基于pytorch实现手写的resnet网络

'''手写的resnet网络'''
import torch
import torch.nn as nn
from torch.utils import model_zoo
from torchvision.models.video.resnet import model_urls


# 定义ResNet基础块
class BasicBlock(nn.Module):
    expansion = 1

    def __init__(self, in_planes, out_planes, stride=1, downsample=None):
        super().__init__()
        self.conv1 = nn.Conv2d(in_planes, out_planes, kernel_size=3, stride=stride, padding=1)
        self.bn1 = nn.BatchNorm2d(out_planes)
        self.relu = nn.ReLU()
        self.conv2 = nn.Conv2d(out_planes, out_planes, kernel_size=3, stride=1, padding=1)
        self.bn2 = nn.BatchNorm2d(out_planes)
        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)
        if self.downsample is not None:
            residual = self.downsample(x)
        out += residual
        out = self.relu(out)
        return out


# 定义ResNet瓶颈块
class Bottleneck(nn.Module):
    expansion = 4

    def __init__(self, in_planes, out_planes, stride=1, downsample=None):
        super().__init__()
        self.conv1 = nn.Conv2d(in_planes, out_planes, kernel_size=1, stride=stride, bias=False)
        self.bn1 = nn.BatchNorm2d(out_planes)
        self.relu = nn.ReLU()
        self.conv2 = nn.Conv2d(out_planes, out_planes, kernel_size=3, stride=1, padding=1, bias=False)
        self.bn2 = nn.BatchNorm2d(out_planes)
        self.conv3 = nn.Conv2d(out_planes, out_planes * self.expansion, kernel_size=1, bias=False)
        self.bn3 = nn.BatchNorm2d(out_planes * self.expansion)
        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 ResNet(nn.Module):
    def __init__(self, block, num_blocks, num_classes=10):
        super().__init__()
        self.in_planes = 64
        self.conv1 = nn.Conv2d(3, 64, kernel_size=3, stride=1, padding=1)
        self.bn1 = nn.BatchNorm2d(64)
        self.relu = nn.ReLU()
        self.maxpool = nn.MaxPool2d(kernel_size=3, stride=2, padding=1)
        self.layer1 = self._make_layer(block, 64, num_blocks[0], stride=1)
        self.layer2 = self._make_layer(block, 128, num_blocks[1], stride=2)
        print(self.layer1)
        print(self.layer2)
        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.avgpool = nn.AdaptiveAvgPool2d((1, 1))
        self.linear = 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)

    def forward(self, x):
        out = self.conv1(x)
        out = self.bn1(out)
        out = self.relu(out)
        out = self.maxpool(out)
        out = self.layer1(out)
        out = self.layer2(out)
        out = self.layer3(out)
        out = self.layer4(out)
        out = self.avgpool(out)
        out = out.view(out.size(0), -1)
        out = self.linear(out)
        return out

    def _make_layer(self, block, out_planes, num_blocks, stride):
        downsample = None
        if stride != 1 or self.in_planes != out_planes * block.expansion:
            downsample = nn.Sequential(
                nn.Conv2d(self.in_planes, out_planes * block.expansion, kernel_size=1, stride=stride, bias=False),
                nn.BatchNorm2d(out_planes * block.expansion)
            )
        layers = []
        layers.append(block(self.in_planes, out_planes, stride, downsample))
        for i in range(1, num_blocks):
            self.in_planes = out_planes * block.expansion
            layers.append(block(self.in_planes, out_planes))
        return nn.Sequential(*layers)


def resnet18(pretrained=False, **kwargs):
    model = ResNet(BasicBlock, [2, 2, 2, 2], **kwargs)
    if pretrained:
        state_dict = model_zoo.load_url(model_urls['resnet18'])
        model.load_state_dict({k: v for k, v in state_dict.items() if k in model.state_dict()})
        for k, v in model.state_dict().items():
            if 'fc' not in k:
                v.requires_grad = False
            else:
                v.requires_grad = True
    return model


def resnet34(pretrained=False, **kwargs):
    model = ResNet(BasicBlock, [3, 4, 6, 3], **kwargs)
    if pretrained:
        state_dict = model_zoo.load_url(model_urls['resnet34'])
        model.load_state_dict({k: v for k, v in state_dict.items() if k in model.state_dict()})
        for k, v in model.state_dict().items():
            if 'fc' not in k:
                v.requires_grad = False
            else:
                v.requires_grad = True
    return model


def resnet50(pretrained=False, **kwargs):
    model = ResNet(Bottleneck, [3, 4, 6, 3], **kwargs)
    if pretrained:
        state_dict = model_zoo.load_url(model_urls['resnet50'])
        model.load_state_dict({k: v for k, v in state_dict.items() if k in model.state_dict()})
        for k, v in model.state_dict().items():
            if 'fc' not in k:
                v.requires_grad = False
            else:
                v.requires_grad = True
    return model


def resnet101(pretrained=False, **kwargs):
    model = ResNet(Bottleneck, [3, 4, 23, 3], **kwargs)
    if pretrained:
        state_dict = model_zoo.load_url(model_urls['resnet101'])
        model.load_state_dict({k: v for k, v in state_dict.items() if k in model.state_dict()})
        for k, v in model.state_dict().items():
            if 'fc' not in k:
                v.requires_grad = False
            else:
                v.requires_grad = True
    return model


def resnet152(pretrained=False, **kwargs):
    model = ResNet(Bottleneck, [3, 8, 36, 3], **kwargs)
    if pretrained:
        state_dict = model_zoo.load_url(model_urls['resnet152'])
        model.load_state_dict({k: v for k, v in state_dict.items() if k in model.state_dict()})
        for k, v in model.state_dict().items():
            if 'fc' not in k:
                v.requires_grad = False
            else:
                v.requires_grad = True
    return model


if __name__ == "__main__":
    import torch

    x = torch.randn((32, 3, 256, 256))
    model = resnet50(False, num_classes=3)
    y = model(x)
    print(y.size())

运行结果：

Sequential(
  (0): Bottleneck(
    (conv1): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
    (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (relu): ReLU()
    (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
    (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
    (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (downsample): Sequential(
      (0): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    )
  )
  (1): Bottleneck(
    (conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
    (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (relu): ReLU()
    (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
    (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
    (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  )
  (2): Bottleneck(
    (conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
    (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (relu): ReLU()
    (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
    (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
    (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  )
)
Sequential(
  (0): Bottleneck(
    (conv1): Conv2d(256, 128, kernel_size=(1, 1), stride=(2, 2), bias=False)
    (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (relu): ReLU()
    (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
    (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
    (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (downsample): Sequential(
      (0): Conv2d(256, 512, kernel_size=(1, 1), stride=(2, 2), bias=False)
      (1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    )
  )
  (1): Bottleneck(
    (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
    (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (relu): ReLU()
    (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
    (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
    (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  )
  (2): Bottleneck(
    (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
    (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (relu): ReLU()
    (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
    (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
    (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  )
  (3): Bottleneck(
    (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
    (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (relu): ReLU()
    (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
    (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
    (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  )
)
torch.Size([32, 3])