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: 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

This is the portion of code I think where the problem is.

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

This is the code I am working on

Answer 1

nn.Conv3d expects the input to have size [batch_size, channels, depth, height, width]. The first convolution expects 3 channels, but with your input having size [100, 16, 16, 16, 3], that would be 16 channels.

Assuming that your data is given as [batch_size, depth, height, width, channels], you need to swap the dimensions around, which can be done with 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)