PyTorch_Biaffine_Dependency_Parsing icon indicating copy to clipboard operation
PyTorch_Biaffine_Dependency_Parsing copied to clipboard

Hello, can I ask you one question. Why use LSTMCell instead of LSTM in LSTM layer?

Open MHaoYun opened this issue 5 years ago • 0 comments

LSTM no need to write a loop to do one pass of backprop through time. And num_layers can be set as LSTM parameter. I think it's much easier to use LSTM. Why not use LSTM?

MHaoYun avatar Dec 26 '19 08:12 MHaoYun