transformers
https://github.com/huggingface/transformers
Python
🤗 Transformers: State-of-the-art Machine Learning for Pytorch, TensorFlow, and JAX.
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 supported55 Subscribers
Add a CodeTriage badge to transformers
Help out
- Issues
- OPT training with torch.compile decomposes SDPA into Triton kernels due to attention_mask
- 4D mask created under torch.compile when padding_mask is None
- Skip mask creation when padding_mask is None under torch.compile
- [Enhancement] Get rid of "paged|" implementations
- MoE models that balance with a router bias never expose their router logits
- MtpModel/MtpLayer topology is too rigid to represent non-uniform MTP architectures
- Image processors silently mutate class defaults when instantiated with min_pixels
- Let `prefix_allowed_tokens_fn` override model `-inf` and raise an exception on unsatisfiable generation constraints.
- Fix: Preserve non-persistent buffer values when moving missing keys off meta device
- Fix: SentencePieceBackend.convert_tokens_to_string dropping byte-fallback tokens
- Docs
- Python not yet supported