Rnn_forward t start from 0 or 1

for t in range(T_x):
a_next, yt_pred, cache = rnn_cell_forward(x[:,:,t],a_next, parameters)
I think t start from 0 and if we use this, we compute a for time step 0th. (RNN cell 0th)
But when I see this paragraph, it doesn’t have RNN cell 0th

Traditionally the examples are numbered starting from 1.
Do not expect the diagrams to be mathematically correct.

1 Like