论文要点：ResNet（模型M+公式F+代码C）

# ResNet

论文：Deep Residual Learning for Image Recognition by Kaiming He, et al

1. 模型

图片来源：动手学深度学习 by 李沐

2. 代码

import torch
from torch import nn
from torch.nn import functional as F
from d2l import torch as d2l

class Residual(nn.Module):
    def __init__(self, input_channels, num_channels, use_1x1conv=False, strides=1):
        super().__init__()
        self.conv1 = nn.Conv2d(input_channels, num_channels, kernel_size=3, padding=1, stride=strides)
        self.conv2 = nn.Conv2d(num_channels, num_channels, kernel_size=3, padding=1)
        if use_1x1conv:
            self.conv3 = nn.Conv2d(input_channels, num_channels, kernel_size=1, stride=strides)
        else:
            self.conv3 = None
        self.bn1 = nn.BatchNorm2d(num_channels)
        self.bn2 = nn.BatchNorm2d(num_channels)
    
    def forward(self, X):
        Y = F.relu(self.bn1(self.conv1(X)))
        Y = self.bn2(self.conv2(Y))
        if self.conv3:
            X = self.conv3(X)
        Y += X
        return F.relu(Y)

# Residual 输入输出维度相同
blk = Residual(3, 3)
X = torch.randn(4, 3, 6, 6)
Y = blk(X)
Y.shape

blk = Residual(3, 3, use_1x1conv=True, strides=2)
blk(X).shape

# res_block包含多个Residual块
def res_block(input_channels, num_channels, num_residual, first_block=False):
    blk = []
    for i in range(num_residual):
        if i == 0 and not first_block:
            blk.append(Residual(input_channels, num_channels, use_1x1conv=True, strides=2))
        else:
            blk.append(Residual(input_channels, num_channels))
    return blk

# ResNet模型, 
# 输出尺寸：224x224，通道数：1
b1 = nn.Sequential(nn.Conv2d(1, 64, kernel_size=7, stride=2, padding=3), nn.BatchNorm2d(64), nn.ReLU(),
                   nn.MaxPool2d(kernel_size=3, stride=2, padding=1))
# 输出尺寸：112x112，通道数：64
b2 = nn.Sequential(*res_block(64, 64, 2, first_block=True))
# 输出尺寸：56x56，通道数：64
b3 = nn.Sequential(*res_block(64, 128, 2))
# 输出尺寸：28x28，通道数：128
b4 = nn.Sequential(*res_block(128, 256, 2))
# 输出尺寸：14x14，通道数：256
b5 = nn.Sequential(*res_block(256, 512, 2))
# 输出尺寸：7x7，通道数：512

net = nn.Sequential(b1, 
                    b2, 
                    b3, 
                    b4, 
                    b5, 
                    nn.AdaptiveAvgPool2d((1, 1)), nn.Flatten(), 
                    nn.Linear(512, 10))

X = torch.randn(1, 1, 224, 224)
for layer in net:
    X = layer(X)
    print(layer.__class__.__name__, ":",  X.shape)