Text to SQL Converter

AI/ML powered application that converts natural language text to SQL queries.

📸 Screenshots

Welcome to Text to SQL Converter

Convert natural language to SQL queries with ease

Database schema visualization

Real-time SQL generation and execution

Manage your query sessions

🎥 Demo Video

Watch the Text to SQL converter in action

Tech Stack

• Frontend: React
• Backend: FastAPI
• Database: MySQL, PostgreSQL & SQLite (user-provided)
• AI/ML: Natural language processing for SQL generation
• Model: mradermacher/natural-sql-7b-i1-GGUF

✨ Features

• Convert plain English to SQL queries
• Real-time query generation
• Clean, intuitive interface
• Multi-database support
• One-click sample database loading

🚀 Deployment

Quick Start

Development

# Start development environment
docker-compose up --build

Production

# Run deployment script (recommended)
./deploy.sh

# Or use make commands
make prod  # Docker deployment
make pm2-start  # PM2 deployment

Access the application at http://localhost:3000

Deployment Options

🐳 Docker (Recommended)

Advantages: Isolated environment, easy scaling

# Production
make prod

# View logs
make logs

# Stop services
make stop

Access:

• Frontend: http://localhost:3000
• Backend API: http://localhost:3000/api (via Nginx proxy)
• Database: Connect to your own PostgreSQL/MySQL/SQLite database

⚡ PM2 Process Manager

Advantages: Process management, auto-restart, load balancing

# Prerequisites
npm install -g pm2 serve pnpm

# Deploy
make pm2-start

# Manage
pm2 logs
pm2 status

🔧 Manual Setup

# Backend
cd backend
pip install -r requirements.txt
python install_llama.py

# Frontend
cd frontend
pnpm install
pnpm build

# Start services
cd backend && uvicorn app.main:app --host 0.0.0.0 --port 8000 &
cd frontend && serve -s dist -l 3000 &

Environment Configuration

Production Variables

# Backend
ENV=production
PYTHONPATH=/app

# Database (if external)
DATABASE_URL=postgresql://user:pass@host:5432/dbname

Database Setup

The application supports multiple database types:

• Sample Database: Click "Load Sample Database" for instant demo with e-commerce data
• PostgreSQL: Create a database and user with appropriate permissions
• MySQL: Create a database and user with appropriate permissions
• SQLite: Upload your own .db file through the interface

Example PostgreSQL setup:

CREATE DATABASE textosql;
CREATE USER texttosql WITH PASSWORD 'your_password';
GRANT ALL PRIVILEGES ON DATABASE texttosql TO texttosql;

Monitoring & Logs

Docker

# View logs
docker-compose -f docker-compose.prod.yml logs -f

# Check status
docker-compose -f docker-compose.prod.yml ps

PM2

# View logs
pm2 logs

# Monitor
pm2 monit

Troubleshooting

Common Issues:

1. Port conflicts: Change ports in docker-compose.prod.yml
2. Memory issues: Increase Docker memory limits
3. Model loading: Ensure sufficient disk space (4GB+ for AI model)

Health Checks:

# Frontend
curl http://localhost:3000

# Backend API
curl http://localhost:3000/api/health

Database Connection

The application supports multiple database types:

• Connect via credentials (PostgreSQL/MySQL)
• Upload SQLite files
• Real-time schema detection and query generation

Architecture

• Frontend: React app with Nginx (port 3000)
• Backend: FastAPI on internal Docker network
• Database: User-provided (PostgreSQL/MySQL/SQLite)
• AI Model: 4GB cached in persistent volume
• Security: Backend only accessible via proxy