diff --git a/python/paddle/v2/dataset/wmt16.py b/python/paddle/v2/dataset/wmt16.py
index bbc28a2da99052308471931122946d0d96b54da5..e2f463be2f7bcd667855f64206d78f387e92ef33 100644
--- a/python/paddle/v2/dataset/wmt16.py
+++ b/python/paddle/v2/dataset/wmt16.py
@@ -171,8 +171,9 @@ def train(src_dict_size, trg_dict_size, src_lang="en"):
         callable: The train reader.
     """
 
-    assert (src_lang in ["en", "de"], ("An error language type.  Only support: "
-                                       "en (for English); de(for Germany)"))
+    if src_lang not in ["en", "de"]:
+        raise ValueError("An error language type.  Only support: "
+                         "en (for English); de(for Germany).")
     src_dict_size, trg_dict_size = __get_dict_size(src_dict_size, trg_dict_size,
                                                    src_lang)
 
@@ -218,9 +219,9 @@ def test(src_dict_size, trg_dict_size, src_lang="en"):
         callable: The test reader.
     """
 
-    assert (src_lang in ["en", "de"],
-            ("An error language type.  "
-             "Only support: en (for English); de(for Germany)"))
+    if src_lang not in ["en", "de"]:
+        raise ValueError("An error language type. "
+                         "Only support: en (for English); de(for Germany).")
 
     src_dict_size, trg_dict_size = __get_dict_size(src_dict_size, trg_dict_size,
                                                    src_lang)
@@ -266,9 +267,9 @@ def validation(src_dict_size, trg_dict_size, src_lang="en"):
     Returns:
         callable: The validation reader.
     """
-    assert (src_lang in ["en", "de"],
-            ("An error language type.  "
-             "Only support: en (for English); de(for Germany)"))
+    if src_lang not in ["en", "de"]:
+        raise ValueError("An error language type. "
+                         "Only support: en (for English); de(for Germany).")
     src_dict_size, trg_dict_size = __get_dict_size(src_dict_size, trg_dict_size,
                                                    src_lang)
 
diff --git a/python/paddle/v2/fluid/layers/nn.py b/python/paddle/v2/fluid/layers/nn.py
index 7b3ad707cc62938a47b65daba852aec7cc7e1caa..a01ccfa635301108e337668188dd14ed0c0b1d8a 100644
--- a/python/paddle/v2/fluid/layers/nn.py
+++ b/python/paddle/v2/fluid/layers/nn.py
@@ -2141,7 +2141,7 @@ def sequence_reshape(input, new_dim):
     return out
 
 
-def transpose(input, perm, name=None):
+def transpose(x, perm, name=None):
     """
     **transpose Layer**
 
@@ -2161,19 +2161,19 @@ def transpose(input, perm, name=None):
         .. code-block:: python
 
             x = fluid.layers.data(name='x', shape=[5, 10, 15], dtype='float32')
-            x_transposed = layers.transpose(input=x, perm=[1, 0, 2])
+            x_transposed = layers.transpose(x, perm=[1, 0, 2])
     """
 
-    if len(perm) != len(input.shape):
+    if len(perm) != len(x.shape):
         raise ValueError(
             "Input(perm) is the permutation of dimensions of Input(input). "
             "It's length shoud be equal to Input(input)'s rank.")
 
     helper = LayerHelper('transpose', **locals())
-    out = helper.create_tmp_variable(helper.input_dtype())
+    out = helper.create_tmp_variable(x.dtype)
     helper.append_op(
         type='transpose',
-        inputs={'X': [input]},
+        inputs={'X': [x]},
         outputs={'Out': [out]},
         attrs={'axis': perm})
     return out
diff --git a/python/paddle/v2/fluid/layers/ops.py b/python/paddle/v2/fluid/layers/ops.py
index b517f8be6a3e5558dd01afe094fb3989cfb3af44..022a94cad440f13383a927233195bb008a688843 100644
--- a/python/paddle/v2/fluid/layers/ops.py
+++ b/python/paddle/v2/fluid/layers/ops.py
@@ -45,10 +45,20 @@ __activations__ = [
 ]
 
 __all__ = [
-    'mean', 'mul', 'reshape', 'scale', 'transpose',
-    'sigmoid_cross_entropy_with_logits', 'elementwise_add', 'elementwise_div',
-    'elementwise_sub', 'elementwise_mul', 'elementwise_max', 'elementwise_min',
-    'clip', 'clip_by_norm', 'sequence_softmax'
+    'mean',
+    'mul',
+    'reshape',
+    'scale',
+    'sigmoid_cross_entropy_with_logits',
+    'elementwise_add',
+    'elementwise_div',
+    'elementwise_sub',
+    'elementwise_mul',
+    'elementwise_max',
+    'elementwise_min',
+    'clip',
+    'clip_by_norm',
+    'sequence_softmax',
 ] + __activations__
 
 for _OP in set(__all__):
diff --git a/python/paddle/v2/fluid/tests/book/test_understand_sentiment_lstm.py b/python/paddle/v2/fluid/tests/book/test_understand_sentiment_lstm.py
index 618191424150eb7c5a24407fc2e106ee8825fedb..117f74c59ad5bf6bb67711801cd7b9a41f39f1f8 100644
--- a/python/paddle/v2/fluid/tests/book/test_understand_sentiment_lstm.py
+++ b/python/paddle/v2/fluid/tests/book/test_understand_sentiment_lstm.py
@@ -65,13 +65,13 @@ def lstm_net(dict_dim, class_dim=2, emb_dim=32, seq_len=80, batch_size=50):
 
     emb = fluid.layers.embedding(input=data, size=[dict_dim, emb_dim])
     emb = fluid.layers.reshape(x=emb, shape=[batch_size, seq_len, emb_dim])
-    emb = fluid.layers.transpose(x=emb, axis=[1, 0, 2])
+    emb = fluid.layers.transpose(x=emb, perm=[1, 0, 2])
 
     c_pre_init = fluid.layers.fill_constant(
         dtype=emb.dtype, shape=[batch_size, emb_dim], value=0.0)
     c_pre_init.stop_gradient = False
     layer_1_out = lstm(emb, c_pre_init=c_pre_init, hidden_dim=emb_dim)
-    layer_1_out = fluid.layers.transpose(x=layer_1_out, axis=[1, 0, 2])
+    layer_1_out = fluid.layers.transpose(x=layer_1_out, perm=[1, 0, 2])
 
     prediction = fluid.layers.fc(input=layer_1_out,
                                  size=class_dim,