attention-is-all-you-need-pytorch
attention-is-all-you-need-pytorch copied to clipboard
add teacher forcing parameter
Firstly,thanks for your improvemen for the original version,it's very helpful!But i got a problem in this code :my train accuracy is very high at the begining(about 55% );and the validation accuracy is very high from starting to end(about 99%),almost having no change.i am really confused with this problem^_^