Automl

A comprehensive AutoML MCP Server that automates the complete machine learning workflow from data upload to model deployment.

📋 Features

• Data Management: Upload and validate CSV datasets
• Exploratory Data Analysis: Generate comprehensive data profiling reports
• Interactive Visualizations: Create plots using Plotly with matplotlib fallback
• AutoML Training: Train multiple models with automatic preprocessing
• Model Comparison: Rank models by performance metrics
• Model Export: Export trained models with metadata
• FastAPI Deployment: Generate production-ready API servers
• Professional Web Interface: Modern dashboard with FastAPI

🚀 Quick Start

Option 1: Web Interface (Recommended)

# Navigate to the automl-mcp directory
cd automl-mcp

# Install dependencies
pip install -e .[dev]

# Start the web interface
python -m automl_mcp.server web

Then open your browser to: http://127.0.0.1:8080

Option 2: Command Line Interface

# Navigate to the automl-mcp directory
cd automl-mcp

# Install dependencies
pip install -e .[dev]

# Start the MCP server
python -m automl_mcp.server run

# Or run a quick demo
python -m automl_mcp.server demo --dataset data/iris.csv --target species

Option 3: Docker

# Build and run with Docker
cd automl-mcp
docker build -t automl-mcp .
docker run -p 8080:8080 automl-mcp

🛠️ Available Commands

• python -m automl_mcp.server web - Start the web interface
• python -m automl_mcp.server run - Start the MCP server in development mode
• python -m automl_mcp.server demo - Run a complete demo pipeline
• python -m automl_mcp.server list-tools - List all available tools

📊 Tool Catalog

The MCP server provides 8 comprehensive tools:

1. upload_csv - Upload CSV datasets
2. validate_schema - Validate dataset schema and infer types
3. profile_data - Generate comprehensive data profiling reports
4. visualize - Create interactive visualizations
5. train_models - Train multiple ML models with preprocessing
6. compare_models - Compare and rank trained models
7. export_model - Export models with metadata
8. deploy_fastapi - Generate production-ready FastAPI applications

🏗️ Project Structure

Data-analysis-Agent-main/
├── automl-mcp/                    # Main AutoML MCP Server
│   ├── src/automl_mcp/
│   │   ├── server.py             # MCP server entrypoint
│   │   ├── web_interface.py      # FastAPI web interface
│   │   ├── tools/                # Tool implementations
│   │   │   ├── datasets.py       # Data management
│   │   │   ├── eda.py           # Exploratory data analysis
│   │   │   ├── viz.py           # Visualizations
│   │   │   ├── training.py      # Model training
│   │   │   ├── compare.py       # Model comparison
│   │   │   ├── export.py        # Model export
│   │   │   └── deploy.py        # FastAPI deployment
│   │   └── templates/           # Web interface templates
│   ├── tests/                   # Test suite
│   ├── docker/                  # Docker configuration
│   ├── pyproject.toml          # Project configuration
│   └── README.md               # Detailed documentation
├── data/                       # Sample datasets
├── requirements.txt            # Legacy dependencies (can be removed)
└── README.md                  # This file

🔧 Development

Installation

cd automl-mcp
python -m venv .venv
source .venv/bin/activate  # On Windows: .venv\Scripts\activate
pip install -e .[dev]

Running Tests

pytest tests/

Code Quality

ruff format src/
ruff check src/
mypy src/

📝 License

MIT License - see LICENSE file for details.

🤝 Contributing

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