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
- Resuming with load_state + skip_first_batches (checkpointing example) does not restore the shuffle order
- Report subprocess failures and enforce requested timeouts
- Fix compile_regions breaking ModuleDict and ModuleList indexing
- Fix skipped-step detection for AMP-aware optimizers
- Fix stateful dataloader checkpoint cache after loading
- Fix shape metadata when copying tensors across devices
- Add an FSDP2-native Ulysses sequence parallelism backend
- `use_stateful_dataloader` + `num_workers>0` draws epoch-0 permutation during `prepare()` before cross-rank RNG sync — corrupts DDP data partition
- Fix stateful dataloader checkpointing across processes
- `datasets` Iterable Dataset sharding support
- Docs
- Python not yet supported