ml-service-framework 0.1.0

A production-ready machine learning framework

MLOps-Boilerplate

A production-ready machine learning framework for building, training, deploying, and monitoring ML models at scale.

🚀 Features

• Modular Architecture: Clean separation of concerns with data layer, ML components, and applications
• Multiple Data Sources: Built-in connectors for PostgreSQL, MongoDB, AWS S3, Azure Blob Storage
• ML Models: Support for Random Forest, XGBoost, and easy extensibility for custom models
• Data Processing: Complete preprocessing pipeline with feature engineering and scaling
• Cross-Validation: K-Fold and Stratified K-Fold validation with comprehensive metrics
• Experiment Tracking: Integration with MLflow for experiment management
• Configuration Management: Pydantic-based configuration with validation
• Testing: Comprehensive test suite with pytest
• Monitoring: Model performance tracking and data drift detection
• API Server: FastAPI-based REST API for model serving
• Production Ready: Docker support, CI/CD pipelines, and best practices

📋 Table of Contents

• Installation
• Quick Start
• Project Structure
• Usage Examples
• Configuration
• Testing
• API Documentation
• Contributing
• License

🔧 Installation

Install from PyPI (Recommended)

pip install mlops-boilerplate

Create a New Project

After installation, create a new ML project using the template:

# Create a new project
ml-create-project my-ml-project

# Navigate to your project
cd my-ml-project

# Set up virtual environment
python -m venv venv
source venv/bin/activate  # On Windows: venv\\Scripts\\activate

# Install dependencies
pip install -r requirements.txt

# Configure environment
cp .env.example .env
# Edit .env with your configuration

Install from Source (For Development)

# Clone the repository
git clone https://github.com/kython220282/MLOps-Boilerplate.git
cd MLOps-Boilerplate

# Create virtual environment
python -m venv venv
source venv/bin/activate  # On Windows: venv\\Scripts\\activate

# Install in development mode
pip install -e .

Install with Docker

docker build -t mlops-boilerplate .
docker run -p 8000:8000 mlops-boilerplate

🏃 Quick Start

1. Setup Environment

# Copy environment template
cp .env.example .env

# Edit .env with your configuration

2. Train a Model

# Using CLI
ml-train --config config/training_config.json

# Or with Python
python -m ml_service.applications.training --config config/training_config.json

3. Run Inference

ml-inference --model-path models/model.joblib \\
             --input-path data/test.csv \\
             --output-path predictions.csv

4. Start API Server

ml-serve --model-path models/model.joblib --port 8000

📁 Project Structure

machine_learning_service/
├── ml_service/                 # Main package
│   ├── applications/          # Application entry points
│   │   ├── training.py       # Training CLI application
│   │   └── inference.py      # Inference CLI application
│   ├── data_layer/           # Data connectors
│   │   ├── data_connector.py # Database connectors
│   │   └── object_connector.py # Cloud storage connectors
│   ├── machine_learning/     # ML components
│   │   ├── data_processor.py # Data preprocessing
│   │   ├── model.py          # Model definitions
│   │   ├── training_pipeline.py # Training orchestration
│   │   └── cross_validator.py # Model validation
│   └── config.py             # Configuration management
├── config/                    # Configuration files
│   ├── training_config.json
│   └── training_config.yaml
├── tests/                     # Test suite
├── docs/                      # Documentation
├── requirements.txt          # Dependencies
├── setup.py                  # Package setup
├── pyproject.toml           # Project configuration
├── .env.example             # Environment template
└── README.md                # This file

💡 Usage Examples

Training with Different Data Sources

From CSV File:

from ml_service.machine_learning.training_pipeline import TrainingPipeline

config = {
    "data_source": {"type": "file", "path": "data/train.csv"},
    "target_column": "target",
    "model": {"type": "random_forest", "n_estimators": 100},
    "task_type": "classification"
}

pipeline = TrainingPipeline(config)
metrics = pipeline.run_pipeline()

From Database:

config = {
    "data_source": {
        "type": "database",
        "connector_type": "postgresql",
        "connection_config": {
            "host": "localhost",
            "database": "ml_db"
        },
        "query": "SELECT * FROM training_data"
    },
    "target_column": "target"
}

Custom Model Development

from ml_service.machine_learning.model import BaseModel

class CustomModel(BaseModel):
    def build_model(self):
        # Your model architecture
        pass
    
    def train(self, X_train, y_train):
        # Training logic
        pass
    
    def predict(self, X):
        # Prediction logic
        pass

Data Processing Pipeline

from ml_service.machine_learning.data_processor import DataProcessor

processor = DataProcessor()
df = processor.load_data("data/train.csv")
X_train, X_test, y_train, y_test = processor.preprocess_pipeline(
    df, target_column="target"
)

⚙️ Configuration

Training Configuration

Create a JSON or YAML configuration file:

{
  "data_source": {
    "type": "file",
    "path": "data/train.csv"
  },
  "target_column": "target",
  "model": {
    "type": "random_forest",
    "n_estimators": 100,
    "max_depth": 10
  },
  "data_processing": {
    "missing_value_strategy": "mean",
    "scaling_method": "standard"
  },
  "cross_validation": {
    "type": "kfold",
    "n_splits": 5
  }
}

Environment Variables

Set in .env file:

# Database
DB_TYPE=postgresql
DB_HOST=localhost
DB_PORT=5432

# MLflow
MLFLOW_TRACKING_URI=http://localhost:5000

# API
API_HOST=0.0.0.0
API_PORT=8000

🧪 Testing

Run the test suite:

# Run all tests
pytest

# Run with coverage
pytest --cov=ml_service --cov-report=html

# Run specific test file
pytest tests/test_model.py

# Run with markers
pytest -m unit
pytest -m integration

📚 API Documentation

Start the API server and visit http://localhost:8000/docs for interactive Swagger documentation.

Example API Endpoints

Health Check:

curl http://localhost:8000/health

Make Prediction:

curl -X POST http://localhost:8000/predict \\
  -H "Content-Type: application/json" \\
  -d '{"features": [1.5, 2.3, 3.1, 4.2, 5.0]}'

🤝 Contributing

Contributions are welcome! Please follow these steps:

1. Fork the repository
2. Create a feature branch (git checkout -b feature/amazing-feature)
3. Commit your changes (git commit -m 'Add amazing feature')
4. Push to the branch (git push origin feature/amazing-feature)
5. Open a Pull Request

Development Setup

# Install dev dependencies
pip install -r requirements.txt
pip install -e ".[dev]"

# Install pre-commit hooks
pre-commit install

# Run code formatting
black ml_service tests
isort ml_service tests

# Run linting
flake8 ml_service tests
mypy ml_service

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

🙏 Acknowledgments

• scikit-learn for ML algorithms
• MLflow for experiment tracking
• FastAPI for API framework
• Pydantic for configuration management

�‍💻 Credits

Created by: Karan
GitHub: @kython220282
Repository: MLOps-Boilerplate

🌟 If You Use This Framework

If you use this framework in your projects, please consider:

• ⭐ Star this repository on GitHub
• 📝 Add credits in your project documentation:

  Built with [MLOps-Boilerplate](https://github.com/kython220282/MLOps-Boilerplate) by Karan
  

• 🔗 Link back to this repository
• 💬 Share your project - Open an issue to showcase what you've built!

Your support helps maintain and improve this framework for everyone. Thank you! 🙏

📞 Support

For questions and support:

• Create an issue on GitHub
• Email: [e-mail]
• Repository: https://github.com/kython220282/MLOps-Boilerplate

---

Happy Model Building! 🎉