Multiply window size by input channels dimension

[StillSpeedo]

Hi, Mathias!
Please, look at the authors' implementation in PyTorch:
self.slide_winsize = self.weight_maskUpdater.shape[1] * self.weight_maskUpdater.shape[2] * self.weight_maskUpdater.shape[3]
and compare with your current
self.window_size = self.kernel_size[0] * self.kernel_size[1]
According to the paper, the scaling factor for all valid (unmasked) pixels in window is 1, but in your case it is 1/(input channels dimension).
I don't think it's a big problem, this additional multiplier is constant for each layer and can be learned by the network, or even batch norm negates it :)
But if you make proposed changes to the code, old trained weights will be invalid.