🦜 LangChain: Chat with SQL Database

Made by Harsh Gidwani

🚀 Live Demo Available Here - Try the application without any setup!

A powerful Streamlit application that allows you to chat with SQL databases using natural language queries powered by LangChain and Groq's Llama model.

Features

• Multiple Database Support:
  • SQLite Student Database
  • SQLite Customer Transactions Database (100+ records)
  • MySQL Database Connection
• Natural Language Queries: Ask questions in plain English
• Real-time Chat Interface: Interactive chat experience with streaming responses
• Secure API Key Handling: Protected API key input
• Database Schema Discovery: Automatic table and column detection
• Transaction History: Maintains chat history with clear option

Database Options

1. Student Database (student.db)

Pre-built SQLite database containing student information.

2. Customer Transactions Database (customer_transactions.db)

Contains 100+ realistic customer transaction records with:

• 50 unique customers with complete profiles
• 100 transactions across multiple categories
• Various payment methods and transaction types
• Realistic date ranges and amounts

3. MySQL Database

Connect to your own MySQL database by providing:

• Host address
• Username and password
• Database name

Installation

1. Clone the repository

   git clone <repository-url>
   cd "6-Chat SQL"

2. Install dependencies

   pip install streamlit langchain langchain-groq sqlalchemy sqlite3 mysql-connector-python faker pathlib

3. Generate Customer Database (if not already present)

   python create_customer_db.py

4. Get Groq API Key

   • Sign up at Groq Console
   • Generate an API key
   • Keep it secure for use in the application

Usage

1. Start the application

   streamlit run app.py

2. Configure the application

   • Select your preferred database from the sidebar
   • Enter your Groq API key
   • For MySQL: provide connection details

3. Start chatting

   • Ask questions like:
     • "Show me all students with grades above 90"
     • "What's the total sales amount by category?"
     • "Which customers made the most purchases?"
     • "Show me all transactions from last month"

Example Queries

For Student Database:

• "How many students are in the database?"
• "Show me students from a specific city"
• "What's the average age of students?"

For Customer Transactions Database:

• "Show me all transactions above $500"
• "Which payment method is most popular?"
• "What's the total revenue by product category?"
• "Show me all refunded transactions"
• "Who are the top 5 customers by transaction amount?"

For Custom MySQL Database:

• Ask any question relevant to your database schema

Project Structure

6-Chat SQL/
├── app.py                          # Main Streamlit application
├── create_customer_db.py           # Script to generate customer database
├── student.db                      # Student database (SQLite)
├── customer_transactions.db        # Customer transactions database (SQLite)
├── assets/
│   └── image.png                   # Demo screenshot
├── .gitignore                      # Git ignore file
└── README.md                       # This file

Technologies Used

• Streamlit: Web application framework
• LangChain: LLM application framework
• Groq: Fast LLM inference
• SQLAlchemy: Database toolkit
• SQLite: Embedded database
• MySQL: Relational database
• Faker: Generate realistic test data

Configuration

Environment Variables

You can set your Groq API key as an environment variable:

export GROQ_API_KEY=your_api_key_here

Database Files

• Ensure student.db and customer_transactions.db are in the project root
• For MySQL connections, ensure the database server is accessible

Troubleshooting

Common Issues

1. API Key Error

   • Ensure you've entered a valid Groq API key
   • Check if the API key has proper permissions

2. Database Connection Failed

   • Verify database files exist in the correct location
   • For MySQL: check connection details and network access

3. Query Errors

   • The AI might need clarification for complex queries
   • Try rephrasing your question
   • Check if the requested data exists in the database

Error Messages

• The api_key client option must be set: Enter your Groq API key
• Please provide all MySQL connection details: Fill in all MySQL fields
• Database connection errors: Check file paths and permissions

Contributing

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Test thoroughly
5. Submit a pull request

License

This project is open source and available under the MIT License.

Support

If you encounter any issues or have questions:

1. Check the troubleshooting section
2. Review the error messages carefully
3. Ensure all dependencies are installed correctly
4. Verify your API key is valid

Future Enhancements

• Support for PostgreSQL databases
• Query result visualization
• Export chat history
• Multiple AI model options
• Database schema visualization
• Query performance metrics

---

Made by Harsh Gidwani