[MNT] Unit testing revamp part 2: classification

.pre-commit-config.yaml

@@ -67,7 +67,7 @@ repos:
         additional_dependencies: [ isort==5.13.2 ]
         args: [ "--nbqa-dont-skip-bad-cells", "--profile=black", "--multi-line=3" ]
       - id: nbqa-black
-        additional_dependencies: [ black==24.2.0 ]
+        additional_dependencies: [ black==24.4.2 ]
         args: [ "--nbqa-dont-skip-bad-cells" ]
       - id: nbqa-flake8
         additional_dependencies: [ flake8==7.0.0 ]

aeon/base/_base.py

@@ -283,7 +283,7 @@ class attribute via nested inheritance. NOT overridden by dynamic
         return deepcopy(collected_tags)
 
     @classmethod
-    def get_class_tag(cls, tag_name, tag_value_default=None):
+    def get_class_tag(cls, tag_name, tag_value_default=None, raise_error=False):
         """
         Get tag value from estimator class (only class tags).
 
@@ -293,12 +293,19 @@ def get_class_tag(cls, tag_name, tag_value_default=None):
             Name of tag value.
         tag_value_default : any type
             Default/fallback value if tag is not found.
+        raise_error : bool
+            Whether a ValueError is raised when the tag is not found.
 
         Returns
         -------
         tag_value :
-            Value of the `tag_name` tag in self. If not found, returns
-            `tag_value_default`.
+            Value of the `tag_name` tag in self. If not found, returns an error if
+            raise_error is True, otherwise it returns `tag_value_default`.
+
+        Raises
+        ------
+        ValueError if raise_error is True i.e. if tag_name is not in self.get_tags(
+        ).keys()
 
         See Also
         --------
@@ -314,7 +321,12 @@ def get_class_tag(cls, tag_name, tag_value_default=None):
         """
         collected_tags = cls.get_class_tags()
 
-        return collected_tags.get(tag_name, tag_value_default)
+        tag_value = collected_tags.get(tag_name, tag_value_default)
+
+        if raise_error and tag_name not in collected_tags.keys():
+            raise ValueError(f"Tag with name {tag_name} could not be found.")
+
+        return tag_value
 
     def get_tags(self):
         """

aeon/base/tests/test_base_collection.py

@@ -4,15 +4,18 @@
 import pytest
 
 from aeon.base import BaseCollectionEstimator
-from aeon.testing.testing_data import EQUAL_LENGTH_UNIVARIATE, UNEQUAL_LENGTH_UNIVARIATE
+from aeon.testing.testing_data import (
+    EQUAL_LENGTH_UNIVARIATE_CLASSIFICATION,
+    UNEQUAL_LENGTH_UNIVARIATE_CLASSIFICATION,
+)
 from aeon.utils import COLLECTIONS_DATA_TYPES
 from aeon.utils.validation import get_type
 
 
 @pytest.mark.parametrize("data", COLLECTIONS_DATA_TYPES)
 def test__get_metadata(data):
     """Test get meta data."""
-    X = EQUAL_LENGTH_UNIVARIATE[data]
+    X = EQUAL_LENGTH_UNIVARIATE_CLASSIFICATION[data]["train"][0]
     meta = BaseCollectionEstimator._get_metadata(X)
     assert not meta["multivariate"]
     assert not meta["missing_values"]
@@ -68,7 +71,7 @@ def test__convert_X(internal_type, data):
     """
     cls = BaseCollectionEstimator()
     # Equal length should default to numpy3D
-    X = EQUAL_LENGTH_UNIVARIATE[data]
+    X = EQUAL_LENGTH_UNIVARIATE_CLASSIFICATION[data]["train"][0]
     cls.metadata_ = cls._check_X(X)
     X2 = cls._convert_X(X)
     assert get_type(X2) == cls.get_tag("X_inner_type")
@@ -88,19 +91,19 @@ def test__convert_X(internal_type, data):
     cls.set_tags(**{"X_inner_type": ["nested_univ", internal_type]})
     X2 = cls._convert_X(X)
     assert get_type(X2) == internal_type
-    if data in UNEQUAL_LENGTH_UNIVARIATE.keys():
-        if internal_type in UNEQUAL_LENGTH_UNIVARIATE.keys():
+    if data in UNEQUAL_LENGTH_UNIVARIATE_CLASSIFICATION.keys():
+        if internal_type in UNEQUAL_LENGTH_UNIVARIATE_CLASSIFICATION.keys():
             cls.set_tags(**{"capability:unequal_length": True})
             cls.set_tags(**{"X_inner_type": ["nested_univ", "np-list", internal_type]})
-            X = UNEQUAL_LENGTH_UNIVARIATE[data]
+            X = UNEQUAL_LENGTH_UNIVARIATE_CLASSIFICATION[data]["train"][0]
             X2 = cls._convert_X(X)
             assert get_type(X2) == "np-list"
 
 
 @pytest.mark.parametrize("data", COLLECTIONS_DATA_TYPES)
 def test_preprocess_collection(data):
     """Test the functionality for preprocessing fit."""
-    data = EQUAL_LENGTH_UNIVARIATE[data]
+    data = EQUAL_LENGTH_UNIVARIATE_CLASSIFICATION[data]["train"][0]
     cls = BaseCollectionEstimator()
     X = cls._preprocess_collection(data)
     assert cls._n_jobs == 1

aeon/classification/convolution_based/_arsenal.py

@@ -295,6 +295,8 @@ def _fit_arsenal(self, X, y, keep_transformed_data=False):
         else:
             raise ValueError(f"Invalid Rocket transformer: {self.rocket_transform}")
 
+        rng = check_random_state(self.random_state)
+
         if time_limit > 0:
             self.n_estimators_ = 0
             self.estimators_ = []
@@ -307,16 +309,7 @@ def _fit_arsenal(self, X, y, keep_transformed_data=False):
                 fit = Parallel(n_jobs=self._n_jobs, prefer="threads")(
                     delayed(self._fit_ensemble_estimator)(
                         _clone_estimator(
-                            base_rocket,
-                            (
-                                None
-                                if self.random_state is None
-                                else (
-                                    255 if self.random_state == 0 else self.random_state
-                                )
-                                * 37
-                                * (i + 1)
-                            ),
+                            base_rocket, rng.randint(np.iinfo(np.int32).max)
                         ),
                         X,
                         y,
@@ -335,16 +328,7 @@ def _fit_arsenal(self, X, y, keep_transformed_data=False):
         else:
             fit = Parallel(n_jobs=self._n_jobs, prefer="threads")(
                 delayed(self._fit_ensemble_estimator)(
-                    _clone_estimator(
-                        base_rocket,
-                        (
-                            None
-                            if self.random_state is None
-                            else (255 if self.random_state == 0 else self.random_state)
-                            * 37
-                            * (i + 1)
-                        ),
-                    ),
+                    _clone_estimator(base_rocket, rng.randint(np.iinfo(np.int32).max)),
                     X,
                     y,
                     keep_transformed_data=keep_transformed_data,