Skip to content
This repository has been archived by the owner on Aug 10, 2023. It is now read-only.

Pre-Release v0.1.1

Pre-release
Pre-release
Compare
Choose a tag to compare
@hfxunlp hfxunlp released this 18 Jan 11:29
· 64 commits to master since this release

Bias in MultiHeadAttn is removed in this release;
Only parameters of the trained model is saved, rather than state dict, trained model with v0.1.0 is not loadable as fine_tune_m without additional convertion;
Try to support RNMT, but recurrent is slow due to less efficient parallelization.