diff --git a/paddle/gserver/layers/FeatureMapExpandLayer.cpp b/paddle/gserver/layers/FeatureMapExpandLayer.cpp
index d023074c52167554358d0d4df7ec40cfba9da2a6..4b685812aac24782e4fe47fa85c7e91bd3494087 100644
--- a/paddle/gserver/layers/FeatureMapExpandLayer.cpp
+++ b/paddle/gserver/layers/FeatureMapExpandLayer.cpp
@@ -95,6 +95,9 @@ void FeatureMapExpandLayer::forward(PassType passType) {
 
 void FeatureMapExpandLayer::backward(const UpdateCallback& callback) {
   MatrixPtr inGrad = getInputGrad(0);
+  if (NULL == inGrad) {
+    return;
+  }
   MatrixPtr outGrad = getOutputGrad();
   size_t batchSize = getInput(0).getBatchSize();
   int imgSize = inGrad->getWidth();
diff --git a/paddle/gserver/layers/GruStepLayer.cpp b/paddle/gserver/layers/GruStepLayer.cpp
index ce692c490881c9330b8113185bbec98b8dfd63e0..4a1006aa941f396c233a0cecfc38228f1f9fafe1 100644
--- a/paddle/gserver/layers/GruStepLayer.cpp
+++ b/paddle/gserver/layers/GruStepLayer.cpp
@@ -68,8 +68,8 @@ bool GruStepLayer::init(const LayerMap& layerMap,
   if (!Layer::init(layerMap, parameterMap)) return false;
   CHECK_EQ(2U, inputLayers_.size());
 
-  CHECK_EQ(getSize() * getSize() * 3, parameters_[1]->getSize());
-  weight_.reset(new Weight(getSize(), getSize() * 3, parameters_[1]));
+  CHECK_EQ(getSize() * getSize() * 3, parameters_[0]->getSize());
+  weight_.reset(new Weight(getSize(), getSize() * 3, parameters_[0]));
 
   if (biasParameter_.get() != NULL) {
     CHECK_EQ(getSize() * 3, biasParameter_->getSize());
diff --git a/paddle/gserver/layers/RotateLayer.cpp b/paddle/gserver/layers/RotateLayer.cpp
index 1cdd8366d5ddff424739e09c01385ce30e383caa..d928f1d93de0ee2de90c6f2e9c778db9e29524bf 100644
--- a/paddle/gserver/layers/RotateLayer.cpp
+++ b/paddle/gserver/layers/RotateLayer.cpp
@@ -25,6 +25,8 @@ bool RotateLayer::init(const LayerMap& layerMap,
   CHECK_EQ(inputLayers_.size(), 1UL);
   height_ = config_.height();
   width_ = config_.width();
+  CHECK_GT(height_, 0);
+  CHECK_GT(width_, 0);
   return true;
 }
 
diff --git a/python/paddle/trainer/config_parser.py b/python/paddle/trainer/config_parser.py
index 62efe5fca0cacae9e4e5258622b9a688a28e831e..b02af991dc577e070dac813cfd18d35ab6dfc3e0 100644
--- a/python/paddle/trainer/config_parser.py
+++ b/python/paddle/trainer/config_parser.py
@@ -1969,13 +1969,13 @@ class ResizeLayer(LayerBase):
 
 @config_layer('rotate')
 class RotateLayer(LayerBase):
-    def __init__(self, name, inputs, height, device=None):
+    def __init__(self, name, inputs, height, width, device=None):
         super(RotateLayer, self).__init__(
             name, 'rotate', 0, inputs=inputs, device=device)
         config_assert(
             len(self.inputs) == 1,
             'RotateLayer must have one and only one input')
-        self.config.height = height
+        self.set_layer_height_width(height, width)
         self.set_layer_size(self.get_input_layer(0).size)
 
 
@@ -3007,7 +3007,7 @@ class GruStepLayer(LayerBase):
         config_assert(input_layer1.size == size,
                       'input_layer1.size != layer.size')
         self.config.active_gate_type = active_gate_type
-        self.create_input_parameter(1, size * size * 3, [size, size * 3])
+        self.create_input_parameter(0, size * size * 3, [size, size * 3])
         self.create_bias_parameter(bias, size * 3)
 
 
diff --git a/python/paddle/trainer_config_helpers/layers.py b/python/paddle/trainer_config_helpers/layers.py
index 19f7ec3186596f2a9b028e71d78a9b2870cea659..2636f6cc1a1653ad6ff676912f3e7206e03bc31c 100755
--- a/python/paddle/trainer_config_helpers/layers.py
+++ b/python/paddle/trainer_config_helpers/layers.py
@@ -1677,22 +1677,23 @@ def trans_layer(input, name=None, layer_attr=None):
 
 @wrap_name_default()
 @layer_support()
-def rotate_layer(input, height, name=None, layer_attr=None):
+def rotate_layer(input, height, width, name=None, layer_attr=None):
     """
-    A layer for rotation (clock-wise), usually used when the input sample
-    is some image or map.
+    A layer for rotating 90 degrees (clock-wise), usually used when the input sample
+    is some image or feature map.
 
     .. math::
-       y(j,i) = x(M-i-1,j)
+       y(j,i,:) = x(M-i-1,j,:)
 
-    where :math:`x` is (M x N) input, and :math:`y` is (N x M) output.
+    where :math:`x` is (M x N x C) input, and :math:`y` is (N x M x C) output.
 
     The example usage is:
 
     .. code-block:: python
 
        rot = rotate_layer(input=layer,
-                          height=100)
+                          height=100,
+                          width=100)
 
     :param input: Input layer.
     :type input: LayerOutput
@@ -1709,6 +1710,7 @@ def rotate_layer(input, height, name=None, layer_attr=None):
     l = Layer(
         name=name,
         height=height,
+        width=width,
         type=LayerType.ROTATE_LAYER,
         inputs=[input.name],
         **ExtraLayerAttribute.to_kwargs(layer_attr))
diff --git a/python/paddle/trainer_config_helpers/tests/layers_test_config.py b/python/paddle/trainer_config_helpers/tests/layers_test_config.py
index 1964e6f113e6423018325ad87bd82b3d333b8868..e6cd35ee761d1acd0b5c1943554c7ea1de6a13f5 100644
--- a/python/paddle/trainer_config_helpers/tests/layers_test_config.py
+++ b/python/paddle/trainer_config_helpers/tests/layers_test_config.py
@@ -39,7 +39,7 @@ z1 = mixed_layer(
 assert z1.size > 0
 
 y2 = fc_layer(input=y, size=15)
-z2 = rotate_layer(input=y2, height=5)
+z2 = rotate_layer(input=y2, height=5, width=3)
 
 cos1 = cos_sim(a=x1, b=y1)
 cos3 = cos_sim(a=x1, b=y2, size=3)