$l4ZQ!t3?DkbZSEd{w?{Nokzrd@EQ<syHciu;wwe_u@uyAC|0QwL+S
zhH-;nl4PnfaOUA7<EUH?7|Dq=APFC&*<dAnkf$s>8OjS}($y%_bu;XG%8+uRE9u88
zCB9^o7w`EvkV|E00^K5`lz9J<{pC{Zgb$+-fBlL6dByJ{gV{LJkV2@IxUJsjXA;WA
zuA+q&>92p=$Fx1>k6+RKdRaEupCLaNRiJw^6qD@e36i0&QS>6tW4Lr6j$8nR5K@FL
zbx^7{CPf_B7EkGrlU{ZISMf+*!I3-ZE}jNtX%S2No(W#V4;9iSAC$D_Jmx>9DxU#;
zP!F&5aH>u+1d594X-Wr4h9KD|lKi*Z|5h=o=c<3kJ5o~rYSL17e%ug`i6BMfb_k>D
zY1U0%H%=vxiKyl8k0Fez=aE$L;&IXjk?afmtHt;cG7W-V<mo!i()!sF6>=M?f?Us!
zB>z>&SCX6JUP1CR!aeDHne2|fQxjfHEozo!H4Bj657!rS;C5t7w~s?wp2m#dj%bC5
z=q-PD!`OEM?1sH*wP~Y@3ix0ef{R*&pk<u#__H=EQmwBK)nsrbQy*S!#e@(Y{E?1G
zm6BRHxVu4-rvpkk>8HN1KyAlL^7Nb$xE&hFsc$q80^Z`Bbs2y181?R{q(PZrW<5Vm
zr?RI<nc0?wKAQoB&Z^nNd6^h)=_MAA7}S5JL_FnfD|$^I2uOOp2Miv0JdTD6Mu6!}
zAuuZC@jssacP)%i@(|~wpC6C3WwHvMT5ZDq)Ihu<<xMjT*wet4)IBDJupxed4G5&S
z6JqRWIaRvEM02IiWQ%8FK@$7bChhvVaChLt@zLU_r&$4bEhk40$7q3`y3TJ0F@k@8
z+@DmKP;iWnhZ+dTh>C@yiiJ+aVnW5jQN=>1Vlkp(>1f4LXT@^Vilw6!OPv+V2`iS4
zDwaAG%TX04jw()cDo#dKoI0vF)u}igRdMF1;!LOFY*fX$ql$B#it|wwZyZ&;(W!Vd
zs^YDqinlryZ%0&|I99zAUDZ1ot$Kf_jw()dDo!U<EF4uVbSf63Dwd8amO2$9wcF7P
z>YU)H4|H^aIuAJN03C~xt|W~XB*${3D@Lc|_zO4ug$92yhQD;fUuy72%Fmfw{+wy@
z=WHy0&fN0nOp`xnWBGIDmOp2j{5c!TpL4hTIoIUR`B?tEal?P3!GAN3zi@xUUuf`0
z^8c+H|F;_cZ^!um<c9x~2LC4+Ul{`!K-)CK;Q&n1vFl4F1_C@GVncvGWic(<M7Ljp
zgAV_nil7#7VGX!Y0T)KVg*D(p1zczWm(~oID#N9L;nJGnQf0U_GF(~%E>*y#0r12c
z@I(bXF#w)g1D>jYrv|_?Yruaq74Xaecy0}Nt^%GL0N+>xzEJ_+7y#c|1HM%O-)aF*
ztPA#uTCh(H1^d(*@KgmnH3BZI0T(LZ!T`9m23)Fu+L~hR8&ua|@C??DLG=p;w_yEw
z)vs&#xYplR{j{ehw1q9&LP1;T(U!JoO9f5)P0wsCJyR?_(_4CGYw3TPV(FRQ(lc92
z&lF3~^p>96T6(Tndak$hjV;<61?`OqZDEVHP|&pbX6ZR)x2g4+KiS&&lVT(5+O?r~
zI$?j%SibH&dif3nzJcTh=9&3&l5o3Gs{Q@bYMYg-P4Sq+j?xx;wLD(Y-96$-|Kzm&
zvq8NO{LvuDh!MqzC`Nxmamx`ckCC#isDIcVQH-33@0zSmDNlbJ%oxt*bWmX|W6R>?
zecp~{4FUXY;_LaO9rn%!c^NT>xC2?~w$QFl#96{Jk>)0bT3(r{>#k|lLrfWRdW`zp
z`{3~J1PLU>k-W^>;B~WIWyN+2>(QLdQAK^spa_T~c+uzc?W=$Kd5;~xhAQeqUrhE>
z1TXq@tzWvwaJAp6-4EqxhzZ2UB)S%nV|AOh{EXuZ6;|ER5{`!H0DO$%IMPWNz6MFl
zzgKLjs;fmB1V598>(sq$@^WNzT-O|DwU6^nT&?OdPV*I<(CS8eDDGnbLqQaUoh0jH
zO1JLB%?3%%hrEBZ+{K5r_M&ko{^qO79WBva(LQNYd!YKt>Pk?fWnP!0n(aG3Mr!xH
z+a}21y4w>8Tu3YVb%h}eRc<~EVk<+OV=li6&&NFZ7Ck?6as&t-^wo>V9LQ^KFms}=
zTzBR{T)IWcf%V|-B^S!l&QlM*!&@p`c$+RZapFI?>SBM=4*9r})#S0~MuR*~Pam=D
z?J??3c@V~RPaKE86DAJN1NODjaI&z!p~F+FXWAOX%#ap_G|nan;CSerzhMpm0<BpV
z=Q6n&X?@U>qsany<ajML1014|g353~{Q#|v33<S_;hp|LL}o1qj&8aHHG604Sa5#H
zz@}$B^Amr(M2v^0fd`Lo`h+!^U3^EfbS#dAQDI}8wxLZsj)pVZMhKctMjIhmfz#n4
z##7UVB&C6)k8hBMi<)!LM%BG^xYsYr`-%P6(v$;*c-<nqAK8EDP7(SNepwoDpdhbX
zY**~z`W+rv5`V$)u5VQLw>2F65|vFJr8(RSV)B33-zdWC6WC>2&ptfuvp9D%=LO=s
zUSZv9-N7SvMg<J;+QxO=+|z-By6&;u8+Jhsq@*XX(1t78XGLvkQSjKVLr7QJxp>>{
zT(sl;BGFy1pzhD~{%H1AJrH@qH0xfDY>j<KN4CMUuhYhre0n-<JjuM=<L?#wo)PkH
zH~oLoQ;RO5TmC+JEXuF+IA?L}7ir;k(9eM8ZeB(lIy^ovDsGn}a6Thx;n|*6x9kWV
zK6i_ntL8;fk#|0TpW0_-&OyTIE>U!TfIX|CR+Iw=(yim?x7|66>})VAvWJbLI|t7-
zw}_9KVckSb-^M(~*|T|&1A{&-gd70W9gtBwuJId9FZj<YcwNftoChyA|1fLX!S|L-
z=lnSI9EI=HgvYgS&bA+N3d3*a!ueV)OPjy_{6>grjS$y^Ok|jKi(+;0#Dw)$?aKcH
H;E~j(^Wt<k

diff --git a/master/releases.html b/master/releases.html
index 4acca8f792..6cc945943f 100644
--- a/master/releases.html
+++ b/master/releases.html
@@ -3,9 +3,8 @@
 <html>
   <head>
     <meta charset="utf-8" />
-    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="generator" content="Docutils 0.17.1: http://docutils.sourceforge.net/" />
-
-    <title>Releases &#8212; AutoSklearn 0.14.2 documentation</title>
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>Releases &#8212; AutoSklearn 0.14.4 documentation</title>
     <link rel="stylesheet" type="text/css" href="_static/pygments.css" />
     <link rel="stylesheet" type="text/css" href="_static/bootstrap-sphinx.css" />
     <link rel="stylesheet" type="text/css" href="_static/sg_gallery.css" />
@@ -55,7 +54,7 @@
         </button>
         <a class="navbar-brand" href="index.html">
           auto-sklearn</a>
-        <span class="navbar-text navbar-version pull-left"><b>0.14.2</b></span>
+        <span class="navbar-text navbar-version pull-left"><b>0.14.4</b></span>
       </div>
 
         <div class="collapse navbar-collapse nav-collapse">
@@ -108,6 +107,14 @@
       <div class="col-md-3">
         <div id="sidebar" class="bs-sidenav" role="complementary"><ul>
 <li><a class="reference internal" href="#">Releases</a><ul>
+<li><a class="reference internal" href="#version-0-14-4">Version 0.14.4</a><ul>
+<li><a class="reference internal" href="#contributors-v0-14-4">Contributors v0.14.4</a></li>
+</ul>
+</li>
+<li><a class="reference internal" href="#version-0-14-3">Version 0.14.3</a><ul>
+<li><a class="reference internal" href="#contributors-v0-14-3">Contributors v0.14.3</a></li>
+</ul>
+</li>
 <li><a class="reference internal" href="#version-0-14-2">Version 0.14.2</a><ul>
 <li><a class="reference internal" href="#contributors-v0-14-2">Contributors v0.14.2</a></li>
 </ul>
@@ -240,21 +247,69 @@
       </div>
     <div class="body col-md-9 content" role="main">
       
-  <section id="releases">
+  <div class="section" id="releases">
 <h1>Releases<a class="headerlink" href="#releases" title="Permalink to this headline">¶</a></h1>
-<section id="version-0-14-2">
+<div class="section" id="version-0-14-4">
+<h2>Version 0.14.4<a class="headerlink" href="#version-0-14-4" title="Permalink to this headline">¶</a></h2>
+<ul class="simple">
+<li><p>Fix #1356: SVR degree hyperparameter now only active with “poly” kernel.</p></li>
+<li><p>Add #1311: Black format checking (non-strict).</p></li>
+<li><p>Maint #1306: Run history is now saved every iteration</p></li>
+<li><p>Doc #1309: Updated the doc faqs to include many use cases and the manual for early introductions</p></li>
+<li><p>Doc #1322: Fix typo in contribution guide</p></li>
+<li><p>Maint #1326: Add isort checker (non-strict)</p></li>
+<li><p>Maint #1238, #1346, #1368, #1370: Update warnings in tests</p></li>
+<li><p>Maint #1325: Test workflow can now be manually triggered</p></li>
+<li><p>Maint #1332: Update docstring and typing of <code class="docutils literal notranslate"><span class="pre">include</span></code> and <code class="docutils literal notranslate"><span class="pre">exclude</span></code> params</p></li>
+<li><p>Add #1260: Support for Python 3.10</p></li>
+<li><p>Add #1318: First update to use the shared backend in a new submodule <a class="reference external" href="https://github.com/automl/automl_common">automl_common</a></p></li>
+<li><p>Fix #1339: Resolve dependancy issues with <code class="docutils literal notranslate"><span class="pre">sphinx_toolbox</span></code></p></li>
+<li><p>Fix #1335: Fix issue where some regression algorithm gave incorrect output dimensions as raised in #1297</p></li>
+<li><p>Doc #1340: Update example for predefined splits</p></li>
+<li><p>Fix #1329: Fix random state not being passed to the ConfigurationSpace</p></li>
+<li><p>Maint #1348: Stop double triggering of github workflows</p></li>
+<li><p>Doc #1349: Rename OSX to macOS in docs</p></li>
+<li><p>Add #1321: Change <code class="docutils literal notranslate"><span class="pre">show_models()</span></code> to produce actual pipeline objects and not a <code class="docutils literal notranslate"><span class="pre">str</span></code></p></li>
+<li><p>Maint #1361: Remove <code class="docutils literal notranslate"><span class="pre">flaky</span></code> dependency</p></li>
+<li><p>Maint #1366: Make <code class="docutils literal notranslate"><span class="pre">SimpleClassificationPipeline</span></code> tests more deterministic</p></li>
+<li><p>Maint #1367: Update test values for <code class="docutils literal notranslate"><span class="pre">MLPRegressor</span></code> with newer numpy</p></li>
+</ul>
+<div class="section" id="contributors-v0-14-4">
+<h3>Contributors v0.14.4<a class="headerlink" href="#contributors-v0-14-4" title="Permalink to this headline">¶</a></h3>
+<ul class="simple">
+<li><p>Eddie Bergman</p></li>
+<li><p>Matthias Feurer</p></li>
+<li><p>Katharina Eggensperger</p></li>
+<li><p>UserFindingSelf</p></li>
+<li><p>partev</p></li>
+</ul>
+</div>
+</div>
+<div class="section" id="version-0-14-3">
+<h2>Version 0.14.3<a class="headerlink" href="#version-0-14-3" title="Permalink to this headline">¶</a></h2>
+<ul class="simple">
+<li><p>HOTFIX #1356: Updates dask to <code class="docutils literal notranslate"><span class="pre">dask.distributed</span> <span class="pre">&gt;=2012.12</span></code>.</p></li>
+</ul>
+<div class="section" id="contributors-v0-14-3">
+<h3>Contributors v0.14.3<a class="headerlink" href="#contributors-v0-14-3" title="Permalink to this headline">¶</a></h3>
+<ul class="simple">
+<li><p>Eddie Bergman</p></li>
+</ul>
+</div>
+</div>
+<div class="section" id="version-0-14-2">
 <h2>Version 0.14.2<a class="headerlink" href="#version-0-14-2" title="Permalink to this headline">¶</a></h2>
 <ul class="simple">
 <li><p>FIX #1290: Fixes a bug where it was not possible to extend Auto-sklearn and run it in parallel.</p></li>
 </ul>
-<section id="contributors-v0-14-2">
+<div class="section" id="contributors-v0-14-2">
 <h3>Contributors v0.14.2<a class="headerlink" href="#contributors-v0-14-2" title="Permalink to this headline">¶</a></h3>
 <ul class="simple">
 <li><p>Matthias Feurer</p></li>
 </ul>
-</section>
-</section>
-<section id="version-0-14-1">
+</div>
+</div>
+<div class="section" id="version-0-14-1">
 <h2>Version 0.14.1<a class="headerlink" href="#version-0-14-1" title="Permalink to this headline">¶</a></h2>
 <ul class="simple">
 <li><p>FIX #1248: Allow for sparse <code class="docutils literal notranslate"><span class="pre">y_test</span></code>.</p></li>
@@ -267,16 +322,16 @@ <h2>Version 0.14.1<a class="headerlink" href="#version-0-14-1" title="Permalink
 <li><p>DOC #1271: Fix a typo in an example in the documentation.</p></li>
 <li><p>DOC #1282: Add a contribution guide.</p></li>
 </ul>
-<section id="contributors-v0-14-1">
+<div class="section" id="contributors-v0-14-1">
 <h3>Contributors v0.14.1<a class="headerlink" href="#contributors-v0-14-1" title="Permalink to this headline">¶</a></h3>
 <ul class="simple">
-<li><p>Edward Bergman</p></li>
+<li><p>Eddie Bergman</p></li>
 <li><p>Michael Becker</p></li>
 <li><p>Katharina Eggensperger</p></li>
 </ul>
-</section>
-</section>
-<section id="version-0-14-0">
+</div>
+</div>
+<div class="section" id="version-0-14-0">
 <h2>Version 0.14.0<a class="headerlink" href="#version-0-14-0" title="Permalink to this headline">¶</a></h2>
 <ul class="simple">
 <li><p>ADD #900: Make data preprocessing more configurable, for example allow to completely disable it.</p></li>
@@ -306,7 +361,7 @@ <h2>Version 0.14.0<a class="headerlink" href="#version-0-14-0" title="Permalink
 github actions.</p></li>
 <li><p>MAINT #1239: Increase the timeout for github actions to 60 minutes.</p></li>
 </ul>
-<section id="contributors-v0-14-0">
+<div class="section" id="contributors-v0-14-0">
 <h3>Contributors v0.14.0<a class="headerlink" href="#contributors-v0-14-0" title="Permalink to this headline">¶</a></h3>
 <ul class="simple">
 <li><p>Pieter Gijsbers</p></li>
@@ -320,9 +375,9 @@ <h3>Contributors v0.14.0<a class="headerlink" href="#contributors-v0-14-0" title
 <li><p>Oli</p></li>
 <li><p>Matthias Feurer</p></li>
 </ul>
-</section>
-</section>
-<section id="version-0-13-0">
+</div>
+</div>
+<div class="section" id="version-0-13-0">
 <h2>Version 0.13.0<a class="headerlink" href="#version-0-13-0" title="Permalink to this headline">¶</a></h2>
 <ul class="simple">
 <li><p>ADD #1100: Provide access to the callbacks of SMAC.</p></li>
@@ -346,7 +401,7 @@ <h2>Version 0.13.0<a class="headerlink" href="#version-0-13-0" title="Permalink
 <li><p>DOC #1158: New example on how to configure the logger for Auto-sklearn.</p></li>
 <li><p>DOC #1165: Improve the readme page.</p></li>
 </ul>
-<section id="contributors-v0-13-0">
+<div class="section" id="contributors-v0-13-0">
 <h3>Contributors v0.13.0<a class="headerlink" href="#contributors-v0-13-0" title="Permalink to this headline">¶</a></h3>
 <ul class="simple">
 <li><p>Matthias Feurer</p></li>
@@ -354,21 +409,21 @@ <h3>Contributors v0.13.0<a class="headerlink" href="#contributors-v0-13-0" title
 <li><p>bitsbuffer</p></li>
 <li><p>Francisco Rivera Valverde</p></li>
 </ul>
-</section>
-</section>
-<section id="version-0-12-8">
+</div>
+</div>
+<div class="section" id="version-0-12-8">
 <h2>Version 0.12.8<a class="headerlink" href="#version-0-12-8" title="Permalink to this headline">¶</a></h2>
 <ul class="simple">
 <li><p>MAINT #1183: Introduce an upper bound on the dask version to retain compatibility with SMAC3.</p></li>
 </ul>
-<section id="contributors-v0-12-8">
+<div class="section" id="contributors-v0-12-8">
 <h3>Contributors v0.12.8<a class="headerlink" href="#contributors-v0-12-8" title="Permalink to this headline">¶</a></h3>
 <ul class="simple">
 <li><p>Eddie Bergman</p></li>
 </ul>
-</section>
-</section>
-<section id="version-0-12-7">
+</div>
+</div>
+<div class="section" id="version-0-12-7">
 <h2>Version 0.12.7<a class="headerlink" href="#version-0-12-7" title="Permalink to this headline">¶</a></h2>
 <ul class="simple">
 <li><p>ADD #1178: Reduce precision if dataset is too large for given memory limit.</p></li>
@@ -382,7 +437,7 @@ <h2>Version 0.12.7<a class="headerlink" href="#version-0-12-7" title="Permalink
 <li><p>MAINT #1168: Upper-bound scipy to <cite>1.6.3</cite> as <cite>1.7.0</cite> is incompatible with <cite>SMAC</cite>.</p></li>
 <li><p>MAINT #1173: Update the license files to be recognized by github.</p></li>
 </ul>
-<section id="contributors-v0-12-7">
+<div class="section" id="contributors-v0-12-7">
 <h3>Contributors v0.12.7<a class="headerlink" href="#contributors-v0-12-7" title="Permalink to this headline">¶</a></h3>
 <ul class="simple">
 <li><p>Francisco Rivera Valverde</p></li>
@@ -394,9 +449,9 @@ <h3>Contributors v0.12.7<a class="headerlink" href="#contributors-v0-12-7" title
 <li><p>Eddie Bergman</p></li>
 <li><p>olehb007</p></li>
 </ul>
-</section>
-</section>
-<section id="version-0-12-6">
+</div>
+</div>
+<div class="section" id="version-0-12-6">
 <h2>Version 0.12.6<a class="headerlink" href="#version-0-12-6" title="Permalink to this headline">¶</a></h2>
 <ul class="simple">
 <li><p>ADD #886: Provide new function which allows fitting only a single configuration.</p></li>
@@ -412,7 +467,7 @@ <h2>Version 0.12.6<a class="headerlink" href="#version-0-12-6" title="Permalink
 <li><p>FIX #1115/#1124: Fix a bug which would prevent Auto-sklearn from computing meta-features in the
 multiprocessing case.</p></li>
 </ul>
-<section id="contributors-v0-12-6">
+<div class="section" id="contributors-v0-12-6">
 <h3>Contributors v0.12.6<a class="headerlink" href="#contributors-v0-12-6" title="Permalink to this headline">¶</a></h3>
 <ul class="simple">
 <li><p>Francisco Rivera Valverde</p></li>
@@ -421,21 +476,21 @@ <h3>Contributors v0.12.6<a class="headerlink" href="#contributors-v0-12-6" title
 <li><p>Matthias Feurer</p></li>
 <li><p>Rohit Agarwal</p></li>
 </ul>
-</section>
-</section>
-<section id="version-0-12-5">
+</div>
+</div>
+<div class="section" id="version-0-12-5">
 <h2>Version 0.12.5<a class="headerlink" href="#version-0-12-5" title="Permalink to this headline">¶</a></h2>
 <ul class="simple">
 <li><p>MAINT: Remove <code class="docutils literal notranslate"><span class="pre">Cython</span></code> and <code class="docutils literal notranslate"><span class="pre">numpy</span></code> as installation requirements.</p></li>
 </ul>
-<section id="contributors-v0-12-5">
+<div class="section" id="contributors-v0-12-5">
 <h3>Contributors v0.12.5<a class="headerlink" href="#contributors-v0-12-5" title="Permalink to this headline">¶</a></h3>
 <ul class="simple">
 <li><p>Matthias Feurer</p></li>
 </ul>
-</section>
-</section>
-<section id="version-0-12-4">
+</div>
+</div>
+<div class="section" id="version-0-12-4">
 <h2>Version 0.12.4<a class="headerlink" href="#version-0-12-4" title="Permalink to this headline">¶</a></h2>
 <ul class="simple">
 <li><p>ADD #660: Enable scikit-learn’s power transformation for input features.</p></li>
@@ -446,7 +501,7 @@ <h2>Version 0.12.4<a class="headerlink" href="#version-0-12-4" title="Permalink
 <li><p>FIX #1072: Fixes a bug where the <code class="docutils literal notranslate"><span class="pre">AutoSklearn2Classifier</span></code> could not be created due to trying to
 cache to the wrong directory.</p></li>
 </ul>
-<section id="contributors-v0-12-4">
+<div class="section" id="contributors-v0-12-4">
 <h3>Contributors v0.12.4<a class="headerlink" href="#contributors-v0-12-4" title="Permalink to this headline">¶</a></h3>
 <ul class="simple">
 <li><p>Matthias Feurer</p></li>
@@ -454,9 +509,9 @@ <h3>Contributors v0.12.4<a class="headerlink" href="#contributors-v0-12-4" title
 <li><p>Maximilian Greil</p></li>
 <li><p>Pepe Berba</p></li>
 </ul>
-</section>
-</section>
-<section id="version-0-12-3">
+</div>
+</div>
+<div class="section" id="version-0-12-3">
 <h2>Version 0.12.3<a class="headerlink" href="#version-0-12-3" title="Permalink to this headline">¶</a></h2>
 <ul class="simple">
 <li><p>FIX #1061: Fixes a bug where the model could not be printed in a jupyter notebook.</p></li>
@@ -469,16 +524,16 @@ <h2>Version 0.12.3<a class="headerlink" href="#version-0-12-3" title="Permalink
 <li><p>MAINT #1076: Uses the correct multiprocessing context for computing meta-features</p></li>
 <li><p>MAINT: Cleanup readme and main directory</p></li>
 </ul>
-<section id="contributors-v0-12-3">
+<div class="section" id="contributors-v0-12-3">
 <h3>Contributors v0.12.3<a class="headerlink" href="#contributors-v0-12-3" title="Permalink to this headline">¶</a></h3>
 <ul class="simple">
 <li><p>Matthias Feurer</p></li>
 <li><p>ROHIT AGARWAL</p></li>
 <li><p>Francisco Rivera</p></li>
 </ul>
-</section>
-</section>
-<section id="version-0-12-2">
+</div>
+</div>
+<div class="section" id="version-0-12-2">
 <h2>Version 0.12.2<a class="headerlink" href="#version-0-12-2" title="Permalink to this headline">¶</a></h2>
 <ul class="simple">
 <li><p>ADD #1045: New example demonstrating how to log multiple metrics during a run of Auto-sklearn.</p></li>
@@ -492,7 +547,7 @@ <h2>Version 0.12.2<a class="headerlink" href="#version-0-12-2" title="Permalink
 <li><p>MAINT #1053: Remove the restriction to guard single-core Auto-sklearn by
 <code class="docutils literal notranslate"><span class="pre">__main__</span> <span class="pre">==</span> <span class="pre">&quot;__name__&quot;</span></code> again.</p></li>
 </ul>
-<section id="contributors-v0-12-2">
+<div class="section" id="contributors-v0-12-2">
 <h3>Contributors v0.12.2<a class="headerlink" href="#contributors-v0-12-2" title="Permalink to this headline">¶</a></h3>
 <ul class="simple">
 <li><p>Matthias Feurer</p></li>
@@ -500,9 +555,9 @@ <h3>Contributors v0.12.2<a class="headerlink" href="#contributors-v0-12-2" title
 <li><p>Francisco Rivera</p></li>
 <li><p>Katharina Eggensperger</p></li>
 </ul>
-</section>
-</section>
-<section id="version-0-12-1">
+</div>
+</div>
+<div class="section" id="version-0-12-1">
 <h2>Version 0.12.1<a class="headerlink" href="#version-0-12-1" title="Permalink to this headline">¶</a></h2>
 <ul class="simple">
 <li><p>ADD: A new heuristic which gives a warning and subsamples the data if it is too large for the
@@ -514,16 +569,16 @@ <h2>Version 0.12.1<a class="headerlink" href="#version-0-12-1" title="Permalink
 used for Auto-sklearn 1.0.</p></li>
 <li><p>MAINT #1040: Minor speed improvements in the ensemble selection algorithm.</p></li>
 </ul>
-<section id="contributors-v0-12-1">
+<div class="section" id="contributors-v0-12-1">
 <h3>Contributors v0.12.1<a class="headerlink" href="#contributors-v0-12-1" title="Permalink to this headline">¶</a></h3>
 <ul class="simple">
 <li><p>Matthias Feurer</p></li>
 <li><p>Katharina Eggensperger</p></li>
 <li><p>Francisco Rivera</p></li>
 </ul>
-</section>
-</section>
-<section id="version-0-12-0">
+</div>
+</div>
+<div class="section" id="version-0-12-0">
 <h2>Version 0.12.0<a class="headerlink" href="#version-0-12-0" title="Permalink to this headline">¶</a></h2>
 <ul class="simple">
 <li><p>BREAKING: Auto-sklearn must now be guarded by <code class="docutils literal notranslate"><span class="pre">__name__</span> <span class="pre">==</span> <span class="pre">&quot;__main__&quot;</span></code> due to the use of the
@@ -538,16 +593,16 @@ <h2>Version 0.12.0<a class="headerlink" href="#version-0-12-0" title="Permalink
 <li><p>FIX #1007, #1012 and #1014: Log multiprocessing output via a new log server. Remove several
 potential deadlocks related to the joint use of multi-processing, multi-threading and logging.</p></li>
 </ul>
-<section id="contributors-v0-12-0">
+<div class="section" id="contributors-v0-12-0">
 <h3>Contributors v0.12.0<a class="headerlink" href="#contributors-v0-12-0" title="Permalink to this headline">¶</a></h3>
 <ul class="simple">
 <li><p>Matthias Feurer</p></li>
 <li><p>ROHIT AGARWAL</p></li>
 <li><p>Francisco Rivera</p></li>
 </ul>
-</section>
-</section>
-<section id="version-0-11-1">
+</div>
+</div>
+<div class="section" id="version-0-11-1">
 <h2>Version 0.11.1<a class="headerlink" href="#version-0-11-1" title="Permalink to this headline">¶</a></h2>
 <ul class="simple">
 <li><p>FIX #989: Fixes a bug where <cite>y</cite> was not passed to all data preprocessors which made 3rd party
@@ -555,14 +610,14 @@ <h2>Version 0.11.1<a class="headerlink" href="#version-0-11-1" title="Permalink
 <li><p>FIX #1001: Fixes a bug which could make Auto-sklearn fail at random.</p></li>
 <li><p>MAINT #1000: Introduce a minimal version for <code class="docutils literal notranslate"><span class="pre">dask.distributed</span></code>.</p></li>
 </ul>
-<section id="contributors-v0-11-1">
+<div class="section" id="contributors-v0-11-1">
 <h3>Contributors v0.11.1<a class="headerlink" href="#contributors-v0-11-1" title="Permalink to this headline">¶</a></h3>
 <ul class="simple">
 <li><p>Matthias Feurer</p></li>
 </ul>
-</section>
-</section>
-<section id="version-0-11-0">
+</div>
+</div>
+<div class="section" id="version-0-11-0">
 <h2>Version 0.11.0<a class="headerlink" href="#version-0-11-0" title="Permalink to this headline">¶</a></h2>
 <ul class="simple">
 <li><p>ADD #992: Move ensemble building from being a separate process to a job submitted to the dask
@@ -577,7 +632,7 @@ <h2>Version 0.11.0<a class="headerlink" href="#version-0-11-0" title="Permalink
 <li><p>MAINT #997: Improve memory handling in the ensemble building process. This allows building
 ensembles for larger datasets.</p></li>
 </ul>
-<section id="contributors-v0-11-0">
+<div class="section" id="contributors-v0-11-0">
 <h3>Contributors v0.11.0<a class="headerlink" href="#contributors-v0-11-0" title="Permalink to this headline">¶</a></h3>
 <ul class="simple">
 <li><p>Matthias Feurer</p></li>
@@ -585,9 +640,9 @@ <h3>Contributors v0.11.0<a class="headerlink" href="#contributors-v0-11-0" title
 <li><p>Karl Leswing</p></li>
 <li><p>ROHIT AGARWAL</p></li>
 </ul>
-</section>
-</section>
-<section id="version-0-10-0">
+</div>
+</div>
+<div class="section" id="version-0-10-0">
 <h2>Version 0.10.0<a class="headerlink" href="#version-0-10-0" title="Permalink to this headline">¶</a></h2>
 <ul class="simple">
 <li><p>ADD #325: Allow to separately optimize metrics for metadata generation.</p></li>
@@ -615,7 +670,7 @@ <h2>Version 0.10.0<a class="headerlink" href="#version-0-10-0" title="Permalink
 <li><p>MAINT #949: Replace <cite>pip</cite> by <cite>pip3</cite> in the installation guidelines.</p></li>
 <li><p>MAINT #280, #535, #956: Update meta-data and include regression meta-data again.</p></li>
 </ul>
-<section id="contributors-v0-10-0">
+<div class="section" id="contributors-v0-10-0">
 <h3>Contributors v0.10.0<a class="headerlink" href="#contributors-v0-10-0" title="Permalink to this headline">¶</a></h3>
 <ul class="simple">
 <li><p>Francisco Rivera</p></li>
@@ -624,9 +679,9 @@ <h3>Contributors v0.10.0<a class="headerlink" href="#contributors-v0-10-0" title
 <li><p>Chu-Cheng Fu</p></li>
 <li><p>Francois Berenger</p></li>
 </ul>
-</section>
-</section>
-<section id="version-0-9-0">
+</div>
+</div>
+<div class="section" id="version-0-9-0">
 <h2>Version 0.9.0<a class="headerlink" href="#version-0-9-0" title="Permalink to this headline">¶</a></h2>
 <ul class="simple">
 <li><p>ADD #157,#889: Improve handling of pandas dataframes, including the possibility to use pandas’
@@ -650,7 +705,7 @@ <h2>Version 0.9.0<a class="headerlink" href="#version-0-9-0" title="Permalink to
 <li><p>FIX #916: Fixes a bug where the data preprocessing configurations were ignored.</p></li>
 <li><p>FIX #925: make internal data preprocessing objects clonable.</p></li>
 </ul>
-<section id="contributors-v0-9-0">
+<div class="section" id="contributors-v0-9-0">
 <h3>Contributors v0.9.0<a class="headerlink" href="#contributors-v0-9-0" title="Permalink to this headline">¶</a></h3>
 <ul class="simple">
 <li><p>Francisco Rivera</p></li>
@@ -658,23 +713,23 @@ <h3>Contributors v0.9.0<a class="headerlink" href="#contributors-v0-9-0" title="
 <li><p>felixleungsc</p></li>
 <li><p>Vladislav Skripniuk</p></li>
 </ul>
-</section>
-</section>
-<section id="version-0-8">
+</div>
+</div>
+<div class="section" id="version-0-8">
 <h2>Version 0.8<a class="headerlink" href="#version-0-8" title="Permalink to this headline">¶</a></h2>
 <ul class="simple">
 <li><p>ADD #803: multi-output regression</p></li>
 <li><p>ADD #893: new Auto-sklearn mode Auto-sklearn 2.0</p></li>
 </ul>
-<section id="contributors-v0-8-0">
+<div class="section" id="contributors-v0-8-0">
 <h3>Contributors v0.8.0<a class="headerlink" href="#contributors-v0-8-0" title="Permalink to this headline">¶</a></h3>
 <ul class="simple">
 <li><p>Chu-Cheng Fu</p></li>
 <li><p>Matthias Feurer</p></li>
 </ul>
-</section>
-</section>
-<section id="version-0-7-1">
+</div>
+</div>
+<div class="section" id="version-0-7-1">
 <h2>Version 0.7.1<a class="headerlink" href="#version-0-7-1" title="Permalink to this headline">¶</a></h2>
 <ul class="simple">
 <li><p>ADD #764: support for automatic per_run_time_limit selection</p></li>
@@ -694,16 +749,16 @@ <h2>Version 0.7.1<a class="headerlink" href="#version-0-7-1" title="Permalink to
 <li><p>FIX #876: allow Auto-sklearn model to be cloned</p></li>
 <li><p>FIX #879: allow 1-D binary predictions</p></li>
 </ul>
-<section id="contributors-v0-7-1">
+<div class="section" id="contributors-v0-7-1">
 <h3>Contributors v0.7.1<a class="headerlink" href="#contributors-v0-7-1" title="Permalink to this headline">¶</a></h3>
 <ul class="simple">
 <li><p>Matthias Feurer</p></li>
 <li><p>Xiaodong DENG</p></li>
 <li><p>Francisco Rivera</p></li>
 </ul>
-</section>
-</section>
-<section id="version-0-7-0">
+</div>
+</div>
+<div class="section" id="version-0-7-0">
 <h2>Version 0.7.0<a class="headerlink" href="#version-0-7-0" title="Permalink to this headline">¶</a></h2>
 <ul class="simple">
 <li><p>ADD #785: user control to reduce the hard drive memory required to store ensembles</p></li>
@@ -720,7 +775,7 @@ <h2>Version 0.7.0<a class="headerlink" href="#version-0-7-0" title="Permalink to
 <li><p>FIX #778: support for python 3.8</p></li>
 <li><p>FIX #781: support for pandas 1.x</p></li>
 </ul>
-<section id="contributors-v0-7-0">
+<div class="section" id="contributors-v0-7-0">
 <h3>Contributors v0.7.0<a class="headerlink" href="#contributors-v0-7-0" title="Permalink to this headline">¶</a></h3>
 <ul class="simple">
 <li><p>Andrew Nader</p></li>
@@ -733,9 +788,9 @@ <h3>Contributors v0.7.0<a class="headerlink" href="#contributors-v0-7-0" title="
 <li><p>Valentin Geffrier</p></li>
 <li><p>Francisco Rivera</p></li>
 </ul>
-</section>
-</section>
-<section id="version-0-6-0">
+</div>
+</div>
+<div class="section" id="version-0-6-0">
 <h2>Version 0.6.0<a class="headerlink" href="#version-0-6-0" title="Permalink to this headline">¶</a></h2>
 <ul class="simple">
 <li><p>MAINT: move from scikit-learn 0.19.X to 0.21.X</p></li>
@@ -743,16 +798,16 @@ <h2>Version 0.6.0<a class="headerlink" href="#version-0-6-0" title="Permalink to
 <li><p>FIX #680: Remove unnecessary print statement</p></li>
 <li><p>FIX #600: Remove unnecessary warning</p></li>
 </ul>
-<section id="contributors-v0-6-0">
+<div class="section" id="contributors-v0-6-0">
 <h3>Contributors v0.6.0<a class="headerlink" href="#contributors-v0-6-0" title="Permalink to this headline">¶</a></h3>
 <ul class="simple">
 <li><p>Guilherme Miotto</p></li>
 <li><p>Matthias Feurer</p></li>
 <li><p>Jin Woo Ahn</p></li>
 </ul>
-</section>
-</section>
-<section id="version-0-5-2">
+</div>
+</div>
+<div class="section" id="version-0-5-2">
 <h2>Version 0.5.2<a class="headerlink" href="#version-0-5-2" title="Permalink to this headline">¶</a></h2>
 <ul class="simple">
 <li><p>FIX #669: Correctly handle arguments to the <code class="docutils literal notranslate"><span class="pre">AutoMLRegressor</span></code></p></li>
@@ -760,7 +815,7 @@ <h2>Version 0.5.2<a class="headerlink" href="#version-0-5-2" title="Permalink to
 <li><p>ADD #676: Allow brackets [ ] inside the temporary and output directory paths.</p></li>
 <li><p>ADD #424: (Experimental) scripts to reproduce the results from the original Auto-sklearn paper.</p></li>
 </ul>
-<section id="contributors-v0-5-2">
+<div class="section" id="contributors-v0-5-2">
 <h3>Contributors v0.5.2<a class="headerlink" href="#contributors-v0-5-2" title="Permalink to this headline">¶</a></h3>
 <ul class="simple">
 <li><p>Jin Woo Ahn</p></li>
@@ -768,9 +823,9 @@ <h3>Contributors v0.5.2<a class="headerlink" href="#contributors-v0-5-2" title="
 <li><p>Matthias Feurer</p></li>
 <li><p>yazanobeidi</p></li>
 </ul>
-</section>
-</section>
-<section id="version-0-5-1">
+</div>
+</div>
+<div class="section" id="version-0-5-1">
 <h2>Version 0.5.1<a class="headerlink" href="#version-0-5-1" title="Permalink to this headline">¶</a></h2>
 <ul class="simple">
 <li><p>ADD #650: Auto-sklearn will immediately stop if prediction using scikit-learn’s dummy predictor
@@ -782,7 +837,7 @@ <h2>Version 0.5.1<a class="headerlink" href="#version-0-5-1" title="Permalink to
 <li><p>FIX #626: Fixes an issue where losses were not minimized, but maximized.</p></li>
 <li><p>MAINT #646: Do no longer restrict the numpy version to be less than 1.14.5.</p></li>
 </ul>
-<section id="contributors-v0-5-1">
+<div class="section" id="contributors-v0-5-1">
 <h3>Contributors v0.5.1<a class="headerlink" href="#contributors-v0-5-1" title="Permalink to this headline">¶</a></h3>
 <ul class="simple">
 <li><p>Jin Woo Ahn</p></li>
@@ -790,9 +845,9 @@ <h3>Contributors v0.5.1<a class="headerlink" href="#contributors-v0-5-1" title="
 <li><p>Matthias Feurer</p></li>
 <li><p>jianswang</p></li>
 </ul>
-</section>
-</section>
-<section id="version-0-5-0">
+</div>
+</div>
+<div class="section" id="version-0-5-0">
 <h2>Version 0.5.0<a class="headerlink" href="#version-0-5-0" title="Permalink to this headline">¶</a></h2>
 <ul class="simple">
 <li><p>ADD #593: Auto-sklearn supports the <code class="docutils literal notranslate"><span class="pre">n_jobs</span></code> argument for parallel
@@ -802,7 +857,7 @@ <h2>Version 0.5.0<a class="headerlink" href="#version-0-5-0" title="Permalink to
 <li><p>TEST #614: Test installation with clean Ubuntu on travis-ci.</p></li>
 <li><p>MAINT: Fixed broken link and typo in the documentation.</p></li>
 </ul>
-<section id="contributors-v0-5-0">
+<div class="section" id="contributors-v0-5-0">
 <h3>Contributors v0.5.0<a class="headerlink" href="#contributors-v0-5-0" title="Permalink to this headline">¶</a></h3>
 <ul class="simple">
 <li><p>Mohd Shahril</p></li>
@@ -811,9 +866,9 @@ <h3>Contributors v0.5.0<a class="headerlink" href="#contributors-v0-5-0" title="
 <li><p>Jirka Borovec</p></li>
 <li><p>Pradeep Reddy Raamana</p></li>
 </ul>
-</section>
-</section>
-<section id="version-0-4-2">
+</div>
+</div>
+<div class="section" id="version-0-4-2">
 <h2>Version 0.4.2<a class="headerlink" href="#version-0-4-2" title="Permalink to this headline">¶</a></h2>
 <ul class="simple">
 <li><p>Fixes #538: Remove rounding errors when giving a training set fraction for
@@ -830,7 +885,7 @@ <h2>Version 0.4.2<a class="headerlink" href="#version-0-4-2" title="Permalink to
 <li><p>MAINT: remove dependency on the six package</p></li>
 <li><p>MAINT: upgrade to XGBoost 0.80</p></li>
 </ul>
-<section id="contributors-v0-4-2">
+<div class="section" id="contributors-v0-4-2">
 <h3>Contributors v0.4.2<a class="headerlink" href="#contributors-v0-4-2" title="Permalink to this headline">¶</a></h3>
 <ul class="simple">
 <li><p>Taneli Mielikäinen</p></li>
@@ -840,9 +895,9 @@ <h3>Contributors v0.4.2<a class="headerlink" href="#contributors-v0-4-2" title="
 <li><p>Teresa Conceição</p></li>
 <li><p>Jin Woo Ahn</p></li>
 </ul>
-</section>
-</section>
-<section id="version-0-4-1">
+</div>
+</div>
+<div class="section" id="version-0-4-1">
 <h2>Version 0.4.1<a class="headerlink" href="#version-0-4-1" title="Permalink to this headline">¶</a></h2>
 <ul class="simple">
 <li><p>Added documentation on <a class="reference external" href="https://github.com/automl/auto-sklearn/pull/510">how to extend Auto-sklearn</a>
@@ -865,7 +920,7 @@ <h2>Version 0.4.1<a class="headerlink" href="#version-0-4-1" title="Permalink to
 <li><p>Travis-ci now runs flake8 to enforce pep8 style guide, and uses travis-ci instead of circle-ci
 for deployment.</p></li>
 </ul>
-<section id="contributors-v0-4-1">
+<div class="section" id="contributors-v0-4-1">
 <h3>Contributors v0.4.1<a class="headerlink" href="#contributors-v0-4-1" title="Permalink to this headline">¶</a></h3>
 <ul class="simple">
 <li><p>Matthias Feurer</p></li>
@@ -874,9 +929,9 @@ <h3>Contributors v0.4.1<a class="headerlink" href="#contributors-v0-4-1" title="
 <li><p>Katharina Eggensperger</p></li>
 <li><p>Jin Woo Ahn</p></li>
 </ul>
-</section>
-</section>
-<section id="version-0-4-0">
+</div>
+</div>
+<div class="section" id="version-0-4-0">
 <h2>Version 0.4.0<a class="headerlink" href="#version-0-4-0" title="Permalink to this headline">¶</a></h2>
 <ul class="simple">
 <li><p>Fixes <a class="reference external" href="https://github.com/automl/auto-sklearn/issues/409">#409</a>: fixes
@@ -904,7 +959,7 @@ <h2>Version 0.4.0<a class="headerlink" href="#version-0-4-0" title="Permalink to
 minimization problem.</p></li>
 <li><p>Implements <a class="reference external" href="https://github.com/automl/auto-sklearn/issues/271">#271</a>:
 XGBoost is available again, even configuring the new dropout functionality.</p></li>
-<li><p>New documentation section <a class="reference internal" href="manual.html#inspecting-the-results"><span class="std std-ref">Inspecting the results</span></a>.</p></li>
+<li><p>New documentation section <a class="reference internal" href="manual.html#inspect"><span class="std std-ref">Inspecting the results</span></a>.</p></li>
 <li><p>Fixes <a class="reference external" href="https://github.com/automl/auto-sklearn/issues/444">#444</a>:
 Auto-sklearn now only loads models for refit which are actually relevant
 for the ensemble.</p></li>
@@ -914,7 +969,7 @@ <h2>Version 0.4.0<a class="headerlink" href="#version-0-4-0" title="Permalink to
 <li><p>Safeguard Auto-sklearn against deleting directories it did not create (Issue
 <a class="reference external" href="https://github.com/automl/auto-sklearn/issues/317">#317</a>.</p></li>
 </ul>
-<section id="contributors-v0-4-0">
+<div class="section" id="contributors-v0-4-0">
 <h3>Contributors v0.4.0<a class="headerlink" href="#contributors-v0-4-0" title="Permalink to this headline">¶</a></h3>
 <ul class="simple">
 <li><p>Matthias Feurer</p></li>
@@ -928,9 +983,9 @@ <h3>Contributors v0.4.0<a class="headerlink" href="#contributors-v0-4-0" title="
 <li><p>Marius Lindauer</p></li>
 <li><p>Yassine Morakakam</p></li>
 </ul>
-</section>
-</section>
-<section id="version-0-3-0">
+</div>
+</div>
+<div class="section" id="version-0-3-0">
 <h2>Version 0.3.0<a class="headerlink" href="#version-0-3-0" title="Permalink to this headline">¶</a></h2>
 <ul class="simple">
 <li><p>Upgrade to scikit-learn 0.19.1.</p></li>
@@ -945,15 +1000,15 @@ <h2>Version 0.3.0<a class="headerlink" href="#version-0-3-0" title="Permalink to
 <li><p>Allow passing an SMBO object into the <code class="docutils literal notranslate"><span class="pre">AutoSklearnClassifier</span></code> and
 <code class="docutils literal notranslate"><span class="pre">AutoSklearnRegressor</span></code>.</p></li>
 </ul>
-<section id="contributors-v0-3-0">
+<div class="section" id="contributors-v0-3-0">
 <h3>Contributors v0.3.0<a class="headerlink" href="#contributors-v0-3-0" title="Permalink to this headline">¶</a></h3>
 <ul class="simple">
 <li><p>Matthias Feurer</p></li>
 <li><p>Jesper van Engelen</p></li>
 </ul>
-</section>
-</section>
-<section id="version-0-2-1">
+</div>
+</div>
+<div class="section" id="version-0-2-1">
 <h2>Version 0.2.1<a class="headerlink" href="#version-0-2-1" title="Permalink to this headline">¶</a></h2>
 <ul class="simple">
 <li><p>Allows the usage of scikit-learn 0.18.2.</p></li>
@@ -966,7 +1021,7 @@ <h2>Version 0.2.1<a class="headerlink" href="#version-0-2-1" title="Permalink to
 <li><p>Typo fixes in print statements.</p></li>
 <li><p>New method to retrieve the models used in the final ensemble.</p></li>
 </ul>
-<section id="contributors-v0-2-1">
+<div class="section" id="contributors-v0-2-1">
 <h3>Contributors v0.2.1<a class="headerlink" href="#contributors-v0-2-1" title="Permalink to this headline">¶</a></h3>
 <ul class="simple">
 <li><p>Matthias Feurer</p></li>
@@ -977,9 +1032,9 @@ <h3>Contributors v0.2.1<a class="headerlink" href="#contributors-v0-2-1" title="
 <li><p>Vicente Alencar</p></li>
 <li><p>Lukas Großberger</p></li>
 </ul>
-</section>
-</section>
-<section id="version-0-2-0">
+</div>
+</div>
+<div class="section" id="version-0-2-0">
 <h2>Version 0.2.0<a class="headerlink" href="#version-0-2-0" title="Permalink to this headline">¶</a></h2>
 <ul class="simple">
 <li><p><strong>auto-sklearn supports custom metrics and all metrics included in
@@ -992,7 +1047,7 @@ <h2>Version 0.2.0<a class="headerlink" href="#version-0-2-0" title="Permalink to
 <li><p><em>auto-sklearn</em> can use multiprocessing in calls to <code class="docutils literal notranslate"><span class="pre">predict()</span></code> and
 <code class="docutils literal notranslate"><span class="pre">predict_proba</span></code>. By <a class="reference external" href="https://github.com/lsorber">Laurent Sorber</a>.</p></li>
 </ul>
-<section id="contributors-v0-2-0">
+<div class="section" id="contributors-v0-2-0">
 <h3>Contributors v0.2.0<a class="headerlink" href="#contributors-v0-2-0" title="Permalink to this headline">¶</a></h3>
 <ul class="simple">
 <li><p>Matthias Feurer</p></li>
@@ -1000,12 +1055,12 @@ <h3>Contributors v0.2.0<a class="headerlink" href="#contributors-v0-2-0" title="
 <li><p>Laurent Sorber</p></li>
 <li><p>Rafael Calsaverini</p></li>
 </ul>
-</section>
-</section>
-<section id="version-0-1-x">
+</div>
+</div>
+<div class="section" id="version-0-1-x">
 <h2>Version 0.1.x<a class="headerlink" href="#version-0-1-x" title="Permalink to this headline">¶</a></h2>
 <p>There are no release notes for auto-sklearn prior to version 0.2.0.</p>
-<section id="contributors-v0-1-x">
+<div class="section" id="contributors-v0-1-x">
 <h3>Contributors v0.1.x<a class="headerlink" href="#contributors-v0-1-x" title="Permalink to this headline">¶</a></h3>
 <ul class="simple">
 <li><p>Matthias Feurer</p></li>
@@ -1025,9 +1080,9 @@ <h3>Contributors v0.1.x<a class="headerlink" href="#contributors-v0-1-x" title="
 <li><p>_329_</p></li>
 <li><p>Iver Jordal</p></li>
 </ul>
-</section>
-</section>
-</section>
+</div>
+</div>
+</div>
 
 
     </div>
@@ -1048,7 +1103,7 @@ <h3>Contributors v0.1.x<a class="headerlink" href="#contributors-v0-1-x" title="
       
     </p>
     <p>
-        &copy; Copyright 2014-2021, Machine Learning Professorship Freiburg.<br/>
+        &copy; Copyright 2014-2022, Machine Learning Professorship Freiburg.<br/>
       Created using <a href="http://sphinx-doc.org/">Sphinx</a> 4.2.0.<br/>
     </p>
   </div>
diff --git a/master/search.html b/master/search.html
index 5dce38d208..75ba2fb3ec 100644
--- a/master/search.html
+++ b/master/search.html
@@ -4,7 +4,7 @@
   <head>
     <meta charset="utf-8" />
     <meta name="viewport" content="width=device-width, initial-scale=1.0" />
-    <title>Search &#8212; AutoSklearn 0.14.2 documentation</title>
+    <title>Search &#8212; AutoSklearn 0.14.4 documentation</title>
     <link rel="stylesheet" type="text/css" href="_static/pygments.css" />
     <link rel="stylesheet" type="text/css" href="_static/bootstrap-sphinx.css" />
     <link rel="stylesheet" type="text/css" href="_static/sg_gallery.css" />
@@ -60,7 +60,7 @@
         </button>
         <a class="navbar-brand" href="index.html">
           auto-sklearn</a>
-        <span class="navbar-text navbar-version pull-left"><b>0.14.2</b></span>
+        <span class="navbar-text navbar-version pull-left"><b>0.14.4</b></span>
       </div>
 
         <div class="collapse navbar-collapse nav-collapse">
@@ -153,7 +153,7 @@ <h1 id="search-documentation">Search</h1>
       
     </p>
     <p>
-        &copy; Copyright 2014-2021, Machine Learning Professorship Freiburg.<br/>
+        &copy; Copyright 2014-2022, Machine Learning Professorship Freiburg.<br/>
       Created using <a href="http://sphinx-doc.org/">Sphinx</a> 4.2.0.<br/>
     </p>
   </div>
diff --git a/master/searchindex.js b/master/searchindex.js
index c25d5cf87a..75c7ecefca 100644
--- a/master/searchindex.js
+++ b/master/searchindex.js
@@ -1 +1 @@
-Search.setIndex({docnames:["api","examples/20_basic/example_classification","examples/20_basic/example_multilabel_classification","examples/20_basic/example_multioutput_regression","examples/20_basic/example_regression","examples/20_basic/sg_execution_times","examples/40_advanced/example_calc_multiple_metrics","examples/40_advanced/example_debug_logging","examples/40_advanced/example_feature_types","examples/40_advanced/example_get_pipeline_components","examples/40_advanced/example_inspect_predictions","examples/40_advanced/example_interpretable_models","examples/40_advanced/example_metrics","examples/40_advanced/example_pandas_train_test","examples/40_advanced/example_resampling","examples/40_advanced/example_single_configuration","examples/40_advanced/sg_execution_times","examples/60_search/example_parallel_manual_spawning_cli","examples/60_search/example_parallel_manual_spawning_python","examples/60_search/example_parallel_n_jobs","examples/60_search/example_random_search","examples/60_search/example_sequential","examples/60_search/example_successive_halving","examples/60_search/sg_execution_times","examples/80_extending/example_extending_classification","examples/80_extending/example_extending_data_preprocessor","examples/80_extending/example_extending_preprocessor","examples/80_extending/example_extending_regression","examples/80_extending/example_restrict_number_of_hyperparameters","examples/80_extending/sg_execution_times","examples/index","extending","faq","index","installation","manual","releases"],envversion:{"sphinx.domains.c":2,"sphinx.domains.changeset":1,"sphinx.domains.citation":1,"sphinx.domains.cpp":4,"sphinx.domains.index":1,"sphinx.domains.javascript":2,"sphinx.domains.math":2,"sphinx.domains.python":3,"sphinx.domains.rst":2,"sphinx.domains.std":2,"sphinx.ext.viewcode":1,sphinx:56},filenames:["api.rst","examples/20_basic/example_classification.rst","examples/20_basic/example_multilabel_classification.rst","examples/20_basic/example_multioutput_regression.rst","examples/20_basic/example_regression.rst","examples/20_basic/sg_execution_times.rst","examples/40_advanced/example_calc_multiple_metrics.rst","examples/40_advanced/example_debug_logging.rst","examples/40_advanced/example_feature_types.rst","examples/40_advanced/example_get_pipeline_components.rst","examples/40_advanced/example_inspect_predictions.rst","examples/40_advanced/example_interpretable_models.rst","examples/40_advanced/example_metrics.rst","examples/40_advanced/example_pandas_train_test.rst","examples/40_advanced/example_resampling.rst","examples/40_advanced/example_single_configuration.rst","examples/40_advanced/sg_execution_times.rst","examples/60_search/example_parallel_manual_spawning_cli.rst","examples/60_search/example_parallel_manual_spawning_python.rst","examples/60_search/example_parallel_n_jobs.rst","examples/60_search/example_random_search.rst","examples/60_search/example_sequential.rst","examples/60_search/example_successive_halving.rst","examples/60_search/sg_execution_times.rst","examples/80_extending/example_extending_classification.rst","examples/80_extending/example_extending_data_preprocessor.rst","examples/80_extending/example_extending_preprocessor.rst","examples/80_extending/example_extending_regression.rst","examples/80_extending/example_restrict_number_of_hyperparameters.rst","examples/80_extending/sg_execution_times.rst","examples/index.rst","extending.rst","faq.rst","index.rst","installation.rst","manual.rst","releases.rst"],objects:{"autosklearn.classification":{AutoSklearnClassifier:[0,0,1,""]},"autosklearn.classification.AutoSklearnClassifier":{fit:[0,1,1,""],fit_ensemble:[0,1,1,""],fit_pipeline:[0,1,1,""],get_configuration_space:[0,1,1,""],get_models_with_weights:[0,1,1,""],get_params:[0,1,1,""],leaderboard:[0,1,1,""],predict:[0,1,1,""],predict_proba:[0,1,1,""],refit:[0,1,1,""],score:[0,1,1,""],set_params:[0,1,1,""],show_models:[0,1,1,""],sprint_statistics:[0,1,1,""]},"autosklearn.experimental.askl2":{AutoSklearn2Classifier:[0,0,1,""]},"autosklearn.experimental.askl2.AutoSklearn2Classifier":{fit:[0,1,1,""],fit_ensemble:[0,1,1,""],fit_pipeline:[0,1,1,""],get_configuration_space:[0,1,1,""],get_models_with_weights:[0,1,1,""],get_params:[0,1,1,""],leaderboard:[0,1,1,""],predict:[0,1,1,""],predict_proba:[0,1,1,""],refit:[0,1,1,""],score:[0,1,1,""],set_params:[0,1,1,""],show_models:[0,1,1,""],sprint_statistics:[0,1,1,""]},"autosklearn.metrics":{accuracy:[0,2,1,""],average_precision:[0,2,1,""],balanced_accuracy:[0,2,1,""],f1:[0,2,1,""],f1_macro:[0,2,1,""],f1_micro:[0,2,1,""],f1_samples:[0,2,1,""],f1_weighted:[0,2,1,""],log_loss:[0,2,1,""],make_scorer:[0,3,1,""],mean_absolute_error:[0,2,1,""],mean_squared_error:[0,2,1,""],median_absolute_error:[0,2,1,""],precision:[0,2,1,""],precision_macro:[0,2,1,""],precision_micro:[0,2,1,""],precision_samples:[0,2,1,""],precision_weighted:[0,2,1,""],r2:[0,2,1,""],recall:[0,2,1,""],recall_macro:[0,2,1,""],recall_micro:[0,2,1,""],recall_samples:[0,2,1,""],recall_weighted:[0,2,1,""],roc_auc:[0,2,1,""]},"autosklearn.pipeline.components.base":{AutoSklearnClassificationAlgorithm:[0,0,1,""],AutoSklearnPreprocessingAlgorithm:[0,0,1,""],AutoSklearnRegressionAlgorithm:[0,0,1,""]},"autosklearn.pipeline.components.base.AutoSklearnClassificationAlgorithm":{get_estimator:[0,1,1,""],predict:[0,1,1,""],predict_proba:[0,1,1,""]},"autosklearn.pipeline.components.base.AutoSklearnPreprocessingAlgorithm":{get_preprocessor:[0,1,1,""],transform:[0,1,1,""]},"autosklearn.pipeline.components.base.AutoSklearnRegressionAlgorithm":{get_estimator:[0,1,1,""],predict:[0,1,1,""]},"autosklearn.pipeline.components.classification":{add_classifier:[31,3,1,""]},"autosklearn.pipeline.components.feature_preprocessing":{add_preprocessor:[31,3,1,""]},"autosklearn.pipeline.components.regression":{add_regressor:[31,3,1,""]},"autosklearn.regression":{AutoSklearnRegressor:[0,0,1,""]},"autosklearn.regression.AutoSklearnRegressor":{fit:[0,1,1,""],fit_ensemble:[0,1,1,""],fit_pipeline:[0,1,1,""],get_configuration_space:[0,1,1,""],get_models_with_weights:[0,1,1,""],get_params:[0,1,1,""],leaderboard:[0,1,1,""],predict:[0,1,1,""],refit:[0,1,1,""],score:[0,1,1,""],set_params:[0,1,1,""],show_models:[0,1,1,""],sprint_statistics:[0,1,1,""]}},objnames:{"0":["py","class","Python class"],"1":["py","method","Python method"],"2":["py","attribute","Python attribute"],"3":["py","function","Python function"]},objtypes:{"0":"py:class","1":"py:method","2":"py:attribute","3":"py:function"},terms:{"0":[0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,18,19,20,21,22,23,24,25,26,27,28,29,34,35],"00":[5,16,23,29],"000":10,"000000":[2,6,11,26,27],"00001":27,"000010":12,"0001":[1,3,9,20,21,22,24,26,28],"00010083678209087315":[1,21],"00010495766342828794":25,"00010868227587711292":20,"0001363185819149026":[1,21,22],"0001384744065375479":20,"00014078218950356456":1,"00014375616988222174":9,"00015788887334393744":9,"0001745391328519669":4,"00017888891850981892":[1,21],"00018009776276177523":[1,21,22],"0002346515712987664":22,"0002600768160857831":[1,21,22],"00027271287919467994":21,"0002944604673080368":[1,21],"00031132126574883666":20,"0003336259810364362":11,"0003701926442639788":4,"000421568792103947":21,"0004684917334431039":[1,21,22],"0005485701552075972":20,"0006517033225329654":4,"0007301343236220855":4,"001":[3,10,28],"0010364937923091088":20,"0010413452644415357":4,"0011804284312897009":21,"0012559712452161484":15,"0014466154794086487":20,"0015130257264171173":9,"0015543757334360092":20,"0017940473175767063":[1,21,22],"0017964371726429177":20,"0018055343233337954":22,"0018475934154961247":20,"002431731981071206":4,"0024386428371511353":[1,21],"002766772136115771":9,"003":10,"004071801722749603":21,"004635634934629576":9,"005003173341372008":[1,21],"005069923784737444":9,"005326508887463406":[21,22],"005428587241449129":22,"005821344021856017":20,"006":10,"006197":1,"00684163":9,"006915066463246923":11,"007017543859649145":9,"0070580904199417415":[1,21,22],"0074025063474275335":20,"0076293":9,"008015420020402715":9,"008542188583124829":[1,21,22],"00894849":9,"009193751229756878":20,"01":[2,3,5,9,12,16,20,22,23,24,26,27,28,29],"010000000000000004":[1,9,21,22],"010000000000000014":11,"01006215":9,"0104432":9,"011283688651384545":22,"011333066835975528":9,"011901034843417571":22,"012143858010275621":15,"012150149892783745":4,"012199160249499144":11,"01223987":9,"012322274881516604":15,"012391313886912093":9,"012607824914758717":4,"013399889539803811":20,"014034416932373166":20,"014184":[1,6],"015996368052062886":1,"016444224834275295":4,"01667385805274091":20,"01684553":9,"017116661677715188":21,"018038793247120707":20,"01851537":9,"02":[1,4,5,9,16,23],"020000":[1,4,20,21,22],"020501":1,"0208475757765491":21,"021277":[1,6],"021944240404849075":4,"02258105767523384":25,"02260942170772907":[1,21],"022834561782884507":22,"023878029144121318":20,"025":10,"02641737e":9,"028368794326241176":9,"028369":[1,6],"02847755502162456":21,"03":[9,23],"030776":1,"03125":28,"03208923":9,"032719158639429445":[1,21,22],"0333981795670597e":9,"03409213e":9,"034330895903523305":[1,21],"03457277874230573":9,"03546099290780147":9,"035461":[1,6],"035543606049804645":20,"037":10,"03737258e":9,"038":[20,23],"039":[13,16],"03961232028373377":22,"04":[1,9,12,16,23,36],"040000":[1,9,20,21,22],"040063255316952e":20,"04074":33,"04210526315789476":9,"042553":[1,6],"045":12,"04561403508771933":9,"045974":1,"04865917":9,"049645":[1,6],"04986510953238":[1,21],"05":[1,3,4,9,16,20,21,22,25,27,28],"05082904256838701":22,"05141281638752715":22,"0517578125e":[3,28],"052862156055921525":22,"054648061621061846":24,"056":10,"05673758865248224":9,"056738":[1,6],"057778875694463e":[1,21,22],"05972079854295879":22,"06":[1,3,4,5,12,16,20,21],"060000":[1,4,20,21,22],"060800813211425456":[21,22],"06290764761689918":20,"063":12,"06315789473684208":9,"06332899":9,"06334655":9,"065":[10,16],"067":12,"06707365823758892":20,"068":12,"06835899987255477":20,"06846183238648647":24,"07":[1,4,9,12,20,21,22,28,29],"0706216214952021":[1,21],"07092198581560283":9,"070922":6,"071":12,"07213284":9,"07563955664363409":[1,21],"078014":[1,6],"07801418439716312":9,"07827509e":9,"0792482460648931":20,"08":[1,3,9,16,20,21,22,28],"080000":[1,20,21,22],"08087614244138486":9,"08125391652261632":22,"085106":[1,6],"09":[4,7,12,18,28],"09111888":9,"09122807017543855":9,"09179220974965213":[1,21,22],"0938587":9,"0945814167023392e":22,"09580337973953734":21,"09809101":9,"09809681787962342":21,"09814953":9,"09884140378258977":22,"0th":9,"0x7f9379a40f10":9,"0x7f937b0d0a00":15,"0x7f937b0d0e50":15,"0x7f937c5a9ca0":15,"1":[0,1,2,3,4,6,7,8,11,12,13,14,15,18,19,20,21,22,24,25,26,27,28,33,35],"10":[0,1,3,4,5,8,9,10,16,17,18,20,21,22,23,24,27,28,34],"100":[1,3,9,14,20,21,22,27,28],"1000":[3,28,36],"10000":[3,28],"100000":[3,20,21,22],"1001":36,"1004":36,"1007":36,"1008":36,"101":[1,21,22],"1012":36,"1014":36,"1017":36,"1024":[17,18,36],"1025":32,"1026":36,"1038":36,"103855734598575e":9,"10392737384710127":20,"1040":36,"1045":36,"1046":36,"1050":36,"1052":36,"1053":36,"1059":36,"1061":36,"1062":36,"1063829787234043":9,"1070":36,"1072":36,"1075":36,"1076":36,"1076975":9,"1079":36,"1085501":9,"11":[1,4,7,12,15,17,18,20,21,22,27,29,34],"1100":36,"1102":[1,21,22],"11042628136263043":22,"1108":36,"1109":36,"1112":36,"1114":36,"1115":36,"1117":[20,36],"111869":1,"1121":36,"1122":36,"1123":36,"1124":36,"1126":36,"1128":36,"1133":36,"1134":36,"11347517730496459":9,"1135":36,"1136":[4,36],"1142":36,"1143":36,"114391364050916":20,"1144":36,"1149":36,"115":[1,21,22,34],"1152":36,"1154":36,"1157":36,"1158":36,"1159":36,"1165":36,"1168":36,"1169":36,"117":12,"1170":36,"1173":36,"11737571e":9,"1178":36,"1179":36,"118":29,"1183":36,"1185":36,"1187":36,"1190":36,"1195":36,"1198":36,"11ec":[7,17,18,25],"12":[1,17,18,20,22,23],"120":[1,3,4,6,10,11,13,14,15,19,21,25],"1200":36,"120000":[1,9,20,22],"1201":36,"1204":36,"1209":36,"1210":36,"1214":36,"1229":36,"123":21,"1231":36,"1233":36,"1235":36,"1239":36,"1242":36,"1245":36,"1248":36,"12585871697357337":20,"1259":36,"126":10,"1261":36,"1263":36,"1269":36,"1270":36,"1271":36,"128":5,"1280839244790386":27,"1282":36,"1290":36,"13":[1,15,20,21,22],"13300503334706695":22,"13366756044911932":20,"13704890e":9,"1387664107898294":20,"14":[1,3,4,15,18,21,22],"140":[1,12,21,32,36],"140000":[4,9,22],"14069368736662313":9,"142":35,"14268277711454813":[1,21,22],"14295295806077554":[1,21,22],"1450":[1,21],"14555017892884228":24,"149":36,"15":[1,2,4,5,13,20,22,28],"15162626175424537":20,"152":[21,22,23],"1567":[1,21],"157":36,"157218":3,"15890408760884822":[1,21],"16":[0,1,20,21,24,28,36],"160000":[9,22],"16262682406125173":22,"16265262021972576":22,"1636487580":15,"1636487583":15,"1636488092":9,"1636488094":9,"1636488096":9,"1636488098":9,"1636488099":9,"1636488102":9,"1636488103":9,"1636488105":9,"1636488106":9,"1636488107":9,"1636488110":9,"163809":1,"1644":20,"166":[6,16],"16649281":9,"17":[1,6,20,21],"170213":6,"17122959":9,"17152095":9,"17184142431776436":[1,21],"1749860017522":[1,21],"17677801820619737":24,"18":[1,3,12,20,21,22,24,26,29,36],"180":9,"180000":[3,22,25],"18010277":9,"18012072":9,"182":[1,21,22],"184":12,"187":24,"1894":20,"19":[1,7,9,12,16,20,21,22,36],"191":12,"195":[4,5],"1b4c952643c3":[7,17,18,25],"1e":[1,3,4,9,20,21,22,27,28],"1e2":27,"1st":32,"1th":9,"2":[0,1,2,3,4,6,7,13,15,18,19,20,21,22,25,26,27,28],"20":[1,2,3,8,9,12,17,18,20,21,22,28,29,36],"200":[1,21,22,28],"2000":[3,28],"200000":3,"2004":[15,35],"2006":35,"20068746139723115":22,"2007":33,"200900":1,"2015":[32,33,36],"2020":33,"2021":[7,12,18],"2037591637980971e":4,"204":15,"2047":28,"207332":3,"20759805e":9,"20_basic":2,"21":[16,36],"2118001735899097e":4,"21521743568582094":4,"216":24,"216493":4,"2192999307296778":20,"22":[1,6,7,14,16,22,36],"226":12,"229":9,"23":[0,12,20,25,29,36],"232":12,"2329674":9,"23434217313954":[1,21],"2345103479872152":20,"23746960178084334":22,"238533890074848e":4,"23910875621465733":20,"24":[7,12,15,16,19,22,23,36],"240000":22,"241306":4,"24966112905354662":20,"24982831110057324":4,"25":[0,3,4,12,16,28],"2531207":9,"255":[1,4,20,21,22,28],"25545052141264185":1,"25595970768123566":[21,22],"26":[1,6],"260000":22,"2613be54":25,"263":21,"264":28,"2641013":9,"269":12,"27":[1,4,12],"271":36,"273":20,"2758095":9,"278":12,"279105226196228":20,"27951285357994865":20,"28":[7,33],"280":[10,36],"280000":[22,25],"282688":3,"2841884333778574e":21,"285":20,"285306370007781":20,"28817393948552533":20,"29":[1,6,12,21],"293":36,"2949722712387337":20,"295":36,"2958033930435781e":[21,22],"2962":33,"29662315421901003":20,"2970":33,"2nd":32,"2th":9,"3":[1,2,3,4,6,7,8,9,11,13,14,15,18,20,21,22,24,26,27,28,33,34],"30":[1,2,3,4,6,7,8,9,10,11,12,13,14,17,18,19,21,22,24,26,27,28,35,36],"300":4,"3072":[0,19],"30755227194768237":[1,21,22],"3094962987325228":21,"31":[12,20,28],"310":[3,5],"312":[25,29],"317":36,"318861":3,"32":[0,1,4,6,9,20,21,22,24,28,36],"320000":4,"32106218519214325":22,"32225143e":9,"325":36,"326":[12,36],"32768":28,"328":21,"328568999403367e":25,"32956881e":9,"33":[0,1,2,20],"338":21,"34":[1,6,9,12,16,18],"340000":[9,22],"3440590734654458e":20,"34481806774236556":20,"348":12,"35":[1,6,12,17,18,20,23],"351":36,"355":[1,21],"358":36,"358378124977852e":4,"36":20,"3600":0,"3621762718897781":4,"36245140e":9,"363":[17,23],"368":20,"369":[24,29],"369811497206404":9,"37":[1,6],"3716748930467322e":22,"3720251865973017":20,"375":36,"37641366e":9,"37705188916038523":22,"3793256905143867":[1,21],"38":[12,20],"38430977e":9,"38461538461538464":14,"3871420537981852":[1,21],"387912939529945e":[1,21,22],"388":12,"39":[1,12,17,21,22,23],"3959585042866587":4,"39749085e":9,"397684":7,"39844669585806":[1,21,22],"3f":[10,12],"3gb":[19,32,35],"3rd":[32,36],"3th":9,"4":[1,4,9,10,14,19,20,21,22,27,28,34],"40":[1,12,22],"400":[3,4,28],"4037113":9,"40594":2,"40665293991189455":15,"409":36,"40922":10,"4096":15,"40981":[8,13],"40_advanc":8,"41":21,"410":36,"410144741041167e":22,"410651487451908e":20,"411":12,"412":7,"413":12,"414":20,"41808321658160696":22,"41826215858914706":[21,22],"419":12,"4196":7,"4199":[17,18],"419a":25,"420000":24,"421745686388309e":20,"424":[14,36],"425":[14,36],"42693600390988135":[1,22],"428":12,"42961621e":9,"43":20,"431":34,"436679":4,"438":36,"43999367631975456":22,"442":[19,23],"4433d19e":17,"444":36,"44455023733729254":20,"44685232":9,"447294":2,"45":12,"454935":3,"455042":4,"45597150689539434":20,"459119":3,"4595393":9,"45954164":9,"46":4,"460000":4,"4622203211010016e":20,"462249":4,"4635442279519353":21,"468":[9,16],"475":[12,16],"4775492074518431":1,"4791448484072812":[1,21],"48":[19,23],"482":36,"48507733455436836":[1,21],"4882838781863879e":20,"4886183647550176":25,"49":[12,19],"491":36,"49138075723513286":[1,22],"49326086e":9,"49503611":9,"498":36,"4th":9,"5":[1,2,3,4,9,10,11,12,13,14,15,19,20,21,22,24,25,26,27,28],"50":[0,3,7,12,17,18,20,21,28,32],"500":[14,16,28,36],"5000":10,"5027651786804199":9,"503":25,"503079":1,"504988116581138e":4,"50650698230178e":4,"507400":4,"512":15,"5121":[1,21],"5137138356508442":1,"518673":4,"519158":1,"52":[7,20,23],"520111487139148e":20,"521":22,"526":9,"5286041327277":4,"530":20,"5312755":15,"532":36,"535":36,"5358099937438965":9,"537":36,"537016":9,"538":[12,36],"54":[12,21],"540000":25,"55":[3,5,11,16],"550960330919455":[21,22],"552706":2,"5550223982458062e":21,"5561389303694175":[1,21],"55773043e":9,"558":36,"56":[1,12,22],"562":18,"562561668029056":22,"5629768371582031":9,"56297684":9,"566":36,"5662900693317384":1,"569":36,"5707983257382487":[1,21,22],"571":[11,16],"57629717e":9,"58":[1,4,5,21],"580000":24,"5846457091661748e":9,"585":36,"58542976103":4,"58615414999356":20,"588":36,"5905006626969":9,"592":36,"593":36,"5942185":15,"5944780427522034":4,"5b2e548d":18,"5gb":32,"5th":9,"6":[1,4,9,14,20,21,22],"60":[1,2,12,15,20,21,22,36],"600":36,"604":2,"6041510105133057":9,"606":[12,20],"6066544055938721":9,"609412172481434e":22,"611":36,"6128603428070196":22,"614":36,"6140191555023193":9,"61401916":9,"615":12,"618":36,"62":3,"620":[8,16],"620000":3,"6228726042266908":20,"626":36,"6277363920171745":4,"628683":4,"636021":2,"639516":4,"64":[0,20,32,34],"6401973":9,"64437157661465":20,"645808":4,"646":36,"648":36,"650":[1,5,36],"6549544334411621":9,"655":36,"6587261797143242":20,"66":[0,22],"660":36,"661377361743419":[1,21],"6661824659281315":9,"667":36,"669":36,"67":[0,14,22],"670":36,"6702926158905029":9,"67029262":9,"673455":1,"676":36,"6779841015398226":22,"680":36,"6836522210129492":20,"688":36,"691132":1,"692":16,"69413781":9,"6941378116607666":9,"6a2":36,"6gb":35,"6th":9,"7":[1,3,4,6,9,20,22,25,28,34],"701":36,"7056696489196201":20,"70629898e":9,"7104048919054863":20,"715":36,"7171678618990129":21,"718799":1,"726182961545702":[1,21],"7305615609807856":[21,22],"731":[12,23],"732":36,"7323115919225983":22,"733":36,"737":12,"7385828495025635":9,"73858285":9,"739":36,"7393440007191528":[1,21],"74116751e":9,"74597367e":9,"7462342366289323e":4,"75":[3,22,28],"7530273530198002":20,"7563245":9,"75655689e":9,"7570b3":4,"761":[26,29],"764":36,"7664515192592053e":4,"76699224":9,"769":36,"7708651296941559":22,"771":36,"772":36,"773":12,"774":36,"777":[17,18],"7772788812704434":9,"778":36,"7797232":9,"781":36,"7818500358383581":22,"78251":4,"7845396961078424":1,"785":36,"7852218490061461":20,"786517":6,"7895711479212801":9,"79057296e":9,"791":36,"791639810538127":20,"794":36,"795":36,"795890":1,"796":[18,23],"7978382221839412":20,"799":[28,29],"7990398406982422":9,"7th":9,"8":[1,3,4,6,9,20,21,22,28],"80":[6,12,20,36],"803":36,"8047274080856589":9,"8057830372269097":4,"80724345e":9,"807816607300627":[1,21],"8134515743047006":22,"814":[12,36],"815":36,"8163307":9,"821530":1,"8219044":9,"822":36,"824072":1,"828":20,"829787":6,"8303675651550293":15,"8355012":9,"840":36,"8428972335335263e":4,"843":36,"843515731149308":20,"845182":6,"8465036896999549":24,"85086661900421e":20,"851598":9,"852":36,"853659":6,"854":36,"856":32,"8572731018066406":9,"8579710144927536":13,"860":34,"8601586365248128":22,"862":36,"86214879612422":20,"863":36,"864":36,"86528503e":9,"8683277328634489":3,"869":36,"8694334":9,"87":20,"870":36,"8704240322113037":9,"872":36,"87397441278958":21,"874":36,"876":[27,29,36],"877":[15,16],"8772654126094648":3,"8786127167630058":8,"879":36,"881":36,"881188":6,"882":36,"884615":6,"886":36,"88652482":9,"88675402e":9,"887640":6,"8888":34,"8888447284698486":9,"88884473":9,"889":36,"890000":6,"891":36,"8911872287672372":20,"893":36,"89361702":9,"893634d8":7,"894231":6,"897":36,"8a32":[7,17,18,25],"8b67af6":36,"9":[1,4,9,20,21,22,28,34],"90":[1,10,21],"900":36,"901":36,"902":36,"905":36,"907":36,"909":36,"911":[7,16,36],"91287172e":9,"914":36,"914306402206421":15,"914894":6,"916":36,"918":36,"919":36,"92198582":9,"921986":6,"923833":6,"924":36,"924589":6,"925":36,"926283631486858":22,"927":36,"927211764000264":20,"9290439925152777":9,"929078":6,"92907801":9,"9292309396985746":[1,21,22],"929412":6,"93":[1,21,22],"931":[22,36],"933":36,"9331254454871041":9,"933333":6,"9350605":9,"935069":6,"936":36,"936170":[6,25],"936690":6,"936842":6,"937":[12,22],"9370629370629371":[25,26],"940":[22,23,36],"940687":6,"941799":6,"943064":6,"943262":6,"94326241":9,"943820":6,"944":12,"9440559440559441":[1,6,14,20,22],"945055":6,"945652":6,"945787":7,"946":36,"94611519e":9,"947":36,"948":36,"948682":6,"949":36,"949611":1,"950303":6,"950355":6,"9503612101678187":20,"950961":1,"951":36,"951048951048951":[9,21,22,24],"952":4,"952558":1,"953":36,"954300":6,"954545":6,"9545562267303467":9,"95455623":9,"955056":6,"955556":6,"956":36,"956044":6,"9565048776795676":20,"956989":6,"957447":6,"958041958041958":[11,14],"9586161969293623":20,"959918":6,"960452":6,"961":36,"961326":6,"961538":6,"961749":6,"963915":6,"9639365673065186":9,"96393657":9,"964":7,"964539":6,"96453901":9,"965034965034965":[14,17,18,22],"965517":6,"965536":6,"965909":6,"966292":6,"966667":6,"967033":6,"967136":22,"968546":1,"9694955791292086":1,"969533":6,"970":36,"971348315581384":1,"971631":[6,9],"97163121":9,"971751":6,"971831":22,"972067":6,"973":[2,5,36],"975151":6,"975309":6,"9755563101648113":9,"9772109830437746":22,"977273":6,"977528":6,"977778":6,"978022":6,"978723":[6,22],"979149":6,"97947373958566":[1,22],"98":33,"98080571":9,"9826666666666667":10,"983051":6,"983240":6,"9836355993007913":[1,21],"984":36,"984767":6,"985816":[6,19,21,22],"988636":6,"9886363636363636":7,"988764":6,"989":36,"989374401993":[1,21],"989729761503726":21,"9899874843554443":7,"99":28,"990":36,"9911616161616161":7,"992":36,"992908":[17,18,20],"9929881254946676":[1,21,22],"993803313878608":[21,22],"9954285714285714":10,"9967857433838874":9,"9968675728600899":20,"997":36,"9984367650965825":[1,21,22],"999":[1,3,9,20,21,22,28],"9999":[3,9,28],"abstract":[0,31],"boolean":[0,35],"break":36,"case":[0,8,9,11,14,17,22,30,32,34,35,36],"class":[0,2,7,10,14,24,25,26,27,28,31,35,36],"concei\u00e7\u00e3o":36,"default":[0,3,7,14,17,21,22,24,26,27,28,35,36],"do":[0,2,7,8,17,18,21,31,32,34,35,36],"export":35,"final":[0,9,17,20,22,32,35,36],"float":[0,8,24,26,27,28,32,35],"function":[0,6,12,13,31,32,36],"gro\u00dfberg":36,"import":[0,1,2,3,4,6,7,8,9,11,12,13,14,15,18,19,20,21,22,24,25,26,27,28,32,33,35,36],"int":[0,2,24,27,28],"long":32,"mielik\u00e4inen":36,"new":[0,15,17,18,24,27,28,31,32,35,36],"public":33,"return":[0,6,12,13,15,20,22,24,25,26,27,28,31,35,36],"short":32,"true":[0,1,2,3,4,6,7,8,9,11,12,13,14,15,17,18,19,20,21,22,24,25,26,27,28,32,33,35],"try":[32,34,36],"while":[0,9,17,32,35],A:[0,4,9,10,11,13,20,33,34,35,36],As:[9,17,18,32,35],At:35,By:[0,7,21,32,35,36],For:[0,7,10,13,14,32,34,35],If:[0,2,8,13,17,18,32,33,34,35],In:[0,6,8,9,13,14,17,31,32,34,35],It:[0,9,14,17,18,22,33,34,35,36],NOT:32,No:36,Not:[0,2,7,8,24,25,26,27,28,34],Of:31,One:[0,4,14,32,35],The:[0,1,3,4,6,7,8,9,10,11,12,13,14,15,17,18,20,21,24,25,26,27,28,31,32,34,35,36],There:[10,13,32,34,36],These:[0,31,32,35],To:[0,2,8,9,10,14,15,17,18,19,22,25,31,32,33,34,36],Will:0,_329_:36,_:[7,15,17,18,19],__:0,__choice__:[1,2,3,4,6,9,11,15,20,21,22,24,25,26,27,28],__init__:[24,25,26,27,28],__main__:[17,18,19,20,22,32,33,36],__name__:[17,18,19,20,22,32,33,36],_gatheringfutur:18,_metric:12,_repeatedsplit:0,_test_scor:13,_valu:15,a10:13,a11:13,a12:13,a13:13,a14:13,a1:13,a2:13,a3:13,a4:13,a5:13,a6:13,a7:13,a8:13,a9:13,aaron:[33,36],abbrevi:[22,31],abil:35,abl:[32,35],about:[0,7,15,17,19,20,22,31,33,34,35,36],abov:[4,32,33,35],acceleration_i:10,acceleration_x:10,acceleration_z:10,acceleromet:10,accept:[0,12,32,35],access:[9,35,36],accidentali:36,accord:[10,14,35],accordingli:8,accu:12,accu_add:12,accuraci:[0,1,2,6,8,9,11,13,14,17,18,19,20,21,22,25,26,32,33,35,36],accuracy_scor:[1,2,6,8,9,11,12,13,14,17,18,20,21,22,24,25,26,33],accuracy_wk:12,achiev:[0,34],across:[17,18],action:36,activ:[1,9,20,21,22,24,28,34],actual:[17,18,36],ad:[0,34,36],adaboost:[6,9,11,20,28],adam:[1,9,20,21,22,24,28],adapt:[20,22],add:[24,25,26,27,28,31,32,34,36],add_classifi:[24,28,31],add_condit:[26,27],add_hyperparamet:[24,26,27,28],add_preprocessor:[25,26,31],add_regressor:[27,31],addit:[0,2,7,9,31,32,36],addition:[8,9,20,35],additional_info:[9,15],address:[17,18],adjust:32,adrian:36,advanc:[0,18,22,33,35],advantag:33,advic:32,advis:33,affect:[0,10],affin:[1,3,21,22,28],after:[0,4,14,21,31,35,36],afterward:21,again:[10,36],against:[4,36],agarw:36,aggress:20,ahn:36,al:[33,35],alencar:36,alexand:36,algorithm:[0,1,2,3,9,15,17,18,19,20,21,22,25,26,28,31,32,33,35,36],alia:[0,2,8],align:32,all:[0,9,11,14,15,18,21,30,33,34,35,36],alleavi:32,allevi:35,alloc:[0,19,32],allow:[0,10,17,18,31,32,34,35,36],along:0,alpha:[1,3,4,9,11,20,21,22,24,25,27,28],alpha_1:4,alpha_2:4,alreadi:[13,32,35],also:[0,6,7,9,10,12,13,14,15,17,18,20,21,28,31,32,33,34,35,36],altern:[20,34],alwai:[0,7,35,36],amount:35,an:[0,4,9,10,11,12,13,14,17,21,25,29,30,31,32,33,34,35,36],analysi:26,anatolii:36,andrew:36,ani:[0,2,7,8,15,35,36],anoth:[20,36],answer:32,api:[12,15,31,32,33],appear:32,append:[17,18],appli:[0,32],applic:[0,17],appreci:33,approach:35,appropri:[0,36],apt:34,ar:[0,2,7,8,9,10,13,14,15,18,20,24,25,26,27,28,31,33,34,35,36],arbitrarili:0,ard_regress:4,area:32,arff:36,arg:[0,17,18],argsort:10,argument:[0,8,13,14,22,32,35,36],around:[26,31,35],arrai:[0,2,8,9,13,14,31,35,36],articl:33,arxiv20a:33,arxiv:[33,36],as_fram:[2,7,8,13,15],ascend:0,asctim:7,askl2:[0,33],aspect:35,assert:[12,25,28],assum:32,astyp:[2,13],async:18,asyncio:18,attach:0,attribut:[0,9,13,36],attributeerror:36,australian:[8,13],author:33,auto:[0,6,7,8,10,12,13,14,15,19,20,29,30,36],autom:[33,35],automat:[0,8,13,18,20,33,35,36],automl:[1,2,3,4,7,9,10,11,13,14,15,17,18,19,20,21,22,33,35,36],automl_:[9,12],automlclassifi:36,automlregressor:36,autosklearn2classifi:[0,33,36],autosklearn:[0,1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,22,24,25,26,27,28,31,32,33,35,36],autosklearn_classification_example_tmp:1,autosklearn_inspect_predictions_example_tmp:10,autosklearn_interpretable_models_example_tmp:11,autosklearn_multioutput_regression_example_tmp:3,autosklearn_parallel_1_example_tmp:19,autosklearn_parallel_2_example_tmp:18,autosklearn_parallel_3_example_tmp:17,autosklearn_random_search_example_tmp:20,autosklearn_regression_example_tmp:4,autosklearn_resampling_example_tmp:14,autosklearn_sequential_example_tmp:21,autosklearn_sh_example_tmp:22,autosklearn_sh_example_tmp_01:22,autosklearn_sh_example_tmp_03:22,autosklearn_sh_example_tmp_04:22,autosklearn_sh_example_tmp_cv_02:22,autosklearn_tmp_:[0,32],autosklearnchoic:9,autosklearnclassifi:[0,1,2,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,22,24,25,26,28,32,33,35,36],autosklearnclassificationalgorithm:[0,24,28,31],autosklearnpreprocessingalgorithm:[0,25,26,31],autosklearnregressionalgorithm:[0,27,31],autosklearnregressor:[0,3,4,27,36],avail:[0,7,9,14,17,18,33,34,35,36],availbl:0,averag:[0,1,3,4,9,11,20,21,22,25,28],average_precis:[0,12],avoid:[0,21,32,33],await:18,axi:10,back:[0,34],backend:36,backward:36,bad:[14,36],bae:36,balanc:[0,1,2,9,11,15,20,21,22,24,25,26,28,32],balanced_accuraci:[0,6,12],balanced_accuracy_scor:36,balancing_strategi:0,base:[0,10,24,25,26,27,31,35],basecrossvalid:0,basepipelin:0,baseshufflesplit:0,bash:[17,34],basic:9,basto:36,batch_siz:[0,1,9,20,21,22,28],bayesian:[17,18,33,35,36],becaus:[0,9,32],becker:36,becom:[17,18],been:34,befor:[14,32,33,36],began:0,behavior:[2,8,32,36],behind:33,being:[0,4,9,32,36],believ:32,bellow:[2,7,8,24,25,26,27,28],belong:35,below:[4,7,11,13,14,20,21,28,32],ben:36,berba:36,bereng:36,bergman:36,berman:36,bernoulli_nb:[6,9,11,20,21,28],best:[0,2,4,11,15,17,18,19,20,21,22,25,32,35,36],beta_1:[1,9,20,21,22,28],beta_2:[1,9,20,21,22,28],better:[0,15,32,35,36],between:[0,4,10,17,18,35,36],bibtex:33,bin:34,binari:[0,34,35,36],binder:[1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,22,24,25,26,27,28],bit:[18,32,36],bitsbuff:36,bla:36,blis_num_thread:32,block:32,blum:[33,36],bohb:22,booktitl:33,bool:[0,13,31,36],boost:36,bootstrap:[1,2,3,4,9,15,20,21,22,25,26,28],borovec:36,bosh:22,bot:36,both:[0,10,15,20,24,31,32,33,35],bound:[35,36],boxplot:10,bracket:36,branch:36,breast_canc:[1,9,11,14,19,20,21,22],broken:36,browser:[1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,22,24,25,26,27,28,34],bsd:33,budget:[0,9,15],budget_typ:22,bug:[33,36],build:[0,32,34,36],builder:[7,32,35,36],built:[6,12],builtin:[2,8],bump:36,bunch:8,c:[1,4,9,20,21,22,25,28,32,34],cach:36,calcul:[0,2,6,7,8,24,25,26,27,28,36],calculate_scor:36,call:[0,8,9,11,14,21,31,32,35,36],callabl:0,callback:[0,20,36],calsaverini:36,can:[0,2,4,7,8,9,10,11,12,13,14,15,17,20,21,31,33,34,35,36],cancellederror:18,cannot:[0,32,34,36],caoyi0905:36,cap:15,care:19,caruana:[15,35],categor:[0,3,8,13,24,26,27,28,35,36],categori:[8,13,36],categorical_encod:[1,2,3,4,9,11,15,20,21,22,24,26,27,28],categorical_transform:[1,2,3,4,9,11,15,20,21,22,24,26,27,28],categoricalhyperparamet:[24,26,27],category_coalesc:[1,2,3,4,9,11,15,20,21,22,24,26,27,28],cater:33,caus:36,certain:[32,35],certainti:0,cff:36,challeng:36,chanc:0,chang:[10,11,15,25,30,32,36],channel:34,channel_prior:34,characterist:0,check:[2,7,17,18,19,25,32,33,34,35,36],check_non:26,cheng:36,chi2:[1,20,21,22,28],choic:[3,9,11,15,24,26,27,28,32],choos:[11,35],chosen:[0,36],chu:36,chung:36,ci:36,circl:36,citat:36,cl:[6,7,8,12,13,15,33],clarifi:36,classif:[5,6,7,8,9,10,12,13,14,15,17,18,19,20,21,22,25,26,28,29,30,32,33,35,36],classifi:[0,15,25,26,31,35,36],classification_metr:12,classifierchoic:[11,15],claus:[3,28,33],clean:36,cleanup:36,clear:35,clf:[24,25,26,28],cli:17,cli_start_schedul:17,cli_start_work:17,click:[1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,22,24,25,26,27,28],client:[0,7,18,32,35],clip:36,clonabl:36,clone:36,close:[18,36],cluster:[0,17,18,19,32,36],code:[1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,22,24,25,26,27,28,30,31,32,33,35,36],coef0:[3,9,20,27,28],coef0_condit:27,coeffici:0,col:6,collect:10,column:[0,8,13,35,36],com:[7,31],combin:[10,15,32],command:[13,15,18,19,23,30,32,34,35,36],common:[14,26,34,35],commun:18,compat:[9,15,17,36],competit:36,compil:[32,34],complet:[1,3,21,28,36],compli:[2,15],compliant:36,compon:[0,11,15,17,18,29,30,35,36],compos:35,comput:[18,19,32,36],concept:0,conda:[32,36],condit:[3,26,27,28,32],config:[0,15,18,34],config_id:[0,9],configspac:[0,15,24,25,26,27,28,31],configur:[0,7,9,16,17,18,24,25,27,30,31,33,35,36],configuration_origin:[9,15],configuration_spac:[0,15,24,25,26,27,28,31],configurationspac:[0,24,25,26,27,28,31],connect:[17,18],consid:[0,14,32,35],consist:[0,36],consol:7,constant:[0,3,9,15,17,18,22,24,25,26,27,28,31,36],constraint:[0,15],construct:[9,14,17,18,20,22,31,32,33,35],constructor:0,consum:35,consumpt:36,contact:33,contain:[0,9,13,17,18,28,33,35,36],context:[18,36],continu:0,contrast:0,contrib:32,contribut:36,control:[0,14,32,35,36],convert:[31,36],cookbook:7,copi:[14,32],core:[0,7,19,21,32,35,36],correct:[32,36],correctli:[0,32,34,36],correspond:[0,10,28],cosin:[1,3,20,21,25,28],cost:[0,1,2,3,4,9,12,15],could:[22,34,36],cours:31,cpu:[17,18,32],crash:[0,2,9,17,18,19,20,21,22,25,36],creat:[0,13,15,18,27,31,32,35,36],creation:0,criterion:[1,2,3,4,9,15,20,21,22,25,26,28],cross:[0,32,35,36],crossvalid:0,crucial:[20,35],cs:[15,22,24,26,27,28,33],csr_matrix:35,cube:[3,20,28],cumbersom:18,curc:32,current:[0,2,32,34,35],custom:[0,6,7,14,28,30,35,36],custom_error:6,customrandomforest:28,cutoff:15,cv:[0,14,22],cv_result:[0,6,36],cv_results_:[0,6,9,35],cython:36,d95f02:4,d:[2,7,8,13,35,36],daemon:18,dai:[32,35,36],dask:[0,7,17,18,19,32,35,36],dask_client:[0,17,18,20,22],dask_distributed__worker__daemon:17,data:[0,29,30,31,36],data_id:[2,7,8,10,13,15],data_preprocess:[15,25],data_preprocessor:[0,1,2,3,4,9,11,15,20,21,22,24,25,26,27,28],datafram:[0,6,8,13,30,35,36],datapreprocessorchoic:15,dataset:[0,1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,22,24,27,28,30,33,35,36],dataset_nam:[0,1,2,3,4,9,10,11,14,15,17,18,19,20,21,22],dataset_properti:[1,2,3,4,9,11,20,21,22,24,25,26,27,28],date:32,deadlock:[32,36],debug:[16,30,32],decid:[0,32,36],decis:[0,32],decision_tre:[3,9,11,20,28],decreas:[10,36],deem:11,deep:0,def:[6,12,17,18,20,22,24,25,26,27,28],default_config:22,default_data_contain:32,default_valu:[24,26,27,28],defin:[0,6,10,12,14,31,35],definit:[31,35],deflat:[3,20,28],degre:[1,3,4,9,11,15,20,21,22,27,28],degree_condit:27,delet:[0,7,32,36],delete_tmp_folder_after_termin:[0,7,15,17,18,21,32],demand:32,demonstr:[8,13,14,20,22,24,25,26,27,28,36],deng:36,dens:[24,25,26,27,28,31],densifi:11,depend:[0,31,32,34,36],depict:4,deploy:36,deprec:[2,8],deprecationwarn:[2,8],descend:0,describ:[0,9,10,31,33,35],deshpand:36,design:9,desired_boolean_column:13,desired_categorical_column:13,desired_numerical_column:13,despit:36,destroi:[14,32],detail:[0,2,8,13,32,34,35],determin:[0,35],determinist:36,dev:34,devdoc:[2,8],develop:[33,34,36],diabet:4,diagon:4,dict:[0,35],dictionari:[0,31,36],did:36,diego:36,differ:[14,17,18,28,32,35,36],digit:[17,18],diogo:36,dir:34,directli:[0,9,13,14,17,31,35,36],directori:[0,7,32,36],disabl:[0,17,18,35,36],disable_evaluator_output:[0,9,14,22],disable_existing_logg:7,disc:[0,35],discrimin:26,discriminant_analysi:26,discuss:32,disk:[7,15,36],displai:0,disregard:0,distanc:[3,9,28],distribut:[0,7,17,18,19,32,34,35,36],do_work:18,doc:[7,10,17,18,32,36],docker:36,dockerfil:36,dockerhub:[34,36],document:[0,7,15,33,34,35,36],doe:[0,10,17,18,25,36],domashnev:36,don:32,done:[8,17,18,19,32,34,35],down:[18,19],download:[1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,22,24,25,26,27,28,30,34,36],downsid:32,drawn:0,drive:[32,36],drop:[33,36],dropout:36,dtype:[2,8,9,13,35],dual:[1,9,20,21,22,28],due:[0,32,36],dummi:[4,36],dummyclassifi:36,dummyfutur:7,dummyregressor:36,duplic:33,durat:[0,1,2,3,4,9,15],dure:[0,9,14,32,35,36],e:[0,2,3,10,13,14,32,34,35,36],each:[0,8,9,10,17,18,19,31,32,35],early_stop:[1,4,9,20,21,22,28],easili:31,eddi:36,edg:36,edward:36,effici:[33,35],effort:32,eggensperg:[33,36],eigen:[26,28],either:[0,6,32,35],elasticnet:[4,28],element:0,elif:13,els:[0,8,13,22,26,28],emploi:35,empti:25,en:[17,18,19],enabl:[15,34,36],encod:[0,1,3,9,13,20,21,28,32,35,36],end:[0,9,11,18],end_tim:0,endtim:[9,15],enforc:[34,35,36],engelen:36,enough:0,enrico:36,ensembl:[0,7,9,15,20,22,28,33,36],ensemble_nbest:[0,17,18,35],ensemble_onli:0,ensemble_optimization_scor:[7,13],ensemble_s:[0,11,17,18,21,35],ensemble_test_scor:[7,13],ensemble_weight:[0,1,2,3,4],ensemblebuild:7,ensur:35,entri:[9,33],entropi:[1,20,21,22,28],enumer:9,environ:[32,34,35,36],epsilon:[1,4,9,20,21,22,28],epsilon_insensit:4,equal:[0,12,36],equalscondit:27,error:[0,6,18,35,36],error_add:12,error_r:[6,12],error_wk:12,especi:[32,35,36],essenti:34,estim:[0,4,7,13,24,27,28,32,33,35,36],et:[33,35],eta0:[4,9,11,22,28],eta:22,etc:9,euclidean:[1,3,21,22,28],evalu:[0,9,12,13,17,18,35,36],even:[7,15,32,36],everi:35,everyon:32,everyth:[17,18,32],everytim:32,exampl:[0,1,2,3,4,6,7,8,9,10,11,13,14,15,17,18,19,20,21,22,24,25,26,27,28,31,32,34,36],example_calc_multiple_metr:[6,16],example_classif:[1,5],example_debug_log:[7,16],example_extending_classif:[24,29],example_extending_data_preprocessor:[25,29],example_extending_preprocessor:[26,29],example_extending_regress:[27,29],example_feature_typ:[8,16],example_get_pipeline_compon:[9,16],example_inspect_predict:[10,16],example_interpretable_model:[11,16],example_metr:[12,16],example_multilabel_classif:[2,5],example_multioutput_regress:[3,5],example_pandas_train_test:[13,16],example_parallel_manual_spawning_cli:[17,23],example_parallel_manual_spawning_python:[18,23],example_parallel_n_job:[19,23],example_random_search:[20,23],example_regress:[4,5],example_resampl:[14,16],example_restrict_number_of_hyperparamet:[28,29],example_sequenti:[21,23],example_single_configur:[15,16],example_successive_halv:[22,23],examples_20_bas:5,examples_40_advanc:16,examples_60_search:23,examples_80_extend:29,examples_jupyt:30,examples_python:30,exce:32,exceed:[0,2,9,17,18,19,20,21,22,25],except:[0,18,20,22,36],exclud:[0,28,35],exclus:[31,35],execut:[0,5,16,17,18,23,29,32],exemplifi:[7,22,35],exercis:0,exhaust:35,exist:[29,30,31],exp:[1,3,20,21,22,28],expect:[0,32],experienc:33,experiment:[0,33,36],explain:[9,34,35],explained_variance_ratio_:9,explan:[16,30,34],explicit:17,explicitli:[14,22],ext:7,extend:[0,6,29,30,32,33,36],extra:0,extra_argu:12,extra_tre:[1,3,6,9,11,21,22,28],extra_trees_preproc_for_classif:[1,11,20,21,22,28],extra_trees_preproc_for_regress:[3,11],extratreesclassifi:36,f1:[0,6,12,32],f1_macro:[0,2,12],f1_micro:[0,12],f1_sampl:[0,12],f1_weight:[0,12],f:[2,10,12,17],f_classif:[1,21,22,28],f_regress:4,facad:[20,22],factor:36,factori:0,fail:[2,9,19,20,22,32,34,36],failur:[0,36],falkner:[33,36],fals:[0,1,2,3,4,6,7,9,11,12,13,14,15,17,18,20,21,22,24,25,26,27,28],familiar:0,faq:[33,36],fashion:9,fast_ica:[1,3,11,20,21,22,28],faster:36,favorit:32,fdr:28,feat_typ:[0,8,13,35],featur:[0,13,14,16,20,28,30,33,36],feature_agglomer:[1,3,11,21,22,28],feature_nam:10,feature_preprocess:[11,15,26,31],feature_preprocessor:[0,1,2,3,4,9,11,15,20,21,22,24,25,26,27,28,35],feature_typ:[1,2,3,4,9,11,15,20,21,22,24,26,27,28],featurepreprocessorchoic:[11,15],feedback:36,feedstock:34,felix:36,felixleungsc:36,fetch:2,fetch_openml:[2,7,8,10,13,15],feurer:[33,36],few:36,field:31,file:[0,5,7,16,17,18,23,29,34,35,36],filenam:[7,17,35],fill_valu:9,find:[0,9,17,18,31,32,36],fine:[0,32],finish:[0,9,18,21,32,35],first:[10,11,14,17,18,21,32,35],fit:[0,2,10,12,14,16,17,18,30,31,32,33,35,36],fit_ensembl:[0,17,18,21],fit_intercept:[1,4,9,11,20,21,22,25,28],fit_pipelin:[0,15],fit_prior:[9,20,21,28],fix:[32,36],flag:[2,7,8,24,25,26,27,28,35],flake8:36,flip:0,float64:[8,9,13],fold:[0,14],folder:0,follow:[0,1,3,4,6,7,8,9,10,11,12,14,15,17,24,25,26,27,28,31,32,33,34,35,36],forbidden:[3,28],forest:[15,35,36],fork:32,forkserv:[32,36],form:[0,15,31,35],format:[2,7,8,9,13,35,36],formatt:7,found:[0,10,12,13,15,25,32,35,36],fpr:[1,20,21,28],frac:0,fraction:[35,36],frame:0,francisco:36,francoi:36,frank:33,free:[17,33],friedman_ms:3,from:[0,2,3,6,7,8,9,10,11,13,15,19,20,22,23,24,25,26,27,28,30,33,34,35,36],from_dict:6,fu:36,full:[1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,22,24,25,26,27,28,31,32,35],fun:[1,3,20,21,22,28],further:[0,13,14,15,18,32,35],furthermor:[4,9,11,20],futur:[18,36],fwe:[4,20,22,28],g:[2,34],galleri:[1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,22,24,25,26,27,28,30,36],gamma:[1,3,9,20,21,27,28],gaussian_nb:[6,11,28],gaussian_process:3,gcc:34,gcc_linux:34,geffrier:36,gener:[0,1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,22,24,25,26,28,30,31,35,36],get:[0,9,15,20,22,33,34,35],get_compon:11,get_configuration_spac:[0,3,15,28],get_default_configur:22,get_estim:0,get_event_loop:18,get_hyperparameter_search_spac:[24,25,26,27,28],get_metric_result:6,get_models_with_weight:[0,9,25],get_param:0,get_preprocessor:0,get_properti:[24,25,26,27,28],get_random_search_object_callback:20,get_roar_object_callback:20,get_smac_object:22,get_smac_object_callback:[0,20,22],get_trials_callback:0,getting_start:15,geurt:28,gijsber:36,gini:[1,2,9,15,20,21,22,25,26,28],github:[7,15,31,32,33,34,36],give:[0,9,30,31,32,35,36],given:[0,10,11,15,17,32,36],glazachev:36,glob:7,gmean:36,go:32,goal:10,good:[0,2,4,13,32,35,36],got:12,gradient:36,gradient_boost:[1,4,6,9,11,20,21,22,28],gradientboostingclassifi:36,grain:[0,32],greater:0,greater_is_bett:[0,6,12],greil:36,grid:[13,35],grid_resolut:10,grow:32,guarante:32,guard:[17,18,19,20,22,32,36],gui:36,guid:[32,33,35,36],guidanc:[2,8],guidelin:[32,35,36],guilherm:36,gxx_linux:34,gyro_i:10,gyro_x:10,gyro_z:10,gyroscop:10,ha:[0,4,9,15,28,31,34,35,36],halv:[9,23,30,36],hand:[14,33],handl:[0,9,31,35,36],handler:7,handles_classif:[24,25,26,27,28,31],handles_multiclass:[24,25,26,27,28,31],handles_multilabel:[24,25,26,27,28,31],handles_multioutput:[24,25,26,27,28],handles_regress:[24,25,26,27,28,31],happen:[13,32,35,36],hard:[32,36],harder:35,harsh:0,hash:0,have:[0,9,10,14,15,17,18,31,32,34,35,36],header:[0,35],heavili:[32,34],hector:36,help:32,here:[1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,22,24,25,26,27,28,32,34,35],herilalaina:36,heurist:[28,36],hidden_layer_depth:[1,9,20,21,22,24,28],hidden_layer_s:24,high:0,higher:[0,4,10,32,36],highli:33,highlight:[7,8,13],hing:[1,20,21,22,25,28],histori:9,hnishi:36,hoc:9,hold:34,holdout:[0,22,32,35,36],home:[2,8],honor:[0,15],host:36,hostedtoolcach:22,hot:[0,32,35],hour:[33,35],how:[0,1,2,3,4,6,7,8,9,10,11,12,13,15,17,18,19,20,21,22,24,25,26,27,28,31,34,35,36],howev:[11,12,21,36],howto:7,html:[2,7,8,15,17,18,19],http:[2,7,8,13,15,17,18,19,36],hutter:33,hyper:15,hyperband:22,hyperparamet:[0,3,9,20,24,26,27,29,30,33,35,36],i:[0,3,9,10,13,14,24,32,35,36],id:[0,9,32],ideal:18,idempot:36,ident:24,identifi:[0,35],idl:17,ids_config:9,ignor:[9,17,18,36],illustr:[9,18,22],imag:36,immedi:36,impact:[10,36],implement:[0,2,7,8,10,24,25,26,28,31,32,35,36],impli:31,importances_mean:10,importances_std:10,improv:[15,32,33,36],imput:[1,2,3,4,9,11,15,20,21,22,24,26,27,28,32,35],includ:[0,9,11,15,22,24,25,26,27,32,33,35,36],include_bia:[1,3,4,11,15,21,22,28],incompat:[0,34,36],incondit:26,increas:[0,32,36],independ:32,index:[0,6,10,19,36],indic:[2,13,32],individu:[0,7,14,35],inf:4,infer:[8,35],influenc:0,info:[7,36],inform:[0,2,7,10,16,17,18,22,30,32,33,34,35,36],inherit:32,initi:[0,9,35,36],initial_budget:22,initial_configur:22,initial_configurations_via_metalearn:[0,2,7,8,20,24,25,26,27,28,35],initial_incumb:20,inproceed:33,input:[0,10,13,24,25,26,27,28,30,31,36],insensit:8,insert:0,insid:[9,14,31,32,36],inspect:[4,10,11,30,32,36],inspir:0,inspur:36,instal:[0,32,33,36],instanc:[0,9,15,20,31,36],instance_id:9,instance_specif:15,instead:[0,7,17,20,22,28,32,34,35,36],instruct:[34,36],int32:2,int64:2,integ:[0,2,32,35],intend:[20,22],intensif:[0,22],intensifi:22,intensifier_kwarg:22,interact:[9,10,22,34],interaction_onli:[1,3,4,11,15,21,22,28],intercept_sc:[1,9,20,21,22,28],interest:[7,9,32],interfac:[10,17],intermedi:7,intern:[7,9,17,18,25,36],interpret:[10,16,30,35],introduc:[0,20,36],invok:[20,22],involv:35,invscal:[4,11,28],io:15,ip:34,ipynb:[1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,22,24,25,26,27,28],is_determinist:[24,25,26,27,28,31],is_valid_configur:15,isaac:36,issu:[25,33,34,35,36],item:[0,9,25],iter:[0,2,7,9,19,20,24,32,36],iterative_cv:36,its:[17,18,31,32,35],itself:[2,8,9,17,18,32,33],iver:36,j:36,jaidev:36,jeong:36,jesper:36,jha:36,jianswang:36,jin:36,jirka:36,jj:36,job:[0,7,17,18,19,36],joblib:32,joel:36,join:[12,17,18],joint:36,jongheon:36,jordal:36,jose:36,joseph:36,josh:36,jost:[33,36],journal:33,json:[7,17],judgement:32,julian:36,jupyt:[1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,22,24,25,26,27,28,30,34,36],just:0,k:[0,4,14],k_nearest_neighbor:[3,9,11,28],kaa:36,karl:36,katharina:[33,36],keep:[0,9,32,35],keep_vari:[1,3,9,20,28],kei:[0,6,9,25,35],kept:[0,35],kernel:[0,1,3,9,20,21,25,28],kernel_pca:[3,11,20,25,28],kernel_ridg:27,kernelridg:27,kernelridgeregress:27,kind:[10,13,35],kitchen_sink:[3,11,28],klein:[33,36],know:[32,33,34],known:32,kobylkin:36,kp:[7,15],kr:[7,15],kwarg:[0,25],l1:[20,21,22,28],l1_ratio:[4,9,28],l2:[1,9,11,20,21,28],l2_regular:[1,4,9,20,21,22,28],label:[0,4,5,10,13,30,36],labelencod:[0,35],lambda_1:4,lambda_2:4,laptop:32,larg:[35,36],larger:[32,36],largest:10,lastli:32,later:0,latest:[17,18,19,33,34,36],latter:0,launch:0,laurent:36,lbfg:24,lda:[9,11,20,25,28,36],lead:[32,36],leaderboard:[0,1,2,3,4,36],learn:[0,2,9,10,13,15,17,18,21,31,33,35,36],learning_r:[1,4,9,11,20,21,22,28],learning_rate_init:[1,9,20,21,22,28],least:[34,36],least_squar:4,left:0,leftov:36,legend:[4,13,35],len:[0,9,20,22,25],length:0,less:36,lesw:36,let:[9,32],leung:36,level:[7,32,36],levelnam:7,leverag:33,lg:33,liac:36,lib:22,liblinear_svc:[1,6,9,11,20,21,28],liblinear_svc_preprocessor:[11,20,21,22,28],librari:[0,32,35],libsvm_svc:[1,6,9,11,21,28],licens:36,lifetim:[17,18],light:[17,18],like:[0,11,22,31,32,35],limit:[0,2,9,15,17,18,19,20,21,22,25,34,36],lindauer:[33,36],line:[10,13,18,19,23,30,32,35],linear:[9,26,32,36],lineardiscriminantanalysi:26,link:[2,36],linkag:[1,3,21,22,28],linkcheck:36,linux:34,list:[0,8,9,11,13,32,34,35],liter:0,ll:10,load:[0,32,36],load_breast_canc:[1,6,9,11,12,14,17,18,19,20,21,22,24,25,26,28,36],load_diabet:[4,27],load_digit:[33,36],load_model:0,local:36,localclust:18,locat:[17,32],lockfil:36,log2:28,log:[0,3,11,12,16,22,27,28,30,36],log_loss:[0,12],logcosh:[3,28],logger:[0,7,36],logging_config:[0,7,36],logging_configur:7,logic:[17,18],logist:24,loglevel:36,logloss:36,longer:[32,36],look:[9,10,15,31,32,35],loop:18,loss:[0,1,4,9,11,20,21,22,25,28,32,36],losses_and_configur:9,lot:[32,36],love:32,low:0,lower:[4,24,26,27,32],lowest:9,lsqr:[26,28],lt:36,luca:36,luka:36,m:28,mabri:36,machin:[0,15,17,18,21,23,30,31,33,34,35,36],made:36,mae:[3,4],mai:[0,2],main:[32,36],maint:36,maintain:[9,32],make:[0,2,15,20,22,25,32,34,35,36],make_regress:3,make_scor:[0,6,12],manag:[17,18,19,35,36],manhattan:[1,3,21,28],mani:[0,32,35],manner:[32,34],manual:[13,17,18,20,28,31,36],manuel:[33,36],map:[0,2],mariu:[33,36],mark:36,mask:[0,9],masked_arrai:9,master:[15,34,36],matplotlib:[4,10,13],matric:31,matrix:0,matthia:[33,36],max:[3,28],max_bin:[1,4,9,20,21,22,28],max_budget:22,max_depth:[1,2,3,4,9,15,20,21,22,25,26,28],max_depth_factor:[3,9,20,28],max_featur:[1,2,3,4,9,15,20,21,22,25,26,28],max_it:[1,9,21,28],max_leaf_nod:[1,2,3,4,9,15,20,21,22,25,26,28],max_models_on_disc:[0,32,35],maxim:[9,35,36],maximilian:36,maximum:[0,28,35],mb:[0,5,16,23,29,32],md5:0,mean:[0,1,2,3,4,6,9,12,14,18,20,21,22,24,26,27,28,31,32,36],mean_absolute_error:[0,12],mean_fit_tim:9,mean_squared_error:[0,12,36],mean_squared_log_error:12,mean_test_scor:[6,9],measur:[13,35],mechan:[0,13,32,35],median:[1,3,11,15,20,21,22,28],median_absolute_error:[0,12],memori:[0,2,9,17,18,19,20,21,22,25,34,36],memory_limit:[0,15,17,18,19,36],mendoza:36,mention:32,merg:33,messag:[7,36],meta:[33,35,36],metadata:[0,35,36],metadata_directori:0,metafeatur:36,metalearn:0,metalearning_configur:[20,22],method:[0,8,15,31,32,35,36],metric:[1,2,3,4,7,8,9,11,13,14,15,16,17,18,19,20,21,22,24,25,26,27,30,32,33,35,36],metric_:6,metric_balanced_accuraci:6,metric_custom_error:6,metric_f1:6,metric_nam:12,metric_precis:6,metric_recal:6,mfeurer:34,michael:36,microsoft:34,might:0,min_chal:22,min_impurity_decreas:[1,2,3,4,9,15,20,21,22,25,26,28],min_samples_leaf:[1,2,3,4,9,15,20,21,22,25,26,28],min_samples_split:[1,2,3,4,9,15,20,21,22,25,26,28],min_weight_fraction_leaf:[1,2,3,4,9,15,20,21,22,25,26,28],minim:[9,36],minimum:[0,35,36],minimum_fract:[1,2,3,4,9,11,15,20,21,22,24,26,27,28],minmax:[1,3,4,9,20,21,22,28],minor:36,minority_coalesc:[1,2,3,4,9,11,15,20,21,22,24,26,27,28],minr:20,minut:[1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,22,24,25,26,27,28,35,36],miotto:36,miss:[32,34,35,36],mkdir:34,mkl_num_thread:[32,35],ml:36,mlp:[1,6,9,11,20,21,22,28],mlpclassifi:[24,36],mlpregressor:36,mode:[1,4,19,20,21,22,28,32,35,36],model:[0,6,12,13,14,15,16,19,20,21,22,30,31,36],model_1:0,model_id:[0,1,2,3,4],model_n:0,model_select:[0,1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,22,24,25,26,27,28,33,36],modern:35,modifi:[2,8],modified_hub:[9,28],modul:[2,10,15,17,18,34,36],mohd:36,morakakam:36,more:[0,2,7,8,9,13,17,21,22,32,33,34,35,36],moreov:6,most:[0,9,32,35],most_frequ:[1,3,4,9,20,21,22,28],mostli:[32,35],move:[32,36],mse:3,msg:7,much:[0,11,32],multi:[0,5,7,30,36],multi_class:[1,9,20,21,22,28],multiclass:[0,1,2,9,11,20,21,22,24,25,26,31,35,36],multiclass_classif:[17,18],multidimension:35,multilabel:[0,1,2,9,11,20,21,22,24,25,26,31,35,36],multinomial_nb:[11,28],multioutput:[0,3,4,27,35],multioutputregressor:0,multipl:[6,10,13,17,18,19,31,32,35,36],multiprocess:[7,17,18,35,36],multitud:33,must:[0,13,17,18,19,24,31,32,34,35,36],mutual:31,mutual_info:[1,20,22,28],mutual_info_classif:[20,21,24,25,28],mybind:36,mypi:36,n:[9,10,12,35],n_class:0,n_cluster:[1,3,21,22,28],n_compon:[3,20,25,28],n_estim:[1,3,9,15,20,21,22,28],n_featur:[0,3],n_inform:3,n_iter:4,n_iter_no_chang:[1,9,20,21,22,28],n_job:[0,15,17,18,19,20,22,35,36],n_jobs_:19,n_label:0,n_neighbor:[3,9,28],n_output:0,n_quantil:[1,3,4,9,15,20,21,22,28],n_repeat:10,n_sampl:[0,3],n_samples_fit:0,n_split:0,n_target:[0,3],n_worker:18,nader:36,name:[0,2,6,7,8,9,11,12,13,15,17,18,19,20,21,22,24,25,26,27,28,31,32,33,35,36],named_step:[9,15],nanni:18,natur:0,nb:34,ndarrai:[0,35],necessari:[0,14,17,32,35],need:[0,9,14,17,18,31,32,34,36],needs_proba:[0,6,12],needs_threshold:[0,6,12],neg:[0,2,31],neither:36,nest:0,neural:33,neural_network:24,neurip:36,neurips15a:33,never:18,nevertheless:[32,35],next:35,nicer:0,nildaimon:36,nip:33,no_coalescens:[1,3,4,9,20,21,22,28],no_encod:[1,3,4,9,15,20,21,22,28],no_preprocess:[1,2,3,4,11,20,21,22,24,25,27,28,35],non:36,none:[0,1,2,3,4,7,9,11,12,15,17,20,21,22,24,25,26,27,28,31,35,36],nopreprocess:28,nor:36,normal:[1,3,9,15,20,21,22,28,32,35],note:[0,2,8,10,11,17,32,35,36],notebook:[1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,22,24,25,26,27,28,30,32,34,36],noth:36,notimplementederror:[24,26,27,28],now:[4,10,11,17,18,32,36],np:[2,6,8,12,13,14,15,35],nthread:[17,18,20,22],nuanc:32,num_nodes_per_lay:[1,9,20,21,22,24,28],num_param:0,num_run:[9,15],num_work:22,number:[0,2,9,10,17,18,19,20,21,22,25,29,30,35,36],numer:[0,8,13,35],numerical_transform:[1,2,3,4,9,11,15,20,21,22,24,26,27,28],numpi:[0,2,3,6,8,12,13,14,15,32,35,36],nystroem_sampl:[3,11,20,28],object:[0,3,9,13,15,20,24,26,27,28,31,32,35,36],observ:28,obtain:[0,16,30,35,36],obviou:0,obvious:14,occur:34,off:[4,22,25,28,35],offici:[7,34],offlin:36,often:32,old:36,olehb007:36,oli:36,omp:36,omp_num_thread:[32,35],onc:[0,18,19,34,36],one:[7,9,14,15,17,18,19,21,28,31,32,33,35,36],one_hot_encod:[1,2,3,4,9,11,20,21,22,24,26,27,28],ones:[2,15,30,32],onli:[0,9,11,15,17,18,28,31,35,36],onlin:20,open:[32,33],openblas_num_thread:[32,35],openml:[2,7,8,10,13,32],openmp:32,oper:[34,36],opposit:36,opt:[22,34],optim:[0,6,9,11,17,18,19,20,22,28,30,32,33,35,36],optimizatino:20,optimum:[0,6,12],option:[0,7,32,34,35],optionalarrai:0,order:[0,13,17,18,31,35,36],ordereddict:9,ordin:13,ordinalencod:35,org:[2,7,8,13,15,17,18,19],origin:[32,36],osx:36,other:[0,9,18,32,34,35],otherwis:[0,22],ouput:36,our:[15,28,32,33],ourselv:27,out:[0,1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,22,24,25,26,27,28,31,33,34],outcom:0,output:[0,5,7,9,14,24,25,26,27,28,30,31,36],output_directori:32,output_distribut:[1,3,4,9,15,20,21,22,28],outsid:32,over:[0,6,8,9,11,14,16,30,32,35,36],overestim:4,overfit:[0,13,32],overhead:36,overview:0,ovr:[1,9,20,21,22,28],own:[17,18,32],p:[3,9,28,34],packag:[15,22,32,34,36],page:[33,36],pair:35,panda:[0,6,8,13,30,35,36],paper:[32,33,36],parallel:[0,1,3,20,21,22,23,28,30,36],parallel_schedul:22,param:[0,9],param_balanc:9,param_classifi:[6,9],param_data_preprocessor:9,param_feature_preprocessor:9,paramet:[0,15,25,36],parametr:28,part:[34,36],parti:[32,36],partial:0,particip:36,particular:[9,35],particularli:22,pass:[0,7,15,17,18,20,32,35,36],passive_aggress:[1,6,9,11,20,21,22,25,28],path:[0,7,36],pathlib:7,pca:[1,3,9,11,20,28,35],pd:[0,6,13,35],penalti:[1,4,9,11,20,21,22,28],pep8:36,pepe:36,per:[0,6,8,9,13,35],per_run_time_limit:[0,1,2,3,4,6,8,9,10,11,12,13,14,15,17,18,19,20,21,22,24,26,27,28,36],percentag:35,percentil:[1,20,21,22,28],perceptron:28,perfect:4,perform:[0,8,15,16,30,32,35,36],performance_over_time_:[0,13,35],perman:0,permutation_import:10,person:10,peter:36,phone:10,pick:33,picklabl:36,pid_:0,pieter:36,pip3:[34,36],pip:[34,36],pipe:17,pipelin:[0,7,9,11,15,24,25,26,27,28,31,32,35,36],pitfal:10,plai:36,platform:34,pleas:[0,17,18,20,22,31,32,33,34,35],plot:[0,8,16,30,35,36],plot_partial_depend:10,plt:[4,10,13,35],point:[0,4,17,18],poli:[3,9,20,28],polynomi:[1,3,4,11,15,21,22,27,28],pooling_func:[1,3,21,22,28],port:[17,34],posit:[2,31],possibl:[0,4,6,8,12,14,15,20,21,22,27,32,34,35,36],post:9,pot:13,potenti:[32,36],power:36,power_t:[4,9,11,28],power_transform:[3,9,20,28],pr:36,practic:[14,30,32],pradeep:36,pre:32,precis:[0,2,6,12,32,36],precision_macro:[0,12],precision_micro:[0,12],precision_sampl:[0,12],precision_weight:[0,12],precomput:0,predefin:14,predefinedsplit:14,predict:[0,1,2,3,6,8,10,11,12,13,14,17,18,20,21,22,24,28,31,32,33,35,36],predict_proba:[0,24,28,36],predictor:[31,36],preprocess:[0,36],preprocessor:[0,9,29,30,31,35,36],present:[0,36],preserv:7,pretti:[14,32],prevent:36,previou:21,previous:[0,17,18,36],princip:9,print:[6,7,8,9,10,13,14,15,17,18,19,20,22,33,35,36],prior:[0,35,36],prioriti:35,probabl:[0,10,36],problem:[0,2,6,9,12,32,34,35,36],proc:17,procedur:[30,35],procedurc:9,process:[0,17,18,21,31,33,34,36],process_cli_work:17,process_python_work:[17,18],processor:0,produc:[31,36],progress:35,properli:[2,8,13,17,18,35],properti:31,proport:0,provid:[0,2,7,8,13,17,24,25,26,27,28,32,34,35,36],prune:[0,35,36],publish:33,pull:34,purpos:[7,13,14,18,22,31],push:36,put:0,pwd:34,py:[1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,35,36],pyninsh:36,pypi:[32,36],pyplot:[4,10,13],pyrfr:[32,34,36],python3:[22,34,36],python:[1,2,3,4,6,7,8,9,10,11,12,13,14,15,19,20,21,22,23,24,25,26,27,28,30,32,34,35,36],q_max:[1,3,4,9,20,21,22,24,28],q_min:[1,3,4,9,20,21,22,24,28],qda:[9,11,20,28,36],quantifi:4,quantile_transform:[1,3,4,9,15,20,21,22,28],queri:[9,31],question:36,r2:[0,3,4,12,27],r2_score:[0,3,4,27],r:[0,10,28],raamana:36,race:20,rafael:36,rais:[0,20,22,24,26,27,28,35,36],rakotoarison:36,ram:32,random:[9,15,23,30,31,32,35,36],random_forest:[1,2,3,4,6,9,11,15,20,21,22,25,26,28,35],random_numb:0,random_st:[1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,22,24,26,27,28,33],random_trees_embed:[3,11,20,22,28],randomforest:15,randomforestclassifi:[15,28,36],randomli:[10,15],rang:[3,4,13,17,18,24,26,27,28,35],rank:[0,1,2,3,4,36],rank_test_scor:[6,9],rather:32,ratio:9,rbf:[1,3,21,27,28],re:[2,14,32,34,36],reach:32,read:[0,32,33,34,36],readi:18,readm:36,real:[2,7,8,17,18,24,25,26,27,28],reason:[32,35],recal:[0,6,12],recall_macro:[0,12],recall_micro:[0,12],recall_sampl:[0,12],recall_weight:[0,12],recent:[33,34],recogn:[8,36],recommend:[2,7,8,15,24,25,26,27,28,34,35],reddi:36,redirect:0,reduc:[3,32,36],redund:32,refer:[0,13,33,35,36],refit:[0,22,32,36],reg:[27,32],reg_param:[9,20,28],regard:35,regist:[31,36],registri:36,regress:[5,11,12,29,30,32,35,36],regression_metr:12,regressor:[0,27,31,35,36],regular:[33,35],rel:[0,36],relat:[10,18,36],releas:[2,8,32],relev:[17,22,32,36],reli:34,relu:[1,20,21,22,24,28],remain:9,remov:[0,20,22,32,36],renam:36,render:[17,18,36],replac:[0,2,12,28,33,34,36],report:33,repositori:32,repres:0,represent:[0,31],reproduc:36,request:32,requir:[0,2,17,18,32,35,36],rerun:32,resampl:[0,16,30,36],resampling_strategi:[0,14,22],resampling_strategy_argu:[0,14,22,36],rescal:[1,2,3,4,9,11,15,20,21,22,24,26,27,28],research:33,residu:0,resort:0,resourc:[32,34,35],respect:32,respons:10,restrict:[9,11,27,29,30,32,36],restructur:36,result:[0,2,4,6,9,17,18,19,20,21,22,25,30,31,33,36],retain:36,retriev:[18,36],return_x_i:[1,2,4,6,7,9,11,12,13,14,15,17,18,19,20,21,22,24,25,26,27,28,33],returncod:17,reuter:2,revers:[17,18],review:2,rf:28,rh2epm:22,rivera:36,rng:[20,22],roar_facad:20,robust:[33,35,36],robust_scal:[1,3,4,9,20,21,22,24,28],roc_auc:[0,12,36],rohit:36,rong:36,root:[7,34],root_mean_squared_error:12,rotat:10,round:36,run:[0,1,3,4,6,7,8,10,11,12,13,14,15,16,17,18,19,20,21,22,24,25,26,27,28,30,32,33,34,35,36],run_id:[20,22],run_info:[0,15],run_kei:9,run_or_walk_inform:10,run_until_complet:18,run_valu:[0,9,15],runcount_limit:[2,7,8,24,25,26,27,28],runhistori:[0,9,12,22],runhistory2epm4logcost:[12,22],runhistory2epm:[12,22],runhistory_:9,runinfo:[0,15],runkei:9,runner:[2,8],runtim:[9,32],runvalu:[0,9,15],ryul:36,s:[0,7,9,10,12,14,17,22,32,34,35,36],safe:[2,8],safeguard:36,sake:[17,18],same:[0,9,28,31,32,33],samm:[20,28],sampl:[0,4,10,15,35],sample_configur:15,sample_weight:0,santo:36,sapronov:36,satyam:36,save:0,scalar:[0,8],scale:[3,27,28,36],scatter:4,scenario:[0,20,22],scenario_dict:[0,20,22],schedul:18,scheduler_address:18,scheduler_fil:17,scheduler_file_nam:17,scheduler_ip:18,scientif:[33,35],scikit:[0,2,9,10,13,15,31,33,35,36],scipi:[0,35,36],score:[0,9,10,12,17,18,19,20,22,28,32,33,35,36],score_func:[0,1,4,6,12,20,21,22,24,25,28],scorer:0,scoring_funct:[0,6],scratch:0,script:[1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,22,24,25,26,27,28,32,36],search:[0,15,23,32,34,35,36],searchspac:32,second:[0,1,2,3,4,6,7,8,9,10,11,13,14,15,17,18,19,20,21,22,24,25,26,27,28,32,35,36],section:[12,32,34,35,36],see:[0,10,11,17,18,31,32,34,36],seed:[0,9,12,15,17,18,19,20,22,31],seen:[0,35],select:[0,9,14,15,22,33,35,36],select_percentile_classif:[1,11,20,21,22,28],select_percentile_regress:11,select_rates_classif:[1,11,20,21,22,24,25,28],select_rates_regress:[4,11],selectpercentileregress:36,selectr:36,self:[0,6,12,17,18,24,25,26,27,28],sensibl:0,separ:36,sequenc:36,sequenti:[23,30,35],seri:[0,35],server:36,servic:34,session:34,set:[0,13,14,17,18,22,25,34,35,36],set_param:0,setattr:25,setup:[18,36],sever:[31,32,35,36],sgd:[4,6,9,11,22,24,28,36],shahril:36,shape:[0,28],share:[35,36],shell:[17,34],ship:34,shortnam:[24,25,26,27,28,31],should:[0,9,17,18,32,33,36],show:[1,2,3,4,6,7,9,10,12,13,14,15,17,18,19,20,21,35,36],show_model:[0,1,2,3,4,9,11,20,21,22,24,25,26,27,35],showcas:13,shown:[11,13,35],shrink:[1,9,21,28],shrinkag:[9,20,25,26,28,36],shrinkage_condit:26,shrinkage_factor:[9,20,28],shuffl:[0,1,9,20,21,22,28,36],shut:[18,19],shutdown:36,sigmoid:[3,20,28],sign:[0,1,2,3,4,9,11,20,21,22,24,25,26,27,31],signatur:[0,36],signed_data:[24,26,27,31],significantli:32,silenc:[2,8],silva:36,similar:11,simpl:[0,1,4,10,32,35],simpleclassificationpipelin:[1,2,9,11,20,21,22,24,25,26],simpleregressionpipelin:[3,4,27],simpli:22,simplifi:[9,32,36],sinc:[0,10,12],singl:[0,11,14,16,17,18,23,30,32,35,36],single_best_optimization_scor:13,single_best_test_scor:13,single_best_train_scor:13,singlethreadedcli:7,site:22,six:36,size:[0,11,35,36],skearn:25,sklean:32,sklearn:[0,6,7,8,10,12,13,14,15,19,20,29,30,36],skripniuk:36,sleep:[17,18],smac3:[7,34,36],smac4ac:22,smac:[0,7,9,12,13,20,22,35,36],smac_ac_facad:22,smac_scenario_arg:[0,2,7,8,24,25,26,27,28],small:32,smaller:12,smbo:[0,36],so:[0,9,10,17,18,31,32,34,35,36],soft:14,solut:[6,12,32,34],solv:[9,34],solver:[1,9,20,21,22,24,26,28],some:[0,10,25,32,35,36],sorber:36,sort:[0,9,36],sort_bi:0,sort_idx:10,sort_ord:0,sort_valu:0,sourc:[0,1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,22,24,25,26,27,28,30,31,32,35],source_id:15,space:[0,7,15,24,27,31,35,36],spars:[0,1,2,3,4,9,11,20,21,22,24,25,26,27,28,31,35,36],sparsiti:0,spawn:[19,23,30,32,35,36],specif:[8,32,34],specifi:[0,2,7,8,14,22,28,31,32,35],speed:[2,7,8,10,24,25,26,27,28,33,36],spend:33,sphinx:[1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,22,24,25,26,27,28,30,36],split:[0,14,30,35,36],splitter:36,spmatrix:0,springenberg:[33,36],sprint_statist:[0,2,9,17,18,19,20,21,22,25,32,35],sqrt:28,squar:0,squared_hing:[1,9,20,21,22,28],stabl:[2,15],stage:9,stage_nam:9,stale:36,standard:[1,2,3,4,9,11,20,21,22,24,26,27,28],start:[0,7,9,10,19,32,33,34,36],start_python_work:18,start_tim:0,startswith:6,starttim:[9,15],startup:36,state:36,statement:[18,36],staticmethod:[24,25,26,27,28],statist:[0,19,20,22,35],statu:[0,6,9,15],statustyp:[9,13,15],stderr:17,stdout:[7,17],stefan:[33,36],step:[0,25,32,35],still:[32,35],stock90975:36,stop:[0,9,18,32,35,36],store:[0,7,9,15,17,32,35,36],str:[0,13,28,31],strategi:[0,1,2,3,4,9,11,15,16,20,21,22,24,25,26,27,28,30,36],stream:7,streamhandl:7,streuhof:36,strict:34,string:[0,2,8,14,32,35],strong:36,structur:7,style:36,subclass:[0,31],submit:[17,18,36],subobject:0,subprocess:[17,18,32,36],subsampl:[10,36],subsect:31,subset:[0,11,32],success:[0,2,6,9,15,17,18,19,20,21,23,25,30,35,36],successive_halv:22,sudo:34,suffer:[13,32],suffici:35,suggest:[32,34,35],sum:0,support:[0,12,13,34,36],sure:[2,15,20,22,25,32,34,36],svd:[26,28],svm:3,swap:35,swig:[32,34],sy:7,symptom:32,synthet:3,system:[33,35,36],t:[10,12,32],ta:[0,20,22],ta_kwarg:[20,22],tabl:0,tae:[0,13],tae_runn:[20,22],tae_runner_kwarg:[20,22],tag:[0,8],take:[0,17,19,35],tane:36,tanh:[1,9,20,21,22,24,28],target:[0,2,8,9,10,13,17,18,19,20,21,22,25,35,36],target_typ:[1,2,3,4,9,11,20,21,22,24,25,26,27],task:[0,1,2,3,4,9,11,15,17,18,20,21,22,24,25,26,27,31,32,35],task_id:9,technic:32,techniqu:36,technolog:33,tell:31,templat:36,tempor:7,temporari:[32,36],temporarili:32,teresa:36,termin:[0,32],test:[0,4,10,13,15,32,35,36],test_fold:14,test_predict:4,test_siz:[10,13,15],testa:36,th:9,than:[0,4,21,22,32,36],thei:[17,18,32,35,36],them:[0,10,11,17,35,36],therebi:32,therefor:[0,32,34,35],thetal:3,thetau:3,thi:[0,1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,22,24,25,26,27,28,31,32,33,34,35,36],thick:10,thin:10,third:32,thread:[35,36],threadpoolctl:[35,36],threads_per_work:18,three:[17,18,32],threshold_lambda:4,through:[20,31],throughout:35,thu:[10,32],tie:36,tight:32,tight_layout:[4,10],time:[0,1,2,3,4,6,7,8,9,10,11,12,14,15,17,18,19,20,21,22,24,25,26,27,28,30,31,32,33,36],time_left_for_this_task:[0,1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,22,24,25,26,27,28],timeout:[17,36],timeseri:35,timestamp:[7,13,35],timothi:36,titl:[13,33,35],tmp:[0,1,3,4,10,11,14,17,18,19,20,21,22,32,36],tmp_dir:0,tmp_directori:32,tmp_folder:[0,1,3,4,7,10,11,14,17,18,19,20,21,22],to_numer:13,to_numpi:8,to_str:6,tobia:36,togeth:[0,15,36],tol:[1,4,9,11,20,21,22,25,26,28],too:[9,32,36],toolkit:33,top:[0,9],top_k:0,total:[0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,32,35],track:[0,36],tracker:33,tradeoff:35,train:[0,1,4,9,10,13,14,15,20,21,22,28,30,32,35,36],train_loss:[0,9,15],train_predict:4,train_siz:[0,14,22],train_test_split:[1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,22,24,25,26,27,28,33],trajectori:[32,36],transform:[0,9,25,26,31,35,36],travi:36,tree:32,tri:0,truncatedsvd:11,tune:[28,33,35,36],tupl:[0,9,15,24,31,36],turn:[10,25,35,36],two:[2,7,8,10,18,21,24,25,26,27,28,31,32,34,35,36],txt:36,type:[0,1,2,3,4,9,13,16,22,24,26,27,28,30,31,35,36],type_of_target:2,typic:0,typo:36,u12:9,u13:9,u16:9,u18:9,u19:9,u20:9,u32:9,u3:9,u5:9,u9:9,u:0,ubuntu:36,un:36,under:[7,32,34],underestim:4,underli:[0,9],understand:10,unidimension:35,uniform:[1,3,4,9,20,21,22,28],uniform_averag:0,uniformfloat:[3,24,26,27,28],uniformfloathyperparamet:[24,26,27,28],uniforminteg:[3,24,27,28],uniformintegerhyperparamet:[24,27,28],union:0,unit:36,unix:34,unless:0,unlik:33,unlimit:36,unnecessari:36,unseen:[13,14,35],unsign:31,unsigned_data:[24,25,26,27,28,31],unsuccess:35,until:[18,35],up:[2,7,8,9,10,11,17,18,24,25,26,27,28,33,34,36],updat:[0,32,34,36],upgrad:36,upper:[0,24,26,27,35,36],us:[0,2,7,8,10,11,13,14,15,17,18,19,21,24,28,30,31,33,34,35,36],usag:[23,30,35,36],user:[0,8,11,13,31,32,33,35,36],userwarn:22,usual:[32,34],util:[0,2,26],v:[0,34],val:25,valentin:36,valid:[0,1,2,9,15,17,18,19,20,21,25,28,30,32,35,36],validation_fract:[1,9,20,21,22,28],valu:[0,3,4,9,10,15,28,31,35,36],valueerror:[20,22],valverd:36,van:36,variabl:[0,32,35,36],varianc:9,variou:[0,12],ve:33,vector:36,veri:[11,32],verifi:34,version:[0,7,32,34],via:[0,1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,22,24,25,26,27,28,32,33,34,35,36],vicent:36,view:0,virtual:34,visual:[4,36],vladimir:36,vladislav:36,vote:14,vs:[7,15],wa:[0,9,15,18,34,36],wai:[2,8,32,33],wait:18,walk:[10,31],want:[8,9,17,18,32,33],ward:[1,3,21,22,28],warm_start:15,warn:[2,8,12,22,36],we:[4,7,9,10,11,12,13,14,15,17,18,19,27,28,32,33,34,35],web:36,weight:[0,1,3,9,15,17,18,20,21,22,24,25,28],weight_1:0,weight_n:0,weird:36,well:[0,9,11,34,35,36],wen:36,were:[0,4,9,25,32,36],what:[0,9,10],wheel:[34,36],when:[0,2,8,10,13,17,18,25,31,32,35,36],where:[0,14,22,28,34,35,36],wherea:13,whether:[0,9,10,13,31,36],which:[0,7,9,10,11,12,13,14,17,18,21,22,28,31,33,35,36],whiten:[1,3,9,20,21,22,28],whole:[14,32],why:32,window:36,wish:2,within:[13,17,23,30,32,36],without:[17,32,35,36],woo:36,work:[0,2,7,8,13,15,17,18,31,32,33,34,35,36],worker:[0,19,22,23,30,32,35,36],worker_process:[17,18],workflow:36,workload:[17,18],workstat:32,wors:0,worst:36,worst_possible_result:0,would:[0,11,14,17,32,33,35,36],wrap:[0,32],wrapper:[9,26,31,35],written:[7,32],wrong:36,wrongli:36,wrt:[0,10],www:[2,7,8,13],x64:22,x:[0,1,2,3,4,6,7,8,9,11,12,13,14,15,17,18,19,20,21,22,24,25,26,27,28,33,35],x_test:[0,1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,22,24,25,26,27,28,33,35],x_train:[1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,22,24,25,26,27,28,33,35],xgboost:36,xiaodong:36,xlabel:4,xlim:4,xtick:10,y:[0,1,2,3,4,6,7,8,9,11,12,13,14,15,17,18,19,20,21,22,24,25,26,27,28,33,36],y_hat:33,y_optim:0,y_pred:[0,24,25,26,27],y_test:[0,1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,22,24,25,26,27,28,33,35,36],y_train:[1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,22,24,25,26,27,28,33,35],y_true:0,yaml:[0,36],yassin:36,yazanobeidi:36,year:33,yet:[0,20,32,35],yield:28,ylabel:4,ylim:4,you:[0,2,7,8,9,13,15,17,18,31,32,33,34,35],young:36,your:[1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,22,24,25,26,27,28,32,34,35],yourself:32,zero:[2,35],zeyi:36,zip:30,zorder:4},titles:["APIs","Classification","Multi-label Classification","Multi-output Regression","Regression","Computation times","Metrics","Logging and debugging","Feature Types","Obtain run information","Model Explanation","Interpretable models","Metrics","Performance-over-time plot","Resampling Strategies","Fit a single configuration","Computation times","Parallel Usage: Spawning workers from the command line","Parallel Usage: Spawning workers from within Python","Parallel Usage  on a single machine","Random Search","Sequential Usage","Successive Halving","Computation times","Extending Auto-Sklearn with Classification Component","Extending Auto-Sklearn with Data Preprocessor Component","Extending Auto-Sklearn with Preprocessor Component","Extending Auto-Sklearn with Regression Component","Restricting the number of hyperparameters for an existing component","Computation times","Examples","Extending auto-sklearn","FAQ","auto-sklearn","Installation","Manual","Releases"],titleterms:{"0":[17,32,33,36],"1":[9,10,17,32,36],"10":36,"11":36,"12":36,"13":36,"14":36,"2":[9,10,17,32,33,36],"3":[10,17,36],"4":36,"5":36,"6":36,"7":36,"8":36,"9":36,"default":32,"final":[1,2,3,4,6,8,11,13,14,21],"function":18,"import":[10,17],"new":33,Will:32,about:[2,9],accuraci:[12,24],ad:32,addit:12,advanc:30,again:14,all:[17,32],also:22,an:[15,22,28],anaconda:34,api:0,approach:22,ar:[17,32],argument:12,auto:[1,2,3,4,9,11,17,18,21,22,24,25,26,27,28,31,32,33,34,35],avail:[11,12,32],background:[17,18],basic:30,best:9,budget:22,build:[1,2,3,4,6,7,8,9,10,11,13,19,21,22,35],built:0,callback:22,can:[22,32],cite:33,classif:[0,1,2,11,24,31],classifi:[1,2,6,7,8,9,11,13,19,20,21,22,24,28],client:17,close:17,command:17,commun:17,compat:34,compon:[9,24,25,26,27,28,31],comput:[5,10,16,23,29,35],conda:34,condit:10,config:7,configur:[3,15,26,28],construct:[1,2,3,4,11,21],contain:32,contribut:33,contributor:36,crash:32,creat:[7,10,17,24,25,26],cross:[14,22],custom:12,data:[1,2,3,4,6,7,8,9,10,11,12,13,14,15,19,20,21,22,24,25,27,28,32,35],dataset:[25,26,32],debug:7,defin:[15,18,22],depend:10,detail:9,disabl:32,disk:32,docker:34,doe:32,dummi:32,eat:32,ensembl:[1,2,3,4,6,8,11,13,14,21,32,35],error:12,estim:15,evalu:32,exampl:[12,30,33,35],exclud:32,exist:28,expect:10,explan:10,extend:[24,25,26,27,31],extens:[0,30],extrem:32,faq:32,fault:32,featur:[8,10,31,32,35],fifth:12,file:32,first:12,fit:[1,3,4,6,7,8,9,11,13,15,19,20,21,22,24,25,26,27,28],forest:28,forg:34,found:[1,2,3,4,9],fourth:12,from:[17,18,32],gener:[27,32],get:[1,2,3,4,6,8,11,13,14,21],get_hyperparameter_search_spac:31,get_properti:31,halv:22,holdout:14,how:32,hungri:32,hyperparamet:28,ic:10,imag:34,implement:27,individu:10,inform:9,input:35,inspect:[9,35],instal:34,instanti:22,interfac:0,interpret:[11,32],issu:32,iter:22,kernel:27,krr:27,label:2,lda:26,learn:[14,32],licens:33,limit:[32,35],line:17,list:12,load:[1,2,3,4,6,7,8,9,10,11,12,13,14,15,19,20,21,22,24,28],log:[7,32],mac:34,machin:[19,32],main:0,manual:[33,35],memori:[32,35],meta:32,metric:[0,6,12],mix:22,mlp:24,model:[1,2,3,4,9,10,11,25,26,27,32,35],modul:0,more:10,multi:[2,3],my:32,next:22,non:32,nopreprocess:25,number:[28,32],object:14,obtain:9,one:10,onli:32,osx:34,output:[3,32],over:13,oversubscript:32,own:12,parallel:[17,18,19,32,35],part:[9,10],partial:10,pd:10,pdp:10,perform:[13,14,25],permut:10,persist:35,plot:[4,10,13],predefin:12,predict:[4,9,25,26,27],preprocess:[25,31,32,35],preprocessor:[11,25,26],print:[1,2,3,4,11,12,21,24,25,26,27,28],process:[32,35],provid:15,python:[17,18],random:[20,28],refit:14,regress:[0,3,4,27,31],regressor:[3,4],releas:36,report:9,requir:34,resampl:[14,35],restrict:[28,35],result:35,ridg:27,roar:20,run:[2,9],s:28,schedul:17,scikit:[14,32],score:[1,2,3,4,6,8,11,13,14,21,25,26,27],search:[9,20,30],searchspac:35,second:12,see:22,segment:32,sequenti:[21,32],set:32,setup:17,show:11,singl:[15,19],sklearn:[1,2,3,4,9,11,17,18,21,22,24,25,26,27,28,31,32,33,34,35],space:[3,26,28,32],spawn:[17,18],splitter:14,start:[17,18],statement:17,statist:[2,9,24,25,26,27],strategi:[14,35],subclass:28,subsampl:22,success:22,successivehalv:22,support:35,system:34,test:24,than:10,third:12,time:[5,13,16,23,29,35],two:17,type:8,ubuntu:34,until:17,up:32,us:[9,12,20,22,26,27,32],usag:[17,18,19,21,32],user:15,v0:36,valid:[14,22],vanilla:35,version:36,view:[1,2,3,4],wait:17,we:22,where:32,which:32,window:34,within:18,without:25,worker:[17,18],write:31,x:36}})
\ No newline at end of file
+Search.setIndex({docnames:["api","examples/20_basic/example_classification","examples/20_basic/example_multilabel_classification","examples/20_basic/example_multioutput_regression","examples/20_basic/example_regression","examples/20_basic/sg_execution_times","examples/40_advanced/example_calc_multiple_metrics","examples/40_advanced/example_debug_logging","examples/40_advanced/example_feature_types","examples/40_advanced/example_get_pipeline_components","examples/40_advanced/example_inspect_predictions","examples/40_advanced/example_interpretable_models","examples/40_advanced/example_metrics","examples/40_advanced/example_pandas_train_test","examples/40_advanced/example_resampling","examples/40_advanced/example_single_configuration","examples/40_advanced/sg_execution_times","examples/60_search/example_parallel_manual_spawning_cli","examples/60_search/example_parallel_n_jobs","examples/60_search/example_random_search","examples/60_search/example_sequential","examples/60_search/example_successive_halving","examples/60_search/sg_execution_times","examples/80_extending/example_extending_classification","examples/80_extending/example_extending_data_preprocessor","examples/80_extending/example_extending_preprocessor","examples/80_extending/example_extending_regression","examples/80_extending/example_restrict_number_of_hyperparameters","examples/80_extending/sg_execution_times","examples/index","extending","faq","index","installation","manual","releases"],envversion:{"sphinx.domains.c":2,"sphinx.domains.changeset":1,"sphinx.domains.citation":1,"sphinx.domains.cpp":4,"sphinx.domains.index":1,"sphinx.domains.javascript":2,"sphinx.domains.math":2,"sphinx.domains.python":3,"sphinx.domains.rst":2,"sphinx.domains.std":2,"sphinx.ext.viewcode":1,sphinx:56},filenames:["api.rst","examples/20_basic/example_classification.rst","examples/20_basic/example_multilabel_classification.rst","examples/20_basic/example_multioutput_regression.rst","examples/20_basic/example_regression.rst","examples/20_basic/sg_execution_times.rst","examples/40_advanced/example_calc_multiple_metrics.rst","examples/40_advanced/example_debug_logging.rst","examples/40_advanced/example_feature_types.rst","examples/40_advanced/example_get_pipeline_components.rst","examples/40_advanced/example_inspect_predictions.rst","examples/40_advanced/example_interpretable_models.rst","examples/40_advanced/example_metrics.rst","examples/40_advanced/example_pandas_train_test.rst","examples/40_advanced/example_resampling.rst","examples/40_advanced/example_single_configuration.rst","examples/40_advanced/sg_execution_times.rst","examples/60_search/example_parallel_manual_spawning_cli.rst","examples/60_search/example_parallel_n_jobs.rst","examples/60_search/example_random_search.rst","examples/60_search/example_sequential.rst","examples/60_search/example_successive_halving.rst","examples/60_search/sg_execution_times.rst","examples/80_extending/example_extending_classification.rst","examples/80_extending/example_extending_data_preprocessor.rst","examples/80_extending/example_extending_preprocessor.rst","examples/80_extending/example_extending_regression.rst","examples/80_extending/example_restrict_number_of_hyperparameters.rst","examples/80_extending/sg_execution_times.rst","examples/index.rst","extending.rst","faq.rst","index.rst","installation.rst","manual.rst","releases.rst"],objects:{"autosklearn.classification":{AutoSklearnClassifier:[0,0,1,""]},"autosklearn.classification.AutoSklearnClassifier":{fit:[0,1,1,""],fit_ensemble:[0,1,1,""],fit_pipeline:[0,1,1,""],get_configuration_space:[0,1,1,""],get_models_with_weights:[0,1,1,""],get_params:[0,1,1,""],leaderboard:[0,1,1,""],predict:[0,1,1,""],predict_proba:[0,1,1,""],refit:[0,1,1,""],score:[0,1,1,""],set_params:[0,1,1,""],show_models:[0,1,1,""],sprint_statistics:[0,1,1,""]},"autosklearn.experimental.askl2":{AutoSklearn2Classifier:[0,0,1,""]},"autosklearn.experimental.askl2.AutoSklearn2Classifier":{fit:[0,1,1,""],fit_ensemble:[0,1,1,""],fit_pipeline:[0,1,1,""],get_configuration_space:[0,1,1,""],get_models_with_weights:[0,1,1,""],get_params:[0,1,1,""],leaderboard:[0,1,1,""],predict:[0,1,1,""],predict_proba:[0,1,1,""],refit:[0,1,1,""],score:[0,1,1,""],set_params:[0,1,1,""],show_models:[0,1,1,""],sprint_statistics:[0,1,1,""]},"autosklearn.metrics":{accuracy:[0,2,1,""],average_precision:[0,2,1,""],balanced_accuracy:[0,2,1,""],f1:[0,2,1,""],f1_macro:[0,2,1,""],f1_micro:[0,2,1,""],f1_samples:[0,2,1,""],f1_weighted:[0,2,1,""],log_loss:[0,2,1,""],make_scorer:[0,3,1,""],mean_absolute_error:[0,2,1,""],mean_squared_error:[0,2,1,""],median_absolute_error:[0,2,1,""],precision:[0,2,1,""],precision_macro:[0,2,1,""],precision_micro:[0,2,1,""],precision_samples:[0,2,1,""],precision_weighted:[0,2,1,""],r2:[0,2,1,""],recall:[0,2,1,""],recall_macro:[0,2,1,""],recall_micro:[0,2,1,""],recall_samples:[0,2,1,""],recall_weighted:[0,2,1,""],roc_auc:[0,2,1,""]},"autosklearn.pipeline.components.base":{AutoSklearnClassificationAlgorithm:[0,0,1,""],AutoSklearnPreprocessingAlgorithm:[0,0,1,""],AutoSklearnRegressionAlgorithm:[0,0,1,""]},"autosklearn.pipeline.components.base.AutoSklearnClassificationAlgorithm":{get_estimator:[0,1,1,""],predict:[0,1,1,""],predict_proba:[0,1,1,""]},"autosklearn.pipeline.components.base.AutoSklearnPreprocessingAlgorithm":{get_preprocessor:[0,1,1,""],transform:[0,1,1,""]},"autosklearn.pipeline.components.base.AutoSklearnRegressionAlgorithm":{get_estimator:[0,1,1,""],predict:[0,1,1,""]},"autosklearn.pipeline.components.classification":{add_classifier:[30,3,1,""]},"autosklearn.pipeline.components.feature_preprocessing":{add_preprocessor:[30,3,1,""]},"autosklearn.pipeline.components.regression":{add_regressor:[30,3,1,""]},"autosklearn.regression":{AutoSklearnRegressor:[0,0,1,""]},"autosklearn.regression.AutoSklearnRegressor":{fit:[0,1,1,""],fit_ensemble:[0,1,1,""],fit_pipeline:[0,1,1,""],get_configuration_space:[0,1,1,""],get_models_with_weights:[0,1,1,""],get_params:[0,1,1,""],leaderboard:[0,1,1,""],predict:[0,1,1,""],refit:[0,1,1,""],score:[0,1,1,""],set_params:[0,1,1,""],show_models:[0,1,1,""],sprint_statistics:[0,1,1,""]}},objnames:{"0":["py","class","Python class"],"1":["py","method","Python method"],"2":["py","attribute","Python attribute"],"3":["py","function","Python function"]},objtypes:{"0":"py:class","1":"py:method","2":"py:attribute","3":"py:function"},terms:{"0":[0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,18,19,20,21,22,23,24,25,26,27,28,31,33,34],"00":[5,12,16,22,28],"000000":6,"00001":26,"000010":12,"0001":[3,9,23,25,27],"00010495766342828794":24,"0001350687148423338":[1,20],"0001363185819149026":21,"0001384744065375479":19,"00014375616988222174":9,"00015788887334393714":9,"00018009776276177523":21,"00021148999718383549":[1,20],"0002346515712987664":21,"0002600768160857831":21,"0003474820560182061":9,"0003701926442639788":[0,4],"000421568792103947":1,"0004668987145850572":11,"0004684917334431039":[1,20],"0005012365297609799":4,"0005024616279807072":19,"0006046872822704108":9,"0006085470273398056":9,"0006517033225329654":[0,4],"0007452270241186694":[1,20],"000896391087588558":19,"001":[3,10,27],"0010303157708058924":1,"0011804284312897009":[1,20],"0012624032013298571":4,"0015130257264171173":9,"00172583":9,"0017940473175767063":[1,20],"0017964371726429177":19,"0018055343233337954":21,"0018475934154961215":19,"002":10,"002431731981071206":4,"002766772136115771":9,"003":10,"0030875":9,"003142456770095119":19,"003640536588885903":21,"00431052":9,"004635634934629583":9,"005069923784737444":9,"005326508887463406":1,"005821344021856017":19,"0059319":9,"006":10,"0067563":9,"00684163":9,"007":10,"007017543859649145":9,"007092198581560294":[11,19],"007608275318572964":19,"0076293":9,"008":[14,16],"008015420020402715":9,"00858029":9,"00894849":9,"00948519":9,"01":[3,5,7,9,12,16,19,21,22,26,27,28],"010000000000000004":9,"010000000000000014":11,"01006215":9,"010232681621243814":19,"0104432":9,"011333066835975528":9,"012150149892783745":4,"01223987":9,"012391313886912093":9,"01257378":9,"012607824914758717":4,"013399889539803811":19,"014034416932373166":19,"014184":[1,6],"014184397163120588":[1,20,21],"015340961719445064":[1,20],"01589975":9,"016444224834275295":4,"01667385805274091":19,"01684553":9,"018038793247120707":19,"01851537":9,"02":[1,5,9,16,19,21,22],"020000000000000004":20,"021276595744680882":[1,19,20,21,25],"021277":[1,6],"021944240404849075":4,"02258105767523384":24,"023":10,"02462169":9,"025360750168211e":4,"02641737e":9,"027708640006448e":21,"028368794326241176":[1,9,19,20,21],"028369":[1,6],"02847755502162456":1,"03":[9,12],"03033175355450235":15,"03125":27,"03162215674470446":21,"03208923":9,"032840833359764676":[1,20],"032863849765258205":21,"03286384976525822":21,"0333981795670635e":9,"03409213e":9,"03457277874230573":9,"035211267605633784":21,"035211267605633804":21,"03546099290780147":[1,9,19,20,21,24],"035461":[1,6],"035543606049804645":19,"036":10,"03737258e":9,"03986668":9,"039906103286385":21,"04":[1,4,9,16,19,21,28,35],"04000000000000001":20,"0400632553169558e":19,"04074":[31,32],"041":12,"04210526315789476":9,"042553":[1,6],"04255319148936165":[1,19,20,21],"043879389796003546":19,"044":[11,16],"04561403508771933":9,"04634380160611007":21,"04662763358613446":19,"04694835680751174":21,"046948356807511755":21,"047641289117234975":15,"04865917":9,"049645":[1,6],"049645390070921946":[1,19,20,21,23],"05":[1,3,4,9,11,16,19,21,24,26,27],"05082904256838701":21,"051":10,"05164319248826292":21,"0517578125e":[3,27],"054648061621061846":23,"05491437530740365":21,"05673758865248224":[1,9,19,20,21],"056738":[1,6],"057778875694463e":21,"0590467":9,"05972079854295879":21,"06":[1,3,4,5,9,16,19,20,21,28],"060000000000000005":20,"060800813211425456":1,"063":12,"06315789473684208":9,"06332899":9,"06334655":9,"06382978723404253":[1,19,20],"063830":[1,6],"06437417":9,"0647401999412075e":[1,20],"06666666666666665":9,"07":[4,9,12,19,27],"0703733":9,"07063243":9,"07092198581560283":[1,9,19,20],"070922":[1,6],"07213284":9,"078014":6,"07801418439716312":[9,19,21],"07827509e":9,"08":[1,3,4,9,12,16,19,21,27],"08000000000000002":20,"08087614244138486":9,"08215962441314555":21,"08291320147381159":[1,20],"08300813783286698":21,"085106":[1,6],"08510638297872342":[1,19],"09":[4,5,9,22,27],"090":12,"09111888":9,"09122807017543855":9,"09179220974965213":21,"09219858156028371":[1,19,20],"092199":1,"0945814167023392e":21,"09750328007832798":21,"09809101":9,"09809681787962342":[1,20],"09814953":9,"099290780141844":[19,21],"0995499108996053":19,"0th":9,"0x7fc78438b070":20,"0x7fc78438b220":20,"0x7fc7843a12e0":21,"0x7fc7843b7d60":21,"0x7fc7843c0100":21,"0x7fc784547220":21,"0x7fc7845473a0":21,"0x7fc784850e80":21,"0x7fc7848717c0":21,"0x7fc784871e50":21,"0x7fc784a023a0":19,"0x7fc784b361f0":19,"0x7fc784b36370":19,"0x7fc784b39ee0":19,"0x7fc784b3b400":20,"0x7fc784bbe130":19,"0x7fc784bd3a00":21,"0x7fc784bd90d0":19,"0x7fc784bf10a0":21,"0x7fc784bf13a0":21,"0x7fc784bf14f0":19,"0x7fc784bf19d0":21,"0x7fc784bf1d00":21,"0x7fc784c08520":21,"0x7fc784c08610":21,"0x7fc784c0a520":21,"0x7fc784c0adf0":21,"0x7fc784c102e0":20,"0x7fc784d702b0":21,"0x7fc784d70430":21,"0x7fc784e47160":20,"0x7fc784e47520":20,"0x7fc784ec5c40":19,"0x7fc784f51070":20,"0x7fc784f7f190":19,"0x7fc784f7fa00":19,"0x7fc785087190":21,"0x7fc785087220":21,"0x7fc785087df0":21,"0x7fc78509f550":19,"0x7fc78509f910":21,"0x7fc78509f9a0":19,"0x7fc78509fd30":21,"0x7fc78509fe80":21,"0x7fc7850a2760":20,"0x7fc7850a2c70":20,"0x7fc7850a8850":19,"0x7fc7850a8be0":21,"0x7fc7850b5d30":21,"0x7fc7850b5e50":21,"0x7fc785684b50":21,"0x7fc7857b5a30":20,"0x7fc7857f88e0":20,"0x7fc7857f8c40":20,"0x7fc78580d130":19,"0x7fc78580d250":19,"0x7fc78580dac0":21,"0x7fc785814430":21,"0x7fc785820970":21,"0x7fc785820a90":21,"0x7fc78582e1f0":19,"0x7fc78582e3d0":21,"0x7fc78582e940":21,"0x7fc78582ed30":21,"0x7fc78582ee20":19,"0x7fc78582efa0":21,"0x7fc785869040":21,"0x7fc785872910":20,"0x7fc785b6a0a0":19,"0x7fc785b6ac70":19,"0x7fc785b7d5e0":20,"0x7fc785b7d6a0":20,"0x7fc785b7e400":21,"0x7fc785b7e4c0":21,"0x7fc785b994f0":19,"0x7fc785b99af0":19,"0x7fc785c0eb50":19,"0x7fc785c0eb80":19,"0x7fc785e313d0":19,"0x7fc785e49970":19,"0x7fc785e4f4f0":20,"0x7fc785eda250":21,"0x7fc785eda430":21,"0x7fc785edabe0":21,"0x7fc785fb2160":20,"0x7fc7860954f0":19,"0x7fc786095a90":21,"0x7fc786095eb0":20,"0x7fc78612c280":20,"0x7fc78612c5b0":20,"0x7fc7861490d0":19,"0x7fc786149370":19,"0x7fc786149910":19,"0x7fc786184070":19,"0x7fc786184130":21,"0x7fc7861841f0":19,"0x7fc786184580":20,"0x7fc7863feee0":26,"0x7fc78648b5b0":21,"0x7fc78648be80":21,"0x7fc7864a4700":21,"0x7fc7864b8130":21,"0x7fc7864b81f0":21,"0x7fc786554580":19,"0x7fc786564c10":21,"0x7fc786567e80":19,"0x7fc78656e670":21,"0x7fc78656e760":21,"0x7fc78657f1c0":19,"0x7fc78657f370":19,"0x7fc786581c40":19,"0x7fc7865df100":19,"0x7fc786625e20":19,"0x7fc78662d580":19,"0x7fc78662deb0":19,"0x7fc78662fb20":21,"0x7fc78662fc40":21,"0x7fc786633bb0":21,"0x7fc78666a6d0":20,"0x7fc786687c10":9,"0x7fc786796cd0":19,"0x7fc78679fe50":19,"0x7fc7867d8340":24,"0x7fc7867e4640":21,"0x7fc786806670":19,"0x7fc786854070":19,"0x7fc786854430":23,"0x7fc7868546d0":21,"0x7fc786854e20":23,"0x7fc78687a1f0":19,"0x7fc786887dc0":19,"0x7fc786899b50":21,"0x7fc786899d90":21,"0x7fc7868a1220":21,"0x7fc7868a13a0":21,"0x7fc78690f400":21,"0x7fc78690ffd0":21,"0x7fc786912460":21,"0x7fc786912b20":21,"0x7fc786918190":20,"0x7fc786918580":20,"0x7fc7869195e0":21,"0x7fc78698af70":19,"0x7fc7869b8e80":21,"0x7fc786a909a0":20,"0x7fc786a9d310":20,"0x7fc786aaaf10":19,"0x7fc786abb100":21,"0x7fc786abb970":20,"0x7fc786abba00":21,"0x7fc786b361f0":9,"0x7fc786b36910":9,"0x7fc786b74100":20,"0x7fc786b860d0":21,"0x7fc786b94340":21,"0x7fc786b94460":21,"0x7fc786bb7a90":19,"0x7fc786bd1b50":19,"0x7fc786bff640":19,"0x7fc786c78730":20,"0x7fc786c845e0":19,"0x7fc786ed1af0":21,"0x7fc786ed1c10":21,"0x7fc786ed8610":21,"0x7fc786edd8b0":21,"0x7fc786fe30a0":21,"0x7fc786fe33d0":21,"0x7fc786fe3550":21,"0x7fc786fe3af0":21,"0x7fc786ff0970":21,"0x7fc786ff0ca0":21,"0x7fc787005400":20,"0x7fc7870056d0":20,"0x7fc78700b100":19,"0x7fc787011e50":19,"0x7fc7870329a0":20,"0x7fc78706ef70":19,"0x7fc78714dfd0":21,"0x7fc7871571c0":21,"0x7fc7871573a0":21,"0x7fc787157910":21,"0x7fc78715d310":21,"0x7fc78715d3a0":21,"0x7fc78715d850":21,"0x7fc78715d9a0":21,"0x7fc78715d9d0":21,"0x7fc78715dfd0":21,"0x7fc7871ccb50":21,"0x7fc7871e5280":21,"0x7fc7871e5400":21,"0x7fc7871ff3d0":19,"0x7fc7871ff9a0":21,"0x7fc7871ffbe0":21,"0x7fc787282b20":19,"0x7fc78728db80":21,"0x7fc78728de20":21,"0x7fc7872bcd30":21,"0x7fc7872bce20":21,"0x7fc787302940":19,"0x7fc78733d670":19,"0x7fc787368bb0":20,"0x7fc787374400":19,"0x7fc787374fd0":19,"0x7fc7873bf130":20,"0x7fc7873f3460":21,"0x7fc78753b430":21,"0x7fc78753bb20":21,"0x7fc78753bcd0":21,"0x7fc78753bfa0":21,"0x7fc78758e4c0":21,"0x7fc7875b5640":21,"0x7fc7875b5970":21,"0x7fc7875c2c40":21,"0x7fc78764e7c0":21,"0x7fc78764e8b0":21,"0x7fc7876620a0":19,"0x7fc7876a5520":21,"0x7fc7876a55b0":21,"0x7fc7876ba3d0":19,"0x7fc7876d02b0":21,"0x7fc7876d0310":21,"0x7fc7876d8b80":21,"0x7fc7876f1ac0":21,"0x7fc7876f1be0":21,"0x7fc78770aca0":19,"0x7fc78770b160":19,"0x7fc78770ba60":19,"0x7fc78770be20":19,"0x7fc787711040":21,"0x7fc7877111c0":21,"0x7fc787711670":21,"0x7fc787742160":21,"0x7fc787748250":19,"0x7fc787748400":19,"0x7fc78776fac0":19,"0x7fc787772ca0":19,"0x7fc787788bb0":19,"0x7fc787788ca0":19,"0x7fc787793460":19,"0x7fc7877938b0":19,"0x7fc78779a040":19,"0x7fc7877aefa0":19,"0x7fc7877cd460":19,"0x7fc7877cdaf0":19,"0x7fc7877de790":21,"0x7fc787842790":19,"0x7fc787888340":19,"0x7fc787888970":21,"0x7fc787888d00":19,"0x7fc7878b0160":21,"0x7fc7878b0700":21,"0x7fc7878b1850":21,"0x7fc7878b19d0":21,"0x7fc7878b5580":21,"0x7fc78790c850":20,"0x7fc7879e2c10":20,"0x7fc787a63bb0":21,"0x7fc787a63dc0":21,"0x7fc787a7a5b0":21,"0x7fc787a7a7f0":21,"0x7fc787a7a910":21,"0x7fc787a7afa0":21,"0x7fc787a7e670":21,"0x7fc787a8deb0":19,"0x7fc787a9e310":20,"0x7fc787aa5220":19,"0x7fc787ae89d0":19,"0x7fc787ba2040":19,"0x7fc787ba29a0":19,"0x7fc787ba92b0":21,"0x7fc787bb7430":21,"0x7fc787bb7550":21,"0x7fc787c50460":21,"0x7fc787c50580":21,"0x7fc787c79550":21,"0x7fc787c7a280":21,"0x7fc787c7a3d0":21,"0x7fc787e052e0":21,"0x7fc787e05310":19,"0x7fc787e05a30":19,"0x7fc787e1a9a0":20,"0x7fc787e35280":21,"0x7fc787e35400":21,"0x7fc787e35f70":19,"0x7fc787e47400":19,"0x7fc787e47580":19,"0x7fc787e52070":19,"0x7fc787e608e0":21,"0x7fc787e609a0":21,"0x7fc787e71580":20,"0x7fc787ec5b20":21,"0x7fc7880f9eb0":19,"0x7fc78a25e7c0":19,"0x7fc78a25eb20":19,"0x7fc78a3702e0":21,"0x7fc78a370be0":21,"0x7fc78a378070":20,"0x7fc78a38fc40":20,"0x7fc78a38fd30":20,"0x7fc78a4be100":19,"0x7fc78a4bed00":21,"0x7fc78a4c8340":19,"0x7fc78a593c70":9,"0x7fc78a593d90":21,"0x7fc78a593e50":20,"0x7fc78a5d3580":4,"0x7fc78a5d3d60":9,"0x7fc78a5eb790":9,"0x7fc78a5f4f70":20,"0x7fc78a6050d0":21,"0x7fc78a605820":19,"0x7fc78a6db2b0":21,"0x7fc78a791250":19,"0x7fc78a791d30":20,"0x7fc78a791d60":19,"0x7fc78a7d53a0":21,"0x7fc78a7d9bb0":20,"0x7fc78a7e35b0":[9,21],"0x7fc78a7f2c70":19,"0x7fc78a826520":9,"0x7fc78a826580":26,"0x7fc78a826a00":4,"0x7fc78a86d100":19,"0x7fc78a86d9a0":19,"0x7fc78a86deb0":21,"0x7fc78a892370":9,"0x7fc78a8d7610":2,"0x7fc78a8d7790":1,"0x7fc78a8d7a90":1,"0x7fc78a956520":21,"0x7fc78a956c40":21,"0x7fc78a9cc160":21,"0x7fc78a9ea4f0":21,"0x7fc78a9eab80":23,"0x7fc78aac0730":19,"0x7fc78aae48b0":15,"0x7fc78ab3f070":20,"0x7fc78ab3f340":1,"0x7fc78ab3f550":20,"0x7fc78ab3f670":9,"0x7fc78ab77d60":20,"0x7fc78ab79cd0":15,"0x7fc78aba4cd0":1,"0x7fc78acf7d00":1,"0x7fc78ad54a00":1,"0x7fc78ad91f70":1,"0x7fc78adc2d30":1,"0x7fc78adce8b0":1,"0x7fc78adfc730":1,"0x7fc78af3f1c0":21,"0x7fc78af3f910":25,"0x7fc78af79460":19,"0x7fc78af79550":19,"0x7fc78af79730":21,"0x7fc78af797f0":19,"0x7fc78af89d60":19,"0x7fc78af90700":19,"0x7fc78b21c7c0":21,"0x7fc78b21c880":21,"0x7fc78b2344c0":20,"0x7fc78b2451f0":20,"0x7fc78b257790":19,"0x7fc78b2743a0":19,"0x7fc78b289160":9,"0x7fc78b2896a0":23,"0x7fc78b3ce220":19,"0x7fc78b3d35e0":19,"0x7fc78b418430":21,"0x7fc78b44ee50":1,"0x7fc78b48c700":4,"0x7fc78b498ac0":19,"0x7fc78b498c70":21,"0x7fc78b4c3280":21,"0x7fc78b4f2310":20,"0x7fc78b503820":4,"0x7fc78b503880":4,"0x7fc78b503dc0":19,"0x7fc78b503fd0":21,"0x7fc78b5451f0":1,"0x7fc78b545430":1,"0x7fc78b58aa90":1,"0x7fc78b6561c0":19,"0x7fc78b656a90":21,"0x7fc78b6bc040":1,"0x7fc78b714ca0":1,"0x7fc78b73b580":21,"0x7fc78b753850":1,"0x7fc78b75a130":1,"0x7fc78b75a580":1,"0x7fc78b79d940":1,"0x7fc78b7d9340":21,"0x7fc78b7d9970":21,"0x7fc78b800b20":20,"0x7fc78b820eb0":1,"0x7fc78b836370":1,"0x7fc78b836a90":1,"0x7fc78b85f6d0":1,"0x7fc78b915d00":21,"0x7fc78b9260a0":19,"0x7fc78b926c40":19,"0x7fc78b929ac0":9,"0x7fc78b94c490":19,"0x7fc78b9c7250":1,"0x7fc78ba1bdc0":1,"0x7fc78ba1bee0":1,"0x7fc78ba251f0":1,"0x7fc78ba373a0":21,"0x7fc78ba37610":21,"0x7fc78ba3ddf0":1,"0x7fc78ba74a90":1,"0x7fc78baa93d0":21,"0x7fc78bad4130":21,"0x7fc78bae9040":20,"0x7fc78bc2d7c0":21,"0x7fc78bc70c10":20,"0x7fc78bc74550":1,"0x7fc78bcec7c0":20,"0x7fc78bcf7910":9,"0x7fc78bcf7dc0":9,"0x7fc78bd04670":1,"0x7fc78bd13160":19,"0x7fc78be25d90":1,"0x7fc78be43c70":20,"0x7fc78be46b20":1,"0x7fc78be46c10":1,"0x7fc78beb3ac0":21,"0x7fc78beb7b50":21,"0x7fc78beb7c40":21,"0x7fc78bff5f70":1,"0x7fc78c057550":21,"0x7fc78c057850":21,"0x7fc78c099940":21,"0x7fc78c0a7940":21,"0x7fc78c0ad400":21,"0x7fc78c0e9b50":21,"0x7fc78c1023d0":21,"0x7fc78c1f30a0":1,"0x7fc78c1f37f0":1,"0x7fc78c1f3c70":1,"0x7fc78c345ee0":1,"0x7fc78c384be0":20,"0x7fc78c384d90":25,"0x7fc78c3b75e0":1,"0x7fc78c3e6490":1,"0x7fc78c414be0":1,"0x7fc78c43e280":19,"0x7fc78c43e580":21,"0x7fc78c43e5b0":19,"0x7fc78c482100":20,"0x7fc78c482160":11,"0x7fc78c482280":4,"0x7fc78c482310":20,"0x7fc78c4826a0":4,"0x7fc78c482ca0":21,"0x7fc78c482eb0":11,"0x7fc78c4ed730":1,"0x7fc78c4f2e80":21,"0x7fc78c4f9700":24,"0x7fc78c53a5e0":20,"0x7fc78c53a9d0":20,"0x7fc78c53c5e0":21,"0x7fc78c53cb80":21,"0x7fc78c5839d0":19,"0x7fc78c59d910":24,"0x7fc78c5a0cd0":21,"0x7fc78c5a0e80":21,"0x7fc78c5a1e80":19,"0x7fc78c5ab8b0":21,"0x7fc78c5ab9a0":21,"0x7fc78c5ada90":21,"0x7fc78c5d4250":19,"0x7fc78c5d43d0":19,"0x7fc78c5fa5b0":1,"0x7fc78c667580":1,"0x7fc78c6cef10":1,"0x7fc78c70c0a0":1,"0x7fc78c752cd0":19,"0x7fc78c864820":1,"0x7fc78c883070":1,"0x7fc78c883580":1,"0x7fc78c8a0250":25,"0x7fc78c8b7970":21,"0x7fc78c8f8a60":19,"0x7fc78c8f8bb0":19,"0x7fc78c90c400":9,"0x7fc78c90c6d0":9,"0x7fc78c9171f0":19,"0x7fc78c917340":19,"0x7fc78c917c10":1,"0x7fc78c920430":3,"0x7fc78c920640":3,"0x7fc78c93b3a0":21,"0x7fc78c93bdc0":1,"0x7fc78c93bf10":[21,23],"0x7fc78c944160":20,"0x7fc78c9441c0":9,"0x7fc78c944760":1,"0x7fc78c9447c0":21,"0x7fc78c944ca0":20,"0x7fc78c944df0":21,"0x7fc78c94d1f0":19,"0x7fc78c94d400":4,"0x7fc78c94dc10":9,"0x7fc78c98fe20":24,"0x7fc78c9bf490":19,"0x7fc78c9bff70":19,"0x7fc78c9c7700":19,"0x7fc78ca210d0":3,"0x7fc78ca212e0":3,"0x7fc78ca21670":2,"0x7fc78ca21940":19,"0x7fc78ca21b20":21,"0x7fc78ca21f10":2,"0x7fc78ca46460":20,"0x7fc78ca86640":9,"0x7fc78ca86700":21,"0x7fc78ca86b50":21,"0x7fc78ca8e9a0":1,"0x7fc78ca98190":1,"0x7fc78ca98610":1,"0x7fc78cae7460":1,"0x7fc78cb34910":1,"0x7fc78cb3c880":1,"0x7fc78cb3ca30":1,"0x7fc78f69f220":20,"0x7fc78f69fd90":19,"0x7fc78f6db9a0":26,"0x7fc78f7b94c0":3,"0x7fc78f7b94f0":3,"0x7fc78f7b95e0":1,"0x7fc78f7b9640":3,"0x7fc78f7b9f10":3,"0x7fc78f80b160":4,"0x7fc78f80b610":4,"0x7fc78f80b910":11,"0x7fc78f83e280":19,"0x7fc78f83e5e0":19,"0x7fc78f84d070":24,"0x7fc78f84d1f0":19,"0x7fc78f84daf0":24,"0x7fc78f8a2b80":21,"0x7fc78f8d10a0":3,"0x7fc78f8d1970":9,"0x7fc78f913a00":4,"0x7fc78f913df0":4,"0x7fc78f92cc70":23,"0x7fc78f986610":19,"0x7fc78f986940":20,"0x7fc78fa69730":1,"0x7fc78fa69a30":1,"0x7fc78fa97670":19,"0x7fc78fa97970":19,"0x7fc78fa980a0":19,"0x7fc78fa984f0":19,"0x7fc78fa98670":19,"0x7fc78face370":19,"0x7fc78fad64f0":19,"0x7fc78fad68e0":19,"0x7fc78fad6d60":19,"0x7fc78fae4fd0":19,"0x7fc78faf4670":19,"0x7fc78faf4e50":19,"0x7fc78fb05610":20,"0x7fc78fb2b040":19,"0x7fc78fbb9b50":1,"0x7fc78fc1c340":21,"0x7fc78fc1c760":21,"0x7fc78fc1f100":20,"0x7fc78fc1f700":21,"0x7fc78fc1fc40":20,"0x7fc78fd0ac70":1,"0x7fc78fd0ae80":1,"0x7fc78fdee5e0":24,"0x7fc78fdee8b0":21,"0x7fc78fdeec10":24,"0x7fc78fe1ea00":1,"0x7fc78fe73970":1,"0x7fc78fe73d60":1,"0x7fc78febd310":21,"0x7fc78febd7c0":20,"0x7fc78feca6a0":1,"0x7fc78ff433d0":19,"0x7fc78ff43730":19,"0x7fc78ff43b20":21,"0x7fc78ff456d0":19,"0x7fc78ff934c0":20,"0x7fc78ffee460":15,"0x7fc7900401f0":1,"0x7fc790040550":1,"0x7fc790049a90":20,"0x7fc790129ac0":1,"0x7fc79015a340":[1,24],"0x7fc790169d30":9,"0x7fc79016ebe0":1,"0x7fc790171700":19,"0x7fc790179040":19,"0x7fc7901c2250":1,"0x7fc7901c53d0":9,"0x7fc7901c5820":9,"0x7fc7901c5dc0":20,"0x7fc7901eaf70":1,"0x7fc79023cd90":1,"0x7fc790279550":1,"0x7fc790288640":1,"0x7fc790288bb0":1,"0x7fc7902bd4f0":20,"0x7fc7902d58b0":1,"0x7fc7902d5a30":1,"0x7fc79032af10":9,"0x7fc7903465e0":9,"0x7fc7903592e0":21,"0x7fc790362c40":1,"0x7fc790366ac0":20,"0x7fc790366ee0":20,"0x7fc790367be0":[19,21],"0x7fc790382190":1,"0x7fc790382760":9,"0x7fc79039a070":21,"0x7fc79039a790":21,"0x7fc79039a820":21,"0x7fc7903bd040":1,"0x7fc7903bdbb0":9,"0x7fc7903bddf0":19,"0x7fc790682160":21,"0x7fc79068b160":19,"0x7fc79068b250":1,"0x7fc79068b340":19,"0x7fc79068b370":1,"0x7fc79068ba00":20,"0x7fc79068bb80":21,"0x7fc793f99a30":1,"0x7fc793fbc9d0":9,"1":[0,1,2,3,4,6,7,8,11,12,13,14,15,18,19,20,21,23,24,25,26,27,31,32,34],"10":[0,1,3,4,9,10,17,19,20,21,23,26,27,33],"100":[3,9,14,15,21,26,27],"1000":[3,27,35],"10000":[3,27],"100000":3,"10000000000000002":20,"1001":35,"1004":35,"1007":35,"1008":35,"101":[1,20],"1012":35,"1014":35,"1017":35,"1020641326904297":9,"1024":[9,17,35],"1025":31,"1026":35,"1038":35,"103855734598575e":9,"1040":35,"1045":35,"104580":4,"1046":35,"1050":35,"1052":35,"1053":35,"1059":35,"1061":35,"1062":35,"1063829787234043":[9,23],"1070":35,"1072":35,"1075":35,"1076":35,"1079":35,"1085501":9,"11":[1,15,19,20,21,26,28,33],"1100":35,"11042628136263043":21,"1108":35,"1109":35,"11103398852619628":26,"1112":35,"1114":35,"1115":35,"1117":35,"1121":35,"1122":35,"1123":35,"1124":35,"1126":35,"11267605633802817":21,"1128":35,"113":[1,20],"1133":35,"1134":35,"11347517730496459":[9,19,24],"1135":35,"1136":[4,35],"1142":35,"1143":35,"114391364050916":19,"1144":35,"1149":35,"115":[21,33],"1152":35,"1154":35,"1157":35,"1158":35,"1159":35,"1165":35,"1168":35,"1169":35,"1170":35,"1173":35,"11737571e":9,"1178":35,"1179":35,"1183":35,"1185":35,"1187":35,"1190":35,"1195":35,"1198":35,"11ec":[7,17,24],"12":[1,2,5,19,20,21,23,28],"120":[0,1,3,4,6,10,11,13,14,15,18,20,24],"1200":35,"1201":35,"1204":35,"120567":[1,6],"12056737588652477":1,"1209":35,"1210":35,"1214":35,"1229":35,"123":1,"1231":35,"1233":35,"1234":21,"1235":35,"1238":35,"1239":35,"1242":35,"1245":35,"1248":35,"125":16,"12585871697357337":19,"1259":35,"1260":35,"1261":35,"1263":35,"1269":35,"1270":35,"1271":35,"128":[1,9,19,20,21],"1282":35,"1290":35,"1297":35,"13":[1,17,19,20,21],"1306":35,"1309":35,"1311":35,"1318":35,"1321":35,"1322":35,"1325":35,"1325058937072754":15,"1326":35,"1329":35,"1332":35,"1335":35,"1339":35,"1340":35,"1346":35,"1348":35,"1349":35,"1356":35,"1361":35,"136510":3,"13651009800609504":3,"1366":35,"1367":35,"1368":35,"137":28,"1370":35,"13704890e":9,"14":[1,3,7,16,17,19,20,21],"140":[31,35],"14069368736662313":9,"142":[31,34],"142032801716559e":19,"14268277711454813":21,"14281372797350056":3,"142814":3,"14295295806077554":21,"144":10,"149":35,"149227":1,"15":[1,2,12,13,19,20,21,24,27,28],"152":21,"153994":1,"157":35,"16":[0,1,4,7,9,12,19,20,21,23,24,27,35],"16262682406125173":21,"1643127754":15,"1643127756":15,"1643128262":9,"1643128263":9,"1643128264":9,"1643128265":9,"1643128266":9,"1643128267":9,"1643128268":9,"1643128270":9,"1643128271":9,"1643128273":9,"1643128274":9,"1643128277":9,"1643128278":9,"1643128280":9,"1657884":9,"16649281":9,"167":[17,22],"17":[1,19,20,21,25,28],"17122959":9,"17152095":9,"175":[2,5],"177767":1,"1778380870819092":9,"17783809":9,"1799":9,"18":[1,4,7,9,12,19,20,21,35],"180":9,"18010277":9,"18012072":9,"186":12,"18669":4,"186914":1,"187":23,"1894":19,"189696963506304e":19,"19":[1,6,9,19,20,21,35],"1e":[3,9,21,26,27],"1e2":26,"1st":31,"1th":9,"2":[0,1,2,3,4,7,15,18,19,20,21,23,24,25,26,27,31],"20":[1,3,8,9,17,19,20,21,27,35],"200":[21,27],"2000":[3,27],"2004":[15,34],"2006":[31,34],"2007":[31,32],"2012":35,"2015":[31,32,35],"2020":[31,32],"2022":[7,12,19],"2037591637980971e":4,"204":[21,22],"2047":27,"20759805e":9,"21":[1,3,19,20,21,35],"2118001735899097e":4,"213965":7,"21521743568582094":4,"216":23,"218":[3,5],"2192999307296778":19,"22":[1,3,7,19,20,21,35],"224804":3,"22480405782181234":3,"229":9,"23":[0,1,7,16,19,20,21,35],"2301056385040283":15,"231":7,"232426":3,"2368037115065363e":4,"23910875621465733":19,"24":[1,9,12,16,19,20,21,35],"24822695035460995":19,"2494493":9,"2497721880674565e":21,"25":[0,1,3,4,6,7,12,19,20,21,27],"254898":9,"255":[12,27],"256":[1,19],"26":[1,3,9,19,20,21],"263":[1,20],"264":27,"264884":2,"2673461437225342":9,"27":[1,14,16,19,20],"271":35,"272":10,"273":12,"28":[1,6,12,19,20,31,32],"280":35,"2836225":9,"2841884333778574e":[1,20],"285306370007781":19,"29":[1,12,19,20],"293":35,"295":35,"2962":[31,32],"2970":[31,32],"2nd":31,"2th":9,"3":[0,1,2,3,4,6,7,8,9,13,15,19,20,21,23,24,25,26,27,32,33],"30":[1,2,3,4,6,7,8,9,10,11,12,13,14,17,18,19,20,21,23,25,26,27,31,34,35],"305902":1,"3072":[0,18],"30755227194768237":21,"309":19,"31":[1,6,14,19,20,22,27],"313003":1,"3137a7553b01":[7,17,24],"317":35,"32":[0,1,3,9,19,20,21,23,27,35],"3202262453017803e":19,"32225143e":9,"32361629219618826":19,"325":35,"326":35,"32768":27,"328":[1,20],"32824":9,"328568999403367e":24,"32956881e":9,"33":[0,6,12,19],"34":[1,4,12,19,20,21],"3440590734654458e":19,"35":[6,12,17,19],"351":35,"358":35,"358378124977852e":4,"36":[12,19,20],"3600":0,"36170212765957444":24,"36245140e":9,"3667099475860596":9,"36670995":9,"368":[12,19],"368794":6,"36879432624113473":20,"369811497206404":9,"37":[1,6,12,19],"3700860264543197e":11,"3716748930467322e":21,"374":12,"375":35,"37641366e":9,"38":[0,6,12,19],"382":[4,5],"38430977e":9,"387912939529945e":21,"388968":9,"389265":1,"39":[1,4,6,12,16,17,20,21,22],"394691":1,"39749085e":9,"39879073225079487":4,"3a1d2162":24,"3f":[10,12],"3gb":[18,31,34],"3rd":[31,35],"3th":9,"4":[1,2,3,4,9,10,11,12,18,19,20,21,23,24,25,26,27,33],"40":[6,21],"400":[3,4,27],"4012537":15,"40594":2,"409":[15,16,35],"40922":10,"4096":15,"40981":[8,13],"40_advanc":8,"41":[1,6,19,20,21],"410":35,"410144741041167e":21,"4126":[1,20],"414":[1,20,21],"416798":1,"41746352744644133":15,"418":12,"42":[1,3],"423290":1,"424":35,"425":35,"42961621e":9,"43":[1,12,20],"431":33,"43667876507897496":[0,4],"436679":4,"438":35,"44":[4,20,24],"440288":1,"444":35,"44685232":9,"447294":2,"4472941828699525":2,"448":[1,20],"4495397":9,"45":12,"4550418898836528":[0,4],"455042":4,"45597150689539434":19,"457":12,"4595393":9,"45954164":9,"4622203211010016e":19,"4635442279519353":[1,20],"472":12,"4748068089650166":4,"474807":4,"4777560234069824":9,"48":23,"482":35,"48226284980773926":9,"483":12,"4848618507385254":9,"49":[12,18],"491":35,"49326086e":9,"49503611":9,"498":35,"499":[25,28],"4th":9,"5":[1,3,6,9,10,13,14,15,18,19,20,21,23,24,25,26,27],"50":[0,3,7,12,17,20,27,31],"500":[23,27,28,35],"5000":10,"500000":6,"502":12,"504512":1,"506856350040198e":21,"512":[1,2,3,4,9,15,19,20,21,24,25],"5186726734789994":4,"518673":4,"52":[18,23],"5201114871391565e":19,"522":5,"526":9,"5286041327277":4,"53":[12,19,22],"532":35,"533":12,"535":35,"537":35,"538":35,"54":[1,11,12,16,20],"542":12,"5475628":9,"55":[12,13,16],"5501728057861328":9,"55017281":9,"552706":2,"5568986":9,"55773043e":9,"558":[26,28,35],"56":[3,5],"560245":1,"5615131855010986":9,"56151319":9,"5627":17,"564060":1,"566":[12,35],"567":[10,16,22],"569":35,"57":[16,18,22],"5744883902069":[1,20],"57629717e":9,"5765781":9,"5783995454757295e":9,"58":12,"5808424949645996":9,"5846457091661748e":9,"585":35,"5852375030517578":9,"5855373845454157":4,"5860536098480225":9,"58605361":9,"588":35,"59":[4,5,6,16,21,22],"590":12,"5905006626969":9,"592":35,"593":35,"5gb":31,"5th":9,"6":[0,1,4,9,10,14,15,19,20,21,24,27],"60":[2,12,15,19,21,35],"600":35,"601":[18,22],"604":2,"604374":1,"6064541566557635":9,"6066226959228516":9,"608678":4,"609412172481434e":21,"611":35,"614":35,"615":12,"617622":4,"618":35,"619489":4,"626":[12,35],"63":11,"631206":6,"637":[12,16],"64":[0,1,11,19,20,21,31,33],"64437157661465":19,"646":35,"6468265056610107":9,"648":35,"650":[12,35],"65199313632313":26,"652099":1,"655":35,"6555753":15,"6564128398895264":9,"65641284":9,"65820479":9,"6582047939300537":9,"6587261797143242":19,"66":[0,21],"660":35,"665899307194":4,"6661824659281315":9,"667":35,"669":35,"67":[0,14,21],"670":35,"676":35,"676522":3,"676605":9,"68":[1,4,6,9,10,11,12,13,14,15,17,18,20,21],"680":35,"6808576583862305":9,"68085766":9,"685210":1,"688":35,"689027":1,"6a2":35,"6gb":[31,34],"6th":9,"7":[1,3,4,9,19,20,21,24,27,33],"701":35,"7056146361355262":15,"70629898e":9,"715":35,"7154517":9,"715452":9,"716530":1,"718":[19,22],"72":[1,4,6,9,10,11,12,13,14,15,17,18,20,21],"7203934192657471":9,"72039342":9,"7244397014112777e":9,"732":35,"733":35,"738643048253599e":1,"739":35,"74":21,"74116751e":9,"741800":1,"74597367e":9,"7462342366289323e":4,"747":[1,5],"747741":1,"748310":1,"7493172":9,"74d6c3db":17,"75":[3,27],"754":12,"75655689e":9,"7570b3":4,"7576518":9,"763":[8,16],"764":[12,35],"76699224":9,"769":35,"771":35,"772":35,"773913":6,"774":35,"777":17,"7772788812704434":9,"778":[27,28,35],"781":35,"783555":1,"785":35,"786":12,"788":12,"7895711479212801":9,"79057296e":9,"791":35,"794":35,"795":35,"797078":1,"7dfa":7,"7dfd":17,"7dfe":24,"7th":9,"8":[1,3,4,8,9,19,20,21,27],"80":[6,12,19,35],"803":[24,28,35],"8047274080856589":9,"805679":1,"807":12,"80724345e":9,"807887":1,"814":35,"815":35,"816570529036241e":19,"822":35,"832":[1,20],"836":12,"840":35,"840576":3,"8428972335335263e":4,"843":35,"843515731149308":19,"8492753623188406":13,"852":[13,16,35],"854":35,"856":31,"860":33,"862":35,"863":35,"863830":1,"864":35,"86528503e":9,"868518":6,"868573":9,"869":35,"870":35,"872":35,"874":35,"875f":[7,17,24],"876":35,"878":[20,22],"8786127167630058":8,"879":35,"879433":6,"881":35,"881188":6,"882":35,"884615":6,"8858495026618883":3,"886":35,"88652482":9,"88675402e":9,"8888":33,"889":35,"890000":6,"891":35,"893":35,"89361702":9,"894231":6,"897":35,"897617":7,"8b67af6":35,"8th":9,"9":[1,9,11,19,20,21,27,33],"90":10,"900":35,"900000":6,"901":35,"902":35,"905":35,"905028":6,"907":[12,35],"909":35,"910112":6,"911":35,"91287172e":9,"914":35,"914894":6,"915":7,"916":35,"918":35,"919":35,"9195228400065838":3,"919836":6,"92198582":9,"921986":6,"923833":6,"924":35,"925":35,"925454":6,"927":35,"928":[9,16],"9290439925152777":9,"929078":6,"92907801":9,"931":35,"932584":6,"933":35,"9331254454871041":9,"933282":1,"934066":6,"934783":6,"935069":6,"936":[12,35],"936170":6,"936842":6,"937":12,"9370629370629371":24,"937446":6,"9375":9,"940":35,"940687":6,"941799":6,"942":[6,16],"943064":6,"943262":6,"94326241":9,"943820":6,"944":12,"9440559440559441":[19,21],"944444":6,"945055":6,"945652":6,"946":35,"94611519e":9,"947":35,"948":35,"948571":6,"948682":6,"949":35,"950276":6,"950303":6,"950355":6,"951":35,"951048951048951":21,"952679":6,"953":35,"954300":6,"954545":6,"955056":6,"955556":6,"956":35,"956044":6,"956989":6,"957447":6,"958":12,"958041958041958":[1,6,9,14,17,19,20,21,25],"959918":6,"960000":6,"960452":6,"961":35,"961326":6,"961538":6,"963915":6,"964539":[6,24],"96453901":9,"964789":14,"965034965034965":[11,14,21],"965116":6,"965517":6,"965536":6,"965909":6,"966292":6,"966667":6,"967033":6,"967136":21,"9689529284463153":3,"969533":6,"970":35,"971631":[6,9],"97163121":9,"971751":6,"971831":21,"972067":6,"973":35,"973582":1,"9749646614525684e":4,"975":[7,16],"975151":6,"9755563101648113":9,"976257":1,"976744":6,"977273":6,"977528":6,"977778":6,"978022":6,"978723":6,"98":32,"98080571":9,"983240":6,"984":[10,35],"984767":6,"985816":[6,18,20,21],"986013986013986":23,"9862375":9,"9886363636363636":7,"988764":6,"989":35,"9899874843554443":7,"99":27,"990":35,"9911616161616161":7,"992":35,"992908":[17,19],"9941379232526211":9,"9952888":9,"9957142857142857":10,"9967857433838874":9,"997":35,"999":[1,3,9,19,20,21,27],"9999":[3,9,27],"999999999999991e":11,"abstract":[0,30],"boolean":[0,31,34],"break":35,"case":[0,8,9,11,14,17,21,29,31,33,34,35],"class":[0,2,7,10,14,23,24,25,26,27,30,31,34,35],"concei\u00e7\u00e3o":35,"default":[0,3,7,14,17,20,21,23,25,26,27,31,34,35],"do":[0,7,8,17,20,30,31,33,34,35],"export":[31,34],"final":[0,9,17,19,21,31,34,35],"float":[0,8,23,25,26,27,31,34],"function":[0,6,12,13,30,31,35],"gro\u00dfberg":35,"import":[0,1,2,3,4,6,7,8,9,11,12,13,14,15,18,19,20,21,23,24,25,26,27,31,32,34,35],"int":[0,2,14,23,26,27],"long":31,"mielik\u00e4inen":35,"new":[0,15,17,23,26,27,30,31,34,35],"public":[31,32],"return":[0,6,12,13,15,19,21,23,24,25,26,27,30,31,34,35],"short":[31,34],"true":[0,1,2,3,4,6,7,8,9,11,12,13,14,15,17,18,19,20,21,23,24,25,26,27,31,32,34],"try":[31,33,35],"while":[0,9,17,31,34],A:[0,4,9,10,11,13,19,32,33,34,35],As:[9,17,31,34],At:[31,34],But:9,By:[0,7,20,31,34,35],For:[0,7,10,13,14,31,33,34],If:[0,8,13,17,31,32,33,34],In:[0,6,8,9,13,14,17,30,31,33,34],It:[0,9,14,17,21,31,32,33,34,35],NOT:31,No:35,Not:[0,2,7,8,23,24,25,26,27,33],Of:30,One:[0,4,14,31,34],Or:31,The:[0,1,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,23,24,25,26,27,30,33,35],There:[10,13,31,33,35],These:[0,30,31,34],To:[0,8,9,10,14,15,17,18,21,24,30,31,32,33,34,35],Will:0,_329_:35,_:[7,15,17,18],__:0,__choice__:[3,6,9,15,27],__init__:[23,24,25,26,27],__main__:[17,18,19,21,31,32,35],__name__:[17,18,19,21,31,32,35],_gatheringfutur:19,_metric:12,_repeatedsplit:0,_test_scor:13,_valu:15,a10:13,a11:13,a12:13,a13:13,a14:13,a1:13,a2:13,a3:13,a4:13,a5:13,a6:13,a7:13,a8:13,a9:13,aaron:[31,32,35],abbrevi:[21,30],abil:[31,34],abl:[31,34],about:[0,7,15,17,18,19,21,30,31,32,33,34,35],abov:[4,17,31,32,34],acceleration_i:10,acceleration_x:10,acceleration_z:10,acceleromet:10,accept:[0,12,31,34],access:[0,9,31,34,35],accidentali:35,accord:[10,14,31,34],accordingli:8,accu:12,accu_add:12,accuraci:[0,1,2,6,8,9,11,13,14,17,18,19,20,21,24,25,31,32,34,35],accuracy_scor:[1,2,6,8,9,11,12,13,14,17,19,20,21,23,24,25,32],accuracy_wk:12,achiev:[0,33],across:17,action:35,activ:[1,9,20,21,23,27,33,35],actual:[17,35],ad:[0,31,33,35],adaboost:[6,9,11,27],adaboostclassifi:19,adam:[9,23,27],adapt:[19,21,34],add:[23,24,25,26,27,30,31,33,35],add_classifi:[23,27,30],add_condit:[25,26],add_hyperparamet:[23,25,26,27],add_preprocessor:[24,25,30],add_regressor:[26,30],addit:[0,7,9,30,31,35],addition:[8,9,19,31,34],additional_info:[9,15],address:17,adjust:31,adrian:35,advanc:[0,21,31,32,34],advantag:32,advic:31,advis:32,affect:[0,10],affin:[3,27],after:[0,4,14,20,30,34,35],afterward:20,again:[10,35],against:[4,35],agarw:35,aggress:19,ahn:35,al:[31,32,34],alencar:35,alexand:35,algorithm:[0,2,3,9,14,15,17,18,19,20,21,24,25,27,30,31,32,34,35],algorithm_run:[1,4,6,9,10,11,12,13,14,15,17,18,20,21],alia:[0,8],align:31,all:[0,9,11,14,15,20,29,31,32,33,34,35],alleavi:31,allevi:[31,34],alloc:[0,18,31],allow:[0,10,17,30,31,33,34,35],along:[0,9],alpha:[0,1,3,4,9,11,19,20,21,23,26,27],alpha_1:[0,4],alpha_2:4,alreadi:[13,31],also:[0,6,7,9,10,12,13,14,15,17,19,20,27,30,31,32,33,34,35],altern:[19,33],alwai:[0,7,31,34,35],amount:[31,34],an:[0,4,9,10,11,12,13,14,17,20,24,28,29,30,31,32,33,34,35],analysi:25,anatolii:35,andrew:35,ani:[0,7,8,15,31,34,35],anoth:[19,35],answer:31,api:[12,15,30,31,32],appear:31,append:[1,4,6,9,10,11,12,13,14,15,17,18,20,21],appli:[0,31],applic:[0,17],appreci:[31,32],approach:[31,34],appropri:[0,35],apt:33,ar:[0,2,7,8,9,10,13,14,15,19,23,24,25,26,27,30,31,32,33,34,35],arbitrarili:0,ard_regress:4,ardregress:[0,4],area:31,arff:35,arg:[0,17],argsort:10,argument:[0,8,13,14,21,31,34,35],around:[25,30,31,34],arrai:[0,2,8,9,13,14,30,31,34,35],articl:[31,32],arxiv20a:[31,32],arxiv:[31,32,35],as_fram:[2,7,8,13,15],ascend:0,asctim:7,askl2:[0,32],aspect:34,assert:[12,24,27],assum:[31,34],astyp:[2,13,14],asyncio:19,attach:0,attribut:[0,9,13,35],attributeerror:35,australian:[8,13],author:[31,32],auto:[0,6,7,8,10,12,13,14,15,18,19,28,29,31,34,35],autom:[31,32,34],automat:[0,8,13,19,31,32,34,35],automl:[0,1,2,3,4,7,9,10,11,13,14,15,17,18,19,20,21,31,32,33,34,35],automl_:[9,12],automl_common:35,automlclassifi:35,automlregressor:35,autosklearn2classifi:[0,32,35],autosklearn:[0,1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,23,24,25,26,27,30,31,32,34,35],autosklearn_classification_example_tmp:1,autosklearn_inspect_predictions_example_tmp:10,autosklearn_interpretable_models_example_tmp:11,autosklearn_multioutput_regression_example_tmp:3,autosklearn_parallel_1_example_tmp:18,autosklearn_parallel_3_example_tmp:17,autosklearn_random_search_example_tmp:19,autosklearn_regression_example_tmp:4,autosklearn_resampling_example_tmp:14,autosklearn_sequential_example_tmp:20,autosklearn_sh_example_tmp:21,autosklearn_sh_example_tmp_01:21,autosklearn_sh_example_tmp_03:21,autosklearn_sh_example_tmp_04:21,autosklearn_sh_example_tmp_cv_02:21,autosklearn_tmp_:[0,31],autosklearnchoic:9,autosklearnclassifi:[0,1,2,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,23,24,25,27,31,32,34,35],autosklearnclassificationalgorithm:[0,23,27,30],autosklearnclasssifi:0,autosklearnpreprocessingalgorithm:[0,24,25,30],autosklearnregressionalgorithm:[0,26,30],autosklearnregressor:[0,3,4,26,35],avail:[0,7,9,14,17,31,32,33,34,35],averag:[0,1,3,9,19,21,27],average_precis:[0,12],avoid:[0,20,31,32],axi:10,back:[0,33],backend:35,backward:35,bad:[14,35],bae:35,balanc:[0,1,2,9,11,15,19,20,21,23,24,25,27,31],balanced_accuraci:[0,6,12],balanced_accuracy_scor:35,balancing_strategi:0,base:[0,10,23,24,25,26,30,34],base_estim:19,basecrossvalid:0,basepipelin:0,baseshufflesplit:0,bash:[17,33],basic:9,basto:35,batch_siz:[0,9,27],bayesian:[17,31,32,34,35],becaus:[0,9,31],becker:35,becom:17,been:33,befor:[14,31,32,35],began:0,behavior:[8,31,35],behind:32,being:[0,4,31,35],believ:31,bellow:[2,7,8,23,24,25,26,27],belong:[31,34],below:[4,7,11,13,14,19,20,27,31,33],ben:35,berba:35,bereng:35,bergman:35,berman:35,bernoulli_nb:[9,11,27],bernoullinb:19,best:[0,2,4,11,14,15,17,18,19,20,21,24,31,34,35],beta_1:[1,9,19,20,21,27],beta_2:[1,9,19,20,21,27],better:[0,15,31,34,35],between:[0,4,10,17,31,34,35],bibtex:32,bin:33,binari:[0,31,33,34,35],binder:[1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,23,24,25,26,27],bit:[31,35],bitsbuff:35,bla:35,black:35,blis_num_thread:31,block:31,blum:[31,32,35],bohb:21,booktitl:[31,32],bool:[0,13,30,35],boost:35,bootstrap:[1,3,9,15,19,20,21,27],borovec:35,bosh:21,bot:35,both:[0,10,15,19,23,30,31,32,34],bound:[31,34,35],boxplot:10,bracket:35,branch:35,breast_canc:[1,9,11,14,18,19,20,21],broken:35,browser:[1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,23,24,25,26,27,33],bsd:32,budget:[0,9,15],budget_typ:21,bug:[32,35],build:[0,31,33,34,35],builder:[7,31,34,35],built:[6,12],builtin:8,bump:35,bunch:8,c:[1,4,9,19,20,21,24,27,31,33],cach:35,cache_s:9,calcul:[0,2,6,7,8,23,24,25,26,27,35],calculate_scor:35,call:[0,8,9,11,14,20,30,31,34,35],callabl:0,callback:[0,19,35],calsaverini:35,can:[0,2,4,7,8,9,10,11,12,13,14,15,17,19,20,30,31,32,33,34,35],cancellederror:19,cannot:[0,33,35],caoyi0905:35,cap:15,care:18,caruana:[15,31,34],categor:[0,3,8,13,23,25,26,27,31,34,35],categori:[8,13,35],categorical_encod:[3,9,15,27],categorical_transform:[3,9,15,27],categoricalhyperparamet:[23,25,26],category_coalesc:[3,9,15,27],cater:32,caus:35,cd:33,certain:31,certainti:0,cff:35,challeng:35,chanc:0,chang:[10,11,15,24,29,31,35],channel:33,channel_prior:33,characterist:0,check:[7,17,18,24,31,32,33,34,35],check_non:25,checker:35,cheng:35,chi2:27,choic:[3,9,11,15,23,25,26,27,31],choos:[11,31,34],chosen:[0,35],chu:35,chung:35,ci:35,circl:35,citat:[31,35],cl:[6,7,8,12,13,15,32],clarifi:35,class_weight:[1,9,20],classif:[5,6,7,8,9,10,12,13,14,15,17,18,19,20,21,24,25,27,28,29,31,32,34,35],classifi:[0,15,24,25,30,31,34,35],classification_metr:12,classifierchoic:[1,2,9,11,15,19,20,21,23,24,25],claus:[3,27,32],clean:35,cleanup:35,clear:[31,34],clf:[23,24,25,27],cli:17,cli_start_schedul:17,cli_start_work:17,click:[1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,23,24,25,26,27,34],client:[0,7,31,34],clip:35,clonabl:35,clone:[33,35],close:35,cluster:[0,17,18,31,35],code:[1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,23,24,25,26,27,29,30,31,32,34,35],coef0:[3,9,26,27],coef0_condit:26,coeffici:0,col:6,collect:10,column:[0,8,13,31,34,35],com:[7,17,30,33],combin:[10,15,31],command:[13,15,18,22,29,31,33,34,35],common:[14,25,31,33,34],compat:[9,15,17,35],competit:35,compil:[31,33],complet:[3,27,35],compli:[2,15],compliant:35,compon:[0,1,2,3,4,11,15,17,19,20,21,28,29,31,34,35],compos:[31,34],comput:[18,31,35],concat:[1,4,6,9,10,11,12,13,14,15,17,18,20,21],concept:0,conda:[31,35],condit:[3,25,26,27,31],config:[0,15,33],config_id:[0,9],configspac:[0,15,23,24,25,26,27,30],configur:[0,7,9,16,17,23,24,26,29,30,32,34,35],configuration_origin:[9,15],configuration_spac:[0,15,23,24,25,26,27,30],configurationspac:[0,23,24,25,26,27,30,35],connect:17,consid:[0,14,31,34],consist:[0,35],consol:7,constant:[0,3,9,15,17,19,21,23,24,25,26,27,30,35],constraint:[0,15],construct:[9,14,17,19,21,30,31,32,34],consum:[31,34],consumpt:35,contact:32,contain:[0,9,13,17,27,31,32,34,35],context:35,continu:0,contrast:0,contrib:31,contribut:[31,35],control:[0,14,31,34,35],convert:[30,35],cookbook:7,copi:[14,31],copy_x:4,core:[0,7,18,20,31,34,35],correct:[31,35],correctli:[0,31,33,35],correspond:[0,10,27],cosin:[3,27],cost:[0,1,2,3,4,9,11,12,15,19,20,21,23,24,25,26],could:[21,33,35],cours:30,cpu:[17,31],crash:[0,2,9,14,17,18,19,20,21,24,31,35],creat:[0,13,15,26,30,31,34,35],creation:0,criterion:[1,3,9,15,19,20,21,27],cross:[0,31,34,35],crossvalid:0,crucial:[19,34],cs:[15,21,23,25,26,27,31,32],csr_matrix:[31,34],cube:[3,27],curc:31,current:[0,31,33,34],custom:[0,6,7,14,27,29,31,34,35],custom_error:6,customrandomforest:27,cutoff:15,cv:[0,9,14,21],cv_result:[0,6,35],cv_results_:[0,6,9,31,34],cv_voting_ensembl:0,cython:35,d95f02:4,d:[2,7,8,13,31,34,35],dai:[31,34,35],dask:[0,7,17,18,31,34,35],dask_client:[0,17,19,21],dask_distributed__worker__daemon:17,data:[0,28,29,30,31,34,35],data_id:[2,7,8,10,13,15],data_preprocess:[0,1,2,3,4,9,11,15,19,20,21,23,24,25,26],data_preprocessor:[0,1,2,3,4,9,11,15,19,20,21,23,24,25,26,27],datafram:[0,6,8,13,29,31,34,35],datapreprocessorchoic:[1,2,3,4,9,11,15,19,20,21,23,24,25,26],dataset:[0,1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,23,26,27,29,31,32,34,35],dataset_nam:[0,1,2,3,4,9,10,11,14,15,17,18,19,20,21],dataset_properti:[23,24,25,26,27],date:31,deadlock:[31,35],debug:[16,29],decid:[0,31,35],decis:[0,31],decision_tre:[1,3,6,9,11,27],decisiontreeclassifi:[1,9,19,20],decreas:[10,35],deem:11,deep:0,def:[6,12,17,19,21,23,24,25,26,27],default_config:21,default_data_contain:31,default_valu:[23,25,26,27],defin:[0,6,10,12,14,30,31,34],definit:[30,31,34],deflat:[3,27],degre:[3,9,26,27,35],degree_condit:26,delet:[0,7,31,35],delete_tmp_folder_after_termin:[0,7,15,17,20,31],demand:31,demonstr:[8,13,14,19,21,23,24,25,26,27,35],deng:35,dens:[23,24,25,26,27,30],densifi:11,depend:[0,30,31,33,35],depict:4,deploy:35,deprec:[1,4,6,8,9,10,11,12,13,14,15,17,18,20,21],deprecationwarn:8,descend:0,describ:[0,9,10,30,31,32,34],deshpand:35,design:9,desired_boolean_column:13,desired_categorical_column:13,desired_numerical_column:13,despit:35,destroi:14,detail:[0,2,8,13,31,33,34],determin:[0,31,34],determinist:35,dev:33,devdoc:8,develop:[32,33,35],df:31,diabet:[0,4],diagon:4,dict:[0,9,31,34],dictionari:[0,9,30,35],did:35,diego:35,differ:[14,17,27,31,34,35],digit:17,dimens:35,diogo:35,dir:33,directli:[0,9,13,14,17,30,31,33,34,35],directori:[0,7,31,35],disabl:[0,17,31,34,35],disable_evaluator_output:[0,9,14,21],disable_existing_logg:7,disc:[0,31,34],discrimin:25,discriminant_analysi:25,discuss:31,disk:[7,15,31,35],displai:0,disregard:0,distanc:[3,9,27],distribut:[0,7,17,18,31,33,34,35],doc:[7,10,17,31,33,34,35],docker:35,dockerfil:35,dockerhub:[33,35],docstr:35,document:[0,7,15,31,32,33,34,35],doe:[0,10,17,24,31,35],domashnev:35,don:31,done:[8,17,18,31,33,34],doubl:35,down:18,download:[1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,23,24,25,26,27,29,33,35],downsid:31,drawn:0,drive:[31,35],drop:[32,35],dropout:35,dtype:[2,8,9,13,31,34],dual:[1,9,19,20,27],due:[0,17,31,35],dummi:[4,35],dummyclassifi:35,dummyfutur:7,dummyregressor:35,duplic:32,durat:[0,1,2,3,4,9,15],dure:[0,9,14,31,34,35],e:[0,3,10,13,14,31,33,34,35],each:[0,8,9,10,17,18,30,31,34],earli:35,early_stop:[1,9,19,20,21,27],easili:30,eat:31,eddi:35,edg:35,edit:33,effici:[31,32,34],effort:31,eggensperg:[31,32,35],eigen:[25,27],either:[0,6,31,34],elasticnet:[4,9,19,27],element:0,elif:13,els:[0,8,13,21,25,27],emploi:[31,34],empti:24,en:[17,18],enabl:[15,33,35],encod:[0,3,9,13,27,31,34,35],end:[0,9,11],end_tim:0,endtim:[9,15],enforc:[31,33,34,35],engelen:35,enough:0,enrico:35,ensembl:[0,7,9,15,19,21,27,32,35],ensemble_dict:0,ensemble_nbest:[0,17,31,34],ensemble_onli:[0,31],ensemble_optimization_scor:[7,13],ensemble_s:[0,11,17,20,31,34],ensemble_test_scor:[7,13],ensemble_weight:[0,1,2,3,4,9,11,19,20,21,23,24,25,26],ensemblebuild:7,ensur:[31,34],entri:[9,32],entropi:[1,15,19,20,21,27],enumer:9,environ:[31,33,34,35],epsilon:[4,9,11,27],epsilon_insensit:4,equal:[0,12,35],equalscondit:26,error:[0,6,19,31,34,35],error_add:12,error_r:[6,12],error_wk:12,especi:[31,34,35],essenti:33,estim:[0,4,7,9,13,21,23,26,27,31,32,34,35],et:[31,32,34],eta0:[4,9,11,19,21,27],eta:21,etc:9,euclidean:[3,27],evalu:[0,9,12,13,17,31,34,35],even:[7,15,31,35],everi:35,everyon:31,everyth:[17,31],everytim:31,exampl:[0,1,2,3,4,6,7,8,9,10,11,13,14,15,17,18,19,20,21,23,24,25,26,27,30,31,33,34,35],example_calc_multiple_metr:[6,16],example_classif:[1,5],example_debug_log:[7,16],example_extending_classif:[23,28],example_extending_data_preprocessor:[24,28],example_extending_preprocessor:[25,28],example_extending_regress:[26,28],example_feature_typ:[8,16],example_get_pipeline_compon:[9,16],example_inspect_predict:[10,16],example_interpretable_model:[11,16],example_metr:[12,16],example_multilabel_classif:[2,5],example_multioutput_regress:[3,5],example_pandas_train_test:[13,16],example_parallel_manual_spawning_cli:[17,22],example_parallel_n_job:[18,22],example_random_search:[19,22],example_regress:[4,5],example_resampl:[14,16],example_restrict_number_of_hyperparamet:[27,28],example_sequenti:[20,22],example_single_configur:[15,16],example_successive_halv:[21,22],examples_20_bas:5,examples_40_advanc:16,examples_60_search:22,examples_80_extend:28,examples_jupyt:29,examples_python:29,exce:31,exceed:[0,2,9,14,17,18,19,20,21,24],except:[0,19,21,35],exclud:[0,27,31,34,35],exclus:[30,31,34],execut:[0,5,16,17,22,28,31],exemplifi:[7,21,31,34],exercis:0,exhaust:[31,34],exist:[28,29,30],exp:[3,27],expand:34,expect:[0,31],experienc:32,experiment:[0,32,35],explain:[9,33],explained_variance_ratio_:9,explan:[16,29,33,34],explicit:17,explicitli:[14,21],ext:7,extend:[0,6,28,29,31,32,35],extra:0,extra_argu:12,extra_tre:[1,3,6,9,11,21,27],extra_trees_preproc_for_classif:[11,15,27],extra_trees_preproc_for_regress:[3,11],extratreesclassifi:[1,19,20,21,35],extratreesregressor:3,f1:[0,6,31],f1_macro:[0,2,12],f1_micro:[0,12],f1_sampl:[0,12],f1_weight:[0,12],f52bd18f:7,f:[2,10,12,17],f_classif:27,facad:[19,21],factor:35,factori:0,fail:[2,9,18,19,21,31,33,35],failur:[0,35],falkner:[31,32,35],fals:[0,1,2,3,4,6,7,9,12,13,14,15,17,19,20,21,23,25,26,27,31],familiar:0,faq:[32,35],fashion:9,fast_ica:[3,11,27],faster:35,favorit:31,fdr:27,feat_typ:[0,8,13,31,34],featur:[0,13,14,16,19,27,29,31,32,34,35],feature_agglomer:[3,11,27],feature_nam:10,feature_preprocess:[1,2,3,4,9,11,15,19,20,21,23,24,25,26,30],feature_preprocessor:[0,1,2,3,4,9,11,15,19,20,21,23,24,25,26,27,31,34],feature_typ:[3,9,15,27],featurepreprocessorchoic:[1,2,3,4,9,11,15,19,20,21,23,24,25,26],feedback:35,feedstock:33,felix:35,felixleungsc:35,fetch:2,fetch_openml:[2,7,8,10,13,15],feurer:[31,32,35],few:35,field:30,file:[0,5,7,16,17,22,28,33,34,35],filenam:[7,17,31,34],fill_valu:9,find:[0,9,17,30,31,34,35],fine:[0,31],finish:[0,9,19,20,31],first:[10,11,14,17,20,31,34,35],fit:[0,2,10,12,14,16,17,29,30,31,32,34,35],fit_ensembl:[0,17,20],fit_intercept:[9,27],fit_pipelin:[0,15],fit_prior:[9,19,27],fix:[31,35],flag:[2,7,8,23,24,25,26,27,31,34],flake8:35,flaki:35,flip:0,float64:[8,9,13],fold:[0,14],folder:0,follow:[0,1,3,4,6,7,8,9,10,11,12,14,15,17,23,24,25,26,27,30,31,32,33,34,35],forbidden:[3,27],forest:[15,31,34,35],fork:31,forkserv:[31,35],form:[0,15,30,31,33,34],format:[2,7,8,9,13,31,34,35],formatt:7,found:[0,10,12,13,15,24,31,34,35],fpr:27,frac:0,fraction:[31,34,35],frame:[0,1,4,6,9,10,11,12,13,14,15,17,18,20,21],francisco:35,francoi:35,frank:[31,32],free:[17,31,32],friedman_ms:3,from:[0,1,2,3,4,6,7,8,9,10,11,12,13,14,15,18,19,20,21,22,23,24,25,26,27,29,31,33,34,35],from_dict:6,fu:35,full:[1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,23,24,25,26,27,30,31,34],fun:[3,27],further:[0,13,14,15,31,34],furthermor:[4,9,11,19],futur:[1,4,6,9,10,11,12,13,14,15,17,18,19,20,21,35],futurewarn:[1,4,6,9,10,11,12,13,14,15,17,18,20,21],fwe:27,g:[31,33,34],galleri:[1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,23,24,25,26,27,29,35],gamma:[3,9,26,27],gaussian_nb:[11,27],gaussian_process:3,gaussiannb:19,gave:35,gcc:33,gcc_linux:33,geffrier:35,gener:[0,1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,23,24,25,27,29,30,34,35],get:[0,9,15,19,21,31,32,33,34],get_compon:11,get_configuration_spac:[0,3,15,27],get_default_configur:21,get_estim:0,get_hyperparameter_search_spac:[23,24,25,26,27],get_metric_result:6,get_models_with_weight:[0,9,24],get_param:0,get_preprocessor:0,get_properti:[23,24,25,26,27],get_random_search_object_callback:19,get_roar_object_callback:19,get_smac_object:21,get_smac_object_callback:[0,19,21],get_trials_callback:0,getting_start:15,geurt:27,gijsber:35,gini:[9,15,27],git:33,github:[7,15,17,30,31,32,33,35],give:[0,9,29,30,31,34,35],given:[0,9,10,11,15,17,31,35],glazachev:35,glob:7,gmean:35,go:31,goal:10,good:[0,2,4,13,31,34,35],got:12,gradient:35,gradient_boost:[1,4,6,9,11,21,27],gradientboostingclassifi:35,grain:[0,31],great:31,greater:0,greater_is_bett:[0,6,12],greil:35,grid:[13,31,34],grid_resolut:10,grow:31,guarante:31,guard:[17,18,19,21,31,35],gui:35,guid:[31,32,34,35],guidanc:8,guidelin:[31,34,35],guilherm:35,gxx_linux:33,gyro_i:10,gyro_x:10,gyro_z:10,gyroscop:10,ha:[0,4,9,15,27,30,31,33,34,35],halv:[9,22,29,35],hand:[14,31,32],handl:[0,9,30,31,34,35],handler:7,handles_classif:[23,24,25,26,27,30],handles_multiclass:[23,24,25,26,27,30],handles_multilabel:[23,24,25,26,27,30],handles_multioutput:[23,24,25,26,27],handles_regress:[23,24,25,26,27,30],happen:[13,31,34,35],hard:[31,35],harder:[31,34],harsh:0,hash:0,have:[0,9,10,14,15,17,30,31,33,34,35],header:[0,31,34],heavili:[31,33],hector:35,help:31,here:[1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,23,24,25,26,27,31,32,33,34],herilalaina:35,heurist:[27,35],hidden_layer_depth:[9,23,27],hidden_layer_s:[1,9,19,20,21,23],high:0,higher:[0,4,10,31,35],highli:32,highlight:[7,8,13],hing:27,histgradientboostingclassifi:[1,19,20,21],histgradientboostingregressor:4,histori:[9,35],hnishi:35,hoc:[9,31,34],hold:33,holdout:[0,21,31,34,35],home:[1,4,6,8,9,10,11,12,13,14,15,17,18,20,21],honor:[0,15],host:35,hostedtoolcach:21,hot:[0,31,34],hotfix:35,hour:[32,34],how:[0,1,2,3,4,6,7,8,9,10,11,12,13,15,17,18,19,20,21,23,24,25,26,27,30,31,33,34,35],howev:[11,12,20,34,35],howto:7,html:[2,7,8,15,17,18],http:[2,7,8,13,15,17,18,35],hungri:31,hutter:[31,32],hyper:15,hyperband:21,hyperparamet:[0,3,9,19,23,25,26,28,29,31,32,34,35],i:[0,3,9,10,13,14,23,31,34,35],id:[0,9,31],idempot:35,ident:23,identifi:[0,31,34],idl:17,ids_config:9,ignor:[9,17,35],illustr:[9,21],imag:35,immedi:35,impact:[10,35],implement:[0,2,7,8,10,23,24,25,27,30,31,34,35],impli:30,importances_mean:10,importances_std:10,improv:[15,31,32,35],imput:[3,9,15,27,31,34],includ:[0,9,11,15,21,23,24,25,26,31,32,34,35],include_bia:[3,27],incompat:[0,33,35],incondit:25,incorrect:35,increas:[0,31,35],indent:[1,2,3,4,9,11,19,20,21,23,24,25,26],independ:31,index:[0,6,10,18,35],indic:[2,13,31],individu:[0,7,14,31,34],inf:4,infer:[8,31,34],influenc:0,info:[7,35],inform:[0,2,7,10,16,17,21,29,31,32,33,34,35],inherit:31,init:33,initi:[0,9,31,33,34,35],initial_budget:21,initial_configur:21,initial_configurations_via_metalearn:[0,2,7,8,19,23,24,25,26,27,31,34],initial_incumb:19,inproceed:[31,32],input:[0,10,13,23,24,25,26,27,29,30,35],insensit:8,insert:0,insid:[9,14,30,31,35],inspect:[4,10,11,29,31,35],inspir:0,inspur:35,instal:[0,31,32,35],instanc:[0,9,15,19,30,35],instance_id:9,instance_specif:15,instead:[0,1,4,6,7,9,10,11,12,13,14,15,17,18,19,20,21,27,31,33,35],instruct:[33,35],integ:[0,2,31,34],intend:[19,21],intensif:[0,21],intensifi:21,intensifier_kwarg:21,interact:[9,10,21,33],interaction_onli:[3,27],intercept_sc:[1,9,19,20,27],interest:[7,9,31],interfac:[10,17],interleav:34,intermedi:7,intern:[7,9,17,24,35],interpret:[10,16,29,31,34],introduc:[0,19,35],introduct:35,invok:[19,21],involv:[31,34],invscal:[9,11,27],io:15,ip:33,ipynb:[1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,23,24,25,26,27],is_determinist:[23,24,25,26,27,30],is_valid_configur:15,isaac:35,isort:35,issu:[17,24,32,33,34,35],item:[0,9,24],iter:[0,2,7,9,18,19,23,31,35],iterative_cv:35,its:[0,9,17,30,31,34],itself:[8,9,17,31,32],iver:35,j:35,jaidev:35,jeong:35,jesper:35,jha:35,jianswang:35,jin:35,jirka:35,jj:35,job:[0,7,17,18,35],joblib:31,joel:35,join:[12,17],joint:35,jongheon:35,jordal:35,jose:35,joseph:35,josh:35,jost:[31,32,35],journal:32,json:[7,17],judgement:31,julian:35,jupyt:[1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,23,24,25,26,27,29,33,35],just:0,k:[0,4,14],k_nearest_neighbor:[3,9,11,27],kaa:35,karl:35,katharina:[31,32,35],keep:[0,9,31,34],keep_vari:[3,9,27],kei:[0,6,9,24,31,34],kept:[0,31,34],kernel:[0,3,9,27,35],kernel_pca:[3,11,27],kernel_ridg:26,kernelridg:26,kernelridgeregress:26,kind:[10,13,31,34],kitchen_sink:[3,11,27],klein:[31,32,35],kneighborsclassifi:[9,19],know:[31,32,33],known:31,kobylkin:35,kp:[7,15],kr:[7,15],kwarg:[0,24],l1:[21,27],l1_ratio:[4,9,19,27],l2:[9,27],l2_regular:[1,4,9,19,20,21,27],label:[0,4,5,10,13,29,35],labelencod:[0,31,34],lambda_1:4,lambda_2:4,laptop:31,larg:[34,35],larger:[31,35],largest:10,lastli:31,later:0,latest:[17,18,32,33,35],latter:0,launch:0,laurent:35,lbfg:23,lda:[1,6,9,11,27,35],lead:[31,35],leaderboard:[0,1,2,3,4,31,34,35],learn:[0,2,9,10,13,15,17,20,30,32,34,35],learning_r:[1,4,9,11,19,20,21,27],learning_rate_init:[1,9,19,20,21,27],least:[33,35],left:0,leftov:35,legend:[4,13,31,34],len:[0,9,19,21,24],length:0,less:35,lesw:35,let:[9,31],leung:35,level:[7,31,35],levelnam:7,leverag:32,lg:[31,32],liac:35,lib:21,liblinear_svc:[1,6,9,11,27],liblinear_svc_preprocessor:[11,27],librari:[0,31,34],libsvm_svc:[9,11,27],licens:35,lifetim:17,light:17,like:[0,11,21,30,31,34],limit:[0,2,9,14,15,17,18,19,20,21,24,31,33,35],lindauer:[31,32,35],line:[10,13,18,22,29,31,34],linear:[9,25,31,35],lineardiscriminantanalysi:[1,19,20,24,25],linearsvc:[1,9,19,20],link:35,linkag:[3,27],linkcheck:35,linux:33,list:[0,8,9,11,13,31,33,34],liter:0,ll:10,load:[0,31,35],load_breast_canc:[1,6,9,11,12,14,17,18,19,20,21,23,24,25,27,35],load_diabet:[0,4,26],load_digit:[32,35],load_model:0,local:35,locat:[17,31],lockfil:35,log2:27,log:[0,3,12,16,21,26,27,29,35],log_loss:[0,12],logcosh:[3,27],logger:[0,7,35],logging_config:[0,7,35],logging_configur:7,logic:17,logist:23,loglevel:35,logloss:35,longer:[31,35],look:[9,10,15,30,31,34],loss:[0,1,4,9,11,19,21,27,31,35],losses_and_configur:9,lot:[31,35],love:31,low:0,lower:[4,23,25,26,31],lowest:9,lsqr:[19,24,25,27],lt:35,luca:35,luka:35,m:27,mabri:35,machin:[0,15,17,20,22,29,30,31,32,33,34,35],maco:35,made:35,mae:3,mai:0,main:[31,35],maint:35,maintain:[9,31],make:[0,2,15,19,21,24,31,33,34,35],make_regress:3,make_scor:[0,6,12],manag:[17,18,34,35],manhattan:[3,27],mani:[0,31,34,35],manner:[31,33],manual:[13,17,27,30,31,35],manuel:[31,32,35],map:[0,2],mariu:[31,32,35],mark:35,mask:[0,9],masked_arrai:9,master:[15,33,35],matplotlib:[4,10,13],matric:30,matrix:0,matthia:[31,32,35],max:[3,27],max_bin:[9,27],max_budget:21,max_depth:[1,3,9,15,19,20,27],max_depth_factor:[3,9,27],max_featur:[1,2,3,9,15,19,20,21,24,25,27],max_it:[1,4,9,11,19,20,21,24,27],max_leaf_nod:[1,3,4,9,15,19,20,21,27],max_models_on_disc:[0,31,34],maxim:[9,31,34,35],maximilian:35,maximum:[0,27,31,34],mb:[0,5,16,22,28,31],md5:0,mean:[0,3,4,6,9,12,14,15,27,30,31,35],mean_absolute_error:[0,12],mean_fit_tim:9,mean_squared_error:[0,12,35],mean_squared_log_error:12,mean_test_scor:[6,9],measur:[13,31,34],mechan:[0,13,31,34],median:[3,27],median_absolute_error:[0,12],member:34,memori:[0,2,9,14,17,18,19,20,21,24,31,33,34,35],memory_limit:[0,15,17,18,35],mendoza:35,mention:31,merg:32,messag:[7,35],meta:[32,34,35],meta_bas:[1,4,6,9,10,11,12,13,14,15,17,18,20,21],metadata:[0,31,34,35],metadata_directori:0,metafeatur:[1,4,6,9,10,11,12,13,14,15,17,18,20,21,35],metalearn:[0,1,4,6,9,10,11,12,13,14,15,17,18,20,21],metalearning_configur:[19,21],method:[0,1,4,6,8,9,10,11,12,13,14,15,17,18,20,21,30,31,34,35],metric:[1,2,3,4,7,8,9,10,11,13,14,15,16,17,18,19,20,21,23,24,25,26,29,31,32,34,35],metric_:6,metric_balanced_accuraci:6,metric_custom_error:6,metric_f1:6,metric_nam:12,metric_precis:6,metric_recal:6,mfeurer:33,michael:35,microsoft:33,might:[0,34],min_chal:21,min_impurity_decreas:[3,9,15,27],min_samples_leaf:[1,3,4,9,15,19,20,21,27],min_samples_split:[1,3,9,15,19,20,21,27],min_weight_fraction_leaf:[3,9,15,27],minim:[9,35],minimum:[0,31,34,35],minimum_fract:[3,9,27],minmax:[3,9,27],minor:35,minority_coalesc:[3,9,27],minr:19,minut:[1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,23,24,25,26,27,31,34,35],miotto:35,miss:[31,33,34,35],mkdir:33,mkl_num_thread:[31,34],ml:35,mlp:[1,6,9,11,27],mlpclassifi:[1,9,19,20,21,23,35],mlpregressor:35,mode:[18,27,31,33,34,35],model:[0,6,12,13,14,15,16,18,19,20,21,29,30,31,35],model_1:0,model_id:[0,1,2,3,4,9,11,19,20,21,23,24,25,26],model_n:0,model_select:[0,1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,23,24,25,26,27,32,35],modern:[31,34],modifi:8,modified_hub:[9,11,27],modul:[2,10,15,17,33,35],mohd:35,morakakam:35,more:[0,2,7,8,9,13,17,20,21,31,32,33,34,35],moreov:6,most:[0,9,31,34],most_frequ:[3,9,27],mostli:[31,34],move:[31,35],mse:3,msg:7,much:[0,11,31],multi:[0,5,7,29,35],multi_class:[9,27],multiclass:[0,2,30,31,34,35],multiclass_classif:17,multidimension:[31,34],multilabel:[0,2,30,31,34,35],multinomial_nb:[11,27],multioutput:[0,3,31,34],multioutputregressor:0,multipl:[6,10,13,17,18,30,31,34,35],multiprocess:[7,17,31,34,35],multitud:32,must:[0,13,17,18,23,30,31,33,34,35],mutual:30,mutual_info:27,mutual_info_classif:27,mybind:35,mypi:35,n:[9,10,12,31,34],n_class:0,n_cluster:[3,27],n_compon:[3,27],n_estim:[1,2,3,9,15,19,20,21,24,25,27],n_featur:[0,3],n_inform:3,n_iter_no_chang:[1,4,9,19,20,21,27],n_job:[0,1,2,3,9,15,17,18,19,20,21,24,25,31,34,35],n_jobs_:18,n_label:0,n_neighbor:[3,9,19,27],n_output:0,n_quantil:[3,9,27],n_repeat:10,n_sampl:[0,3],n_samples_fit:0,n_split:0,n_target:[0,3],nader:35,name:[0,2,6,7,8,9,11,12,13,14,15,17,18,19,20,21,23,24,25,26,27,30,31,32,34,35],named_step:[9,15],natur:0,nb:33,ndarrai:[0,31,34],necessari:[0,14,17,31,34],need:[0,9,14,17,30,31,33,35],needs_proba:[0,6,12],needs_threshold:[0,6,12],neg:[0,2,30],neither:35,nest:0,neural:[31,32],neural_network:23,neurip:[32,35],neurips15a:[31,32],never:19,nevertheless:[31,34],newer:35,next:[31,34],nicer:0,nildaimon:35,nip:32,no_coalescens:[3,9,15,27],no_encod:[3,9,27],no_preprocess:[0,3,11,21,27,31,34],non:35,none:[0,1,3,4,7,9,12,15,17,19,20,21,23,24,25,26,27,30,31,34,35],nopreprocess:27,nor:35,normal:[3,9,27,31,34],note:[0,8,10,11,17,31,34,35],notebook:[1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,23,24,25,26,27,29,31,32,33,35],noth:35,notimplementederror:[23,25,26,27],now:[4,10,11,17,31,35],np:[2,6,8,12,13,14,15,31,34],nthread:[17,19,21],nuanc:31,num_nodes_per_lay:[9,23,27],num_param:0,num_run:[9,15],num_work:21,number:[0,2,9,10,14,17,18,19,20,21,24,28,29,31,34,35],numer:[0,8,13,31,34],numerical_transform:[3,9,15,27],numpi:[0,2,3,6,8,12,13,14,15,31,34,35],nystroem_sampl:[3,11,27],object:[0,1,2,3,4,9,11,13,15,19,20,21,23,24,25,26,27,30,31,34,35],observ:27,obtain:[0,16,29,31,34,35],obviou:0,obvious:14,occur:33,off:[24,27,31,34],offer:34,offici:[7,33],offlin:35,often:31,old:35,olehb007:35,oli:35,omp:35,omp_num_thread:[31,34],onc:[0,18,33,35],one:[0,7,9,14,15,17,18,20,27,30,31,32,34,35],one_hot_encod:[3,9,15,27],ones:[2,15,29,31],onli:[0,9,11,15,17,27,30,31,34,35],onlin:19,open:[31,32],openblas_num_thread:[31,34],openml:[2,7,8,10,13,31],openmp:31,oper:[33,35],opposit:35,opt:[21,33],optim:[0,6,9,11,17,18,19,27,29,31,32,34,35],optimizatino:19,optimum:[0,6,12],option:[0,7,31,33,34],optionalarrai:0,order:[0,13,17,30,31,34,35],ordereddict:9,ordin:13,ordinalencod:[31,34],org:[2,7,8,13,15,17,18],origin:[31,35],osx:35,other:[0,9,33],otherwis:[0,21],ouput:35,our:[15,27,31,32],ourselv:26,out:[0,1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,23,24,25,26,27,30,32,33,34],outcom:0,output:[0,5,7,9,14,23,24,25,26,27,29,30,35],output_directori:31,output_distribut:[3,9,27],outsid:31,over:[0,6,8,9,11,14,16,29,31,34,35],overestim:4,overfit:[0,13,31],overhead:35,overview:[0,34],ovr:[9,27],own:[17,31],p:[3,9,27,33],packag:[15,21,31,33,35],page:[31,32,35],pair:[31,34],panda:[0,1,4,6,8,9,10,11,12,13,14,15,17,18,20,21,29,31,34,35],paper:[31,32,35],parallel:[0,3,20,22,27,29,31,35],parallel_schedul:21,param:[0,9,35],param_balanc:9,param_classifi:[6,9],param_data_preprocessor:9,param_feature_preprocessor:9,paramet:[0,15,24,35],parametr:27,part:[33,35],partev:35,parti:[31,35],partial:0,particip:35,particular:[9,31,34],particularli:21,pass:[0,7,15,17,19,31,34,35],passive_aggress:[1,6,9,11,21,27],passiveaggressiveclassifi:[1,19,21,24],path:[0,7,35],pathlib:7,pca:[3,9,11,27,31,34],pd:[0,6,13,31,34],penalti:[4,9,19,21,27],pep8:35,pepe:35,per:[0,6,8,9,13,31,34],per_run_time_limit:[0,1,2,3,4,6,8,9,10,11,12,13,14,15,17,18,19,20,21,23,25,26,27,35],percentag:[31,34],percentil:27,perceptron:[19,27],perfect:4,perform:[0,8,15,16,29,31,34,35],performance_over_time_:[0,13,31,34],perman:0,permutation_import:10,persist:[31,34],person:10,peter:35,phone:10,pick:32,picklabl:35,pid_:0,pieter:35,pip3:[33,35],pip:[33,35],pipe:17,pipelin:[0,1,2,3,4,7,9,11,15,19,20,21,23,24,25,26,27,30,31,34,35],pitfal:10,plai:35,platform:33,pleas:[0,17,19,21,30,31,32,33,34],plot:[0,8,16,29,31,34,35],plot_partial_depend:10,plt:[4,10,13,31,34],point:[0,4,17],poli:[3,9,27,35],polynomi:[3,11,26,27],pooling_func:[3,27],port:[17,33],posit:[2,30],possibl:[0,4,6,8,12,14,15,19,20,21,26,31,33,34,35],post:[9,31,34],pot:13,potenti:[31,35],power:35,power_t:[4,9,27],power_transform:[3,9,27],pprint:[1,2,3,4,9,11,19,20,21,23,24,25,26],pr:35,practic:[14,29,31],pradeep:35,pre:31,precis:[0,6,31,35],precision_macro:[0,12],precision_micro:[0,12],precision_sampl:[0,12],precision_weight:[0,12],precomput:0,predefin:[14,35],predefinedsplit:14,predict:[0,1,2,3,6,8,10,11,12,13,14,17,19,20,21,23,27,30,31,32,34,35],predict_proba:[0,23,27,35],predictor:[30,35],preprocess:[0,31,34,35],preprocessor:[0,9,28,29,30,31,34,35],present:[0,31,34,35],preserv:7,pretti:[14,31],prevent:35,previou:20,previous:[0,17,35],princip:9,print:[0,6,7,8,9,10,13,14,15,17,18,19,21,31,32,34,35],prior:[0,31,34,35],prioriti:[31,34],probabl:[0,10,35],problem:[0,2,6,9,12,31,33,34,35],proc:17,procedur:[29,34],procedurc:9,process:[0,17,20,30,31,32,33,34,35],process_cli_work:17,process_python_work:17,processor:0,produc:[30,35],progress:34,properli:[2,8,13,17,31,34],properti:30,proport:0,provid:[0,2,7,8,13,17,23,24,25,26,27,31,32,33,34,35],prune:[0,31,34,35],publish:32,pull:33,purpos:[7,13,14,21,30],push:35,put:0,pwd:33,py:[1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,31,34,35],pyninsh:35,pypi:[31,35],pyplot:[4,10,13],pyrfr:[31,33,35],python3:[21,33,35],python:[1,2,3,4,6,7,8,9,10,11,12,13,14,15,18,19,20,21,23,24,25,26,27,29,31,33,34,35],q_max:[3,9,27],q_min:[3,9,27],qda:[9,11,27,35],quadraticdiscriminantanalysi:19,quantifi:4,quantile_transform:[3,9,27],queri:[9,30],question:[31,35],r2:[0,3,4,12,26],r2_score:[0,3,4,26],r:[0,10,27],raamana:35,race:19,rafael:35,rais:[0,19,21,23,25,26,27,31,34,35],rakotoarison:35,ram:31,random:[9,15,22,29,30,31,34,35],random_forest:[0,1,2,3,6,9,11,15,21,27,31,34],random_numb:0,random_st:[1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,23,24,25,26,27,32],random_trees_embed:[3,11,27],randomforest:15,randomforestclassifi:[1,2,9,15,19,20,21,24,25,27,35],randomforestregressor:3,randomli:[10,15],rang:[3,4,13,17,23,25,26,27,31,34],rank:[0,1,2,3,4,9,11,19,20,21,23,24,25,26,35],rank_test_scor:[6,9],rather:31,ratio:9,rbf:[3,26,27],re:[2,14,31,33,35],reach:31,read:[0,31,32,33,35],readm:35,real:[2,7,8,17,23,24,25,26,27],reason:[31,34],recal:[0,6],recall_macro:[0,12],recall_micro:[0,12],recall_sampl:[0,12],recall_weight:[0,12],recent:[32,33],recogn:[8,35],recommend:[2,7,8,15,23,24,25,26,27,31,33,34],recurs:33,reddi:35,redirect:0,reduc:[3,31,35],redund:31,ref:17,refer:[0,13,31,32,34,35],refit:[0,21,31,35],reg:[26,31],reg_param:[9,19,27],regard:[31,34],regist:[30,35],registri:35,regress:[5,11,12,28,29,31,34,35],regression_metr:12,regressor:[0,9,26,30,31,34,35],regressorchoic:[3,4,26],regular:[31,32,34],regularli:31,rel:[0,35],relat:[10,35],releas:[8,31],relev:[17,21,31,35],reli:33,relu:[23,27],remain:9,remov:[0,1,4,6,9,10,11,12,13,14,15,17,18,19,20,21,31,35],renam:35,render:[17,35],replac:[0,12,27,32,33,35],report:32,repositori:31,repres:0,represent:30,reproduc:35,request:31,requir:[0,2,17,31,34,35],rerun:31,resampl:[0,9,16,29,34,35],resampling_strategi:[0,14,21],resampling_strategy_argu:[0,14,21,35],rescal:[3,9,15,27],research:32,residu:0,resolv:35,resort:0,resourc:33,respect:31,respons:10,restrict:[9,11,26,28,29,31,34,35],restructur:35,result:[0,2,4,6,9,14,17,18,19,20,21,24,29,30,32,35],retain:35,retriev:[19,35],return_x_i:[0,1,2,4,6,7,9,11,12,13,14,15,17,18,19,20,21,23,24,25,26,27,32],returncod:17,reuter:2,revers:17,rf:27,rh2epm:21,rivera:35,rng:[19,21],roar_facad:19,robust:[31,32,34,35],robust_scal:[3,9,27],roc_auc:[0,12,35],rohit:35,rong:35,root:[7,33],root_mean_squared_error:12,rotat:10,round:35,run:[0,1,3,4,6,7,8,10,11,12,13,14,15,16,17,18,19,20,21,23,24,25,26,27,29,31,32,33,34,35],run_id:[19,21],run_info:[0,15],run_kei:9,run_or_walk_inform:10,run_valu:[0,9,15],runcount_limit:[2,7,8,23,24,25,26,27],runhistori:[0,9,12,21],runhistory2epm4logcost:[12,21],runhistory2epm:[12,21],runhistory_:9,runinfo:[0,15],runkei:9,runner:[1,4,6,8,9,10,11,12,13,14,15,17,18,20,21],runtim:[9,31],runvalu:[0,9,15],ryul:35,s:[0,7,9,10,12,14,17,21,31,33,34,35],safe:8,safeguard:35,sake:17,same:[0,9,27,30,31,32],samm:[19,27],sampl:[0,4,10,15,31,34],sample_configur:15,sample_weight:0,santo:35,sapronov:35,satyam:35,save:[0,35],scalar:[0,8],scale:[3,26,27,35],scatter:4,scenario:[0,19,21],scenario_dict:[0,19,21],scheduler_fil:17,scheduler_file_nam:17,scientif:[31,32,34],scikit:[0,2,9,10,13,15,30,31,32,34,35],scipi:[0,31,34,35],score:[0,9,10,12,17,18,19,21,27,31,32,34,35],score_func:[0,6,12,27],scorer:0,scoring_funct:[0,6],scratch:0,script:[1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,23,24,25,26,27,31,35],search:[0,15,22,33,35],searchspac:34,second:[0,1,2,3,4,6,7,8,9,10,11,13,14,15,17,18,19,20,21,23,24,25,26,27,31,35],section:[12,31,33,34,35],see:[0,10,11,17,30,31,33,34,35],seed:[0,9,12,15,17,18,19,21,30],seen:[0,31,34],select:[0,9,14,15,21,31,32,35],select_percentile_classif:[11,27],select_percentile_regress:11,select_rates_classif:[11,27],select_rates_regress:11,selected_indic:14,selectpercentileregress:35,selectr:35,self:[0,1,4,6,9,10,11,12,13,14,15,17,18,20,21,23,24,25,26,27],sensibl:0,separ:35,sequenc:35,sequenti:[22,29,31,34],seri:[0,31,34],server:35,servic:33,session:33,set:[0,13,14,17,21,24,31,33,34,35],set_param:0,setattr:24,setup:35,sever:[30,31,34,35],sgd:[4,6,9,11,21,23,27,35],sgdclassifi:[9,11,19,21],sgdregressor:[0,4],shahril:35,shape:[0,27],share:[31,34,35],shell:[17,33],ship:33,shortnam:[23,24,25,26,27,30],should:[0,9,17,31,32,35],show:[1,2,3,4,6,7,9,10,12,13,14,15,17,18,19,20,31,34,35],show_model:[0,1,2,3,4,9,11,19,20,21,23,24,25,26,31,34,35],showcas:13,shown:[11,13,31,34],shrink:[9,27],shrinkag:[9,19,24,25,27,35],shrinkage_condit:25,shrinkage_factor:[9,27],shuffl:[0,9,21,27,35],shut:18,shutdown:35,sigmoid:[3,27],sign:[0,30],signatur:[0,35],signed_data:[23,25,26,30],significantli:31,silenc:8,silva:35,similar:11,simpl:[0,1,4,10,31,34],simpleclassificationpipelin:35,simpli:21,simplifi:[9,31,35],sinc:[0,10,12],singl:[0,11,14,16,17,22,29,31,34,35],single_best_optimization_scor:13,single_best_test_scor:13,single_best_train_scor:13,singlethreadedcli:7,site:21,six:35,size:[0,11,31,34,35],skearn:24,sklean:31,sklearn:[0,6,7,8,10,12,13,14,15,18,19,28,29,31,34,35],sklearn_classifi:[0,1,2,9,11,19,20,21,23,24,25],sklearn_regressor:[0,3,4,9,26],skripniuk:35,sleep:17,slide:32,smac3:[7,33,35],smac4ac:21,smac:[0,7,9,12,13,19,21,31,34,35],smac_ac_facad:21,smac_scenario_arg:[0,2,7,8,23,24,25,26,27],small:31,smaller:12,smbo:[0,35],so:[0,9,10,17,30,31,33,34,35],soft:[14,21],solut:[6,12,31,33],solv:[9,33],solver:[9,19,23,24,25,27],some:[0,10,24,31,34,35],sorber:35,sort:[0,9,35],sort_bi:0,sort_idx:10,sort_ord:0,sort_valu:0,sound:31,sourc:[0,1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,23,24,25,26,27,29,30,31,34],source_id:15,space:[0,7,15,23,26,30,35],spars:[0,24,26,27,30,31,34,35],sparsiti:0,spawn:[18,22,29,31,34,35],specif:[8,31,33],specifi:[0,7,8,14,21,27,30,31,34],speed:[2,7,8,10,23,24,25,26,27,32,35],spend:32,sphinx:[1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,23,24,25,26,27,29,35],sphinx_toolbox:35,sphx_glr_examples_60_search_example_parallel_manual_spawning_python:17,split:[0,14,29,31,34,35],splitter:35,spmatrix:0,springenberg:[31,32,35],sprint_statist:[0,2,9,14,17,18,19,20,21,24,31,34],sqrt:27,squar:0,squared_hing:[1,9,19,21,27],stabl:[2,15],stage:9,stage_nam:9,stale:35,standard:[3,9,27],start:[0,7,9,10,18,31,32,33,35],start_tim:0,startswith:6,starttim:[9,15],startup:35,state:35,statement:35,staticmethod:[23,24,25,26,27],statist:[0,18,19,21,34],statu:[0,6,9,15],statustyp:[9,13,15],stderr:17,stdout:[7,17],stefan:[31,32,35],step:[0,24,31,34],still:[31,34],stock90975:35,stop:[0,9,31,34,35],store:[0,7,9,15,17,31,34,35],str:[0,13,27,30,35],strategi:[0,1,3,9,15,16,19,20,21,23,24,27,29,34,35],stream:7,streamhandl:7,streuhof:35,strict:[33,35],string:[0,2,8,14,31,34],strong:35,structur:7,style:35,subclass:[0,30],submit:[17,35],submodul:[33,35],subobject:0,subprocess:[17,31,35],subsampl:[10,35],subsect:30,subset:[0,11,31],success:[0,2,6,9,14,15,17,18,19,20,22,24,29,31,34,35],successive_halv:21,sudo:33,suffer:[13,31],suffici:[31,34],suggest:[31,33,34],sum:0,support:[0,12,13,31,33,34,35],sure:[2,15,19,21,24,31,33,35],svc:9,svd:[25,27],svm:3,svr:35,swap:34,swig:[31,33],sy:7,symptom:31,synthet:3,system:[31,32,34,35],t:[10,12,31],ta:[0,19,21],ta_kwarg:[19,21],tabl:0,tae:[0,13],tae_runn:[19,21],tae_runner_kwarg:[19,21],tag:[0,8],take:[0,17,18,31,34],talk:[31,32,34],tane:35,tanh:[1,9,20,21,23,27],target:[0,2,8,9,10,13,14,17,18,19,20,21,24,31,34,35],task:[0,15,17,30,31,34],task_id:9,technic:31,techniqu:35,technolog:32,tell:30,templat:35,tempor:7,temporari:[31,35],temporarili:31,teresa:35,termin:[0,31],test:[0,4,10,13,15,31,33,34,35],test_fold:14,test_predict:4,test_siz:[10,13,15],testa:35,text:34,th:9,than:[0,4,20,21,31,35],thank:31,thei:[17,31,34,35],them:[0,10,11,17,31,34,35],therebi:31,therefor:[0,31,33,34],thetal:3,thetau:3,thi:[0,1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,23,24,25,26,27,30,31,32,33,34,35],thick:10,thin:10,third:31,thread:[31,34,35],threadpoolctl:[31,34,35],three:[17,31],threshold_lambda:4,through:[19,30],throughout:[31,34],thu:[10,31],tie:35,tight:31,tight_layout:[4,10],time:[0,1,2,3,4,6,7,8,9,10,11,12,14,15,17,18,19,20,21,23,24,25,26,27,29,30,31,32,34,35],time_left_for_this_task:[0,1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,23,24,25,26,27],timeout:[17,35],timeseri:[31,34],timestamp:[7,13,31,34],timothi:35,titl:[13,31,32,34],tmp:[0,1,3,4,10,11,14,17,18,19,20,21,31,35],tmp_dir:0,tmp_folder:[0,1,3,4,7,10,11,14,17,18,19,20,21,31],to_numer:13,to_numpi:8,to_str:6,tobia:35,togeth:[0,15,35],tol:[1,4,9,11,19,20,21,24,25,27],too:[9,31,35],toolkit:32,top:[0,9],top_k:0,total:[0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,31,34],track:[0,35],tracker:[31,32],tradeoff:34,train:[0,4,9,10,13,14,15,20,27,29,31,34,35],train_loss:[0,9,15],train_predict:4,train_siz:[0,14,21],train_test_split:[1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,23,24,25,26,27,32],trajectori:[31,35],transform:[0,9,24,25,30,31,34,35],travi:35,tree:31,tri:0,trigger:35,truncatedsvd:11,tune:[27,31,32,34,35],tupl:[0,15,23,30,35],turn:[10,24,31,34,35],tutori:[31,34],two:[2,7,8,10,20,23,24,25,26,27,30,31,33,34,35],txt:35,type:[0,1,2,3,4,9,13,16,21,23,25,26,27,29,30,31,34,35],type_of_target:2,typic:0,typo:35,u12:9,u13:9,u16:9,u18:9,u19:9,u20:9,u32:9,u3:9,u5:9,u9:9,u:0,ubuntu:35,un:35,under:[7,31,33,34],underestim:4,underli:[0,9],understand:10,unidimension:[31,34],uniform:[3,9,27],uniform_averag:0,uniformfloat:[3,23,25,26,27],uniformfloathyperparamet:[23,25,26,27],uniforminteg:[3,23,26,27],uniformintegerhyperparamet:[23,26,27],union:0,unit:35,unix:33,unless:0,unlik:32,unlimit:35,unnecessari:35,unseen:[13,14,31,34],unsign:30,unsigned_data:[23,24,25,26,27,30],unsuccess:[31,34],until:[31,34],up:[2,7,8,9,10,11,17,23,24,25,26,27,31,32,33,35],updat:[0,31,33,35],upgrad:35,upper:[0,23,25,26,31,34,35],us:[0,1,2,4,6,7,8,10,11,13,14,15,17,18,20,23,27,29,30,31,32,33,34,35],usag:[0,22,29,31,34,35],user:[0,8,11,13,30,31,32,34,35],userfindingself:35,userwarn:21,usual:[31,33],util:[0,2,25],v:[0,33],val:24,valentin:35,valid:[0,2,9,15,17,18,19,20,24,27,29,31,34,35],validation_fract:[1,4,9,19,20,21,27],valu:[0,3,4,9,10,15,27,30,31,34,35],valueerror:[19,21],valverd:35,van:35,vanilla:[31,34],variabl:[0,31,34,35],varianc:9,variou:[0,12],ve:[31,32],vector:35,verbos:[1,9,19,20,21],veri:[11,31],verifi:33,version:[0,1,4,6,7,9,10,11,12,13,14,15,17,18,20,21,31,33],via:[0,1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,23,24,25,26,27,31,32,33,34,35],vicent:35,view:0,virtual:33,visual:[4,35],vladimir:35,vladislav:35,vote:[14,21],voting_model:[0,9,21],votingclassifi:21,vs:[7,15],wa:[0,9,15,19,33,35],wai:[2,8,31,32,34],walk:[10,30],want:[8,9,17,31,32],ward:[3,27],warm_start:[1,2,3,4,9,11,15,19,20,21,24,25],warn:[8,12,21,35],we:[4,7,9,10,11,12,13,14,15,17,18,26,27,31,32,33,34],web:35,weight:[0,1,3,9,17,19,20,21,23,24,27],weight_1:0,weight_n:0,weird:35,well:[0,9,11,31,33,34,35],wen:35,were:[0,4,9,24,31,35],what:[0,9,10],wheel:[33,35],when:[0,8,10,13,17,24,30,31,34,35],where:[0,9,21,27,31,33,35],wherea:13,whether:[0,9,10,13,30,35],which:[0,7,9,10,11,12,13,14,17,20,21,27,30,31,32,34,35],whiten:[3,9,27],whole:[14,31],why:31,window:35,within:[13,17,31,35],without:[17,31,34,35],woo:35,work:[0,1,4,6,7,8,9,10,11,12,13,14,15,17,18,20,21,30,31,32,33,34,35],worker:[0,18,21,22,29,31,34,35],worker_process:17,workflow:35,workload:17,workstat:31,wors:0,worst:35,worst_possible_result:0,would:[0,11,14,17,31,32,34,35],wrap:[0,31],wrapper:[9,25,30,31,34],written:[7,31],wrong:35,wrongli:35,wrt:[0,10],www:[2,7,8,13],x64:21,x:[0,1,2,3,4,6,7,8,9,11,12,13,14,15,17,18,19,20,21,23,24,25,26,27,31,32,34],x_test:[0,1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,23,24,25,26,27,31,32,34],x_train:[0,1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,23,24,25,26,27,31,32,34],xgboost:35,xiaodong:35,xlabel:4,xlim:4,xtick:10,y:[0,1,2,3,4,6,7,8,9,11,12,13,14,15,17,18,19,20,21,23,24,25,26,27,32,35],y_hat:32,y_optim:0,y_pred:[0,23,24,25,26],y_test:[0,1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,23,24,25,26,27,31,32,34,35],y_train:[0,1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,23,24,25,26,27,31,32,34],y_true:0,yaml:[0,35],yassin:35,yazanobeidi:35,year:[31,32],yet:[0,19,31,34],yield:27,ylabel:4,ylim:4,you:[0,7,8,9,13,15,17,30,31,32,33,34],young:35,your:[1,2,3,4,6,7,8,9,10,11,12,13,14,15,17,18,19,20,21,23,24,25,26,27,31,33,34],yourself:31,zero:[2,31,34],zeyi:35,zip:29,zorder:4},titles:["APIs","Classification","Multi-label Classification","Multi-output Regression","Regression","Computation times","Metrics","Logging and debugging","Feature Types","Obtain run information","Model Explanation","Interpretable models","Metrics","Performance-over-time plot","Resampling Strategies","Fit a single configuration","Computation times","Parallel Usage: Spawning workers from the command line","Parallel Usage  on a single machine","Random Search","Sequential Usage","Successive Halving","Computation times","Extending Auto-Sklearn with Classification Component","Extending Auto-Sklearn with Data Preprocessor Component","Extending Auto-Sklearn with Preprocessor Component","Extending Auto-Sklearn with Regression Component","Restricting the number of hyperparameters for an existing component","Computation times","Examples","Extending auto-sklearn","FAQ","auto-sklearn","Installation","Manual","Releases"],titleterms:{"0":[17,32,35],"1":[9,10,17,35],"10":35,"11":35,"12":35,"13":35,"14":35,"2":[9,10,17,32,35],"3":[10,17,35],"4":35,"5":35,"6":35,"7":35,"8":35,"9":35,"final":[1,2,3,4,6,8,11,13,14,20],"import":[10,17],"new":32,The:[31,34],about:[2,9],accuraci:[12,23],addit:12,advanc:29,again:14,all:17,also:21,an:[15,21,27],anaconda:33,api:0,approach:21,ar:17,argument:12,auto:[1,2,3,4,9,11,17,20,21,23,24,25,26,27,30,32,33],avail:[11,12],background:17,basic:29,best:9,budget:21,build:[1,2,3,4,6,7,8,9,10,11,13,18,20,21],built:0,callback:21,can:21,cite:32,classif:[0,1,2,11,23,30],classifi:[1,2,6,7,8,9,11,13,18,19,20,21,23,27],client:17,close:17,command:17,commun:17,compat:33,compon:[9,23,24,25,26,27,30],comput:[5,10,16,22,28,34],conda:33,condit:10,config:7,configur:[3,15,25,27,31],construct:[1,2,3,4,11,20],contribut:32,contributor:35,creat:[7,10,17,23,24,25],cross:[14,21],custom:12,data:[1,2,3,4,6,7,8,9,10,11,12,13,14,15,18,19,20,21,23,24,26,27],dataset:[24,25],debug:[7,31],defin:[15,21],depend:10,detail:9,docker:33,ensembl:[1,2,3,4,6,8,11,13,14,20,31,34],error:12,estim:15,exampl:[12,29,32],exist:27,expect:10,explan:10,extend:[23,24,25,26,30],extens:[0,29],faq:31,featur:[8,10,30],fifth:12,file:31,first:12,fit:[1,3,4,6,7,8,9,11,13,15,18,19,20,21,23,24,25,26,27],forest:27,forg:33,found:[1,2,3,4,9],fourth:12,from:[17,32],gener:[26,31],get:[1,2,3,4,6,8,11,13,14,20],get_hyperparameter_search_spac:30,get_properti:30,halv:21,holdout:14,hyperparamet:27,ic:10,imag:33,implement:26,individu:10,inform:9,inspect:[9,34],instal:33,instanti:21,interfac:0,interpret:11,issu:31,iter:21,kernel:26,krr:26,label:2,lda:25,learn:[14,31],licens:32,limit:34,line:17,list:12,load:[1,2,3,4,6,7,8,9,10,11,12,13,14,15,18,19,20,21,23,27],log:[7,31],machin:18,maco:33,main:0,manag:31,manual:[32,34],materi:32,meta:31,metric:[0,6,12],mix:21,mlp:23,model:[1,2,3,4,9,10,11,24,25,26,34],modul:0,more:10,multi:[2,3],next:21,nopreprocess:24,number:27,object:14,obtain:9,one:10,other:[31,34],output:[3,31],over:13,own:12,parallel:[17,18,34],part:[9,10],partial:10,pd:10,pdp:10,perform:[13,14,24],permut:10,plot:[4,10,13],predefin:12,predict:[4,9,24,25,26],preprocess:[24,30],preprocessor:[11,24,25],present:32,print:[1,2,3,4,11,12,20,23,24,25,26,27],procedur:31,provid:15,python:17,random:[19,27],refit:14,regress:[0,3,4,26,30],regressor:[3,4],releas:35,report:9,requir:33,resampl:14,resourc:[31,34],restrict:27,result:[31,34],ridg:26,roar:19,run:[2,9],s:27,schedul:17,scikit:14,score:[1,2,3,4,6,8,11,13,14,20,24,25,26],search:[9,19,29,31,34],second:12,see:21,select:34,sequenti:20,setup:17,show:11,singl:[15,18],sklearn:[1,2,3,4,9,11,17,20,21,23,24,25,26,27,30,32,33],sourc:33,space:[3,25,27,31,34],spawn:17,splitter:14,start:17,statement:17,statist:[2,9,23,24,25,26],strategi:14,subclass:27,subsampl:21,success:21,successivehalv:21,system:33,test:23,than:10,third:12,time:[5,13,16,22,28],tutori:32,two:17,type:8,ubuntu:33,until:17,us:[9,12,19,21,25,26],usag:[17,18,20],user:15,v0:35,valid:[14,21],version:35,view:[1,2,3,4],wait:17,we:21,window:33,without:24,worker:17,write:30,x:35}})
\ No newline at end of file