CIFAR10——小土堆学习笔记

10分类问题

采用的网络模型

构建该模型：

函数Conv2d()中的padding需自己计算，根据输入图片和输出图片以及kernel的尺寸计算。
nn.Flatten()是将6444的数据展成一维。

# 构建模型
class CFIAR10_Model(nn.Module):
    def __init__(self):
        super().__init__()
        self.model = nn.Sequential(
            nn.Conv2d(3, 32, 5, 1, 2),
            nn.MaxPool2d(2),
            nn.Conv2d(32, 32, 5, 1, 2),
            nn.MaxPool2d(2),
            nn.Conv2d(32, 64, 5, 1, 2),
            nn.MaxPool2d(2),
            nn.Flatten(),
            nn.Linear(64 * 4 * 4, 64),
            nn.Linear(64, 10)
        )

    def forward(self, x):
        x = self.model(x)
        return x

整体代码（cpu）：

import torchvision
from torch.utils.tensorboard import SummaryWriter
from model import *
from torch.utils.data import DataLoader

# 准备数据集
compose = torchvision.transforms.Compose([
    torchvision.transforms.ToTensor()
])
train_dataset = torchvision.datasets.CIFAR10(root="../datasetCIFAR10", train=True, transform=compose, download=True)
test_dataset = torchvision.datasets.CIFAR10(root="../datasetCIFAR10", train=True, transform=compose, download=True)

# 数据集大小
train_size = len(train_dataset)
test_size = len(test_dataset)
print("训练数据集的大小为{}".format(train_size))
print("测试数据集的大小为{}".format(test_size))

# 用dataloader来加载数据
train_dataloader = DataLoader(train_dataset, batch_size=64)
test_dataloader = DataLoader(test_dataset, batch_size=64)

# 创建网络模型
model = CFIAR10_Model()

# 损失函数
loss_fn = nn.CrossEntropyLoss()

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

# 设置网络的一些参数
# 记录训练的次数
train_step = 0
# 记录测试的次数
test_step = 0
# 训练的轮数
epoch = 10

# 使用tensorboard
writer = SummaryWriter("logs_train")

# 训练步骤
for i in range(epoch):
    print("------第{}轮训练------".format(i+1))
    # 训练步骤开始
    for data in train_dataloader:
        imgs, targets = data
        output = model(imgs)

        loss = loss_fn(output, targets)
        # 优化器优化参数
        optim.zero_grad()
        loss.backward()
        optim.step()

        train_step = train_step + 1
        if train_step % 100 == 0:
            print("训练次数{}，损失为{}".format(train_step, loss.item()))
            writer.add_scalar("train_loss", loss.item(), train_step)
            # eg：直接输出loss，会显示tensor(5), loss.item()会显示5

    # 测试步骤开始
    total_loss_test = 0
    total_test_accuracy = 0
    with torch.no_grad():
        for data in test_dataloader:
            imgs, targets = data
            outputs = model(imgs)
            loss = loss_fn(outputs, targets)
            total_loss_test = total_loss_test + loss.item()
            accuracy = (outputs.argmax(1) == targets).sum()
            total_test_accuracy = total_test_accuracy + accuracy
    print("整体测试集上的loss:{}".format(total_loss_test))
    print("整体测试集上的accuracy:{}".format(total_test_accuracy/test_size))
    writer.add_scalar("test_loss", total_loss_test, test_step)
    writer.add_scalar("test_accuracy", total_test_accuracy, test_step)
    test_step = test_step + 1

    # 模型保存
    torch.save(model, "model{}.pth".format(i))
    # torch.save(model.state_dict(), "model_{}.pth".format(i))
    print("模型已保存")

writer.close()

GPU版：

方式一：

采用GPU的地方：
·网络模型

if torch.cuda.is_available():
    model = model.cuda()

·数据（输入，标注）

   if torch.cuda.is_available():
            imgs = imgs.cuda()
            targets = targets.cuda()

·损失函数

if torch.cuda.is_available():
    loss_fn = loss_fn.cuda()

整体代码

import torchvision
from torch.utils.tensorboard import SummaryWriter
from model import *
from torch.utils.data import DataLoader

# 准备数据集
compose = torchvision.transforms.Compose([
    torchvision.transforms.ToTensor()
])
train_dataset = torchvision.datasets.CIFAR10(root="../datasetCIFAR10", train=True, transform=compose, download=True)
test_dataset = torchvision.datasets.CIFAR10(root="../datasetCIFAR10", train=True, transform=compose, download=True)

# 数据集大小
train_size = len(train_dataset)
test_size = len(test_dataset)
print("训练数据集的大小为{}".format(train_size))
print("测试数据集的大小为{}".format(test_size))

# 用dataloader来加载数据
train_dataloader = DataLoader(train_dataset, batch_size=64)
test_dataloader = DataLoader(test_dataset, batch_size=64)

# 创建网络模型
model = CFIAR10_Model()
if torch.cuda.is_available():
    model = model.cuda()

# 损失函数
loss_fn = nn.CrossEntropyLoss()
if torch.cuda.is_available():
    loss_fn = loss_fn.cuda()
# 优化器
learning_rate = 1e-2
optim = torch.optim.SGD(model.parameters(), lr=learning_rate)

# 设置网络的一些参数
# 记录训练的次数
train_step = 0
# 记录测试的次数
test_step = 0
# 训练的轮数
epoch = 10

# 使用tensorboard
writer = SummaryWriter("logs_train")

# 训练步骤
for i in range(epoch):
    print("------第{}轮训练------".format(i+1))

    # 训练步骤开始
    for data in train_dataloader:
        imgs, targets = data
        
        if torch.cuda.is_available():
            imgs = imgs.cuda()
            targets = targets.cuda()
            
        output = model(imgs)

        loss = loss_fn(output, targets)
        # 优化器优化参数
        optim.zero_grad()
        loss.backward()
        optim.step()

        train_step = train_step + 1
        if train_step % 100 == 0:
            print("训练次数{}，损失为{}".format(train_step, loss.item()))
            writer.add_scalar("train_loss", loss.item(), train_step)
            # eg：直接输出loss，会显示tensor(5), loss.item()会显示5

    # 测试步骤开始
    total_loss_test = 0
    total_test_accuracy = 0
    with torch.no_grad():
        for data in test_dataloader:
            imgs, targets = data

            if torch.cuda.is_available():
                imgs = imgs.cuda()
                targets = targets.cuda()
                
            outputs = model(imgs)
            loss = loss_fn(outputs, targets)
            total_loss_test = total_loss_test + loss.item()
            accuracy = (outputs.argmax(1) == targets).sum()
            total_test_accuracy = total_test_accuracy + accuracy
    print("整体测试集上的loss:{}".format(total_loss_test))
    print("整体测试集上的accuracy:{}".format(total_test_accuracy/test_size))
    writer.add_scalar("test_loss", total_loss_test, test_step)
    writer.add_scalar("test_accuracy", total_test_accuracy, test_step)
    test_step = test_step + 1

    # 模型保存
    torch.save(model, "model{}.pth".format(i))
    # torch.save(model.state_dict(), "model_{}.pth".format(i))
    print("模型已保存")

writer.close()

方式二

.to(device)
device = torch.device(“cpu”)
device = torch.device(“cuda”)
device = torch.device(“cuda:0”) # 有多个显卡的情况

整体代码：

import torchvision
from torch.utils.tensorboard import SummaryWriter
from model import *
from torch.utils.data import DataLoader

# 定义训练的设备
device = torch.device("gpu")

# 准备数据集
compose = torchvision.transforms.Compose([
    torchvision.transforms.ToTensor()
])
train_dataset = torchvision.datasets.CIFAR10(root="../datasetCIFAR10", train=True, transform=compose, download=True)
test_dataset = torchvision.datasets.CIFAR10(root="../datasetCIFAR10", train=True, transform=compose, download=True)

# 数据集大小
train_size = len(train_dataset)
test_size = len(test_dataset)
print("训练数据集的大小为{}".format(train_size))
print("测试数据集的大小为{}".format(test_size))

# 用dataloader来加载数据
train_dataloader = DataLoader(train_dataset, batch_size=64)
test_dataloader = DataLoader(test_dataset, batch_size=64)

# 创建网络模型
model = CFIAR10_Model()
model.to(device)

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

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

# 设置网络的一些参数
# 记录训练的次数
train_step = 0
# 记录测试的次数
test_step = 0
# 训练的轮数
epoch = 10

# 使用tensorboard
writer = SummaryWriter("logs_train")

# 训练步骤
for i in range(epoch):
    print("------第{}轮训练------".format(i+1))

    # 训练步骤开始
    for data in train_dataloader:
        imgs, targets = data

        imgs = imgs.to(device)
        targets = targets.to(device)

        output = model(imgs)
        loss = loss_fn(output, targets)

        # 优化器优化参数
        optim.zero_grad()
        loss.backward()
        optim.step()

        train_step = train_step + 1
        if train_step % 100 == 0:
            print("训练次数{}，损失为{}".format(train_step, loss.item()))
            writer.add_scalar("train_loss", loss.item(), train_step)
            # eg：直接输出loss，会显示tensor(5), loss.item()会显示5

    # 测试步骤开始
    total_loss_test = 0
    total_test_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 = loss_fn(outputs, targets)
            total_loss_test = total_loss_test + loss.item()
            accuracy = (outputs.argmax(1) == targets).sum()
            total_test_accuracy = total_test_accuracy + accuracy
    print("整体测试集上的loss:{}".format(total_loss_test))
    print("整体测试集上的accuracy:{}".format(total_test_accuracy/test_size))
    writer.add_scalar("test_loss", total_loss_test, test_step)
    writer.add_scalar("test_accuracy", total_test_accuracy, test_step)
    test_step = test_step + 1

    # 模型保存
    torch.save(model, "model{}.pth".format(i))
    # torch.save(model.state_dict(), "model_{}.pth".format(i))
    print("模型已保存")

writer.close()