모두를 위한 딥러닝 시즌 2: ResNet

Uomnf97·2021년 8월 2일
0

ResNet Network 만들기

torchvision.models.resnet

  • resnet (18,34,50,101,152)를 만들 수 있도록 되어있음
  • 3x224x224 입력을 기준으로 만들도록 되어 있음

BasicBlock

  • downsample을 이용해서 연산이 이뤄질 수 잇도록 사이즈를 맞춰줌
    downsample이 없다면 pooling 과 stride로 shpae이 변화할 때 사이즈가 변해 제대로 연산할 수 없게 에러를 출력하게 됨

  • BasicBlock은 in_channel이 들어오면 오른쪽으로 빠지는 identity가 코드상에 구현되어있고 밑으로 convolution, ReLU 사이에 BatchNorm이 한번있고 다시 ReLU, Conv, BatchNorm 다음에 나와서 identity후에 out, ReLU를 통과시켜 Basic Block이 완성된다.

BottleNeck

  • downsample을 이용해서 연산이 이뤄질 수 잇도록 사이즈를 맞춰줌
    downsample이 없다면 pooling 과 stride로 shpae이 변화할 때 사이즈가 변해 제대로 연산할 수 없게 에러를 출력하게 됨

ResNet 소스코드

import torch.nn as nn
import torch.utils.model_zoo as model_zoo


__all__ = ['ResNet', 'resnet18', 'resnet34', 'resnet50', 'resnet101',
           'resnet152']


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',
}

def conv3x3(in_planes, out_planes, stride=1):
    """3x3 convolution with padding"""
    return nn.Conv2d(in_planes, out_planes, kernel_size=3, stride=stride,
                     padding=1, bias=False)


def conv1x1(in_planes, out_planes, stride=1):
    """1x1 convolution"""
    return nn.Conv2d(in_planes, out_planes, kernel_size=1, stride=stride, bias=False)
    
class BasicBlock(nn.Module):
    expansion = 1

    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.relu = nn.ReLU(inplace=True)
        self.conv2 = conv3x3(planes, planes)
        self.bn2 = nn.BatchNorm2d(planes)
        self.downsample = downsample
        self.stride = stride

    def forward(self, x):

        identity = x

        out = self.conv1(x) # 3x3 stride = 2
        out = self.bn1(out)
        out = self.relu(out)

        out = self.conv2(out) # 3x3 stride = 1
        out = self.bn2(out)

        if self.downsample is not None:
            identity = self.downsample(x)

        out += identity
        out = self.relu(out)

        return out
        
        
class Bottleneck(nn.Module):
    expansion = 4

    def __init__(self, inplanes, planes, stride=1, downsample=None):
        super(Bottleneck, self).__init__()
        self.conv1 = conv1x1(inplanes, planes) #conv1x1(64,64)
        self.bn1 = nn.BatchNorm2d(planes)
        self.conv2 = conv3x3(planes, planes, stride)#conv3x3(64,64)
        self.bn2 = nn.BatchNorm2d(planes)
        self.conv3 = conv1x1(planes, planes * self.expansion) #conv1x1(64,256)
        self.bn3 = nn.BatchNorm2d(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) # 1x1 stride = 1
        out = self.bn1(out)
        out = self.relu(out)

        out = self.conv2(out) # 3x3 stride = stride 
        out = self.bn2(out)
        out = self.relu(out)

        out = self.conv3(out) # 1x1 stride = 1
        out = self.bn3(out)

        if self.downsample is not None:
            identity = self.downsample(x)

        out += identity
        out = self.relu(out)

        return out
        
class ResNet(nn.Module):
    # model = ResNet(Bottleneck, [3, 4, 6, 3], **kwargs) #resnet 50 
    def __init__(self, block, layers, num_classes=1000, zero_init_residual=False):
        super(ResNet, self).__init__()
        
        self.inplanes = 64
               
        self.conv1 = nn.Conv2d(3, 64, kernel_size=7, stride=2, padding=3, bias=False)
        
        self.bn1 = nn.BatchNorm2d(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]'''3''')
        self.layer2 = self._make_layer(block, 128, layers[1]'''4''', stride=2)
        self.layer3 = self._make_layer(block, 256, layers[2]'''6''', stride=2)
        self.layer4 = self._make_layer(block, 512, layers[3]'''3''', stride=2)
        
        self.avgpool = nn.AdaptiveAvgPool2d((1, 1))
        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.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 _make_layer(self, block, planes, blocks, stride=1):
        
        downsample = None
        
        if stride != 1 or self.inplanes != planes * block.expansion: 
            
            downsample = nn.Sequential(
                conv1x1(self.inplanes, planes * block.expansion, stride), #conv1x1(256, 512, 2)
                nn.BatchNorm2d(planes * block.expansion), #batchnrom2d(512)
            )

        layers = []
        layers.append(block(self.inplanes, planes, stride, downsample))
        
        self.inplanes = planes * block.expansion #self.inplanes = 128 * 4
        
        for _ in range(1, blocks): 
            layers.append(block(self.inplanes, planes)) # * 3

        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)
        x = x.view(x.size(0), -1)
        x = self.fc(x)

        return x
        
def resnet18(pretrained=False, **kwargs):
    model = ResNet(BasicBlock, [2, 2, 2, 2], **kwargs) #=> 2*(2+2+2+2) +1(conv1) +1(fc)  = 16 +2 =resnet 18
    return model
    
    
def resnet50(pretrained=False, **kwargs):
    model = ResNet(Bottleneck, [3, 4, 6, 3], **kwargs) #=> 3*(3+4+6+3) +(conv1) +1(fc) = 48 +2 = 50
    return model
    
def resnet152(pretrained=False, **kwargs):
    model = ResNet(Bottleneck, [3, 8, 36, 3], **kwargs) # 3*(3+8+36+3) +2 = 150+2 = resnet152    
    return mode
    
import torchvision.models.resnet as resnet

res = resnet.resnet50()

res
profile
사회적 가치를 실현하는 프로그래머

0개의 댓글