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
- accelerator.reduce(tensor, "mean") fails on integer tensors when more than one process is used
- Fix reduce(..., "mean") on integer tensors in distributed runs
- Preserve single-tensor inputs when clipping gradients
- get_max_memory() budgets CPU offload from host RAM, not the container memory limit
- Clean up temporary checkpoint offloads when loading fails
- Handle iterable loaders safely when joining uneven inputs
- Cap the CPU budget of get_max_memory at the cgroup memory limit
- Release first-process ordering barriers when the block raises
- Enforce execute_subprocess_async timeout and fail on signal kills
- fix: support windows path separators when sorting automatic checkpoints (#4454)
- Docs
- Python not yet supported