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
- Log Train/Samples/train_loss to monitors again
- SuperOffload: checkpoint the CPU Adam state, fix unused parameters and rollbacks, speculate with clipping
- Honor explicit dtype in ZeRO tensor constructors
- Fix ZeRO-1/2 lazy optimizer-state linking for all parameter groups
- [ROCm] Fix the ROCm build of inference_core_ops
- [ops] Warn when DeepSpeedCPUAdam is limited to one OpenMP thread
- Fix legacy MoE dispatch capacity agreement across expert-parallel ranks
- [RFC] Agentic RL training on the DeepSpeed hybrid engine
- [checkpoint] Batch replica selection for affine transfer plans
- [RFC] Batch replica selection for affine transfer plans
- Docs
- Python not yet supported