import torch
import torch.nn as nn
from torch.hub import load_state_dict_from_url
model_urls = {
'resnet18': 'https://download.pytorch.org/models/resnet18-5c106cde.pth',
'resnet34': 'https://download.pytorch.org/models/resnet34-333f7ec4.pth',
'resnet50': 'https://download.pytorch.org/models/resnet50-19c8e357.pth',
'resnet101': 'https://download.pytorch.org/models/resnet101-5d3b4d8f.pth',
'resnet152': 'https://download.pytorch.org/models/resnet152-b121ed2d.pth',
'resnext50_32x4d': 'https://download.pytorch.org/models/resnext50_32x4d-7cdf4587.pth',
'resnext101_32x8d': 'https://download.pytorch.org/models/resnext101_32x8d-8ba56ff5.pth',
'wide_resnet50_2': 'https://download.pytorch.org/models/wide_resnet50_2-95faca4d.pth',
'wide_resnet101_2': 'https://download.pytorch.org/models/wide_resnet101_2-32ee1156.pth',
}
# 封装标准卷积
def conv3x3(in_planes, out_planes, stride=1,padding=1):
return nn.Conv2d(in_planes, out_planes, kernel_size=3, stride=stride, padding=padding, bias=False)
def conv1x1(in_planes, out_planes, stride=1):
# 当卷积层后面跟着bn层时，卷积层不需要bias
# 因为在bn层重新学习均值和方差
return nn.Conv2d(in_planes, out_planes, kernel_size=1, stride=stride, bias=False)
class BasicBolck(nn.Module):
# 通道放大倍数
expansion = 1
def __init__(self, inplanes, planes, stride=1, downsample=None, norm_layer=None):
# 调用父类的初始化函数
super(BasicBlock, self).__init__()
# 如果没有指定Batchnormalization
if norm_layer is None:
# 使用标准的BatchNorm
norm_layer = nn.BatchNorm2d
self.conv1 = conv3x3(inplanes, planes, stride)
self.bn1 = norm_layer(planes)
self.relu = nn.ReLU(inplace=True)
self.conv2 = conv3x3(planes, planes)
self.bn2 = norm_layer(planes)
self.downsample = downsample
self.stride = stride
# forward用来调用，x是输入
def forward(self, x):
indentity = x
out = self.conv1(x)
out = self.bn1(out)
out = self.relu(out)
out = self.conv2(out)
out = self.bn2(out)
# 下采样
if self.downsample is not None:
identity = self.downsample(x)
# 为什么out不下采样呢？stride=1就不下采样，stride=2已经下采样了
# f(x)+x
out += identity
# 在加和之后再调用激活函数
out = self.relu(out)
return out
class BottleNeck(nn.Module):
# 放大倍数
expansion = 4
def __init__(self, inplanes, planes, stride=1, downsample=None, norm_layer=None):
super(BottleNeck, self).__init___()
if norm_layer is None:
norm_layer = nn.BatchNorm2d
self.conv1 = conv1x1(inplanes, planes)
self.bn1 = norm_layer(planes)
self.conv2 = conv3x3(planes, planes)
self.bn2 = norm_layer(planes)
self.conv3 = conv1x1(planes, planes * self.expansion)
self.bn3 = norm_layer(planes * self.expansion)
self.relu = nn.ReLU(inplace=True)
self.downsample = downsample
self.stride = stride
def forward(self, x):
identity = 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.downasample is not None:
identity = self.downsample(x)
out += identity
out = self.relu(out)
return out
class ResNet(nn.Module):
# num_class 分多少类
def __init__(self, block, layers, num_class=1000, norm_layer=None):
super(ResNet, self).__init__()
if norm_layer is None:
norm_layer = nn.BatchNorm2d
self.inplanes = 64
self.conv1 = nn.Conv2d(3, self.inplanes, kernel_size=7, stride=2, padding=3, bias=False)
self.bn1 = norm_layer(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)
self.layer3 = self._make_layer(block, 256, layers[2], stride=2)
self.layer4 = self._make_layer(block, 512, layers[3], stride=2)
# 平均池化
self.avgpool = nn.AdaptiveAvgPool2d((1,1))
# 分类层
self.fc = nn.Linear(512*block.expansion, num_class)
# 参数初始化
for m in self.modules():
# 如果是卷积层
if isinstance(m, nn.Conv2d):
# 凯明初始化
# 主要的应用场景是搭配RElu激活函数
nn.init.kaiming_normal_(m.weight, mode='fan_out', nonlinearity='relu')
# 如果是Batch层
elif isinstance(m, (nn.BatchNorm2d, nn.GroupNorm)):
# 常量初始化 把权重和误差初始化为0,1
nn.init.constant_(m.weight, 1)
nn.init.constant_(m.bias, 0)
def _make_layer(self, block, stride=1):
# 生成stage里的内容
norm_layer = self._norm_layer
downsample = None
# 判断是否需要下采样 stride!=1发生下采样
if stride != 1 or self.inplanes != planes*bloack.expansion:
# downsample 用1x1的卷积来调整维度
# downsample是一个小的网络，由1x1卷积层和normlazition组成
downsample = nn.Sequential(conv1x1(self.inplanes, planes*block.expansion, stride),norm_layer(planes*block.expansion))
layers=[]
layers.append(block(self.inplanes, planes, stride, downsample, norm_layer))
self.inplanes = planes * self.expansion
for _ in range(1, blocks):
layer.append(block(self.inplanes, planes, norm_layer=norm_layer))
# *list 是一个语法， 对list降维
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)
x = self.avgpool(x)
# 展平
x = torch.flattern(x,1)
x = self.fc(x)
return x
# 封装预加载训练
def _resnet(arch, block, layers, pretrained, progress, **kwargs):
model = ResNet(block, layers, **kwargs)
# 如果需要预训练
if pretrained:
state_dict = load_state_dict_from_url(model_urls[arch], progress=progress)
model.load_state_dict(state_dict)
return model
def resnet(pretrained=False, progress=True, **kwargs):
return _resnet('resnet152', BottleNeck, [3,8,36,3], pretrained, progress, **kwargs)
model = resnet152(pretrained = True)
model.eval()
