def __init__(self, block, layers, num_classes=1000, **kwargs):
self.inplanes = 64
super(ResNet_whiten_3n, self).__init__()
self.conv1 = nn.Conv2d(3,
64,
kernel_size=7,
stride=2,
padding=3,
bias=False)
self.bn1 = my.Norm(64)
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.avgpool = nn.AvgPool2d(7, stride=1)
if kwargs.setdefault('last', False):
self.last_bn = my.Norm(512 * block.expansion, dim=2)
else:
self.last_bn = None
drop_ratio = kwargs.setdefault('dropout', 0)
self.dropout = nn.Dropout(p=drop_ratio) if drop_ratio > 0 else None
self.fc = nn.Linear(512 * block.expansion, 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, nn.BatchNorm2d):
m.weight.data.fill_(1)
m.bias.data.zero_()
def __init__(self,
inplanes,
planes,
stride=1,
downsample=None,
groups=1,
base_width=64,
dilation=1,
norm_layer=None):
super(Bottleneck, self).__init__()
if norm_layer is None:
norm_layer = nn.BatchNorm2d
width = int(planes * (base_width / 64.)) * groups
# Both self.conv2 and self.downsample layers downsample the input when stride != 1
self.conv1 = conv1x1(inplanes, width)
#self.bn1 = norm_layer(width)
self.bn1 = my.Norm(width)
self.conv2 = conv3x3(width, width, stride, groups, dilation)
#self.bn2 = norm_layer(width)
self.bn2 = my.Norm(width)
self.conv3 = conv1x1(width, planes * self.expansion)
#self.bn3 = norm_layer(planes * self.expansion)
self.bn3 = my.Norm(planes * self.expansion)
self.relu = nn.ReLU(inplace=True)
self.downsample = downsample
self.stride = stride
def __init__(self,
inplanes,
planes,
stride=1,
downsample=None,
groups=1,
base_width=64,
dilation=1,
norm_layer=None):
super(BasicBlock, self).__init__()
if norm_layer is None:
norm_layer = nn.BatchNorm2d
if groups != 1 or base_width != 64:
raise ValueError(
'BasicBlock only supports groups=1 and base_width=64')
if dilation > 1:
raise NotImplementedError(
"Dilation > 1 not supported in BasicBlock")
# Both self.conv1 and self.downsample layers downsample the input when stride != 1
self.conv1 = conv3x3(inplanes, planes, stride)
#self.bn1 = norm_layer(planes)
self.bn1 = my.Norm(planes)
self.relu = nn.ReLU(inplace=True)
self.conv2 = conv3x3(planes, planes)
#self.bn2 = norm_layer(planes)
self.bn2 = my.Norm(planes)
self.downsample = downsample
self.stride = stride
def __init__(self, block, layers, num_classes=1000, zero_init_residual=False,
groups=1, width_per_group=64, replace_stride_with_dilation=None,
norm_layer=None,**kwargs):
super(ResNext, self).__init__()
if norm_layer is None:
norm_layer = nn.BatchNorm2d
self._norm_layer = norm_layer
self.inplanes = 64
self.dilation = 1
if replace_stride_with_dilation is None:
# each element in the tuple indicates if we should replace
# the 2x2 stride with a dilated convolution instead
replace_stride_with_dilation = [False, False, False]
if len(replace_stride_with_dilation) != 3:
raise ValueError("replace_stride_with_dilation should be None "
"or a 3-element tuple, got {}".format(replace_stride_with_dilation))
self.groups = groups
self.base_width = width_per_group
self.conv1 = nn.Conv2d(3, self.inplanes, kernel_size=7, stride=2, padding=3,
bias=False)
#self.bn1 = norm_layer(self.inplanes)
self.bn1 = my.Norm(self.inplanes)
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,
dilate=replace_stride_with_dilation[0])
self.layer3 = self._make_layer(block, 256, layers[2], stride=2,
dilate=replace_stride_with_dilation[1])
self.layer4 = self._make_layer(block, 512, layers[3], stride=2,
dilate=replace_stride_with_dilation[2])
self.avgpool = nn.AdaptiveAvgPool2d((1, 1))
if kwargs.setdefault('last', False):
self.last_bn = my.Norm(512 * block.expansion, dim=2)
else:
self.last_bn = None
drop_ratio=kwargs.setdefault('dropout', 0)
self.dropout = nn.Dropout(p=drop_ratio) if drop_ratio > 0 else None
self.fc = nn.Linear(512 * block.expansion, num_classes)
for m in self.modules():
if isinstance(m, nn.Conv2d):
nn.init.kaiming_normal_(m.weight, mode='fan_out', nonlinearity='relu')
elif isinstance(m, (nn.BatchNorm2d, nn.GroupNorm)):
nn.init.constant_(m.weight, 1)
nn.init.constant_(m.bias, 0)
# Zero-initialize the last BN in each residual branch,
# so that the residual branch starts with zeros, and each residual block behaves like an identity.
# This improves the model by 0.2~0.3% according to https://arxiv.org/abs/1706.02677
if zero_init_residual:
for m in self.modules():
if isinstance(m, Bottleneck):
nn.init.constant_(m.bn3.weight, 0)
elif isinstance(m, BasicBlock):
nn.init.constant_(m.bn2.weight, 0)
def __init__(self, inplanes, planes, stride=1, downsample=None):
super(BasicBlock, self).__init__()
self.conv1 = conv3x3(inplanes, planes, stride)
#self.bn1 = nn.BatchNorm2d(planes)
self.bn1 = my.Norm(planes)
self.relu = nn.ReLU(inplace=True)
self.conv2 = conv3x3(planes, planes)
self.bn2 = my.Norm(planes)
self.downsample = downsample
self.stride = stride
def __init__(self, depth=4, width=100, **kwargs):
super(MLP, self).__init__()
layers = [
ext.View(28 * 28),
nn.Linear(28 * 28, width),
ext.Norm(width),
nn.ReLU(True)
]
for index in range(depth - 1):
layers.append(nn.Linear(width, width))
layers.append(ext.Norm(width))
layers.append(nn.ReLU(True))
layers.append(nn.Linear(width, 10))
self.net = nn.Sequential(*layers)
def __init__(self, inplanes, planes, stride=1, downsample=None):
super(BasicBlock, self).__init__()
#self.conv1 = conv3x3(inplanes, planes, stride)
self.conv1 = my.NormConv(inplanes, planes, 3, stride, padding=1, bias=False)
#self.bn1 = nn.BatchNorm2d(planes)
self.bn1 = my.Norm(planes)
self.relu = nn.ReLU(inplace=True)
#self.conv2 = conv3x3(planes, planes)
#self.conv2 = Conv2d_ONI(planes, planes, 3, stride=1, padding=1, bias=False, T=7)
self.conv2 = my.NormConv(planes, planes, 3, stride=1, padding=1, bias=False)
#self.bn2 = nn.BatchNorm2d(planes)
self.bn2 = my.Norm(planes)
self.downsample = downsample
self.stride = stride
def __init__(self, depth=28, widen_factor=1, num_classes=10, dropout=0.0):
super(WideResNet, self).__init__()
nChannels = [
16, 16 * widen_factor, 32 * widen_factor, 64 * widen_factor
]
assert ((depth - 4) % 6 == 0)
n = (depth - 4) / 6
block = BasicBlock
# 1st conv before any network block
self.conv1 = nn.Conv2d(3, nChannels[0], 3, 1, 1, bias=False)
# 1st block
self.block1 = self._make_layer(n, block, nChannels[0], nChannels[1], 1,
dropout)
# 2nd block
self.block2 = self._make_layer(n, block, nChannels[1], nChannels[2], 2,
dropout)
# 3rd block
self.block3 = self._make_layer(n, block, nChannels[2], nChannels[3], 2,
dropout)
# global average pooling and classifier
self.bn = my.Norm(nChannels[3])
self.relu = nn.ReLU(inplace=True)
self.pool = nn.AvgPool2d(8, 1)
self.fc = nn.Linear(nChannels[3], num_classes)
self.nChannels = nChannels[3]
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, nn.BatchNorm2d):
m.weight.data.fill_(1)
m.bias.data.zero_()
elif isinstance(m, nn.Linear):
m.bias.data.zero_()
def _make_layer(self, block, planes, blocks, stride=1, dilate=False):
norm_layer = self._norm_layer
downsample = None
previous_dilation = self.dilation
if dilate:
self.dilation *= stride
stride = 1
if stride != 1 or self.inplanes != planes * block.expansion:
downsample = nn.Sequential(
conv1x1(self.inplanes, planes * block.expansion, stride),
my.Norm(planes * block.expansion),
)
layers = []
layers.append(
block(self.inplanes, planes, stride, downsample, self.groups,
self.base_width, previous_dilation, norm_layer))
self.inplanes = planes * block.expansion
for _ in range(1, blocks):
layers.append(
block(self.inplanes,
planes,
groups=self.groups,
base_width=self.base_width,
dilation=self.dilation,
norm_layer=norm_layer))
return nn.Sequential(*layers)
def __init__(self, inplanes, planes, stride=1, downsample=None):
super(Bottleneck, self).__init__()
self.conv1 = nn.Conv2d(inplanes, planes, kernel_size=1, bias=False)
self.bn1 = my.Norm(planes)
self.conv2 = nn.Conv2d(planes,
planes,
kernel_size=3,
stride=stride,
padding=1,
bias=False)
self.bn2 = my.Norm(planes)
self.conv3 = nn.Conv2d(planes, planes * 4, kernel_size=1, bias=False)
self.bn3 = nn.BatchNorm2d(planes * 4)
self.relu = nn.ReLU(inplace=True)
self.downsample = downsample
self.stride = stride
def __init__(self, in_channels, out_channels, stride, drop_ratio=0.0):
super(BasicBlock, self).__init__()
self.bn1 = my.Norm(in_channels)
self.relu1 = nn.ReLU(inplace=True)
self.conv1 = nn.Conv2d(in_channels,
out_channels,
3,
stride,
1,
bias=False)
self.bn2 = my.Norm(out_channels)
self.relu2 = nn.ReLU(inplace=True)
self.dropout = nn.Dropout2d(p=drop_ratio) if drop_ratio > 0 else None
self.conv2 = nn.Conv2d(out_channels, out_channels, 3, 1, 1, bias=False)
if in_channels != out_channels:
self.shortcut = nn.Conv2d(in_channels,
out_channels,
1,
stride,
0,
bias=False)
else:
self.shortcut = None
def _make_layer(self, block, planes, blocks, stride=1):
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),
my.NormConv(self.inplanes, planes * block.expansion, kernel_size=1, stride=stride, bias=False, ONIRow_Fix=True),
#nn.BatchNorm2d(planes * block.expansion), )
my.Norm(planes * block.expansion), )
layers = []
layers.append(block(self.inplanes, planes, stride, downsample))
self.inplanes = planes * block.expansion
for i in range(1, blocks):
layers.append(block(self.inplanes, planes))
return nn.Sequential(*layers)
def make_layers(cfg, batch_norm=True):
layers = []
in_channels = 3
for v in cfg:
if v == 'M':
layers += [nn.MaxPool2d(kernel_size=2, stride=2)]
else:
conv2d = nn.Conv2d(in_channels,
v,
kernel_size=3,
padding=1,
bias=False)
if batch_norm:
layers += [conv2d, my.Norm(v), nn.ReLU(inplace=True)]
else:
layers += [conv2d, nn.ReLU(inplace=True)]
in_channels = v
return nn.Sequential(*layers)
def __init__(self, block, layers, num_classes=1000, **kwargs):
self.inplanes = 64
super(ResNetDebug, self).__init__()
#self.conv1 = nn.Conv2d(3, 64, kernel_size=7, stride=2, padding=3, bias=False)
self.conv1 = my.NormConv(3, 64, kernel_size=7, stride=2, padding=3, bias=False)
self.bn1 = my.Norm(64)
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.avgpool = nn.AvgPool2d(7, stride=1)
self.fc = nn.Linear(512 * block.expansion, 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, nn.BatchNorm2d):
m.weight.data.fill_(1)
m.bias.data.zero_()
