gentaiscool/distfuse

A library to calculate similarity scores between two collections of text sequences encoded using transformer models for bitext mining, dense retrieval, retrieval-based classification, and retrieval-augmented generation (RAG).

39
/ 100
Emerging

Supports ensemble scoring by combining multiple heterogeneous encoders (Hugging Face, OpenAI, Cohere) with configurable weights and distance metrics (cosine, euclidean, manhattan), including instruction-tuned models. Implements the MINERS approach for multilingual dense retrieval, enabling both pairwise similarity computation and multi-reference evaluation scenarios where predictions are scored against variable numbers of reference texts.

No commits in the last 6 months. Available on PyPI.

Stale 6m
Maintenance 0 / 25
Adoption 7 / 25
Maturity 18 / 25
Community 14 / 25

How are scores calculated?

Stars

5

Forks

3

Language

Python

License

Apache-2.0

Last pushed

Jun 22, 2024

Monthly downloads

20

Commits (30d)

0

Dependencies

5

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/embeddings/gentaiscool/distfuse"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.