I want to train an LSTM network for time-series predictions, and want to get to the bottom of LSTM's. In my understanding, the number of cells in a single LSTM layer can vary. However, since each cell takes an input at time-step t, wouldn't the number of cells need to be equal to t?
For example (from TensorFlow tutorial):
t=0 t=1 t=2 t=3 t=4
[The, brown, fox, is, quick]
[The, red, fox, jumped, high]
words_in_dataset[0] = [The, The]
words_in_dataset[1] = [brown, red]
words_in_dataset[2] = [fox, fox]
words_in_dataset[3] = [is, jumped]
words_in_dataset[4] = [quick, high]
batch_size = 2, time_steps = 5
Wouldn't the maximum number of cells in each layer be 5? since after the 5th input we no longer have any other information to input. However, I've seen many networks with a higher number of cells than that. Therefore, why is this a possibility?