sentence-transformers
https://github.com/ukplab/sentence-transformers
Python
Sentence Embeddings with BERT & XLNet
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 supported15 Subscribers
Add a CodeTriage badge to sentence-transformers
Help out
- Issues
- TSDAE uses the wrong decoder token IDs when BERT vocabularies differ
- Fix TSDAE retokenization for untied decoder vocabularies
- feat: Cut peak GPU memory and runtime of `community_detection`, with the same output
- `community_detection`: Cut peak GPU memory 13–39× and runtime 1.25–1.64×, with the same output
- [fix] Keep the (0, dim) shape when encoding an empty input
- [fix] Accept tensor and multi-vector calibration in quantize_embeddings
- Respect per-call tokenization when excluding prompt tokens
- Keep sparse embedding normalization gradients finite for zero rows
- Fix per-input hidden states in raw encode outputs
- [Proposal] Nano evaluators for any Nano-set in the NanoBEIR format
- Docs
- Python not yet supported