feat: numpy 2.0rc support (#651)

* numpy 2.0rc support * pep compliant specification * typo * extra requirement with matrix * add marker info
koaning · Apr 20, 2024 · 8cabda3 · 8cabda3
1 parent 35dd279
commit 8cabda3
Show file tree

Hide file tree

Showing 12 changed files with 58 additions and 25 deletions.
diff --git a/.github/workflows/schedule-dependencies.yml b/.github/workflows/schedule-dependencies.yml
@@ -7,7 +7,7 @@ on:
 
 
 jobs:
-  cron:
+  cron-base:
     runs-on: ${{ matrix.os }}
     strategy:
       matrix:
@@ -30,9 +30,36 @@ jobs:
     - name: Install dependencies
       run: |
         uv pip install wheel --system
-        uv pip install ${{ matrix.pre-release-dependencies }} scikit-lego --system
+        uv pip install ${{ matrix.pre-release-dependencies }} --system -e ".[test-dep]"
         uv pip freeze
     - name: Test with pytest
+      run: pytest -n auto --disable-warnings --cov=sklego -m "not cvxpy and not formulaic and not umap"
+
+  cron-extra:
+    runs-on: ${{ matrix.os }}
+    strategy:
+      matrix:
+        python-version: ["3.10"]
+        os: [macos-latest, ubuntu-latest, windows-latest]
+        pre-release-dependencies: ["--pre", ""]
+        extra: ["cvxpy", "formulaic", "umap"]
+    steps:
+    - name: Checkout source code
+      uses: actions/checkout@v4
+    - name: Install uv (Unix)
+      if: runner.os != 'Windows'
+      run: curl -LsSf https://astral.sh/uv/install.sh | sh
+    - name: Install uv (Windows)
+      if: runner.os == 'Windows'
+      run: powershell -c "irm https://astral.sh/uv/install.ps1 | iex"
+    - name: Set up Python ${{ matrix.python-version }}
+      uses: actions/setup-python@v5
+      with:
+        python-version: ${{ matrix.python-version }}
+    - name: Install dependencies
       run: |
-        uv pip install -e ".[test]" --system
-        make test
+        uv pip install wheel --system
+        uv pip install ${{ matrix.pre-release-dependencies }} -e ".[test-dep,${{ matrix.extra }}]" --system
+        uv pip freeze
+    - name: Test with pytest
+      run: pytest -n auto --disable-warnings --cov=sklego -m "${{ matrix.extra }}"
diff --git a/.github/workflows/test.yml b/.github/workflows/test.yml
@@ -33,6 +33,6 @@ jobs:
       with:
         python-version: ${{ matrix.python-version }}
     - name: Install dependencies
-      run: uv pip install -e ".[test]" --system
+      run: uv pip install -e ".[test-all]" --system
     - name: Test with pytest
       run: make test
diff --git a/pyproject.toml b/pyproject.toml
@@ -45,9 +45,9 @@ issue-tracker = "https://github.com/koaning/scikit-lego/issues"
 documentation = "https://koaning.github.io/scikit-lego/"
 
 [project.optional-dependencies]
-cvxpy = ["cmake", "osqp", "cvxpy>=1.1.8"]
+cvxpy = ["cmake", "osqp", "cvxpy>=1.1.8", "numpy<2.0"]
 formulaic = ["formulaic>=0.6.0"]
-umap = ["umap-learn>=0.4.6"]
+umap = ["umap-learn>=0.4.6", "numpy<2.0"]
 
 all = ["scikit-lego[cvxpy,formulaic,umap]"]
 
@@ -60,14 +60,17 @@ docs = [
     "mkdocstrings-python>=1.7.3",
 ]
 
-test = [
-    "scikit-lego[all]",
+test-dep = [
     "pytest>=6.2.5",
     "pytest-xdist>=1.34.0",
     "pytest-cov>=2.6.1",
     "pytest-mock>=1.6.3",
 ]
 
+test-all = [
+    "scikit-lego[all,test-dep]",
+]
+
 utils = [
     "matplotlib>=3.0.2",
     "jupyter>=1.0.0",
@@ -104,5 +107,8 @@ ignore = [
 
 [tool.pytest.ini_options]
 markers = [
-    "cvxpy: tests that require cvxpy (deselect with '-m \"not cvxpy\"')"
+    "cvxpy: tests that require cvxpy (deselect with '-m \"not cvxpy\"')",
+    "formulaic: tests that require formulaic (deselect with '-m \"not formulaic\"')",
+    "umap: tests that require umap (deselect with '-m \"not umap\"')"
 ]
+
diff --git a/sklego/meta/_shrinkage_utils.py b/sklego/meta/_shrinkage_utils.py
@@ -152,7 +152,7 @@ def _set_shrinkage_function(self):
 
             def no_shrinkage_function(x):
                 n = len(self.fitted_levels_[-1])
-                return np.lib.pad([1], (len(x) - 1, n - len(x)), "constant", constant_values=(0))
+                return np.pad([1], (len(x) - 1, n - len(x)), "constant", constant_values=(0))
 
             shrinkage_function_ = no_shrinkage_function
 

diff --git a/sklego/meta/hierarchical_predictor.py b/sklego/meta/hierarchical_predictor.py
@@ -324,7 +324,7 @@ def _predict_estimators(self, X, method_name):
                 raw_pred = getattr(_estimator, method_name)(grp_frame.drop(columns=self.groups_))
 
                 preds[np.ix_(grp_idx, [level_idx], last_dim_ix)] = np.atleast_3d(raw_pred[:, None])
-                shrinkage[np.ix_(grp_idx)] = np.lib.pad(
+                shrinkage[np.ix_(grp_idx)] = np.pad(
                     _shrinkage_factor, (0, self.n_levels_ - len(_shrinkage_factor)), "constant", constant_values=(0)
                 )
 

diff --git a/sklego/pandas_utils.py b/sklego/pandas_utils.py
@@ -298,7 +298,7 @@ def _add_lagged_numpy_columns(X, cols, lags, drop_na):
     if not all([col < X.shape[1] for col in cols]):
         raise KeyError("The column does not exist")
 
-    combos = (shift(X[:, col], -lag, cval=np.NaN) for col in cols for lag in lags)
+    combos = (shift(X[:, col], -lag, cval=np.nan) for col in cols for lag in lags)
 
     # In integer-based ndarrays, NaN values are represented as
     # -9223372036854775808, so we convert back and forth from

diff --git a/tests/test_estimators/test_demographic_parity.py b/tests/test_estimators/test_demographic_parity.py
@@ -12,6 +12,8 @@
 from sklego.metrics import p_percent_score
 from tests.conftest import classifier_checks, general_checks, nonmeta_checks, select_tests
 
+pytestmark = pytest.mark.cvxpy
+
 
 @pytest.mark.parametrize(
     "test_fn",
@@ -20,7 +22,6 @@
         exclude=["check_sample_weights_invariance", "check_sample_weights_list", "check_sample_weights_pandas_series"],
     ),
 )
-@pytest.mark.cvxpy
 def test_standard_checks(test_fn):
     trf = DemographicParityClassifier(
         covariance_threshold=None,
@@ -70,7 +71,6 @@ def _test_same(dataset):
         assert np.sum(lr.predict(X_without_sens) != fair.predict(X)) / len(X) < 0.01
 
 
-@pytest.mark.cvxpy
 def test_same_logistic(random_xy_dataset_clf):
     """
     Tests whether the fair classifier performs similar to logistic regression
@@ -80,7 +80,6 @@ def test_same_logistic(random_xy_dataset_clf):
     _test_same(random_xy_dataset_clf)
 
 
-@pytest.mark.cvxpy
 def test_same_logistic_multiclass(random_xy_dataset_multiclf):
     """
     Tests whether the fair classifier performs similar to logistic regression
@@ -90,7 +89,6 @@ def test_same_logistic_multiclass(random_xy_dataset_multiclf):
     _test_same(random_xy_dataset_multiclf)
 
 
-@pytest.mark.cvxpy
 def test_regularization(sensitive_classification_dataset):
     """Tests whether increasing regularization decreases the norm of the coefficient vector"""
     X, y = sensitive_classification_dataset
@@ -103,7 +101,6 @@ def test_regularization(sensitive_classification_dataset):
         prev_theta_norm = theta_norm
 
 
-@pytest.mark.cvxpy
 def test_fairness(sensitive_classification_dataset):
     """tests whether fairness (measured by p percent score) increases as we decrease the covariance threshold"""
     X, y = sensitive_classification_dataset

diff --git a/tests/test_estimators/test_equal_opportunity.py b/tests/test_estimators/test_equal_opportunity.py
@@ -7,6 +7,8 @@
 from sklego.metrics import equal_opportunity_score
 from tests.conftest import classifier_checks, general_checks, nonmeta_checks, select_tests
 
+pytestmark = pytest.mark.cvxpy
+
 
 @pytest.mark.parametrize(
     "test_fn",
@@ -15,7 +17,6 @@
         exclude=["check_sample_weights_invariance", "check_sample_weights_list", "check_sample_weights_pandas_series"],
     ),
 )
-@pytest.mark.cvxpy
 def test_standard_checks(test_fn):
     trf = EqualOpportunityClassifier(
         covariance_threshold=None,
@@ -69,7 +70,6 @@ def _test_same(dataset):
     assert np.sum(lr.predict(X_without_sens) != fair.predict(X)) / len(X) < 0.01
 
 
-@pytest.mark.cvxpy
 def test_same_logistic(random_xy_dataset_clf):
     """
     Tests whether the fair classifier performs similar to logistic regression
@@ -79,7 +79,6 @@ def test_same_logistic(random_xy_dataset_clf):
     _test_same(random_xy_dataset_clf)
 
 
-@pytest.mark.cvxpy
 def test_same_logistic_multiclass(random_xy_dataset_multiclf):
     """
     Tests whether the fair classifier performs similar to logistic regression
@@ -89,7 +88,6 @@ def test_same_logistic_multiclass(random_xy_dataset_multiclf):
     _test_same(random_xy_dataset_multiclf)
 
 
-@pytest.mark.cvxpy
 def test_regularization(sensitive_classification_dataset):
     """Tests whether increasing regularization decreases the norm of the coefficient vector"""
     X, y = sensitive_classification_dataset
@@ -104,7 +102,6 @@ def test_regularization(sensitive_classification_dataset):
         prev_theta_norm = theta_norm
 
 
-@pytest.mark.cvxpy
 def test_fairness(sensitive_classification_dataset):
     """tests whether fairness (measured by p percent score) increases as we decrease the covariance threshold"""
     X, y = sensitive_classification_dataset

diff --git a/tests/test_estimators/test_probweight_regression.py b/tests/test_estimators/test_probweight_regression.py
@@ -5,6 +5,8 @@
 from sklego.linear_model import ProbWeightRegression
 from tests.conftest import general_checks, nonmeta_checks, regressor_checks, select_tests
 
+pytestmark = pytest.mark.cvxpy
+
 
 @pytest.mark.parametrize(
     "test_fn",
@@ -13,15 +15,13 @@
         exclude=["check_sample_weights_invariance", "check_sample_weights_list", "check_sample_weights_pandas_series"],
     ),
 )
-@pytest.mark.cvxpy
 def test_estimator_checks(test_fn):
     regr_min_zero = ProbWeightRegression(non_negative=True)
     test_fn(ProbWeightRegression.__name__ + "_min_zero_true", regr_min_zero)
     regr_not_min_zero = ProbWeightRegression(non_negative=False)
     test_fn(ProbWeightRegression.__name__ + "_min_zero_true_false", regr_not_min_zero)
 
 
-@pytest.mark.cvxpy
 def test_shape_trained_model(random_xy_dataset_regr):
     X, y = random_xy_dataset_regr
     mod_no_intercept = ProbWeightRegression()

diff --git a/tests/test_estimators/test_umap_reconstruction.py b/tests/test_estimators/test_umap_reconstruction.py
@@ -5,6 +5,8 @@
 from sklego.decomposition import UMAPOutlierDetection
 from tests.conftest import general_checks, nonmeta_checks, outlier_checks, select_tests
 
+pytestmark = pytest.mark.umap
+
 
 @pytest.mark.parametrize(
     "test_fn",

diff --git a/tests/test_preprocessing/test_formulaic_transformer.py b/tests/test_preprocessing/test_formulaic_transformer.py
@@ -9,6 +9,8 @@
 
 from sklego.preprocessing import FormulaicTransformer
 
+pytestmark = pytest.mark.formulaic
+
 
 @pytest.fixture()
 def df():

diff --git a/tests/test_preprocessing/test_interval_encoder.py b/tests/test_preprocessing/test_interval_encoder.py
@@ -6,6 +6,8 @@
 from sklego.preprocessing import IntervalEncoder
 from tests.conftest import general_checks, transformer_checks
 
+pytestmark = pytest.mark.cvxpy
+
 
 @pytest.mark.parametrize(
     "test_fn",