neural network - Backpropagation through time in stateful RNNs -


if use stateful rnn in keras processing sequence of length n divided n parts (each time step processed individually),

  1. how backpropagation handled? affect last time step, or backpropagate through entire sequence?
  2. if not propagate through entire sequence, there way this?

the propagation horizon limited second dimension of input sequence. i.e. if data of type (num_sequences, num_time_steps_per_seq, data_dim) prop done on time horizon of value num_time_steps_per_seq take @

https://github.com/fchollet/keras/issues/3669


Comments

Popular posts from this blog

php - Auto increment employee ID -

php - isset function not working properly -

python - Evaluating the next line in a For Loop while in the current iteration -