FIX remove additional fit method

jmschrei · jmschrei · commit 6da6718e35fa · 2015-09-11T15:49:07.000+02:00
diff --git a/sklearn/ensemble/gradient_boosting.py b/sklearn/ensemble/gradient_boosting.py
@@ -929,6 +929,13 @@ def fit(self, X, y, sample_weight=None, monitor=None):
             computing held-out estimates, early stopping, model introspect, and
             snapshoting.
 
+        presort : bool, optional (default=False)
+
+            Whether to presort the data to speed up the finding of best splits in
+            fitting. By default gradient boosting uses presorting, but this may
+            slow down the training process on large datasets, or with deep trees.
+            This option is not available for sparse data.
+
         Returns
         -------
         self : object
@@ -1285,6 +1292,14 @@ class GradientBoostingClassifier(BaseGradientBoosting, ClassifierMixin):
         If None, the random number generator is the RandomState instance used
         by `np.random`.
 
+    presort : bool, optional (default=False)
+
+        Whether to presort the data to speed up the finding of best splits in
+        fitting. By default this is turned on for gradient boosting. However,
+        if the dataset is very large, or the trees being built are deep, this
+        may have performance costs. This option is not available for sparse
+        data.
+
     Attributes
     ----------
     feature_importances_ : array, shape = [n_features]
@@ -1336,7 +1351,8 @@ def __init__(self, loss='deviance', learning_rate=0.1, n_estimators=100,
                  min_samples_leaf=1, min_weight_fraction_leaf=0.,
                  max_depth=3, init=None, random_state=None,
                  max_features=None, verbose=0,
-                 max_leaf_nodes=None, warm_start=False):
+                 max_leaf_nodes=None, warm_start=False,
+                 presort=True):
 
         super(GradientBoostingClassifier, self).__init__(
             loss=loss, learning_rate=learning_rate, n_estimators=n_estimators,
@@ -1346,7 +1362,8 @@ def __init__(self, loss='deviance', learning_rate=0.1, n_estimators=100,
             max_depth=max_depth, init=init, subsample=subsample,
             max_features=max_features,
             random_state=random_state, verbose=verbose,
-            max_leaf_nodes=max_leaf_nodes, warm_start=warm_start)
+            max_leaf_nodes=max_leaf_nodes, warm_start=warm_start,
+            presort=presort)
 
     def _validate_y(self, y):
         self.classes_, y = np.unique(y, return_inverse=True)
@@ -1611,6 +1628,13 @@ class GradientBoostingRegressor(BaseGradientBoosting, RegressorMixin):
         If None, the random number generator is the RandomState instance used
         by `np.random`.
 
+    presort : bool, optional (default=False)
+
+        Whether to presort the data to speed up the finding of best splits in
+        fitting. By default this is turned on for gradient boosting. However,
+        if the dataset is very large, or the trees being built are deep, this
+        may have performance costs. This option is not available for sparse
+        data.
 
     Attributes
     ----------
@@ -1660,7 +1684,7 @@ def __init__(self, loss='ls', learning_rate=0.1, n_estimators=100,
                  min_samples_leaf=1, min_weight_fraction_leaf=0.,
                  max_depth=3, init=None, random_state=None,
                  max_features=None, alpha=0.9, verbose=0, max_leaf_nodes=None,
-                 warm_start=False):
+                 warm_start=False, presort=True):
 
         super(GradientBoostingRegressor, self).__init__(
             loss=loss, learning_rate=learning_rate, n_estimators=n_estimators,
@@ -1670,7 +1694,8 @@ def __init__(self, loss='ls', learning_rate=0.1, n_estimators=100,
             max_depth=max_depth, init=init, subsample=subsample,
             max_features=max_features,
             random_state=random_state, alpha=alpha, verbose=verbose,
-            max_leaf_nodes=max_leaf_nodes, warm_start=warm_start)
+            max_leaf_nodes=max_leaf_nodes, warm_start=warm_start,
+            presort=True)
 
     def predict(self, X):
         """Predict regression target for X.
