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 supported21 Subscribers
Add a CodeTriage badge to deepspeed
Help out
- Issues
- [REQUEST] Add ReFlow: Asynchronous CPU-Offloaded Optimizers for ZeRO-3
- Honor --module and --no_python in the SLURM launcher
- Forward core binding flags through the pdsh launcher
- Write CPU optimizer results back to non-contiguous params
- Clean up FLOPs profiling when inference or reporting raises
- Add ReFlow asynchronous CPU-offloaded optimizers for ZeRO-3
- Handle unknown CPU count when setting NVCC threads
- [BUG] get_model_profile leaks profiler state when inference or reporting raises
- [BUG] Flaky: TestZero2CPUOffloadGradAccumNorm::test fails intermittently (accumulated_grads_in_cpu should be non-zero after first backward)
- Fix: [BUG] Flaky: TestZero2CPUOffloadGradAccumNorm::test fails intermittent
- Docs
- Python not yet supported