After running python train_OAAT.py --use_defaults CIFAR100_WRN error BasicBlock

[headscott]

I got this error after the training finished:

File "OAAT/perceptual_advex/perceptual_attack_adv.py", line 65, in _make_layer
    layers.append(block(i == 0 and in_planes or out_planes, out_planes, i == 0 and stride or 1, dropRate))
TypeError: BasicBlock.__init__() takes from 3 to 4 positional arguments but 5 were given

And I think the problem is, that there are two definitions von BasicBlock in perceptual_attack_adv.py:
The one for WideResNet:

class BasicBlock(nn.Module):
    def __init__(self, in_planes, out_planes, stride, dropRate=0.0):
        super(BasicBlock, self).__init__()
        self.bn1 = nn.BatchNorm2d(in_planes)
        self.relu1 = nn.ReLU(inplace=True)
        ...

The one for ResNet:

class BasicBlock(nn.Module):
    expansion = 1

    def __init__(self, in_planes, planes, stride=1):
        super(BasicBlock, self).__init__()
        self.conv1 = nn.Conv2d(in_planes, planes, kernel_size=3, stride=stride, padding=1, bias=False)
        ...

This should be fixed by renaming one or both of them to something like WideResBasicBlock or ResBasicBlock. I'll try it myself now, and then I will tell my result.