neural-network - 初学者 PyTorch - 运行时错误 : shape '[16, 400]' is invalid for input of size 9600

Question

我正在尝试构建 CNN，但出现此错误:

---> 52         x = x.view(x.size(0), 5 * 5 * 16)
RuntimeError: shape '[16, 400]' is invalid for input of size 9600

我不清楚“x.view”行的输入应该是什么。另外，我真的不明白我应该在我的代码中使用这个“x.view”函数多少次。是不是只有一次，在 3 个卷积层和 2 个线性层之后？还是5次，每层一次？

这是我的代码:

美国有线电视新闻网

import torch.nn.functional as F

# Convolutional neural network
class ConvNet(nn.Module):

    def __init__(self, num_classes=10):
        super(ConvNet, self).__init__()

        self.conv1 = nn.Conv2d(
            in_channels=3, 
            out_channels=16, 
            kernel_size=3)

        self.conv2 = nn.Conv2d(
            in_channels=16, 
            out_channels=24, 
            kernel_size=4)

        self.conv3 = nn.Conv2d(
            in_channels=24, 
            out_channels=32, 
            kernel_size=4)

        self.dropout = nn.Dropout2d(p=0.3)

        self.pool = nn.MaxPool2d(2)

        self.fc1 = nn.Linear(16 * 5 * 5, 120)
        self.fc2 = nn.Linear(512, 10)

        self.final = nn.Softmax(dim=1)

    def forward(self, x):

        print('shape 0 ' + str(x.shape))

        x = F.max_pool2d(F.relu(self.conv1(x)), 2)  
        x = self.dropout(x)

        print('shape 1 ' + str(x.shape))

        x = F.max_pool2d(F.relu(self.conv2(x)), 2)  
        x = self.dropout(x)

        print('shape 2 ' + str(x.shape))

        # x = F.max_pool2d(F.relu(self.conv3(x)), 2)  
        # x = self.dropout(x)

        x = F.interpolate(x, size=(5, 5))  
        x = x.view(x.size(0), 5 * 5 * 16)

        x = self.fc1(x) 

        return x

net = ConvNet()

有人可以帮助我理解问题吗？

'x.shape' 的输出是:

形状 0 torch.Size([16, 3, 256, 256])

形状 1 torch.Size([16, 16, 127, 127])

形状 2 torch.Size([16, 24, 62, 62])

谢谢

Answer 1

这意味着 channel 和空间维度的乘积不是 5*5*16 .要展平张量，请替换 x = x.view(x.size(0), 5 * 5 * 16)和:

x = x.view(x.size(0), -1)

和 self.fc1 = nn.Linear(600, 120)和:

self.fc1 = nn.Linear(600, 120)