LLM-Based SQL Query Generator

Project Overview

This project is an AI-powered SQL Query Generator that translates natural language queries into SQL statements using GPT-4. It helps users generate, validate, and execute SQL queries against a specified database schema.

User Interface

Features

• Natural Language to SQL: Uses OpenAI's GPT-4 to generate SQL queries.
• Query Validation: Ensures generated queries are syntactically correct.
• Database Execution: Runs SQL queries against an SQLite database.
• Custom Schema Support: GPT generates SQL based on your database schema.
• User-Friendly UI: Built with Streamlit for ease of use.
• Secure API Key Management: Users can input their OpenAI API Key manually or load it from a .env file.

Tech Stack

• Backend: Python, OpenAI API, SQLite
• Frontend: Streamlit
• Libraries: SQL Parsing, python-dotenv

🚀 Installation & Setup

1️. Clone the Repository

git clone https://github.com/ghaihitasha/llm-sql-query-generator.git
cd llm-sql-query-generator

2. Install Dependencies

pip install -r requirements.txt

3. Set Up the .env File

Create a .env file in the root directory and add your OpenAI API Key:

OPENAI_API_KEY=your_openai_api_key_here

4. Run the Streamlit App

streamlit run src/app.py

Usage

1. Enter your OpenAI API Key in the Project Settings section or let it load from .env.
2. Provide the database path to your SQLite file.
3. Enter a natural language query (e.g., "Show all employees who joined after 2020").
4. Click Generate SQL to get the SQL query.
5. Validate & Execute the query to see the results.

Contributing

Pull requests are welcome! If you’d like to contribute, please open an issue first to discuss your changes.