0.7.1 patch (#81)

* testing and tags fix

* patch

Author: MatthewMiddlehurst

.github/workflows/release.yml

@@ -52,7 +52,7 @@ jobs:
     strategy:
       matrix:
         os: [ ubuntu-24.04, macos-14, windows-2022 ]
-        python-version: [ "3.9", "3.10", "3.11", "3.12" ]
+        python-version: [ "3.9", "3.10", "3.11", "3.12", "3.13" ]
 
     steps:
       - uses: actions/checkout@v4

pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "tsml"
-version = "0.7.0"
+version = "0.7.1"
 description = "A development sandbox for time series machine learning algorithms which dont fit into aeon."
 authors = [
     {name = "Matthew Middlehurst", email = "[email protected]"},

tsml/__init__.py

@@ -1,3 +1,3 @@
 """tsml."""
 
-__version__ = "0.7.0"
+__version__ = "0.7.1"

tsml/base.py

@@ -247,9 +247,6 @@ def _check_n_features(self, X: Union[np.ndarray, List[np.ndarray]], reset: bool)
                 f"is expecting {self.n_features_in_[1]} series length as input."
             )
 
-    def _get_tags(self) -> dict:
-        return _safe_tags(self)
-
     def _more_tags(self) -> dict:
         return _DEFAULT_TAGS
 

tsml/tests/test_estimators_sklearn.py

@@ -16,6 +16,7 @@
 from sklearn.model_selection import train_test_split
 from sklearn.pipeline import make_pipeline
 from sklearn.preprocessing import scale
+from sklearn.utils._tags import _safe_tags as _safe_tags_sklearn
 from sklearn.utils._testing import (
     SkipTest,
     assert_allclose,
@@ -28,7 +29,7 @@
 from sklearn.utils.validation import _num_samples, check_is_fitted
 
 import tsml.utils.testing as test_utils
-from tsml.utils._tags import _DEFAULT_TAGS, _safe_tags
+from tsml.utils._tags import _safe_tags
 
 
 @ignore_warnings(category=FutureWarning)
@@ -1409,8 +1410,8 @@ def check_estimator_get_tags_default_keys(name, estimator_orig):
     if not hasattr(estimator, "_get_tags"):
         return
 
+    default_tags_keys = set(_safe_tags_sklearn(estimator).keys())
     tags_keys = set(estimator._get_tags().keys())
-    default_tags_keys = set(_DEFAULT_TAGS.keys())
     assert tags_keys.intersection(default_tags_keys) == default_tags_keys, (
         f"{name}._get_tags() is missing entries for the following default tags: "
         f"{default_tags_keys - tags_keys.intersection(default_tags_keys)}"

tsml/tests/test_sklearn_compatability.py

@@ -0,0 +1,119 @@
+"""Unit tests for aeon classifier compatability with sklearn interfaces."""
+
+__maintainer__ = []
+__all__ = [
+    "test_sklearn_cross_validation",
+    "test_sklearn_cross_validation_iterators",
+    "test_sklearn_parameter_tuning",
+    "test_sklearn_composite_classifiers",
+]
+
+import numpy as np
+import pytest
+from sklearn.calibration import CalibratedClassifierCV
+from sklearn.ensemble import VotingClassifier
+from sklearn.experimental import enable_halving_search_cv  # noqa
+from sklearn.model_selection import (
+    GridSearchCV,
+    GroupKFold,
+    GroupShuffleSplit,
+    HalvingGridSearchCV,
+    HalvingRandomSearchCV,
+    KFold,
+    LeaveOneOut,
+    LeavePGroupsOut,
+    LeavePOut,
+    RandomizedSearchCV,
+    RepeatedKFold,
+    ShuffleSplit,
+    StratifiedKFold,
+    StratifiedShuffleSplit,
+    TimeSeriesSplit,
+    cross_val_score,
+)
+from sklearn.pipeline import Pipeline
+
+from tsml.dummy import DummyClassifier
+from tsml.transformations import PeriodogramTransformer
+from tsml.utils.testing import generate_3d_test_data
+
+# StratifiedGroupKFold(n_splits=2), removed because it is not available in sklearn 0.24
+CROSS_VALIDATION_METHODS = [
+    KFold(n_splits=2),
+    RepeatedKFold(n_splits=2, n_repeats=2),
+    LeaveOneOut(),
+    LeavePOut(p=5),
+    ShuffleSplit(n_splits=2, test_size=0.25),
+    StratifiedKFold(n_splits=2),
+    StratifiedShuffleSplit(n_splits=2, test_size=0.25),
+    GroupKFold(n_splits=2),
+    LeavePGroupsOut(n_groups=5),
+    GroupShuffleSplit(n_splits=2, test_size=0.25),
+    TimeSeriesSplit(n_splits=2),
+]
+PARAMETER_TUNING_METHODS = [
+    GridSearchCV,
+    RandomizedSearchCV,
+    HalvingGridSearchCV,
+    HalvingRandomSearchCV,
+]
+COMPOSITE_ESTIMATORS = [
+    Pipeline(
+        [
+            ("transform", PeriodogramTransformer()),
+            ("clf", DummyClassifier()),
+        ]
+    ),
+    VotingClassifier(
+        estimators=[
+            ("clf1", DummyClassifier()),
+            ("clf2", DummyClassifier()),
+            ("clf3", DummyClassifier()),
+        ]
+    ),
+    CalibratedClassifierCV(
+        estimator=DummyClassifier(),
+        cv=2,
+    ),
+]
+
+
+def test_sklearn_cross_validation():
+    """Test sklearn cross-validation works with tsml data and classifiers."""
+    clf = DummyClassifier()
+    X, y = generate_3d_test_data(n_samples=20, n_channels=2, series_length=30)
+    scores = cross_val_score(clf, X, y=y, cv=KFold(n_splits=2))
+    assert isinstance(scores, np.ndarray)
+
+
+@pytest.mark.parametrize("cross_validation_method", CROSS_VALIDATION_METHODS)
+def test_sklearn_cross_validation_iterators(cross_validation_method):
+    """Test if sklearn cross-validation iterators can handle tsml data."""
+    X, y = generate_3d_test_data(n_samples=20, n_channels=2, series_length=30)
+    groups = [1, 1, 2, 2, 3, 3, 4, 4, 5, 5, 6, 6, 7, 7, 8, 8, 9, 9, 10, 10]
+
+    for train, test in cross_validation_method.split(X=X, y=y, groups=groups):
+        assert isinstance(train, np.ndarray) and isinstance(test, np.ndarray)
+
+
+@pytest.mark.parametrize("parameter_tuning_method", PARAMETER_TUNING_METHODS)
+def test_sklearn_parameter_tuning(parameter_tuning_method):
+    """Test if sklearn parameter tuners can handle tsml data and classifiers."""
+    clf = DummyClassifier()
+    param_grid = {"strategy": ["prior", "constant"], "constant": [0, 1]}
+    X, y = generate_3d_test_data(n_samples=20, n_channels=2, series_length=30)
+
+    parameter_tuning_method = parameter_tuning_method(
+        clf, param_grid, cv=KFold(n_splits=2)
+    )
+    parameter_tuning_method.fit(X, y)
+    assert isinstance(parameter_tuning_method.best_estimator_, DummyClassifier)
+
+
+@pytest.mark.parametrize("composite_classifier", COMPOSITE_ESTIMATORS)
+def test_sklearn_composite_classifiers(composite_classifier):
+    """Test if sklearn composite classifiers can handle tsml data and classifiers."""
+    X, y = generate_3d_test_data(n_samples=20, n_channels=2, series_length=30)
+    composite_classifier.fit(X, y)
+    preds = composite_classifier.predict(X=X)
+    assert isinstance(preds, np.ndarray)