Python 训练营打卡 Day 47

注意力热力图可视化

在day 46代码的基础上，对比不同卷积层热力图可视化的结果

import torch
import torch.nn as nn
import torch.optim as optim
from torchvision import datasets, transforms
from torch.utils.data import DataLoader
import matplotlib.pyplot as plt
import numpy as np# 设置中文字体支持
plt.rcParams["font.family"] = ["SimHei"]
plt.rcParams['axes.unicode_minus'] = False  # 解决负号显示问题# 检查GPU是否可用
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
print(f"使用设备: {device}")# 1. 数据预处理
# 训练集：使用多种数据增强方法提高模型泛化能力
train_transform = transforms.Compose([# 随机裁剪图像，从原图中随机截取32x32大小的区域transforms.RandomCrop(32, padding=4),# 随机水平翻转图像（概率0.5）transforms.RandomHorizontalFlip(),# 随机颜色抖动：亮度、对比度、饱和度和色调随机变化transforms.ColorJitter(brightness=0.2, contrast=0.2, saturation=0.2, hue=0.1),# 随机旋转图像（最大角度15度）transforms.RandomRotation(15),# 将PIL图像或numpy数组转换为张量transforms.ToTensor(),# 标准化处理：每个通道的均值和标准差，使数据分布更合理transforms.Normalize((0.4914, 0.4822, 0.4465), (0.2023, 0.1994, 0.2010))
])# 测试集：仅进行必要的标准化，保持数据原始特性，标准化不损失数据信息，可还原
test_transform = transforms.Compose([transforms.ToTensor(),transforms.Normalize((0.4914, 0.4822, 0.4465), (0.2023, 0.1994, 0.2010))
])# 2. 加载CIFAR-10数据集
train_dataset = datasets.CIFAR10(root='./data',train=True,download=True,transform=train_transform  # 使用增强后的预处理
)test_dataset = datasets.CIFAR10(root='./data',train=False,transform=test_transform  # 测试集不使用增强
)# 3. 创建数据加载器
batch_size = 64
train_loader = DataLoader(train_dataset, batch_size=batch_size, shuffle=True)
test_loader = DataLoader(test_dataset, batch_size=batch_size, shuffle=False)# ===================== 新增：通道注意力模块（SE模块） =====================
class ChannelAttention(nn.Module):"""通道注意力模块(Squeeze-and-Excitation)"""def __init__(self, in_channels, reduction_ratio=16):"""参数:in_channels: 输入特征图的通道数reduction_ratio: 降维比例，用于减少参数量"""super(ChannelAttention, self).__init__()# 全局平均池化 - 将空间维度压缩为1x1，保留通道信息self.avg_pool = nn.AdaptiveAvgPool2d(1)# 全连接层 + 激活函数，用于学习通道间的依赖关系self.fc = nn.Sequential(# 降维：压缩通道数，减少计算量nn.Linear(in_channels, in_channels // reduction_ratio, bias=False),nn.ReLU(inplace=True),# 升维：恢复原始通道数nn.Linear(in_channels // reduction_ratio, in_channels, bias=False),# Sigmoid将输出值归一化到[0,1]，表示通道重要性权重nn.Sigmoid())def forward(self, x):"""参数:x: 输入特征图，形状为 [batch_size, channels, height, width]返回:加权后的特征图，形状不变"""batch_size, channels, height, width = x.size()# 1. 全局平均池化：[batch_size, channels, height, width] → [batch_size, channels, 1, 1]avg_pool_output = self.avg_pool(x)# 2. 展平为一维向量：[batch_size, channels, 1, 1] → [batch_size, channels]avg_pool_output = avg_pool_output.view(batch_size, channels)# 3. 通过全连接层学习通道权重：[batch_size, channels] → [batch_size, channels]channel_weights = self.fc(avg_pool_output)# 4. 重塑为二维张量：[batch_size, channels] → [batch_size, channels, 1, 1]channel_weights = channel_weights.view(batch_size, channels, 1, 1)# 5. 将权重应用到原始特征图上（逐通道相乘）return x * channel_weights  # 输出形状：[batch_size, channels, height, width]# 4. 模型的定义
class CNN(nn.Module):def __init__(self):super(CNN, self).__init__()  # ---------------------- 第一个卷积块 ----------------------self.conv1 = nn.Conv2d(3, 32, 3, padding=1)self.bn1 = nn.BatchNorm2d(32)self.relu1 = nn.ReLU()# 新增：插入通道注意力模块（SE模块）self.ca1 = ChannelAttention(in_channels=32, reduction_ratio=16)  self.pool1 = nn.MaxPool2d(2, 2)  # ---------------------- 第二个卷积块 ----------------------self.conv2 = nn.Conv2d(32, 64, 3, padding=1)self.bn2 = nn.BatchNorm2d(64)self.relu2 = nn.ReLU()# 新增：插入通道注意力模块（SE模块）self.ca2 = ChannelAttention(in_channels=64, reduction_ratio=16)  self.pool2 = nn.MaxPool2d(2)  # ---------------------- 第三个卷积块 ----------------------self.conv3 = nn.Conv2d(64, 128, 3, padding=1)self.bn3 = nn.BatchNorm2d(128)self.relu3 = nn.ReLU()# 新增：插入通道注意力模块（SE模块）self.ca3 = ChannelAttention(in_channels=128, reduction_ratio=16)  self.pool3 = nn.MaxPool2d(2)  # ---------------------- 全连接层（分类器） ----------------------self.fc1 = nn.Linear(128 * 4 * 4, 512)self.dropout = nn.Dropout(p=0.5)self.fc2 = nn.Linear(512, 10)def forward(self, x):# ---------- 卷积块1处理 ----------x = self.conv1(x)       x = self.bn1(x)         x = self.relu1(x)       x = self.ca1(x)  # 应用通道注意力x = self.pool1(x)       # ---------- 卷积块2处理 ----------x = self.conv2(x)       x = self.bn2(x)         x = self.relu2(x)       x = self.ca2(x)  # 应用通道注意力x = self.pool2(x)       # ---------- 卷积块3处理 ----------x = self.conv3(x)       x = self.bn3(x)         x = self.relu3(x)       x = self.ca3(x)  # 应用通道注意力x = self.pool3(x)       # ---------- 展平与全连接层 ----------x = x.view(-1, 128 * 4 * 4)  x = self.fc1(x)           x = self.relu3(x)       x = self.dropout(x)       x = self.fc2(x)           return x  # 重新初始化模型，包含通道注意力模块
model = CNN()
model = model.to(device)  # 将模型移至GPU（如果可用）criterion = nn.CrossEntropyLoss()  # 交叉熵损失函数
optimizer = optim.Adam(model.parameters(), lr=0.001)  # Adam优化器# 引入学习率调度器，在训练过程中动态调整学习率--训练初期使用较大的 LR 快速降低损失，训练后期使用较小的 LR 更精细地逼近全局最优解。
# 在每个 epoch 结束后，需要手动调用调度器来更新学习率，可以在训练过程中调用 scheduler.step()
scheduler = optim.lr_scheduler.ReduceLROnPlateau(optimizer,        # 指定要控制的优化器（这里是Adam）mode='min',       # 监测的指标是"最小化"（如损失函数）patience=3,       # 如果连续3个epoch指标没有改善，才降低LRfactor=0.5        # 降低LR的比例（新LR = 旧LR × 0.5）
)# 5. 训练模型（记录每个 iteration 的损失）
def train(model, train_loader, test_loader, criterion, optimizer, scheduler, device, epochs):model.train()  # 设置为训练模式# 记录每个 iteration 的损失all_iter_losses = []  # 存储所有 batch 的损失iter_indices = []     # 存储 iteration 序号# 记录每个 epoch 的准确率和损失train_acc_history = []test_acc_history = []train_loss_history = []test_loss_history = []for epoch in range(epochs):running_loss = 0.0correct = 0total = 0for batch_idx, (data, target) in enumerate(train_loader):data, target = data.to(device), target.to(device)  # 移至GPUoptimizer.zero_grad()  # 梯度清零output = model(data)  # 前向传播loss = criterion(output, target)  # 计算损失loss.backward()  # 反向传播optimizer.step()  # 更新参数# 记录当前 iteration 的损失iter_loss = loss.item()all_iter_losses.append(iter_loss)iter_indices.append(epoch * len(train_loader) + batch_idx + 1)# 统计准确率和损失running_loss += iter_loss_, predicted = output.max(1)total += target.size(0)correct += predicted.eq(target).sum().item()# 每100个批次打印一次训练信息if (batch_idx + 1) % 100 == 0:print(f'Epoch: {epoch+1}/{epochs} | Batch: {batch_idx+1}/{len(train_loader)} 'f'| 单Batch损失: {iter_loss:.4f} | 累计平均损失: {running_loss/(batch_idx+1):.4f}')# 计算当前epoch的平均训练损失和准确率epoch_train_loss = running_loss / len(train_loader)epoch_train_acc = 100. * correct / totaltrain_acc_history.append(epoch_train_acc)train_loss_history.append(epoch_train_loss)# 测试阶段model.eval()  # 设置为评估模式test_loss = 0correct_test = 0total_test = 0with torch.no_grad():for data, target in test_loader:data, target = data.to(device), target.to(device)output = model(data)test_loss += criterion(output, target).item()_, predicted = output.max(1)total_test += target.size(0)correct_test += predicted.eq(target).sum().item()epoch_test_loss = test_loss / len(test_loader)epoch_test_acc = 100. * correct_test / total_testtest_acc_history.append(epoch_test_acc)test_loss_history.append(epoch_test_loss)# 更新学习率调度器scheduler.step(epoch_test_loss)# 早停机制实现if epoch == 0:best_acc = epoch_test_accpatience_counter = 0else:if epoch_test_acc > best_acc:best_acc = epoch_test_accpatience_counter = 0# 保存最佳模型torch.save(model.state_dict(), 'best_model.pth')else:patience_counter += 1if patience_counter >= 5:  # 连续5个epoch测试准确率没有提升则停止训练print(f'Early stopping at epoch {epoch+1}, best accuracy: {best_acc:.2f}%')breakprint(f'Epoch {epoch+1}/{epochs} 完成 | 训练准确率: {epoch_train_acc:.2f}% | 测试准确率: {epoch_test_acc:.2f}%')# 绘制所有 iteration 的损失曲线plot_iter_losses(all_iter_losses, iter_indices)# 绘制每个 epoch 的准确率和损失曲线plot_epoch_metrics(train_acc_history, test_acc_history, train_loss_history, test_loss_history)return epoch_test_acc  # 返回最终测试准确率# 6. 绘制每个 iteration 的损失曲线
def plot_iter_losses(losses, indices):plt.figure(figsize=(10, 4))plt.plot(indices, losses, 'b-', alpha=0.7, label='Iteration Loss')plt.xlabel('Iteration（Batch序号）')plt.ylabel('损失值')plt.title('每个 Iteration 的训练损失')plt.legend()plt.grid(True)plt.tight_layout()plt.show()# 7. 绘制每个 epoch 的准确率和损失曲线
def plot_epoch_metrics(train_acc, test_acc, train_loss, test_loss):epochs = range(1, len(train_acc) + 1)plt.figure(figsize=(12, 4))# 绘制准确率曲线plt.subplot(1, 2, 1)plt.plot(epochs, train_acc, 'b-', label='训练准确率')plt.plot(epochs, test_acc, 'r-', label='测试准确率')plt.xlabel('Epoch')plt.ylabel('准确率 (%)')plt.title('训练和测试准确率')plt.legend()plt.grid(True)# 绘制损失曲线plt.subplot(1, 2, 2)plt.plot(epochs, train_loss, 'b-', label='训练损失')plt.plot(epochs, test_loss, 'r-', label='测试损失')plt.xlabel('Epoch')plt.ylabel('损失值')plt.title('训练和测试损失')plt.legend()plt.grid(True)plt.tight_layout()plt.show()# 8. 执行训练和测试
epochs = 50  # 增加训练轮次为了确保收敛
print("开始使用CNN训练模型...")
final_accuracy = train(model, train_loader, test_loader, criterion, optimizer, scheduler, device, epochs)
print(f"训练完成！最终测试准确率: {final_accuracy:.2f}%")# # 保存模型
# torch.save(model.state_dict(), 'cifar10_cnn_model.pth')
# print("模型已保存为: cifar10_cnn_model.pth")

我们关注的不只是精度的差异，还包含了同精度下训练时长的差异等，在大规模数据集上推理时长、训练时长都非常重要。因为资源是有限的。可视化部分同理，在训练完成后通过钩子函数取出权重or梯度，即可进行特征图的可视化、Grad-CAM可视化、注意力热图可视化

# 可视化空间注意力热力图（显示模型关注的图像区域）
def visualize_attention_map(model, test_loader, device, class_names, num_samples=3):"""可视化模型的注意力热力图，展示模型关注的图像区域"""model.eval()  # 设置为评估模式with torch.no_grad():for i, (images, labels) in enumerate(test_loader):if i >= num_samples:  # 只可视化前几个样本breakimages, labels = images.to(device), labels.to(device)# 创建一个钩子，捕获中间特征图activation_maps = []def hook(module, input, output):activation_maps.append(output.cpu())# 为最后一个卷积层注册钩子（获取特征图）hook_handle = model.conv3.register_forward_hook(hook)# 前向传播，触发钩子outputs = model(images)# 移除钩子hook_handle.remove()# 获取预测结果_, predicted = torch.max(outputs, 1)# 获取原始图像img = images[0].cpu().permute(1, 2, 0).numpy()# 反标准化处理img = img * np.array([0.2023, 0.1994, 0.2010]).reshape(1, 1, 3) + np.array([0.4914, 0.4822, 0.4465]).reshape(1, 1, 3)img = np.clip(img, 0, 1)# 获取激活图（最后一个卷积层的输出）feature_map = activation_maps[0][0].cpu()  # 取第一个样本# 计算通道注意力权重（使用SE模块的全局平均池化）channel_weights = torch.mean(feature_map, dim=(1, 2))  # [C]# 按权重对通道排序sorted_indices = torch.argsort(channel_weights, descending=True)# 创建子图fig, axes = plt.subplots(1, 4, figsize=(16, 4))# 显示原始图像axes[0].imshow(img)axes[0].set_title(f'原始图像\n真实: {class_names[labels[0]]}\n预测: {class_names[predicted[0]]}')axes[0].axis('off')# 显示前3个最活跃通道的热力图for j in range(3):channel_idx = sorted_indices[j]# 获取对应通道的特征图channel_map = feature_map[channel_idx].numpy()# 归一化到[0,1]channel_map = (channel_map - channel_map.min()) / (channel_map.max() - channel_map.min() + 1e-8)# 调整热力图大小以匹配原始图像from scipy.ndimage import zoomheatmap = zoom(channel_map, (32/feature_map.shape[1], 32/feature_map.shape[2]))# 显示热力图axes[j+1].imshow(img)axes[j+1].imshow(heatmap, alpha=0.5, cmap='jet')axes[j+1].set_title(f'注意力热力图 - 通道 {channel_idx}')axes[j+1].axis('off')plt.tight_layout()plt.show()# 调用可视化函数
visualize_attention_map(model, test_loader, device, class_names, num_samples=3)