██╗  ██╗ ██████╗ ██████╗  ██████╗  ██████╗ ███████╗████████╗    ███╗   ██╗ ██████╗ ██████╗ ███████╗
╚██╗██╔╝██╔════╝ ██╔══██╗██╔═══██╗██╔═══██╗██╔════╝╚══██╔══╝    ████╗  ██║██╔═══██╗██╔══██╗██╔════╝
 ╚███╔╝ ██║  ███╗██████╔╝██║   ██║██║   ██║███████╗   ██║       ██╔██╗ ██║██║   ██║██║  ██║█████╗  
 ██╔██╗ ██║   ██║██╔══██╗██║   ██║██║   ██║╚════██║   ██║       ██║╚██╗██║██║   ██║██║  ██║██╔══╝  
██╔╝ ██╗╚██████╔╝██████╔╝╚██████╔╝╚██████╔╝███████║   ██║       ██║ ╚████║╚██████╔╝██████╔╝███████╗
╚═╝  ╚═╝ ╚═════╝ ╚═════╝  ╚═════╝  ╚═════╝ ╚══════╝   ╚═╝       ╚═╝  ╚═══╝ ╚═════╝ ╚═════╝ ╚══════╝

XGBoost Node

Fast, native Node.js bindings for XGBoost (Linux & MAC)

Features

• 🚀 Native C++ bindings for maximum performance
• 🧵 Multi-threaded prediction support
• 🔄 Async/Promise-based API
• 💪 Type definitions included
• 🐧 Linux support (more platforms coming soon)

Prerequisites

• Linux OS or Mac OS
• Node.js >= 14.0.0
• python 3
• GCC/G++ compiler

Sometimes you will need to install these python packages if not present on you system.

• setuptools -- Builds/install Python packages, especially ones with C extensions
• distutils -- Legacy build helper, still assumed by some packages
• libomp -- Enables OpenMP support for multithreaded C++ libraries like XGBoost

Installation

npm install xgboost_node

Quick Start

import xgboost from 'xgboost_node';

// Training example
const features = [
[1200, 8, 10, 0, 1, 1], // example data could be housing or flight
[800, 14, 15, 1, 2, 0],
[950, 10, 12, 1, 1, 0],
[1000, 9, 11, 0, 0, 1],
[1100, 13, 14, 0, 2, 1],
];
const labels = [250, 180];  // Prices

const params = {
    max_depth: 3,
    eta: 0.1,
    objective: 'reg:squarederror',
    eval_metric: 'rmse'
};

async function main() {
    // Train model
    await xgboost.train(features, labels, params);
    
    // Save the trained model
    await xgboost.saveModel('model.xgb');
    
    // Load model for predictions
    await xgboost.loadModel('model.xgb');
    
    // Make predictions
    const predictions = await xgboost.predict([[1300, 9, 11, 0, 1, 1]]);
    console.log('Predicted price:', predictions[0]);
}

main().catch(console.error);

API Reference

train(features: number[][], labels: number[], params: object): Promise

Trains an XGBoost model with the provided features and labels.

Parameters:

• features: 2D array of training features
• labels: Array of training labels
• params: XGBoost parameters object

predict(features: number[][]): Promise<number[]>

Makes predictions using the trained model.

Parameters:

• features: 2D array of features to predict

Returns:

• Array of predictions

saveModel(path: string): Promise

Saves the trained model to disk.

Parameters:

• path: File path to save the model

loadModel(path: string): Promise

Loads a trained model from disk.

Parameters:

• path: Path to the saved model file

Building from Source

1. Clone the repository:

git clone https://github.com/yourusername/xgboost-node.git

2. Install dependencies:

npm install

3. Build the native module:

npm run build

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

MIT

Disclaimer

This software is provided "as is", without warranty of any kind, express or implied, including but not limited to the warranties of merchantability, fitness for a particular purpose, and noninfringement. In no event shall the authors or copyright holders be liable for any claim, damages, or other liability, whether in an action of contract, tort, or otherwise, arising from, out of, or in connection with the software or the use or other dealings in the software.

Acknowledgments

• XGBoost team for the amazing gradient boosting library
• N-API team for the native addon interface