update doc_string

hub_module/modules/text/embedding/word2vec_skipgram/module.py

@@ -70,7 +70,7 @@ class Word2vecSkipGram(hub.Module):
         Returns:
              inputs(dict): the input variables of word2vec_skipgram (words)
              outputs(dict): the output variables of input words (word embeddings)
-             main_program(Program): the main_program of Senta with pretrained prameters
+             main_program(Program): the main_program of word2vec_skipgram with pretrained prameters
         """
         assert num_data >= 1 and num_data <= 3, "num_data(%d) must be 1, 2, or 3" % num_data
         main_program = fluid.Program()

hub_module/modules/text/sentiment_analysis/emotion_detection_textcnn/module.py

@@ -61,7 +61,7 @@ class EmotionDetectionTextCNN(hub.NLPPredictionModule):
              inputs(dict): the input variables of emotion_detection_textcnn (words)
              outputs(dict): the output variables of input words (word embeddings and label probilities);
                  the sentence embedding and sequence length of the first input text.
-             main_program(Program): the main_program of Senta with pretrained prameters
+             main_program(Program): the main_program of emotion_detection_textcnn with pretrained prameters
         """
         assert num_data >= 1 and num_data <= 3, "num_data(%d) must be 1, 2, or 3" % num_data
         main_program = fluid.Program()

hub_module/modules/text/sentiment_analysis/senta_bilstm/net.py

@@ -2,8 +2,8 @@
 import paddle.fluid as fluid
 
 
-def bilstm_net(data,
-               dict_dim,
+def bilstm_net(emb,
+               seq_len,
                emb_dim=128,
                hid_dim=128,
                hid_dim2=96,
@@ -12,15 +12,12 @@ def bilstm_net(data,
     """
     Bi-Lstm net
     """
-    # embedding layer
-    emb = fluid.layers.embedding(
-        input=data,
-        size=[dict_dim, emb_dim],
-    )
+    # unpad the token_feature
+    unpad_feature = fluid.layers.sequence_unpad(emb, length=seq_len)
 
     # bi-lstm layer
-    fc0 = fluid.layers.fc(input=emb, size=hid_dim * 4)
-    rfc0 = fluid.layers.fc(input=emb, size=hid_dim * 4)
+    fc0 = fluid.layers.fc(input=unpad_feature, size=hid_dim * 4)
+    rfc0 = fluid.layers.fc(input=unpad_feature, size=hid_dim * 4)
     lstm_h, c = fluid.layers.dynamic_lstm(
         input=fc0, size=hid_dim * 4, is_reverse=False)
     rlstm_h, c = fluid.layers.dynamic_lstm(

hub_module/modules/text/sentiment_analysis/senta_bow/net.py

@@ -2,15 +2,15 @@
 import paddle.fluid as fluid
 
 
-def bow_net(data, dict_dim, emb_dim=128, hid_dim=128, hid_dim2=96, class_dim=2):
+def bow_net(emb, seq_len, hid_dim=128, hid_dim2=96, class_dim=2):
     """
     Bow net
     """
-    # embedding layer
-    emb = fluid.layers.embedding(input=data, size=[dict_dim, emb_dim])
+    # unpad the token_feature
+    unpad_feature = fluid.layers.sequence_unpad(emb, length=seq_len)
 
     # bow layer
-    bow = fluid.layers.sequence_pool(input=emb, pool_type='sum')
+    bow = fluid.layers.sequence_pool(input=unpad_feature, pool_type='sum')
     bow_tanh = fluid.layers.tanh(bow)
     # full connect layer
     fc_1 = fluid.layers.fc(input=bow_tanh, size=hid_dim, act="tanh")

hub_module/modules/text/sentiment_analysis/senta_cnn/net.py

@@ -2,22 +2,16 @@
 import paddle.fluid as fluid
 
 
-def cnn_net(data,
-            dict_dim,
-            emb_dim=128,
-            hid_dim=128,
-            hid_dim2=96,
-            class_dim=2,
-            win_size=3):
+def cnn_net(emb, seq_len, hid_dim=128, hid_dim2=96, class_dim=2, win_size=3):
     """
     Conv net
     """
-    # embedding layer
-    emb = fluid.layers.embedding(input=data, size=[dict_dim, emb_dim])
+    # unpad the token_feature
+    unpad_feature = fluid.layers.sequence_unpad(emb, length=seq_len)
 
     # convolution layer
     conv_3 = fluid.nets.sequence_conv_pool(
-        input=emb,
+        input=unpad_feature,
         num_filters=hid_dim,
         filter_size=win_size,
         act="tanh",

hub_module/modules/text/sentiment_analysis/senta_gru/net.py

@@ -2,8 +2,8 @@
 import paddle.fluid as fluid
 
 
-def gru_net(data,
-            dict_dim,
+def gru_net(emb,
+            seq_len,
             emb_dim=128,
             hid_dim=128,
             hid_dim2=96,
@@ -12,10 +12,10 @@ def gru_net(data,
     """
     gru net
     """
-    # embedding layer
-    emb = fluid.layers.embedding(input=data, size=[dict_dim, emb_dim])
+    # unpad the token_feature
+    unpad_feature = fluid.layers.sequence_unpad(emb, length=seq_len)
 
-    fc0 = fluid.layers.fc(input=emb, size=hid_dim * 3)
+    fc0 = fluid.layers.fc(input=unpad_feature, size=hid_dim * 3)
 
     # GRU layer
     gru_h = fluid.layers.dynamic_gru(input=fc0, size=hid_dim, is_reverse=False)

hub_module/modules/text/sentiment_analysis/senta_lstm/net.py

@@ -2,20 +2,14 @@
 import paddle.fluid as fluid
 
 
-def lstm_net(data,
-             dict_dim,
-             emb_dim=128,
-             hid_dim=128,
-             hid_dim2=96,
-             class_dim=2,
-             emb_lr=30.0):
+def lstm_net(emb, seq_len, hid_dim=128, hid_dim2=96, class_dim=2, emb_lr=30.0):
     """
     Lstm net
     """
-    # embedding layer
-    emb = fluid.layers.embedding(input=data, size=[dict_dim, emb_dim])
+    # unpad the token_feature
+    unpad_feature = fluid.layers.sequence_unpad(emb, length=seq_len)
     # Lstm layer
-    fc0 = fluid.layers.fc(input=emb, size=hid_dim * 4)
+    fc0 = fluid.layers.fc(input=unpad_feature, size=hid_dim * 4)
     lstm_h, c = fluid.layers.dynamic_lstm(
         input=fc0, size=hid_dim * 4, is_reverse=False)
     # max pooling layer