Sungchul Kim

Results 2 issues of Sungchul Kim

I'm reproducing this paper and code and I have one question. At `model/resnet.py`, I think that bn-relu are duplicated in PreAct ResNet18. ```python def CIFAR_ResNet18(pretrained=False, **kwargs): return CIFAR_ResNet(PreActBlock, [2,2,2,2], **kwargs)...

# Description When fine-tuning seg-preview model, `num_select` seems to be set to 200, but it is set to 300 like other detection models. It is due to that this parameter...