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
- Fix LukeTokenizer dropping <s>/</s> on the plain-text path
- Fall back to model type for unresolvable image processor names
- [RFC] Zero-Copy In-Place MoE Dynamic Token Routing for Mixtral/DeepSeek (1.77x Speedup, 0 Bytes Aux VRAM)
- Fix GPT-NeoX family tokenization when config flags and serialized post-processor disagree
- LukeTokenizer drops <s>/</s> on the plain-text path (no post-processor installed)
- Decide whether the checkpoint shard prefetch should be on by default
- Expert-parallel token dispatch: each rank trains on its own part of the batch
- test_get_cached_module_file_local_handles_symlinked_hub_cache fails on Windows without symlink privilege
- Fix odd head_dim validation for RoPE configurations
- fix(audio): Route the ALM conversion scripts through from_pretrained
- Docs
- Python not yet supported