Attention_Network_With_Keras

Attention_Network_With_Keras copied to clipboard

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

z595054650

Hi, Thank you for your example. I'm trying to use this Attention example in my LSTM model. However, in `def attention layer`, the line `h = Lambda(lambda X: K.zeros(shape=(K.shape(X)[0], n_h)))(X)`...

wandli