accelerate
https://github.com/huggingface/accelerate
Python
🚀 A simple way to train and use PyTorch models with multi-GPU, TPU, mixed-precision
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 supported1 Subscribers
Add a CodeTriage badge to accelerate
Help out
- Issues
- handle weight sharing with init_on_device
- Unable to specify HYBRID_SHARD for FSDP which requires process group or device_mesh to be passed
- Why does accelerate's MegatronLMPlugin use its own megatron?
- Enable cpu offload with weights inside the module
- WebLoader Support
- stderr: AssertionError: Main process was not first
- Loading weights straight to GPU only works for inference
- Feature request: support for multi-node, multi-GPU distributed inference
- accelerate autocast on mps device
- Feature request: FSDP for TPUs
- Docs
- Python not yet supported