Hugging Face's logo

Spaces:
onnx-community
/
convert-to-onnx
Running

App Files Community
6
convert-to-onnx / README.md
Felladrin's picture
Felladrin
Use the Normal HF header
9bfcf12
preview code
|
raw
history blame contribute delete
2.19 kB

A newer version of the Streamlit SDK is available: 1.41.1

Upgrade
metadata
title: Convert to ONNX
emoji: ☯
colorFrom: indigo
colorTo: yellow
sdk: streamlit
sdk_version: 1.40.1
app_file: app.py
pinned: true
license: apache-2.0
short_description: Convert a Hugging Face model to ONNX format

Convert to ONNX

Overview

This project provides a Streamlit application that converts Hugging Face models to ONNX (Open Neural Network Exchange) format, enabling broader model compatibility and deployment options. The application streamlines the process of downloading, converting, and uploading models to Hugging Face.

Features

  • One-Click Model Conversion: Convert Hugging Face models to ONNX format with minimal configuration
  • User-Friendly Interface: Intuitive Streamlit-based web interface
  • Quantization Support: Automatic model quantization for reduced size and faster inference

Prerequisites

  • Python 3.8+
  • Hugging Face account and API token
  • Git

Installation

  1. Clone the repository:

    git clone https://huggingface.co/spaces/onnx-community/convert-to-onnx
cd convert-to-onnx

  2. Install dependencies:

    pip install -r requirements.txt

  3. Configure Hugging Face credentials:

    # Option 1: Environment variables
export HF_TOKEN="your_token"

# Option 2: Create .streamlit/secrets.toml
echo 'HF_TOKEN = "your_token"' > .streamlit/secrets.toml

Usage

  1. Start the application:

    streamlit run app.py

  2. Access the web interface at http://localhost:8501

  3. Enter a Hugging Face model ID (e.g., EleutherAI/pythia-14m)

  4. Click "Proceed" to start the conversion

The converted model will be available in your Hugging Face account as {username}/{model-name}-ONNX.

Development

Contributing

  1. Fork the repository
  2. Create a feature branch
  3. Implement changes with tests
  4. Submit a pull request

Troubleshooting

Common issues and solutions:

  • Authentication Errors: Verify your Hugging Face credentials
  • Conversion Failures: Check model compatibility and available disk space
  • Upload Issues: Ensure stable internet connection and valid permissions

License

Apache 2.0 License