简单的CNN网络模型搭建（以Lenet-5网络和VGG网络测试MNIST和cifar10数据集）

 LeNet-5

LeNet-5是一个较简单的卷积神经网络。输入的二维图像，先经过两次卷积层到池化层，再经过全连接层，最后使用softmax分类作为输出层。

1、INPUT层-输入层

输入图像的尺寸统一归一化为32*32。

2、C1层-卷积层

3、S2层-池化层（下采样层）

采样方式：4个输入相加，乘以一个可训练参数，再加上一个可训练偏置。结果通过sigmoid

4、C3层-卷积层

第二次卷积的输出是C3，16个10x10的特征图，卷积核大小是 5*5.

C3与S2中前3个图相连的卷积结构如图所示：

采用上述这样的组合了？论文中说有两个原因：1）减少参数，2）这种不对称的组合连接的方式有利于提取多种组合特征。

5、S4层-池化层（下采样层）

S4是pooling层，窗口大小仍然是2*2，C3层的16个10x10的图分别进行以2x2为单位的池化得到16个5x5的特征图。有5x5x5x16=2000个连接。连接的方式与S2层类似。

6、C5层-卷积层

7、F6层-全连接层

8、Output层-全连接层

采用的是径向基函数（RBF）的网络连接方式

环境配置

anaconda配置python3.6环境

数据准备

import torchvision.datasets as datasets

root='data', # 表示数据的加载的根目录

train=True,  # True表示加载数据库的训练集，false加载测试集

transform=torchvision.transform.Totensor()) # 表示对数据进行预处理转换为Totensor类型，none为不进行预处理

download=True, # True表示自动下载数据集

完整代码 如下（Lenet-5网络测试MNIST数据集）

import torch
import torch.nn as nn
import torchvision
from setuptools import dist
from torch.nn import Sequential, Conv2d, BatchNorm2d, ReLU, MaxPool2d
from torch.utils.data import DataLoader
from torch.utils.tensorboard import SummaryWriter
import time

#定义训练的设备
device = torch.device("cuda:0")

#准备数据集
train_data = torchvision.datasets.MNIST(root='Mnist',train=True,transform=torchvision.transforms.ToTensor(),
                                        download=True) #训练数据集

test_data = torchvision.datasets.MNIST(root='./mnist/', train=False, transform=torchvision.transforms.ToTensor(),
                                        download=True) #测试数据集

# length 长度
train_data_size = len(train_data)
test_data_size = len(test_data)
print("训练数据集的长度为：{}".format(train_data_size))
print("测试数据集的长度为：{}".format(test_data_size))

# 利用 DataLoader 来加载数据集
train_dataloader = DataLoader(train_data, batch_size=64)
test_dataloader = DataLoader(test_data, batch_size=64)

# 搭建神经网络
class LetNet5(nn.Module):
    def __init__(self):
        super(LetNet5, self).__init__()
        self.model1 = Sequential(
            Conv2d(1, 6, 5, 1, padding=2),
            BatchNorm2d(6),
            ReLU(),
            MaxPool2d(kernel_size=2, stride=2),
            Conv2d(6, 16, 5),
            BatchNorm2d(16),
            ReLU(),
            MaxPool2d(kernel_size=2, stride=2),
            Conv2d(16, 120, 5),
            BatchNorm2d(120),
            ReLU()
        )

        self.model2 = Sequential(
            nn.Linear(120, 84),
            nn.ReLU(),
            nn.Linear(84, 10),
            nn.LogSoftmax()
        )

    def forward(self, x):
        x = self.model1(x)
        x = x.reshape(x.size(0), -1)  # 降维
        x = self.model2(x)
        return x

# 搭建网络模型
model = LetNet5()
model = model.to(device)

# 损失函数
criterion = nn.CrossEntropyLoss()
criterion = criterion.to(device)

# 优化器
learning_rate = 1e-3
optimizer = torch.optim.Adam(model.parameters(), lr=learning_rate)

# 添加tensorboard
writer = SummaryWriter('logs_LeNet')

# 测试的总次数
total_step = len(train_dataloader)

# 记录训练的次数
total_train_step = 0
# 记录测试的次数
total_test_step = 0
# 训练的轮数
epoch = 50

start_time = time.time()  #记录开始时间
for i in range(epoch):
    print("-----第 {} 次训练开始-----".format(i+1))

    # 训练步骤开始
    model.train()
    for data in train_dataloader:
        imgs, targets = data
        imgs = imgs.to(device)
        targets = targets.to(device)
        outputs = model(imgs)
        loss = criterion(outputs, targets)

        # 优化器优化模型
        optimizer.zero_grad()
        loss.backward()
        optimizer.step()

        total_train_step = total_train_step + 1
        if total_train_step % 100 == 0:
            end_time = time.time() #记录结束时间
            print('训练的时间： ' + str(end_time - start_time))
            print("训练次数： {}， loss： {}".format(total_train_step, loss.item()))
            writer.add_scalar("train_loss", loss.item(), total_train_step)

    # 测试步骤开始
    model.eval()
    total_test_loss = 0
    total_accuracy = 0
    with torch.no_grad():
        for data in test_dataloader:
            imgs, targets = data
            imgs = imgs.to(device)
            targets = targets.to(device)
            outputs = model(imgs)
            loss = criterion(outputs, targets)
            total_test_loss = total_test_loss + loss.item()
            accuracy = (outputs.argmax(1) == targets).sum()
            total_accuracy = total_accuracy + accuracy

    print("整体测试集上的loss： {}".format(total_test_loss))
    print("整体测试集上的正确率： {}".format(total_accuracy/test_data_size))
    writer.add_scalar("test_loss", total_test_loss, total_test_step)
    writer.add_scalar("test_accuracy", total_accuracy/test_data_size, total_test_step)
    total_test_step = total_test_step + 1

    torch.save(model, "img_{}.pth".format(i))
    print("模型已保存")

writer.close()

VGG网络训练cifar10数据集

import torchvision
import torch
from torch.nn import Sequential, Conv2d, MaxPool2d, Flatten, Linear
from torch.utils.data import DataLoader
import torch.nn.functional as F
from torch import nn
from torch.utils.tensorboard import SummaryWriter

#定义训练的设备
device = torch.device("cuda:0")

#准备数据集
train_data = torchvision.datasets.CIFAR10(root= "data", train=True, transform=torchvision.transforms.ToTensor(),
                                           download=True)   #训练数据集

test_data = torchvision.datasets.CIFAR10(root="data", train=False, transform=torchvision.transforms.ToTensor(),
                                         download=True)     #测试数据集

# length 长度
train_data_size = len(train_data)
test_data_size = len(test_data)
print("训练数据集的长度为：{}".format(train_data_size))
print("测试数据集的长度为：{}".format(test_data_size))

# 利用 DataLoader 来加载数据集
train_dataloader = DataLoader(train_data, batch_size=64)
test_dataloader = DataLoader(test_data, batch_size=64)

# 搭建神经网络
class Img_clarrification(nn.Module):

    def __init__(self):
        super(Img_clarrification,self).__init__()
        self.conv1 = nn.Conv2d(3,64,3,padding=1)
        self.conv2 = nn.Conv2d(64,64,3,padding=1)
        self.pool1 = nn.MaxPool2d(2, 2)
        self.bn1 = nn.BatchNorm2d(64)
        self.relu1 = nn.ReLU()

        self.conv3 = nn.Conv2d(64,128,3,padding=1)
        self.conv4 = nn.Conv2d(128, 128, 3,padding=1)
        self.pool2 = nn.MaxPool2d(2, 2, padding=1)
        self.bn2 = nn.BatchNorm2d(128)
        self.relu2 = nn.ReLU()

        self.conv5 = nn.Conv2d(128,128, 3,padding=1)
        self.conv6 = nn.Conv2d(128, 128, 3,padding=1)
        self.conv7 = nn.Conv2d(128, 128, 1,padding=1)
        self.pool3 = nn.MaxPool2d(2, 2, padding=1)
        self.bn3 = nn.BatchNorm2d(128)
        self.relu3 = nn.ReLU()

        self.conv8 = nn.Conv2d(128, 256, 3,padding=1)
        self.conv9 = nn.Conv2d(256, 256, 3, padding=1)
        self.conv10 = nn.Conv2d(256, 256, 1, padding=1)
        self.pool4 = nn.MaxPool2d(2, 2, padding=1)
        self.bn4 = nn.BatchNorm2d(256)
        self.relu4 = nn.ReLU()

        self.conv11 = nn.Conv2d(256, 512, 3, padding=1)
        self.conv12 = nn.Conv2d(512, 512, 3, padding=1)
        self.conv13 = nn.Conv2d(512, 512, 1, padding=1)
        self.pool5 = nn.MaxPool2d(2, 2, padding=1)
        self.bn5 = nn.BatchNorm2d(512)
        self.relu5 = nn.ReLU()

        self.fc14 = nn.Linear(512*4*4,1024)
        self.drop1 = nn.Dropout2d()
        self.fc15 = nn.Linear(1024,1024)
        self.drop2 = nn.Dropout2d()
        self.fc16 = nn.Linear(1024,10)


    def forward(self,x):
        x = self.conv1(x)
        x = self.conv2(x)
        x = self.pool1(x)
        x = self.bn1(x)
        x = self.relu1(x)


        x = self.conv3(x)
        x = self.conv4(x)
        x = self.pool2(x)
        x = self.bn2(x)
        x = self.relu2(x)

        x = self.conv5(x)
        x = self.conv6(x)
        x = self.conv7(x)
        x = self.pool3(x)
        x = self.bn3(x)
        x = self.relu3(x)

        x = self.conv8(x)
        x = self.conv9(x)
        x = self.conv10(x)
        x = self.pool4(x)
        x = self.bn4(x)
        x = self.relu4(x)

        x = self.conv11(x)
        x = self.conv12(x)
        x = self.conv13(x)
        x = self.pool5(x)
        x = self.bn5(x)
        x = self.relu5(x)

        x = x.view(-1,512*4*4)
        x = F.relu(self.fc14(x))
        x = self.drop1(x)
        x = F.relu(self.fc15(x))
        x = self.drop2(x)
        x = self.fc16(x)

        return x

# 搭建网络模型
img_clarrification = Img_clarrification()
img_clarrification = img_clarrification.to(device)

# 损失函数
loss_fn = nn.CrossEntropyLoss()
loss_fn = loss_fn.to(device)

# 优化器
learning_rate = 1e-2
optimizer = torch.optim.SGD(img_clarrification.parameters(), lr=learning_rate)

# 设置训练网络的一些参数
# 记录训练的次数
total_train_step = 0
# 记录测试的次数
total_test_step = 0
# 训练的轮数
epoch = 25

# 添加tensorboard
writer = SummaryWriter("logs_vggnet")

for i in range(epoch):
    print("-----第 {} 次训练开始-----".format(i+1))

    # 训练步骤开始
    img_clarrification.train()
    for data in train_dataloader:
        imgs, targets = data
        imgs = imgs.to(device)
        targets = targets.to(device)
        outputs = img_clarrification(imgs)
        loss = loss_fn(outputs, targets)

        # 优化器优化模型
        optimizer.zero_grad()
        loss.backward()
        optimizer.step()

        total_train_step = total_train_step + 1
        if total_train_step % 100 == 0:
            print("训练次数： {}， loss： {}".format(total_train_step, loss.item()))
            writer.add_scalar("train_loss", loss.item(), total_train_step)

    # 测试步骤开始
    img_clarrification.eval()
    total_test_loss = 0
    total_accuracy = 0
    with torch.no_grad():
        for data in test_dataloader:
            imgs, targets = data
            imgs = imgs.to(device)
            targets = targets.to(device)
            outputs = img_clarrification(imgs)
            loss = loss_fn(outputs, targets)
            total_test_loss = total_test_loss + loss.item()
            accuracy = (outputs.argmax(1) == targets).sum()
            total_accuracy = total_accuracy + accuracy

    print("整体测试集上的loss： {}".format(total_test_loss))
    print("整体测试集上的正确率： {}".format(total_accuracy/test_data_size))
    writer.add_scalar("test_loss", total_test_loss, total_test_step)
    writer.add_scalar("test_accuracy", total_accuracy/test_data_size, total_test_step)
    total_test_step = total_test_step + 1

    torch.save(img_clarrification, "img_{}.pth".format(i))
    print("模型已保存")

writer.close()