MNIST、torchvision中的输出和广播形状不匹配

import torchvision.transforms as transforms
from torchvision.datasets import MNIST

# 定义预处理
transform = transforms.Compose([
    transforms.ToTensor(),  # 将图像转换为PyTorch张量
    transforms.Normalize((0.1307,), (0.3081,))  # 标准化图像
])

# 加载MNIST数据集
train_dataset = MNIST(root='./data', train=True, download=True, transform=transform)

import torch.nn as nn
import torch.nn.functional as F

class SimpleCNN(nn.Module):
    def __init__(self):
        super(SimpleCNN, self).__init__()
        self.conv1 = nn.Conv2d(1, 32, kernel_size=3)
        self.fc1 = nn.Linear(32 * 26 * 26, 10)

    def forward(self, x):
        x = F.relu(self.conv1(x))
        x = x.view(x.size(0), -1)  # 展平张量
        x = self.fc1(x)
        return x

model = SimpleCNN()
criterion = nn.CrossEntropyLoss()  # 适用于分类问题的损失函数

# 假设output是模型的输出，target是目标标签
output = model(input_tensor)
loss = criterion(output, target_tensor)  # 这里output的形状应该是(batch_size, num_classes)

# 假设output的形状是(batch_size, num_classes)，而target的形状是(batch_size,)
# 如果需要，可以将target转换为one-hot编码
target_one_hot = F.one_hot(target_tensor, num_classes=10).float()