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
- ds_inference success but OOM when use tp_presharded_mode=True[BUG]
- How zero3 releases the GPU memory occupied by parameters?
- pipe_parallel model continue eval_batch train_batch eval_batch.
- [Question]How to training on different node with specific gpu in each node
- [REQUEST] quantize_training document update
- Poetry can't add DeepSpeed
- [BUG] if use zero3+tp, use save_16bit_model to save weights, only save 1/tp weights
- [BUG] ZeRO 3 error: expected the next 4 parameters in the parameter fetch queue to be ... but got ()
- Barely any speedup when running Llama for inference (with Kernel Injection) [BUG]
- [BUG] hybrid_engine for zero 3 seems invalid
- Docs
- Python not yet supported