finetuning yolo 8 candle model

What is the correct way to finetune yolo8 model to be used here ? Finetuning model using candle is not straightforward.
candle\candle-examples\examples\yolo-v8\main.rs
// model model architecture points at ultralytics : https://github.com/ultralytics/ultralytics/issues/189
But my model trained using ultralytics and converted to safetensors yield tensor errors when used in candle ylo 8 example. Renaming the tensors to match the candle yolo model did not work.

I see DarkNet struct in the model.rs so I wonder if one should rather use [Darknet](https://github.com/hank-ai/darknet) instead (@LaurentMazare ) ?