deepspeed
https://github.com/microsoft/deepspeed
Python
DeepSpeed is a deep learning optimization library that makes distributed training easy, efficient, and effective.
Triage Issues!
When you volunteer to triage issues, you'll receive an email each day with a link to an open issue that needs help in this project. You'll also receive instructions on how to triage issues.
Triage Docs!
Receive a documented method or class from your favorite GitHub repos in your inbox every day. If you're really pro, receive undocumented methods or classes and supercharge your commit history.
Python not yet supported21 Subscribers
Add a CodeTriage badge to deepspeed
Help out
- Issues
- Fix flop counts for the matmul family in the flops profiler
- Skip narrow dtypes in the mixed-precision cast
- Fix autotuner crash when an experiment produces no metric
- Log the skipped optimizer step on gradient overflow
- fix: remove dead gradient norm fallback
- Forward the inputs DSUNet._forward accepts to the UNet
- Use the val argument in mask_nan_or_inf_with_val_inplace
- ZeRO-3 HybridMegatron QKV layout mutations are not persisted
- Size the interpolate flops from the whole output tensor
- Size the elastic v0.2 micro batch from the data-parallel world
- Docs
- Python not yet supported