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
- Drop the documented grad_hooks ZeRO option, which does not exist
- [AutoTP] Replace tp_shard process-wide globals with per-model AutoTPMeta
- Filter --include against the real slots, not against itself
- Add adaptive gather-buffer pooling for DeepCompile ZeRO-3
- Split DeepCompile ZeRO-3 memory scheduler
- Update versions of steps in workflows
- Update pytest version per latest dependabot
- ZeRO-2 + bf16 silently computes incorrect gradients at multi-rank when a submodule is used more than once per step, a regression from #7665, still present in 0.19.3
- [RFC] Universal checkpoint: non-semantic per-parameter geometric shard map
- Cleanup: remove tp_shard process-wide scalar globals (num_kv_heads / tp_grain_size / ...), thread explicitly
- Docs
- Python not yet supported