Attention_Network_With_Keras h, _, c = at_LSTM(context, initial

h, _, c = at_LSTM(context, initial_state=[h, c])

Open z595054650 opened this issue 5 years ago • 1 comments

Why not take the output of the previous time step as the input of the next time step, together with context as the input?

Dec 13 '19 11:12 z595054650

While it is technically correct, it is more idiomatic to separate the previous output and true input: context

Does this answer your question? c:

Mar 31 '21 07:03 Choco31415