[Track v1.4] New PR branch to serve as submodule for scikit-tree

.github/workflows/check-changelog.yml

@@ -4,65 +4,30 @@ name: Check Changelog
 # To bypass this check, label the PR with "No Changelog Needed".
 on:
   pull_request:
-    types: [opened, edited, labeled, unlabeled, synchronize]
+    types: [opened, synchronize, labeled, unlabeled]
 
 jobs:
   check:
     name: A reviewer will let you know if it is required or can be bypassed
     runs-on: ubuntu-latest
-    if: ${{ contains(github.event.pull_request.labels.*.name, 'No Changelog Needed') == 0 }}
     steps:
-      - name: Get PR number and milestone
-        run: |
-          echo "PR_NUMBER=${{ github.event.pull_request.number }}" >> $GITHUB_ENV
-          echo "TAGGED_MILESTONE=${{ github.event.pull_request.milestone.title }}" >> $GITHUB_ENV
       - uses: actions/checkout@v4
         with:
           fetch-depth: '0'
-      - name: Check the changelog entry
+      - name: Check if tests have changed
+        id: tests_changed
         run: |
           set -xe
           changed_files=$(git diff --name-only origin/main)
           # Changelog should be updated only if tests have been modified
-          if [[ ! "$changed_files" =~ tests ]]
-          then
-            exit 0
-          fi
-          all_changelogs=$(cat ./doc/whats_new/v*.rst)
-          if [[ "$all_changelogs" =~ :pr:\`$PR_NUMBER\` ]]
+          if [[ "$changed_files" =~ tests ]]
           then
-            echo "Changelog has been updated."
-            # If the pull request is milestoned check the correspondent changelog
-            if exist -f ./doc/whats_new/v${TAGGED_MILESTONE:0:4}.rst
-            then
-              expected_changelog=$(cat ./doc/whats_new/v${TAGGED_MILESTONE:0:4}.rst)
-              if [[ "$expected_changelog" =~ :pr:\`$PR_NUMBER\` ]]
-              then
-                echo "Changelog and milestone correspond."
-              else
-                echo "Changelog and milestone do not correspond."
-                echo "If you see this error make sure that the tagged milestone for the PR"
-                echo "and the edited changelog filename properly match."
-                exit 1
-              fi
-            fi
-          else
-            echo "A Changelog entry is missing."
-            echo ""
-            echo "Please add an entry to the changelog at 'doc/whats_new/v*.rst'"
-            echo "to document your change assuming that the PR will be merged"
-            echo "in time for the next release of scikit-learn."
-            echo ""
-            echo "Look at other entries in that file for inspiration and please"
-            echo "reference this pull request using the ':pr:' directive and"
-            echo "credit yourself (and other contributors if applicable) with"
-            echo "the ':user:' directive."
-            echo ""
-            echo "If you see this error and there is already a changelog entry,"
-            echo "check that the PR number is correct."
-            echo ""
-            echo "If you believe that this PR does not warrant a changelog"
-            echo "entry, say so in a comment so that a maintainer will label"
-            echo "the PR with 'No Changelog Needed' to bypass this check."
-            exit 1
+            echo "check_changelog=true" >> $GITHUB_OUTPUT
           fi
+
+      - name: Check changelog entry
+        if: steps.tests_changed.outputs.check_changelog == 'true'
+        uses: scientific-python/action-towncrier-changelog@v1
+        env:
+          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+          BOT_USERNAME: changelog-bot

.github/workflows/cuda-ci.yml

@@ -16,7 +16,7 @@ jobs:
       - uses: actions/checkout@v4
 
       - name: Build wheels
-        uses: pypa/cibuildwheel@v2.20.0
+        uses: pypa/cibuildwheel@v2.21.3
         env:
           CIBW_BUILD: cp312-manylinux_x86_64
           CIBW_MANYLINUX_X86_64_IMAGE: manylinux2014

.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@8a08d616893759ef8e1aa1f2785787c0b97e20d6 # v1.10.0
+      uses: pypa/gh-action-pypi-publish@fb13cb306901256ace3dab689990e13a5550ffaa # v1.11.0
       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@8a08d616893759ef8e1aa1f2785787c0b97e20d6 # v1.10.0
+      uses: pypa/gh-action-pypi-publish@fb13cb306901256ace3dab689990e13a5550ffaa # v1.11.0
       if: ${{ github.event.inputs.pypi_repo == 'pypi' }}
       with:
         print-hash: true

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

@@ -6,6 +6,7 @@ on:
 
 permissions:
   contents: write
+  statuses: write
 
 jobs:
   update-lock-files:
@@ -23,12 +24,30 @@ jobs:
         env:
           GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
         run: |
-          pr_info=$(gh pr view ${{ github.event.issue.number }} --repo ${{ github.repository }} --json headRefName,headRepository,headRepositoryOwner)
+          pr_info=$(gh pr view ${{ github.event.issue.number }} --repo ${{ github.repository }} --json headRefName,headRefOid,headRepository,headRepositoryOwner)
           pr_head_ref=$(echo "$pr_info" | jq -r '.headRefName')
+          pr_head_sha=$(echo "$pr_info" | jq -r '.headRefOid')
           pr_head_repository=$(echo "$pr_info" | jq -r '.headRepositoryOwner.login + "/" + .headRepository.name')
           echo "pr_head_ref=$pr_head_ref" >> $GITHUB_OUTPUT
+          echo "pr_head_sha=$pr_head_sha" >> $GITHUB_OUTPUT
           echo "pr_head_repository=$pr_head_repository" >> $GITHUB_OUTPUT
 
+      # Set the status of the latest commit in the PR to indicate that the update is in progress
+      # https://docs.github.com/en/rest/commits/statuses?apiVersion=2022-11-28#create-a-commit-status
+      - name: Set pending status
+        env:
+          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+        run: |
+          gh api \
+            --method POST \
+            -H "Accept: application/vnd.github+json" \
+            -H "X-GitHub-Api-Version: 2022-11-28" \
+            /repos/${{ github.repository }}/statuses/${{ steps.pr-head-info.outputs.pr_head_sha }} \
+            -f "state=pending" \
+            -f "target_url=https://github.com/${{ github.repository }}/actions/runs/${{ github.run_id }}" \
+            -f "description=Updating lock files..." \
+            -f "context=update-lock-files-pr"
+
       - name: Check out the PR branch
         uses: actions/checkout@v4
         with:
@@ -57,3 +76,18 @@ jobs:
           conda install -c conda-forge "$(get_dep conda-lock min)" -y
 
           python build_tools/on_pr_comment_update_environments_and_lock_files.py
+
+      - name: Set completion status
+        if: ${{ always() }}
+        env:
+          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+        run: |
+          gh api \
+            --method POST \
+            -H "Accept: application/vnd.github+json" \
+            -H "X-GitHub-Api-Version: 2022-11-28" \
+            /repos/${{ github.repository }}/statuses/${{ steps.pr-head-info.outputs.pr_head_sha }} \
+            -f "state=${{ job.status == 'success' && 'success' || 'error' }}" \
+            -f "target_url=https://github.com/${{ github.repository }}/actions/runs/${{ github.run_id }}" \
+            -f "description=Lock files ${{ job.status == 'success' && 'updated' || 'failed to update' }}." \
+            -f "context=update-lock-files-pr"

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

@@ -6,10 +6,6 @@ on:
   schedule:
     - cron: '0 5 * * 1'
 
-# In order to add the "CUDA CI" label we need to have write permissions for PRs
-permissions:
-  pull-requests: write
-
 jobs:
   update_lock_files:
     if: github.repository == 'scikit-learn/scikit-learn'
@@ -46,7 +42,7 @@ jobs:
 
       - name: Create Pull Request
         id: cpr
-        uses: peter-evans/create-pull-request@v6
+        uses: peter-evans/create-pull-request@v7
         with:
           token: ${{ secrets.BOT_GITHUB_TOKEN }}
           push-to-fork: scikit-learn-bot/scikit-learn
@@ -66,9 +62,16 @@ jobs:
       - name: Trigger additional tests
         if: steps.cpr.outputs.pull-request-number != '' && matrix.name == 'array-api'
         env:
-          GH_TOKEN: ${{ github.token }}
+          GH_TOKEN: ${{ secrets.BOT_GITHUB_TOKEN }}
+          PR_NUMBER: ${{steps.cpr.outputs.pull-request-number}}
         run: |
-          gh pr edit ${{steps.cpr.outputs.pull-request-number}} --add-label "CUDA CI"
+          curl -L \
+            -X POST \
+            -H "Accept: application/vnd.github+json" \
+            -H "Authorization: Bearer $GH_TOKEN" \
+            -H "X-GitHub-Api-Version: 2022-11-28" \
+            https://api.github.com/repos/scikit-learn/scikit-learn/issues/$PR_NUMBER/labels \
+            -d '{"labels":["CUDA CI"]}'
 
       - name: Check Pull Request
         if: steps.cpr.outputs.pull-request-number != ''

.github/workflows/wheels.yml

@@ -44,6 +44,11 @@ jobs:
   build_wheels:
     name: Build wheel for cp${{ matrix.python }}-${{ matrix.platform_id }}-${{ matrix.manylinux_image }}
     runs-on: ${{ matrix.os }}
+
+    # For conda-incubator/setup-miniconda to work
+    defaults:
+      run:
+        shell: bash -el {0}
     needs: check_build_trigger
     if: needs.check_build_trigger.outputs.build
 
@@ -65,6 +70,9 @@ jobs:
           - os: windows-latest
             python: 312
             platform_id: win_amd64
+          - os: windows-latest
+            python: 313
+            platform_id: win_amd64
 
           # Linux 64 bit manylinux2014
           - os: ubuntu-latest
@@ -86,27 +94,36 @@ jobs:
             python: 312
             platform_id: manylinux_x86_64
             manylinux_image: manylinux2014
+          - os: ubuntu-latest
+            python: 313
+            platform_id: manylinux_x86_64
+            manylinux_image: manylinux2014
           - os: ubuntu-latest
             python: 313t
             platform_id: manylinux_x86_64
             manylinux_image: manylinux2014
-            # TODO: remove next line when Python 3.13 is released
-            prerelease_pythons: True
             free_threaded_support: True
 
           # MacOS x86_64
-          - os: macos-12
+          - os: macos-13
             python: 39
             platform_id: macosx_x86_64
-          - os: macos-12
+          - os: macos-13
             python: 310
             platform_id: macosx_x86_64
-          - os: macos-12
+          - os: macos-13
             python: 311
             platform_id: macosx_x86_64
-          - os: macos-12
+          - os: macos-13
             python: 312
             platform_id: macosx_x86_64
+          - os: macos-13
+            python: 313
+            platform_id: macosx_x86_64
+          - os: macos-13
+            python: 313t
+            platform_id: macosx_x86_64
+            free_threaded_support: True
 
           # MacOS arm64
           - os: macos-14
@@ -121,6 +138,13 @@ jobs:
           - os: macos-14
             python: 312
             platform_id: macosx_arm64
+          - os: macos-14
+            python: 313
+            platform_id: macosx_arm64
+          - os: macos-14
+            python: 313t
+            platform_id: macosx_arm64
+            free_threaded_support: True
 
     steps:
       - name: Checkout scikit-learn
@@ -131,33 +155,8 @@ jobs:
         with:
           python-version: "3.11" # update once build dependencies are available
 
-      - name: Install conda for macos arm64
-        if: ${{ matrix.platform_id == 'macosx_arm64' }}
-        run: |
-          set -ex
-          # macos arm64 runners do not have conda installed. Thus we much install conda manually
-          EXPECTED_SHA="dd832d8a65a861b5592b2cf1d55f26031f7c1491b30321754443931e7b1e6832"
-          MINIFORGE_URL="https://github.com/conda-forge/miniforge/releases/download/23.11.0-0/Mambaforge-23.11.0-0-MacOSX-arm64.sh"
-          curl -L --retry 10 $MINIFORGE_URL -o miniforge.sh
-
-          # Check SHA
-          file_sha=$(shasum -a 256 miniforge.sh | awk '{print $1}')
-          if [ "$EXPECTED_SHA" != "$file_sha" ]; then
-              echo "SHA values did not match!"
-              exit 1
-          fi
-
-          # Install miniforge
-          MINIFORGE_PATH=$HOME/miniforge
-          bash ./miniforge.sh -b -p $MINIFORGE_PATH
-          echo "$MINIFORGE_PATH/bin" >> $GITHUB_PATH
-          echo "CONDA_HOME=$MINIFORGE_PATH" >> $GITHUB_ENV
-
-      - name: Set conda environment for non-macos arm64 environments
-        if: ${{ matrix.platform_id != 'macosx_arm64' }}
-        run: |
-          # Non-macos arm64 envrionments already have conda installed
-          echo "CONDA_HOME=/usr/local/miniconda" >> $GITHUB_ENV
+      - uses: conda-incubator/setup-miniconda@v3
+        if: ${{ startsWith(matrix.platform_id, 'macosx') }}
 
       - name: Build and test wheels
         env:
@@ -173,11 +172,17 @@ jobs:
           # toolchain
           CIBW_CONFIG_SETTINGS_WINDOWS: "setup-args=--vsenv"
           CIBW_REPAIR_WHEEL_COMMAND_WINDOWS: bash build_tools/github/repair_windows_wheels.sh {wheel} {dest_dir}
+          CIBW_BEFORE_BUILD: bash {project}/build_tools/wheels/cibw_before_build.sh {project}
           CIBW_BEFORE_TEST_WINDOWS: bash build_tools/github/build_minimal_windows_image.sh ${{ matrix.python }}
           CIBW_BEFORE_TEST: bash {project}/build_tools/wheels/cibw_before_test.sh
+          CIBW_ENVIRONMENT_PASS_LINUX: RUNNER_OS
           CIBW_TEST_REQUIRES: pytest pandas
-          CIBW_TEST_COMMAND: bash {project}/build_tools/wheels/test_wheels.sh
-          CIBW_TEST_COMMAND_WINDOWS: bash {project}/build_tools/github/test_windows_wheels.sh ${{ matrix.python }}
+          # On Windows, we use a custom Docker image and CIBW_TEST_REQUIRES_WINDOWS
+          # does not make sense because it would install dependencies in the host
+          # rather than inside the Docker image
+          CIBW_TEST_REQUIRES_WINDOWS: ""
+          CIBW_TEST_COMMAND: bash {project}/build_tools/wheels/test_wheels.sh {project}
+          CIBW_TEST_COMMAND_WINDOWS: bash {project}/build_tools/github/test_windows_wheels.sh ${{ matrix.python }} {project}
           CIBW_BUILD_VERBOSITY: 1
 
         run: bash build_tools/wheels/build_wheels.sh

.gitignore

@@ -25,6 +25,9 @@ doc/developers/maintainer.rst
 doc/index.rst
 doc/min_dependency_table.rst
 doc/min_dependency_substitutions.rst
+# release notes generated by towncrier
+doc/whats_new/notes-towncrier.rst
+
 *.pdf
 pip-log.txt
 scikit_learn.egg-info/
@@ -90,3 +93,6 @@ sklearn-env/
 
 # Default JupyterLite content
 jupyterlite_contents
+
+# file recognised by vscode IDEs containing env variables
+.env

SECURITY.md

@@ -4,8 +4,8 @@
 
 | Version       | Supported          |
 | ------------- | ------------------ |
-| 1.5.1         | :white_check_mark: |
-| < 1.5.1       | :x:                |
+| 1.5.2         | :white_check_mark: |
+| < 1.5.2       | :x:                |
 
 ## Reporting a Vulnerability
 

azure-pipelines.yml

@@ -253,7 +253,7 @@ jobs:
 - template: build_tools/azure/posix.yml
   parameters:
     name: macOS
-    vmImage: macOS-12
+    vmImage: macOS-13
     dependsOn: [linting, git_commit, Ubuntu_Jammy_Jellyfish]
     # Runs when dependencies succeeded or skipped
     condition: |

benchmarks/bench_20newsgroups.py

@@ -21,7 +21,7 @@
     "extra_trees": ExtraTreesClassifier(max_features="sqrt", min_samples_split=10),
     "logistic_regression": LogisticRegression(),
     "naive_bayes": MultinomialNB(),
-    "adaboost": AdaBoostClassifier(n_estimators=10, algorithm="SAMME"),
+    "adaboost": AdaBoostClassifier(n_estimators=10),
 }