Hugging Face's logo

Spaces:
Krippa
/
sql-books-rag
Runtime error

App Files Community
sql-books-rag / README.md
Krippa
Deploy RAG app with Groq LFS
eb22b1f
preview code
|
Raw
History Blame Contribute Delete
1.12 kB

A newer version of the Gradio SDK is available: 6.26.0

Upgrade
metadata 
title: SQL Books RAG
emoji: 📚
colorFrom: blue
colorTo: indigo
sdk: gradio
sdk_version: 6.6.0
app_file: app.py
pinned: false

📚 SQL Books RAG

A Retrieval-Augmented Generation (RAG) system that answers SQL questions using content from 5 SQL textbooks.

How It Works

  1. Retrieval – Your question is embedded using all-MiniLM-L6-v2 and matched against a FAISS index of ~500-word chunks extracted from the books.
  2. Generation – The top-5 most relevant chunks are fed as context to Llama 3.1-8B via the Groq API to produce a detailed, grounded answer.

Data Sources

Book Author
Practical SQL: A Beginner's Guide to Storytelling with Data Anthony DeBarros
SQL for Data Scientists Renee M. Teate
SQL for Data Analysis Cathy Tanimura
The Art of SQL Stéphane Faroult
Learning SQL: Generate, Manipulate, and Retrieve Data Alan Beaulieu

Tech Stack

  • Embeddings: sentence-transformers/all-MiniLM-L6-v2
  • Vector Store: FAISS (IndexFlatL2)
  • LLM: Llama 3.1-8B-Instant via Groq API (free tier)
  • UI: Gradio