Closed mnslarcher closed 4 years ago
Hi,
thank you for your raising this issue. I think the code is correct but the comment is wrong. When implementing, I assumed the labels is a tensor with size (M, 5) consisting of class, x1, y1, x2, y2.
However, in the comment it says x1, y1, x2, y2, class. So, I should change the comment. It will be updated in 1-2 days.
Thank you.
Hi @kaanakan,
From https://github.com/kaanakan/object_detection_confusion_matrix/blob/fe9cd4a54f5bf5ef592467af83f9073a6a2155cc/confusion_matrix.py#L52
I see that you are using the first column as classes but from your description: https://github.com/kaanakan/object_detection_confusion_matrix/blob/fe9cd4a54f5bf5ef592467af83f9073a6a2155cc/confusion_matrix.py#L47
classes are in the last column (-1), right?