如何在pytorch图像处理模型中处理包含多个图像的样本？

import torch
from torch.utils.data import Dataset, DataLoader
from torchvision import transforms
from PIL import Image

# 自定义数据集类
class MultiImageDataset(Dataset):
    def __init__(self, image_paths, transform=None):
        self.image_paths = image_paths
        self.transform = transform

    def __len__(self):
        return len(self.image_paths)

    def __getitem__(self, idx):
        # 假设每对图像的路径是连续的
        img1_path = self.image_paths[idx]
        img2_path = self.image_paths[idx + 1]
        
        img1 = Image.open(img1_path)
        img2 = Image.open(img2_path)
        
        if self.transform:
            img1 = self.transform(img1)
            img2 = self.transform(img2)
        
        # 返回一个样本，包含两张图像
        return img1, img2

# 图像变换
transform = transforms.Compose([
    transforms.Resize((256, 256)),
    transforms.ToTensor()
])

# 假设我们有一个图像路径列表
image_paths = ['path/to/image1_1.jpg', 'path/to/image1_2.jpg', 'path/to/image2_1.jpg', 'path/to/image2_2.jpg']

# 创建数据集实例
dataset = MultiImageDataset(image_paths, transform=transform)

# 创建数据加载器
dataloader = DataLoader(dataset, batch_size=2, shuffle=True)

# 遍历数据加载器
for img1_batch, img2_batch in dataloader:
    # 在这里进行模型训练或其他处理
    print(img1_batch.shape, img2_batch.shape)