11 Commits

Author SHA1 Message Date
Fangjun Kuang
b1b21eb1e4 Fix decoder padding mask. 2021-08-04 14:57:06 +08:00
Fangjun Kuang
a6d9b3c9ab Minor fixes. 2021-08-03 22:16:34 +08:00
Fangjun Kuang
2be7a0a555 Remove unused code. 2021-08-03 17:24:06 +08:00
Fangjun Kuang
f6091b10c0 Refactor transformer.py 2021-08-02 23:48:26 +08:00
Fangjun Kuang
1fa30998da WIP: Refactoring 2021-07-31 20:24:47 +08:00
Fangjun Kuang
398ed80d7a Minor fixes to support DDP training. 2021-07-31 15:26:57 +08:00
Fangjun Kuang
b94d97da37 Disable gradient computation in evaluation mode. 2021-07-29 20:37:31 +08:00
Fangjun Kuang
acc63a9172 WIP: Add BPE training code. 2021-07-29 20:23:52 +08:00
Fangjun Kuang
bd69e4be32 Use attention decoder for rescoring. 2021-07-28 12:22:09 +08:00
Fangjun Kuang
f65854cca5 Add BPE decoding results. 2021-07-27 17:38:47 +08:00
Fangjun Kuang
4ccae509d3 WIP: Begin to add BPE decoding 2021-07-26 20:06:58 +08:00