P5_Pytorch实现运动鞋识别

最新推荐文章于 2024-09-12 09:11:48 发布

BCI_0

最新推荐文章于 2024-09-12 09:11:48 发布

阅读量654

点赞数 17

分类专栏：深度学习笔记文章标签： pytorch 人工智能 python

本文链接：https://blog.csdn.net/qq_44611551/article/details/141250980

版权

深度学习笔记专栏收录该内容

8 篇文章 0 订阅

订阅专栏

🍨 本文为🔗365天深度学习训练营中的学习记录博客
🍖 原作者：K同学啊

🍺要求：

了解如何设置动态学习率（重点）
调整代码使测试集accuracy到达84%。

🍻拔高（可选）：

保存训练过程中的最佳模型权重
调整代码使测试集accuracy到达86%。

🏡 我的环境：

● 语言环境：Python3.8
● 编译器：Jupyter Lab
● 深度学习环境：Pytorch
○ torch == 2.2.2 + cu121
○ torchvision == 0.17.2 + cu121

一、前期准备

1.设置GPU

import torch
import torch.nn as nn
import torchvision.transforms as transforms
import torchvision
from torchvision import datasets

import PIL, os, pathlib

device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')

device

device(type='cuda')

2.导入数据

import os, PIL, random, pathlib

data_dir = './data/P5_data/'
data_dir = pathlib.Path(data_dir)

data_paths = list(data_dir.glob('*'))
classNames = [str(path).split('\\')[2] for path in data_paths]
classNames

['test', 'train']

train_transforms = transforms.Compose([
    transforms.Resize([224, 224]),
    # transforms.RandomHorizontalFlip(), # 随机水平翻转
    transforms.ToTensor(),
    transforms.Normalize(
        mean=[0.485, 0.456, 0.406], 
        std=[0.229, 0.224, 0.225])
])

test_transform = transforms.Compose([
    transforms.Resize([224, 224]),
    transforms.ToTensor(),
    transforms.Normalize(
        mean=[0.485, 0.456, 0.406], 
        std=[0.229, 0.224, 0.225])
])

train_dataset = datasets.ImageFolder("./data/P5_data/train/",transform=train_transforms)
test_dataset  = datasets.ImageFolder("./data/P5_data/test/",transform=train_transforms)

train_dataset.class_to_idx

{'adidas': 0, 'nike': 1}

batch_size = 32

train_dl = torch.utils.data.DataLoader(train_dataset,
                                           batch_size=batch_size,
                                           shuffle=True,
                                           num_workers=1)
test_dl = torch.utils.data.DataLoader(test_dataset,
                                          batch_size=batch_size,
                                          shuffle=True,
                                          num_workers=1)

for X, y in test_dl:
    print('Shape of [N, C, H, W]', X.shape)
    print('Shape of y', y.shape, y.dtype)
    break

Shape of [N, C, H, W] torch.Size([32, 3, 224, 224])
Shape of y torch.Size([32]) torch.int64

二、构建简单的CNN网络

import torch.functional as F

class Model(nn.Module):
    def __init__(self):
        super(Model, self).__init__()
        self.conv1 = nn.Sequential(
        nn.Conv2d(3, 12, kernel_size=5, padding=0),
        nn.BatchNorm2d(12),
        nn.ReLU())
        
        self.conv2 = nn.Sequential(
        nn.Conv2d(12, 12, kernel_size=5, padding = 0),
        nn.BatchNorm2d(12),
        nn.ReLU())
        
        self.pool3 = nn.Sequential(
        nn.MaxPool2d(2))
        
        self.conv4 = nn.Sequential(
        nn.Conv2d(12, 24, kernel_size=5, padding=0),
        nn.BatchNorm2d(24),
        nn.ReLU())
        
        self.conv5 = nn.Sequential(
        nn.Conv2d(24, 24, kernel_size=5, padding=0),
        nn.BatchNorm2d(24),
        nn.ReLU())
        
        self.pool6 = nn.Sequential(
        nn.MaxPool2d(2))
        
        self.dropout = nn.Sequential(
        nn.Dropout(0.2))
        
        self.fc = nn.Sequential(
        nn.Linear(24*50*50, len(classNames)))
        
    def forward(self, x):
        
        batch_size = x.size(0)
        x = self.conv1(x)  # 卷积-BN-激活
        x = self.conv2(x)  # 卷积-BN-激活
        x = self.pool3(x)  # 池化
        x = self.conv4(x)  # 卷积-BN-激活
        x = self.conv5(x)  # 卷积-BN-激活
        x = self.pool6(x)  # 池化
        x = self.dropout(x)
        x = x.view(batch_size, -1)  # flatten 变成全连接网络需要的输入 (batch, 24*50*50) ==> (batch, -1), -1 此处自动算出的是24*50*50
        x = self.fc(x)
       
        return x

device = 'cuda' if torch.cuda.is_available() else 'cpu'
print('Using {} device'.format(device))

model = Model().to(device)
model

Using cuda device 

Model(
  (conv1): Sequential(
    (0): Conv2d(3, 12, kernel_size=(5, 5), stride=(1, 1))
    (1): BatchNorm2d(12, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (2): ReLU()
  )
  (conv2): Sequential(
    (0): Conv2d(12, 12, kernel_size=(5, 5), stride=(1, 1))
    (1): BatchNorm2d(12, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (2): ReLU()
  )
  (pool3): Sequential(
    (0): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
  )
  (conv4): Sequential(
    (0): Conv2d(12, 24, kernel_size=(5, 5), stride=(1, 1))
    (1): BatchNorm2d(24, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (2): ReLU()
  )
  (conv5): Sequential(
    (0): Conv2d(24, 24, kernel_size=(5, 5), stride=(1, 1))
    (1): BatchNorm2d(24, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (2): ReLU()
  )
  (pool6): Sequential(
    (0): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
  )
  (dropout): Sequential(
    (0): Dropout(p=0.2, inplace=False)
  )
  (fc): Sequential(
    (0): Linear(in_features=60000, out_features=2, bias=True)
  )
)

三、训练模型

1.编写训练函数

# 训练循环
def train(dataloader, model, loss_fn, optimizer):
    size = len(dataloader.dataset)
    num_batches = len(dataloader) # 批次数目，(size / batch_size)
    
    train_loss, train_acc = 0, 0
    
    for X, y in dataloader:
        X, y = X.to(device), y.to(device)
        
        # 计算预测误差
        pred = model(X)
        loss = loss_fn(pred, y)
        
        # 反向传播
        optimizer.zero_grad() # 梯度归零
        loss.backward() # 反向传播
        optimizer.step() # 每一步自动更新
        
        train_acc += (pred.argmax(1) == y).type(torch.float).sum().item()
        train_loss += loss.item()
        
    train_acc /= size
    train_loss /= num_batches
    
    return train_acc, train_loss

# 测试循环
def test(dataloader, model, loss_fn):
    size = len(dataloader.dataset)
    num_batches = len(dataloader)
    
    test_loss, test_acc = 0, 0
    
    with torch.no_grad(): #停止梯度更新
        for imgs, target in dataloader:
            imgs, target = imgs.to(device), target.to(device)
            
            # 计算loss
            target_pred = model(imgs)
            loss = loss_fn(target_pred, target)
            
            test_loss += loss.item()
            test_acc += (target_pred.argmax(1) == target).type(torch.float).sum().item()
            
    test_acc /= size
    test_loss /= num_batches
    
    return test_acc, test_loss

3.设置动态学习率

def adjust_learning_rate(optimizer, epoch, start_lr):
    # 每2个epoch衰减到原来的0.92
    lr = start_lr * (0.92 ** (epoch // 2))
    for param_group in optimizer.param_groups:
        param_group['lr'] = lr
        
learn_rate = 1e-4
optimizer = torch.optim.SGD(model.parameters(), lr = learn_rate)

# # 调用官方动态学习率接口时使用
# lambda1 = lambda epoch: (0.92 ** (epoch // 2))
# optimizer = torch.optim.SGD(model.parameters(), lr=learn_rate)
# scheduler = torch.optim.lr_scheduler.LambdaLR(optimizer, lr_lambda=lambda1) #选定调整方法

4.正式训练

loss_fn    = nn.CrossEntropyLoss() # 创建损失函数
epochs     = 40

train_loss = []
train_acc  = []
test_loss  = []
test_acc   = []

for epoch in range(epochs):
    # 更新学习率（使用自定义学习率时使用）
    adjust_learning_rate(optimizer, epoch, learn_rate)
    
    model.train()
    epoch_train_acc, epoch_train_loss = train(train_dl, model, loss_fn, optimizer)
    # scheduler.step() # 更新学习率（调用官方动态学习率接口时使用）
    
    model.eval()
    epoch_test_acc, epoch_test_loss = test(test_dl, model, loss_fn)
    
    train_acc.append(epoch_train_acc)
    train_loss.append(epoch_train_loss)
    test_acc.append(epoch_test_acc)
    test_loss.append(epoch_test_loss)
    
    # 获取当前的学习率
    lr = optimizer.state_dict()['param_groups'][0]['lr']
    
    template = ('Epoch:{:2d}, Train_acc:{:.1f}%, Train_loss:{:.3f}, Test_acc:{:.1f}%, Test_loss:{:.3f}, Lr:{:.2E}')
    print(template.format(epoch+1, epoch_train_acc*100, epoch_train_loss, 
                          epoch_test_acc*100, epoch_test_loss, lr))
print('Done')

Epoch: 1, Train_acc:49.2%, Train_loss:0.818, Test_acc:56.6%, Test_loss:0.682, Lr:1.00E-04
Epoch: 2, Train_acc:59.2%, Train_loss:0.688, Test_acc:65.8%, Test_loss:0.651, Lr:1.00E-04
Epoch: 3, Train_acc:66.1%, Train_loss:0.625, Test_acc:69.7%, Test_loss:0.633, Lr:9.20E-05
Epoch: 4, Train_acc:68.1%, Train_loss:0.602, Test_acc:69.7%, Test_loss:0.614, Lr:9.20E-05
Epoch: 5, Train_acc:72.9%, Train_loss:0.572, Test_acc:71.1%, Test_loss:0.613, Lr:8.46E-05
Epoch: 6, Train_acc:70.9%, Train_loss:0.558, Test_acc:67.1%, Test_loss:0.539, Lr:8.46E-05
Epoch: 7, Train_acc:74.7%, Train_loss:0.523, Test_acc:73.7%, Test_loss:0.538, Lr:7.79E-05
Epoch: 8, Train_acc:75.7%, Train_loss:0.507, Test_acc:78.9%, Test_loss:0.550, Lr:7.79E-05
Epoch: 9, Train_acc:78.7%, Train_loss:0.476, Test_acc:75.0%, Test_loss:0.583, Lr:7.16E-05
Epoch:10, Train_acc:82.1%, Train_loss:0.451, Test_acc:73.7%, Test_loss:0.546, Lr:7.16E-05
......
Epoch:31, Train_acc:93.0%, Train_loss:0.310, Test_acc:77.6%, Test_loss:0.502, Lr:2.86E-05
Epoch:32, Train_acc:94.2%, Train_loss:0.300, Test_acc:76.3%, Test_loss:0.492, Lr:2.86E-05
Epoch:33, Train_acc:94.6%, Train_loss:0.296, Test_acc:75.0%, Test_loss:0.501, Lr:2.63E-05
Epoch:34, Train_acc:92.4%, Train_loss:0.298, Test_acc:75.0%, Test_loss:0.474, Lr:2.63E-05
Epoch:35, Train_acc:94.4%, Train_loss:0.293, Test_acc:75.0%, Test_loss:0.453, Lr:2.42E-05
Epoch:36, Train_acc:94.8%, Train_loss:0.288, Test_acc:75.0%, Test_loss:0.489, Lr:2.42E-05
Epoch:37, Train_acc:94.4%, Train_loss:0.286, Test_acc:75.0%, Test_loss:0.476, Lr:2.23E-05
Epoch:38, Train_acc:93.8%, Train_loss:0.297, Test_acc:75.0%, Test_loss:0.467, Lr:2.23E-05
Epoch:39, Train_acc:94.2%, Train_loss:0.278, Test_acc:75.0%, Test_loss:0.471, Lr:2.05E-05
Epoch:40, Train_acc:94.8%, Train_loss:0.275, Test_acc:76.3%, Test_loss:0.502, Lr:2.05E-05
Done

四、结果可视化

1.Loss与Accuracy图

import matplotlib.pyplot as plt
#隐藏警告
import warnings
warnings.filterwarnings("ignore")               #忽略警告信息
plt.rcParams['font.sans-serif']    = ['SimHei'] # 用来正常显示中文标签
plt.rcParams['axes.unicode_minus'] = False      # 用来正常显示负号
plt.rcParams['figure.dpi']         = 100        #分辨率

epochs_range = range(epochs)

plt.figure(figsize=(12, 3))
plt.subplot(1, 2, 1)

plt.plot(epochs_range, train_acc, label='Training Accuracy')
plt.plot(epochs_range, test_acc, label='Test Accuracy')
plt.legend(loc='lower right')
plt.title('Training and Validation Accuracy')

plt.subplot(1, 2, 2)
plt.plot(epochs_range, train_loss, label='Training Loss')
plt.plot(epochs_range, test_loss, label='Test Loss')
plt.legend(loc='upper right')
plt.title('Training and Validation Loss')
plt.show()

在这里插入图片描述

2.指定图片进行预测

from PIL import Image 

classes = list(train_dataset.class_to_idx)

def predict_one_image(image_path, model, transform, classes):
    
    test_img = Image.open(image_path).convert('RGB')
    plt.imshow(test_img)  # 展示预测的图片

    test_img = transform(test_img)
    img = test_img.to(device).unsqueeze(0)
    
    model.eval()
    output = model(img)

    _,pred = torch.max(output,1)
    pred_class = classes[pred]
    print(f'预测结果是：{pred_class}')

# 预测训练集中的某张照片
predict_one_image(image_path='./data/P5_data/test/adidas/1.jpg', 
                  model=model, 
                  transform=train_transforms, 
                  classes=classes)

预测结果是：adidas

在这里插入图片描述

五、保存并加载模型

# 模型保存
PATH = './model/model_p5.pth'  # 保存的参数文件名
torch.save(model.state_dict(), PATH)

# 将参数加载到model当中
model.load_state_dict(torch.load(PATH, map_location=device))

<All keys matched successfully>

六、动态学习率

1. torch.optim.lr_scheduler.StepLR

torch.optim.lr_scheduler.StepLR(optimizer, step_size, gamma)
    # 等间隔动态调整方法，每经过step_size个epoch，做一次学习率decay，以gamma值为缩小倍数。
    """
    optimizer(Optimizer)：是之前定义好的需要优化的优化器的实例名
    step_size(int)：是学习率衰减的周期，每经过每个epoch，做一次学习率decay
    gamma(float)：学习率衰减的乘法因子。Default：0.1
    """

optimizer = torch.optim.SGD(net.parameters(), lr = 0.001)
scheduler = torch.optim.lr_scheduler.StepLR(optimizer, step_size=5, gamma=0.1)

2. lr_scheduler.LambdaLR

torch.optim.lr_scheduler.LambdaLR(optimizer, lr_lambda)
    # 根据自己定义的函数更新学习率。
    """
    optimizer(Optimizer)：是之前定义好的需要优化的优化器的实例名
    lr_lambda(function)：更新学习率的函数
    """

lambda1 = lambda epoch: (0.92 ** (epoch // 2) # 第二组参数的调整方法
optimizer = torch.optim.SGD(model.parameters(), lr=learn_rate)
scheduler = torch.optim.lr_scheduler.LambdaLR(optimizer, lr_lambda=lambda1)

3. lr_scheduler.MultiStepLR

torch.optim.lr_scheduler.MultiStepLR(optimizer, milestones, gamma=0.1)
    # 在特定的 epoch 中调整学习率
    """
    optimizer(Optimizer)：是之前定义好的需要优化的优化器的实例名
    milestones(list)：是一个关于epoch数值的list，表示在达到哪个epoch范围内开始变化，必须是升序排列
    gamma(float)：学习率衰减的乘法因子。Default：0.1
    """

optimizer = torch.optim.SGD(net.parameters(), lr=0.001 )
scheduler = torch.optim.lr_scheduler.MultiStepLR(optimizer, 
                                                 milestones=[2,6,15], #调整学习率的epoch数
                                                 gamma=0.1)

七、总结

使用nn.Sequential() 函数可以让代码可读性更好，要学会使用该函数。
动态学习率其作用有如下：动态调整学习率通过在训练过程中适时地调整学习率，可以提高模型的收敛速度、稳定性和泛化能力。
2.1 收敛速度：合理调整学习率可以加快模型的收敛速度，减少训练所需时间。
2.2 稳定性：通过自适应调整学习率，可以使模型在训练过程中更加稳定，减少训练过程中的振荡和离散。
2.3 泛化能力：合理调整学习率可以提高模型的泛化能力。