Downsampling odenet for resnet

following is my code snippet I want to classify images of CIFAR10 dataset with a ODENET can someone help with the downsampling I am getting the following error
RuntimeError: Given groups=1, weight of size [64, 1, 3, 3], expected input[128, 3, 110, 110] to have 1 channels, but got 3 channels instead

if args.downsampling_method == ‘conv’:
downsampling_layers = [
nn.Conv2d(1, 64, 3, 1),
norm(64),
nn.ReLU(inplace=True),
nn.Conv2d(64, 64, 4, 2, 1),
norm(64),
nn.ReLU(inplace=True),
nn.Conv2d(64, 64, 4, 2, 1),
]
elif args.downsampling_method == ‘res’:
downsampling_layers = [
nn.Conv2d(1, 64, 3, 1),
ResBlock(64, 64, stride=2, downsample=conv1x1(64, 64, 2)),
ResBlock(64, 64, stride=2, downsample=conv1x1(64, 64, 2)),
]

Based on the error message the input to the conv layer has 3 channels (I guess it could be the image tensor) while the conv layer expects an input with a single input channel. You could change in_channels to 3 in: nn.Conv2d(3, 64, 3, 1) to fix this.

Thanks a ton for the reply sir