Stroia Dacian

I'm not even sure taking `[::d]` is the right way to go. The following example ```python t = torch.tensor( [ [0, 0, 0, 0, 0, 1, 1, 1, 1, 1,...

A possible solution would be (using einops): ```python from einops import rearrange t = torch.tensor( [ [0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 2, 2, 2, 2,...