Darknet/yolo what is the best onnx runtime to use?

1

Using darknet/yolo: CCodeRun/darknet: Darknet/YOLO object detection framework - Codeberg.org

I have trained my model and I have converted my model to onnx format.

I want to use rust with the onnx model however not sure what to use? Would something like burn ai be most suitable or something else?

2

What do you mean by Rust-based? The framework should be written in Rust or are you only interested in Rust bindings so that you can write your code in Rust? I'd recommend looking at Burn, Candle and the Rust Torch bindings and see what you like / performs best.

3

Rust with the onnx model

4

Burn, candle and torch all support the conversion of ONNX models to their native representations. Burn and candle are written in Rust. Torch—arguably the most widely used deep learning framework in the world—is a C++ library with a Rust wrapper.

5

Regardless if I used ultralytics, darknet etc, if I exported my model to onnx, would it be automatically compatible with Burn or Candle etc?

6

I've looked at the docs a bit more and looks like Candle can only run inference on ONNX models. Burn has a converter that converts ONNX models to their native tensor format. I've not read anywhere that either only supports a subset of the ONNX format.

7

So if I exported my model to onnx from darknet it should then work on either candle or burn (obviosuly I convert it to native for burn)?

8

Yes.

9

Which one is better to use then? What are the main differences?