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
- Resume from Zero OOM，because this line let every SINGLE DP process load ALL optimizer states, Memory Needed * N
- SparseSelfAttention doesn't work with max_seq_length=4096
- checkpoint type error
- Bert training model failed when add --deepspeed_transformer_kernel
- Distillation implementation for 3D parallelism.
- ZERO Stage 3 with NVME Offload CPU OOM
- `sub_group_size` is not documented
- [fp16] need to automatically convert fp32 inputs to fp16
- [performance] fusing zero.Init and pretrained weights loading
- Feature Request: cleanup() or close() method
- Docs
- Python not yet supported