chuangchuangtan

Can you release the details of training strategy for CAM, Has, ACol, CutMix, SPG and ADL? When implementing those methods, I get different results comparing with Table 3 of SPA...

Did you use 23151 or 50000 images to calculate results of Peak-T, Peak-IoU and GT-Known Loc in Table 3?

I implement CutMix twice according to [evaluation script](https://drive.google.com/drive/folders/1rd3iV9Xif2tRgofQWrL3qH1_lIFkicdI), [train script](https://github.com/clovaai/wsolevaluation) and [evaluation hyperparameter](https://docs.google.com/spreadsheets/d/1O4gu69FOOooPoTTtAEmFdfjs2K0EtFneYWQFk8rNqzw/edit#gid=0), and con't get results of Peak-T, Peak-IoU in Table 3. I achieve Peak-IoU 52.96 and 52.80, instead...

It is confusing in section 4.1 that "For fair comparisons, we adopt the same training strategy with SEM."

Thank you for your attention.