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 supported20 Subscribers
Add a CodeTriage badge to deepspeed
Help out
- Issues
- [WIP][AutoTP] Complete uneven sharding and universal checkpoint support
- Add managed_gradient_accumulation for ZeRO stage 0/1
- Add Triton grouped-GEMM for MoE experts on Ampere/Ada
- Non-reentrant activation checkpoint CPU offload
- Make PipelineModule.set_checkpoint_interval actually change the interval
- Stabilize DeepCompile ZeRO-3 memory scheduling
- [RFC] Support sharding LM heads and adopting Online Softmax
- [AutoTP] Fix ZeRO-3 checkpoint consolidation to gather across TP and DP
- fix: avoid ZeroDivisionError in compute_elastic_config return_microbatch path
- [Pipeline] Add Ray-backed stage placement and execution with pluggable transportsFeature/ray pipeline parallelism
- Docs
- Python not yet supported