Caffe2 implementation of Open Neural Network Exchange (ONNX)

Usage

• ONNX to Caffe2
• Caffe2 to ONNX
• other end-to-end tutorials

Installation

onnx-caffe2 is installed as a part of Caffe2. Please follow the instructions to install Caffe2.

Folder Structure

• ./: the main folder that all code lies under
  • frontend.py: translate from caffe2 model to onnx model
  • backend.py: execution engine that runs onnx on caffe2
• tests/: test files

Testing

onnx-caffe2 uses pytest as test driver. In order to run tests, first you need to install pytest:

pip install pytest-cov

After installing pytest, do

pytest

to run tests.

Testing coverage issues/status: https://github.com/caffe2/caffe2/blob/master/caffe2/python/onnx/ONNXOpCoverage.md

Development

During development it's convenient to install caffe2 in development mode:

cd /path/to/caffe2
pip install -e caffe2/

License

MIT License