saymrwulf/pytorch
1
0
Fork 0
mirror of https://github.com/saymrwulf/pytorch.git synced 2026-05-15 21:00:47 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions
pytorch/caffe2/python/onnx
History
Yinghai Lu 4be4b4c8b5 Remove weight from input of onnxifi backend op (#10575) 
Summary:
The ONNXIFI backend will absorb the constant weight in Conv, so we should not add it as an input. This is just a test artifacts. Note that Onnxifi transformer will do the right thing when cutting the graph to absorb the weights.

rdzhabarov
Pull Request resolved: https://github.com/pytorch/pytorch/pull/10575

Reviewed By: houseroad

Differential Revision: D9357339

Pulled By: yinghai

fbshipit-source-id: a613fa3acafa687295312f5211f8e9d7f77b39cd
2018-08-16 10:56:25 -07:00
..
bin Remove Apache headers from source. 2018-03-27 13:10:18 -07:00
tests Pass shape infos to ONNX -> Caffe2 C++ conversion backend (#9870) 2018-07-26 12:00:32 -07:00
__init__.py
backend.py Add aten_op to caffe2 onnx (python) backend 2018-08-16 00:39:30 -07:00
backend_cpp_rep.py Remove Apache headers from source. 2018-03-27 13:10:18 -07:00
backend_rep.py Disallow using the OOP api workspace as context managers (#6456) 2018-04-09 22:13:54 -07:00
error.py Remove Apache headers from source. 2018-03-27 13:10:18 -07:00
frontend.py Bump up the C2 onnx frontend opset to 8 (#9006) 2018-06-29 11:56:11 -07:00
helper.py [Caffe2] Scoped dummy name generator (#6458) 2018-04-16 11:58:02 -07:00
onnxifi.py ONNXIFI transform (#9569) 2018-07-20 15:09:59 -07:00
ONNXOpCoverage.md Update the ONNX op coverage in C2 2018-06-29 17:25:19 -07:00
README.md Add README and ONNXOpCoverage doc back (#2102) 2018-03-01 17:05:25 -08:00
test_onnxifi.py Remove weight from input of onnxifi backend op (#10575) 2018-08-16 10:56:25 -07:00
workspace.py Disallow using the OOP api workspace as context managers (#6456) 2018-04-09 22:13:54 -07:00

README.md

Caffe2 implementation of Open Neural Network Exchange (ONNX)

Usage

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