scikit-learn
diff --git a/‎sklearn/ensemble/_gradient_boosting.c
Lines changed: 18 additions & 3 deletions b/‎sklearn/ensemble/_gradient_boosting.c
Lines changed: 18 additions & 3 deletions
diff --git a/‎sklearn/ensemble/gradient_boosting.py
Lines changed: 1 addition & 6 deletions b/‎sklearn/ensemble/gradient_boosting.py
Lines changed: 1 addition & 6 deletions
diff --git a/‎sklearn/ensemble/tests/test_gradient_boosting.py
Lines changed: 0 additions & 16 deletions b/‎sklearn/ensemble/tests/test_gradient_boosting.py
Lines changed: 0 additions & 16 deletions
@@ -45,7 +45,6 @@
 
 from ..tree.tree import DecisionTreeRegressor
 from ..tree._tree import DTYPE, TREE_LEAF
-from ..tree._splitter import PresortBestSplitter
 from ..tree._criterion import FriedmanMSE
 
 from ._gradient_boosting import predict_stages
@@ -1018,11 +1017,7 @@ def _fit_stages(self, X, y, y_pred, sample_weight, random_state,
 
         # init criterion and splitter
         criterion = FriedmanMSE(1)
-        splitter = PresortBestSplitter(criterion,
-                                       self.max_features_,
-                                       self.min_samples_leaf,
-                                       min_weight_leaf,
-                                       random_state)
+        splitter = 'best'
 
         if self.verbose:
             verbose_reporter = VerboseReporter(self.verbose)
 
@@ -984,22 +984,6 @@ def test_non_uniform_weights_toy_edge_case_reg():
         assert_greater(gb.predict([[1, 0]])[0], 0.5)
 
 
-def test_non_uniform_weights_toy_min_weight_leaf():
-    # Regression test for issue #4447
-    X = [[1, 0],
-         [1, 0],
-         [1, 0],
-         [0, 1],
-         ]
-    y = [0, 0, 1, 0]
-    # ignore the first 2 training samples by setting their weight to 0
-    sample_weight = [0, 0, 1, 1]
-    gb = GradientBoostingRegressor(n_estimators=5, min_weight_fraction_leaf=0.1)
-    gb.fit(X, y, sample_weight=sample_weight)
-    assert_true(gb.predict([[1, 0]])[0] > 0.5)
-    assert_almost_equal(gb.estimators_[0, 0].splitter.min_weight_leaf, 0.2)
-
-
 def test_non_uniform_weights_toy_edge_case_clf():
     X = [[1, 0],
          [1, 0],