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
- Security: Unverified remote wheel installation enables supply-chain compromise
- DataLoader shuffle sequence replays from epoch 0 after resuming from a checkpoint
- device_map="auto": silent corruption of tensors captured in register_forward_hook (3+ GPUs, inference_mode, stale bare references)
- Isssue when using torch.compile
- [Feature] Save model-only feature in `save_state`
- Save model-only feature in `save_state`
- Feat: Support dynamic batch size in BatchSamplerShard with even_batches
- `datasets` Iterable Dataset sharding support
- Deadlock when running multi tasks
- Feature Request: Add API to Materialize Meta Tensors for Device Movement
- Docs
- Python not yet supported