WIP, right now it's not a crate yet, but the code works and produces the same results as the reference python yolo 26e implementation.

getting started

• first get the onnx file, by running (working directory = root of the repo):

uv run .\py-yolo\export_onnx.py

Then run:

cargo run --bin run_onnx_v2 --profile release

[When using load-dynamic feature] ONNX Runtime Library Not Found

OnnxRuntime is dynamically loaded, so if it's not found correctly, then download the correct onnxruntime library from GitHub Releases.

Then put the dll/so/dylib location in your PATH, or point the ORT_DYLIB_PATH env var to it.

PowerShell example:

• Adjust path to where the dll is.

$env:ORT_DYLIB_PATH = "C:/Apps/onnxruntime/lib/onnxruntime.dll"

Shell example:

export ORT_DYLIB_PATH="/usr/local/lib/libonnxruntime.so"