Merge pull request scikit-learn#5531 from arjoly/float-min_samples

[MRG +2 ]  min_samples_split and min_samples_leaf now accept a percentage

doc/modules/ensemble.rst

@@ -165,20 +165,20 @@ in bias::
     >>> X, y = make_blobs(n_samples=10000, n_features=10, centers=100,
     ...     random_state=0)
 
-    >>> clf = DecisionTreeClassifier(max_depth=None, min_samples_split=1,
+    >>> clf = DecisionTreeClassifier(max_depth=None, min_samples_split=2,
     ...     random_state=0)
     >>> scores = cross_val_score(clf, X, y)
     >>> scores.mean()                             # doctest: +ELLIPSIS
     0.97...
 
     >>> clf = RandomForestClassifier(n_estimators=10, max_depth=None,
-    ...     min_samples_split=1, random_state=0)
+    ...     min_samples_split=2, random_state=0)
     >>> scores = cross_val_score(clf, X, y)
     >>> scores.mean()                             # doctest: +ELLIPSIS
     0.999...
 
     >>> clf = ExtraTreesClassifier(n_estimators=10, max_depth=None,
-    ...     min_samples_split=1, random_state=0)
+    ...     min_samples_split=2, random_state=0)
     >>> scores = cross_val_score(clf, X, y)
     >>> scores.mean() > 0.999
     True

doc/modules/tree.rst

@@ -343,7 +343,8 @@ Tips on practical use
   * Use ``min_samples_split`` or ``min_samples_leaf`` to control the number of
     samples at a leaf node.  A very small number will usually mean the tree
     will overfit, whereas a large number will prevent the tree from learning
-    the data.  Try ``min_samples_leaf=5`` as an initial value.
+    the data. Try ``min_samples_leaf=5`` as an initial value. If the sample size
+    varies greatly, a float number can be used as percentage in these two parameters.
     The main difference between the two is that ``min_samples_leaf`` guarantees
     a minimum number of samples in a leaf, while ``min_samples_split`` can
     create arbitrary small leaves, though ``min_samples_split`` is more common

doc/whats_new.rst

@@ -21,8 +21,8 @@ New features
      implementation supports kernel engineering, gradient-based hyperparameter optimization or
      sampling of functions from GP prior and GP posterior. Extensive documentation and
      examples are provided. By `Jan Hendrik Metzen`_.
-     
-   - Added the :class:`ensemble.IsolationForest` class for anomaly detection based on 
+
+   - Added the :class:`ensemble.IsolationForest` class for anomaly detection based on
      random forests. By `Nicolas Goix`_.
 
 Enhancements
@@ -39,8 +39,18 @@ Enhancements
      method ``decision_path`` which returns the decision path of samples in
      the tree. By `Arnaud Joly`_
 
-   - A new example has been added unveling the decision tree structure.
-     By `Arnaud Joly`_
+
+    - The random forest, extra tree and decision tree estimators now has a
+      method ``decision_path`` which returns the decision path of samples in
+      the tree. By `Arnaud Joly`_
+
+    - A new example has been added unveling the decision tree structure.
+      By `Arnaud Joly`_
+
+    - Random forest, extra trees, decision trees and gradient boosting estimator
+      accept the parameter ``min_samples_split`` and ``min_samples_leaf``
+      provided as a percentage of the training samples. By
+      `yelite`_ and `Arnaud Joly`_
 
 Bug fixes
 .........
@@ -65,6 +75,10 @@ Bug fixes
       :class:`decomposition.KernelPCA`, :class:`manifold.LocallyLinearEmbedding`,
       and :class:`manifold.SpectralEmbedding`. By `Peter Fischer`_.
 
+    - Random forest, extra trees, decision trees and gradient boosting
+      won't accept anymore ``min_samples_split=1`` as at least 2 samples
+      are required to split a decision tree node. By `Arnaud Joly`_
+
 API changes summary
 -------------------
 
@@ -3854,3 +3868,4 @@ David Huard, Dave Morrill, Ed Schofield, Travis Oliphant, Pearu Peterson.
 .. _Graham Clenaghan: https://github.com/gclenaghan
 .. _Giorgio Patrini: https://github.com/giorgiop
 .. _Elvis Dohmatob: https://github.com/dohmatob
+.. _yelite https://github.com/yelite

examples/ensemble/plot_gradient_boosting_regression.py

@@ -33,7 +33,7 @@
 
 ###############################################################################
 # Fit regression model
-params = {'n_estimators': 500, 'max_depth': 4, 'min_samples_split': 1,
+params = {'n_estimators': 500, 'max_depth': 4, 'min_samples_split': 2,
           'learning_rate': 0.01, 'loss': 'ls'}
 clf = ensemble.GradientBoostingRegressor(**params)
 

sklearn/ensemble/forest.py

@@ -777,36 +777,38 @@ class RandomForestClassifier(ForestClassifier):
         Note: the search for a split does not stop until at least one
         valid partition of the node samples is found, even if it requires to
         effectively inspect more than ``max_features`` features.
-        Note: this parameter is tree-specific.
 
     max_depth : integer or None, optional (default=None)
         The maximum depth of the tree. If None, then nodes are expanded until
         all leaves are pure or until all leaves contain less than
         min_samples_split samples.
         Ignored if ``max_leaf_nodes`` is not None.
-        Note: this parameter is tree-specific.
 
-    min_samples_split : integer, optional (default=2)
-        The minimum number of samples required to split an internal node.
-        Note: this parameter is tree-specific.
+    min_samples_split : int, float, optional (default=2)
+        The minimum number of samples required to split an internal node:
 
-    min_samples_leaf : integer, optional (default=1)
-        The minimum number of samples in newly created leaves.  A split is
-        discarded if after the split, one of the leaves would contain less then
-        ``min_samples_leaf`` samples.
-        Note: this parameter is tree-specific.
+        - If int, then consider `min_samples_split` as the minimum number.
+        - If float, then `min_samples_split` is a percentage and
+          `ceil(min_samples_split * n_samples)` are the minimum
+          number of samples for each split.
+
+    min_samples_leaf : int, float, optional (default=1)
+        The minimum number of samples required to be at a leaf node:
+
+        - If int, then consider `min_samples_leaf` as the minimum number.
+        - If float, then `min_samples_leaf` is a percentage and
+          `ceil(min_samples_leaf * n_samples)` are the minimum
+          number of samples for each node.
 
     min_weight_fraction_leaf : float, optional (default=0.)
         The minimum weighted fraction of the input samples required to be at a
         leaf node.
-        Note: this parameter is tree-specific.
 
     max_leaf_nodes : int or None, optional (default=None)
         Grow trees with ``max_leaf_nodes`` in best-first fashion.
         Best nodes are defined as relative reduction in impurity.
         If None then unlimited number of leaf nodes.
         If not None then ``max_depth`` will be ignored.
-        Note: this parameter is tree-specific.
 
     bootstrap : boolean, optional (default=True)
         Whether bootstrap samples are used when building trees.
@@ -834,7 +836,6 @@ class RandomForestClassifier(ForestClassifier):
         new forest.
 
     class_weight : dict, list of dicts, "balanced", "balanced_subsample" or None, optional
-
         Weights associated with classes in the form ``{class_label: weight}``.
         If not given, all classes are supposed to have weight one. For
         multi-output problems, a list of dicts can be provided in the same
@@ -844,8 +845,9 @@ class RandomForestClassifier(ForestClassifier):
         weights inversely proportional to class frequencies in the input data
         as ``n_samples / (n_classes * np.bincount(y))``
 
-        The "balanced_subsample" mode is the same as "balanced" except that weights are
-        computed based on the bootstrap sample for every tree grown.
+        The "balanced_subsample" mode is the same as "balanced" except that
+        weights are computed based on the bootstrap sample for every tree
+        grown.
 
         For multi-output, the weights of each column of y will be multiplied.
 
@@ -952,7 +954,6 @@ class RandomForestRegressor(ForestRegressor):
     criterion : string, optional (default="mse")
         The function to measure the quality of a split. The only supported
         criterion is "mse" for the mean squared error.
-        Note: this parameter is tree-specific.
 
     max_features : int, float, string or None, optional (default="auto")
         The number of features to consider when looking for the best split:
@@ -969,36 +970,38 @@ class RandomForestRegressor(ForestRegressor):
         Note: the search for a split does not stop until at least one
         valid partition of the node samples is found, even if it requires to
         effectively inspect more than ``max_features`` features.
-        Note: this parameter is tree-specific.
 
     max_depth : integer or None, optional (default=None)
         The maximum depth of the tree. If None, then nodes are expanded until
         all leaves are pure or until all leaves contain less than
         min_samples_split samples.
         Ignored if ``max_leaf_nodes`` is not None.
-        Note: this parameter is tree-specific.
 
-    min_samples_split : integer, optional (default=2)
-        The minimum number of samples required to split an internal node.
-        Note: this parameter is tree-specific.
+    min_samples_split : int, float, optional (default=2)
+        The minimum number of samples required to split an internal node:
 
-    min_samples_leaf : integer, optional (default=1)
-        The minimum number of samples in newly created leaves.  A split is
-        discarded if after the split, one of the leaves would contain less then
-        ``min_samples_leaf`` samples.
-        Note: this parameter is tree-specific.
+        - If int, then consider `min_samples_split` as the minimum number.
+        - If float, then `min_samples_split` is a percentage and
+          `ceil(min_samples_split * n_samples)` are the minimum
+          number of samples for each split.
+
+    min_samples_leaf : int, float, optional (default=1)
+        The minimum number of samples required to be at a leaf node:
+
+        - If int, then consider `min_samples_leaf` as the minimum number.
+        - If float, then `min_samples_leaf` is a percentage and
+          `ceil(min_samples_leaf * n_samples)` are the minimum
+          number of samples for each node.
 
     min_weight_fraction_leaf : float, optional (default=0.)
         The minimum weighted fraction of the input samples required to be at a
         leaf node.
-        Note: this parameter is tree-specific.
 
     max_leaf_nodes : int or None, optional (default=None)
         Grow trees with ``max_leaf_nodes`` in best-first fashion.
         Best nodes are defined as relative reduction in impurity.
         If None then unlimited number of leaf nodes.
         If not None then ``max_depth`` will be ignored.
-        Note: this parameter is tree-specific.
 
     bootstrap : boolean, optional (default=True)
         Whether bootstrap samples are used when building trees.
@@ -1110,7 +1113,6 @@ class ExtraTreesClassifier(ForestClassifier):
     criterion : string, optional (default="gini")
         The function to measure the quality of a split. Supported criteria are
         "gini" for the Gini impurity and "entropy" for the information gain.
-        Note: this parameter is tree-specific.
 
     max_features : int, float, string or None, optional (default="auto")
         The number of features to consider when looking for the best split:
@@ -1127,36 +1129,38 @@ class ExtraTreesClassifier(ForestClassifier):
         Note: the search for a split does not stop until at least one
         valid partition of the node samples is found, even if it requires to
         effectively inspect more than ``max_features`` features.
-        Note: this parameter is tree-specific.
 
     max_depth : integer or None, optional (default=None)
         The maximum depth of the tree. If None, then nodes are expanded until
         all leaves are pure or until all leaves contain less than
         min_samples_split samples.
         Ignored if ``max_leaf_nodes`` is not None.
-        Note: this parameter is tree-specific.
 
-    min_samples_split : integer, optional (default=2)
-        The minimum number of samples required to split an internal node.
-        Note: this parameter is tree-specific.
+    min_samples_split : int, float, optional (default=2)
+        The minimum number of samples required to split an internal node:
 
-    min_samples_leaf : integer, optional (default=1)
-        The minimum number of samples in newly created leaves.  A split is
-        discarded if after the split, one of the leaves would contain less then
-        ``min_samples_leaf`` samples.
-        Note: this parameter is tree-specific.
+        - If int, then consider `min_samples_split` as the minimum number.
+        - If float, then `min_samples_split` is a percentage and
+          `ceil(min_samples_split * n_samples)` are the minimum
+          number of samples for each split.
+
+    min_samples_leaf : int, float, optional (default=1)
+        The minimum number of samples required to be at a leaf node:
+
+        - If int, then consider `min_samples_leaf` as the minimum number.
+        - If float, then `min_samples_leaf` is a percentage and
+          `ceil(min_samples_leaf * n_samples)` are the minimum
+          number of samples for each node.
 
     min_weight_fraction_leaf : float, optional (default=0.)
         The minimum weighted fraction of the input samples required to be at a
         leaf node.
-        Note: this parameter is tree-specific.
 
     max_leaf_nodes : int or None, optional (default=None)
         Grow trees with ``max_leaf_nodes`` in best-first fashion.
         Best nodes are defined as relative reduction in impurity.
         If None then unlimited number of leaf nodes.
         If not None then ``max_depth`` will be ignored.
-        Note: this parameter is tree-specific.
 
     bootstrap : boolean, optional (default=False)
         Whether bootstrap samples are used when building trees.
@@ -1184,7 +1188,6 @@ class ExtraTreesClassifier(ForestClassifier):
         new forest.
 
     class_weight : dict, list of dicts, "balanced", "balanced_subsample" or None, optional
-
         Weights associated with classes in the form ``{class_label: weight}``.
         If not given, all classes are supposed to have weight one. For
         multi-output problems, a list of dicts can be provided in the same
@@ -1266,7 +1269,8 @@ def __init__(self,
             n_estimators=n_estimators,
             estimator_params=("criterion", "max_depth", "min_samples_split",
                               "min_samples_leaf", "min_weight_fraction_leaf",
-                              "max_features", "max_leaf_nodes", "random_state"),
+                              "max_features", "max_leaf_nodes",
+                              "random_state"),
             bootstrap=bootstrap,
             oob_score=oob_score,
             n_jobs=n_jobs,
@@ -1302,7 +1306,6 @@ class ExtraTreesRegressor(ForestRegressor):
     criterion : string, optional (default="mse")
         The function to measure the quality of a split. The only supported
         criterion is "mse" for the mean squared error.
-        Note: this parameter is tree-specific.
 
     max_features : int, float, string or None, optional (default="auto")
         The number of features to consider when looking for the best split:
@@ -1319,44 +1322,44 @@ class ExtraTreesRegressor(ForestRegressor):
         Note: the search for a split does not stop until at least one
         valid partition of the node samples is found, even if it requires to
         effectively inspect more than ``max_features`` features.
-        Note: this parameter is tree-specific.
 
     max_depth : integer or None, optional (default=None)
         The maximum depth of the tree. If None, then nodes are expanded until
         all leaves are pure or until all leaves contain less than
         min_samples_split samples.
         Ignored if ``max_leaf_nodes`` is not None.
-        Note: this parameter is tree-specific.
 
-    min_samples_split : integer, optional (default=2)
-        The minimum number of samples required to split an internal node.
-        Note: this parameter is tree-specific.
+    min_samples_split : int, float, optional (default=2)
+        The minimum number of samples required to split an internal node:
 
-    min_samples_leaf : integer, optional (default=1)
-        The minimum number of samples in newly created leaves.  A split is
-        discarded if after the split, one of the leaves would contain less then
-        ``min_samples_leaf`` samples.
-        Note: this parameter is tree-specific.
+        - If int, then consider `min_samples_split` as the minimum number.
+        - If float, then `min_samples_split` is a percentage and
+          `ceil(min_samples_split * n_samples)` are the minimum
+          number of samples for each split.
+
+    min_samples_leaf : int, float, optional (default=1)
+        The minimum number of samples required to be at a leaf node:
+
+        - If int, then consider `min_samples_leaf` as the minimum number.
+        - If float, then `min_samples_leaf` is a percentage and
+          `ceil(min_samples_leaf * n_samples)` are the minimum
+          number of samples for each node.
 
     min_weight_fraction_leaf : float, optional (default=0.)
         The minimum weighted fraction of the input samples required to be at a
         leaf node.
-        Note: this parameter is tree-specific.
 
     max_leaf_nodes : int or None, optional (default=None)
         Grow trees with ``max_leaf_nodes`` in best-first fashion.
         Best nodes are defined as relative reduction in impurity.
         If None then unlimited number of leaf nodes.
         If not None then ``max_depth`` will be ignored.
-        Note: this parameter is tree-specific.
 
     bootstrap : boolean, optional (default=False)
         Whether bootstrap samples are used when building trees.
-        Note: this parameter is tree-specific.
 
     oob_score : bool
-        Whether to use out-of-bag samples to estimate
-        the generalization error.
+        Whether to use out-of-bag samples to estimate the generalization error.
 
     n_jobs : integer, optional (default=1)
         The number of jobs to run in parallel for both `fit` and `predict`.
@@ -1471,13 +1474,21 @@ class RandomTreesEmbedding(BaseForest):
         min_samples_split samples.
         Ignored if ``max_leaf_nodes`` is not None.
 
-    min_samples_split : integer, optional (default=2)
-        The minimum number of samples required to split an internal node.
+    min_samples_split : int, float, optional (default=2)
+        The minimum number of samples required to split an internal node:
+
+        - If int, then consider `min_samples_split` as the minimum number.
+        - If float, then `min_samples_split` is a percentage and
+          `ceil(min_samples_split * n_samples)` is the minimum
+          number of samples for each split.
+
+    min_samples_leaf : int, float, optional (default=1)
+        The minimum number of samples required to be at a leaf node:
 
-    min_samples_leaf : integer, optional (default=1)
-        The minimum number of samples in newly created leaves.  A split is
-        discarded if after the split, one of the leaves would contain less then
-        ``min_samples_leaf`` samples.
+        - If int, then consider `min_samples_leaf` as the minimum number.
+        - If float, then `min_samples_leaf` is a percentage and
+          `ceil(min_samples_leaf * n_samples)` is the minimum
+          number of samples for each node.
 
     min_weight_fraction_leaf : float, optional (default=0.)
         The minimum weighted fraction of the input samples required to be at a