diff --git a/doc/api/v2/config/layer.rst b/doc/api/v2/config/layer.rst
index 2a02baf17ba0d1119a8d222024616ef8ae33f8d5..4e3589ebc47039a4bc4e04b130c0aa536dda73a1 100644
--- a/doc/api/v2/config/layer.rst
+++ b/doc/api/v2/config/layer.rst
@@ -419,6 +419,11 @@ hsigmoid
 ..  autoclass:: paddle.v2.layer.hsigmoid
     :noindex:
 
+smooth_l1
+---------
+..  automodule:: paddle.v2.layer.smooth_l1
+    :noindex:
+
 Check Layer 
 ============
 
diff --git a/python/paddle/trainer_config_helpers/layers.py b/python/paddle/trainer_config_helpers/layers.py
index 56fca13d37242a091f09b35ca2e139071c644cad..1796e48f09ae50642897af7e783f020580772ac4 100755
--- a/python/paddle/trainer_config_helpers/layers.py
+++ b/python/paddle/trainer_config_helpers/layers.py
@@ -116,7 +116,7 @@ __all__ = [
     'spp_layer',
     'pad_layer',
     'eos_layer',
-    'smooth_l1_cost',
+    'smooth_l1',
     'layer_support',
 ]
 
@@ -5283,7 +5283,7 @@ def multi_binary_label_cross_entropy(input,
 
 @wrap_name_default()
 @layer_support()
-def smooth_l1_cost(input, label, name=None, layer_attr=None):
+def smooth_l1(input, label, name=None, layer_attr=None):
     """
     This is a L1 loss but more smooth. It requires that the
     size of input and label are equal. The formula is as follows,
@@ -5307,8 +5307,8 @@ def smooth_l1_cost(input, label, name=None, layer_attr=None):
 
     .. code-block:: python
 
-       cost = smooth_l1_cost(input=input_layer,
-                             label=label_layer)
+       cost = smooth_l1(input=input_layer,
+                        label=label_layer)
 
     :param input: The input layer.
     :type input: LayerOutput