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
- Function sample_negatives is the performance bottleneck of multi-GPU distributed training.
- NLP / GSLM: Did anyone succeed in making K-means clustering model? (S2U)
- Implemented simple perturbations described in "Rethinking Perturbations in Encoder-Decoders for Fast Training"
- Test failures after upgrading to PyTorch 1.13.0
- Request to provide torchscript-ed model
- Hallucination with numbers : NLLB English to Spanish translation
- Data preprocessing and training scripts to reproduce the "Efficient Large Scale Language Modeling with Mixtures of Experts"
- how to use generate() or translation() in Simultaneous Translation
- Different vocab size between xlmr-xxlarge and xlmr-large
- Average losses with the correct ntokens due to prefix-size > 0
- Docs
- Python not yet supported