init + inference.py патч

This commit is contained in:
2026-09-12 03:29:38 +03:00
commit 6f7f6031d1
50 changed files with 7410 additions and 0 deletions

View File

@ -0,0 +1,13 @@
from model.deep_lab_model.backbone import resnet, xception, drn, mobilenet
def build_backbone(backbone, output_stride, BatchNorm):
if backbone == 'resnet':
return resnet.ResNet101(output_stride, BatchNorm)
elif backbone == 'xception':
return xception.AlignedXception(output_stride, BatchNorm)
elif backbone == 'drn':
return drn.drn_d_54(BatchNorm)
elif backbone == 'mobilenet':
return mobilenet.MobileNetV2(output_stride, BatchNorm)
else:
raise NotImplementedError

View File

@ -0,0 +1,402 @@
import torch.nn as nn
import math
import torch.utils.model_zoo as model_zoo
from model.deep_lab_model.sync_batchnorm.batchnorm import SynchronizedBatchNorm2d
webroot = 'http://dl.yf.io/drn/'
model_urls = {
'resnet50': 'https://download.pytorch.org/models/resnet50-19c8e357.pth',
'drn-c-26': webroot + 'drn_c_26-ddedf421.pth',
'drn-c-42': webroot + 'drn_c_42-9d336e8c.pth',
'drn-c-58': webroot + 'drn_c_58-0a53a92c.pth',
'drn-d-22': webroot + 'drn_d_22-4bd2f8ea.pth',
'drn-d-38': webroot + 'drn_d_38-eebb45f0.pth',
'drn-d-54': webroot + 'drn_d_54-0e0534ff.pth',
'drn-d-105': webroot + 'drn_d_105-12b40979.pth'
}
def conv3x3(in_planes, out_planes, stride=1, padding=1, dilation=1):
return nn.Conv2d(in_planes, out_planes, kernel_size=3, stride=stride,
padding=padding, bias=False, dilation=dilation)
class BasicBlock(nn.Module):
expansion = 1
def __init__(self, inplanes, planes, stride=1, downsample=None,
dilation=(1, 1), residual=True, BatchNorm=None):
super(BasicBlock, self).__init__()
self.conv1 = conv3x3(inplanes, planes, stride,
padding=dilation[0], dilation=dilation[0])
self.bn1 = BatchNorm(planes)
self.relu = nn.ReLU(inplace=True)
self.conv2 = conv3x3(planes, planes,
padding=dilation[1], dilation=dilation[1])
self.bn2 = BatchNorm(planes)
self.downsample = downsample
self.stride = stride
self.residual = residual
def forward(self, x):
residual = x
out = self.conv1(x)
out = self.bn1(out)
out = self.relu(out)
out = self.conv2(out)
out = self.bn2(out)
if self.downsample is not None:
residual = self.downsample(x)
if self.residual:
out += residual
out = self.relu(out)
return out
class Bottleneck(nn.Module):
expansion = 4
def __init__(self, inplanes, planes, stride=1, downsample=None,
dilation=(1, 1), residual=True, BatchNorm=None):
super(Bottleneck, self).__init__()
self.conv1 = nn.Conv2d(inplanes, planes, kernel_size=1, bias=False)
self.bn1 = BatchNorm(planes)
self.conv2 = nn.Conv2d(planes, planes, kernel_size=3, stride=stride,
padding=dilation[1], bias=False,
dilation=dilation[1])
self.bn2 = BatchNorm(planes)
self.conv3 = nn.Conv2d(planes, planes * 4, kernel_size=1, bias=False)
self.bn3 = BatchNorm(planes * 4)
self.relu = nn.ReLU(inplace=True)
self.downsample = downsample
self.stride = stride
def forward(self, x):
residual = x
out = self.conv1(x)
out = self.bn1(out)
out = self.relu(out)
out = self.conv2(out)
out = self.bn2(out)
out = self.relu(out)
out = self.conv3(out)
out = self.bn3(out)
if self.downsample is not None:
residual = self.downsample(x)
out += residual
out = self.relu(out)
return out
class DRN(nn.Module):
def __init__(self, block, layers, arch='D',
channels=(16, 32, 64, 128, 256, 512, 512, 512),
BatchNorm=None):
super(DRN, self).__init__()
self.inplanes = channels[0]
self.out_dim = channels[-1]
self.arch = arch
if arch == 'C':
self.conv1 = nn.Conv2d(3, channels[0], kernel_size=7, stride=1,
padding=3, bias=False)
self.bn1 = BatchNorm(channels[0])
self.relu = nn.ReLU(inplace=True)
self.layer1 = self._make_layer(
BasicBlock, channels[0], layers[0], stride=1, BatchNorm=BatchNorm)
self.layer2 = self._make_layer(
BasicBlock, channels[1], layers[1], stride=2, BatchNorm=BatchNorm)
elif arch == 'D':
self.layer0 = nn.Sequential(
nn.Conv2d(3, channels[0], kernel_size=7, stride=1, padding=3,
bias=False),
BatchNorm(channels[0]),
nn.ReLU(inplace=True)
)
self.layer1 = self._make_conv_layers(
channels[0], layers[0], stride=1, BatchNorm=BatchNorm)
self.layer2 = self._make_conv_layers(
channels[1], layers[1], stride=2, BatchNorm=BatchNorm)
self.layer3 = self._make_layer(block, channels[2], layers[2], stride=2, BatchNorm=BatchNorm)
self.layer4 = self._make_layer(block, channels[3], layers[3], stride=2, BatchNorm=BatchNorm)
self.layer5 = self._make_layer(block, channels[4], layers[4],
dilation=2, new_level=False, BatchNorm=BatchNorm)
self.layer6 = None if layers[5] == 0 else \
self._make_layer(block, channels[5], layers[5], dilation=4,
new_level=False, BatchNorm=BatchNorm)
if arch == 'C':
self.layer7 = None if layers[6] == 0 else \
self._make_layer(BasicBlock, channels[6], layers[6], dilation=2,
new_level=False, residual=False, BatchNorm=BatchNorm)
self.layer8 = None if layers[7] == 0 else \
self._make_layer(BasicBlock, channels[7], layers[7], dilation=1,
new_level=False, residual=False, BatchNorm=BatchNorm)
elif arch == 'D':
self.layer7 = None if layers[6] == 0 else \
self._make_conv_layers(channels[6], layers[6], dilation=2, BatchNorm=BatchNorm)
self.layer8 = None if layers[7] == 0 else \
self._make_conv_layers(channels[7], layers[7], dilation=1, BatchNorm=BatchNorm)
self._init_weight()
def _init_weight(self):
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, SynchronizedBatchNorm2d):
m.weight.data.fill_(1)
m.bias.data.zero_()
elif isinstance(m, nn.BatchNorm2d):
m.weight.data.fill_(1)
m.bias.data.zero_()
def _make_layer(self, block, planes, blocks, stride=1, dilation=1,
new_level=True, residual=True, BatchNorm=None):
assert dilation == 1 or dilation % 2 == 0
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),
BatchNorm(planes * block.expansion),
)
layers = list()
layers.append(block(
self.inplanes, planes, stride, downsample,
dilation=(1, 1) if dilation == 1 else (
dilation // 2 if new_level else dilation, dilation),
residual=residual, BatchNorm=BatchNorm))
self.inplanes = planes * block.expansion
for i in range(1, blocks):
layers.append(block(self.inplanes, planes, residual=residual,
dilation=(dilation, dilation), BatchNorm=BatchNorm))
return nn.Sequential(*layers)
def _make_conv_layers(self, channels, convs, stride=1, dilation=1, BatchNorm=None):
modules = []
for i in range(convs):
modules.extend([
nn.Conv2d(self.inplanes, channels, kernel_size=3,
stride=stride if i == 0 else 1,
padding=dilation, bias=False, dilation=dilation),
BatchNorm(channels),
nn.ReLU(inplace=True)])
self.inplanes = channels
return nn.Sequential(*modules)
def forward(self, x):
if self.arch == 'C':
x = self.conv1(x)
x = self.bn1(x)
x = self.relu(x)
elif self.arch == 'D':
x = self.layer0(x)
x = self.layer1(x)
x = self.layer2(x)
x = self.layer3(x)
low_level_feat = x
x = self.layer4(x)
x = self.layer5(x)
if self.layer6 is not None:
x = self.layer6(x)
if self.layer7 is not None:
x = self.layer7(x)
if self.layer8 is not None:
x = self.layer8(x)
return x, low_level_feat
class DRN_A(nn.Module):
def __init__(self, block, layers, BatchNorm=None):
self.inplanes = 64
super(DRN_A, self).__init__()
self.out_dim = 512 * block.expansion
self.conv1 = nn.Conv2d(3, 64, kernel_size=7, stride=2, padding=3,
bias=False)
self.bn1 = BatchNorm(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], BatchNorm=BatchNorm)
self.layer2 = self._make_layer(block, 128, layers[1], stride=2, BatchNorm=BatchNorm)
self.layer3 = self._make_layer(block, 256, layers[2], stride=1,
dilation=2, BatchNorm=BatchNorm)
self.layer4 = self._make_layer(block, 512, layers[3], stride=1,
dilation=4, BatchNorm=BatchNorm)
self._init_weight()
def _init_weight(self):
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, SynchronizedBatchNorm2d):
m.weight.data.fill_(1)
m.bias.data.zero_()
elif isinstance(m, nn.BatchNorm2d):
m.weight.data.fill_(1)
m.bias.data.zero_()
def _make_layer(self, block, planes, blocks, stride=1, dilation=1, BatchNorm=None):
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),
BatchNorm(planes * block.expansion),
)
layers = []
layers.append(block(self.inplanes, planes, stride, downsample, BatchNorm=BatchNorm))
self.inplanes = planes * block.expansion
for i in range(1, blocks):
layers.append(block(self.inplanes, planes,
dilation=(dilation, dilation, ), BatchNorm=BatchNorm))
return nn.Sequential(*layers)
def forward(self, x):
x = self.conv1(x)
x = self.bn1(x)
x = self.relu(x)
x = self.maxpool(x)
x = self.layer1(x)
x = self.layer2(x)
x = self.layer3(x)
x = self.layer4(x)
return x
def drn_a_50(BatchNorm, pretrained=True):
model = DRN_A(Bottleneck, [3, 4, 6, 3], BatchNorm=BatchNorm)
if pretrained:
model.load_state_dict(model_zoo.load_url(model_urls['resnet50']))
return model
def drn_c_26(BatchNorm, pretrained=True):
model = DRN(BasicBlock, [1, 1, 2, 2, 2, 2, 1, 1], arch='C', BatchNorm=BatchNorm)
if pretrained:
pretrained = model_zoo.load_url(model_urls['drn-c-26'])
del pretrained['fc.weight']
del pretrained['fc.bias']
model.load_state_dict(pretrained)
return model
def drn_c_42(BatchNorm, pretrained=True):
model = DRN(BasicBlock, [1, 1, 3, 4, 6, 3, 1, 1], arch='C', BatchNorm=BatchNorm)
if pretrained:
pretrained = model_zoo.load_url(model_urls['drn-c-42'])
del pretrained['fc.weight']
del pretrained['fc.bias']
model.load_state_dict(pretrained)
return model
def drn_c_58(BatchNorm, pretrained=True):
model = DRN(Bottleneck, [1, 1, 3, 4, 6, 3, 1, 1], arch='C', BatchNorm=BatchNorm)
if pretrained:
pretrained = model_zoo.load_url(model_urls['drn-c-58'])
del pretrained['fc.weight']
del pretrained['fc.bias']
model.load_state_dict(pretrained)
return model
def drn_d_22(BatchNorm, pretrained=True):
model = DRN(BasicBlock, [1, 1, 2, 2, 2, 2, 1, 1], arch='D', BatchNorm=BatchNorm)
if pretrained:
pretrained = model_zoo.load_url(model_urls['drn-d-22'])
del pretrained['fc.weight']
del pretrained['fc.bias']
model.load_state_dict(pretrained)
return model
def drn_d_24(BatchNorm, pretrained=True):
model = DRN(BasicBlock, [1, 1, 2, 2, 2, 2, 2, 2], arch='D', BatchNorm=BatchNorm)
if pretrained:
pretrained = model_zoo.load_url(model_urls['drn-d-24'])
del pretrained['fc.weight']
del pretrained['fc.bias']
model.load_state_dict(pretrained)
return model
def drn_d_38(BatchNorm, pretrained=True):
model = DRN(BasicBlock, [1, 1, 3, 4, 6, 3, 1, 1], arch='D', BatchNorm=BatchNorm)
if pretrained:
pretrained = model_zoo.load_url(model_urls['drn-d-38'])
del pretrained['fc.weight']
del pretrained['fc.bias']
model.load_state_dict(pretrained)
return model
def drn_d_40(BatchNorm, pretrained=True):
model = DRN(BasicBlock, [1, 1, 3, 4, 6, 3, 2, 2], arch='D', BatchNorm=BatchNorm)
if pretrained:
pretrained = model_zoo.load_url(model_urls['drn-d-40'])
del pretrained['fc.weight']
del pretrained['fc.bias']
model.load_state_dict(pretrained)
return model
def drn_d_54(BatchNorm, pretrained=True):
model = DRN(Bottleneck, [1, 1, 3, 4, 6, 3, 1, 1], arch='D', BatchNorm=BatchNorm)
if pretrained:
pretrained = model_zoo.load_url(model_urls['drn-d-54'])
del pretrained['fc.weight']
del pretrained['fc.bias']
model.load_state_dict(pretrained)
return model
def drn_d_105(BatchNorm, pretrained=True):
model = DRN(Bottleneck, [1, 1, 3, 4, 23, 3, 1, 1], arch='D', BatchNorm=BatchNorm)
if pretrained:
pretrained = model_zoo.load_url(model_urls['drn-d-105'])
del pretrained['fc.weight']
del pretrained['fc.bias']
model.load_state_dict(pretrained)
return model
if __name__ == "__main__":
import torch
model = drn_a_50(BatchNorm=nn.BatchNorm2d, pretrained=True)
input = torch.rand(1, 3, 512, 512)
output, low_level_feat = model(input)
print(output.size())
print(low_level_feat.size())

View File

@ -0,0 +1,151 @@
import torch
import torch.nn.functional as F
import torch.nn as nn
import math
from model.deep_lab_model.sync_batchnorm.batchnorm import SynchronizedBatchNorm2d
import torch.utils.model_zoo as model_zoo
def conv_bn(inp, oup, stride, BatchNorm):
return nn.Sequential(
nn.Conv2d(inp, oup, 3, stride, 1, bias=False),
BatchNorm(oup),
nn.ReLU6(inplace=True)
)
def fixed_padding(inputs, kernel_size, dilation):
kernel_size_effective = kernel_size + (kernel_size - 1) * (dilation - 1)
pad_total = kernel_size_effective - 1
pad_beg = pad_total // 2
pad_end = pad_total - pad_beg
padded_inputs = F.pad(inputs, (pad_beg, pad_end, pad_beg, pad_end))
return padded_inputs
class InvertedResidual(nn.Module):
def __init__(self, inp, oup, stride, dilation, expand_ratio, BatchNorm):
super(InvertedResidual, self).__init__()
self.stride = stride
assert stride in [1, 2]
hidden_dim = round(inp * expand_ratio)
self.use_res_connect = self.stride == 1 and inp == oup
self.kernel_size = 3
self.dilation = dilation
if expand_ratio == 1:
self.conv = nn.Sequential(
# dw
nn.Conv2d(hidden_dim, hidden_dim, 3, stride, 0, dilation, groups=hidden_dim, bias=False),
BatchNorm(hidden_dim),
nn.ReLU6(inplace=True),
# pw-linear
nn.Conv2d(hidden_dim, oup, 1, 1, 0, 1, 1, bias=False),
BatchNorm(oup),
)
else:
self.conv = nn.Sequential(
# pw
nn.Conv2d(inp, hidden_dim, 1, 1, 0, 1, bias=False),
BatchNorm(hidden_dim),
nn.ReLU6(inplace=True),
# dw
nn.Conv2d(hidden_dim, hidden_dim, 3, stride, 0, dilation, groups=hidden_dim, bias=False),
BatchNorm(hidden_dim),
nn.ReLU6(inplace=True),
# pw-linear
nn.Conv2d(hidden_dim, oup, 1, 1, 0, 1, bias=False),
BatchNorm(oup),
)
def forward(self, x):
x_pad = fixed_padding(x, self.kernel_size, dilation=self.dilation)
if self.use_res_connect:
x = x + self.conv(x_pad)
else:
x = self.conv(x_pad)
return x
class MobileNetV2(nn.Module):
def __init__(self, output_stride=8, BatchNorm=None, width_mult=1., pretrained=True):
super(MobileNetV2, self).__init__()
block = InvertedResidual
input_channel = 32
current_stride = 1
rate = 1
interverted_residual_setting = [
# t, c, n, s
[1, 16, 1, 1],
[6, 24, 2, 2],
[6, 32, 3, 2],
[6, 64, 4, 2],
[6, 96, 3, 1],
[6, 160, 3, 2],
[6, 320, 1, 1],
]
# building first layer
input_channel = int(input_channel * width_mult)
self.features = [conv_bn(3, input_channel, 2, BatchNorm)]
current_stride *= 2
# building inverted residual blocks
for t, c, n, s in interverted_residual_setting:
if current_stride == output_stride:
stride = 1
dilation = rate
rate *= s
else:
stride = s
dilation = 1
current_stride *= s
output_channel = int(c * width_mult)
for i in range(n):
if i == 0:
self.features.append(block(input_channel, output_channel, stride, dilation, t, BatchNorm))
else:
self.features.append(block(input_channel, output_channel, 1, dilation, t, BatchNorm))
input_channel = output_channel
self.features = nn.Sequential(*self.features)
self._initialize_weights()
if pretrained:
self._load_pretrained_model()
self.low_level_features = self.features[0:4]
self.high_level_features = self.features[4:]
def forward(self, x):
low_level_feat = self.low_level_features(x)
x = self.high_level_features(low_level_feat)
return x, low_level_feat
def _load_pretrained_model(self):
pretrain_dict = model_zoo.load_url('http://jeff95.me/models/mobilenet_v2-6a65762b.pth')
model_dict = {}
state_dict = self.state_dict()
for k, v in pretrain_dict.items():
if k in state_dict:
model_dict[k] = v
state_dict.update(model_dict)
self.load_state_dict(state_dict)
def _initialize_weights(self):
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))
torch.nn.init.kaiming_normal_(m.weight)
elif isinstance(m, SynchronizedBatchNorm2d):
m.weight.data.fill_(1)
m.bias.data.zero_()
elif isinstance(m, nn.BatchNorm2d):
m.weight.data.fill_(1)
m.bias.data.zero_()
if __name__ == "__main__":
input = torch.rand(1, 3, 512, 512)
model = MobileNetV2(output_stride=16, BatchNorm=nn.BatchNorm2d)
output, low_level_feat = model(input)
print(output.size())
print(low_level_feat.size())

View File

@ -0,0 +1,170 @@
import math
import torch.nn as nn
import torch.utils.model_zoo as model_zoo
from model.deep_lab_model.sync_batchnorm.batchnorm import SynchronizedBatchNorm2d
class Bottleneck(nn.Module):
expansion = 4
def __init__(self, inplanes, planes, stride=1, dilation=1, downsample=None, BatchNorm=None):
super(Bottleneck, self).__init__()
self.conv1 = nn.Conv2d(inplanes, planes, kernel_size=1, bias=False)
self.bn1 = BatchNorm(planes)
self.conv2 = nn.Conv2d(planes, planes, kernel_size=3, stride=stride,
dilation=dilation, padding=dilation, bias=False)
self.bn2 = BatchNorm(planes)
self.conv3 = nn.Conv2d(planes, planes * 4, kernel_size=1, bias=False)
self.bn3 = BatchNorm(planes * 4)
self.relu = nn.ReLU(inplace=True)
self.downsample = downsample
self.stride = stride
self.dilation = dilation
def forward(self, x):
residual = x
out = self.conv1(x)
out = self.bn1(out)
out = self.relu(out)
out = self.conv2(out)
out = self.bn2(out)
out = self.relu(out)
out = self.conv3(out)
out = self.bn3(out)
if self.downsample is not None:
residual = self.downsample(x)
out += residual
out = self.relu(out)
return out
class ResNet(nn.Module):
def __init__(self, block, layers, output_stride, BatchNorm, pretrained=True):
self.inplanes = 64
super(ResNet, self).__init__()
blocks = [1, 2, 4]
if output_stride == 16:
strides = [1, 2, 2, 1]
dilations = [1, 1, 1, 2]
elif output_stride == 8:
strides = [1, 2, 1, 1]
dilations = [1, 1, 2, 4]
else:
raise NotImplementedError
# Modules
self.conv1 = nn.Conv2d(3, 64, kernel_size=7, stride=2, padding=3,
bias=False)
self.bn1 = BatchNorm(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], stride=strides[0], dilation=dilations[0], BatchNorm=BatchNorm)
self.layer2 = self._make_layer(block, 128, layers[1], stride=strides[1], dilation=dilations[1], BatchNorm=BatchNorm)
self.layer3 = self._make_layer(block, 256, layers[2], stride=strides[2], dilation=dilations[2], BatchNorm=BatchNorm)
self.layer4 = self._make_MG_unit(block, 512, blocks=blocks, stride=strides[3], dilation=dilations[3], BatchNorm=BatchNorm)
# self.layer4 = self._make_layer(block, 512, layers[3], stride=strides[3], dilation=dilations[3], BatchNorm=BatchNorm)
self._init_weight()
# if pretrained:
# self._load_pretrained_model()
def _make_layer(self, block, planes, blocks, stride=1, dilation=1, BatchNorm=None):
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),
BatchNorm(planes * block.expansion),
)
layers = []
layers.append(block(self.inplanes, planes, stride, dilation, downsample, BatchNorm))
self.inplanes = planes * block.expansion
for i in range(1, blocks):
layers.append(block(self.inplanes, planes, dilation=dilation, BatchNorm=BatchNorm))
return nn.Sequential(*layers)
def _make_MG_unit(self, block, planes, blocks, stride=1, dilation=1, BatchNorm=None):
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),
BatchNorm(planes * block.expansion),
)
layers = []
layers.append(block(self.inplanes, planes, stride, dilation=blocks[0]*dilation,
downsample=downsample, BatchNorm=BatchNorm))
self.inplanes = planes * block.expansion
for i in range(1, len(blocks)):
layers.append(block(self.inplanes, planes, stride=1,
dilation=blocks[i]*dilation, BatchNorm=BatchNorm))
return nn.Sequential(*layers)
def forward(self, input):
x = self.conv1(input)
x = self.bn1(x)
x = self.relu(x)
x = self.maxpool(x)
x = self.layer1(x)
low_level_feat = x
x = self.layer2(x)
x = self.layer3(x)
x = self.layer4(x)
return x, low_level_feat
def _init_weight(self):
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, SynchronizedBatchNorm2d):
m.weight.data.fill_(1)
m.bias.data.zero_()
elif isinstance(m, nn.BatchNorm2d):
m.weight.data.fill_(1)
m.bias.data.zero_()
def _load_pretrained_model(self):
import urllib.request
import ssl
ssl._create_default_https_context = ssl._create_unverified_context
response = urllib.request.urlopen('https://download.pytorch.org/models/resnet101-5d3b4d8f.pth')
pretrain_dict = model_zoo.load_url('https://download.pytorch.org/models/resnet101-5d3b4d8f.pth')
model_dict = {}
state_dict = self.state_dict()
for k, v in pretrain_dict.items():
if k in state_dict:
# if 'conv1' in k:
# continue
model_dict[k] = v
state_dict.update(model_dict)
self.load_state_dict(state_dict)
def ResNet101(output_stride, BatchNorm, pretrained=True):
"""Constructs a ResNet-101 model.
Args:
pretrained (bool): If True, returns a model pre-trained on ImageNet
"""
model = ResNet(Bottleneck, [3, 4, 23, 3], output_stride, BatchNorm, pretrained=pretrained)
return model
if __name__ == "__main__":
import torch
model = ResNet101(BatchNorm=nn.BatchNorm2d, pretrained=True, output_stride=8)
input = torch.rand(1, 3, 512, 512)
output, low_level_feat = model(input)
print(output.size())
print(low_level_feat.size())

View File

@ -0,0 +1,288 @@
import math
import torch
import torch.nn as nn
import torch.nn.functional as F
import torch.utils.model_zoo as model_zoo
from model.deep_lab_model.sync_batchnorm.batchnorm import SynchronizedBatchNorm2d
def fixed_padding(inputs, kernel_size, dilation):
kernel_size_effective = kernel_size + (kernel_size - 1) * (dilation - 1)
pad_total = kernel_size_effective - 1
pad_beg = pad_total // 2
pad_end = pad_total - pad_beg
padded_inputs = F.pad(inputs, (pad_beg, pad_end, pad_beg, pad_end))
return padded_inputs
class SeparableConv2d(nn.Module):
def __init__(self, inplanes, planes, kernel_size=3, stride=1, dilation=1, bias=False, BatchNorm=None):
super(SeparableConv2d, self).__init__()
self.conv1 = nn.Conv2d(inplanes, inplanes, kernel_size, stride, 0, dilation,
groups=inplanes, bias=bias)
self.bn = BatchNorm(inplanes)
self.pointwise = nn.Conv2d(inplanes, planes, 1, 1, 0, 1, 1, bias=bias)
def forward(self, x):
x = fixed_padding(x, self.conv1.kernel_size[0], dilation=self.conv1.dilation[0])
x = self.conv1(x)
x = self.bn(x)
x = self.pointwise(x)
return x
class Block(nn.Module):
def __init__(self, inplanes, planes, reps, stride=1, dilation=1, BatchNorm=None,
start_with_relu=True, grow_first=True, is_last=False):
super(Block, self).__init__()
if planes != inplanes or stride != 1:
self.skip = nn.Conv2d(inplanes, planes, 1, stride=stride, bias=False)
self.skipbn = BatchNorm(planes)
else:
self.skip = None
self.relu = nn.ReLU(inplace=True)
rep = []
filters = inplanes
if grow_first:
rep.append(self.relu)
rep.append(SeparableConv2d(inplanes, planes, 3, 1, dilation, BatchNorm=BatchNorm))
rep.append(BatchNorm(planes))
filters = planes
for i in range(reps - 1):
rep.append(self.relu)
rep.append(SeparableConv2d(filters, filters, 3, 1, dilation, BatchNorm=BatchNorm))
rep.append(BatchNorm(filters))
if not grow_first:
rep.append(self.relu)
rep.append(SeparableConv2d(inplanes, planes, 3, 1, dilation, BatchNorm=BatchNorm))
rep.append(BatchNorm(planes))
if stride != 1:
rep.append(self.relu)
rep.append(SeparableConv2d(planes, planes, 3, 2, BatchNorm=BatchNorm))
rep.append(BatchNorm(planes))
if stride == 1 and is_last:
rep.append(self.relu)
rep.append(SeparableConv2d(planes, planes, 3, 1, BatchNorm=BatchNorm))
rep.append(BatchNorm(planes))
if not start_with_relu:
rep = rep[1:]
self.rep = nn.Sequential(*rep)
def forward(self, inp):
x = self.rep(inp)
if self.skip is not None:
skip = self.skip(inp)
skip = self.skipbn(skip)
else:
skip = inp
x = x + skip
return x
class AlignedXception(nn.Module):
"""
Modified Alighed Xception
"""
def __init__(self, output_stride, BatchNorm,
pretrained=True):
super(AlignedXception, self).__init__()
if output_stride == 16:
entry_block3_stride = 2
middle_block_dilation = 1
exit_block_dilations = (1, 2)
elif output_stride == 8:
entry_block3_stride = 1
middle_block_dilation = 2
exit_block_dilations = (2, 4)
else:
raise NotImplementedError
# Entry flow
self.conv1 = nn.Conv2d(3, 32, 3, stride=2, padding=1, bias=False)
self.bn1 = BatchNorm(32)
self.relu = nn.ReLU(inplace=True)
self.conv2 = nn.Conv2d(32, 64, 3, stride=1, padding=1, bias=False)
self.bn2 = BatchNorm(64)
self.block1 = Block(64, 128, reps=2, stride=2, BatchNorm=BatchNorm, start_with_relu=False)
self.block2 = Block(128, 256, reps=2, stride=2, BatchNorm=BatchNorm, start_with_relu=False,
grow_first=True)
self.block3 = Block(256, 728, reps=2, stride=entry_block3_stride, BatchNorm=BatchNorm,
start_with_relu=True, grow_first=True, is_last=True)
# Middle flow
self.block4 = Block(728, 728, reps=3, stride=1, dilation=middle_block_dilation,
BatchNorm=BatchNorm, start_with_relu=True, grow_first=True)
self.block5 = Block(728, 728, reps=3, stride=1, dilation=middle_block_dilation,
BatchNorm=BatchNorm, start_with_relu=True, grow_first=True)
self.block6 = Block(728, 728, reps=3, stride=1, dilation=middle_block_dilation,
BatchNorm=BatchNorm, start_with_relu=True, grow_first=True)
self.block7 = Block(728, 728, reps=3, stride=1, dilation=middle_block_dilation,
BatchNorm=BatchNorm, start_with_relu=True, grow_first=True)
self.block8 = Block(728, 728, reps=3, stride=1, dilation=middle_block_dilation,
BatchNorm=BatchNorm, start_with_relu=True, grow_first=True)
self.block9 = Block(728, 728, reps=3, stride=1, dilation=middle_block_dilation,
BatchNorm=BatchNorm, start_with_relu=True, grow_first=True)
self.block10 = Block(728, 728, reps=3, stride=1, dilation=middle_block_dilation,
BatchNorm=BatchNorm, start_with_relu=True, grow_first=True)
self.block11 = Block(728, 728, reps=3, stride=1, dilation=middle_block_dilation,
BatchNorm=BatchNorm, start_with_relu=True, grow_first=True)
self.block12 = Block(728, 728, reps=3, stride=1, dilation=middle_block_dilation,
BatchNorm=BatchNorm, start_with_relu=True, grow_first=True)
self.block13 = Block(728, 728, reps=3, stride=1, dilation=middle_block_dilation,
BatchNorm=BatchNorm, start_with_relu=True, grow_first=True)
self.block14 = Block(728, 728, reps=3, stride=1, dilation=middle_block_dilation,
BatchNorm=BatchNorm, start_with_relu=True, grow_first=True)
self.block15 = Block(728, 728, reps=3, stride=1, dilation=middle_block_dilation,
BatchNorm=BatchNorm, start_with_relu=True, grow_first=True)
self.block16 = Block(728, 728, reps=3, stride=1, dilation=middle_block_dilation,
BatchNorm=BatchNorm, start_with_relu=True, grow_first=True)
self.block17 = Block(728, 728, reps=3, stride=1, dilation=middle_block_dilation,
BatchNorm=BatchNorm, start_with_relu=True, grow_first=True)
self.block18 = Block(728, 728, reps=3, stride=1, dilation=middle_block_dilation,
BatchNorm=BatchNorm, start_with_relu=True, grow_first=True)
self.block19 = Block(728, 728, reps=3, stride=1, dilation=middle_block_dilation,
BatchNorm=BatchNorm, start_with_relu=True, grow_first=True)
# Exit flow
self.block20 = Block(728, 1024, reps=2, stride=1, dilation=exit_block_dilations[0],
BatchNorm=BatchNorm, start_with_relu=True, grow_first=False, is_last=True)
self.conv3 = SeparableConv2d(1024, 1536, 3, stride=1, dilation=exit_block_dilations[1], BatchNorm=BatchNorm)
self.bn3 = BatchNorm(1536)
self.conv4 = SeparableConv2d(1536, 1536, 3, stride=1, dilation=exit_block_dilations[1], BatchNorm=BatchNorm)
self.bn4 = BatchNorm(1536)
self.conv5 = SeparableConv2d(1536, 2048, 3, stride=1, dilation=exit_block_dilations[1], BatchNorm=BatchNorm)
self.bn5 = BatchNorm(2048)
# Init weights
self._init_weight()
# Load pretrained model
if pretrained:
self._load_pretrained_model()
def forward(self, x):
# Entry flow
x = self.conv1(x)
x = self.bn1(x)
x = self.relu(x)
x = self.conv2(x)
x = self.bn2(x)
x = self.relu(x)
x = self.block1(x)
# add relu here
x = self.relu(x)
low_level_feat = x
x = self.block2(x)
x = self.block3(x)
# Middle flow
x = self.block4(x)
x = self.block5(x)
x = self.block6(x)
x = self.block7(x)
x = self.block8(x)
x = self.block9(x)
x = self.block10(x)
x = self.block11(x)
x = self.block12(x)
x = self.block13(x)
x = self.block14(x)
x = self.block15(x)
x = self.block16(x)
x = self.block17(x)
x = self.block18(x)
x = self.block19(x)
# Exit flow
x = self.block20(x)
x = self.relu(x)
x = self.conv3(x)
x = self.bn3(x)
x = self.relu(x)
x = self.conv4(x)
x = self.bn4(x)
x = self.relu(x)
x = self.conv5(x)
x = self.bn5(x)
x = self.relu(x)
return x, low_level_feat
def _init_weight(self):
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, SynchronizedBatchNorm2d):
m.weight.data.fill_(1)
m.bias.data.zero_()
elif isinstance(m, nn.BatchNorm2d):
m.weight.data.fill_(1)
m.bias.data.zero_()
def _load_pretrained_model(self):
pretrain_dict = model_zoo.load_url('http://data.lip6.fr/cadene/pretrainedmodels/xception-b5690688.pth')
model_dict = {}
state_dict = self.state_dict()
for k, v in pretrain_dict.items():
if k in state_dict:
if 'pointwise' in k:
v = v.unsqueeze(-1).unsqueeze(-1)
if k.startswith('block11'):
model_dict[k] = v
model_dict[k.replace('block11', 'block12')] = v
model_dict[k.replace('block11', 'block13')] = v
model_dict[k.replace('block11', 'block14')] = v
model_dict[k.replace('block11', 'block15')] = v
model_dict[k.replace('block11', 'block16')] = v
model_dict[k.replace('block11', 'block17')] = v
model_dict[k.replace('block11', 'block18')] = v
model_dict[k.replace('block11', 'block19')] = v
elif k.startswith('block12'):
model_dict[k.replace('block12', 'block20')] = v
elif k.startswith('bn3'):
model_dict[k] = v
model_dict[k.replace('bn3', 'bn4')] = v
elif k.startswith('conv4'):
model_dict[k.replace('conv4', 'conv5')] = v
elif k.startswith('bn4'):
model_dict[k.replace('bn4', 'bn5')] = v
else:
model_dict[k] = v
state_dict.update(model_dict)
self.load_state_dict(state_dict)
if __name__ == "__main__":
import torch
model = AlignedXception(BatchNorm=nn.BatchNorm2d, pretrained=True, output_stride=16)
input = torch.rand(1, 3, 512, 512)
output, low_level_feat = model(input)
print(output.size())
print(low_level_feat.size())