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
- AssertionError: AutoTP not supported for model. Please use kernel injection since container policy for model exists.
- [BUG] Inference failed serveral times
- Fix handling of (CUDA,ROCR)_VISIBLE_DEVICES
- [BUG] Memory increase consistently when using multiple NCCL_IB_HCA.
- [REQUEST] Add Aim to the Monitor Module
- [REQUEST] Add torchdynamo disable decorators to graph-break on collectives
- [Performance] Excessive small reduce operations scheduled in backward pass with ZeRO2
- [REQUEST] BF16 in DeepSpeed Inference
- [BUG] Cannot get elastic scaling to work
- [BUG] terminate called after throwing an instance of 'std::bad_alloc'
- Docs
- Python not yet supported