ENH Adds n_features_in_ checking in gaussian_process (scikit-learn#18743)

Author: thomasjpfan

sklearn/gaussian_process/_gpc.py

@@ -14,7 +14,7 @@
 from ..base import BaseEstimator, ClassifierMixin, clone
 from .kernels \
     import RBF, CompoundKernel, ConstantKernel as C
-from ..utils.validation import check_is_fitted, check_array
+from ..utils.validation import check_is_fitted
 from ..utils import check_random_state
 from ..utils.optimize import _check_optimize_result
 from ..preprocessing import LabelEncoder
@@ -689,9 +689,11 @@ def predict(self, X):
         check_is_fitted(self)
 
         if self.kernel is None or self.kernel.requires_vector_input:
-            X = check_array(X, ensure_2d=True, dtype="numeric")
+            X = self._validate_data(X, ensure_2d=True, dtype="numeric",
+                                    reset=False)
         else:
-            X = check_array(X, ensure_2d=False, dtype=None)
+            X = self._validate_data(X, ensure_2d=False, dtype=None,
+                                    reset=False)
 
         return self.base_estimator_.predict(X)
 
@@ -717,9 +719,11 @@ def predict_proba(self, X):
                              "one_vs_rest mode instead.")
 
         if self.kernel is None or self.kernel.requires_vector_input:
-            X = check_array(X, ensure_2d=True, dtype="numeric")
+            X = self._validate_data(X, ensure_2d=True, dtype="numeric",
+                                    reset=False)
         else:
-            X = check_array(X, ensure_2d=False, dtype=None)
+            X = self._validate_data(X, ensure_2d=False, dtype=None,
+                                    reset=False)
 
         return self.base_estimator_.predict_proba(X)
 

sklearn/gaussian_process/_gpr.py

@@ -15,7 +15,6 @@
 from ..base import MultiOutputMixin
 from .kernels import RBF, ConstantKernel as C
 from ..utils import check_random_state
-from ..utils.validation import check_array
 from ..utils.optimize import _check_optimize_result
 from ..utils.validation import _deprecate_positional_args
 
@@ -320,9 +319,11 @@ def predict(self, X, return_std=False, return_cov=False):
                 "returning full covariance.")
 
         if self.kernel is None or self.kernel.requires_vector_input:
-            X = check_array(X, ensure_2d=True, dtype="numeric")
+            X = self._validate_data(X, ensure_2d=True, dtype="numeric",
+                                    reset=False)
         else:
-            X = check_array(X, ensure_2d=False, dtype=None)
+            X = self._validate_data(X, ensure_2d=False, dtype=None,
+                                    reset=False)
 
         if not hasattr(self, "X_train_"):  # Unfitted;predict based on GP prior
             if self.kernel is None:

sklearn/tests/test_common.py

@@ -272,7 +272,6 @@ def test_search_cv(estimator, check, request):
     'ensemble',
     'feature_extraction',
     'feature_selection',
-    'gaussian_process',
     'isotonic',
     'linear_model',
     'manifold',