第P5周：Pytorch实现运动鞋识别

最新推荐文章于 2025-05-27 23:20:12 发布

小羊的 utopia

最新推荐文章于 2025-05-27 23:20:12 发布

阅读量174

点赞数 6

文章标签： pytorch 人工智能 python

本文链接：https://blog.youkuaiyun.com/2405_87523719/article/details/145931373

版权

🍨 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

🚀我的环境：

语言环境：python 3.12.6
编译器：jupyter lab
深度学习环境：Pytorch

前期准备

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

import os,PIL,pathlib

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

device

C:\Users\yxy\AppData\Local\Programs\Python\Python312\Lib\site-packages\torch\cuda\__init__.py:129: UserWarning: CUDA initialization: The NVIDIA driver on your system is too old (found version 10020). Please update your GPU driver by downloading and installing a new version from the URL: http://www.nvidia.com/Download/index.aspx Alternatively, go to: https://pytorch.org to install a PyTorch version that has been compiled with your version of the CUDA driver. (Triggered internally at C:\actions-runner\_work\pytorch\pytorch\builder\windows\pytorch\c10\cuda\CUDAFunctions.cpp:108.)
  return torch._C._cuda_getDeviceCount() > 0





device(type='cpu')

import os,PIL,random,pathlib

data_dir = 'd:/Users/yxy/Desktop/46-data'
data_dir = pathlib.Path(data_dir)

data_paths  = list(data_dir.glob('*'))
classeNames = [path.name for path in data_paths]
classeNames

['test', 'train']

# 关于transforms.Compose的更多介绍可以参考：https://blog.youkuaiyun.com/qq_38251616/article/details/124878863
train_transforms = transforms.Compose([
    transforms.Resize([224, 224]),  # 将输入图片resize成统一尺寸
    # transforms.RandomHorizontalFlip(), # 随机水平翻转
    transforms.ToTensor(),          # 将PIL Image或numpy.ndarray转换为tensor，并归一化到[0,1]之间
    transforms.Normalize(           # 标准化处理-->转换为标准正太分布（高斯分布），使模型更容易收敛
        mean=[0.485, 0.456, 0.406], 
        std=[0.229, 0.224, 0.225])  # 其中 mean=[0.485,0.456,0.406]与std=[0.229,0.224,0.225] 从数据集中随机抽样计算得到的。
])

test_transform = transforms.Compose([
    transforms.Resize([224, 224]),  # 将输入图片resize成统一尺寸
    transforms.ToTensor(),          # 将PIL Image或numpy.ndarray转换为tensor，并归一化到[0,1]之间
    transforms.Normalize(           # 标准化处理-->转换为标准正太分布（高斯分布），使模型更容易收敛
        mean=[0.485, 0.456, 0.406], 
        std=[0.229, 0.224, 0.225])  # 其中 mean=[0.485,0.456,0.406]与std=[0.229,0.224,0.225] 从数据集中随机抽样计算得到的。
])

train_dataset = datasets.ImageFolder('d:/Users/yxy/Desktop/46-data/train',transform=train_transforms)
test_dataset  = datasets.ImageFolder('d:/Users/yxy/Desktop/46-data/test',transform=test_transform)

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 X [N, C, H, W]: ", X.shape)
    print("Shape of y: ", y.shape, y.dtype)
    break

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

构建CNN

import torch.nn.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), # 12*220*220
            nn.BatchNorm2d(12),
            nn.ReLU())
        
        self.conv2=nn.Sequential(
            nn.Conv2d(12, 12, kernel_size=5, padding=0), # 12*216*216
            nn.BatchNorm2d(12),
            nn.ReLU())
        
        self.pool3=nn.Sequential(
            nn.MaxPool2d(2))                              # 12*108*108
        
        self.conv4=nn.Sequential(
            nn.Conv2d(12, 24, kernel_size=5, padding=0), # 24*104*104
            nn.BatchNorm2d(24),
            nn.ReLU())
        
        self.conv5=nn.Sequential(
            nn.Conv2d(24, 24, kernel_size=5, padding=0), # 24*100*100
            nn.BatchNorm2d(24),
            nn.ReLU())
        
        self.pool6=nn.Sequential(
            nn.MaxPool2d(2))                              # 24*50*50

        self.dropout = nn.Sequential(
            nn.Dropout(0.2))
        
        self.fc=nn.Sequential(
            nn.Linear(24*50*50, len(classeNames)))
        
    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 cpu 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)
  )
)

训练模型

# 训练循环
def train(dataloader, model, loss_fn, optimizer):
    size = len(dataloader.dataset)  # 训练集的大小
    num_batches = len(dataloader)   # 批次数目, (size/batch_size，向上取整)

    train_loss, train_acc = 0, 0  # 初始化训练损失和正确率
    
    for X, y in dataloader:  # 获取图片及其标签
        X, y = X.to(device), y.to(device)
        
        # 计算预测误差
        pred = model(X)          # 网络输出
        loss = loss_fn(pred, y)  # 计算网络输出和真实值之间的差距，targets为真实值，计算二者差值即为损失
        
        # 反向传播
        optimizer.zero_grad()  # grad属性归零
        loss.backward()        # 反向传播
        optimizer.step()       # 每一步自动更新
        
        # 记录acc与loss
        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

def test (dataloader, model, loss_fn):
    size        = len(dataloader.dataset)  # 测试集的大小
    num_batches = len(dataloader)          # 批次数目, (size/batch_size，向上取整)
    test_loss, test_acc = 0, 0
    
    # 当不进行训练时，停止梯度更新，节省计算内存消耗
    with torch.no_grad():
        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

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 # 初始学习率
optimizer  = torch.optim.SGD(model.parameters(), lr=learn_rate)

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:55.2%, Train_loss:0.754, Test_acc:59.2%, Test_loss:0.692, Lr:1.00E-04
Epoch: 2, Train_acc:62.9%, Train_loss:0.664, Test_acc:60.5%, Test_loss:0.659, Lr:1.00E-04
Epoch: 3, Train_acc:70.1%, Train_loss:0.590, Test_acc:64.5%, Test_loss:0.617, Lr:9.20E-05
Epoch: 4, Train_acc:71.3%, Train_loss:0.552, Test_acc:68.4%, Test_loss:0.581, Lr:9.20E-05
Epoch: 5, Train_acc:70.9%, Train_loss:0.536, Test_acc:71.1%, Test_loss:0.605, Lr:8.46E-05
Epoch: 6, Train_acc:76.3%, Train_loss:0.502, Test_acc:61.8%, Test_loss:0.608, Lr:8.46E-05
Epoch: 7, Train_acc:78.1%, Train_loss:0.473, Test_acc:69.7%, Test_loss:0.538, Lr:7.79E-05
Epoch: 8, Train_acc:79.7%, Train_loss:0.455, Test_acc:72.4%, Test_loss:0.534, Lr:7.79E-05
Epoch: 9, Train_acc:84.3%, Train_loss:0.421, Test_acc:72.4%, Test_loss:0.542, Lr:7.16E-05
Epoch:10, Train_acc:84.9%, Train_loss:0.404, Test_acc:73.7%, Test_loss:0.534, Lr:7.16E-05
Epoch:11, Train_acc:85.7%, Train_loss:0.389, Test_acc:73.7%, Test_loss:0.506, Lr:6.59E-05
Epoch:12, Train_acc:86.5%, Train_loss:0.392, Test_acc:75.0%, Test_loss:0.490, Lr:6.59E-05
Epoch:13, Train_acc:86.3%, Train_loss:0.381, Test_acc:75.0%, Test_loss:0.526, Lr:6.06E-05
Epoch:14, Train_acc:89.2%, Train_loss:0.360, Test_acc:76.3%, Test_loss:0.489, Lr:6.06E-05
Epoch:15, Train_acc:89.6%, Train_loss:0.347, Test_acc:73.7%, Test_loss:0.523, Lr:5.58E-05
Epoch:16, Train_acc:91.2%, Train_loss:0.347, Test_acc:73.7%, Test_loss:0.501, Lr:5.58E-05
Epoch:17, Train_acc:89.6%, Train_loss:0.339, Test_acc:76.3%, Test_loss:0.548, Lr:5.13E-05
Epoch:18, Train_acc:89.2%, Train_loss:0.347, Test_acc:73.7%, Test_loss:0.467, Lr:5.13E-05
Epoch:19, Train_acc:90.8%, Train_loss:0.320, Test_acc:78.9%, Test_loss:0.489, Lr:4.72E-05
Epoch:20, Train_acc:91.0%, Train_loss:0.325, Test_acc:77.6%, Test_loss:0.483, Lr:4.72E-05
Epoch:21, Train_acc:92.6%, Train_loss:0.309, Test_acc:75.0%, Test_loss:0.510, Lr:4.34E-05
Epoch:22, Train_acc:90.6%, Train_loss:0.313, Test_acc:80.3%, Test_loss:0.484, Lr:4.34E-05
Epoch:23, Train_acc:92.2%, Train_loss:0.300, Test_acc:78.9%, Test_loss:0.519, Lr:4.00E-05
Epoch:24, Train_acc:92.8%, Train_loss:0.298, Test_acc:76.3%, Test_loss:0.450, Lr:4.00E-05
Epoch:25, Train_acc:92.6%, Train_loss:0.294, Test_acc:76.3%, Test_loss:0.530, Lr:3.68E-05
Epoch:26, Train_acc:92.2%, Train_loss:0.290, Test_acc:77.6%, Test_loss:0.509, Lr:3.68E-05
Epoch:27, Train_acc:93.8%, Train_loss:0.291, Test_acc:78.9%, Test_loss:0.461, Lr:3.38E-05
Epoch:28, Train_acc:94.8%, Train_loss:0.271, Test_acc:80.3%, Test_loss:0.466, Lr:3.38E-05
Epoch:29, Train_acc:93.4%, Train_loss:0.275, Test_acc:77.6%, Test_loss:0.502, Lr:3.11E-05
Epoch:30, Train_acc:93.8%, Train_loss:0.276, Test_acc:76.3%, Test_loss:0.449, Lr:3.11E-05
Epoch:31, Train_acc:95.4%, Train_loss:0.257, Test_acc:80.3%, Test_loss:0.456, Lr:2.86E-05
Epoch:32, Train_acc:94.0%, Train_loss:0.276, Test_acc:76.3%, Test_loss:0.521, Lr:2.86E-05
Epoch:33, Train_acc:93.4%, Train_loss:0.266, Test_acc:80.3%, Test_loss:0.455, Lr:2.63E-05
Epoch:34, Train_acc:94.2%, Train_loss:0.262, Test_acc:78.9%, Test_loss:0.486, Lr:2.63E-05
Epoch:35, Train_acc:94.6%, Train_loss:0.262, Test_acc:80.3%, Test_loss:0.441, Lr:2.42E-05
Epoch:36, Train_acc:96.0%, Train_loss:0.252, Test_acc:78.9%, Test_loss:0.482, Lr:2.42E-05
Epoch:37, Train_acc:96.6%, Train_loss:0.257, Test_acc:78.9%, Test_loss:0.432, Lr:2.23E-05
Epoch:38, Train_acc:94.8%, Train_loss:0.257, Test_acc:80.3%, Test_loss:0.472, Lr:2.23E-05
Epoch:39, Train_acc:94.6%, Train_loss:0.253, Test_acc:78.9%, Test_loss:0.447, Lr:2.05E-05
Epoch:40, Train_acc:94.6%, Train_loss:0.254, Test_acc:78.9%, Test_loss:0.442, Lr:2.05E-05
Done

结果可视化

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

from datetime import datetime
current_time = datetime.now() # 获取当前时间

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.xlabel(current_time) # 打卡请带上时间戳，否则代码截图无效

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

>>> x = torch.zeros(2, 1, 2, 1, 2)
>>> x.size()
torch.Size([2, 1, 2, 1, 2])
>>> y = torch.squeeze(x)
>>> y.size()
torch.Size([2, 2, 2])
>>> y = torch.squeeze(x, 0)
>>> y.size()
torch.Size([2, 1, 2, 1, 2])
>>> y = torch.squeeze(x, 1)
>>> y.size()
torch.Size([2, 2, 1, 2])

torch.Size([2, 2, 1, 2])

import torch  # 确保导入 torch

x = torch.tensor([1, 2, 3, 4])

# 在维度 0 上增加一个维度
x_unsqueezed_0 = torch.unsqueeze(x, 0)
print(x_unsqueezed_0)  
# 输出: tensor([[1, 2, 3, 4]])

# 在维度 1 上增加一个维度
x_unsqueezed_1 = torch.unsqueeze(x, 1)
print(x_unsqueezed_1)  
# 输出:
# tensor([[1],
#         [2],
#         [3],
#         [4]])

tensor([[1, 2, 3, 4]])
tensor([[1],
        [2],
        [3],
        [4]])

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='d:/Users/yxy/Desktop/46-data/test/adidas/0.jpg', 
                  model=model, 
                  transform=train_transforms, 
                  classes=classes)

预测结果是：adidas

# 模型保存
PATH = 'd:/Users/yxy/Desktop/46-data/model.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>

import torch
import torch.nn as nn
import torch.optim as optim

# 定义一个简单的神经网络
class SimpleNet(nn.Module):
    def __init__(self):
        super(SimpleNet, self).__init__()
        self.fc1 = nn.Linear(10, 5)  # 输入维度10，输出维度5

    def forward(self, x):
        return self.fc1(x)

# 创建模型实例
net = SimpleNet()

# 定义优化器
optimizer = optim.SGD(net.parameters(), lr=0.001)

# 定义学习率调度器
scheduler = optim.lr_scheduler.StepLR(optimizer, step_size=5, gamma=0.1)

print("优化器和学习率调度器已正确初始化！")

优化器和学习率调度器已正确初始化！

from torch.optim import SGD
from torch.optim.lr_scheduler import ExponentialLR

# 创建一个可训练参数
model = [nn.Parameter(torch.randn(2, 2, requires_grad=True))]

# 定义优化器
optimizer = SGD(model, lr=0.1)

# 定义指数学习率调度器
scheduler = ExponentialLR(optimizer, gamma=0.9)

print("优化器和学习率调度器已正确初始化！")

优化器和学习率调度器已正确初始化！

总结
本文使用了 torch.optim.lr_scheduler.StepLR
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。函数原型是torch.optim.lr_scheduler.StepLR(optimizer, step_size, gamma=0.1, last_epoch=-1)