facebookresearch/fairseq

key_padding_mask is not used in transformer decoder layer?

Open

#537 aperta il 27 feb 2019

Vedi su GitHub
 (1 commento) (0 reazioni) (0 assegnatari)Python (6224 fork)batch import
bughelp wanted

Metriche repository

Star
 (29.107 star)
Metriche merge PR
 (Nessuna PR mergiata in 30 g)

Descrizione

when reading the source code i found that key_padding_mask is not used when calculating self attention. There is no problem when the target is padded right by default because attn_mask could do the same thing. But how about left padding on the target?

Guida contributor