60天python训练计划----day43

复习日

作业：

kaggle找到一个图像数据集，用cnn网络进行训练并且用grad-cam做可视化

进阶：并拆分成多个文件

我选择图像分类，该数据集分为六类，包含建筑、森林、冰川、山脉、海洋和街道。

import torch
import torch.nn as nn
import torch.nn.functional as F
import torchvision.transforms as transforms
from torchvision.datasets import ImageFolder
import numpy as np
import matplotlib.pyplot as plt
import os
from PIL import Image
import warnings
 
# 设置随机种子
torch.manual_seed(42)
np.random.seed(42)
 
# 数据预处理：转换为张量 + 归一化 + Resize到64x64
transform = transforms.Compose([
    transforms.Resize((64, 64)),
    transforms.ToTensor(),
    transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))
])
 
# 加载数据集
trainset = ImageFolder(root='./intel_image_classification/seg_train/seg_train', transform=transform)
testset = ImageFolder(root='./intel_image_classification/seg_test/seg_test', transform=transform)
classes = trainset.classes
 
 
# 构建CNN模型
class SimpleCNN(nn.Module):
    def __init__(self):
        super(SimpleCNN, self).__init__()
        self.conv1 = nn.Conv2d(3, 32, kernel_size=3, padding=1)
        self.conv2 = nn.Conv2d(32, 64, kernel_size=3, padding=1)
        self.conv3 = nn.Conv2d(64, 128, kernel_size=3, padding=1)
        self.pool = nn.MaxPool2d(2, 2)
        self.fc1 = nn.Linear(128 * 8 * 8, 512)
        self.fc2 = nn.Linear(512, 6)
 
    def forward(self, x):
        x = self.pool(F.relu(self.conv1(x)))  # 64 -> 32
        x = self.pool(F.relu(self.conv2(x)))  # 32 -> 16
        x = self.pool(F.relu(self.conv3(x)))  # 16 -> 8
        x = x.view(-1, 128 * 8 * 8)
        x = F.relu(self.fc1(x))
        x = self.fc2(x)
        return x
 
 
# 创建模型及设备
def create_model():
    model = SimpleCNN()
    print("模型已创建")
    device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
    return model.to(device), device
 
 
# 模型训练函数
def train_model(model, device, trainset, epochs=2):
    trainloader = torch.utils.data.DataLoader(trainset, batch_size=64, shuffle=True, num_workers=2)
    criterion = nn.CrossEntropyLoss()
    optimizer = torch.optim.Adam(model.parameters(), lr=0.001)
 
    model.train()
    for epoch in range(epochs):
        running_loss = 0.0
        for i, (inputs, labels) in enumerate(trainloader):
            inputs, labels = inputs.to(device), labels.to(device)
            optimizer.zero_grad()
            outputs = model(inputs)
            loss = criterion(outputs, labels)
            loss.backward()
            optimizer.step()
            running_loss += loss.item()
 
            if i % 100 == 99:
                print(f'[{epoch + 1}, {i + 1}] 损失: {running_loss / 100:.3f}')
                running_loss = 0.0
    print("训练完成")
 
 
# Grad-CAM 实现类
class GradCAM:
    def __init__(self, model, target_layer):
        self.model = model
        self.target_layer = target_layer
        self.gradients = None
        self.activations = None
        self.register_hooks()
 
    def register_hooks(self):
        def forward_hook(module, input, output):
            self.activations = output.detach()
 
        def backward_hook(module, grad_input, grad_output):
            self.gradients = grad_output[0].detach()
 
        self.target_layer.register_forward_hook(forward_hook)
        self.target_layer.register_backward_hook(backward_hook)
 
    def generate_cam(self, input_image, target_class=None):
        model_output = self.model(input_image)
        if target_class is None:
            target_class = torch.argmax(model_output, dim=1).item()
 
        self.model.zero_grad()
        one_hot = torch.zeros_like(model_output)
        one_hot[0, target_class] = 1
        model_output.backward(gradient=one_hot)
 
        gradients = self.gradients
        activations = self.activations
        weights = torch.mean(gradients, dim=(2, 3), keepdim=True)
        cam = torch.sum(weights * activations, dim=1, keepdim=True)
        cam = F.relu(cam)
        cam = F.interpolate(cam, size=(64, 64), mode='bilinear', align_corners=False)
        cam = cam - cam.min()
        cam = cam / cam.max() if cam.max() > 0 else cam
 
        return cam.cpu().squeeze().numpy(), target_class
 
 
# 图像张量转换为numpy数组，用于可视化
def tensor_to_np(tensor):
    img = tensor.cpu().numpy().transpose(1, 2, 0)
    mean = np.array([0.5, 0.5, 0.5])
    std = np.array([0.5, 0.5, 0.5])
    img = std * img + mean
    img = np.clip(img, 0, 1)
    return img
 
 
# 主函数入口（避免Windows多次导入问题）
if __name__ == "__main__":
    warnings.filterwarnings("ignore")
    plt.rcParams["font.family"] = ["SimHei"]
    plt.rcParams['axes.unicode_minus'] = False
 
    model, device = create_model()
 
    # 是否训练模型
    train_model(model, device, trainset, epochs=20)
    torch.save(model.state_dict(), 'intel_cnn.pth')
    model.eval()
 
    # Grad-CAM 可视化
    idx = 300
    image, label = testset[idx]
    print(f"选择的图像类别: {classes[label]}")
 
    input_tensor = image.unsqueeze(0).to(device)
    grad_cam = GradCAM(model, model.conv3)
    heatmap, pred_class = grad_cam.generate_cam(input_tensor)
 
    # 可视化图像 + CAM
    plt.figure(figsize=(12, 4))
 
    plt.subplot(1, 3, 1)
    plt.imshow(tensor_to_np(image))
    plt.title(f"原始图像: {classes[label]}")
    plt.axis('off')
 
    plt.subplot(1, 3, 2)
    plt.imshow(heatmap, cmap='jet')
    plt.title(f"Grad-CAM热力图: {classes[pred_class]}")
    plt.axis('off')
 
    plt.subplot(1, 3, 3)
    img = tensor_to_np(image)
    heatmap_resized = np.uint8(255 * heatmap)
    heatmap_colored = plt.cm.jet(heatmap_resized)[:, :, :3]
    superimposed_img = heatmap_colored * 0.4 + img * 0.6
    plt.imshow(superimposed_img)
    plt.title("叠加热力图")
    plt.axis('off')
 
    plt.tight_layout()
    plt.savefig('grad_cam_result.png')
    plt.show()

 

@浙大疏锦行