Let's say I have one input feature with 10 sequences of length 25 to predict the next value. So keras will receive an input vector of (10, 25, 1). If I use LSTM models with stateful=True the cell states will be remembered in between batches.

However, if I have multiple sequences in each batch, is the training equivalent to having stateful=False and a single batch of length 10?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.