[MNT] Raise version bound for scikit-learn 1.6 (aeon-toolkit#2486)

* update ver and new tags

* default tags

* toml

* Update _shapelets.py

Fix linear estimator coefs issue

* expected results

* Change expected results

* update

* only linux

* remove mixins just to see test

* revert

---------

Co-authored-by: Antoine Guillaume <antoine.guillaume45@gmail.com>

Author: MatthewMiddlehurst

aeon/base/_base.py

@@ -415,6 +415,18 @@ def __sklearn_is_fitted__(self):
         """Check fitted status and return a Boolean value."""
         return self.is_fitted
 
+    def __sklearn_tags__(self):
+        """Return sklearn style tags for the estimator."""
+        aeon_tags = self.get_tags()
+        sklearn_tags = super().__sklearn_tags__()
+        sklearn_tags.non_deterministic = aeon_tags.get("non_deterministic", False)
+        sklearn_tags.target_tags.one_d_labels = True
+        sklearn_tags.input_tags.three_d_array = True
+        sklearn_tags.input_tags.allow_nan = aeon_tags.get(
+            "capability:missing_values", False
+        )
+        return sklearn_tags
+
     def _validate_data(self, **kwargs):
         """Sklearn data validation."""
         raise NotImplementedError(

aeon/classification/base.py

@@ -26,6 +26,7 @@ class name: BaseClassifier
 
 import numpy as np
 import pandas as pd
+from sklearn.base import ClassifierMixin
 from sklearn.metrics import get_scorer, get_scorer_names
 from sklearn.model_selection import cross_val_predict
 
@@ -35,7 +36,7 @@ class name: BaseClassifier
 from aeon.utils.validation.labels import check_classification_y
 
 
-class BaseClassifier(BaseCollectionEstimator):
+class BaseClassifier(ClassifierMixin, BaseCollectionEstimator):
     """
     Abstract base class for time series classifiers.
 
@@ -66,7 +67,6 @@ def __init__(self):
         self.classes_ = []  # classes seen in y, unique labels
         self.n_classes_ = -1  # number of unique classes in y
         self._class_dictionary = {}
-        self._estimator_type = "classifier"
 
         super().__init__()
 

aeon/clustering/base.py

@@ -7,11 +7,12 @@
 from typing import final
 
 import numpy as np
+from sklearn.base import ClusterMixin
 
 from aeon.base import BaseCollectionEstimator
 
 
-class BaseClusterer(BaseCollectionEstimator):
+class BaseClusterer(ClusterMixin, BaseCollectionEstimator):
     """Abstract base class for time series clusterers.
 
     Parameters
@@ -26,10 +27,6 @@ class BaseClusterer(BaseCollectionEstimator):
 
     @abstractmethod
     def __init__(self):
-        # required for compatibility with some sklearn interfaces e.g.
-        # CalibratedClassifierCV
-        self._estimator_type = "clusterer"
-
         super().__init__()
 
     @final
@@ -132,24 +129,6 @@ def fit_predict(self, X, y=None) -> np.ndarray:
             to return.
         y: ignored, exists for API consistency reasons.
 
-        Returns
-        -------
-        np.ndarray (1d array of shape (n_cases,))
-            Index of the cluster each time series in X belongs to.
-        """
-        return self._fit_predict(X, y)
-
-    def _fit_predict(self, X, y=None) -> np.ndarray:
-        """Fit predict using base methods.
-
-        Parameters
-        ----------
-        X : np.ndarray (2d or 3d array of shape (n_cases, n_timepoints) or shape
-            (n_cases, n_channels, n_timepoints)).
-            Time series instances to train clusterer and then have indexes each belong
-            to return.
-        y: ignored, exists for API consistency reasons.
-
         Returns
         -------
         np.ndarray (1d array of shape (n_cases,))

aeon/regression/base.py

@@ -25,6 +25,7 @@ class name: BaseRegressor
 
 import numpy as np
 import pandas as pd
+from sklearn.base import RegressorMixin
 from sklearn.metrics import get_scorer, get_scorer_names
 from sklearn.model_selection import cross_val_predict
 from sklearn.utils.multiclass import type_of_target
@@ -33,7 +34,7 @@ class name: BaseRegressor
 from aeon.base._base import _clone_estimator
 
 
-class BaseRegressor(BaseCollectionEstimator):
+class BaseRegressor(RegressorMixin, BaseCollectionEstimator):
     """Abstract base class for time series regressors.
 
     The base regressor specifies the methods and method signatures that all
@@ -54,9 +55,6 @@ class BaseRegressor(BaseCollectionEstimator):
 
     @abstractmethod
     def __init__(self):
-        # required for compatibility with some sklearn interfaces
-        self._estimator_type = "regressor"
-
         super().__init__()
 
     @final

aeon/regression/feature_based/_catch22.py

@@ -94,8 +94,8 @@ class Catch22Regressor(BaseRegressor):
     >>> reg.fit(X, y)
     Catch22Regressor(...)
     >>> reg.predict(X)
-    array([0.63821896, 1.0906666 , 0.58323551, 1.57550709, 0.48413489,
-           0.70976176, 1.33206165, 1.09927538, 1.51673405, 0.31683308])
+    array([0.63821896, 1.0906666 , 0.64351536, 1.57550709, 0.46036267,
+           0.79297397, 1.32882497, 1.12603087, 1.51673405, 0.31683308])
     """
 
     _tags = {

aeon/regression/sklearn/tests/test_rotation_forest_regressor.py

@@ -24,21 +24,21 @@ def test_rotf_output():
     rotf.fit(X_train, y_train)
 
     expected = [
-        0.02694297,
-        0.02694297,
-        0.01997832,
-        0.04276962,
-        0.09027588,
-        0.02706564,
-        0.02553648,
-        0.04075808,
-        0.02900289,
-        0.04248546,
-        0.02694297,
-        0.03667328,
-        0.0235855,
-        0.03444119,
-        0.0235855,
+        0.026,
+        0.0245,
+        0.0224,
+        0.0453,
+        0.0892,
+        0.0314,
+        0.026,
+        0.0451,
+        0.0287,
+        0.04,
+        0.026,
+        0.0378,
+        0.0265,
+        0.0356,
+        0.0281,
     ]
 
     np.testing.assert_array_almost_equal(expected, rotf.predict(X_test[:15]), decimal=4)

aeon/testing/estimator_checking/_yield_classification_checks.py

@@ -31,7 +31,7 @@
 def _yield_classification_checks(estimator_class, estimator_instances, datatypes):
     """Yield all classification checks for an aeon classifier."""
     # only class required
-    if sys.platform != "darwin":  # We cannot guarantee same results on ARM macOS
+    if sys.platform == "linux":  # We cannot guarantee same results on ARM macOS
         # Compare against results for both UnitTest and BasicMotions if available
         yield partial(
             check_classifier_against_expected_results,

aeon/testing/estimator_checking/_yield_regression_checks.py

@@ -26,7 +26,7 @@
 def _yield_regression_checks(estimator_class, estimator_instances, datatypes):
     """Yield all regression checks for an aeon regressor."""
     # only class required
-    if sys.platform != "darwin":  # We cannot guarantee same results on ARM macOS
+    if sys.platform == "linux":  # We cannot guarantee same results on ARM macOS
         # Compare against results for both Covid3Month and CardanoSentiment if available
         yield partial(
             check_regressor_against_expected_results,

aeon/testing/estimator_checking/_yield_transformation_checks.py

@@ -26,7 +26,8 @@
 def _yield_transformation_checks(estimator_class, estimator_instances, datatypes):
     """Yield all transformation checks for an aeon transformer."""
     # only class required
-    if sys.platform != "darwin":
+    if sys.platform == "linux":  # We cannot guarantee same results on ARM macOS
+        # Compare against results for both UnitTest and BasicMotions if available
         yield partial(
             check_transformer_against_expected_results,
             estimator_class=estimator_class,

aeon/testing/expected_results/expected_classifier_outputs.py

@@ -67,16 +67,16 @@
 )
 unit_test_proba["TemporalDictionaryEnsemble"] = np.array(
     [
-        [0.2778, 0.7222],
-        [0.7222, 0.2778],
+        [0.3307, 0.6693],
+        [0.6693, 0.3307],
         [0.0, 1.0],
-        [0.6251, 0.3749],
-        [0.3749, 0.6251],
+        [0.5538, 0.4462],
+        [0.6693, 0.3307],
         [1.0, 0.0],
-        [0.3749, 0.6251],
+        [0.4462, 0.5538],
         [0.0, 1.0],
-        [0.4653, 0.5347],
-        [0.3749, 0.6251],
+        [0.5538, 0.4462],
+        [0.4462, 0.5538],
     ]
 )
 unit_test_proba["WEASEL"] = np.array(
@@ -263,16 +263,16 @@
 )
 unit_test_proba["HIVECOTEV2"] = np.array(
     [
-        [0.0613, 0.9387],
-        [0.5531, 0.4479],
-        [0.0431, 0.9569],
+        [0.2239, 0.7761],
+        [0.6732, 0.3268],
+        [0.1211, 0.8789],
         [1.0, 0.0],
-        [0.9751, 0.0249],
+        [0.9818, 0.0182],
         [1.0, 0.0],
-        [0.7398, 0.2602],
-        [0.0365, 0.9635],
-        [0.7829, 0.2171],
-        [0.9236, 0.0764],
+        [0.7201, 0.2799],
+        [0.2058, 0.7942],
+        [0.8412, 0.1588],
+        [0.9441, 0.0559],
     ]
 )
 unit_test_proba["CanonicalIntervalForestClassifier"] = np.array(
@@ -293,12 +293,12 @@
     [
         [0.1, 0.9],
         [0.8, 0.2],
-        [0.0, 1.0],
+        [0.1, 0.9],
         [1.0, 0.0],
         [0.7, 0.3],
         [0.9, 0.1],
         [0.8, 0.2],
-        [0.4, 0.6],
+        [0.5, 0.5],
         [0.9, 0.1],
         [1.0, 0.0],
     ]
@@ -379,11 +379,11 @@
         [0.3505, 0.6495],
         [0.1753, 0.8247],
         [0.8247, 0.1753],
-        [0.3505, 0.6495],
+        [0.6495, 0.3505],
         [0.701, 0.299],
         [0.6495, 0.3505],
         [0.1753, 0.8247],
-        [0.5258, 0.4742],
+        [0.8247, 0.1753],
         [1.0, 0.0],
     ]
 )
@@ -656,12 +656,12 @@
 )
 basic_motions_proba["FreshPRINCEClassifier"] = np.array(
     [
-        [0.0, 0.0, 0.1, 0.9],
+        [0.0, 0.0, 0.2, 0.8],
         [0.9, 0.1, 0.0, 0.0],
         [0.0, 0.0, 0.8, 0.2],
         [0.1, 0.9, 0.0, 0.0],
-        [0.1, 0.0, 0.0, 0.9],
-        [0.0, 0.0, 0.1, 0.9],
+        [0.1, 0.0, 0.1, 0.8],
+        [0.0, 0.0, 0.2, 0.8],
         [0.7, 0.3, 0.0, 0.0],
         [0.0, 0.0, 1.0, 0.0],
         [0.0, 1.0, 0.0, 0.0],
@@ -782,15 +782,15 @@
 )
 basic_motions_proba["DrCIFClassifier"] = np.array(
     [
+        [0.1, 0.0, 0.2, 0.7],
+        [0.5, 0.4, 0.0, 0.1],
+        [0.0, 0.0, 0.8, 0.2],
+        [0.1, 0.9, 0.0, 0.0],
+        [0.1, 0.0, 0.3, 0.6],
         [0.0, 0.0, 0.2, 0.8],
-        [0.4, 0.5, 0.1, 0.0],
-        [0.0, 0.0, 0.7, 0.3],
-        [0.2, 0.8, 0.0, 0.0],
-        [0.0, 0.0, 0.3, 0.7],
-        [0.0, 0.0, 0.3, 0.7],
-        [0.7, 0.2, 0.1, 0.0],
-        [0.0, 0.0, 0.7, 0.3],
-        [0.1, 0.7, 0.1, 0.1],
+        [0.5, 0.3, 0.0, 0.2],
+        [0.0, 0.0, 0.8, 0.2],
+        [0.2, 0.7, 0.0, 0.1],
         [0.0, 0.9, 0.0, 0.1],
     ]
 )