RuntimeError: Given groups=1, weight of size [32, 3, 16, 16, 16], expected input[100, 16, 16, 16, 3] to have 3 channels, but got 16 channels instead

Question

RuntimeError：给定组数=1，权重大小为 [32, 3, 16, 16, 16]，预期输入 [100, 16, 16, 16, 3] 有 3 个通道，但得到的却是 16 个通道

这是我认为问题所在的代码部分。

def __init__(self):
        super(Lightning_CNNModel, self).__init__()

        self.conv_layer1 = self._conv_layer_set(3, 32)
        self.conv_layer2 = self._conv_layer_set(32, 64)
        self.fc1 = nn.Linear(2**3*64, 128)
        self.fc2 = nn.Linear(128, 10)   # num_classes = 10
        self.relu = nn.LeakyReLU()
        self.batch=nn.BatchNorm1d(128)
        self.drop=nn.Dropout(p=0.15)

    def _conv_layer_set(self, in_c, out_c):
        conv_layer = nn.Sequential(
            nn.Conv3d(in_c, out_c, kernel_size=(3, 3, 3), padding=0),
            nn.LeakyReLU(),
            nn.MaxPool3d((2, 2, 2)),
        )

        return conv_layer



    def forward(self, x):
        out = self.conv_layer1(x)
        out = self.conv_layer2(out)
        out = out.view(out.size(0), -1)
        out = self.fc1(out)
        out = self.relu(out)
        out = self.batch(out)
        out = self.drop(out)
        out = self.fc2(out)

        return out

这是我正在处理的代码

Answer 1

nn.Conv3d 期望输入的大小为 [batch_size、通道、深度、高度、宽度]。第一个卷积需要 3 个通道，但是如果您的输入大小为 [100, 16, 16, 16, 3]，那将是 16 个通道.

假设你的数据是[batch_size, depth, height, width, channels]，你需要交换维度，这是可以做到的torch.Tensor.permute:

# From: [batch_size, depth, height, width, channels]
# To: [batch_size, channels, depth, height, width]
input = input.permute(0, 4, 1, 2, 3)

RuntimeError: Given groups=1, weight of size [32, 3, 16, 16, 16], expected input[100, 16, 16, 16, 3] to have 3 channels, but got 16 channels instead

RuntimeError: Given groups=1, weight of size [32, 3, 16, 16, 16], expected input[100, 16, 16, 16, 3] to have 3 channels, but got 16 channels instead

python

pytorch

conv-neural-network

pytorch-lightning