ENH Add fitted check for kde (scikit-learn#16762)

Author: quangngd

sklearn/neighbors/_kde.py

@@ -8,7 +8,7 @@
 from scipy.special import gammainc
 from ..base import BaseEstimator
 from ..utils import check_array, check_random_state
-from ..utils.validation import _check_sample_weight
+from ..utils.validation import _check_sample_weight, check_is_fitted
 
 from ..utils.extmath import row_norms
 from ._ball_tree import BallTree, DTYPE
@@ -184,6 +184,7 @@ def score_samples(self, X):
             probability densities, so values will be low for high-dimensional
             data.
         """
+        check_is_fitted(self)
         # The returned density is normalized to the number of points.
         # For it to be a probability, we must scale it.  For this reason
         # we'll also scale atol.
@@ -241,6 +242,7 @@ def sample(self, n_samples=1, random_state=None):
         X : array_like, shape (n_samples, n_features)
             List of samples.
         """
+        check_is_fitted(self)
         # TODO: implement sampling for other valid kernel shapes
         if self.kernel not in ['gaussian', 'tophat']:
             raise NotImplementedError()

sklearn/neighbors/tests/test_kde.py

@@ -9,6 +9,7 @@
 from sklearn.datasets import make_blobs
 from sklearn.model_selection import GridSearchCV
 from sklearn.preprocessing import StandardScaler
+from sklearn.exceptions import NotFittedError
 import joblib
 
 
@@ -235,3 +236,15 @@ def test_pickling(tmpdir, sample_weight):
     scores_pickled = kde.score_samples(X)
 
     assert_allclose(scores, scores_pickled)
+
+
+@pytest.mark.parametrize('method', ['score_samples', 'sample'])
+def test_check_is_fitted(method):
+    # Check that predict raises an exception in an unfitted estimator.
+    # Unfitted estimators should raise a NotFittedError.
+    rng = np.random.RandomState(0)
+    X = rng.randn(10, 2)
+    kde = KernelDensity()
+
+    with pytest.raises(NotFittedError):
+        getattr(kde, method)(X)