vario function scaled by 2

[FLBuchner]

Hi,

I am trying out your package and while testing I found that the output of the vario function is scaled by factor 2. I think it would be faster to only evaluate from i = 1; i<d; i++ and j = i; j<d; j++ and scale by 2 afterwards, since the weight matrix should be symmetric.

https://github.com/sallen12/MultivCalibration/blob/91371c1f00f4e2632759fbf84b9d83b36117ab00/src/variogram_func.cpp#L39C1-L50C2

Best

[sallen12]

Hi, thanks a lot for pointing this out. I think this faster implementation is what I had in mind, I just didn't change the starting point for j. The ordering of the pre-ranks is insensitive to scaling by a constant, so this won't affect the subsequent results, but I've now corrected this using your suggestion. Thanks!