Tutel v0.2.0
What's New in v0.2.0:
- Support Windows Python3 + Torch Installation;
- Add examples to enable Tutel MoE in Fairseq;
- Refactor MoE Layer implementation, letting all features (e.g. top-X, overlap, parallel_type, capacity, ..) be able to change at different forward interations;
- New features: load_importance_loss, cosine router, inequivalent_tokens;
- Extend capacity_factor value that includes zero value and negative values for smarter capacity estimation;
- Add tutel.checkpoint conversion tools to reformat checkpoint files, making it able to use existing checkpoints to train/infer with a different world size.
How to Setup:
python3 -m pip install --user https://github.com/microsoft/tutel/archive/refs/tags/v0.2.0.tar.gz