human-motion-prediction
human-motion-prediction copied to clipboard
Encoder inputs are less than source_seq_len by 1
Hi @una-dinosauria I'm wondering why did you choose encoder_inputs to be less than source_seq_len by 1? Line 453 While you choose decoder_outputs and decoder_inputs to be the same as target_seq_len. https://github.com/una-dinosauria/human-motion-prediction/blob/c9a2774fe59bb78c6ec6d04e15f2cb7243c4348c/src/seq2seq_model.py#L435-L455
Because the last frame of encoder_inputs
is given as a seed to the decoder.