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 supported9 Subscribers
Add a CodeTriage badge to deepspeed
Help out
- Issues
- [QUESTION] Pipeline explaination needed
- Feature request: Ability to disable autocast locally
- [Question] How to preshard a model for tensor parallism
- Unify benchmark scripts and knowledge
- Schema for ds_config.json
- [REQUEST] An option to only save the model state_dict when save_checkpoint(), and how to manually save & load the model state_dict when using ZERO3
- DeepSpeed still gives CUDA-out-of-memory issue
- [Question] why are overlap and contiguous grads meaningless in stage 1 and are ignored
- [BUG] Deepspeed inference engine seems to only support tensor parallelism and not model parallelism; Can we partition a model into pipeline stages via the inference engine support in deep speed ?
- ZeroQuant not compressing and making BERT slower
- Docs
- Python not yet supported