yolov8n.yaml keeps getting downloaded regradless of the given model

[shayandaneshvar]

Hi,
I use the following command to train a custom dataset on the pre-trained Yolov8 with yolov8s weights, but the yolov8n is downloaded every time and I am guessing that the input for the model is being ignored.
Am I doing it wrong?

yolo detect train data={DATASET_YML} model=yolov8s.yaml pretrained=True epochs={epochs} imgsz={image_size} verbose=True

I also wanted to know what parameters should I set, so that I can train yolov8s similarly to yolov5s so that the same IoU and augmentation are used.

Also, what should I set to make the model run on GPU? The default value for the device param is null but it is missing from the document what will be chosen in the default mode.

[glenn-jocher]

@shayandaneshvar for YOLOv8-related issues please raise an issue on https://github.com/ultralytics/ultralytics