写作动机
第一篇主要从数学建模的角度总结了近期自学神经网络的成果,但纸上得来终觉浅,不通过代码实现一遍很多东西终究只是空中楼阁。所以本篇总结主要从代码角度回顾总结自学成果。
尽量做到每日一次总结。
行文思路
仿照鱼书的风格,先给出计算图,再一步一步根据计算图搭建代码。
使用框架
因为笔者是一名医学生,所以选择现有的框架——pytorch进行代码的实现。
参考书籍
《动手学深度学习2.0》
2024/3/5——ResNet
残差网络的核心思想是:每个附加层都应该更容易地包含原始函数作为其元素之一。
同时加入批量规范化层。
确定模型
Fashion-MNIST。
计算图
Residual块(两种,第二种加入1×1的卷积层是为了调整输出通道数):
ResNet-18:
代码实现
import torch
from torch import nn
from torch.nn import functional as F
from d2l import torch as d2l
ResNet块:
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)
ResNet网络:
因为ResNet块要用很多次,所以先定义函数来简便操作:
def resnet_block(input_channels, num_channels, num_residuals, first_block=False):
blk = []
for i in range(num_residuals):
if i == 0 and not first_block:
blk.append(Residual(input_channels, num_channels, use_1x1conv=True, strides=2))
else:
blk.append(Residual(num_channels, num_channels))
return blk
接着搭建网络:
ResNet = 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),
*resnet_block(64, 64, 2, first_block=True),
*resnet_block(64, 128, 2),
*resnet_block(128, 256, 2),
*resnet_block(256, 512, 2),
nn.AdaptiveAvgPool2d((1,1)),
nn.Flatten(),
nn.Linear(512, 10)
)
应用
先确定输入和输出:
batch_size = 256
train_iter, test_iter = d2l.load_data_fashion_mnist(batch_size, resize=96)
万事俱备,只欠数据:
lr = 0.05
num_epochs = 10
d2l.train_ch6(net, train_iter, test_iter, num_epochs, lr, d2l.try_gpu())
结果是: