unexclude SFA

aeon/testing/estimator_checking/_yield_estimator_checks.py

@@ -9,7 +9,6 @@
 
 import joblib
 import numpy as np
-import pandas as pd
 from sklearn.exceptions import NotFittedError
 
 from aeon.anomaly_detection.base import BaseAnomalyDetector
@@ -605,61 +604,6 @@ def check_raises_not_fitted_error(estimator, datatype):
                 _run_estimator_method(estimator, method, datatype, "test")
 
 
-def _equal_outputs(output1, output2):
-    """Test whether two outputs from an estimator are logically identical.
-
-    Valid data structures are:
-    1. float: returns a single value (e.g. forecasting)
-    2. numpy array:
-        scalars: stores an equal length collection or series (default)
-        objects: an array of arrays stored as objects (e.g. SimilaritySearch)
-    3. dict: a histogram of counts, usually of discretised sub-series (e.g. SFA)
-    4. pd.DataFrame: series stored in dataframe (e.g. Dobin)
-    5. list: stores possibly unequal length series in a format 2-4
-    6. tuple: stores two or more series/collections in a format 2-4 (e.g. imbalance
-    transformers)
-
-    """
-    if type(output1) is not type(output2):
-        return False
-    if np.issubdtype(type(output1), np.floating):
-        return np.isclose(output1, output2)
-    if np.issubdtype(type(output1), np.bool_):
-        return output1 == output2
-    if isinstance(output1, np.ndarray):  # 1. X an equal length collection or series
-        if np.isscalar(output1):
-            return np.allclose(output1, output2, equal_nan=True)
-        for i in range(len(output1)):
-            if not _equal_outputs(output1[i], output2[i]):
-                return False
-            return True
-    if isinstance(output1, dict):  # 2. X a dictionary, dense collection or series
-        if output1.keys() != output2.keys():
-            return False
-        for k in output1.keys():
-            if not _equal_outputs(output1[k], output2[k]):
-                return False
-        return True
-    if isinstance(output1, pd.DataFrame) or isinstance(output1, pd.Series):
-        # 3. X a dataframe
-        return np.allclose(output1.values, output2.values, equal_nan=True)
-    if isinstance(output1, list):  # X a possibly unequal length collection
-        if len(output1) != len(output2):
-            return False
-        for i in range(len(output1)):
-            if not _equal_outputs(output1[i], output2[i]):
-                return False
-        return True
-    if isinstance(output1, tuple):  # returns (X,y)
-        if len(output1) != len(output2):
-            return False
-        for i in range(len(output1)):
-            if not _equal_outputs(output1[i], output2[i]):
-                return False
-        return True
-    return False
-
-
 def check_persistence_via_pickle(estimator, datatype):
     """Check that we can pickle all estimators."""
     estimator = _clone_estimator(estimator, random_state=0)

aeon/testing/testing_config.py

@@ -50,7 +50,7 @@
     "RSASTClassifier": ["check_fit_deterministic"],
     "SAST": ["check_fit_deterministic"],
     "RSAST": ["check_fit_deterministic"],
-    "SFA": ["check_persistence_via_pickle", "check_fit_deterministic"],
+    #    "SFA": ["check_persistence_via_pickle", "check_fit_deterministic"],
     # missed in legacy testing, changes state in predict/transform
     "FLUSSSegmenter": ["check_non_state_changing_method"],
     "InformationGainSegmenter": ["check_non_state_changing_method"],