fairseq
https://github.com/facebookresearch/fairseq
Python
Facebook AI Research Sequence-to-Sequence Toolkit written in Python.
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 supported17 Subscribers
Add a CodeTriage badge to fairseq
Help out
- Issues
- Wav2Vec2 pipeline feature extractor normalizes input over batch dimension, is it a feature or bug in design?
- How to save the fine-tune checkpoint
- NLLB-200 & LLama 4: confusion about Nordic languages
- Add an option to disable appending eos to the end of each document while preprocessing
- Add option to disable EOS appending in preprocessing (#5615)
- payload['cfg'] has no 'model' attribute or it is None.
- Using RoBERTa with more than 512 tokens
- Installation fails on Windows: Missing stddef.h while building fairseq C++ extension (libbleu.cpp)
- Why am I running nllb-200 on A100 with only 30% GPU utilization? Can you help me solve it
- Question regarding the model size, token limits and model optimizaion?
- Docs
- Python not yet supported