img, boxes, labels = self.transform(img, target[:, :4], target[:, 4]) IndexError: too many indices for array

[SpringUp]

[neverstoplearn]

i meet the same error have you solve it?

[SpringUp]

I remember it seems that there is a problem with the annotation of an original image,you can try to print the image index to determine which image has the problem.

[GuoJingWeng]

If the training test has no objects, then the negative samples are not labeled, showing how to resolve this error

[ZQ-bit]

I checked the index when IndexError was raised, and found out that it's parameter #difficult#'s fault in my case. If you click "difficult" when labeling your data, you'll have difficult = int(obj.find('difficult').text) == 1 True, and it turns out " res = []"