P4-猴痘病图片识别(pytorch+CNN)

P4：猴痘病识别

• 🍨 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

目标：

• 训练过程中保存效果最好的模型参数
• 加载最佳模型参数识别本地的一张图片
• 调整网络结构使test accuracy达到88%

提升：

• 调整模型参数并观察test acc的变化
• 尝试设置动态学习率
• test acc到达90%

环境设置:

• 语言环境：python 3.10.12
• 编译器：Google Colab
• 学习框架：PyTorch;torch version:2.2.1+cu121

import torch
import torch.nn as nn
import torchvision.transforms as transforms
import torchvision
from torchvision import datasets
import os, PIL, pathlib, random

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

cpu





'2.2.0+cpu'

导入数据

from google.colab import drive
drive.mount("/content/drive/")

Mounted at /content/drive/

%cd "/content/drive/MyDrive/Colab Notebooks/jupyter notebook/data/P4"
!ls

/content/drive/Othercomputers/My laptop/jupyter notebook/data/P4
Monkeypox  Others

path = "./"
data_dir = pathlib.Path(path)

data_paths = list(data_dir.glob("*"))

classNames = [str(path) for path in data_paths] #这里分割符号使用“/",用什么符号split可以print来检查
classNames

['Monkeypox', 'Others']

total_data = "./"

# 关于transforms.Compose的更多介绍可以参考：https://blog.csdn.net/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.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。

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

Dataset ImageFolder
    Number of datapoints: 2142
    Root location: ./
    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}

total_data.class_to_idx是一个存储了数据集类别和对应索引的字典。在PyTorch的ImageFolder数据加载器中，根据数据集文件夹的组织结构，每个文件夹代表一个类别，class_to_idx字典将每个类别名称映射为一个数字索引。

具体来说，如果数据集文件夹包含两个子文件夹，比如Monkeypox和Others，class_to_idx字典将返回类似以下的映射关系：{'Monkeypox': 0, 'Others': 1}

划分数据集

train_size = int(0.8*(len(total_data)))
test_size = len(total_data) - train_size
train_set, test_set = torch.utils.data.random_split(total_data,[train_size,test_size])#注意torch.utils.data的使用
train_set, test_set

(<torch.utils.data.dataset.Subset at 0x7f3f7c1a3610>,
 <torch.utils.data.dataset.Subset at 0x7f3f7c1a1b10>)

train_size, test_size

(1713, 429)

batch_size = 32

train_data = torch.utils.data.DataLoader(train_set, batch_size = batch_size, shuffle=True, num_workers=1,drop_last=True)
test_data = torch.utils.data.DataLoader(test_set,batch_size = batch_size, shuffle=True, num_workers=1,drop_last=True)

for X, y in test_data:
    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 Network_bn(nn.Module):
    def __init__(self):
        super(Network_bn, self).__init__()
        """
        nn.Conv2d()函数：
        第一个参数（in_channels）是输入的channel数量
        第二个参数（out_channels）是输出的channel数量
        第三个参数（kernel_size）是卷积核大小
        第四个参数（stride）是步长，默认为1
        第五个参数（padding）是填充大小，默认为0
        """
        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(classNames))

    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=2, bias=True)
)

训练模型

设置超参数

loss_fn = nn.CrossEntropyLoss()
learn_rate = 1e-4
opt = torch.optim.SGD(model.parameters(), lr=learn_rate)

编写训练函数

# 训练循环
def train(dataloader, model, loss_fn, optimizer):
    size = len(dataloader.dataset)  # 训练集的大小，一共1713张图片
    num_batches = len(dataloader)   # 批次数目，1713/32=53.53≈53，drop_last=true
    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

1. optimizer.zero_grad()
   
   函数会遍历模型的所有参数，通过内置方法截断反向传播的梯度流，再将每个参数的梯度值设为0，即上一次的梯度记录被清空。

2. loss.backward()
   
   PyTorch的反向传播(即tensor.backward())是通过autograd包来实现的，autograd包会根据tensor进行过的数学运算来自动计算其对应的梯度。
   
   
   具体来说，torch.tensor是autograd包的基础类，如果你设置tensor的requires_grads为True，就会开始跟踪这个tensor上面的所有运算，如果你做完运算后使用tensor.backward()，所有的梯度就会自动运算，tensor的梯度将会累加到它的.grad属性里面去。
   
   
   更具体地说，损失函数loss是由模型的所有权重w经过一系列运算得到的，若某个w的requires_grads为True，则w的所有上层参数（后面层的权重w）的.grad_fn属性中就保存了对应的运算，然后在使用loss.backward()后，会一层层的反向传播计算每个w的梯度值，并保存到该w的.grad属性中。
   
   
   如果没有进行tensor.backward()的话，梯度值将会是None，因此loss.backward()要写在optimizer.step()之前。

3. optimizer.step()
   
   step()函数的作用是执行一次优化步骤，通过梯度下降法来更新参数的值。因为梯度下降是基于梯度的，所以在执行optimizer.step()函数前应先执行loss.backward()函数来计算梯度。
   
   注意：optimizer只负责通过梯度下降进行优化，而不负责产生梯度，梯度是tensor.backward()方法产生的。

编写测试函数

def test (dataloader, model, loss_fn):
    size = len(dataloader.dataset)  # 测试集的大小，一共429张图片
    num_batches = len(dataloader)    # 批次数目（429/32=13.406..，drop_last=true）
    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

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_data, model, loss_fn, opt)

    model.eval()
    epoch_test_acc, epoch_test_loss = test(test_data, 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:94.2%, Train_loss:0.158, Test_acc:86.0%，Test_loss:0.463
Epoch: 2, Train_acc:94.6%, Train_loss:0.132, Test_acc:86.2%，Test_loss:0.521
Epoch: 3, Train_acc:94.8%, Train_loss:0.130, Test_acc:86.5%，Test_loss:0.500
Epoch: 4, Train_acc:95.0%, Train_loss:0.128, Test_acc:86.5%，Test_loss:0.508
Epoch: 5, Train_acc:95.4%, Train_loss:0.104, Test_acc:87.2%，Test_loss:0.450
Epoch: 6, Train_acc:95.1%, Train_loss:0.109, Test_acc:86.9%，Test_loss:0.470
Epoch: 7, Train_acc:95.0%, Train_loss:0.101, Test_acc:86.5%，Test_loss:0.483
Epoch: 8, Train_acc:95.3%, Train_loss:0.112, Test_acc:86.7%，Test_loss:0.491
Epoch: 9, Train_acc:95.4%, Train_loss:0.109, Test_acc:86.5%，Test_loss:0.482
Epoch:10, Train_acc:95.7%, Train_loss:0.100, Test_acc:87.4%，Test_loss:0.489
Epoch:11, Train_acc:95.9%, Train_loss:0.086, Test_acc:86.2%，Test_loss:0.500
Epoch:12, Train_acc:95.1%, Train_loss:0.108, Test_acc:86.9%，Test_loss:0.454
Epoch:13, Train_acc:95.9%, Train_loss:0.092, Test_acc:86.7%，Test_loss:0.486
Epoch:14, Train_acc:95.8%, Train_loss:0.094, Test_acc:85.5%，Test_loss:0.487
Epoch:15, Train_acc:95.7%, Train_loss:0.097, Test_acc:85.8%，Test_loss:0.476
Epoch:16, Train_acc:95.8%, Train_loss:0.087, Test_acc:85.8%，Test_loss:0.479
Epoch:17, Train_acc:96.1%, Train_loss:0.089, Test_acc:85.8%，Test_loss:0.483
Epoch:18, Train_acc:96.2%, Train_loss:0.085, Test_acc:86.2%，Test_loss:0.486
Epoch:19, Train_acc:96.0%, Train_loss:0.094, Test_acc:85.8%，Test_loss:0.487
Epoch:20, Train_acc:95.9%, Train_loss:0.092, Test_acc:86.0%，Test_loss:0.478
Done

结果可视化

loss与accuracy

import matplotlib.pyplot as plt
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()

指定图片进行预测

⭐**torch.squeeze()**详解

函数原型

torch.squeeze(input, dim = None,*, out = None)
将维度为1的去除

关键参数

• input(Tensor):输入Tensor
• dim(int,optional):如果给定，输入将只在这个维度上被压缩

x = torch.zeros(2,1,2,1,2)
print(x.size())
y=torch.squeeze(x)
print(y.size())
y1=torch.squeeze(x,0)
print(y1.size())
y2 = torch.squeeze(x, 1)
print(y2.size())

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

⭐**torch.unsqueeze()**

对数据维度进行扩充。给指定位置加上维度为1的维度

函数原型

torch.unsqueeze(input,dim)

关键参数

• input(Tensor)
• dim(int):插入单例维度的索引

#例：
x = torch.tensor([1,2,3,4])
x.unsqueeze(0)

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

torch.unsqueeze(x,1)
x.unsqueeze(1)

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

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='./Others/NM60_01_03.jpg',
                  model=model,
                  transform=train_transforms,
                  classes=classes)

预测结果是：Others

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当中
model.load_state_dict(torch.load(PATH, map_location=device))

<All keys matched successfully>

总结

构建识别猴痘病的CNN模型，并达到test acc为86%-87%

后续计划继续调整模型结构及参数使得test acc大于88%