Test Error

[AmingWu]

Dear Authors,

I have trained the model. But when I make a test, there is an error.

python ./tools_cam/test_cam.py --config_file ./configs/ILSVRC/deit_tscam_base_patch16_224.yaml --resume /home/user/Desktop/wam/TS-CAM/ckpt/ImageNet/deit_tscam_base_patch16_224_CAM-NORMAL_SEED26_CAM-THR0.12_BS256_2021-07-29-21-38/ckpt/model_best.pth TEST.SAVE_BOXED_IMAGE True MODEL.CAM_THR 0.12

[vasgaowei]

From your scripts, you have already trained ts-cam based on deit-base backbone, and saved corresponding model weights. You may try adding a line in test_cam.py:

from models.vgg import vgg16_cam

[AmingWu]

Thanks for your reply. I run the test code. How to see the test result?

[vasgaowei]

From your screenshot, the top1 (Cls@1) and top5 (Cls@5) classification accuracy are 68.7% and 88.9% respectively. And top1 (Loc@1) and top5 (Loc@5) localization accuracy are 47.4% and 59.4%, respectively. The Loc_gt, which means the image target label is known, is 63.9%.

[punk1]

From your screenshot, the top1 (Cls@1) and top5 (Cls@5) classification accuracy are 68.7% and 88.9% respectively. And top1 (Loc@1) and top5 (Loc@5) localization accuracy are 47.4% and 59.4%, respectively. The Loc_gt, which means the image target label is known, is 63.9%.

hello i have a question ，Shouldn't the accuracy of the final test be average?