Yolo .onnx can not be handle

[timimm]

Hi guys. I wonder if you have the situation that executing ezkl gen-setting -M yolov8n-face.onnx with a yolo onnx model file. the Error was:

Can anyone help me? I want to prove a humanbeing check model with the yolov8n-face model.....

[alexander-camuto]

what version of ezkl are you using ?

[timimm]

Both the CLI and pip package are the latest, v11.0.14.

[alexander-camuto]

can you post a link to the model here :)

[timimm]

Yep, certainly!
https://github.com/akanametov/yolov8-face

It needs a transformation, here is the code:

from ultralytics import YOLO

# Load a model
model = YOLO(r"E:\download\yolov8n_100e.pt")  # load an official model
# model = YOLO('path/to/best.pt')  # load a custom trained

# Use the model
success = model.export(format="onnx",simplify=True,opset=12)

[timimm]

So, is this a bug or not? Anyone help plz :)

[alexander-camuto]

This should be patched in part now