What would be the label format?

[SeungyounShin]

Training fails with multiples of image size and normalized label format with [cx,cy,w,h]

Do I have to resize the images before training?

[knoppmyth]

No, you shouldn't have to resize the images. You should post any error you're getting.

[glenn-jocher]

@knoppmyth for YOLOv8, the label format should be in the normalized [class, cx, cy, w, h] format, where cx, cy, w, and h are the center coordinates, width, and height of the bounding box, all normalized between 0 and 1 relative to the image size. The training process will automatically handle image resizing for you.

If you're encountering issues, please ensure your labels are correctly formatted and check your dataset's integrity. If the problem persists, feel free to share the error message you're receiving for further assistance. 🛠️