model is taken from the source: Yolov8 in extensions and with support for pre/postprocessing does not detect objects.

[Pesekot1]

Hello. Please help me solve the problem.
I am trying to run an object detection application for the ONNX runtime on Android. The model in the example yolov8n_with_pre_post_processing.onnx works without problems.
But the model is taken from the source: Yolov8 in extensions and with support for pre/postprocessing does not detect objects.
What can I do wrong?

[9479421]

i am encountering the problem now, Did you have resolved it?

[Christian-lyc]

I used my trained utralytics model and convert it to onnx format. But after replace the yolov8n_with_pre_post_processing.onnx model, the code doesn't work. Do you know how to convert the normal onnx yolov8 to their one?

[anatolii20]

Hello. I did not bother to understand the Microsoft problem. I just found another, in my opinion, better option. I trained the model here https://colab.research.google.com/github/roboflow-ai/notebooks/blob/main/notebooks/train-yolov8-object-detection-on-custom-dataset.ipynb and then imported it into Android. You can see an example of the application in Google Play (it is called "What is this sign?").