Merge pull request scikit-learn#4588 from amueller/random_forest_predict_check

[MRG+2] don't to input validation in each tree for RandomForest.predict

Author: GaelVaroquaux

sklearn/ensemble/forest.py

@@ -365,7 +365,8 @@ def _set_oob_score(self, X, y):
             mask = np.ones(n_samples, dtype=np.bool)
             mask[estimator.indices_] = False
             mask_indices = sample_indices[mask]
-            p_estimator = estimator.predict_proba(X[mask_indices, :])
+            p_estimator = estimator.predict_proba(X[mask_indices, :],
+                                                  check_input=False)
 
             if self.n_outputs_ == 1:
                 p_estimator = [p_estimator]
@@ -508,7 +509,7 @@ class in a leaf.
         # Parallel loop
         all_proba = Parallel(n_jobs=n_jobs, verbose=self.verbose,
                              backend="threading")(
-            delayed(_parallel_helper)(e, 'predict_proba', X)
+            delayed(_parallel_helper)(e, 'predict_proba', X, check_input=False)
             for e in self.estimators_)
 
         # Reduce
@@ -614,6 +615,10 @@ def predict(self, X):
 
         # Check data
         X = check_array(X, dtype=DTYPE, accept_sparse="csr")
+        if issparse(X) and (X.indices.dtype != np.intc or
+                            X.indptr.dtype != np.intc):
+            raise ValueError("No support for np.int64 index based "
+                             "sparse matrices")
 
         # Assign chunk of trees to jobs
         n_jobs, n_trees, starts = _partition_estimators(self.n_estimators,
@@ -622,7 +627,7 @@ def predict(self, X):
         # Parallel loop
         all_y_hat = Parallel(n_jobs=n_jobs, verbose=self.verbose,
                              backend="threading")(
-            delayed(_parallel_helper)(e, 'predict', X)
+            delayed(_parallel_helper)(e, 'predict', X, check_input=False)
             for e in self.estimators_)
 
         # Reduce
@@ -642,7 +647,7 @@ def _set_oob_score(self, X, y):
             mask = np.ones(n_samples, dtype=np.bool)
             mask[estimator.indices_] = False
             mask_indices = sample_indices[mask]
-            p_estimator = estimator.predict(X[mask_indices, :])
+            p_estimator = estimator.predict(X[mask_indices, :], check_input=False)
 
             if self.n_outputs_ == 1:
                 p_estimator = p_estimator[:, np.newaxis]

sklearn/tree/tree.py

@@ -309,7 +309,7 @@ def fit(self, X, y, sample_weight=None, check_input=True):
 
         return self
 
-    def predict(self, X):
+    def predict(self, X, check_input=True):
         """Predict class or regression value for X.
 
         For a classification model, the predicted class for each sample in X is
@@ -323,16 +323,21 @@ def predict(self, X):
             ``dtype=np.float32`` and if a sparse matrix is provided
             to a sparse ``csr_matrix``.
 
+        check_input : boolean, (default=True)
+            Allow to bypass several input checking.
+            Don't use this parameter unless you know what you do.
+
         Returns
         -------
         y : array of shape = [n_samples] or [n_samples, n_outputs]
             The predicted classes, or the predict values.
         """
-        X = check_array(X, dtype=DTYPE, accept_sparse="csr")
-        if issparse(X) and (X.indices.dtype != np.intc or
-                            X.indptr.dtype != np.intc):
-            raise ValueError("No support for np.int64 index based "
-                             "sparse matrices")
+        if check_input:
+            X = check_array(X, dtype=DTYPE, accept_sparse="csr")
+            if issparse(X) and (X.indices.dtype != np.intc or
+                                X.indptr.dtype != np.intc):
+                raise ValueError("No support for np.int64 index based "
+                                 "sparse matrices")
 
         n_samples, n_features = X.shape
 
@@ -568,12 +573,16 @@ def __init__(self,
             class_weight=class_weight,
             random_state=random_state)
 
-    def predict_proba(self, X):
+    def predict_proba(self, X, check_input=True):
         """Predict class probabilities of the input samples X.
 
         The predicted class probability is the fraction of samples of the same
         class in a leaf.
 
+        check_input : boolean, (default=True)
+            Allow to bypass several input checking.
+            Don't use this parameter unless you know what you do.
+
         Parameters
         ----------
         X : array-like or sparse matrix of shape = [n_samples, n_features]
@@ -589,11 +598,12 @@ class in a leaf.
             classes corresponds to that in the attribute `classes_`.
         """
         check_is_fitted(self, 'n_outputs_')
-        X = check_array(X, dtype=DTYPE, accept_sparse="csr")
-        if issparse(X) and (X.indices.dtype != np.intc or
-                            X.indptr.dtype != np.intc):
-            raise ValueError("No support for np.int64 index based "
-                             "sparse matrices")
+        if check_input:
+            X = check_array(X, dtype=DTYPE, accept_sparse="csr")
+            if issparse(X) and (X.indices.dtype != np.intc or
+                                X.indptr.dtype != np.intc):
+                raise ValueError("No support for np.int64 index based "
+                                 "sparse matrices")
 
         n_samples, n_features = X.shape