Merging main

Signed-off-by: Adam Li <adam2392@gmail.com>

Author: adam2392

.github/workflows/cuda-ci.yml

@@ -8,15 +8,41 @@ on:
       - labeled
 
 jobs:
+  build_wheel:
+    if: contains(github.event.pull_request.labels.*.name, 'CUDA CI')
+    runs-on: "ubuntu-latest"
+    name: Build wheel for Pull Request
+    steps:
+      - uses: actions/checkout@v4
+
+      - name: Build wheels
+        uses: pypa/cibuildwheel@v2.20.0
+        env:
+          CIBW_BUILD: cp312-manylinux_x86_64
+          CIBW_MANYLINUX_X86_64_IMAGE: manylinux2014
+          CIBW_BUILD_VERBOSITY: 1
+          CIBW_ARCHS: x86_64
+
+      - uses: actions/upload-artifact@v4
+        with:
+          name: cibw-wheels
+          path: ./wheelhouse/*.whl
+
   tests:
     if: contains(github.event.pull_request.labels.*.name, 'CUDA CI')
+    needs: [build_wheel]
     runs-on:
       group: cuda-gpu-runner-group
     # Set this high enough so that the tests can comforatble run. We set a
     # timeout to make abusing this workflow less attractive.
     timeout-minutes: 20
     name: Run Array API unit tests
     steps:
+      - uses: actions/download-artifact@v4
+        with:
+          pattern: cibw-wheels
+          path: ~/dist
+
       - uses: actions/setup-python@v5
         with:
           # XXX: The 3.12.4 release of Python on GitHub Actions is corrupted:
@@ -37,9 +63,14 @@ jobs:
         run: |
           source "${HOME}/conda/etc/profile.d/conda.sh"
           conda activate sklearn
-          pip install --verbose --no-build-isolation --config-settings editable-verbose=true --editable .
+          pip install ~/dist/cibw-wheels/$(ls ~/dist/cibw-wheels)
+
       - name: Run array API tests
         run: |
           source "${HOME}/conda/etc/profile.d/conda.sh"
           conda activate sklearn
-          SCIPY_ARRAY_API=1 pytest -k 'array_api'
+          python -c "import sklearn; sklearn.show_versions()"
+
+          SCIPY_ARRAY_API=1 pytest --pyargs sklearn -k 'array_api'
+        # Run in /home/runner to not load sklearn from the checkout repo
+        working-directory: /home/runner

.github/workflows/publish_pypi.yml

@@ -39,13 +39,13 @@ jobs:
       run: |
         python build_tools/github/check_wheels.py
     - name: Publish package to TestPyPI
-      uses: pypa/gh-action-pypi-publish@ec4db0b4ddc65acdf4bff5fa45ac92d78b56bdf0 # v1.9.0
+      uses: pypa/gh-action-pypi-publish@8a08d616893759ef8e1aa1f2785787c0b97e20d6 # v1.10.0
       with:
         repository-url: https://test.pypi.org/legacy/
         print-hash: true
       if: ${{ github.event.inputs.pypi_repo == 'testpypi' }}
     - name: Publish package to PyPI
-      uses: pypa/gh-action-pypi-publish@ec4db0b4ddc65acdf4bff5fa45ac92d78b56bdf0 # v1.9.0
+      uses: pypa/gh-action-pypi-publish@8a08d616893759ef8e1aa1f2785787c0b97e20d6 # v1.10.0
       if: ${{ github.event.inputs.pypi_repo == 'pypi' }}
       with:
         print-hash: true

.github/workflows/update-lock-files-pr.yml

@@ -10,7 +10,8 @@ permissions:
 jobs:
   update-lock-files:
     if: >-
-      github.event.issue.pull_request
+      github.repository == 'scikit-learn/scikit-learn'
+      && github.event.issue.pull_request
       && startsWith(github.event.comment.body, '@scikit-learn-bot update lock-files')
     runs-on: ubuntu-latest
 
@@ -39,9 +40,9 @@ jobs:
       # malicious changes to these scripts in PRs to be executed
       - name: Download scripts from main
         run: |
-          curl https://raw.githubusercontent.com/${{ github.repository }}/comment-update-lock/build_tools/shared.sh --retry 5 -o ./build_tools/shared.sh
-          curl https://raw.githubusercontent.com/${{ github.repository }}/comment-update-lock/build_tools/update_environments_and_lock_files.py --retry 5 -o ./build_tools/update_environments_and_lock_files.py
-          curl https://raw.githubusercontent.com/${{ github.repository }}/comment-update-lock/build_tools/on_pr_comment_update_environments_and_lock_files.py --retry 5 -o ./build_tools/on_pr_comment_update_environments_and_lock_files.py
+          curl https://raw.githubusercontent.com/${{ github.repository }}/main/build_tools/shared.sh --retry 5 -o ./build_tools/shared.sh
+          curl https://raw.githubusercontent.com/${{ github.repository }}/main/build_tools/update_environments_and_lock_files.py --retry 5 -o ./build_tools/update_environments_and_lock_files.py
+          curl https://raw.githubusercontent.com/${{ github.repository }}/main/build_tools/on_pr_comment_update_environments_and_lock_files.py --retry 5 -o ./build_tools/on_pr_comment_update_environments_and_lock_files.py
 
       - name: Update lock files
         env:

CITATION.cff

@@ -0,0 +1,48 @@
+cff-version: 1.2.0
+title: scikit-learn
+type: software
+authors:
+  - name: "The scikit-learn developers"
+message: "If you use scikit-learn in a scientific publication, we would appreciate citations to the following paper:"
+preferred-citation:
+  type: article
+  title: "Scikit-learn: Machine Learning in Python"
+  authors:
+  - family-names: "Pedregosa"
+    given-names: "Fabian"
+  - family-names: "Varoquaux"
+    given-names: "Gaël"
+  - family-names: "Gramfort"
+    given-names: "Alexandre"
+  - family-names: "Michel"
+    given-names: "Vincent"
+  - family-names: "Thirion"
+    given-names: "Bertrand"
+  - family-names: "Grisel"
+    given-names: "Olivier"
+  - family-names: "Blondel"
+    given-names: "Mathieu"
+  - family-names: "Prettenhofer"
+    given-names: "Peter"
+  - family-names: "Weiss"
+    given-names: "Ron"
+  - family-names: "Dubourg"
+    given-names: "Vincent"
+  - family-names: "Vanderplas"
+    given-names: "Jake"
+  - family-names: "Passos"
+    given-names: "Alexandre"
+  - family-names: "Cournapeau"
+    given-names: "David"
+  - family-names: "Brucher"
+    given-names: "Matthieu"
+  - family-names: "Perrot"
+    given-names: "Matthieu"
+  - family-names: "Duchesnay"
+    given-names: "Édouard"
+  journal: "Journal of Machine Learning Research"
+  volume: 12
+  start: 2825
+  end: 2830
+  year: 2011
+  url: "https://jmlr.csail.mit.edu/papers/v12/pedregosa11a.html"

azure-pipelines.yml

@@ -236,11 +236,11 @@ jobs:
         not(contains(dependencies['git_commit']['outputs']['commit.message'], '[ci skip]'))
       )
     matrix:
-      debian_atlas_32bit:
-        DOCKER_CONTAINER: 'i386/debian:12'
+      debian_32bit:
+        DOCKER_CONTAINER: 'i386/debian:trixie'
         DISTRIB: 'debian-32'
         COVERAGE: "true"
-        LOCK_FILE: './build_tools/azure/debian_atlas_32bit_lock.txt'
+        LOCK_FILE: './build_tools/azure/debian_32bit_lock.txt'
         # disable pytest xdist due to unknown bug with 32-bit container
         PYTEST_XDIST_VERSION: 'none'
         SKLEARN_TESTS_GLOBAL_RANDOM_SEED: '4'  # non-default seed

build_tools/azure/debian_32bit_lock.txt

@@ -0,0 +1,37 @@
+#
+# This file is autogenerated by pip-compile with Python 3.12
+# by the following command:
+#
+#    pip-compile --output-file=build_tools/azure/debian_32bit_lock.txt build_tools/azure/debian_32bit_requirements.txt
+#
+coverage[toml]==7.6.1
+    # via pytest-cov
+cython==3.0.11
+    # via -r build_tools/azure/debian_32bit_requirements.txt
+iniconfig==2.0.0
+    # via pytest
+joblib==1.4.2
+    # via -r build_tools/azure/debian_32bit_requirements.txt
+meson==1.5.1
+    # via meson-python
+meson-python==0.16.0
+    # via -r build_tools/azure/debian_32bit_requirements.txt
+ninja==1.11.1.1
+    # via -r build_tools/azure/debian_32bit_requirements.txt
+packaging==24.1
+    # via
+    #   meson-python
+    #   pyproject-metadata
+    #   pytest
+pluggy==1.5.0
+    # via pytest
+pyproject-metadata==0.8.0
+    # via meson-python
+pytest==8.3.2
+    # via
+    #   -r build_tools/azure/debian_32bit_requirements.txt
+    #   pytest-cov
+pytest-cov==5.0.0
+    # via -r build_tools/azure/debian_32bit_requirements.txt
+threadpoolctl==3.5.0
+    # via -r build_tools/azure/debian_32bit_requirements.txt

build_tools/azure/debian_atlas_32bit_requirements.txt renamed to build_tools/azure/debian_32bit_requirements.txt

@@ -1,10 +1,10 @@
 # DO NOT EDIT: this file is generated from the specification found in the
 # following script to centralize the configuration for CI builds:
 # build_tools/update_environments_and_lock_files.py
-cython==3.0.10  # min
-joblib==1.2.0  # min
-threadpoolctl==3.1.0
-pytest==7.1.2  # min
-pytest-cov==2.9.0  # min
+cython
+joblib
+threadpoolctl
+pytest
+pytest-cov
 ninja
 meson-python

build_tools/azure/debian_atlas_32bit_lock.txt

@@ -24,6 +24,9 @@ setup_ccache() {
         done
         export PATH="${CCACHE_LINKS_DIR}:${PATH}"
         ccache -M 256M
+
+        # Zeroing statistics so that ccache statistics are shown only for this build
+        ccache -z
     fi
 }
 
@@ -36,7 +39,7 @@ pre_python_environment_install() {
     elif [[ "$DISTRIB" == "debian-32" ]]; then
         apt-get update
         apt-get install -y python3-dev python3-numpy python3-scipy \
-                python3-matplotlib libatlas3-base libatlas-base-dev \
+                python3-matplotlib libopenblas-dev \
                 python3-virtualenv python3-pandas ccache git
 
     # TODO for now we use CPython 3.13 from Ubuntu deadsnakes PPA. When CPython