Find answers from the community

Updated 3 months ago

Hi Everyone

Hi Everyone

I still don't understand the Hybrid Search function on PgVectorStore, can anyone help me to understand how that different than regular Vector Similarity Search?

Also, if lets say I already have huge VectorIndex, if I want to enable Hybrid Search, do I need to re-ingest the wrole index?

Thank you

4 comments

TTeemu

It combines keyword + semantic search. I think you need to pass it when creating the index: https://gpt-index.readthedocs.io/en/latest/examples/vector_stores/postgres.html

RRendy Febry

Thanks @Teemu , I know that far. I'm more interested to know how it works under the hood.

TTeemu

I'm not 100% sure but I think they're running both vector + keyword search simultaneously and then using a cross-encoder to rerank those results.

TTeemu

https://github.com/pgvector/pgvector-python/blob/master/examples/hybrid_search.py

Add a reply