https://github.com/robertocenteno/wrapture
Wrapture lets you go from a Python-trained model to deployable JavaScript with a single command. It generates TypeScript bindings and a Web/Node-compatible wrapper, using WebGPU/WASM-ready ONNX runtimes.
https://github.com/robertocenteno/wrapture
javascript machine-learning model model-conversion onnx pytorch quantization ruby rubygems simplifier typescript webgpu wrapture
Last synced: about 1 month ago
JSON representation
Wrapture lets you go from a Python-trained model to deployable JavaScript with a single command. It generates TypeScript bindings and a Web/Node-compatible wrapper, using WebGPU/WASM-ready ONNX runtimes.
- Host: GitHub
- URL: https://github.com/robertocenteno/wrapture
- Owner: robertocenteno
- License: mit
- Created: 2025-05-17T04:34:44.000Z (about 1 year ago)
- Default Branch: main
- Last Pushed: 2026-05-01T04:59:02.000Z (about 2 months ago)
- Last Synced: 2026-05-01T06:34:48.705Z (about 2 months ago)
- Topics: javascript, machine-learning, model, model-conversion, onnx, pytorch, quantization, ruby, rubygems, simplifier, typescript, webgpu, wrapture
- Language: TypeScript
- Size: 715 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- Changelog: CHANGELOG.md
- Contributing: CONTRIBUTING.md
- License: LICENSE
- Code of conduct: CODE_OF_CONDUCT.md
- Security: SECURITY.md
Awesome Lists containing this project
README
# 🌟 Wrapture: Seamless Model Deployment from Python to JavaScript
Welcome to the **Wrapture** repository! This project simplifies the process of deploying machine learning models trained in Python to JavaScript environments. With just a single command, you can generate TypeScript bindings and create a Web/Node-compatible wrapper using ONNX runtimes that are ready for WebGPU and WASM.
## Table of Contents
- [Features](#features)
- [Getting Started](#getting-started)
- [Installation](#installation)
- [Usage](#usage)
- [Topics](#topics)
- [Contributing](#contributing)
- [License](#license)
- [Releases](#releases)
- [Contact](#contact)
## Features
- **One Command Deployment**: Transition from Python to JavaScript effortlessly.
- **TypeScript Bindings**: Automatically generate TypeScript bindings for your models.
- **Web/Node Compatibility**: Create wrappers that work seamlessly in both web and Node.js environments.
- **WebGPU and WASM Ready**: Utilize the latest technologies for efficient model execution.
- **Support for ONNX**: Leverage the ONNX runtime for model inference.
## Getting Started
To get started with Wrapture, follow these steps:
1. **Install Dependencies**: Ensure you have the necessary tools installed.
2. **Prepare Your Model**: Train your model in Python and export it in ONNX format.
3. **Run Wrapture**: Use the command line to generate your JavaScript deployment.
## Installation
You can install Wrapture using npm. Open your terminal and run:
```bash
npm install wrapture
```
## Usage
After installing, you can deploy your model with a single command. Here’s how:
1. **Export Your Model**: Ensure your model is exported as an ONNX file.
2. **Run Wrapture**: Use the following command:
```bash
wrapture deploy path/to/your/model.onnx
```
This command will generate the necessary TypeScript bindings and wrappers.
## Topics
Wrapture covers a range of topics relevant to modern machine learning and deployment:
- JavaScript
- Machine Learning
- Model Conversion
- ONNX
- PyTorch
- Quantization
- Simplifier
- TypeScript
- WASM
- WebGPU
## Contributing
We welcome contributions! If you want to contribute to Wrapture, please follow these steps:
1. Fork the repository.
2. Create a new branch for your feature or bug fix.
3. Make your changes and commit them.
4. Push your branch to your forked repository.
5. Open a pull request.
## License
Wrapture is licensed under the MIT License. See the [LICENSE](LICENSE) file for details.
## Releases
To check for the latest releases, visit the [Releases](https://github.com/robertocenteno/wrapture/releases) section. You can download the latest version and execute it to start deploying your models.
## Contact
For any inquiries or feedback, feel free to reach out:
- **GitHub**: [Wrapture GitHub](https://github.com/robertocenteno/wrapture)
- **Email**: your-email@example.com
---
Thank you for checking out Wrapture! We hope this tool simplifies your model deployment process. For more details, visit the [Releases](https://github.com/robertocenteno/wrapture/releases) section.