brettlyy/text-to-sql
An application to write and run SQL queries, returning answers to natural language questions, using langchain and open source LLM models through HuggingFace.
Extracts database schemas from SQLite files and uses Chroma vector embeddings to retrieve relevant schemas for context, enabling accurate SQL generation across multi-database environments. Leverages LangChain orchestration with Google's Flan-T5-XXL model via HuggingFace's free inference API, then executes generated queries and analyzes results to answer questions. Built on the Spider dataset for training semantic parsing across complex, cross-domain database structures.
No commits in the last 6 months.
Stars
42
Forks
16
Language
Jupyter Notebook
License
MIT
Category
Last pushed
Sep 13, 2023
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/vector-db/brettlyy/text-to-sql"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related tools
dsba6010-llm-applications/baemax_tc
LLM App to demystify and summarize Terms and Conditions agreements
techdomegh/ai-news-scraper
AI News Scraper & Semantic Search: A Python application that scrapes news articles, uses GenAI...
bhattbhavesh91/pdf-qa-astradb-langchain
Explore how to build a Q&A system on PDF File's using AstraDB's Vector DB with Langchain and OpenAI API's
lightfeed/sdk
Lightfeed SDK to search and filter web data
d1pankarmedhi/legal_summarizer
📃 A contracts clause summarization system using LLM and vector database