In coco.py line200 comment is wrong which is easy to misleading users.

https://github.com/Megvii-BaseDetection/YOLOX/blob/6880e3999eb5cf83037e1818ee63d589384587bd/yolox/data/datasets/coco.py#L200
I try to create a new dataset class in order to feed my own data, and I found that the following comment is wrong:

# coco.py line 200
'''
padded_labels (torch.Tensor): pre-processed label data.
    The shape is :math:`[max_labels, 5]`.
    each label consists of [class, xc, yc, w, h]:
        class (float): class index.
        xc, yc (float) : center of bbox whose values range from 0 to 1.
        w, h (float) : size of bbox whose values range from 0 to 1.
'''

the correct comment is like this:

'''
target (numpy.ndarray): pre-processed label data.
    The shape is :math:`[max_labels, 5]`.
    each label consists of [x1, y1, x2, y2, class]:
        class (float): class index.
        x1, y1 (float) : top-left points whose values range from 0 to 640
        x2, y2 (float) : right-down points whose values range from 0 to 640
'''

Open yolox/data/datasets/coco.py at line 200 and compare the existing documentation with the target array format shown in the issue. Update the comment so it describes target, [x1, y1, x2, y2, class], and the stated coordinate ranges; done when the documentation matches the actual label data.