Merge pull request #49 from scikit-learn-contrib/0.3.X

DOC add random test times to mitigate inconsistent results on different OS

Author: MatthewSZhang

.github/workflows/ci.yml

@@ -8,64 +8,17 @@ on:
     branches: ["main"]
 
 jobs:
-  test:
-    runs-on: ubuntu-latest
-
-    steps:
-      - uses: actions/checkout@v4
-      - uses: prefix-dev/[email protected]
-        with:
-          environments: default
-          cache: true
-
-      - name: Re-install local
-        run: |
-          pixi run rebuild
-
-      - name: Lint with ruff
-        run: |
-          pixi run lint
-      - name: Lint with cython-lint
-        run: |
-          pixi run cython-lint
-      - name: Format with black
-        run: |
-          pixi run fmt
-      - name: Type check with mypy
-        run: |
-          pixi run type
-      - name: Test with pytest
-        run: |
-          pixi run test
-      - name: Test with doctest
-        shell: bash
-        run: |
-          pixi run doc
-          CMD=doctest pixi run doc
-      - name: Test coverage
-        shell: bash
-        run: |
-          FMT=xml pixi run test-coverage
-      - name: Upload coverage reports to Codecov
-        uses: codecov/codecov-action@v5
-        with:
-          token: ${{ secrets.CODECOV_TOKEN }}
-      - name: Build SDist
-        run: |
-          pixi run build-sdist
-      - name: Store artifacts
-        uses: actions/upload-artifact@v4
-        with:
-          name: cibw-sdist
-          path: dist/*.tar.gz
+  call-test:
+    uses: ./.github/workflows/test.yml
+    secrets: inherit
 
   build:
     strategy:
       fail-fast: false
       matrix:
         os: [ubuntu-latest, windows-latest, macos-latest]
     runs-on: ${{ matrix.os }}
-    needs: test
+    needs: call-test
     steps:
       - uses: actions/checkout@v4
       - name: Build wheels

.github/workflows/test.yml

@@ -0,0 +1,60 @@
+name: Test
+
+on:
+  workflow_call:
+  push:
+    branches: ["*.X"]
+  pull_request:
+    branches: ["*.X"]
+
+jobs:
+  test:
+    runs-on: ubuntu-latest
+
+    steps:
+      - uses: actions/checkout@v4
+      - uses: prefix-dev/[email protected]
+        with:
+          environments: default
+          cache: true
+
+      - name: Re-install local
+        run: |
+          pixi run rebuild
+
+      - name: Lint with ruff
+        run: |
+          pixi run lint
+      - name: Lint with cython-lint
+        run: |
+          pixi run cython-lint
+      - name: Format with black
+        run: |
+          pixi run fmt
+      - name: Type check with mypy
+        run: |
+          pixi run type
+      - name: Test with pytest
+        run: |
+          pixi run test
+      - name: Test with doctest
+        shell: bash
+        run: |
+          pixi run doc
+          CMD=doctest pixi run doc
+      - name: Test coverage
+        shell: bash
+        run: |
+          FMT=xml pixi run test-coverage
+      - name: Upload coverage reports to Codecov
+        uses: codecov/codecov-action@v5
+        with:
+          token: ${{ secrets.CODECOV_TOKEN }}
+      - name: Build SDist
+        run: |
+          pixi run build-sdist
+      - name: Store artifacts
+        uses: actions/upload-artifact@v4
+        with:
+          name: cibw-sdist
+          path: dist/*.tar.gz

examples/plot_pruning.py

@@ -26,7 +26,7 @@
 from sklearn.linear_model import LogisticRegression
 
 data, labels = load_iris(return_X_y=True)
-baseline_lr = LogisticRegression(max_iter=110).fit(data, labels)
+baseline_lr = LogisticRegression(max_iter=1000).fit(data, labels)
 
 # %%
 # Random data pruning
@@ -40,7 +40,7 @@
 def _random_pruning(X, y, n_samples_to_select: int, random_state: int):
     rng = np.random.default_rng(random_state)
     ids_random = rng.choice(y.size, n_samples_to_select, replace=False)
-    pruned_lr = LogisticRegression(max_iter=110).fit(X[ids_random], y[ids_random])
+    pruned_lr = LogisticRegression(max_iter=1000).fit(X[ids_random], y[ids_random])
     return pruned_lr.coef_, pruned_lr.intercept_
 
 
@@ -72,9 +72,9 @@ def _fastcan_pruning(
     ).fit(X)
     atoms = kmeans.cluster_centers_
     ids_fastcan = minibatch(
-        X.T, atoms.T, n_samples_to_select, batch_size=batch_size, tol=1e-6, verbose=0
+        X.T, atoms.T, n_samples_to_select, batch_size=batch_size, verbose=0
     )
-    pruned_lr = LogisticRegression(max_iter=110).fit(X[ids_fastcan], y[ids_fastcan])
+    pruned_lr = LogisticRegression(max_iter=1000).fit(X[ids_fastcan], y[ids_fastcan])
     return pruned_lr.coef_, pruned_lr.intercept_
 
 
@@ -111,4 +111,4 @@ def plot_box(X, y, baseline, n_samples_to_select: int, n_random: int):
     plt.show()
 
 
-plot_box(data, labels, baseline_lr, n_samples_to_select=100, n_random=10)
+plot_box(data, labels, baseline_lr, n_samples_to_select=100, n_random=100)