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 supported8 Subscribers
Add a CodeTriage badge to deepspeed
Help out
- Issues
- [REQUEST]Inference Optimized Pipeline Parallelism
- [BUG] PipelineEngine does not store losses on last stage in pipeline buffers
- [BUG] Training speed is slower and consumes more memory than FSDP
- [BUG] Parameter name of `DeepSpeedCPUAdam` optimizer
- [BUG] OPT-66B fails to find layer weights
- [BUG] Galactica-6.7B: Asking to pad but the tokenizer does not have a padding token
- deepspeed stage 3 do not save the lr_scheduler
- [BUG] Flash attention seems cannot be integrated with pipeline parallelism due to absent of input grad
- [BUG]I cannot run use DeepSpeed Chat train my model with enabled "hybrid_engine" in step3
- [BUG] Pipeline Parallelism cannot work with BFloat16 + Optimizer offload
- Docs
- Python not yet supported