def __init__(self, depth, num_classes, is_imagenet):
super(ResNet_cutmix, self).__init__()
self.in_planes = 64
self.is_imagenet = is_imagenet
block, num_blocks = cfg(depth)
self.cutmix = CutMix()
self.conv1 = conv3x3(3, 64)
self.bn1 = nn.BatchNorm2d(64)
self.layer1 = self._make_layer(block, 64, num_blocks[0], stride=1)
self.layer2 = self._make_layer(block, 128, num_blocks[1], stride=2)
self.layer3 = self._make_layer(block, 256, num_blocks[2], stride=2)
self.layer4 = self._make_layer(block, 512, num_blocks[3], stride=2)
self.linear = nn.Linear(512 * block.expansion, num_classes)
def __init__(self, depth, num_classes, dropout_rate, is_immagenet):
super(ResNet_att, self).__init__()
self.in_planes = 64
self.num_classes = num_classes
self.drop_rate = dropout_rate
self.is_imagenet = is_immagenet
block, num_blocks = cfg(depth)
self.conv1 = conv3x3(3, 64)
self.bn1 = nn.BatchNorm2d(64)
self.layer1 = self._make_layer(block, 64, num_blocks[0], stride=1)
self.layer2 = self._make_layer(block, 128, num_blocks[1], stride=2)
self.layer3 = self._make_layer(block, 256, num_blocks[2], stride=2)
self.layer4 = self._make_layer(block, 512, num_blocks[3], stride=2)
self.linear = nn.Linear(512 * block.expansion, num_classes)
self.dropout = CAMDropout(self.linear.weight,
self.linear.bias,
self.drop_rate,
device=torch.device('cuda'))
def __init__(self,
depth,
widen_factor,
conv_bias,
dropout_rate,
num_classes,
is_imagenet=False):
super(Wide_ResNet, self).__init__()
self.in_planes = 16
self.is_imagenet = is_imagenet
assert ((depth - 4) % 6 == 0), 'Wide-resnet depth should be 6n+4'
n = (depth - 4) / 6
k = widen_factor
print('| Wide-Resnet %dx%d' % (depth, k))
nStages = [16, 16 * k, 32 * k, 64 * k]
self.conv1 = conv3x3(3, nStages[0], stride=1, bias=conv_bias)
self.layer1 = self._wide_layer(wide_basic,
nStages[1],
n,
dropout_rate,
stride=1,
conv_bias=conv_bias)
self.layer2 = self._wide_layer(wide_basic,
nStages[2],
n,
dropout_rate,
stride=2,
conv_bias=conv_bias)
self.layer3 = self._wide_layer(wide_basic,
nStages[3],
n,
dropout_rate,
stride=2,
conv_bias=conv_bias)
self.bn1 = nn.BatchNorm2d(nStages[3]) # 20200114 changed here
self.linear = nn.Linear(nStages[3], num_classes)