@@ -694,6 +694,9 @@ class RandomForestClassifier(ForestClassifier):
694694 A random forest is a meta estimator that fits a number of decision tree
695695 classifiers on various sub-samples of the dataset and use averaging to
696696 improve the predictive accuracy and control over-fitting.
697+ The sub-sample size is always the same as the original
698+ input sample size but the samples are drawn with replacement if
699+ `bootstrap=True` (default).
697700
698701 Read more in the :ref:`User Guide <forest>`.
699702
@@ -715,7 +718,7 @@ class RandomForestClassifier(ForestClassifier):
715718 `int(max_features * n_features)` features are considered at each
716719 split.
717720 - If "auto", then `max_features=sqrt(n_features)`.
718- - If "sqrt", then `max_features=sqrt(n_features)`.
721+ - If "sqrt", then `max_features=sqrt(n_features)` (same as "auto") .
719722 - If "log2", then `max_features=log2(n_features)`.
720723 - If None, then `max_features=n_features`.
721724
@@ -883,6 +886,9 @@ class RandomForestRegressor(ForestRegressor):
883886 A random forest is a meta estimator that fits a number of classifying
884887 decision trees on various sub-samples of the dataset and use averaging
885888 to improve the predictive accuracy and control over-fitting.
889+ The sub-sample size is always the same as the original
890+ input sample size but the samples are drawn with replacement if
891+ `bootstrap=True` (default).
886892
887893 Read more in the :ref:`User Guide <forest>`.
888894
0 commit comments