Is there any resource for conv2dtranpose for videos?

I am trying to implement Videogan and trying to figure out how to get upsampling for 3 dimension.
self.conv1b = nn.ConvTranspose2d(zdim, 512, [4,4], [1,1]) I found this code but I am not sure what is it doing. are [4,4] the kernel size and [1,1] padding ?

No, the 4s are the kernel size and the the 1s are the stride.
You can check the docs for the argument names.