LLM Powered SQL Project

Overview

• The LLM Powered SQL project leverages the power of Google PaLM to create the SQL query from natural language.
• Leverage the HuggingFace "all-MiniLM-L6-v2" model to create embeddings for SQL queries. These embeddings are stored in the ChromaDB Database, providing a powerful and efficient way to retrieve information based on semantic similarity.
• Used LangChain to develop this project.

Features

Embeddings Generation: Utilizes the HuggingFace "all-MiniLM-L6-v2" model to generate embeddings for SQL queries.

ChromaDB Database: Stores the generated embeddings in the ChromaDB Database for efficient retrieval and querying.

Streamlit Deployment: The model is deployed using Streamlit, providing a user-friendly interface for interacting with the LLM-powered SQL capabilities.