pytorch-error：output with shape [1, 28, 28] doesn't match the broadcast shape [3, 28, 28]

缺乏、安全感 2023-06-29 12:59 48阅读 0赞

使用《深度学习之pytorch实战计算机视觉》进行学习时，在6.4实战手写数字识别时遇到了该问题，报错情况如下：

Traceback (most recent call last):
      File "G:/pytorch/torch_数字识别器.py", line 19, in <module>
        images, lables = next(iter(data_loder_train))
      File "D:\ProgramData\Anaconda3\lib\site-packages\torch\utils\data\dataloader.py", line 346, in __next__
        data = self._dataset_fetcher.fetch(index)  # may raise StopIteration
      File "D:\ProgramData\Anaconda3\lib\site-packages\torch\utils\data\_utils\fetch.py", line 44, in fetch
        data = [self.dataset[idx] for idx in possibly_batched_index]
      File "D:\ProgramData\Anaconda3\lib\site-packages\torch\utils\data\_utils\fetch.py", line 44, in <listcomp>
        data = [self.dataset[idx] for idx in possibly_batched_index]
      File "D:\ProgramData\Anaconda3\lib\site-packages\torchvision\datasets\mnist.py", line 95, in __getitem__
        img = self.transform(img)
      File "D:\ProgramData\Anaconda3\lib\site-packages\torchvision\transforms\transforms.py", line 61, in __call__
        img = t(img)
      File "D:\ProgramData\Anaconda3\lib\site-packages\torchvision\transforms\transforms.py", line 166, in __call__
        return F.normalize(tensor, self.mean, self.std, self.inplace)
      File "D:\ProgramData\Anaconda3\lib\site-packages\torchvision\transforms\functional.py", line 217, in normalize
        tensor.sub_(mean[:, None, None]).div_(std[:, None, None])
    RuntimeError: output with shape [1, 28, 28] doesn't match the broadcast shape [3, 28, 28]
    
    Process finished with exit code 1

**题目的报错信息显示，输出图像与输入图像在维度上不匹配。因为MNIST是灰度图像，但是书上给出的均值方差有三个，改为一个即可**  
修改后的程序如下：

import torchvision
    import torch.utils.data as da
    import matplotlib.pyplot as plt
    from torchvision import datasets, transforms
    transform = transforms.Compose([
        transforms.ToTensor(),
        transforms.Normalize(mean=[0.1307], std=[0.3081])#修改位置
    ])
    data_train = datasets.MNIST(root="G:/pytorch/data",transform=transform, train=True, download=True)
    data_test = datasets.MNIST(root="G:/pytorch/data", transform=transform, train=False)
    #数据加载
    data_loder_train = da.DataLoader(dataset=data_train, batch_size=64, shuffle = True)
    data_loder_test = da.DataLoader(dataset=data_test, batch_size = 64, shuffle = True)
    images, lables = next(iter(data_loder_train))
    img = torchvision.utils.make_grid(images)
    img = img.numpy().transpose(1,2,0)
    std = [0.5, 0.5, 0.5]
    mean = [0.5, 0.5, 0.5]
    img = img * std + mean
    for i in range(64):
        print(lables[i], end=" ")
        i += 1
        if i%8 is 0:
            print(end='\n')
    plt.imshow(img)
    plt.show()

运行结果如下：

tensor(0) tensor(5) tensor(7) tensor(9) tensor(3) tensor(2) tensor(0) tensor(7) 
    tensor(0) tensor(0) tensor(8) tensor(3) tensor(8) tensor(7) tensor(9) tensor(2) 
    tensor(2) tensor(7) tensor(7) tensor(0) tensor(7) tensor(7) tensor(7) tensor(2) 
    tensor(3) tensor(9) tensor(0) tensor(7) tensor(7) tensor(3) tensor(3) tensor(7) 
    tensor(2) tensor(8) tensor(7) tensor(8) tensor(0) tensor(8) tensor(7) tensor(1) 
    tensor(0) tensor(3) tensor(6) tensor(5) tensor(8) tensor(5) tensor(2) tensor(4) 
    tensor(8) tensor(6) tensor(9) tensor(4) tensor(1) tensor(7) tensor(5) tensor(2) 
    tensor(7) tensor(4) tensor(8) tensor(2) tensor(7) tensor(1) tensor(3) tensor(1)

![在这里插入图片描述][watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3dlaXhpbl80NTg4NTIzMg_size_16_color_FFFFFF_t_70]

[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3dlaXhpbl80NTg4NTIzMg_size_16_color_FFFFFF_t_70]: https://img-blog.csdnimg.cn/20200112231839415.png?x-oss-process=image/watermark,type_ZmFuZ3poZW5naGVpdGk,shadow_10,text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3dlaXhpbl80NTg4NTIzMg==,size_16,color_FFFFFF,t_70