diff --git a/paddle/fluid/API.spec b/paddle/fluid/API.spec
index 13adab474b28bb27c88304aa9a1539953c6fb0f8..03c7f32a1261a184e6bdf4689aa411aa99ea8e68 100644
--- a/paddle/fluid/API.spec
+++ b/paddle/fluid/API.spec
@@ -227,7 +227,7 @@ paddle.fluid.layers.psroi_pool (ArgSpec(args=['input', 'rois', 'output_channels'
 paddle.fluid.layers.teacher_student_sigmoid_loss (ArgSpec(args=['input', 'label', 'soft_max_up_bound', 'soft_max_lower_bound'], varargs=None, keywords=None, defaults=(15.0, -15.0)), ('document', '2f6ff96864054a31aa4bb659c6722c99'))
 paddle.fluid.layers.huber_loss (ArgSpec(args=['input', 'label', 'delta'], varargs=None, keywords=None, defaults=None), ('document', '431a4301c35032166ec029f7432c80a7'))
 paddle.fluid.layers.kldiv_loss (ArgSpec(args=['x', 'target', 'reduction', 'name'], varargs=None, keywords=None, defaults=('mean', None)), ('document', '776d536cac47c89073abc7ee524d5aec'))
-paddle.fluid.layers.tree_conv (ArgSpec(args=['nodes_vector', 'edge_set', 'output_size', 'num_filters', 'max_depth', 'act', 'param_attr', 'bias_attr', 'name'], varargs=None, keywords=None, defaults=(1, 2, 'tanh', None, None, None)), ('document', '34ea12ac9f10a65dccbc50100d12e607'))
+paddle.fluid.layers.tree_conv (ArgSpec(args=['nodes_vector', 'edge_set', 'output_size', 'num_filters', 'max_depth', 'act', 'param_attr', 'bias_attr', 'name'], varargs=None, keywords=None, defaults=(1, 2, 'tanh', None, None, None)), ('document', '2985a372ac897ea4e13aced7f930d6f8'))
 paddle.fluid.layers.npair_loss (ArgSpec(args=['anchor', 'positive', 'labels', 'l2_reg'], varargs=None, keywords=None, defaults=(0.002,)), ('document', '46994d10276dd4cb803b4062b5d14329'))
 paddle.fluid.layers.pixel_shuffle (ArgSpec(args=['x', 'upscale_factor'], varargs=None, keywords=None, defaults=None), ('document', '132b6e74ff642a392bd6b14c10aedc65'))
 paddle.fluid.layers.fsp_matrix (ArgSpec(args=['x', 'y'], varargs=None, keywords=None, defaults=None), ('document', 'b76ccca3735bea4a58a0dbf0d77c5393'))
diff --git a/python/paddle/fluid/layers/nn.py b/python/paddle/fluid/layers/nn.py
index facc5ff11e75511443318f83eca7fd179b2160f7..428692cc63a9a6a75891b74b6581b4fc34388e86 100644
--- a/python/paddle/fluid/layers/nn.py
+++ b/python/paddle/fluid/layers/nn.py
@@ -11051,21 +11051,19 @@ def tree_conv(nodes_vector,
     Examples:
         .. code-block:: python
 
-          nodes_vector = layers.data(name='vectors', shape=[None, 10, 5], dtype='float32)
-          # None for batch size, 10 for max_node_size of dataset, 5 for vector width
-          edge_set = layers.data(name='edge_set', shape=[None, 10, 2], dtype='float32')
-          # None for batch size, 10 for max_node_size of dataset, 2 for every edge has two nodes
+          # 10 for max_node_size of dataset, 5 for vector width
+          nodes_vector = fluid.layers.data(name='vectors', shape=[10, 5], dtype='float32')
+          # 10 for max_node_size of dataset, 2 for every edge has two nodes
           # edges must be directional
-          out_vector = layers.tree_conv(nodes_vector, edge_set, 6, 1, 2, 'tanh',
-              ParamAttr(initializer=Constant(1.0), ParamAttr(initializer=Constant(1.0))
-          # the shape of output will be [None, 10, 6, 1],
-          # None for batch size, 10 for max_node_size of dataset, 6 for output size, 1 for 1 filter
-          out_vector = layers.reshape(out_vector, shape=[None, 10, 6])
+          edge_set = fluid.layers.data(name='edge_set', shape=[10, 2], dtype='float32')
+          # the shape of output will be [10, 6, 1],
+          # 10 for max_node_size of dataset, 6 for output size, 1 for 1 filter
+          out_vector = fluid.layers.tree_conv(nodes_vector, edge_set, 6, 1, 2)
           # After reshape, output tensor could be nodes_vector for next tree convolution
-          out_vector_2 = layers.tree_conv(out_vector, edge_set, 3, 4, 2, 'tanh',
-              ParamAttr(initializer=Constant(1.0), ParamAttr(initializer=Constant(1.0))
+          out_vector = fluid.layers.reshape(out_vector, shape=[-1, 10, 6])
+          out_vector_2 = fluid.layers.tree_conv(out_vector, edge_set, 3, 4, 2)
           # also output tensor could be pooling(the pooling in paper called global pooling)
-          pooled = layers.reduce_max(out_vector, dims=2) # global pooling
+          pooled = fluid.layers.reduce_max(out_vector, dim=2) # global pooling
     """
     helper = LayerHelper("tree_conv", **locals())
     dtype = helper.input_dtype('nodes_vector')