scikit-learn · qinhanmin2014 · Jul 10, 2018 · Apr 29, 2018 · Apr 29, 2018 · May 1, 2018
diff --git a/doc/whats_new/v0.20.rst b/doc/whats_new/v0.20.rst
@@ -830,6 +830,12 @@ Misc
   indices should be rejected.
   :issue:`11327` by :user:`Karan Dhingra <kdhingra307>` and `Joel Nothman`_.
 
+Preprocessing
+
+- In :class:`preprocessing.FunctionTransformer`, the default of ``validate``
+  will be from ``True`` to ``False`` in 0.22.
+  :issue:`10655` by :user:`Guillaume Lemaitre <glemaitre>`.
+
 Changes to estimator checks
 ---------------------------
 

diff --git a/sklearn/preprocessing/_function_transformer.py b/sklearn/preprocessing/_function_transformer.py
@@ -42,10 +42,16 @@ class FunctionTransformer(BaseEstimator, TransformerMixin):
 
     validate : bool, optional default=True
         Indicate that the input X array should be checked before calling
-        func. If validate is false, there will be no input validation.
-        If it is true, then X will be converted to a 2-dimensional NumPy
-        array or sparse matrix. If this conversion is not possible or X
-        contains NaN or infinity, an exception is raised.
+        ``func``. The possibilities are:
+
+        - If False, there is no input validation.
+        - If True, then X will be converted to a 2-dimensional NumPy array or
+          sparse matrix. If the conversion is not possible an exception is
+          raised.
+
+        .. deprecated:: 0.20
+           ``validate=True`` as default will be replaced by
+           ``validate=False`` in 0.22.
 
     accept_sparse : boolean, optional
         Indicate that func accepts a sparse matrix as input. If validate is
@@ -72,7 +78,7 @@ class FunctionTransformer(BaseEstimator, TransformerMixin):
         Dictionary of additional keyword arguments to pass to inverse_func.
 
     """
-    def __init__(self, func=None, inverse_func=None, validate=True,
+    def __init__(self, func=None, inverse_func=None, validate=None,
                  accept_sparse=False, pass_y='deprecated', check_inverse=True,
                  kw_args=None, inv_kw_args=None):
         self.func = func
@@ -84,6 +90,19 @@ def __init__(self, func=None, inverse_func=None, validate=True,
         self.kw_args = kw_args
         self.inv_kw_args = inv_kw_args
 
+    def _check_input(self, X):
+        # FIXME: Future warning to be removed in 0.22
+        if self.validate is None:
+            self._validate = True
+            warnings.warn("The default validate=True will be replaced by "
+                          "validate=False in 0.22.", FutureWarning)
+        else:
+            self._validate = self.validate
+
+        if self._validate:
+            return check_array(X, accept_sparse=self.accept_sparse)
+        return X
+
     def _check_inverse_transform(self, X):
         """Check that func and inverse_func are the inverse."""
         idx_selected = slice(None, None, max(1, X.shape[0] // 100))
@@ -111,8 +130,7 @@ def fit(self, X, y=None):
         -------
         self
         """
-        if self.validate:
-            X = check_array(X, self.accept_sparse)
+        X = self._check_input(X)
         if (self.check_inverse and not (self.func is None or
                                         self.inverse_func is None)):
             self._check_inverse_transform(X)
@@ -165,8 +183,7 @@ def inverse_transform(self, X, y='deprecated'):
                                kw_args=self.inv_kw_args)
 
     def _transform(self, X, y=None, func=None, kw_args=None):
-        if self.validate:
-            X = check_array(X, self.accept_sparse)
+        X = self._check_input(X)
 
         if func is None:
             func = _identity

diff --git a/sklearn/preprocessing/tests/test_function_transformer.py b/sklearn/preprocessing/tests/test_function_transformer.py
@@ -1,3 +1,4 @@
+import pytest
 import numpy as np
 from scipy import sparse
 
@@ -145,7 +146,8 @@ def test_check_inverse():
         trans = FunctionTransformer(func=np.sqrt,
                                     inverse_func=np.around,
                                     accept_sparse=accept_sparse,
-                                    check_inverse=True)
+                                    check_inverse=True,
+                                    validate=True)
         assert_warns_message(UserWarning,
                              "The provided functions are not strictly"
                              " inverse of each other. If you are sure you"
@@ -156,15 +158,38 @@ def test_check_inverse():
         trans = FunctionTransformer(func=np.expm1,
                                     inverse_func=np.log1p,
                                     accept_sparse=accept_sparse,
-                                    check_inverse=True)
+                                    check_inverse=True,
+                                    validate=True)
         Xt = assert_no_warnings(trans.fit_transform, X)
         assert_allclose_dense_sparse(X, trans.inverse_transform(Xt))
 
     # check that we don't check inverse when one of the func or inverse is not
     # provided.
     trans = FunctionTransformer(func=np.expm1, inverse_func=None,
-                                check_inverse=True)
+                                check_inverse=True, validate=True)
     assert_no_warnings(trans.fit, X_dense)
     trans = FunctionTransformer(func=None, inverse_func=np.expm1,
-                                check_inverse=True)
+                                check_inverse=True, validate=True)
     assert_no_warnings(trans.fit, X_dense)
+
+
+@pytest.mark.parametrize("validate, expected_warning",
+                         [(None, FutureWarning),
+                          (True, None),
+                          (False, None)])
+def test_function_transformer_future_warning(validate, expected_warning):
+    # FIXME: to be removed in 0.22
+    X = np.random.randn(100, 10)
+    transformer = FunctionTransformer(validate=validate)
+    with pytest.warns(expected_warning) as results:
+        transformer.fit_transform(X)
+    if expected_warning is None:
+        assert len(results) == 0
+
+
+def test_function_transformer_frame():
+    pd = pytest.importorskip('pandas')
+    X_df = pd.DataFrame(np.random.randn(100, 10))
+    transformer = FunctionTransformer(validate=False)
+    X_df_trans = transformer.fit_transform(X_df)
+    assert hasattr(X_df_trans, 'loc')