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 supported18 Subscribers
Add a CodeTriage badge to deepspeed
Help out
- Issues
- Zero Stage-2 Frozen Layers[BUG]
- Issue with DeepSpeed Inference - Multiple Processes for Model Loading and Memory Allocation
- [REQUEST] [zero_to_fp32.py] Allow user to specify the dtype (e.g. `torch.bfloat16`) in the output file
- [REQUEST] How to use int8 quantization inference without training?
- Exception: Current loss scale already at minimum - cannot decrease scale anymore. Exiting run.
- [BUG]Zero++ quantizer unsupport BFloat16
- Deepspeed Inference not working on llama when input has padding and using kernel injection
- [BUG] model.load_checkpoint out of memory
- deepspeed stage 3 do not save the lr_scheduler
- [BUG] Pipeline Parallelism cannot work with BFloat16 + Optimizer offload
- Docs
- Python not yet supported