diff --git a/sklearn/preprocessing/label.py b/sklearn/preprocessing/label.py
index cb9a2f527b40f..ce62373f780f9 100644
--- a/sklearn/preprocessing/label.py
+++ b/sklearn/preprocessing/label.py
@@ -105,9 +105,12 @@ def fit(self, y):
         -------
         self : returns an instance of self.
         """
-        y = column_or_1d(y, warn=True)
+        # y = column_or_1d(y, warn=True)
         _check_numpy_unicode_bug(y)
-        self.classes_ = np.unique(y)
+        if 'multioutput' in type_of_target(y):
+            self.classes_ = [np.unique(y[:, k]) for k in range(y.shape[1])]
+        else:
+            self.classes_ = np.unique(y)
         return self
 
     def fit_transform(self, y):
@@ -122,9 +125,17 @@ def fit_transform(self, y):
         -------
         y : array-like of shape [n_samples]
         """
-        y = column_or_1d(y, warn=True)
+        # y = column_or_1d(y, warn=True)
         _check_numpy_unicode_bug(y)
-        self.classes_, y = np.unique(y, return_inverse=True)
+        if 'multioutput' in type_of_target(y):
+            self.classes_ = []
+            y_enc = np.empty(y.shape, dtype=np.int)
+            for k in range(y_enc.shape[1]):
+                classes, y_enc[:, k] = np.unique(y[:, k], return_inverse=True)
+                self.classes_.append(classes)
+            y = y_enc
+        else:
+            self.classes_, y = np.unique(y, return_inverse=True)
         return y
 
     def transform(self, y):
@@ -141,12 +152,18 @@ def transform(self, y):
         """
         self._check_fitted()
 
-        classes = np.unique(y)
-        _check_numpy_unicode_bug(classes)
-        if len(np.intersect1d(classes, self.classes_)) < len(classes):
-            diff = np.setdiff1d(classes, self.classes_)
-            raise ValueError("y contains new labels: %s" % str(diff))
-        return np.searchsorted(self.classes_, y)
+        if 'multioutput' in type_of_target(y):
+            y_enc = np.empty(y.shape, dtype=np.int)
+            for k in range(y.shape[1]):
+                y_enc[:, k] = np.searchsorted(self.classes_[k], y[:, k])
+            return y_enc
+        else:
+            classes = np.unique(y)
+            _check_numpy_unicode_bug(classes)
+            if len(np.intersect1d(classes, self.classes_)) < len(classes):
+                diff = np.setdiff1d(classes, self.classes_)
+                raise ValueError("y contains new labels: %s" % str(diff))
+            return np.searchsorted(self.classes_, y)
 
     def inverse_transform(self, y):
         """Transform labels back to original encoding.
@@ -668,9 +685,6 @@ class MultiLabelBinarizer(BaseEstimator, TransformerMixin):
     classes : array-like of shape [n_classes] (optional)
         Indicates an ordering for the class labels
 
-    sparse_output : boolean (default: False),
-        Set to true if output binary array is desired in CSR sparse format
-
     Attributes
     ----------
     classes_ : array of labels
diff --git a/sklearn/preprocessing/tests/test_label.py b/sklearn/preprocessing/tests/test_label.py
index f523870ffd2fd..bdaa62fa53a2e 100644
--- a/sklearn/preprocessing/tests/test_label.py
+++ b/sklearn/preprocessing/tests/test_label.py
@@ -229,6 +229,79 @@ def test_label_encoder():
     assert_raises(ValueError, le.transform, [0, 6])
 
 
+def test_label_encoder_multioutput():
+    """Test LabelEncoder's with multioutput target data"""
+    le = LabelEncoder()
+    y = np.array([[1,2,3,4,0],
+                  [1,0,0,0,0],
+                  [1,1,1,1,1],
+                  [1,-1,2,0,1],
+                  [1,0,1,1,0],
+                  [1,2,3,4,5]])
+    classes = [[1],[-1,0,1,2],[0,1,2,3],[0,1,4],[0,1,5]]
+    y_enc = np.array([[0,3,3,2,0],
+                      [0,1,0,0,0],
+                      [0,2,1,1,1],
+                      [0,0,2,0,1],
+                      [0,1,1,1,0],
+                      [0,3,3,2,2]])
+
+    # Test fit
+    le.fit(y)
+    for i in range(y.shape[1]):
+        assert_array_equal(le.classes_[i], classes[i])
+
+    # Test transform
+    assert_array_equal(le.transform(y), y_enc)
+
+    # Test inverse transform
+    assert_array_equal(le.inverse_transform(y_enc), y)
+
+    # Test fit transform
+    le = LabelEncoder()
+    ret = le.fit_transform(y)
+    assert_array_equal(ret, y_enc)
+
+    # Test unseeen label error
+    assert_raises(ValueError, le.transform, [0, 6])
+
+def test_label_encoder_sparse_multioutput():
+    """Test LabelEncoder's with multioutput target data in sparse formats"""
+        """Test LabelEncoder's with multioutput target data"""
+    le = LabelEncoder()
+    y = sp.csc_matrix(np.array([[0,2,3,4,0],
+                                [0,0,0,0,0],
+                                [0,1,1,1,0],
+                                [0,-1,2,0,1],
+                                [0,0,1,1,0],
+                                [0,2,3,4,5]]))
+    classes = [[0],[-1,0,1,2],[0,1,2,3],[0,1,4],[0,1,5]]
+    y = sp.csc_matrix(np.array([[0,3,3,2,0],
+                                [0,2,0,0,0],
+                                [0,2,1,1,0],
+                                [0,0,2,0,1],
+                                [0,1,1,1,0],
+                                [0,3,3,2,2]]))
+
+    # Test fit
+    le.fit(y)
+    for i in range(y.shape[1]):
+        assert_array_equal(le.classes_[i], classes[i])
+
+    # Test transform
+    assert_array_equal(le.transform(y).toarray(), y_enc)
+
+    # Test inverse transform
+    assert_array_equal(le.inverse_transform(y_enc).toarray(), y)
+
+    # Test fit transform
+    le = LabelEncoder()
+    ret = le.fit_transform(y)
+    assert_array_equal(ret.toarray(), y_enc)
+
+    # Test unseeen label error
+    assert_raises(ValueError, le.transform, [0, 6])
+
 def test_label_encoder_fit_transform():
     """Test fit_transform"""
     le = LabelEncoder()