FS change

EpistasisLab · Aug 16, 2023 · fbbdf60 · fbbdf60
1 parent 854dae2
commit fbbdf60
Show file tree

Hide file tree

Showing 15 changed files with 21 additions and 20 deletions.
diff --git a/_version.py b/_version.py
@@ -1 +1 @@
-__version__ = '0.1.5-alpha'
+__version__ = '0.1.6-alpha'
diff --git a/autoqtl/__pycache__/__init__.cpython-310.pyc b/autoqtl/__pycache__/__init__.cpython-310.pyc
diff --git a/autoqtl/__pycache__/autoqtl.cpython-310.pyc b/autoqtl/__pycache__/autoqtl.cpython-310.pyc
diff --git a/autoqtl/__pycache__/base.cpython-310.pyc b/autoqtl/__pycache__/base.cpython-310.pyc
diff --git a/autoqtl/__pycache__/decorators.cpython-310.pyc b/autoqtl/__pycache__/decorators.cpython-310.pyc
diff --git a/autoqtl/__pycache__/export_utils.cpython-310.pyc b/autoqtl/__pycache__/export_utils.cpython-310.pyc
diff --git a/autoqtl/__pycache__/gp_deap.cpython-310.pyc b/autoqtl/__pycache__/gp_deap.cpython-310.pyc
diff --git a/autoqtl/__pycache__/gp_types.cpython-310.pyc b/autoqtl/__pycache__/gp_types.cpython-310.pyc
diff --git a/autoqtl/__pycache__/operator_utils.cpython-310.pyc b/autoqtl/__pycache__/operator_utils.cpython-310.pyc
diff --git a/autoqtl/base.py b/autoqtl/base.py
@@ -1507,22 +1507,22 @@ def _summary_of_best_pipeline(self, features_dataset1, target_dataset1, features
 
                     print(self.print_pipeline_hyper(pipeline))
 
-                    # print the names of the selected features if FS exists in the pipeline
-                    pipeline_sklearn = self._toolbox.compile(expr=pipeline)
-                    with warnings.catch_warnings():
-                        warnings.simplefilter("ignore")
-                    pipeline_sklearn.fit(selected_features, selected_target)
-                    flag = False
-                    for name, transformer in pipeline_sklearn.steps:
-                        if name=='variancethreshold' or name=='selectpercentile' or name=='featureencodingfrequencyselector':
-                            selected_features_mask = transformer.get_support()
-                            selected_features_names = features_dataset1.columns[selected_features_mask]
-                            print("No of features selected in the pipeline: ", len(selected_features_names))
-                            print("Feature names selected in the pipeline: ", selected_features_names.tolist())
-                            flag = True
-
-                    if flag==False:
-                        print("No feature selectors were selected in the pipeline")
+                    # # print the names of the selected features if FS exists in the pipeline
+                    # pipeline_sklearn = self._toolbox.compile(expr=pipeline)
+                    # with warnings.catch_warnings():
+                    #     warnings.simplefilter("ignore")
+                    # pipeline_sklearn.fit(selected_features, selected_target)
+                    # flag = False
+                    # for name, transformer in pipeline_sklearn.steps:
+                    #     if name=='variancethreshold' or name=='selectpercentile' or name=='featureencodingfrequencyselector':
+                    #         selected_features_mask = transformer.get_support()
+                    #         selected_features_names = features_dataset1.columns[selected_features_mask]
+                    #         print("No of features selected in the pipeline: ", len(selected_features_names))
+                    #         print("Feature names selected in the pipeline: ", selected_features_names.tolist())
+                    #         flag = True
+
+                    # if flag==False:
+                    #     print("No feature selectors were selected in the pipeline")
 
                     self.score1_final_list.append(pipeline_scores.wvalues[0])
                     self.score2_final_list.append(pipeline_scores.wvalues[1])

diff --git a/autoqtl/builtins/__pycache__/__init__.cpython-310.pyc b/autoqtl/builtins/__pycache__/__init__.cpython-310.pyc
diff --git a/autoqtl/builtins/__pycache__/feature_encoding_frequency_selector.cpython-310.pyc b/autoqtl/builtins/__pycache__/feature_encoding_frequency_selector.cpython-310.pyc
diff --git a/autoqtl/builtins/__pycache__/genetic_encoders.cpython-310.pyc b/autoqtl/builtins/__pycache__/genetic_encoders.cpython-310.pyc
diff --git a/autoqtl/config/__pycache__/regressor.cpython-310.pyc b/autoqtl/config/__pycache__/regressor.cpython-310.pyc
diff --git a/tests/autoqtl_test.py b/tests/autoqtl_test.py
@@ -10,7 +10,8 @@
 
 # Insert the path of the currently cloned github repo folder
 #sys.path.append("Insert path")
-sys.path.append("/Users/ghosha/Documents/VSCode Projects/autoqtl-1")
+#sys.path.append("/Users/ghosha/Documents/VSCode Projects/autoqtl-1")
+sys.path.append("/Users/ghosha/Documents/VSCode Projects/autoqtl")
 
 import autoqtl
 
@@ -156,7 +157,7 @@ def test_summary_of_best_pipeline():
         verbosity=3
     )
     #autoqtl_obj.fit(features_dataset1, target_dataset1, features_dataset2, target_dataset2)
-    autoqtl_obj.fit(features_80, target_80,32, 0.5)
+    autoqtl_obj.fit(features_80, target_80,42, 0.5)
     #autoqtl_obj._summary_of_best_pipeline(features_dataset1, target_dataset2, features_dataset2, target_dataset2)
     assert isinstance(autoqtl_obj._optimized_pipeline, creator.Individual)
     #autoqtl_obj.get_feature_importance(features_dataset1, target_dataset1, random_state=0)
@@ -168,7 +169,7 @@ def test_summary_of_best_pipeline():
     # this method outputs a text file named EvaluationOnHoldout.txt with the performance on holdout data for each pareto front pipeline
     #autoqtl_obj.get_test_r2(features_dataset1, target_dataset1, features_dataset2, target_dataset2 ,features_20, target_20, features_80, target_80, test_X, test_y)
     #autoqtl_obj.shap_feature_importance(test_X, test_y, random_state=0)
-    autoqtl_obj.get_final_output(test_X, test_y, features_20, target_20, file_path = '/Users/ghosha/Library/CloudStorage/Box-Box/AutoQTL/Mouse Data/Analysis/AutoQTL Outputs/')
+    autoqtl_obj.get_final_output(test_X, test_y, features_20, target_20, file_path = '/Users/ghosha/Library/CloudStorage/Box-Box/AutoQTL/Mouse Data/Analysis/AutoQTL Outputs/Package Tests/')
     #autoqtl_obj.average_feature_importance(test_X, test_y)
 
     # # individual SHAP plot