PyTorch_21_利用GPU训练

• 利用GPU训练的三要素

• train_gpu_1.py

import torch
import torchvision
from torch.utils.tensorboard import SummaryWriter

# from model import TuDui

from torch import nn
from torch.utils.data import DataLoader

import time

# 下面两个数据是没有cuda方法的
train_data = torchvision.datasets.CIFAR10(root="./dataset", train=True, transform=torchvision.transforms.ToTensor(),
                                          download=True)
test_data = torchvision.datasets.CIFAR10(root="./dataset", train=False, transform=torchvision.transforms.ToTensor(),
                                         download=True)

# 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_data_loader = DataLoader(train_data, batch_size=64)
test_data_loader = DataLoader(test_data, batch_size=64)


# 创建网络模型
class TuDui(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
tudui = TuDui()

# cuda position1:
if torch.cuda.is_available():
    tudui = tudui.cuda()


# 创建损失函数
loss_fn = nn.CrossEntropyLoss()

# cuda position2:
if torch.cuda.is_available():
    loss_fn = loss_fn.cuda()

# 优化器,这是随机梯度下降
# learning_rate = 0.01
# 1e-2 = 1 * (10)^(-2) = 0.01
learning_rate = 1e-2
# 优化器里面也是没有cuda方法的
optimizer = torch.optim.SGD(tudui.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
# 训练轮数
epoch = 10

# 添加tensorboard
writer = SummaryWriter("logs")

start_time = time.time()
for i in range(epoch):
    print("--------第{}轮训练开始--------".format(i+1))
    # 训练开始
    # This has any effect only on certain modules. Dropout, BatchNorm
    tudui.train()
    for data in train_data_loader:
        imgs, targets = data

        # cuda position3:
        if torch.cuda.is_available():
            imgs = imgs.cuda()
            targets = targets.cuda()

        outputs = tudui(imgs)
        loss = loss_fn(outputs, targets)
        # 梯度清零
        optimizer.zero_grad()
        # 反向传播
        loss.backward()
        # 卷积核参数调优
        optimizer.step()

        total_train_step += 1
        # 为了保证输出的信息不是太多而设置的
        if total_train_step % 100 == 0:
            end_time = time.time()
            print("近100次训练总耗时：{}秒".format(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)

    # 测试开始
    # This has any effect only on certain modules.Dropout, BatchNorm
    tudui.eval()
    # 对卷积核调优完成，还是对调优后的卷积核进行测试
    # 测试步骤开始,没有梯度，能够保证不对卷积核进行调优
    total_test_loss = 0
    total_accuracy = 0
    with torch.no_grad():
        for data in test_data_loader:
            imgs, targets = data

            # cuda position4:
            if torch.cuda.is_available():
                imgs = imgs.cuda()
                targets = targets.cuda()

            outputs = tudui(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
    # 但分类问题一般使用正确率来衡量
    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
    # 这是方式1
    torch.save(tudui, "tudui_{}.pth".format(i))
    # 这是方式2
    # torch.save(tudui.state_dict(), "tudui_dict_{}.pth".format(i))
    print("模型{}已保存".format(i))

writer.close()

• train_gpu_2.py

import torch
import torchvision
from torch.utils.tensorboard import SummaryWriter

# from model import TuDui

from torch import nn
from torch.utils.data import DataLoader

import time

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

# 下面两个数据是没有cuda方法的
train_data = torchvision.datasets.CIFAR10(root="./dataset", train=True, transform=torchvision.transforms.ToTensor(),
                                          download=True)
test_data = torchvision.datasets.CIFAR10(root="./dataset", train=False, transform=torchvision.transforms.ToTensor(),
                                         download=True)

# 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_data_loader = DataLoader(train_data, batch_size=64)
test_data_loader = DataLoader(test_data, batch_size=64)


# 创建网络模型
class TuDui(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
tudui = TuDui()

# cuda position1:
# tudui = tudui.to(device)
tudui.to(device)


# 创建损失函数
loss_fn = nn.CrossEntropyLoss()

# cuda position2:
# loss_fn = loss_fn.to(device)
loss_fn.to(device)

# 优化器,这是随机梯度下降
# learning_rate = 0.01
# 1e-2 = 1 * (10)^(-2) = 0.01
learning_rate = 1e-2
# 优化器里面也是没有cuda方法的
optimizer = torch.optim.SGD(tudui.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
# 训练轮数
epoch = 10

# 添加tensorboard
writer = SummaryWriter("logs")

start_time = time.time()
for i in range(epoch):
    print("--------第{}轮训练开始--------".format(i+1))
    # 训练开始
    # This has any effect only on certain modules. Dropout, BatchNorm
    tudui.train()
    for data in train_data_loader:
        imgs, targets = data

        # cuda position3:
        imgs = imgs.to(device)
        targets = targets.to(device)

        outputs = tudui(imgs)
        loss = loss_fn(outputs, targets)
        # 梯度清零
        optimizer.zero_grad()
        # 反向传播
        loss.backward()
        # 卷积核参数调优
        optimizer.step()

        total_train_step += 1
        # 为了保证输出的信息不是太多而设置的
        if total_train_step % 100 == 0:
            end_time = time.time()
            print("近100次训练总耗时：{}秒".format(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)

    # 测试开始
    # This has any effect only on certain modules.Dropout, BatchNorm
    tudui.eval()
    # 对卷积核调优完成，还是对调优后的卷积核进行测试
    # 测试步骤开始,没有梯度，能够保证不对卷积核进行调优
    total_test_loss = 0
    total_accuracy = 0
    with torch.no_grad():
        for data in test_data_loader:
            imgs, targets = data

            # cuda position4:
            imgs = imgs.to(device)
            targets = targets.to(device)

            outputs = tudui(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
    # 但分类问题一般使用正确率来衡量
    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
    # 这是方式1
    torch.save(tudui, "tudui_{}.pth".format(i))
    # 这是方式2
    # torch.save(tudui.state_dict(), "tudui_dict_{}.pth".format(i))
    print("模型{}已保存".format(i))

writer.close()