search

buriburisuri / sugartensor

A slim tensorflow wrapper that provides syntactic sugar for tensor variables. This library will be helpful for practical deep learning researchers not beginners.
MIT License
372 stars 63 forks source link

How to stack LSTM layer after conv1d layer #36

Open galaxyh opened 6 years ago

galaxyh commented 6 years ago

I am experimenting on speech-to-text-wavenet. I tried to add a LSTM layer after conv1d layer, for example: skip .sg_conv1d(size=1, act='tanh', bn=True, name='conv_1') .sg_lstm(last_only=True, name='rnn_1') .sg_dense(dim=emotion_size) The output shape of sg_conv1d is (16, ?, 128)

When running, I got the following error: sg_layer.py", line 499, in sg_rnn for i in range(tensor.get_shape().as_list()[1]): TypeError: range() integer end argument expected, got NoneType.

Is there have any advice? thanks

jesuistay commented 6 years ago

Have had the same problem. I got no idea