[SHOW PR] Submodulev2

.circleci/config.yml

@@ -89,22 +89,23 @@ jobs:
           root: doc/_build/html
           paths: .
 
-  deploy:
-    docker:
-      - image: cimg/python:3.8.12
-    steps:
-      - checkout
-      - run: ./build_tools/circle/checkout_merge_commit.sh
-      # Attach documentation generated in the 'doc' step so that it can be
-      # deployed.
-      - attach_workspace:
-          at: doc/_build/html
-      - run: ls -ltrh doc/_build/html/stable
-      - deploy:
-          command: |
-            if [[ "${CIRCLE_BRANCH}" =~ ^main$|^[0-9]+\.[0-9]+\.X$ ]]; then
-              bash build_tools/circle/push_doc.sh doc/_build/html/stable
-            fi
+  # XXX: in order to make sure our fork passes all the CIs and not remove too many LOC, we don't want to deploy
+  # deploy:
+  #   docker:
+  #     - image: cimg/python:3.8.12
+  #   steps:
+  #     - checkout
+  #     - run: ./build_tools/circle/checkout_merge_commit.sh
+  #     # Attach documentation generated in the 'doc' step so that it can be
+  #     # deployed.
+  #     - attach_workspace:
+  #         at: doc/_build/html
+  #     - run: ls -ltrh doc/_build/html/stable
+  #     - deploy:
+  #         command: |
+  #           if [[ "${CIRCLE_BRANCH}" =~ ^main$|^[0-9]+\.[0-9]+\.X$ ]]; then
+  #             bash build_tools/circle/push_doc.sh doc/_build/html/stable
+  #           fi
 
 workflows:
   version: 2

.cirrus.star

@@ -4,17 +4,17 @@
 load("cirrus", "env", "fs", "http")
 
 def main(ctx):
-    # Only run for scikit-learn/scikit-learn. For debugging on a fork, you can
+    # Only run for neurodata/scikit-learn. For debugging on a fork, you can
     # comment out the following condition.
-    if env.get("CIRRUS_REPO_FULL_NAME") != "scikit-learn/scikit-learn":
+    if env.get("CIRRUS_REPO_FULL_NAME") != "neurodata/scikit-learn":
         return []
 
     arm_wheel_yaml = "build_tools/cirrus/arm_wheel.yml"
     arm_tests_yaml = "build_tools/cirrus/arm_tests.yml"
 
     # Nightly jobs always run
     if env.get("CIRRUS_CRON", "") == "nightly":
-        return fs.read(arm_wheel_yaml)
+        return fs.read(arm_wheel_yaml) + fs.read(arm_tests_yaml)
 
     # Get commit message for event. We can not use `git` here because there is
     # no command line access in starlark. Thus we need to query the GitHub API
@@ -26,10 +26,12 @@ def main(ctx):
     response = http.get(url).json()
     commit_msg = response["message"]
 
-    if "[skip ci]" in commit_msg:
-        return []
+    jobs_to_run = ""
 
     if "[cd build]" in commit_msg or "[cd build cirrus]" in commit_msg:
-        return fs.read(arm_wheel_yaml) + fs.read(arm_tests_yaml)
+        jobs_to_run += fs.read(arm_wheel_yaml)
+
+    if "[cirrus arm]" in commit_msg:
+        jobs_to_run += fs.read(arm_tests_yaml)
 
-    return fs.read(arm_tests_yaml)
+    return jobs_to_run

.github/workflows/check-changelog.yml

@@ -10,12 +10,13 @@ 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 }}
+    if: ${{ contains(github.event.pull_request.labels.*.name, 'No Changelog Needed') == 0 && github.repository == 'scikit-learn/scikit-learn' }}
     steps:
       - name: Get PR number and milestone
         run: |
           echo "PR_NUMBER=${{ github.event.pull_request.number }}" >> $GITHUB_ENV
           echo "TAGGED_MILESTONE=${{ github.event.pull_request.milestone.title }}" >> $GITHUB_ENV
+          echo "${{ github.repository }}"
       - uses: actions/checkout@v3
         with:
           fetch-depth: '0'

.github/workflows/check-manifest.yml

@@ -7,7 +7,7 @@ on:
 jobs:
   check-manifest:
     # Don't run on forks
-    if: github.repository == 'scikit-learn/scikit-learn'
+    if: github.repository == 'neurodata/scikit-learn'
 
     runs-on: ubuntu-latest
     steps:

.github/workflows/check-upstream.yml

@@ -0,0 +1,27 @@
+# Create Github Actions workflow that checks upstream scikit-learn 'main' branch and
+# creates or updates
+# an existing pull request to https://github.com/neurodata/scikit-learn:fork.
+# Runs the check weekly.
+# Creates a pull request if there are changes.
+
+# name: Check upstream scikit-learn
+
+# on:
+#   schedule:
+#     - cron: '0 0 * * 0'
+
+# jobs:
+#   check-upstream:
+#     runs-on: ubuntu-latest
+#     steps:
+#       - uses: actions/checkout@v2
+#       - name: Check upstream scikit-learn
+#         uses: neurodata/check-upstream@main
+#         with:
+#           upstream: scikit-learn/scikit-learn
+#           fork: neurodata/scikit-learn
+#           branch: fork
+#           token: ${{ secrets.GITHUB_TOKEN }}
+
+# # Creates a pull request if there are changes.
+

.github/workflows/labeler-module.yml

@@ -16,7 +16,7 @@ jobs:
     steps:
     - uses: thomasjpfan/labeler@v2.5.1
       continue-on-error: true
-      if: github.repository == 'scikit-learn/scikit-learn'
+      if: github.repository == 'neurodata/scikit-learn'
       with:
         repo-token: "${{ secrets.GITHUB_TOKEN }}"
         max-labels: "3"
@@ -27,7 +27,7 @@ jobs:
     steps:
     - uses: thomasjpfan/labeler@v2.5.1
       continue-on-error: true
-      if: github.repository == 'scikit-learn/scikit-learn'
+      if: github.repository == 'neurodata/scikit-learn'
       with:
         repo-token: "${{ secrets.GITHUB_TOKEN }}"
         configuration-path: ".github/labeler-file-extensions.yml"

.github/workflows/update_tracking_issue.yml

@@ -24,7 +24,7 @@ on:
 jobs:
   update_tracking_issue:
     runs-on: ubuntu-latest
-    if: github.repository == 'scikit-learn/scikit-learn' && github.event_name == 'schedule'
+    if: github.repository == 'neurodata/scikit-learn' && github.event_name == 'schedule'
     steps:
       - uses: actions/checkout@v3
       - uses: actions/setup-python@v4

.github/workflows/wheels.yml

@@ -7,12 +7,12 @@ on:
     - cron: "42 3 */1 * *"
   push:
     branches:
-      - main
+      - fork
       # Release branches
       - "[0-9]+.[0-9]+.X"
   pull_request:
     branches:
-      - main
+      - fork
       - "[0-9]+.[0-9]+.X"
   # Manual run
   workflow_dispatch:
@@ -26,7 +26,7 @@ jobs:
   check_build_trigger:
     name: Check build trigger
     runs-on: ubuntu-latest
-    if: github.repository == 'scikit-learn/scikit-learn'
+    if: github.repository == 'neurodata/scikit-learn'
     outputs:
       build: ${{ steps.check_build_trigger.outputs.build }}
 
@@ -103,6 +103,18 @@ jobs:
             python: 311
             platform_id: macosx_x86_64
 
+          # MacOS arm64
+          # The latest Python version is built and tested on CirrusCI
+          - os: macos-latest
+            python: 38
+            platform_id: macosx_arm64
+          - os: macos-latest
+            python: 39
+            platform_id: macosx_arm64
+          - os: macos-latest
+            python: 310
+            platform_id: macosx_arm64
+
     steps:
       - name: Checkout scikit-learn
         uses: actions/checkout@v3
@@ -178,31 +190,8 @@ jobs:
         with:
           path: dist/*.tar.gz
 
-  # Upload the wheels and the source distribution
-  upload_anaconda:
-    name: Upload to Anaconda
-    runs-on: ubuntu-latest
-    needs: [build_wheels, build_sdist]
-    # The artifacts cannot be uploaded on PRs
-    if: github.event_name != 'pull_request'
-
-    steps:
-      - name: Checkout scikit-learn
-        uses: actions/checkout@v3
-
-      - name: Download artifacts
-        uses: actions/download-artifact@v3
+      - uses: actions/upload-artifact@v3
         with:
           path: dist
+          name: ${{ matrix.python[0] }}-${{ matrix.os[1] }}
 
-      - name: Setup Python
-        uses: actions/setup-python@v4
-
-      - name: Upload artifacts
-        env:
-          # Secret variables need to be mapped to environment variables explicitly
-          SCIKIT_LEARN_NIGHTLY_UPLOAD_TOKEN: ${{ secrets.SCIKIT_LEARN_NIGHTLY_UPLOAD_TOKEN }}
-          SCIKIT_LEARN_STAGING_UPLOAD_TOKEN: ${{ secrets.SCIKIT_LEARN_STAGING_UPLOAD_TOKEN }}
-          ARTIFACTS_PATH: dist/artifact
-        # Force a replacement if the remote file already exists
-        run: bash build_tools/github/upload_anaconda.sh

.gitignore

@@ -10,6 +10,7 @@
 build
 sklearn/datasets/__config__.py
 sklearn/**/*.html
+scikit_learn_tree.egg-info/*
 
 dist/
 MANIFEST
@@ -99,6 +100,9 @@ sklearn/metrics/_pairwise_distances_reduction/_middle_term_computer.pxd
 sklearn/metrics/_pairwise_distances_reduction/_middle_term_computer.pyx
 sklearn/metrics/_pairwise_distances_reduction/_radius_neighbors.pxd
 sklearn/metrics/_pairwise_distances_reduction/_radius_neighbors.pyx
+sklearn/neighbors/_ball_tree.pyx
+sklearn/neighbors/_binary_tree.pxi
+sklearn/neighbors/_kd_tree.pyx
 
 # Default JupyterLite content
 jupyterlite_contents

Makefile

@@ -62,3 +62,6 @@ doc-noplot: inplace
 
 code-analysis:
 	build_tools/linting.sh
+
+build-dev:
+	pip install --verbose --no-build-isolation --editable .