Open
Description
The Dilated layer doesn't take into account k. This can lead to taking more neighbours than intended.
t = torch.tensor([
[0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1],
[0, 1, 2, 3, 4, 5, 0, 1, 2, 3, 4, 5]
])
res = Dilated(k=2, dilation=2)(t)
print(res) # here 3 neighbours are taken even though the constructor specified 2.
# tensor([[0, 0, 0, 1, 1, 1],
# [0, 2, 4, 0, 2, 4]])Metadata
Metadata
Assignees
Labels
No labels