RuntimeError: Given groups=1, weight of size [64, 3, 3, 3], expected input[64, 1, 32, 32] to have 3 channels, but got 1 channels instead

Question

I am working on SVHN dataset, and I got this error, only during training phase. During the instantiation of the model, it works.

RuntimeError: Given groups=1, weight of size [64, 3, 3, 3], expected input[64, 1, 32, 32] to have 3 channels, but got 1 channels instead

To be sure of having 3 channels, I wrote that Grayscale(3) transformation.

# Compose transformations
data_transform = transforms.Compose([
  transforms.Resize((32,32)),
  transforms.Grayscale(num_output_channels=3),
  transforms.RandomHorizontalFlip(),
  transforms.ToTensor(),
])

# Compose transformations
test_transform = transforms.Compose([
  transforms.Resize((32,32)),
  transforms.Grayscale(num_output_channels=3),
  transforms.ToTensor(),
])

# Load MNIST dataset with transforms
train_set_svhn = torchvision.datasets.SVHN(root=base_dir, split='train', download=True, transform=data_transform, target_transform=None)
test_set_svhn = torchvision.datasets.SVHN(root=base_dir, split='test', download=True, transform=test_transform)

class VGG16(nn.Module):

    def __init__(self, num_classes):
        super(VGG16, self).__init__()

        # calculate same padding:
        # (w - k + 2*p)/s + 1 = o
        # => p = (s(o-1) - w + k)/2

        self.block_1 = nn.Sequential(
            nn.Conv2d(in_channels=3,
                      out_channels=64,
                      kernel_size=(3, 3),
                      stride=(1, 1),
                      # (1(32-1)- 32 + 3)/2 = 1
                      padding=1),
            nn.BatchNorm2d(64),
            nn.ReLU(),
            nn.Conv2d(in_channels=64,
                      out_channels=64,
                      kernel_size=(3, 3),
                      stride=(1, 1),
                      padding=1),
            nn.BatchNorm2d(64),
            nn.ReLU(),
            nn.MaxPool2d(kernel_size=(2, 2),
                         stride=(2, 2))
        )

        self.block_2 = nn.Sequential(
            nn.Conv2d(in_channels=64,
                      out_channels=128,
                      kernel_size=(3, 3),
                      stride=(1, 1),
                      padding=1),
            nn.BatchNorm2d(128),
            nn.ReLU(),
            nn.Conv2d(in_channels=128,
                      out_channels=128,
                      kernel_size=(3, 3),
                      stride=(1, 1),
                      padding=1),
            nn.BatchNorm2d(128),
            nn.ReLU(),
            nn.MaxPool2d(kernel_size=(2, 2),
                         stride=(2, 2))
        )
        
        self.block_3 = nn.Sequential(
            nn.Conv2d(in_channels=128,
                      out_channels=256,
                      kernel_size=(3, 3),
                      stride=(1, 1),
                      padding=1),
            nn.BatchNorm2d(256),
            nn.ReLU(),
            nn.Conv2d(in_channels=256,
                      out_channels=256,
                      kernel_size=(3, 3),
                      stride=(1, 1),
                      padding=1),
            nn.BatchNorm2d(256),
            nn.ReLU(),
            nn.Conv2d(in_channels=256,
                      out_channels=256,
                      kernel_size=(3, 3),
                      stride=(1, 1),
                      padding=1),
            nn.BatchNorm2d(256),
            nn.ReLU(),
            nn.MaxPool2d(kernel_size=(2, 2),
                         stride=(2, 2))
        )
           

        self.classifier = nn.Sequential(
            nn.Linear(4096, 4096),
            nn.ReLU(True),
            nn.Dropout(p=0.25),
            nn.Linear(4096, 4096),
            nn.ReLU(True),
            nn.Dropout(p=0.25),
            nn.Linear(4096, num_classes),
        )

        for m in self.modules():
            if isinstance(m, torch.nn.Conv2d) or isinstance(m, torch.nn.Linear):
                nn.init.kaiming_uniform_(m.weight, mode='fan_in', nonlinearity='leaky_relu')
#                 nn.init.xavier_normal_(m.weight)
                if m.bias is not None:
                    m.bias.detach().zero_()

        # self.avgpool = nn.AdaptiveAvgPool2d((7, 7))

    def forward(self, x):

        x = self.block_1(x)
        x = self.block_2(x)
        x = self.block_3(x)
        # x = self.avgpool(x)
        x = x.view(x.size(0), -1)
        logits = self.classifier(x)
        probas = F.softmax(logits, dim=1)
        # probas = nn.Softmax(logits)
        return probas
        # return logits

I have no idea where that 1 comes from

Moreover, this is the shape of the output of the model without classifier (fully connected layers):

output = model1(test_x)
output.shape

torch.Size([1, 256, 4, 4])

And indeed I will pass 256x4x4 as input to the first FC.

Edit:

Answer 1

I encountered similar problems, but I did not use the MNIST dataset. I confirmed that I used the input file of (256256,3), and the following is my script file.

RuntimeError: Given groups=1, weight of size [64, 3, 4, 4], expected input[2, 2, 64, 64] to have 3 channels, but got 2 channels instead[enter image description here][1]