PyTorch学习笔记（十六）——利用GPU训练

小乖路路通2023-08-20 16:49

一、方式一

网络模型、损失函数、数据（包括输入、标注）

找到以上三种变量，调用它们的.cuda()，再返回即可

python 复制代码 
if torch.cuda.is_available():
    mynn = mynn.cuda()
python 复制代码 
if torch.cuda.is_available():
    loss_function = loss_function.cuda()
python 复制代码 
for data in train_dataloader:
    imgs,targets = data
    if torch.cuda.is_available():
        imgs = imgs.cuda()
        targets = targets.cuda()
python 复制代码 
for data in test_dataloader:
    imgs,targets = data
    if torch.cuda.is_available():
        imgs = imgs.cuda()
        targets = targets.cuda()

完整代码：

python 复制代码 
import torch
import torchvision
from torch import nn
from torch.utils.data import DataLoader
from torch.utils.tensorboard import SummaryWriter
import time
# from model import *

# 准备数据集
train_data = torchvision.datasets.CIFAR10(root="../datasets",train=True,transform=torchvision.transforms.ToTensor(),download=False)
test_data = torchvision.datasets.CIFAR10(root="../datasets",train=False,transform=torchvision.transforms.ToTensor(),download=False)

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)

# 创建网络模型
class MyNN(nn.Module):
    def __init__(self):
        super(MyNN, 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
mynn = MyNN()
if torch.cuda.is_available():
    mynn = mynn.cuda()

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

# 设置训练网络的一些参数
total_train_step = 0 # 记录训练次数
total_test_step = 0 # 记录测试次数
epoch = 10 # 训练的轮数

# 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
    mynn.train()
    for data in train_dataloader:
        imgs,targets = data
        if torch.cuda.is_available():
            imgs = imgs.cuda()
            targets = targets.cuda()
        outputs = mynn(imgs)
        loss = loss_function(outputs, targets)
        # 优化器优化模型
        optimizer.zero_grad()
        loss.backward()
        optimizer.step()

        total_train_step += 1
        if total_train_step % 100 == 0:
            end_time = time.time()
            print("所用时间：{}".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)

    # 测试步骤开始
    mynn.eval()
    total_test_loss = 0
    total_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 = mynn(imgs)
            loss = loss_function(outputs, targets)
            total_test_loss += loss
            accuracy = (outputs.argmax(1) == targets).sum()
            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 += 1

    torch.save(mynn,"mynn_{}.pth".format(i))
    # torch.save(mynn.state_dict(),"mynn_{}.pth".format(i))
    print("模型已保存")

writer.close()

比较CPU和GPU的训练时间：

查看GPU信息：

在 终端里输入nvidia-smi
使用Google Colab：Google 为我们提供了一个免费的GPU

修改 ------> 笔记本设置 ------> 硬件加速器选择GPU（每周免费使用30h）

二、方式二（更常用）

定义训练设备

python 复制代码 
device = torch.device("cpu")
python 复制代码 
# 对于单显卡来说，以下两种方式没有区别
device = torch.device("cuda")
device = torch.device("cuda:0")
python 复制代码 
# 一种语法的简写，程序在 CPU 或 GPU/cuda 环境下都能运行
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")

网络模型、损失函数、数据（包括输入、标注）

找到以上三种变量，.to(device)，再返回即可

python 复制代码 
mynn = MyNN()
mynn = mynn.to(device)
# 这里可以不用再赋值给mynn，直接mynn.to(device) 也可以
python 复制代码 
loss_function = nn.CrossEntropyLoss()
loss_function = loss_function.to(device)
# 这里可以不用再赋值给loss_function ，直接loss_function .to(device) 也可以
python 复制代码 
for data in train_dataloader:
    imgs,targets = data
    imgs = imgs.to(device)
    targets = targets.to(device)
    # 这里必须赋值
python 复制代码 
for data in test_dataloader:
    imgs,targets = data
    imgs = imgs.to(device)
    targets = imgs.to(device)
    # 这里必须赋值

完整代码：

python 复制代码 
import torch
import torchvision
from torch import nn
from torch.utils.data import DataLoader
from torch.utils.tensorboard import SummaryWriter
import time
# from model import *

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

# 准备数据集
train_data = torchvision.datasets.CIFAR10(root="../datasets",train=True,transform=torchvision.transforms.ToTensor(),download=False)
test_data = torchvision.datasets.CIFAR10(root="../datasets",train=False,transform=torchvision.transforms.ToTensor(),download=False)

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)

# 创建网络模型
class MyNN(nn.Module):
    def __init__(self):
        super(MyNN, 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
mynn = MyNN()
mynn.to(device)

# 损失函数
loss_function = nn.CrossEntropyLoss()
loss_function.to(device)
# 优化器
learning_rate = 1e-2
optimizer = torch.optim.SGD(mynn.parameters(), lr=learning_rate)

# 设置训练网络的一些参数
total_train_step = 0 # 记录训练次数
total_test_step = 0 # 记录测试次数
epoch = 10 # 训练的轮数

# 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
    mynn.train()
    for data in train_dataloader:
        imgs,targets = data
        imgs = imgs.to(device)
        targets = targets.to(device)
        outputs = mynn(imgs)
        loss = loss_function(outputs, targets)
        # 优化器优化模型
        optimizer.zero_grad()
        loss.backward()
        optimizer.step()

        total_train_step += 1
        if total_train_step % 100 == 0:
            end_time = time.time()
            print("所用时间：{}".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)

    # 测试步骤开始
    mynn.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 = mynn(imgs)
            loss = loss_function(outputs, targets)
            total_test_loss += loss
            accuracy = (outputs.argmax(1) == targets).sum()
            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 += 1

    torch.save(mynn,"mynn_{}.pth".format(i))
    # torch.save(mynn.state_dict(),"mynn_{}.pth".format(i))
    print("模型已保存")

writer.close()
相关推荐
布吉岛呀~1 小时前
Pytorch复习笔记--pytorch常见交叉熵函数的实现
pytorch
fydw_7152 小时前
PyTorch 激活函数及非线性变换详解
人工智能·pytorch·python
天蓝蓝235282 小时前
PyTorch开源的深度学习框架
pytorch·深度学习·开源
#include<菜鸡>5 小时前
动手学深度学习（pytorch土堆）-06损失函数与反向传播、模型训练、GPU训练
人工智能·pytorch·深度学习
5pace16 小时前
PyTorch深度学习快速入门教程【土堆】基础知识篇
人工智能·pytorch·深度学习
AI完全体18 小时前
AI小项目4-用Pytorch从头实现Transformer（详细注解）
人工智能·pytorch·深度学习·机器学习·语言模型·transformer·注意力机制
吉小雨21 小时前
PyTorch经典模型
人工智能·pytorch·python
crownyouyou1 天前
最简单的一文安装Pytorch+CUDA
人工智能·pytorch·python
artificiali2 天前
Anaconda配置pytorch的基本操作
人工智能·pytorch·python
#include<菜鸡>2 天前
动手学深度学习（pytorch土堆）-04torchvision中数据集的使用
人工智能·pytorch·深度学习
热门推荐
01RAG 实践- Ollama+RagFlow 部署本地知识库02组基轨迹建模 GBTM的介绍与实现（Stata 或 R）032024年高教社杯数学建模国赛C题超详细解题思路分析04软件工程从理论到实践客观题汇总(头歌第一章至第八章)05苍穹外卖面试总结06【2024数模国赛赛题思路公开】国赛B题思路丨附可运行代码丨无偿自提07【halcon】halcon字符识别——OCR08yolov8实战第五天——yolov8+ffmpg实时视频流检测并进行实时推流——（推流，保姆教学）09【2024高教社杯全国大学生数学建模竞赛】B题 生产过程中的决策问题——解题思路 代码 论文10比特币钱包库 bitcoinjs-lib 讲解：生成钱包、导入钱包、签名验证、转账交易