第P4周：猴痘病识别

1.设置CPU（如果有GPU就使用GPU，否则使用CPU）

>- **🍨 本文为[🔗365天深度学习训练营](https://mp.weixin.qq.com/s/Z9yL_wt7L8aPOr9Lqb1K3w) 中的学习记录博客**
>- **🍖 原作者：[K同学啊](https://mtyjkh.blog.csdn.net/)**

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

device(type='cpu')

2.导入数据

import os,PIL,random,pathlib

data_dir = './4-data/'
data_dir = pathlib.Path(data_dir)

data_paths = list(data_dir.glob('*'))
classeNames = [str(path).split("/")[1] for path in data_paths]
classeNames

['.DS_Store', 'Others', 'Monkeypox']

total_datadir = './4-data/'

train_transforms = 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转换成tensor并归一化到[0, 1]之间
    transforms.Normalize(                 ## 标准化处理-->转换为正态分布（高斯分布），使模型更加容易收敛
          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])从数据集随机抽样得到的
  ])

total_data = datasets.ImageFolder(total_datadir,transform=train_transforms)
total_data

Dataset ImageFolder
    Number of datapoints: 2142
    Root location: ./4-data/
    StandardTransform
Transform: Compose(
               Resize(size=[224, 224], interpolation=bilinear, max_size=None, antialias=True)
               ToTensor()
               Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
           )

total_data.class_to_idx

{'Monkeypox': 0, 'Others': 1}

3.划分数据集

train_size = int(0.8 * len(total_data))
test_size = len(total_data) - train_size
train_dataset, test_dataset = torch.utils.data.random_split(total_data, [train_size, test_size])
train_dataset, test_dataset

(<torch.utils.data.dataset.Subset at 0x146f09090>,
 <torch.utils.data.dataset.Subset at 0x146f094b0>)

train_size, test_size

(1713, 429)

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

二、构建简单的CNN网络

import torch.nn.functional as F

class Network_bn(nn.Module):
    def __init__(self):
        super(Network_bn, self).__init__()
        
        self.conv1 = nn.Conv2d(in_channels = 3, out_channels = 12, kernel_size = 5, stride = 1, padding = 0)
        self.bn1 = nn.BatchNorm2d(12)
        self.conv2 = nn.Conv2d(in_channels = 12, out_channels = 12, kernel_size = 5, stride = 1, padding = 0)
        self.bn2 = nn.BatchNorm2d(12)
        self.pool = nn.MaxPool2d(2, 2)
        self.conv4 = nn.Conv2d(in_channels = 12, out_channels = 24, kernel_size = 5, stride = 1, padding = 0)
        self.bn4 = nn.BatchNorm2d(24)
        self.conv5 = nn.Conv2d(in_channels = 24, out_channels = 24, kernel_size = 5, stride = 1, padding = 0)
        self.bn5 = nn.BatchNorm2d(24)
        self.fc1 = nn.Linear(24*50*50, len(classeNames))

    def forward(self, x):
        x = F.relu(self.bn1(self.conv1(x)))
        x = F.relu(self.bn2(self.conv2(x)))
        x = self.pool(x)
        x = F.relu(self.bn4(self.conv4(x)))
        x = F.relu(self.bn5(self.conv5(x)))
        x = self.pool(x)
        x = x.view(-1, 24*50*50)
        x = self.fc1(x)
        
        return x
    
device = "cuda" if torch.cuda.is_available() else "cpu"
print("Using {} device".format(device))
    
model = Network_bn().to(device)
model

Using cpu device





Network_bn(
  (conv1): Conv2d(3, 12, kernel_size=(5, 5), stride=(1, 1))
  (bn1): BatchNorm2d(12, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (conv2): Conv2d(12, 12, kernel_size=(5, 5), stride=(1, 1))
  (bn2): BatchNorm2d(12, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (pool): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
  (conv4): Conv2d(12, 24, kernel_size=(5, 5), stride=(1, 1))
  (bn4): BatchNorm2d(24, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (conv5): Conv2d(24, 24, kernel_size=(5, 5), stride=(1, 1))
  (bn5): BatchNorm2d(24, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (fc1): Linear(in_features=60000, out_features=3, bias=True)
)

三、训练模型

1.设置超参数

loss_fn = nn.CrossEntropyLoss()   ## 创建损失函数
learn_rate = 1e-4     ## 学习率
opt = torch.optim.SGD(model.parameters(), lr = learn_rate)

2.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)
    
    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)         ## 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

3.编写测试函数

def test (dataloader, model, loss_fn):
    size = len(dataloader.dataset)
    num_batches = len(dataloader)
    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

4.正式训练

epochs = 20
train_loss = []
train_acc = []
test_loss = []
test_acc = []

for epoch in range(epochs):
    model.train()
    epoch_train_acc, epoch_train_loss = train(train_dl, model, loss_fn, opt)
    
    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)
    
    template = ('Epoch:{:2d}, Train_acc:{:.1f}%, Train_loss:{:.3f}, Test_acc:{:.1f}%, Test_loss:{:.3f}')
    print(template.format(epoch+1, epoch_train_acc*100, epoch_train_loss, epoch_test_acc*100, epoch_test_loss))
print('Done')

Epoch: 1, Train_acc:58.9%, Train_loss:0.745, Test_acc:65.5%, Test_loss:0.637
Epoch: 2, Train_acc:67.7%, Train_loss:0.634, Test_acc:68.5%, Test_loss:0.596
Epoch: 3, Train_acc:71.8%, Train_loss:0.562, Test_acc:70.6%, Test_loss:0.571
Epoch: 4, Train_acc:74.4%, Train_loss:0.536, Test_acc:72.0%, Test_loss:0.571
Epoch: 5, Train_acc:77.6%, Train_loss:0.489, Test_acc:73.0%, Test_loss:0.553
Epoch: 6, Train_acc:80.2%, Train_loss:0.455, Test_acc:75.8%, Test_loss:0.518
Epoch: 7, Train_acc:81.7%, Train_loss:0.439, Test_acc:76.0%, Test_loss:0.500
Epoch: 8, Train_acc:81.9%, Train_loss:0.426, Test_acc:79.3%, Test_loss:0.479
Epoch: 9, Train_acc:84.6%, Train_loss:0.397, Test_acc:80.0%, Test_loss:0.460
Epoch:10, Train_acc:86.3%, Train_loss:0.371, Test_acc:80.4%, Test_loss:0.448
Epoch:11, Train_acc:87.2%, Train_loss:0.358, Test_acc:81.1%, Test_loss:0.436
Epoch:12, Train_acc:88.3%, Train_loss:0.347, Test_acc:82.5%, Test_loss:0.433
Epoch:13, Train_acc:87.9%, Train_loss:0.337, Test_acc:83.2%, Test_loss:0.413
Epoch:14, Train_acc:89.3%, Train_loss:0.323, Test_acc:83.2%, Test_loss:0.424
Epoch:15, Train_acc:89.5%, Train_loss:0.312, Test_acc:84.6%, Test_loss:0.424
Epoch:16, Train_acc:91.9%, Train_loss:0.295, Test_acc:84.1%, Test_loss:0.396
Epoch:17, Train_acc:91.2%, Train_loss:0.288, Test_acc:84.1%, Test_loss:0.387
Epoch:18, Train_acc:91.5%, Train_loss:0.279, Test_acc:83.4%, Test_loss:0.376
Epoch:19, Train_acc:92.0%, Train_loss:0.277, Test_acc:84.4%, Test_loss:0.393
Epoch:20, Train_acc:92.2%, Train_loss:0.270, Test_acc:83.7%, Test_loss:0.377
Done

四、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

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()

   2.指定图片进行预测

from PIL import Image

classes = list(total_data.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图片
predict_one_image(image_path='./4-data/Monkeypox/M01_01_00.jpg',
                  model=model,
                  transform=train_transforms,
                  classes=classes)

预测结果是:Monkeypox

五、保存并加载模型

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

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

<All keys matched successfully>

由于初始模型的Test_acc不是很高，所以以下将尝试一些改进：

改进1: 增加卷积核数量

import torch.nn.functional as F

class Network_bn(nn.Module):
    def __init__(self):
        super(Network_bn, self).__init__()
        
        self.conv1 = nn.Conv2d(in_channels = 3, out_channels = 32, kernel_size = 5, stride = 1, padding = 0)
        self.bn1 = nn.BatchNorm2d(32)
        self.conv2 = nn.Conv2d(in_channels = 32, out_channels = 64, kernel_size = 5, stride = 1, padding = 0)
        self.bn2 = nn.BatchNorm2d(64)
        self.pool = nn.MaxPool2d(2, 2)
        self.conv4 = nn.Conv2d(in_channels = 64, out_channels = 128, kernel_size = 5, stride = 1, padding = 0)
        self.bn4 = nn.BatchNorm2d(128)
        self.conv5 = nn.Conv2d(in_channels = 128, out_channels = 128, kernel_size = 5, stride = 1, padding = 0)
        self.bn5 = nn.BatchNorm2d(128)
        self.fc1 = nn.Linear(128*50*50, len(classeNames))

    def forward(self, x):
        x = F.relu(self.bn1(self.conv1(x)))
        x = F.relu(self.bn2(self.conv2(x)))
        x = self.pool(x)
        x = F.relu(self.bn4(self.conv4(x)))
        x = F.relu(self.bn5(self.conv5(x)))
        x = self.pool(x)
        x = x.view(-1, 128*50*50)
        x = self.fc1(x)
        
        return x
    
device = "cuda" if torch.cuda.is_available() else "cpu"
print("Using {} device".format(device))
    
model = Network_bn().to(device)
model

改进后：训练到epoch5发现效果很不好

原因分析：可能是增加卷积核数量后模型复杂度升高，而数据集规模不是很大，从而导致了过拟合。

改进2: 使用小卷积核（3*3）

import torch.nn.functional as F

class Network_bn(nn.Module):
    def __init__(self):
        super(Network_bn, self).__init__()
        
        self.conv1 = nn.Conv2d(in_channels = 3, out_channels = 12, kernel_size = 3, stride = 1, padding = 0)
        self.bn1 = nn.BatchNorm2d(12)
        self.conv2 = nn.Conv2d(in_channels = 12, out_channels = 12, kernel_size = 3, stride = 1, padding = 0)
        self.bn2 = nn.BatchNorm2d(12)
        self.pool = nn.MaxPool2d(2, 2)
        self.conv4 = nn.Conv2d(in_channels = 12, out_channels = 24, kernel_size = 3, stride = 1, padding = 0)
        self.bn4 = nn.BatchNorm2d(24)
        self.conv5 = nn.Conv2d(in_channels = 24, out_channels = 24, kernel_size = 3, stride = 1, padding = 0)
        self.bn5 = nn.BatchNorm2d(24)
        self.fc1 = nn.Linear(24*53*53, len(classeNames))

    def forward(self, x):
        x = F.relu(self.bn1(self.conv1(x)))
        x = F.relu(self.bn2(self.conv2(x)))
        x = self.pool(x)
        x = F.relu(self.bn4(self.conv4(x)))
        x = F.relu(self.bn5(self.conv5(x)))
        x = self.pool(x)
        x = x.view(-1, 24*53*53)
        x = self.fc1(x)
        
        return x
    
device = "cuda" if torch.cuda.is_available() else "cpu"
print("Using {} device".format(device))
    
model = Network_bn().to(device)
model

改进后结果：

对比之前5*5卷积核，结果没有提高。

原因分析：可能是由于模型感受野的减小，3乘3卷积核单次卷积操作中捕获的图像信息范围相较于5*5有所减小，因此模型对较大或全局特征的捕获能力不足。对于猴痘病识别这样的任务，可能某些大尺度特征对于分类比较重要，可能3乘3卷积核不足以提取这些特征。

改进3: 在原来的基础上增加dropout层：

import torch.nn.functional as F

class Network_bn(nn.Module):
    def __init__(self):
        super(Network_bn, self).__init__()
        
        self.conv1 = nn.Conv2d(in_channels = 3, out_channels = 12, kernel_size = 5, stride = 1, padding = 0)
        self.bn1 = nn.BatchNorm2d(12)
        self.conv2 = nn.Conv2d(in_channels = 12, out_channels = 12, kernel_size = 5, stride = 1, padding = 0)
        self.bn2 = nn.BatchNorm2d(12)
        self.pool = nn.MaxPool2d(2, 2)
        self.conv4 = nn.Conv2d(in_channels = 12, out_channels = 24, kernel_size = 5, stride = 1, padding = 0)
        self.bn4 = nn.BatchNorm2d(24)
        self.conv5 = nn.Conv2d(in_channels = 24, out_channels = 24, kernel_size = 5, stride = 1, padding = 0)
        self.bn5 = nn.BatchNorm2d(24)
        self.dropout = nn.Dropout(0.5)      ## dropout层
        self.fc1 = nn.Linear(24*50*50, len(classeNames))

    def forward(self, x):
        x = F.relu(self.bn1(self.conv1(x)))
        x = F.relu(self.bn2(self.conv2(x)))
        x = self.pool(x)
        x = F.relu(self.bn4(self.conv4(x)))
        x = F.relu(self.bn5(self.conv5(x)))
        x = self.pool(x)
        x = x.view(-1, 24*50*50)
        
        ## 在全连接层之前加入dropout层
        x = self.dropout(x)
        x = self.fc1(x)
        
        return x
    
device = "cuda" if torch.cuda.is_available() else "cpu"
print("Using {} device".format(device))
    
model = Network_bn().to(device)
model

改进后结果：

对比最开始的模型，Test_acc并没有提高。
原因分析：有可能参数选择不好，导致过度正则化，特别是在模型规模不大时。
同时有可能模型结构比较简单，加入 Dropout 后模型的表达能力不足，因此影响了其在测试集上的表现。

改进4: 应用动态学习率

loss_fn = nn.CrossEntropyLoss()   ## 创建损失函数
learn_rate = 1e-4     ## 学习率
opt = torch.optim.Adam(model.parameters(), lr = learn_rate, weight_decay=1e-4)  ## 使用Adam优化器
scheduler = torch.optim.lr_scheduler.StepLR(opt, step_size=10, gamma=0.1)  ## 动态学习率

改进结果：

可以发现Test_acc达到90%改进效果较为显著。

总结：
在改进模型结构时：
1.我们增加其深度即增加卷积核核个数，理论上是可以提高其测试集准确度，但是在训练集规模较小时可能会过拟合。
2.我们在尝试减小卷积核时，理论上是可以通过小的卷积核来更细致地捕捉图像的局部特征。这样有助于模型提取更多有用的细节信息，从而提高识别准确率。但是同时模型感受野会减小，从而导致模型对较大或全局特征的捕获能力不足，最终影响分类性能。
设置动态学习率，可以根据策略自动调整学习率，帮助模型更快收敛并且避免其陷入局部最优解。使用效果也是较为突出。