Pytorch入门需要达到的效果

会搭建深度学习环境和依赖包安装

使用Anaconda创建环境、在pytorch官网安装pytorch、安装依赖包

会使用常见操作，例如`matmul`，`sigmoid`，`softmax`，`relu`，`linear`

matmul操作见文章torch.matmul()的用法
sigmoid，softmax，relu都是常用的激活函数，linear是线性层：

from torch import nn

class Network(nn.Module):
    def __init__(self):
        super().__init__()
        self.model = nn.Sequential(
            nn.Sigmoid(),
            nn.Softmax(),
            nn.ReLU(),
            nn.Linear(1024, 64)
        )

`dataset`，`dataloader`，损失函数，优化器的使用

`dataset`，`dataloader`

官方文档是这么写的：
在这里插入图片描述
当我们自定义一个dataset的时候，需要继承Dataset，重写__getitem__()方法，也可以重写__len__()方法，下面是一个例子，我们的数据集存放成这种形式，每一个image图片都对应一个相同名称的label文件，如0013035.jpg和0013035.txt就分别是一个图片和它的label：
在这里插入图片描述

import torchvision.transforms
from PIL import Image
from torch.utils.data import Dataset, DataLoader


class MyData(Dataset):
    def __init__(self, root_dir, image_dir, label_dir):
        self.root_dir = root_dir
        self.image_dir = image_dir
        self.label_dir = label_dir
        self.image_path = os.path.join(self.root_dir, self.image_dir)
        self.label_path = os.path.join(self.root_dir, self.label_dir)
        self.imgs = os.listdir(self.image_path)
        self.labels = os.listdir(self.label_path)

    def __getitem__(self, item):
        img_name = self.imgs[item]
        img_item_path = os.path.join(self.image_path, img_name)
        label_item_path = os.path.join(self.label_path, self.convert_to_txt_path(img_name))
        img = Image.open(img_item_path)
        with open(label_item_path, 'r') as f:
            label = f.read().strip()
        return img, label

    def convert_to_txt_path(self, image_path):
        # 使用正则表达式匹配文件名中的点和扩展名，并替换为'.txt'
        label_path = re.sub(r'\.[^.]+?$', '.txt', image_path)
        return label_path

    def __len__(self):
        return len(self.imgs)

root_dir = "dataset/train"
ants_image_dir = "ants_image"
bees_image_dir = "bees_image"
ants_label_dir = "ants_label"
bees_label_dir = "bees_label"
ants_dataset = MyData(root_dir, ants_image_dir, ants_label_dir)
bees_dataset = MyData(root_dir, bees_image_dir, bees_label_dir)

train_dataset = ants_dataset + bees_dataset

img, target = train_dataset[0]
transform = torchvision.transforms.ToTensor()
print(transform(img).shape)
print(target)

我们使用dataloader来读取这个数据集，我们需要对jpg格式的dataset进行处理，将其转换为相同大小的tensor，再读取：

import torchvision.transforms
from PIL import Image
from torch.utils.data import Dataset, DataLoader


class MyData(Dataset):
    def __init__(self, root_dir, image_dir, label_dir):
        self.root_dir = root_dir
        self.image_dir = image_dir
        self.label_dir = label_dir
        self.image_path = os.path.join(self.root_dir, self.image_dir)
        self.label_path = os.path.join(self.root_dir, self.label_dir)
        self.imgs = os.listdir(self.image_path)
        self.labels = os.listdir(self.label_path)

    def __getitem__(self, item):
        img_name = self.imgs[item]
        img_item_path = os.path.join(self.image_path, img_name)
        label_item_path = os.path.join(self.label_path, self.convert_to_txt_path(img_name))
        img = Image.open(img_item_path)
        transcompose = torchvision.transforms.Compose([torchvision.transforms.Resize((300, 300)), torchvision.transforms.ToTensor()])
        img = transcompose(img)
        with open(label_item_path, 'r') as f:
            label = f.read().strip()
        return img, label

    def convert_to_txt_path(self, image_path):
        # 使用正则表达式匹配文件名中的点和扩展名，并替换为'.txt'
        label_path = re.sub(r'\.[^.]+?$', '.txt', image_path)
        return label_path

    def __len__(self):
        return len(self.imgs)

root_dir = "dataset/train"
ants_image_dir = "ants_image"
bees_image_dir = "bees_image"
ants_label_dir = "ants_label"
bees_label_dir = "bees_label"
ants_dataset = MyData(root_dir, ants_image_dir, ants_label_dir)
bees_dataset = MyData(root_dir, bees_image_dir, bees_label_dir)

train_dataset = ants_dataset + bees_dataset

img, target = train_dataset[0]
print(img.shape)
print(target)

train_dataloader = DataLoader(train_dataset, batch_size=64, drop_last=True)
for data in train_dataloader:
    try:
        imgs, target = data
    except Exception as e:
        print(f"跳过异常文件:  {e}")

使用公开数据集的示例如下：

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

# 准备的测试数据集
test_data = torchvision.datasets.CIFAR10("./dataset", train=False, transform=torchvision.transforms.ToTensor())

test_loader = DataLoader(test_data, batch_size=64, shuffle=True, num_workers=0, drop_last=True)

# 测试数据集中第一张图片及target
img, target = test_data[0]
print(img.shape)
print(target)

writer = SummaryWriter("dataloader")
for epoch in range(2):
    step = 0
    for data in test_loader:
        imgs, targets = data
        # print(imgs.shape)
        # print(targets)
        writer.add_images("Epoch: {}".format(epoch), imgs, step)
        step = step + 1

writer.close()

损失函数

Loss的用法实际上就两行代码的事情，以下是示例：

import torch
from torch.nn import L1Loss, MSELoss
from torch import nn

inputs = torch.tensor([1, 2, 3], dtype=torch.float)
targets = torch.tensor([1, 2, 5], dtype=torch.float)

inputs = torch.reshape(inputs, (1, 1, 1, 3))
targets = torch.reshape(targets, (1, 1, 1, 3))

loss = L1Loss(reduction='sum')
result = loss(inputs, targets)

loss_mse = MSELoss()
result_mse = loss_mse(inputs, targets)

print(result)
print(result_mse)

x = torch.tensor([0.1, 0.2, 0.3])
y = torch.tensor([1])
x = torch.reshape(x, (1, 3))
loss_cross = nn.CrossEntropyLoss()
result_cross = loss_cross(x, y)
print(result_cross)

优化器

优化器的使用也很简单，但要注意，在每一步训练之前都需要用optim.zero_grad()将梯度置零，避免梯度累加造成问题，用loss.backward()得到梯度以后用optim.step()更新参数

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


dataset = torchvision.datasets.CIFAR10("./dataset", train=False, transform=torchvision.transforms.ToTensor(),
                                       download=True)
dataloader = DataLoader(dataset, batch_size=1)

class Network(nn.Module):
    def __init__(self):
        super().__init__()
        self.model1 = Sequential(
            Conv2d(3, 32,5, padding=2),
            MaxPool2d(2),
            Conv2d(32, 32,5, padding=2),
            MaxPool2d(2),
            Conv2d(32, 64,5,padding=2),
            MaxPool2d(2),
            Flatten(),
            Linear(1024, 64),
            Linear(64,10)
        )

    def forward(self, x):
        x = self.model1(x)
        return x

loss = nn.CrossEntropyLoss()
network = Network()
optim = torch.optim.SGD(network.parameters(), lr=0.01)

for epoch in range(20):
    running_loss = 0.0
    for data in dataloader:
        imgs, targets = data
        outputs = network(imgs)
        result_loss = loss(outputs, targets)
        optim.zero_grad()
        result_loss.backward()
        optim.step()
        running_loss = running_loss + result_loss
    print(running_loss)

`gpu`手写和预测一个模型

`gpu`写模型

这里采用to(device)的方式使用gpu，对模型、损失函数和读数据部分使用to(device)调用gpu，其他和cpu并无区别：

import torch
import torchvision.datasets
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("cuda:0" if torch.cuda.is_available() else "cpu")
# 准备数据集
train_data = torchvision.datasets.CIFAR10("./dataset", train=True, transform=torchvision.transforms.ToTensor(),
                                          download=True)
test_data = torchvision.datasets.CIFAR10("./dataset", train=False, transform=torchvision.transforms.ToTensor(),
                                        download=True)

# length
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(f"测试数据集的长度为: {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
# 搭建神经网络
class Network(nn.Module):
    def __init__(self):
        super().__init__()
        self.model = nn.Sequential(
            nn.Conv2d(3, 32, 5, padding=2),
            nn.MaxPool2d(2),
            nn.Conv2d(32, 32, 5, padding=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

network = Network()
network.to(device)

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

# 优化器
# learning_rate = 0.01
learning_rate = 1e-2
optimizer = torch.optim.SGD(network.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 = 30

# 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
    network.train()
    for data in train_dataloader:
        imgs, targets = data
        imgs = imgs.to(device)
        targets = targets.to(device)
        outputs = network(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:
            end_time = time.time()
            print(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
    network.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 = network(imgs)
            loss = loss_fn(outputs, targets)
            total_test_loss = total_test_loss + loss
            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(network, "network_{}.pth".format(i))
    print("模型已保存")

writer.close()

gpu预测模型

把读取到的模型和数据用to(device)设置成gpu运行

import torch
import torchvision.transforms
from PIL import Image
from torch import nn

# 定义训练的设备
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
img_path = "dog.png"
image = Image.open(img_path)
print(image)

transform = torchvision.transforms.Compose([torchvision.transforms.Resize((32, 32)),
                                            torchvision.transforms.ToTensor()])
image = transform(image)
print(image.shape)

# 搭建神经网络
class Network(nn.Module):
    def __init__(self):
        super().__init__()
        self.model = nn.Sequential(
            nn.Conv2d(3, 32, 5, padding=2),
            nn.MaxPool2d(2),
            nn.Conv2d(32, 32, 5, padding=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 = torch.load("network_29.pth").to(device)
print(model)
image = torch.reshape(image, (1, 3, 32, 32))
image = image.to(device)
model.eval()
with torch.no_grad():
    output = model(image)
print(output)

print(output.argmax(1))