AnswerDotAI / AnswerDotAI/rerankers
Allow arbitrary "document" types
Open
enhancement
- Dominant language
- Python
- Stars
- 1.6k
- Forks
- 104
- PR merge metrics
- No merged PRs in 30d
Description
Just a quick thought: Why restrict the basic re-ranking interface to just strings?
For example, one might want to re-rank images and there might be models that support that.
I believe, the two main model classes of this library, `Result` and `RankedResults`, can pretty easily be relaxed to support generic document types (i.e., by extending `Generic[T]` and swapping `str` for `T`).
Contributor guide
No contributing guide indexed for this repository
Assessment
This issue has not been assessed yet.