feat: update default reranker to RRF (#1580)

- Both LinearCombination (the current default) and RRF are pretty fast
compared to model based rerankers. RRF is slightly faster.
- In our tests RRF has also been slightly more accurate.

This PR:
- Makes RRF the default reranker
- Removed duplicate docs for rerankers
This commit is contained in:
Ayush Chaurasia
2024-09-03 14:00:13 +05:30
committed by GitHub
parent fde636ca2e
commit 03ef1dc081
3 changed files with 12 additions and 190 deletions

View File

@@ -71,6 +71,8 @@ LanceDB comes with some built-in rerankers. Here are some of the rerankers that
- [OpenAI Reranker](./openai.md)
- [Linear Combination Reranker](./linear_combination.md)
- [Jina Reranker](./jina.md)
- [AnswerDotAI Rerankers](./answerdotai.md)
- [Reciprocal Rank Fusion Reranker](./rrf.md)
## Creating Custom Rerankers