LSTM Class - Keras.layers.lstm

[Denis Dal Soler]

Hi

is there someone that can please explain me the architecture that this instruction makes (better graphically)?

model.add(LSTM(128, return_sequences=True, input_shape=(look_back, 1)))

In some discussion I found that 128 are the LSTM units of the hidden LSTM layer, in others I found 128 are the LSTM cells of the layer (and each cell has look_back LSTM unit).

Can someone that knows the python code of this instruction (where I can find it?) tell me which kind of architecture it makes? How many single LSTM (meaning forget gate, input gate, output gate)   are there?? 128?128xlook_back?Others?

Thanks

Denis

[Denis Dal Soler]

Hi

is there someone that can please explain me the architecture that this instruction makes (better graphically)?

model.add(LSTM(128, return_sequences=True, input_shape=(look_back, 1)))

In some discussion I found that 128 are the LSTM units of the hidden LSTM layer, in others I found 128 are the LSTM cells of the layer (and each cell has look_back LSTM unit).

Can someone that knows the python code of this instruction (where I can find it?) tell me which kind of architecture it makes? How many single LSTM (meaning forget state, input state, output state)   are there?? 128?128xlook_back?Others?

Thanks

Denis

--

denisdalsoler.com

+39 3281638511