LSTM implementation

[mees]

Hi,
I wanted to ask about the LSTM implementation as it looks a bit different to the original implementation, for example it is actually not necessary to use an tf.nn.dynamic_rnn for variable sized sequences, as the inputs are already padded to have the same length, right?

[InnerPeace-Wu]

Yes, the input is padded as the same length. I did not give much attention on the implementation, i just found out the one worked for me, so there may be more suitable way to do the job.