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 supported17 Subscribers
Add a CodeTriage badge to deepspeed
Help out
- Issues
- [REQUEST] Model serving via deepspeed's inference module
- [QUESTION] About Pipeline Parallelism save_checkpoint
- [QUESTION] Skipping examples with NaN loss
- [BUG] loss discrepancy among ZeRO-0, 1, 2, 3, when gradient accumulate multiple steps
- [BUG] AttributeError: 'FusedAdam' object has no attribute 'set_grad_none'
- Issues with running inference on quantized model
- DGX-1 / V100-32GB-SMX2 - nvlink is not activated (0 activity)
- Performance degradation after using DeepSpeed (UniLM/layoutlmv2)
- Transformer/fix layer norm
- [REQUEST] Support mixed precision calculation with ZERO
- Docs
- Python not yet supported