scikit-learn · glemaitre · Sep 20, 2019 · Sep 3, 2019 · Sep 3, 2019 · Sep 3, 2019
diff --git a/doc/whats_new/v0.22.rst b/doc/whats_new/v0.22.rst
@@ -551,6 +551,10 @@ Changelog
   and sparse matrix.
   :pr:`14538` by :user:`Jérémie du Boisberranger <jeremiedbb>`.
 
+- |Fix| :func:`utils.check_array` is now raising an error instead of casting
+  NaN to integer.
+  :pr:`14872` by `Roman Yurchak`_.
+
 :mod:`sklearn.metrics`
 ..................................
 

diff --git a/sklearn/utils/tests/test_validation.py b/sklearn/utils/tests/test_validation.py
@@ -202,6 +202,26 @@ def test_check_array_force_all_finite_object():
         check_array(X, dtype=None, force_all_finite=True)
 
 
+@pytest.mark.parametrize(
+    "X, err_msg",
+    [(np.array([[1, np.nan]]),
+      "Input contains NaN, infinity or a value too large for.*int"),
+     (np.array([[1, np.nan]]),
+      "Input contains NaN, infinity or a value too large for.*int"),
+     (np.array([[1, np.inf]]),
+      "Input contains NaN, infinity or a value too large for.*int"),
+     (np.array([[1, np.nan]], dtype=np.object),
+      "cannot convert float NaN to integer")]
+)
+@pytest.mark.parametrize("force_all_finite", [True, False])
+def test_check_array_force_all_finite_object_unsafe_casting(
+        X, err_msg, force_all_finite):
+    # casting a float array containing NaN or inf to int dtype should
+    # raise an error irrespective of the force_all_finite parameter.
+    with pytest.raises(ValueError, match=err_msg):
+        check_array(X, dtype=np.int, force_all_finite=force_all_finite)
+
+
 @ignore_warnings
 def test_check_array():
     # accept_sparse == False

diff --git a/sklearn/utils/validation.py b/sklearn/utils/validation.py
@@ -32,7 +32,7 @@
 warnings.simplefilter('ignore', NonBLASDotWarning)
 
 
-def _assert_all_finite(X, allow_nan=False):
+def _assert_all_finite(X, allow_nan=False, msg_dtype=None):
     """Like assert_all_finite, but only for ndarray."""
     # validation is also imported in extmath
     from .extmath import _safe_accumulator_op
@@ -52,7 +52,11 @@ def _assert_all_finite(X, allow_nan=False):
         if (allow_nan and np.isinf(X).any() or
                 not allow_nan and not np.isfinite(X).all()):
             type_err = 'infinity' if allow_nan else 'NaN, infinity'
-            raise ValueError(msg_err.format(type_err, X.dtype))
+            raise ValueError(
+                    msg_err.format
+                    (type_err,
+                     msg_dtype if msg_dtype is not None else X.dtype)
+            )
     # for object dtype data, we only check for NaNs (GH-13254)
     elif X.dtype == np.dtype('object') and not allow_nan:
         if _object_dtype_isnan(X).any():
@@ -494,7 +498,17 @@ def check_array(array, accept_sparse=False, accept_large_sparse=True,
         with warnings.catch_warnings():
             try:
                 warnings.simplefilter('error', ComplexWarning)
-                array = np.asarray(array, dtype=dtype, order=order)
+                if dtype is not None and np.dtype(dtype).kind in 'iu':
+                    # Conversion float -> int should not contain NaN or
+                    # inf (numpy#14412). We cannot use casting='safe' because
+                    # then conversion float -> int would be disallowed.
+                    array = np.asarray(array, order=order)
+                    if array.dtype.kind == 'f':
+                        _assert_all_finite(array, allow_nan=False,
+                                           msg_dtype=dtype)
+                    array = array.astype(dtype, casting="unsafe", copy=False)
+                else:
+                    array = np.asarray(array, order=order, dtype=dtype)
             except ComplexWarning:
                 raise ValueError("Complex data not supported\n"
                                  "{}\n".format(array))