Merge pull request #260 from BCG-X-Official/dev/2.3rc0

BUILD: release sklearndf 2.3rc0
BCG-X-Official · Nov 14, 2023 · cde262d · cde262d
2 parents 8335d8c + 50f3cc1
commit cde262d
Show file tree

Hide file tree

Showing 37 changed files with 270 additions and 247 deletions.
diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml
@@ -5,7 +5,7 @@ repos:
       - id: isort
 
   - repo: https://github.com/psf/black
-    rev: 22.8.0
+    rev: 23.10.1
     hooks:
       - id: black
         language: python_venv
@@ -41,5 +41,5 @@ repos:
         language: python_venv
         language_version: python39
         additional_dependencies:
-          - numpy~=1.22
+          - numpy~=1.24
           - gamma-pytools~=2.1
diff --git a/README.rst b/README.rst
@@ -31,9 +31,9 @@ To this end, *sklearndf* enhances scikit-learn's estimators as follows:
 The following quickstart guide provides a minimal example workflow to get up and running
 with *sklearndf*.
 For additional tutorials and the API reference,
-see the `sklearndf documentation <https://bcg-gamma.github.io/sklearndf/>`__.
+see the `sklearndf documentation <https://bcg-x-official.github.io/sklearndf/>`__.
 Changes and additions to new versions are summarized in the
-`release notes <https://bcg-gamma.github.io/sklearndf/release_notes.html>`__.
+`release notes <https://bcg-x-official.github.io/sklearndf/release_notes.html>`__.
 
 
 Installation
@@ -243,20 +243,20 @@ Contributing
 
 Contributions to *sklearndf* are welcome and appreciated.
 For any bug reports or feature requests/enhancements please use the appropriate
-`GitHub form <https://github.com/BCG-Gamma/sklearndf/issues>`_, and if you wish to do
+`GitHub form <https://github.com/BCG-X-Official/sklearndf/issues>`_, and if you wish to do
 so, please open a PR addressing the issue.
 
 We do ask that for any major changes please discuss these with us first via an issue.
 
 For further information on contributing please see our
-`contribution guide <https://bcg-gamma.github.io/sklearndf/contribution_guide.html>`__.
+`contribution guide <https://bcg-x-official.github.io/sklearndf/contribution_guide.html>`__.
 
 
 License
 -------
 
 *sklearndf* is licensed under Apache 2.0 as described in the
-`LICENSE <https://github.com/BCG-Gamma/sklearndf/blob/develop/LICENSE>`_ file.
+`LICENSE <https://github.com/BCG-X-Official/sklearndf/blob/develop/LICENSE>`_ file.
 
 
 Acknowledgements
@@ -284,7 +284,7 @@ or have a look at
 .. |pypi| image:: https://badge.fury.io/py/sklearndf.svg
    :target: https://pypi.org/project/sklearndf/
 
-.. |azure_build| image:: https://dev.azure.com/gamma-facet/facet/_apis/build/status/BCG-Gamma.sklearndf?repoName=BCG-Gamma%2Fsklearndf&branchName=develop
+.. |azure_build| image:: https://dev.azure.com/gamma-facet/facet/_apis/build/status/BCG-X-Official.sklearndf?repoName=BCG-X-Official%2Fsklearndf&branchName=develop
    :target: https://dev.azure.com/gamma-facet/facet/_build?definitionId=8&_a=summary
 
 .. |azure_code_cov| image:: https://img.shields.io/azure-devops/coverage/gamma-facet/facet/8/2.1.x
@@ -297,7 +297,7 @@ or have a look at
    :target: https://github.com/psf/black
 
 .. |made_with_sphinx_doc| image:: https://img.shields.io/badge/Made%20with-Sphinx-1f425f.svg
-   :target: https://bcg-gamma.github.io/sklearndf/index.html
+   :target: https://bcg-x-official.github.io/sklearndf/index.html
 
 .. |license_badge| image:: https://img.shields.io/badge/License-Apache%202.0-olivegreen.svg
    :target: https://opensource.org/licenses/Apache-2.0

diff --git a/RELEASE_NOTES.rst b/RELEASE_NOTES.rst
@@ -7,6 +7,22 @@ Release Notes
 .. |nbsp| unicode:: 0xA0
    :trim:
 
+*sklearndf* 2.3
+---------------
+
+
+2.3.0
+~~~~~
+
+*sklearndf* 2.3 adds support for
+`scikit-learn 1.3 <https://scikit-learn.org/1.3>`_
+and drops support for *scikit-learn* |nbsp| 0.24.
+
+- API: add DF wrapper classe :class:`.HDBSCANDF` for native estimator
+  :class:`~sklearn.cluster.HDBSCAN`
+- API: add DF wrapper class :class:`.TargetEncoderDF` for native estimator
+  :class:`~sklearn.preprocessing.TargetEncoder`
+
 
 *sklearndf* 2.2
 ---------------

diff --git a/azure-pipelines.yml b/azure-pipelines.yml
@@ -1,9 +1,9 @@
 trigger:
-  - 2.2.x
+  - 2.3.x
   - release/*
 
 pr:
-  - 2.2.x
+  - 2.3.x
   - release/*
 
 pool:
@@ -24,8 +24,8 @@ resources:
   repositories:
     - repository: pytools
       type: github
-      endpoint: BCG-Gamma
-      name: BCG-Gamma/pytools
+      endpoint: BCG-X-Official
+      name: BCG-X-Official/pytools
       ref: 2.1.x
 
 variables:
@@ -65,7 +65,7 @@ stages:
               versionSpec: '3.9'
             displayName: 'use Python 3.9'
           - script: |
-              python -m pip install black~=22.8
+              python -m pip install black~=23.10.1
               python -m black --check .
             displayName: 'Run black'
       - job:
@@ -655,6 +655,7 @@ stages:
 
                 cd $(System.DefaultWorkingDirectory)
                 echo "Checking out github-pages"
+                git fetch origin github-pages --depth=1
                 git checkout --track origin/github-pages
 
                 # make sure we have a docs directory
@@ -676,6 +677,7 @@ stages:
 
                 cd $(System.DefaultWorkingDirectory)
                 echo "Checking out $(branchName)"
+                git fetch origin $(branchName) --depth=1
                 git checkout $(branchName)
 
                 # install micromamba

diff --git a/condabuild/meta.yaml b/condabuild/meta.yaml
@@ -61,14 +61,14 @@ test:
     - pytest -vs test
 
 about:
-  home: https://github.com/BCG-Gamma/sklearndf
+  home: https://github.com/BCG-X-Official/sklearndf
   license: Apache Software License v2.0
   license_file: LICENSE
   description: |
     sklearndf is an open source library designed to address a common need with
     scikit-learn: the outputs of transformers are numpy arrays, even when the input
     is a data frame. However, to inspect a model it is essential to keep track of
     the feature names.
-  dev_url: https://github.com/BCG-Gamma/sklearndf
-  doc_url: https://bcg-gamma.github.io/sklearndf/
-  doc_source_url: https://github.com/BCG-Gamma/sklearndf/blob/develop/README.rst
+  dev_url: https://github.com/BCG-X-Official/sklearndf
+  doc_url: https://bcg-x-official.github.io/sklearndf/
+  doc_source_url: https://github.com/BCG-X-Official/sklearndf/blob/develop/README.rst
diff --git a/environment.yml b/environment.yml
@@ -11,10 +11,10 @@ dependencies:
   - matplotlib        ~= 3.7
   - numpy             ~= 1.24
   - pandas            ~= 2.0
-  - pip               ~= 23.0
+  - pip               ~= 23.3
   - python            ~= 3.9
   - scikit-learn      ~= 1.2.0
-  - scipy             ~= 1.10
+  - scipy             ~= 1.11
   - xgboost           ~= 1.7
   - pip:
      - arfs           ~= 1.1
@@ -27,8 +27,8 @@ dependencies:
   - sphinx-autodoc-typehints ~= 1.19.2
   - pydata-sphinx-theme      ~= 0.8.1
   # notebooks
-  - ipywidgets ~= 8.0
-  - jupyterlab ~= 3.5
-  - openpyxl   ~= 3.0
-  - seaborn    ~= 0.12
+  - ipywidgets ~= 8.1
+  - jupyterlab ~= 3.6
+  - openpyxl   ~= 3.1
+  - seaborn    ~= 0.13
   - tableone   ~= 0.7
diff --git a/pypi_description.rst b/pypi_description.rst
@@ -23,7 +23,7 @@ License
 ---------------------------
 
 *sklearndf* is licensed under Apache 2.0 as described in the
-`LICENSE <https://github.com/BCG-Gamma/sklearndf/blob/develop/LICENSE>`_ file.
+`LICENSE <https://github.com/BCG-X-Official/sklearndf/blob/develop/LICENSE>`_ file.
 
 .. Begin-Badges
 
@@ -40,7 +40,7 @@ License
     :target: https://github.com/psf/black
 
 .. |made_with_sphinx_doc| image:: https://img.shields.io/badge/Made%20with-Sphinx-1f425f.svg
-    :target: https://bcg-gamma.github.io/sklearndf/index.html
+    :target: https://bcg-x-official.github.io/sklearndf/index.html
 
 .. |license_badge| image:: https://img.shields.io/badge/License-Apache%202.0-olivegreen.svg
     :target: https://opensource.org/licenses/Apache-2.0

diff --git a/pyproject.toml b/pyproject.toml
@@ -8,18 +8,18 @@ exclude = [".idea", "tmp", "dist", ".tox", ".pytest_cache"]
 [tool.flit.metadata]
 module = "sklearndf"
 author = "Boston Consulting Group (BCG)"
-home-page = "https://github.com/BCG-Gamma/sklearndf"
+home-page = "https://github.com/BCG-X-Official/sklearndf"
 description-file = "pypi_description.rst"
 dist-name = "sklearndf"
 license = "Apache Software License v2.0"
 
 requires = [
     "gamma-pytools  ~=2.1",
-    "numpy          >=1.21",  # cannot use ~= due to conda bug
+    "numpy          >=1.21,<2a",  # cannot use ~= due to conda bug
     "packaging      >=20",
     "pandas         >=1",
-    "scikit-learn   >=0.24,<1.3a",
-    "scipy          ~=1.4",
+    "scikit-learn   >=1,<1.4a",
+    "scipy          ~=1.6",
 ]
 
 requires-python = ">=3.7,<4a"
@@ -61,8 +61,8 @@ docs = [
 ]
 
 [tool.flit.metadata.urls]
-Documentation = "https://bcg-gamma.github.io/sklearndf/"
-Repository = "https://github.com/BCG-Gamma/sklearndf"
+Documentation = "https://bcg-x-official.github.io/sklearndf/"
+Repository = "https://github.com/BCG-X-Official/sklearndf"
 
 [build]
 # comma-separated list of packages to be built from source in pip min builds
@@ -73,12 +73,12 @@ no-binary.min = ["matplotlib"]
 boruta         = "~=0.3.0"
 gamma-pytools  = "~=2.1.0"
 lightgbm       = "~=3.0.0"
-numpy          = "==1.21.6"  # cannot use ~= due to conda bug
+numpy          = "==1.21.6"        # cannot use ~= due to conda bug
 packaging      = "~=20.9"
 pandas         = "~=1.1.5"
-python         = ">=3.7.12,<3.8a"    # cannot use ~= due to conda bug
-scipy          = "~=1.4.1"
-scikit-learn   = "~=0.24.2"
+python         = ">=3.7.12,<3.8a"  # cannot use ~= due to conda bug
+scipy          = "~=1.6.3"
+scikit-learn   = "~=1.0.2"
 xgboost        = "~=1.0.2"
 # additional minimum requirements of gamma-pytools
 joblib         = "~=0.14.1"
@@ -90,12 +90,12 @@ typing_inspect = "~=0.4.0"
 arfs           = "~=1.1"
 gamma-pytools  = "~=2.1"
 lightgbm       = "~=3.3"
-numpy          = ">=1.23"  # cannot use ~= due to conda bug
+numpy          = ">=1.24,<2a"     # cannot use ~= due to conda bug
 packaging      = ">=20"
 pandas         = "~=2.0"
-python         = ">=3.9,<3.11a"   # cannot use ~= due to conda bug; arfs <=1.0.7 does not support python >=3.11
-scikit-learn   = "~=1.1,<1.3a" # todo: replace with ~=1.2.0 once it is released
-scipy          = "~=1.8"
+python         = ">=3.11,<3.12a"  # cannot use ~= due to conda bug
+scikit-learn   = "~=1.3.2"
+scipy          = "~=1.11"
 xgboost        = "~=1.5"
 # additional maximum requirements of gamma-pytools
 joblib         = "~=1.1"

diff --git a/sphinx/make.py b/sphinx/make.py
@@ -9,14 +9,13 @@
 
 
 if __name__ == "__main__":
-
     # noinspection PyUnusedLocal
     def run_make(branch: str, working_directory: str) -> None:
         """Stub, overwritten by bootstrap.py"""
 
     # run the common make file available in the pytools repo
     with request.urlopen(
-        f"https://raw.githubusercontent.com/BCG-Gamma/pytools/{BRANCH}"
+        f"https://raw.githubusercontent.com/BCG-X-Official/pytools/{BRANCH}"
         f"/sphinx/base/bootstrap.py"
     ) as response:
         exec(response.read().decode("utf-8"), globals())

diff --git a/sphinx/source/conf.py b/sphinx/source/conf.py
@@ -20,7 +20,7 @@
     html_logo=os.path.join("_images", "gamma_sklearndf_logo.png"),
     intersphinx_mapping={
         "lightgbm": ("https://lightgbm.readthedocs.io/en/latest/", None),
-        "pytools": ("https://bcg-gamma.github.io/pytools/", None),
+        "pytools": ("https://bcg-x-official.github.io/pytools/", None),
         "sklearn": ("https://scikit-learn.org/stable", None),
         "xgboost": ("https://xgboost.readthedocs.io/en/latest/", None),
     },

diff --git a/sphinx/source/contribution_guide.rst b/sphinx/source/contribution_guide.rst
@@ -296,7 +296,7 @@ The ``sphinx/source`` folder contains:
 
 
 The two key scripts are ``make.py`` and ``conf.py``. The base configuration for the
-these scripts can be found in `pytools/sphinx <https://github.com/BCG-Gamma/pytools/tree/develop/sphinx/base>`_.
+these scripts can be found in `pytools/sphinx <https://github.com/BCG-X-Official/pytools/tree/develop/sphinx/base>`_.
 The reason for this is to minimise code given the standardization of the documentation
 build across multiple packages.
 
@@ -358,7 +358,7 @@ The build process for the PyPI and conda distributions uses the following key
 files:
 
 - ``make.py``: generic Python script for package builds. Most configuration is imported
-  from pytools `make.py <https://github.com/BCG-Gamma/pytools/blob/develop/make.py>`__
+  from pytools `make.py <https://github.com/BCG-X-Official/pytools/blob/develop/make.py>`__
   which is a build script that wraps the package build, as well as exposing the matrix
   dependency definitions specified in the ``pyproject.toml`` as environment variables
 - ``pyproject.toml``: metadata for PyPI, build settings and package dependencies

diff --git a/sphinx/source/faqs.rst b/sphinx/source/faqs.rst
@@ -15,7 +15,7 @@ on `stackoverflow <https://stackoverflow.com/>`_.
 1. **What if I find a bug or have an idea for a new feature?**
 
     For bug reports or feature requests please use our
-    `GitHub issue tracker <https://github.com/BCG-Gamma/sklearndf/issues>`_.
+    `GitHub issue tracker <https://github.com/BCG-X-Official/sklearndf/issues>`_.
     For any other enquiries please feel free to contact us at [email protected].
 
 2. **How can I contribute?**

diff --git a/src/sklearndf/__init__.py b/src/sklearndf/__init__.py
@@ -8,4 +8,4 @@
 from ._sklearn_version import *
 from ._sklearndf import *
 
-__version__ = "2.2.1"
+__version__ = "2.3rc0"
diff --git a/src/sklearndf/_sklearn_version.py b/src/sklearndf/_sklearn_version.py
@@ -7,18 +7,14 @@
 
 __all__ = [
     "__sklearn_version__",
-    "__sklearn_0_24__",
-    "__sklearn_1_0__",
     "__sklearn_1_1__",
     "__sklearn_1_2__",
     "__sklearn_1_3__",
     "__sklearn_1_4__",
 ]
 
 __sklearn_version__ = Version(sklearn_version)
-__sklearn_0_24__ = Version("0.24")
-__sklearn_1_0__ = Version("1.0")
 __sklearn_1_1__ = Version("1.1")
-__sklearn_1_2__ = Version("1.2dev")
+__sklearn_1_2__ = Version("1.2")
 __sklearn_1_3__ = Version("1.3")
 __sklearn_1_4__ = Version("1.4")
diff --git a/src/sklearndf/classification/__init__.py b/src/sklearndf/classification/__init__.py
@@ -2,10 +2,7 @@
 Extended versions of all `scikit-learn` classifiers with enhanced support for data
 frames.
 """
-from .. import __sklearn_1_0__, __sklearn_version__
 from ._classification import *
 from ._classification_v0_22 import *
 from ._classification_v0_23 import *
-
-if __sklearn_version__ >= __sklearn_1_0__:
-    from ._classification_v1_0 import *
+from ._classification_v1_0 import *
diff --git a/src/sklearndf/classification/wrapper/_wrapper.py b/src/sklearndf/classification/wrapper/_wrapper.py
@@ -153,7 +153,6 @@ def _prediction_with_class_labels(
         ],
         classes: Optional[Sequence[Any]] = None,
     ) -> Union[pd.Series, pd.DataFrame, List[pd.DataFrame]]:
-
         # if we have a multi-output classifier, prediction of probabilities
         # yields a list of NumPy arrays
         if not isinstance(prediction, list):

diff --git a/src/sklearndf/clustering/__init__.py b/src/sklearndf/clustering/__init__.py
@@ -3,8 +3,11 @@
 frames.
 """
 
-from .. import __sklearn_1_1__, __sklearn_version__
+from .. import __sklearn_1_1__, __sklearn_1_3__, __sklearn_version__
 from ._clustering import *
 
 if __sklearn_version__ >= __sklearn_1_1__:
     from ._clustering_v1_1 import *
+
+if __sklearn_version__ >= __sklearn_1_3__:
+    from ._clustering_v1_3 import *