Merge pull request #21 from PanyiDong/dev

prepare for v0.2.3

.github/workflows/build-nn.yml

@@ -1,9 +1,10 @@
-name: NN
+name: build_nn
 on:
   - pull_request
+  - push
 jobs:
   run:
-    runs-on: self-hosted
+    runs-on: ubuntu-latest
     env:
       OS: ubuntu-latest
       PYTHON: "3.9"
@@ -13,7 +14,8 @@ jobs:
         uses: actions/setup-python@master
         with:
           python-version: 3.9
-      - name: "run tests (with neural network)"
+      - name: "run builds (with neural network)"
         run: |
+          python -m pip install --upgrade pip
           pip install pytest
           pip install -e .[nn]

.github/workflows/build-windows.yml

@@ -14,7 +14,8 @@ jobs:
         uses: actions/setup-python@master
         with:
           python-version: 3.9
-      - name: "run tests (without neural network)"
+      - name: "run builds (without neural network)"
         run: |
+          python -m pip install --upgrade pip
           pip install pytest
           pip install -e .[normal]

.github/workflows/build.yml

@@ -14,7 +14,8 @@ jobs:
         uses: actions/setup-python@master
         with:
           python-version: 3.9
-      - name: "run tests (without neural network)"
+      - name: "run builds (without neural network)"
         run: |
+          python -m pip install --upgrade pip
           pip install pytest
           pip install -e .[normal]

.github/workflows/codecov.yml

@@ -15,11 +15,12 @@ jobs:
           python-version: 3.9
       - name: "generate report"
         run: |
+          python -m pip install --upgrade pip
           pip install coverage pytest
-          pip install -e .[normal]
+          pip install -e .[extended]
           coverage run -m pytest
       - name: Upload coverage to Codecov
-        uses: codecov/codecov-action@v1
+        uses: codecov/codecov-action@v3
         with:
           flags: uitests
           fail_ci_if_error: false

.github/workflows/tests-nn.yml → .github/workflows/tests-dev.yml

@@ -1,4 +1,4 @@
-name: nn_test
+name: dev_test
 on:
   - pull_request
 jobs:
@@ -15,6 +15,7 @@ jobs:
           python-version: 3.9
       - name: "run tests (with neural network)"
         run: |
+          python -m pip install --upgrade pip
           pip install pytest
-          pip install -e .[nn]
-          pytest tests/
+          pip install -e .[dev]
+          pytest tests/test_hpo

.github/workflows/tests-windows.yml

@@ -15,6 +15,7 @@ jobs:
           python-version: 3.9
       - name: "run tests (without neural network)"
         run: |
+          python -m pip install --upgrade pip
           pip install pytest
-          pip install -e .[normal]
+          pip install -e .[extended]
           pytest tests/

.github/workflows/tests.yml

@@ -15,6 +15,7 @@ jobs:
           python-version: 3.9
       - name: "run tests (without neural network)"
         run: |
+          python -m pip install --upgrade pip
           pip install pytest
-          pip install -e .[normal]
+          pip install -e .[extended]
           pytest tests/

.gitignore

@@ -16,4 +16,4 @@ htmlcov/*
 .pytest_cache/
 test.py
 *.so
-VERSION
+*.json

InsurAutoML/VERSION

@@ -0,0 +1 @@
+0.2.3

InsurAutoML/__init__.py

@@ -4,14 +4,14 @@
 GitHub: https://github.com/PanyiDong/
 Mathematics Department, University of Illinois at Urbana-Champaign (UIUC)
 
-Project: My_AutoML
-Latest Version: 0.2.0
-Relative Path: /My_AutoML/__init__.py
-File Created: Friday, 25th February 2022 6:13:42 pm
+Project: InsurAutoML
+Latest Version: 0.2.3
+Relative Path: /InsurAutoML/__init__.py
+File: __init__.py
 Author: Panyi Dong (panyid2@illinois.edu)
 
 -----
-Last Modified: Sunday, 7th August 2022 11:00:45 am
+Last Modified: Tuesday, 15th November 2022 4:04:57 pm
 Modified By: Panyi Dong (panyid2@illinois.edu)
 
 -----
@@ -38,7 +38,10 @@
 SOFTWARE.
 """
 
-__version__ = "0.2.0"
+# read version from file
+from .version import _get_version
+
+__version__ = _get_version()
 
 from ._base import no_processing, load_data
 from ._utils import (
@@ -120,7 +123,7 @@
 #     GeneticAlgorithm,
 # )
 
-# extracted from autosklearn
+# extracted from sklearn
 # not all used in the pipeline
 # from ._feature_selection import (
 #     Densifier,
@@ -229,7 +232,7 @@
     "FeatureFilter",
     "ASFFS",
     "GeneticAlgorithm",
-    "densifier",  # from autosklearn
+    "densifier",  # from sklearn
     "extra_trees_preproc_for_classification",
     "extra_trees_preproc_for_regression",
     "fast_ica",

InsurAutoML/_archive/_multimodal.py

@@ -0,0 +1,181 @@
+"""
+File: _multimodal.py
+Author: Panyi Dong
+GitHub: https://github.com/PanyiDong/
+Mathematics Department, University of Illinois at Urbana-Champaign (UIUC)
+
+Project: InsurAutoML
+Latest Version: 0.2.3
+Relative Path: /InsurAutoML/_archive/_multimodal.py
+File: _multimodal.py
+Author: Panyi Dong (panyid2@illinois.edu)
+
+-----
+Last Modified: Sunday, 13th November 2022 12:38:01 am
+Modified By: Panyi Dong (panyid2@illinois.edu)
+
+-----
+MIT License
+
+Copyright (c) 2022 - 2022, Panyi Dong
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.
+"""
+
+from typing import List, Tuple, Dict, Union, Any
+import numpy as np
+import pandas as pd
+import torch
+import torch.nn.functional as F
+from transformers import AutoTokenizer
+
+
+def TxtTokenize(
+    data: Union[pd.DataFrame, np.ndarray],
+    model_name: str = "bert-base-uncased",
+    max_length: int = 128,
+):
+    # get tokenizer
+    tokenizer = AutoTokenizer.from_pretrained(model_name)
+    # tokenizer mapping function
+    def _tokenizer(text: str) -> List[int]:
+        return tokenizer(
+            text,
+            max_length=max_length,
+            padding="max_length",
+            truncation=True,
+            return_attention_mask=False,
+            return_token_type_ids=False,
+        )["input_ids"]
+
+    # tokenize the data
+    data = data.applymap(lambda x: _tokenizer(x))
+
+    # convert to tensor
+    return torch.cat([torch.tensor(data[col].to_list()) for col in data.columns], dim=1)
+
+
+# The purpose of this function is to make sure the unique values are consecutive
+# Example: [8, 9, 4, 9, 4, 8, 2, 0, 6, 1] -> [5, 6, 3, 6, 3, 5, 2, 0, 4, 1]
+# As seen, the missing 10 classes are reduced to 7 classes
+def CatConsec(
+    data: torch.Tensor,
+):
+    # get sorted (by ascending order) and indices
+    sorted, indices = data.sort(dim=0, descending=False)
+    # replicate the first row to make sure the first/second row diff is 0
+    data_pad = torch.cat([sorted[0, :].unsqueeze(0), sorted], dim=0)
+    # get the difference between each row and reduce by 1
+    # since the normal step is 1, we don't need modification for those
+    data_diff = F.relu(torch.diff(data_pad, dim=0) - 1)
+    # get the cumulative sum of the difference
+    data_cumsum = torch.cumsum(data_diff, dim=0)
+    # initialize the result
+    result = torch.zeros(data.size(), dtype=torch.int64)
+
+    # distribute the indices to the unique values
+    return result.scatter_(0, indices, (sorted - data_cumsum))
+
+
+# Update: Nov. 13, 2022
+# Function version decrypted to avoid inconsistency for train/test split
+def CatOffsetEncoding(
+    data: Union[pd.DataFrame, np.ndarray, torch.Tensor],
+    unique_classes: List[int] = None,
+    starting_offset: int = 0,
+) -> torch.Tensor:
+
+    """
+    This function is used to convert categorical data into Encoding.
+
+    Parameters
+    ----------
+    data: input data, must be a 2D tensor.
+
+    unique_classes: the number of unique classes for each column.
+
+    starting_offset: the starting offset for each column. This is used to make sure
+    different encodings have different range of values.
+    """
+
+    # convert data to torch tensor
+    data = torch.tensor(data.values if isinstance(data, pd.DataFrame) else data)
+
+    # check data dimension
+    if len(data.size()) != 2:
+        raise TypeError(
+            "Data must be a 2D tensor. Got {}D tensor.".format(len(data.size()))
+        )
+
+    # if unique_classes not passed, get unique classes from data
+    if unique_classes is None:
+        # if not passed, check the unique values to start from 0 with step 1
+        # make sure the unique values are consecutive
+        data = CatConsec(data)
+
+        unique_classes = [len(torch.unique(t)) for t in torch.unbind(data.T)]
+
+    # get offset encoding
+    # cumsum of each column number of unique classes
+    cat_offset = F.pad(torch.tensor(unique_classes), (1, 0), value=starting_offset)
+    cat_offset = cat_offset.cumsum(dim=-1)[:-1]
+
+    return data + cat_offset
+
+
+def NumOffsetEncoding(
+    data: Union[pd.DataFrame, np.ndarray, torch.Tensor],
+    num_classes: int = 10,
+    starting_offset: int = 0,
+) -> torch.Tensor:
+
+    """
+    This function is used to convert numerical data into categorical Encoding.
+
+    Parameters
+    ----------
+    data: input data, must be a 2D tensor.
+
+    num_classes: the number of unique classes for each column.
+
+    starting_offset: the starting offset for each column. This is used to make sure
+    different encodings have different range of values.
+    """
+
+    # convert data to torch tensor
+    data = torch.tensor(data.values if isinstance(data, pd.DataFrame) else data)
+
+    # check data dimension
+    if len(data.size()) != 2:
+        raise TypeError(
+            "Data must be a 2D tensor. Got {}D tensor.".format(len(data.size()))
+        )
+
+    # convert numerical data into categorical data
+    # use min-max normalization and multiply by num_classes
+    # then convert to int to get categories
+    vmax = torch.max(data, dim=0)[0]
+    vmin = torch.min(data, dim=0)[0]
+
+    data = (data - vmin.unsqueeze(0)) / (vmax - vmin).unsqueeze(0) * num_classes
+    data = data.int()
+
+    return CatOffsetEncoding(
+        data, [num_classes for _ in range(data.size()[1])], starting_offset
+    )

InsurAutoML/_balancing/__init__.py

@@ -1,17 +1,17 @@
 """
-File: __init__.py
+File Name: __init__.py
 Author: Panyi Dong
 GitHub: https://github.com/PanyiDong/
 Mathematics Department, University of Illinois at Urbana-Champaign (UIUC)
 
-Project: My_AutoML
-Latest Version: 0.2.0
-Relative Path: /My_AutoML/_balancing/__init__.py
-File Created: Wednesday, 6th April 2022 12:20:45 am
+Project: InsurAutoML
+Latest Version: 0.2.3
+Relative Path: /InsurAutoML/_balancing/__init__.py
+File Created: Monday, 24th October 2022 11:56:57 pm
 Author: Panyi Dong (panyid2@illinois.edu)
 
 -----
-Last Modified: Monday, 24th October 2022 10:52:57 pm
+Last Modified: Monday, 14th November 2022 6:57:33 pm
 Modified By: Panyi Dong (panyid2@illinois.edu)
 
 -----

InsurAutoML/_balancing/_legacy.py

@@ -1,17 +1,17 @@
 """
-File: _legacy.py
+File Name: _legacy.py
 Author: Panyi Dong
 GitHub: https://github.com/PanyiDong/
 Mathematics Department, University of Illinois at Urbana-Champaign (UIUC)
 
-Project: My_AutoML
-Latest Version: 0.2.0
-Relative Path: /My_AutoML/_balancing/_legacy.py
-File Created: Friday, 8th April 2022 9:19:36 pm
+Project: InsurAutoML
+Latest Version: 0.2.3
+Relative Path: /InsurAutoML/_balancing/_legacy.py
+File Created: Monday, 24th October 2022 11:56:57 pm
 Author: Panyi Dong (panyid2@illinois.edu)
 
 -----
-Last Modified: Monday, 24th October 2022 10:56:54 pm
+Last Modified: Monday, 14th November 2022 6:57:54 pm
 Modified By: Panyi Dong (panyid2@illinois.edu)
 
 -----