DOC: X_argsorted

Author: glouppe

sklearn/ensemble/weight_boosting.py

@@ -167,6 +167,14 @@ def _boost(self, iboost, X, y, sample_weight, X_argsorted=None):
         sample_weight : array-like of shape = [n_samples]
             The current sample weights.
 
+        X_argsorted : array-like, shape = [n_samples, n_features] (optional)
+            Each column of ``X_argsorted`` holds the row indices of ``X``
+            sorted according to the value of the corresponding feature
+            in ascending order.
+            The argument is supported to enable multiple decision trees
+            to share the data structure and to avoid re-computation in
+            tree ensembles. For maximum efficiency use dtype np.int32.
+
         Returns
         -------
         sample_weight : array-like of shape = [n_samples] or None
@@ -397,6 +405,14 @@ def _boost(self, iboost, X, y, sample_weight, X_argsorted=None):
         sample_weight : array-like of shape = [n_samples]
             The current sample weights.
 
+        X_argsorted : array-like, shape = [n_samples, n_features] (optional)
+            Each column of ``X_argsorted`` holds the row indices of ``X``
+            sorted according to the value of the corresponding feature
+            in ascending order.
+            The argument is supported to enable multiple decision trees
+            to share the data structure and to avoid re-computation in
+            tree ensembles. For maximum efficiency use dtype np.int32.
+
         Returns
         -------
         sample_weight : array-like of shape = [n_samples] or None
@@ -913,6 +929,14 @@ def _boost(self, iboost, X, y, sample_weight, X_argsorted=None):
         sample_weight : array-like of shape = [n_samples]
             The current sample weights.
 
+        X_argsorted : array-like, shape = [n_samples, n_features] (optional)
+            Each column of ``X_argsorted`` holds the row indices of ``X``
+            sorted according to the value of the corresponding feature
+            in ascending order.
+            The argument is supported to enable multiple decision trees
+            to share the data structure and to avoid re-computation in
+            tree ensembles. For maximum efficiency use dtype np.int32.
+
         Returns
         -------
         sample_weight : array-like of shape = [n_samples] or None
@@ -942,6 +966,7 @@ def _boost(self, iboost, X, y, sample_weight, X_argsorted=None):
 
         # Fit on the bootstrapped sample and obtain a prediction
         # for all samples in the training set
+        # X_argsorted is not used since bootstrap copies are used.
         estimator.fit(X[bootstrap_idx], y[bootstrap_idx])
         y_predict = estimator.predict(X)