PyTorch - FashionMNIST + LeNet / AlexNet / VGG / GooLeNet / NiN / ResNet / DenseNet

AI工程仔

已于 2023-03-22 14:25:33 修改

阅读量912

点赞数

文章标签： pytorch FashionMNIST Lenet VGG AlexNet

于 2023-03-18 12:06:14 首次发布

原文链接：https://zh.d2l.ai/chapter_convolutional-modern/alexnet.html

版权

项目说明

参考: D2L - 现代卷积神经网络
https://zh.d2l.ai/chapter_convolutional-modern/alexnet.html

数据集说明 - FashionMNIST

Fashion-MNIST 由10个类别的图像组成，每个类别由训练数据集（train dataset）中的6000张图像和测试数据集（test dataset）中的1000张图像组成。
图像是一个 28*28 的灰度图像。
十个类别是：t-shirt, trouser, pillover, dress, coat, sandal, shirt, sneaker, bag, ankle boot

算法说明 - LeNet-5

这里使用 LeNet-5 来实现 Fashion-MNIST 服装图⽚的分类任务。

LeNet 是最早发布的卷积神经网络之一，是由AT&T贝尔实验室的研究员Yann LeCun在1989年提出的（并以其命名），目的是识别图像中的手写数字。
当时，Yann LeCun发表了第一篇通过反向传播成功训练卷积神经网络的研究，这项工作代表了十多年来神经网络研究开发的成果。
LeNet取得了与支持向量机（support vector machines）性能相媲美的成果，成为监督学习的主流方法。 LeNet被广泛用于自动取款机（ATM）机中，帮助识别处理支票的数字。时至今日，一些自动取款机仍在运行Yann LeCun和他的同事Leon Bottou在上世纪90年代写的代码。

LeNet-5 网络结构

总体来看，LeNet-5 由两个部分组成：

卷积编码器：由两个卷积层组成;
全连接层密集块：由三个全连接层组成。

下图以输入尺寸为 28*28 单通道的图片，输出10分类为例：
在这里插入图片描述

在这里插入图片描述

说明：

每个卷积块中的基本单元是⼀个卷积层、⼀个 sigmoid 激活函数和平均汇聚层。
每个卷积层使⽤ 5 × 5 卷积核和⼀个 sigmoid 激活函数。
第⼀卷积层有 6 个输出通道，第⼆个卷积层有16个输出通道。
每个 2 ×2 的池操作（步幅2）通过空间下采样将维数减少4倍。
卷积的输出形状由批量⼤⼩、通道数、⾼度、宽度决定。
为了将卷积块的输出传递给稠密块，我们必须在⼩批量中展平每个样本。
换⾔之，我们将这个四维输⼊转换成全连接层所期望的⼆维输⼊。
这⾥的⼆维表⽰的第⼀个维度索引⼩批量中的样本，第⼆个维度给出每个样本的平⾯向量表⽰。
LeNet-5 的稠密块有三个全连接层，分别有120、84和10个输出。
因为我们在执⾏分类任务，所以输出层的10维对应于最后输出结果的数量。

代码实现

引入框架

import os
import numpy as np
import pandas as pd
import torch
import torch.nn as nn
import torch.optim as optim
from torch.utils.data import Dataset, DataLoader
from torchvision import transforms
from torchvision import datasets
import matplotlib.pyplot as plt

数据准备

下载数据集


data_transform = transforms.Compose([
    #transforms.ToPILImage(),  
     # 这一步取决于后续的数据读取方式，如果使用内置数据集读取方式则不需要
    transforms.Resize(image_size),
    transforms.ToTensor()])
 
train_data = datasets.FashionMNIST(root=task_dir, train=True, download=True, transform=data_transform)
test_data = datasets.FashionMNIST(root=task_dir, train=False, download=True, transform=data_transform)

train_loader = DataLoader(train_data, batch_size=batch_size, shuffle=True, num_workers=num_workers, drop_last=True)
test_loader = DataLoader(test_data, batch_size=batch_size, shuffle=False, num_workers=num_workers)

查看数据

print(len(mnist_test))
print(len(mnist_train))

mnist_train[0][0].shape # 取出第一张图片
# torch.Size([1, 28, 28])  # 黑白图片 28 * 28


image, label = next(iter(train_loader))
print(image.shape, label.shape)
plt.imshow(image[0][0], cmap="gray")

在这里插入图片描述

定义网络

class Net(nn.Module):
    def __init__(self):
        super(Net, self).__init__()
        self.conv = nn.Sequential(
            nn.Conv2d(1, 32, 5),
            nn.ReLU(),
            nn.MaxPool2d(2, stride=2),
            nn.Dropout(0.3),
            
            nn.Conv2d(32, 64, 5),
            nn.ReLU(),
            nn.MaxPool2d(2, stride=2),
            nn.Dropout(0.3)
        )
        
        self.fc = nn.Sequential(
            nn.Linear(64*4*4, 512),
            nn.ReLU(),
            nn.Linear(512, 10)
        )
        
    def forward(self, x):
        x = self.conv(x)
        x = x.view(-1, 64*4*4)
        x = self.fc(x)
        x = nn.functional.normalize(x)
        return x

也有教程将网络定义为：

class LeNet(nn.Module):
    def __init__(self):
        super(LeNet, self).__init__()
        self.conv = nn.Sequential(
            nn.Conv2d(1, 6, 5), # in_channels, out_channels, kernel_size
            nn.Sigmoid(),
            nn.MaxPool2d(2, 2), # kernel_size, stride
            
            nn.Conv2d(6, 16, 5),
            nn.Sigmoid(),
            nn.MaxPool2d(2, 2)
        )
        
        self.fc = nn.Sequential(
            nn.Linear(16*4*4, 120),
            nn.Sigmoid(),
            nn.Linear(120, 84),
            nn.Sigmoid(),
            nn.Linear(84, 10)
        )

    def forward(self, img):
        feature = self.conv(img)
        output = self.fc(feature.view(img.shape[0], -1))
        return output

训练

设置参数

device = "cuda:0" if torch.cuda.is_available() else "cpu"
 
batch_size = 256
num_workers = 0   # 对于Windows用户，这里应设置为0，否则会出现多线程错误
lr = 1e-4
epochs = 5
image_size = 28

model = Net()
model.to(device)
print(model) 
criterion = nn.CrossEntropyLoss()
optimizer = optim.Adam(model.parameters(), lr=0.001)

# model 
Net(
  (conv): Sequential(
    (0): Conv2d(1, 32, kernel_size=(5, 5), stride=(1, 1))
    (1): ReLU()
    (2): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
    (3): Dropout(p=0.3, inplace=False)
    (4): Conv2d(32, 64, kernel_size=(5, 5), stride=(1, 1))
    (5): ReLU()
    (6): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
    (7): Dropout(p=0.3, inplace=False)
  )
  (fc): Sequential(
    (0): Linear(in_features=1024, out_features=512, bias=True)
    (1): ReLU()
    (2): Linear(in_features=512, out_features=10, bias=True)
  )
)

训练方法


def train(epoch):

    model.train()
    train_loss = 0
    
    for data, label in train_loader:
        data = data.to(device)
        label = label.to(device)
        
        optimizer.zero_grad()
        output = model(data)
        
        loss = criterion(output, label)
        loss.backward() 
        optimizer.step()
        
        train_loss += loss.item()*data.size(0)
    
    train_loss = train_loss/len(train_loader.dataset)
    print('\n-- Epoch: {} \t Training Loss: {:.6f}'.format(epoch, train_loss))

验证方法

def val(epoch):       
    model.eval()
    val_loss = 0
    gt_labels = []
    pred_labels = []
    
    with torch.no_grad():
        for data, label in test_loader:
            #data, label = data.cuda(), label.cuda()
            output = model(data)
            preds = torch.argmax(output, 1)
            
            gt_labels.append(label.cpu().data.numpy())
            pred_labels.append(preds.cpu().data.numpy())
            
            loss = criterion(output, label)
            val_loss += loss.item()*data.size(0)
            
    val_loss = val_loss/len(test_loader.dataset)
    gt_labels, pred_labels = np.concatenate(gt_labels), np.concatenate(pred_labels)
    acc = np.sum(gt_labels==pred_labels)/len(pred_labels)
    
    print('Epoch: {} \tValidation Loss: {:.6f}, Accuracy: {:6f}'.format(epoch, val_loss, acc))

执行训练

for epoch in range(1, epochs+1):
      train(epoch)
      val(epoch)

其他网络实现

AlexNet

class AlexNet(nn.Module):
    def __init__(self):
        super(AlexNet, self).__init__()
        self.conv = nn.Sequential(
            nn.Conv2d(1, 96, 11, 4), # in_channels, out_channels, kernel_size, stride, padding
            nn.ReLU(),
            nn.MaxPool2d(3, 2), # kernel_size, stride
            # 减小卷积窗口，使用填充为2来使得输入与输出的高和宽一致，且增大输出通道数
            nn.Conv2d(96, 256, 5, 1, 2),
            nn.ReLU(),
            nn.MaxPool2d(3, 2),
            # 连续3个卷积层，且使用更小的卷积窗口。除了最后的卷积层外，进一步增大了输出通道数。
            # 前两个卷积层后不使用池化层来减小输入的高和宽
            nn.Conv2d(256, 384, 3, 1, 1),
            nn.ReLU(),
            nn.Conv2d(384, 384, 3, 1, 1),
            nn.ReLU(),
            nn.Conv2d(384, 256, 3, 1, 1),
            nn.ReLU(),
            nn.MaxPool2d(3, 2)
        )
         # 这里全连接层的输出个数比LeNet中的大数倍。使用丢弃层来缓解过拟合
        self.fc = nn.Sequential(
            nn.Linear(256*5*5, 4096),
            nn.ReLU(),
            nn.Dropout(0.5),
            nn.Linear(4096, 4096),
            nn.ReLU(),
            nn.Dropout(0.5),
            # 输出层。由于这里使用Fashion-MNIST，所以用类别数为10，而非论文中的1000
            nn.Linear(4096, 10),
        )

    def forward(self, img):
        feature = self.conv(img)
        output = self.fc(feature.view(img.shape[0], -1))
        return output

net = AlexNet()
print(net)

AlexNet(
  (conv): Sequential(
    (0): Conv2d(1, 96, kernel_size=(11, 11), stride=(4, 4))
    (1): ReLU()
    (2): MaxPool2d(kernel_size=3, stride=2, padding=0, dilation=1, ceil_mode=False)
    (3): Conv2d(96, 256, kernel_size=(5, 5), stride=(1, 1), padding=(2, 2))
    (4): ReLU()
    (5): MaxPool2d(kernel_size=3, stride=2, padding=0, dilation=1, ceil_mode=False)
    (6): Conv2d(256, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
    (7): ReLU()
    (8): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
    (9): ReLU()
    (10): Conv2d(384, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
    (11): ReLU()
    (12): MaxPool2d(kernel_size=3, stride=2, padding=0, dilation=1, ceil_mode=False)
  )
  (fc): Sequential(
    (0): Linear(in_features=6400, out_features=4096, bias=True)
    (1): ReLU()
    (2): Dropout(p=0.5)
    (3): Linear(in_features=4096, out_features=4096, bias=True)
    (4): ReLU()
    (5): Dropout(p=0.5)
    (6): Linear(in_features=4096, out_features=10, bias=True)
  )
)

LeNet

class LeNet(nn.Module):
    def __init__(self):
        super(LeNet, self).__init__()
        self.conv = nn.Sequential(
            nn.Conv2d(1, 6, 5), # in_channels, out_channels, kernel_size
            nn.Sigmoid(),
            nn.MaxPool2d(2, 2), # kernel_size, stride
            nn.Conv2d(6, 16, 5),
            nn.Sigmoid(),
            nn.MaxPool2d(2, 2)
        )
        self.fc = nn.Sequential(
            nn.Linear(16*4*4, 120),
            nn.Sigmoid(),
            nn.Linear(120, 84),
            nn.Sigmoid(),
            nn.Linear(84, 10)
        )

    def forward(self, img):
        feature = self.conv(img)
        output = self.fc(feature.view(img.shape[0], -1))
        return output

VGG

VGG块

def vgg_block(num_convs, in_channels, out_channels):
    blk = []
    
    for i in range(num_convs):
        if i == 0:
            blk.append(nn.Conv2d(in_channels, out_channels, kernel_size=3, padding=1))
        else:
            blk.append(nn.Conv2d(out_channels, out_channels, kernel_size=3, padding=1))
            
        blk.append(nn.ReLU())
        
    blk.append(nn.MaxPool2d(kernel_size=2, stride=2))
    return nn.Sequential(*blk)

VGG网络

conv_arch = ((1, 1, 64), (1, 64, 128), (2, 128, 256), (2, 256, 512), (2, 512, 512))
fc_features = 512 * 7 * 7 # 根据卷积层的输出算出来的
fc_hidden_units = 4096 # 任意

def vgg(conv_arch, fc_features, fc_hidden_units=4096):
    net = nn.Sequential()
    # 卷积层部分
    for i, (num_convs, in_channels, out_channels) in enumerate(conv_arch):
        net.add_module("vgg_block_" + str(i+1), vgg_block(num_convs, in_channels, out_channels))
    # 全连接层部分
    net.add_module("fc", nn.Sequential(d2l.FlattenLayer(),
                                 nn.Linear(fc_features, fc_hidden_units),
                                 nn.ReLU(),
                                 nn.Dropout(0.5),
                                 nn.Linear(fc_hidden_units, fc_hidden_units),
                                 nn.ReLU(),
                                 nn.Dropout(0.5),
                                 nn.Linear(fc_hidden_units, 10)
                                ))
    return net

net = vgg(conv_arch, fc_features, fc_hidden_units)
X = torch.rand(1, 1, 224, 224)

# named_children获取一级子模块及其名字(named_modules会返回所有子模块,包括子模块的子模块)
for name, blk in net.named_children(): 
    X = blk(X)
    print(name, 'output shape: ', X.shape)

打印输出：

vgg_block_1 output shape:  torch.Size([1, 64, 112, 112])
vgg_block_2 output shape:  torch.Size([1, 128, 56, 56])
vgg_block_3 output shape:  torch.Size([1, 256, 28, 28])
vgg_block_4 output shape:  torch.Size([1, 512, 14, 14])
vgg_block_5 output shape:  torch.Size([1, 512, 7, 7])
fc output shape:  torch.Size([1, 10])

ratio = 8
small_conv_arch = [(1, 1, 64//ratio), (1, 64//ratio, 128//ratio), (2, 128//ratio, 256//ratio), 
                   (2, 256//ratio, 512//ratio), (2, 512//ratio, 512//ratio)]

net = vgg(small_conv_arch, fc_features // ratio, fc_hidden_units // ratio)
print(net)

打印网络结构

Sequential(
  (vgg_block_1): Sequential(
    (0): Conv2d(1, 8, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
    (1): ReLU()
    (2): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
  )
  (vgg_block_2): Sequential(
    (0): Conv2d(8, 16, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
    (1): ReLU()
    (2): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
  )
  (vgg_block_3): Sequential(
    (0): Conv2d(16, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
    (1): ReLU()
    (2): Conv2d(32, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
    (3): ReLU()
    (4): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
  )
  (vgg_block_4): Sequential(
    (0): Conv2d(32, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
    (1): ReLU()
    (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
    (3): ReLU()
    (4): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
  )
  (vgg_block_5): Sequential(
    (0): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
    (1): ReLU()
    (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
    (3): ReLU()
    (4): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
  )
  (fc): Sequential(
    (0): FlattenLayer()
    (1): Linear(in_features=3136, out_features=512, bias=True)
    (2): ReLU()
    (3): Dropout(p=0.5)
    (4): Linear(in_features=512, out_features=512, bias=True)
    (5): ReLU()
    (6): Dropout(p=0.5)
    (7): Linear(in_features=512, out_features=10, bias=True)
  )
)

GoogLeNet

GoogLeNet ：含并行连结的网络

Inception 块

class Inception(nn.Module):
    # c1 - c4为每条线路里的层的输出通道数
    def __init__(self, in_c, c1, c2, c3, c4):
        super(Inception, self).__init__()
        # 线路1，单1 x 1卷积层
        self.p1_1 = nn.Conv2d(in_c, c1, kernel_size=1)
        # 线路2，1 x 1卷积层后接3 x 3卷积层
        self.p2_1 = nn.Conv2d(in_c, c2[0], kernel_size=1)
        self.p2_2 = nn.Conv2d(c2[0], c2[1], kernel_size=3, padding=1)
        # 线路3，1 x 1卷积层后接5 x 5卷积层
        self.p3_1 = nn.Conv2d(in_c, c3[0], kernel_size=1)
        self.p3_2 = nn.Conv2d(c3[0], c3[1], kernel_size=5, padding=2)
        # 线路4，3 x 3最大池化层后接1 x 1卷积层
        self.p4_1 = nn.MaxPool2d(kernel_size=3, stride=1, padding=1)
        self.p4_2 = nn.Conv2d(in_c, c4, kernel_size=1)

    def forward(self, x):
        p1 = F.relu(self.p1_1(x))
        p2 = F.relu(self.p2_2(F.relu(self.p2_1(x))))
        p3 = F.relu(self.p3_2(F.relu(self.p3_1(x))))
        p4 = F.relu(self.p4_2(self.p4_1(x)))
        return torch.cat((p1, p2, p3, p4), dim=1)  # 在通道维上连结输出

GoogLeNet模型

b1 = nn.Sequential(nn.Conv2d(1, 64, kernel_size=7, stride=2, padding=3),
                   nn.ReLU(),
                   nn.MaxPool2d(kernel_size=3, stride=2, padding=1))

b2 = nn.Sequential(nn.Conv2d(64, 64, kernel_size=1),
                   nn.Conv2d(64, 192, kernel_size=3, padding=1),
                   nn.MaxPool2d(kernel_size=3, stride=2, padding=1))
                   
b3 = nn.Sequential(Inception(192, 64, (96, 128), (16, 32), 32),
                   Inception(256, 128, (128, 192), (32, 96), 64),
                   nn.MaxPool2d(kernel_size=3, stride=2, padding=1))
                   
b4 = nn.Sequential(Inception(480, 192, (96, 208), (16, 48), 64),
                   Inception(512, 160, (112, 224), (24, 64), 64),
                   Inception(512, 128, (128, 256), (24, 64), 64),
                   Inception(512, 112, (144, 288), (32, 64), 64),
                   Inception(528, 256, (160, 320), (32, 128), 128),
                   nn.MaxPool2d(kernel_size=3, stride=2, padding=1))

b5 = nn.Sequential(Inception(832, 256, (160, 320), (32, 128), 128),
                   Inception(832, 384, (192, 384), (48, 128), 128),
                   d2l.GlobalAvgPool2d())

net = nn.Sequential(b1, b2, b3, b4, b5, d2l.FlattenLayer(), nn.Linear(1024, 10))
X = torch.rand(1, 1, 96, 96)
for blk in net.children(): 
    X = blk(X)
    print('output shape: ', X.shape)

output shape:  torch.Size([1, 64, 24, 24])
output shape:  torch.Size([1, 192, 12, 12])
output shape:  torch.Size([1, 480, 6, 6])
output shape:  torch.Size([1, 832, 3, 3])
output shape:  torch.Size([1, 1024, 1, 1])
output shape:  torch.Size([1, 1024])
output shape:  torch.Size([1, 10])

NiN

网络中的网络

NiN块

def nin_block(in_channels, out_channels, kernel_size, stride, padding):
    blk = nn.Sequential(nn.Conv2d(in_channels, out_channels, kernel_size, stride, padding),
                        nn.ReLU(),
                        nn.Conv2d(out_channels, out_channels, kernel_size=1),
                        nn.ReLU(),
                        nn.Conv2d(out_channels, out_channels, kernel_size=1),
                        nn.ReLU())
    return blk

NiN模型

net = nn.Sequential(
    nin_block(1, 96, kernel_size=11, stride=4, padding=0),
    nn.MaxPool2d(kernel_size=3, stride=2),
    nin_block(96, 256, kernel_size=5, stride=1, padding=2),
    nn.MaxPool2d(kernel_size=3, stride=2),
    nin_block(256, 384, kernel_size=3, stride=1, padding=1),
    nn.MaxPool2d(kernel_size=3, stride=2), 
    nn.Dropout(0.5),
    # 标签类别数是10
    nin_block(384, 10, kernel_size=3, stride=1, padding=1),
    # 全局平均池化层可通过将窗口形状设置成输入的高和宽实现
    nn.AvgPool2d(kernel_size=5),
    # 将四维的输出转成二维的输出，其形状为(批量大小, 10)
    d2l.FlattenLayer())

X = torch.rand(1, 1, 224, 224)

for name, blk in net.named_children(): 
    X = blk(X)
    print(name, 'output shape: ', X.shape)

0 output shape:  torch.Size([1, 96, 54, 54])
1 output shape:  torch.Size([1, 96, 26, 26])
2 output shape:  torch.Size([1, 256, 26, 26])
3 output shape:  torch.Size([1, 256, 12, 12])
4 output shape:  torch.Size([1, 384, 12, 12])
5 output shape:  torch.Size([1, 384, 5, 5])
6 output shape:  torch.Size([1, 384, 5, 5])
7 output shape:  torch.Size([1, 10, 5, 5])
8 output shape:  torch.Size([1, 10, 1, 1])
9 output shape:  torch.Size([1, 10])

batch-norm

def batch_norm(is_training, X, gamma, beta, moving_mean, moving_var, eps, momentum):
    # 判断当前模式是训练模式还是预测模式
    if not is_training:
        # 如果是在预测模式下，直接使用传入的移动平均所得的均值和方差
        X_hat = (X - moving_mean) / torch.sqrt(moving_var + eps)
    else:
        assert len(X.shape) in (2, 4)
        if len(X.shape) == 2:
            # 使用全连接层的情况，计算特征维上的均值和方差
            mean = X.mean(dim=0)
            var = ((X - mean) ** 2).mean(dim=0)
        else:
            # 使用二维卷积层的情况，计算通道维上（axis=1）的均值和方差。这里我们需要保持
            # X的形状以便后面可以做广播运算
            mean = X.mean(dim=0, keepdim=True).mean(dim=2, keepdim=True).mean(dim=3, keepdim=True)
            var = ((X - mean) ** 2).mean(dim=0, keepdim=True).mean(dim=2, keepdim=True).mean(dim=3, keepdim=True)
        # 训练模式下用当前的均值和方差做标准化
        X_hat = (X - mean) / torch.sqrt(var + eps)
        # 更新移动平均的均值和方差
        moving_mean = momentum * moving_mean + (1.0 - momentum) * mean
        moving_var = momentum * moving_var + (1.0 - momentum) * var
    Y = gamma * X_hat + beta  # 拉伸和偏移
    return Y, moving_mean, moving_var

class BatchNorm(nn.Module):
    def __init__(self, num_features, num_dims):
        super(BatchNorm, self).__init__()
        if num_dims == 2:
            shape = (1, num_features)
        else:
            shape = (1, num_features, 1, 1)
        # 参与求梯度和迭代的拉伸和偏移参数，分别初始化成0和1
        self.gamma = nn.Parameter(torch.ones(shape))
        self.beta = nn.Parameter(torch.zeros(shape))
        # 不参与求梯度和迭代的变量，全在内存上初始化成0
        self.moving_mean = torch.zeros(shape)
        self.moving_var = torch.zeros(shape)

    def forward(self, X):
        # 如果X不在内存上，将moving_mean和moving_var复制到X所在显存上
        if self.moving_mean.device != X.device:
            self.moving_mean = self.moving_mean.to(X.device)
            self.moving_var = self.moving_var.to(X.device)
        # 保存更新过的moving_mean和moving_var, Module实例的traning属性默认为true, 调用.eval()后设成false
        Y, self.moving_mean, self.moving_var = batch_norm(self.training, 
            X, self.gamma, self.beta, self.moving_mean,
            self.moving_var, eps=1e-5, momentum=0.9)
        return Y

使用批量归一化层的LeNet

net = nn.Sequential(
            nn.Conv2d(1, 6, 5), # in_channels, out_channels, kernel_size
            BatchNorm(6, num_dims=4),
            nn.Sigmoid(),
            nn.MaxPool2d(2, 2), # kernel_size, stride
            nn.Conv2d(6, 16, 5),
            BatchNorm(16, num_dims=4),
            nn.Sigmoid(),
            nn.MaxPool2d(2, 2),
            d2l.FlattenLayer(),
            nn.Linear(16*4*4, 120),
            BatchNorm(120, num_dims=2),
            nn.Sigmoid(),
            nn.Linear(120, 84),
            BatchNorm(84, num_dims=2),
            nn.Sigmoid(),
            nn.Linear(84, 10)
        )

batch_size = 256
train_iter, test_iter = d2l.load_data_fashion_mnist(batch_size=batch_size)

lr, num_epochs = 0.001, 5
optimizer = torch.optim.Adam(net.parameters(), lr=lr)
d2l.train_ch5(net, train_iter, test_iter, batch_size, optimizer, device, num_epochs)

training on  cuda
epoch 1, loss 0.0039, train acc 0.790, test acc 0.835, time 2.9 sec
epoch 2, loss 0.0018, train acc 0.866, test acc 0.821, time 3.2 sec
epoch 3, loss 0.0014, train acc 0.879, test acc 0.857, time 2.6 sec
epoch 4, loss 0.0013, train acc 0.886, test acc 0.820, time 2.7 sec
epoch 5, loss 0.0012, train acc 0.891, test acc 0.859, time 2.8 sec

net[1].gamma.view((-1,)), net[1].beta.view((-1,))

(tensor([ 1.2537,  1.2284,  1.0100,  1.0171,  0.9809,  1.1870], device='cuda:0'),
 tensor([ 0.0962,  0.3299, -0.5506,  0.1522, -0.1556,  0.2240], device='cuda:0'))

batch-norm 简洁实现

net = nn.Sequential(
            nn.Conv2d(1, 6, 5), # in_channels, out_channels, kernel_size
            nn.BatchNorm2d(6),
            nn.Sigmoid(),
            nn.MaxPool2d(2, 2), # kernel_size, stride
            nn.Conv2d(6, 16, 5),
            nn.BatchNorm2d(16),
            nn.Sigmoid(),
            nn.MaxPool2d(2, 2),
            d2l.FlattenLayer(),
            nn.Linear(16*4*4, 120),
            nn.BatchNorm1d(120),
            nn.Sigmoid(),
            nn.Linear(120, 84),
            nn.BatchNorm1d(84),
            nn.Sigmoid(),
            nn.Linear(84, 10)
        )

batch_size = 256
train_iter, test_iter = d2l.load_data_fashion_mnist(batch_size=batch_size)

lr, num_epochs = 0.001, 5
optimizer = torch.optim.Adam(net.parameters(), lr=lr)
d2l.train_ch5(net, train_iter, test_iter, batch_size, optimizer, device, num_epochs)

training on  cuda
epoch 1, loss 0.0054, train acc 0.767, test acc 0.795, time 2.0 sec
epoch 2, loss 0.0024, train acc 0.851, test acc 0.748, time 2.0 sec
epoch 3, loss 0.0017, train acc 0.872, test acc 0.814, time 2.2 sec
epoch 4, loss 0.0014, train acc 0.883, test acc 0.818, time 2.1 sec
epoch 5, loss 0.0013, train acc 0.889, test acc 0.734, time 1.8 sec

ResNet 残差网络

残差块

class Residual(nn.Module):  # 本类已保存在d2lzh_pytorch包中方便以后使用
    def __init__(self, in_channels, out_channels, use_1x1conv=False, stride=1):
        super(Residual, self).__init__()
        self.conv1 = nn.Conv2d(in_channels, out_channels, kernel_size=3, padding=1, stride=stride)
        self.conv2 = nn.Conv2d(out_channels, out_channels, kernel_size=3, padding=1)
        if use_1x1conv:
            self.conv3 = nn.Conv2d(in_channels, out_channels, kernel_size=1, stride=stride)
        else:
            self.conv3 = None
        self.bn1 = nn.BatchNorm2d(out_channels)
        self.bn2 = nn.BatchNorm2d(out_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)
        return F.relu(Y + X)

blk = Residual(3, 3)
X = torch.rand((4, 3, 6, 6))
blk(X).shape # torch.Size([4, 3, 6, 6])

blk = Residual(3, 6, use_1x1conv=True, stride=2)
blk(X).shape # torch.Size([4, 6, 3, 3])

net = 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)
      )

def resnet_block(in_channels, out_channels, num_residuals, first_block=False):
    if first_block:
        assert in_channels == out_channels # 第一个模块的通道数同输入通道数一致
    blk = []
    for i in range(num_residuals):
        if i == 0 and not first_block:
            blk.append(Residual(in_channels, out_channels, use_1x1conv=True, stride=2))
        else:
            blk.append(Residual(out_channels, out_channels))
    return nn.Sequential(*blk)

net.add_module("resnet_block1", resnet_block(64, 64, 2, first_block=True))
net.add_module("resnet_block2", resnet_block(64, 128, 2))
net.add_module("resnet_block3", resnet_block(128, 256, 2))
net.add_module("resnet_block4", resnet_block(256, 512, 2))

net.add_module("global_avg_pool", d2l.GlobalAvgPool2d()) # GlobalAvgPool2d的输出: (Batch, 512, 1, 1)
net.add_module("fc", nn.Sequential(d2l.FlattenLayer(), nn.Linear(512, 10)))

X = torch.rand((1, 1, 224, 224))
for name, layer in net.named_children():
    X = layer(X)
    print(name, ' output shape:\t', X.shape)

0  output shape:	 torch.Size([1, 64, 112, 112])
1  output shape:	 torch.Size([1, 64, 112, 112])
2  output shape:	 torch.Size([1, 64, 112, 112])
3  output shape:	 torch.Size([1, 64, 56, 56])
resnet_block1  output shape:	 torch.Size([1, 64, 56, 56])
resnet_block2  output shape:	 torch.Size([1, 128, 28, 28])
resnet_block3  output shape:	 torch.Size([1, 256, 14, 14])
resnet_block4  output shape:	 torch.Size([1, 512, 7, 7])
global_avg_pool  output shape:	 torch.Size([1, 512, 1, 1])
fc  output shape:	 torch.Size([1, 10])

DenseNet

稠密块

def conv_block(in_channels, out_channels):
    blk = nn.Sequential(nn.BatchNorm2d(in_channels), 
                        nn.ReLU(),
                        nn.Conv2d(in_channels, out_channels, kernel_size=3, padding=1))
    return blk

class DenseBlock(nn.Module):
    def __init__(self, num_convs, in_channels, out_channels):
        super(DenseBlock, self).__init__()
        net = []
        for i in range(num_convs):
            in_c = in_channels + i * out_channels
            net.append(conv_block(in_c, out_channels))
        self.net = nn.ModuleList(net)
        self.out_channels = in_channels + num_convs * out_channels # 计算输出通道数

    def forward(self, X):
        for blk in self.net:
            Y = blk(X)
            X = torch.cat((X, Y), dim=1)  # 在通道维上将输入和输出连结
        return X

blk = DenseBlock(2, 3, 10)
X = torch.rand(4, 3, 8, 8)
Y = blk(X)
Y.shape  # torch.Size([4, 23, 8, 8])

过渡层

def transition_block(in_channels, out_channels):
    blk = nn.Sequential(
            nn.BatchNorm2d(in_channels), 
            nn.ReLU(),
            nn.Conv2d(in_channels, out_channels, kernel_size=1),
            nn.AvgPool2d(kernel_size=2, stride=2))
    return blk

blk = transition_block(23, 10)
blk(Y).shape  # torch.Size([4, 10, 4, 4])

DenseNet模型

net = 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))

num_channels, growth_rate = 64, 32  # num_channels为当前的通道数
num_convs_in_dense_blocks = [4, 4, 4, 4]

for i, num_convs in enumerate(num_convs_in_dense_blocks):
    DB = DenseBlock(num_convs, num_channels, growth_rate)
    net.add_module("DenseBlosk_%d" % i, DB)
    # 上一个稠密块的输出通道数
    num_channels = DB.out_channels
    # 在稠密块之间加入通道数减半的过渡层
    if i != len(num_convs_in_dense_blocks) - 1:
        net.add_module("transition_block_%d" % i, transition_block(num_channels, num_channels // 2))
        num_channels = num_channels // 2

net.add_module("BN", nn.BatchNorm2d(num_channels))
net.add_module("relu", nn.ReLU())
net.add_module("global_avg_pool", d2l.GlobalAvgPool2d()) # GlobalAvgPool2d的输出: (Batch, num_channels, 1, 1)
net.add_module("fc", nn.Sequential(d2l.FlattenLayer(), nn.Linear(num_channels, 10)))

X = torch.rand((1, 1, 96, 96))
for name, layer in net.named_children():
    X = layer(X)
    print(name, ' output shape:\t', X.shape)

0  output shape:	 torch.Size([1, 64, 48, 48])
1  output shape:	 torch.Size([1, 64, 48, 48])
2  output shape:	 torch.Size([1, 64, 48, 48])
3  output shape:	 torch.Size([1, 64, 24, 24])
DenseBlosk_0  output shape:	 torch.Size([1, 192, 24, 24])
transition_block_0  output shape:	 torch.Size([1, 96, 12, 12])
DenseBlosk_1  output shape:	 torch.Size([1, 224, 12, 12])
transition_block_1  output shape:	 torch.Size([1, 112, 6, 6])
DenseBlosk_2  output shape:	 torch.Size([1, 240, 6, 6])
transition_block_2  output shape:	 torch.Size([1, 120, 3, 3])
DenseBlosk_3  output shape:	 torch.Size([1, 248, 3, 3])
BN  output shape:	 torch.Size([1, 248, 3, 3])
relu  output shape:	 torch.Size([1, 248, 3, 3])
global_avg_pool  output shape:	 torch.Size([1, 248, 1, 1])
fc  output shape:	 torch.Size([1, 10])

训练和评测的实现

来自 d2l 代码

# 本函数已保存在d2lzh_pytorch包中方便以后使用。该函数将被逐步改进：它的完整实现将在“图像增广”一节中描述

def evaluate_accuracy(data_iter, net, device=None):
    
    if device is None and isinstance(net, torch.nn.Module):
        # 如果没指定device就使用net的device
        device = list(net.parameters())[0].device
        
    acc_sum, n = 0.0, 0
    
    with torch.no_grad():
        
        for X, y in data_iter:
            
            if isinstance(net, torch.nn.Module):
                
                net.eval() # 评估模式, 这会关闭dropout
                acc_sum += (net(X.to(device)).argmax(dim=1) == y.to(device)).float().sum().cpu().item()
                net.train() # 改回训练模式
                
            else: # 自定义的模型, 3.13节之后不会用到, 不考虑GPU
                
                if('is_training' in net.__code__.co_varnames): # 如果有is_training这个参数
                    # 将is_training设置成False
                    acc_sum += (net(X, is_training=False).argmax(dim=1) == y).float().sum().item() 
                else:
                    acc_sum += (net(X).argmax(dim=1) == y).float().sum().item() 
                    
            n += y.shape[0]
            
    return acc_sum / n

# 本函数已保存在d2lzh_pytorch包中方便以后使用

def train_ch5(net, train_iter, test_iter, batch_size, optimizer, device, num_epochs):
    
    net = net.to(device)
    print("training on ", device)
    loss = torch.nn.CrossEntropyLoss()
    batch_count = 0

    for epoch in range(num_epochs):
        train_l_sum, train_acc_sum, n, start = 0.0, 0.0, 0, time.time()
    
        for X, y in train_iter:
        
            X = X.to(device)
            y = y.to(device)
            
            y_hat = net(X)
            
            l = loss(y_hat, y)
            
            optimizer.zero_grad()
            l.backward()
            
            optimizer.step()
            
            train_l_sum += l.cpu().item()
            train_acc_sum += (y_hat.argmax(dim=1) == y).sum().cpu().item()
            n += y.shape[0]
            batch_count += 1
            
        test_acc = evaluate_accuracy(test_iter, net)
        print('epoch %d, loss %.4f, train acc %.3f, test acc %.3f, time %.1f sec'
              % (epoch + 1, train_l_sum / batch_count, train_acc_sum / n, test_acc, time.time() - start))

batch_size = 256

train_iter, test_iter = d2l.load_data_fashion_mnist(batch_size=batch_size)

lr, num_epochs = 0.001, 5

optimizer = torch.optim.Adam(net.parameters(), lr=lr)

train_ch5(net, train_iter, test_iter, batch_size, optimizer, device, num_epochs)