MNT parameter validation for covariance.empirical_covariance (scikit-learn#25146)

shogohida · web-flow · commit 17b82783422b · 2022-12-09T08:15:05.000-05:00
diff --git a/sklearn/covariance/_empirical_covariance.py b/sklearn/covariance/_empirical_covariance.py
@@ -17,6 +17,7 @@
 from .. import config_context
 from ..base import BaseEstimator
 from ..utils import check_array
+from ..utils._param_validation import validate_params
 from ..utils.extmath import fast_logdet
 from ..metrics.pairwise import pairwise_distances
 
@@ -48,6 +49,12 @@ def log_likelihood(emp_cov, precision):
     return log_likelihood_
 
 
+@validate_params(
+    {
+        "X": ["array-like"],
+        "assume_centered": ["boolean"],
+    }
+)
 def empirical_covariance(X, *, assume_centered=False):
     """Compute the Maximum likelihood covariance estimator.
 
diff --git a/sklearn/tests/test_public_functions.py b/sklearn/tests/test_public_functions.py
@@ -96,6 +96,7 @@ def _check_function_param_validation(
 PARAM_VALIDATION_FUNCTION_LIST = [
     "sklearn.cluster.estimate_bandwidth",
     "sklearn.cluster.kmeans_plusplus",
+    "sklearn.covariance.empirical_covariance",
     "sklearn.feature_extraction.grid_to_graph",
     "sklearn.feature_extraction.img_to_graph",
     "sklearn.metrics.accuracy_score",
