https://github.com/fluxml/onnx.jl

Read ONNX graphs in Julia
https://github.com/fluxml/onnx.jl

deep-learning flux julia machine-learning onnx

Last synced: 12 days ago
JSON representation

Read ONNX graphs in Julia

• Host: GitHub
• URL: https://github.com/fluxml/onnx.jl
• Owner: FluxML
• License: other
• Created: 2018-02-09T13:53:32.000Z (about 7 years ago)
• Default Branch: master
• Last Pushed: 2025-03-03T15:13:15.000Z (about 2 months ago)
• Last Synced: 2025-04-03T23:09:04.774Z (20 days ago)
• Topics: deep-learning, flux, julia, machine-learning, onnx
• Language: Julia
• Size: 8.79 MB
• Stars: 144
• Watchers: 16
• Forks: 26
• Open Issues: 28
• Metadata Files:
  • Readme: README.md
  • Funding: .github/FUNDING.yml
  • License: LICENSE.md

Awesome Lists containing this project

README

        
# ONNX

ONNX.jl is in the process of a total reconstruction and currently supports saving & loading graphs as a [`Umlaut.Tape`](https://dfdx.github.io/Umlaut.jl/dev/reference/#Umlaut.Tape). When possible, functions from `NNlib` or standard library are used, but no conversion to Flux is implemented yet. See [resnet18.jl](examples/resnet18.jl) for a practical example of graph loading.

Not all ONNX operators are implemented. See [How to contribute](CONTRIBUTE.md) for details of adding new operators.