def __init__(self,
inplanes,
planes,
stride=1,
downsample=None,
using_moving_average=True,
last_gamma=True):
super(Bottleneck, self).__init__()
self.conv1 = nn.Conv2d(inplanes, planes, kernel_size=1, bias=False)
self.sn1 = sn.SwitchNorm(planes,
using_moving_average=using_moving_average)
self.conv2 = nn.Conv2d(planes,
planes,
kernel_size=3,
stride=stride,
padding=1,
bias=False)
self.sn2 = sn.SwitchNorm(planes,
using_moving_average=using_moving_average)
self.conv3 = nn.Conv2d(planes, planes * 4, kernel_size=1, bias=False)
self.sn3 = sn.SwitchNorm(planes * 4,
using_moving_average=using_moving_average,
last_gamma=last_gamma)
self.relu = nn.ReLU(inplace=True)
self.downsample = downsample
self.stride = stride
def __init__(self, block, layers, num_classes=1000, using_moving_average=True, drop_prob=0.5):
self.inplanes = 64
self.using_moving_average = using_moving_average
super(ResNetV2SN, self).__init__()
self.conv1 = nn.Conv2d(3, 64, kernel_size=7, stride=2, padding=3,
bias=False)
self.sn1 = sn.SwitchNorm(64, using_moving_average=self.using_moving_average)
self.relu = nn.ReLU(inplace=True)
self.maxpool = nn.MaxPool2d(kernel_size=3, stride=2, padding=1)
self.layer1 = self._make_layer(block, 64, layers[0])
self.layer2 = self._make_layer(block, 128, layers[1], stride=2)
self.layer3 = self._make_layer(block, 256, layers[2], stride=2)
self.layer4 = self._make_layer(block, 512, layers[3], stride=2)
self.sn_out = sn.SwitchNorm(512 * 4, using_moving_average=self.using_moving_average)
self.spp = SPPLayer(3)
self.drouput = nn.Dropout(p=drop_prob)
self.fc = nn.Linear(512 * block.expansion * (64 + 16 + 4), num_classes)
for m in self.modules():
if isinstance(m, nn.Conv2d):
n = m.kernel_size[0] * m.kernel_size[1] * m.out_channels
m.weight.data.normal_(0, math.sqrt(2. / n))
elif isinstance(m, sn.SwitchNorm):
m.weight.data.fill_(1)
m.bias.data.zero_()
def __init__(self, inplanes, planes, stride=1, downsample=None, using_moving_average=True):
super(BasicBlock, self).__init__()
self.sn1 = sn.SwitchNorm(inplanes, using_moving_average=using_moving_average)
self.conv1 = conv3x3(inplanes, planes, stride)
self.relu = nn.ReLU(inplace=True)
self.sn2 = sn.SwitchNorm(planes, using_moving_average=using_moving_average)
self.conv2 = conv3x3(planes, planes)
self.downsample = downsample
self.stride = stride
def _make_layer(self, block, planes, blocks, stride=1, last_gamma=True):
downsample = None
if stride != 1 or self.inplanes != planes * block.expansion:
downsample = nn.Sequential(
nn.Conv2d(self.inplanes,
planes * block.expansion,
kernel_size=1,
stride=stride,
bias=False),
sn.SwitchNorm(planes * block.expansion,
using_moving_average=self.using_moving_average),
)
layers = [
block(self.inplanes,
planes,
stride,
downsample,
using_moving_average=self.using_moving_average,
last_gamma=last_gamma)
]
self.inplanes = planes * block.expansion
for i in range(1, blocks):
layers.append(
block(self.inplanes,
planes,
using_moving_average=self.using_moving_average,
last_gamma=last_gamma))
return nn.Sequential(*layers)