https://github.com/chainer/onnx-chainer

Add-on package for ONNX format support in Chainer
https://github.com/chainer/onnx-chainer

caffe chainer deep-learning onnx onnx-chainer onnx-format onnx-support

Last synced: 10 months ago
JSON representation

Add-on package for ONNX format support in Chainer

• Host: GitHub
• URL: https://github.com/chainer/onnx-chainer
• Owner: chainer
• License: mit
• Created: 2017-11-21T05:23:33.000Z (over 8 years ago)
• Default Branch: master
• Last Pushed: 2019-11-06T09:38:25.000Z (over 6 years ago)
• Last Synced: 2025-07-14T20:23:42.974Z (12 months ago)
• Topics: caffe, chainer, deep-learning, onnx, onnx-chainer, onnx-format, onnx-support
• Language: Python
• Homepage:
• Size: 1.18 MB
• Stars: 85
• Watchers: 16
• Forks: 24
• Open Issues: 13
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# ONNX-Chainer

[![PyPI](https://img.shields.io/pypi/v/onnx-chainer.svg)](https://pypi.org/project/onnx-chainer/)

[![GitHub license](https://img.shields.io/github/license/chainer/onnx-chainer.svg)](https://github.com/chainer/onnx-chainer)

[![Build Status](https://travis-ci.org/chainer/onnx-chainer.svg?branch=master)](https://travis-ci.org/chainer/onnx-chainer)

[![codecov](https://codecov.io/gh/chainer/onnx-chainer/branch/master/graph/badge.svg)](https://codecov.io/gh/chainer/onnx-chainer)

[![Documentation Status](https://readthedocs.org/projects/onnx-chainer/badge/?version=latest)](https://onnx-chainer.readthedocs.io/en/latest/?badge=latest)

All code and functionalities of ONNX-Chainer have been merged into [Chainer](https://chainer.org/) v7rc1 and this repository supports only bug fixes.

This is an add-on package for ONNX support by Chainer.

## Tested environment

see [Tested environments](https://onnx-chainer.readthedocs.io/en/latest/introduction/index.html#tested-environments)

## Installation

```bash

pip install onnx-chainer

```

## Run Test

### 1. Install test modules

```bash

$ pip install onnx-chainer[test-cpu]

```

Or, on GPU environment

```bash

$ pip install cupy  # or cupy-cudaXX is useful

$ pip install onnx-chainer[test-gpu]

```

### 2. Run tests

```bash

$ pytest -m "not gpu"

```

Or, on GPU environment

```bash

$ pytest

```

## Quick Start

First, install [ChainerCV](https://github.com/chainer/chainercv) to get the pre-trained models.

```python

import numpy as np

import chainer

import chainercv.links as C

import onnx_chainer

model = C.VGG16(pretrained_model='imagenet')

# Pseudo input

x = np.zeros((1, 3, 224, 224), dtype=np.float32)

onnx_chainer.export(model, x, filename='vgg16.onnx')

```

## Supported Functions

see [Supported Functions](https://onnx-chainer.readthedocs.io/en/latest/introduction/index.html#supported-functions)

## Contribution

Any contribution to ONNX-Chainer is welcome!

- Python codes follow [Chainer Coding Guidelines](https://docs.chainer.org/en/stable/contribution.html#coding-guidelines)