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
- DeepSpeed has its own process group
- Severe GPU training performance degradation due to incorrectly set `KMP_AFFINITY`
- Enable evaluation during deepspeed Sequence Parallel
- `save_function` arg in `save_pretrained` is dropped in transformers 5 RCs
- RuntimeError: Tensor on device meta is not on the expected device cuda:0!
- Accelerate and DataLoaders
- Why TP can't be used with pure DP?
- [Docs] Update low-precision training docs for MS-AMP
- Support stateful_dataloader for IterableDatasetShard
- Gradient accumulation gives worse results when using DeepSpeed ZeRO 2
- Docs
- Python not yet supported