Neural search library in Python for medium-sized corpora
https://github.com/raphaelsty/cherche
Cherche (search in French) allows you to create a neural search pipeline using retrievers and pre-trained language models as rankers. Cherche is meant to be used with small to medium sized corpora. Cherche's main strength is its ability to build diverse and end-to-end pipelines.
Comments
Interesting.
1) Why do you need both lunr and elasticsearch as dependencies?
2) You shouldn't pin the exact versions of your dependencies in setup.py. Otherwise it makes it very hard to integrate.
Hi,
1) The dependency on the Elasticsearch python client allows Elasticsearch to be used as a retriever. The same goes for Lunr. It might be interesting to separate the different dependencies.
2) Of course I'll update it.
It is better if you submit and put your text as a comment. How do you define mid sized corpora? Comparison with other solutions?
I'm more used to reading than posting on Hacker News. I'll do better next time. :)