For seq2seq, got any examples like CTC or attention?


(Rchen Andrews) #1

get layer like CTC or attention?


#2

You could look at https://github.com/pytorch/examples/tree/master/OpenNMT


(Adam Paszke) #3

Also see https://github.com/pytorch/pytorch/issues/570


#4

Now DeepSpeech (with CTC loss) is implemented in PyTorch here:


(Rchen Andrews) #5

thks, guys!:slight_smile: