manzar96
manzar96
> @zhao1402072392 teacher forcing raito may be used during training. You can have a look at the model.py, where you can find out that the ratio is actually 1. If...
Did you solve the bug on beam search?
@dimi1357 Did you finally make it work? Can you provide me the "full changes" in some way? I am also interested in using the GPT2 model as decoder.
Hello, I also mailed the concerned author but i got no reply too! Did you find the dataset?
@ShuaiBai623 @JustinLin610 the download link provided in github seems inactive. Could you kindly share access to the dataset or an updated link?