How to evaluate retrieval in RAG pipelines

In this video, we'll learn how to use a library called ranx to evaluate the quality of the search results returned in a RAG pipeline. We'll compare Full-Text Search, Vector Search, and Hybrid Search, all using DuckDB.
#duckdb #rag
Code - github.com/mne...
Ranx - amenra.github....
llama.cpp - llama-cpp-pyth...

[@user-ce7vu3ct3y]

I wish to use Qdrant vector store, can you give some guidance on how to switch from duckdb to qdrant for your given example here?

[@learndatawithmark]

If you look at this video we use Qdrant - ruclips.net/video/gigip1Pxf88/видео.html
Let me know if you need more than that and maybe I can do another video.