diff --git a/paddle/fluid/operators/shape_op.cc b/paddle/fluid/operators/shape_op.cc
index c75fce7959d1af51afd52af23fe657d10a2f3988..b44d5f898013a5d27467bd80118c29a886d5e8b3 100644
--- a/paddle/fluid/operators/shape_op.cc
+++ b/paddle/fluid/operators/shape_op.cc
@@ -36,10 +36,13 @@ class ShapeOpMaker : public framework::OpProtoAndCheckerMaker {
  public:
   void Make() override {
     AddInput("Input", "(Tensor), The input tensor.");
-    AddOutput("Out", "(Tensor), The shape of input tensor.");
+    AddOutput("Out",
+              "(Tensor), The shape of input tensor, the data type of the shape"
+              " is int64_t, will be on the same device with the input Tensor.");
     AddComment(R"DOC(
-Shape Operator. 
-Get the shape of input tensor.
+Shape Operator
+
+Get the shape of input tensor. Only support CPU input Tensor now.
 )DOC");
   }
 };
diff --git a/python/paddle/fluid/layers/nn.py b/python/paddle/fluid/layers/nn.py
index a3b2d2b777fdae6a24c77636f823576b3f5ca718..40e72aa4884015e5724fd057ea85c58e23a4bd4b 100644
--- a/python/paddle/fluid/layers/nn.py
+++ b/python/paddle/fluid/layers/nn.py
@@ -4650,7 +4650,7 @@ def image_resize_short(input, out_short_len, resample='BILINEAR'):
 
     Returns:
         Variable: The output is a 4-D tensor of the shape
-                        (num_batches, channls, out_h, out_w).
+        (num_batches, channls, out_h, out_w).
     """
     in_shape = input.shape
     if len(in_shape) != 4: