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
- [BUG] Hybrid Engine with DeepSpeed Stage 3 results and Llama V2 results in gibberish outputs
- can deepspeed fast-gen support int8 weightonly inference
- [BUG] Can't perform gradient accumulation on pipeline parallelism with inputs of different lengths
- [BUG] Failed to checkpoint with deepspeed 0.12.4
- [BUG] The forward hook function of Zero stege3 might misjudge the current step of the training process whether it is in the forward computation or in the backward recomputation
- [BUG]
- Deepspeed fails with frozen weights (e.g. only train llama2 embedding layer)
- Fix for HF integrations CI
- [REQUEST] Autocast for BF16
- [BUG] `DeepSpeedCPUAdam` compatibility on ROCm
- Docs
- Python not yet supported