【Pytorch】完整的模型训练套路与利用GPU训练 - 学习笔记

Lucy@IshtarXu

已于 2022-02-18 14:18:26 修改

阅读量1.6k

点赞数

分类专栏：深度学习 Pytorch Python 文章标签： pytorch 深度学习

于 2022-02-18 13:49:55 首次发布

本文链接：https://blog.csdn.net/FelicityXu/article/details/123001109

版权

深度学习同时被 3 个专栏收录

43 篇文章 17 订阅

订阅专栏

Python

32 篇文章 1 订阅

订阅专栏

Pytorch

21 篇文章 2 订阅

订阅专栏

本文介绍了如何在PyTorch中进行模型训练与测试，包括数据预处理、网络模型构建、损失函数选择、优化器设置以及训练过程的实现。同时，展示了两种GPU训练的方法，通过`.cuda()`和`.to(device)`将模型与数据迁移到GPU上，以提升计算速度。此外，还利用TensorBoard进行训练和测试过程的可视化，便于监控模型性能。

摘要由CSDN通过智能技术生成

视频地址

1 - 模型训练与测试代码

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

from model import *  # 引入model文件

# 准备数据集
from torch import nn
from torch.utils.data import DataLoader

train_data = torchvision.datasets.CIFAR10('./dataset', train=True, transform=torchvision.transforms.ToTensor())
test_data = torchvision.datasets.CIFAR10('./dataset', train=False, transform=torchvision.transforms.ToTensor())

# length求长度
train_data_size = len(train_data)
test_data_size = len(test_data)
# 如果train_data_size=10, 训练数据集的长度为：10
print("训练数据集的长度为:{}".format(train_data_size))
# 光标移到代码行上按ctrl+d，可以复制这一行代码
print("测试数据集的长度为:{}".format(test_data_size))
# 输出 训练数据集长度为：50000   测试数据集长度为：10000


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

# 创建网络模型
model = Model()

# 损失函数 使用交叉熵
loss_fn = nn.CrossEntropyLoss()

# 优化器
learning_rate = 1e-2
# learning_rate = 0.01
optimizer = torch.optim.SGD(model.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 = 10

# 添加tensorboard
writer = SummaryWriter("./logs_train")

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

    # 训练步骤开始
    for data in train_dataloader:
        imgs, targets = data
        outputs = model(imgs)
        loss = loss_fn(outputs, targets)

        # 梯度清零
        optimizer.zero_grad()
        # 反向传播
        loss.backward()
        # 使用优化器优化模型
        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()))  # loss.item 转化为数字
            writer.add_scalar("train_loss", loss.item(), total_train_step)

    # 测试步骤开始
    total_test_loss = 0
    total_accuracy = 0  # 整体正确个数
    with torch.no_grad():  # 没有梯度，因为是测试，不要再调整梯度参数了
        for data in test_dataloader:
            imgs, targets = data
            outputs = model(imgs)
            loss = loss_fn(outputs, targets)
            total_test_loss = total_test_loss +loss.item()
            # 计算准确率
            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

        # 保存训练之后的结果
        torch.save(model, "model_{}.pth".format(i))
        print("模型已保存")

writer.close()

其中.model文件为

# 搭建神经网络 10分类网络 单独用python文件
import torch
from torch import nn


class Model(nn.Module):
    def __init__(self):
        super(Model, self).__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


# 测试网络的正确性
if __name__ == '__main__':
    model = Model()
    input = torch.ones((64, 3, 32, 32)) # 64张图片
    output = model(input)
    print(output.shape)

# 输出 torch.Size([64, 10])

输出结果为
在这里插入图片描述

2 - GPU训练

2.1 - GPU训练方式1

找到：

网络模型
（训练、测试步骤的）数据
损失函数

在后面加上.cuda()即可

import time

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

# from model import *  # 引入model文件

# 准备数据集
from torch import nn
from torch.utils.data import DataLoader


train_data = torchvision.datasets.CIFAR10('./dataset', train=True, transform=torchvision.transforms.ToTensor())
test_data = torchvision.datasets.CIFAR10('./dataset', train=False, transform=torchvision.transforms.ToTensor())

# length求长度
train_data_size = len(train_data)
test_data_size = len(test_data)
# 如果train_data_size=10, 训练数据集的长度为：10
print("训练数据集的长度为:{}".format(train_data_size))
# 光标移到代码行上按ctrl+d，可以复制这一行代码
print("测试数据集的长度为:{}".format(test_data_size))
# 输出 训练数据集长度为：50000   测试数据集长度为：10000


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


# 创建网络模型
class Model(nn.Module):
    def __init__(self):
        super(Model, self).__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


model = Model()
# 使用cuda
model = model.cuda()

# 损失函数 使用交叉熵
loss_fn = nn.CrossEntropyLoss()
# 使用cuda
loss_fn = loss_fn.cuda()


# 优化器
learning_rate = 1e-2
# learning_rate = 0.01
optimizer = torch.optim.SGD(model.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 = 10

# 添加tensorboard
writer = SummaryWriter("./logs_train")
start_time = time.time()

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

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

        # 使用cuda
        imgs = imgs.cuda()
        targets = targets.cuda()

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

        # 梯度清零
        optimizer.zero_grad()
        # 反向传播
        loss.backward()
        # 使用优化器优化模型
        optimizer.step()

        total_train_step = total_train_step + 1
        if total_train_step % 100 == 0:
            end_time = time.time()
            print(end_time - start_time)
            print("训练次数：{}, Loss：{}".format(total_train_step, loss.item()))  # loss.item 转化为数字
            writer.add_scalar("train_loss", loss.item(), total_train_step)

    # 测试步骤开始
    total_test_loss = 0
    total_accuracy = 0  # 整体正确个数
    with torch.no_grad():  # 没有梯度，因为是测试，不要再调整梯度参数了
        for data in test_dataloader:
            imgs, targets = data

            # 使用cuda
            imgs = imgs.cuda()
            targets = targets.cuda()

            outputs = model(imgs)
            loss = loss_fn(outputs, targets)
            total_test_loss = total_test_loss +loss.item()
            # 计算准确率
            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

        # 保存训练之后的结果
        torch.save(model, "model_{}.pth".format(i))
        print("模型已保存")

writer.close()

2.2 - 使用Google Colab

Google CoLab
在这里插入图片描述

2.3 - GPU训练方式2（常用）

.to(device)
device=torch.device(‘CPU’)
torch.device(‘cuda’)

import time

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

# from model import *  # 引入model文件

# 准备数据集
from torch import nn
from torch.utils.data import DataLoader

# 定义训练的设备
device = torch.device('cuda')

train_data = torchvision.datasets.CIFAR10('./dataset', train=True, transform=torchvision.transforms.ToTensor())
test_data = torchvision.datasets.CIFAR10('./dataset', train=False, transform=torchvision.transforms.ToTensor())

# length求长度
train_data_size = len(train_data)
test_data_size = len(test_data)
# 如果train_data_size=10, 训练数据集的长度为：10
print("训练数据集的长度为:{}".format(train_data_size))
# 光标移到代码行上按ctrl+d，可以复制这一行代码
print("测试数据集的长度为:{}".format(test_data_size))
# 输出 训练数据集长度为：50000   测试数据集长度为：10000


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


# 创建网络模型
class Model(nn.Module):
    def __init__(self):
        super(Model, self).__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


model = Model()
# 使用cuda
model = model.to(device)

# 损失函数 使用交叉熵
loss_fn = nn.CrossEntropyLoss()
# 使用cuda
loss_fn = loss_fn.to(device)


# 优化器
learning_rate = 1e-2
# learning_rate = 0.01
optimizer = torch.optim.SGD(model.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 = 10

# 添加tensorboard
writer = SummaryWriter("./logs_train")
start_time = time.time()

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

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

        # 使用cuda
        imgs = imgs.to(device)
        targets = targets.to(device)

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

        # 梯度清零
        optimizer.zero_grad()
        # 反向传播
        loss.backward()
        # 使用优化器优化模型
        optimizer.step()

        total_train_step = total_train_step + 1
        if total_train_step % 100 == 0:
            end_time = time.time()
            print(end_time - start_time)
            print("训练次数：{}, Loss：{}".format(total_train_step, loss.item()))  # loss.item 转化为数字
            writer.add_scalar("train_loss", loss.item(), total_train_step)

    # 测试步骤开始
    total_test_loss = 0
    total_accuracy = 0  # 整体正确个数
    with torch.no_grad():  # 没有梯度，因为是测试，不要再调整梯度参数了
        for data in test_dataloader:
            imgs, targets = data

            # 使用cuda
            imgs = imgs.to(device)
            targets = targets.to(device)

            outputs = model(imgs)
            loss = loss_fn(outputs, targets)
            total_test_loss = total_test_loss +loss.item()
            # 计算准确率
            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

        # 保存训练之后的结果
        torch.save(model, "model_{}.pth".format(i))
        print("模型已保存")

writer.close()