Is the IOU_1 value higher than the paper？

[deepzzz123]

hi,I am using the pretrained model in the checkpoint file folder,then I run: sh scripts/eval.sh,
the result dispaly:
acc: 0.98980 miou: 0.90326 mf1: 0.94702 iou_0: 0.98931 iou_1: 0.81721 F1_0: 0.99463 F1_1: 0.89941 precision_0: 0.99440 precision_1: 0.90329 recall_0: 0.99485 recall_1: 0.89557
It seem that the result is different with the paper,is this result accuracy or something else wrong?Appreciate getting your reply.

[Deleted user]

I want to ask the same question, in the paper any value of iou, F1, precision , recall has one value but when applying code there are two values for 0 and 1.
how to calculate the result in the paper? is it the average?

[hugo_han]

I want to ask the same question, in the paper any value of iou, F1, precision , recall has one value but when applying code there are two values for 0 and 1. how to calculate the result in the paper? is it the average?

I think these two values represent the scores of the two categories of areas without change and areas with changes, we focus on scores of areas with changes, which means focus on value XXX_1.

[Moonquakes0]

Hi, I have used the pretrained model in the checkpoint file folder, but I got poor results on the LEVIR-CD dataset. Can you tell me why this is so?

[yxzz1121]

hi,I am using the pretrained model in the checkpoint file folder,then I run: sh scripts/eval.sh,
the result dispaly:
acc: 0.98980 miou: 0.90326 mf1: 0.94702 iou_0: 0.98931 iou_1: 0.81721 F1_0: 0.99463 F1_1: 0.89941 precision_0: 0.99440 precision_1: 0.90329 recall_0: 0.99485 recall_1: 0.89557
It seem that the result is different with the paper,is this result accuracy or something else wrong?Appreciate getting your reply.
i also got the same results with you. is exactly the same