computation time of DirectionalVariogram

[mserouar]

Hello there !

I was wondering if it was possible to make this function faster : skg.DirectionalVariogram(list(zip(data.x, data.y)), data.m, azimuth=90, maxlag=100, n_lags=5).
Knowing that length of list(zip(data.x, data.y)) is only of 1500 and takes ~4-5s.

Thanks !

[mmaelicke]

Hi there,

I can confirm that the directional variogram is really slow. It used to be faster. I am not sure when the performance started dropping; I will have to investigate some test-logos and check when the last changes have been made to that class.
Do you just wish to have it faster, or do you also have an idea, how that could be achieved?

I have to admit, that I don't have much time right now to make substantial progress in this package.

[mserouar]

I just wanted to know if it was a known issue, first of all, if there was a reason for that computational delay faced to the 'simple' Variogram class.
I didn't expect any action from you, simply curious. But if I also have some time, I may have a look.
Thanks a lot !

[mmaelicke]

This is a new issue for me, and I feel like this should be faster, so I think I might consider this a bug.

Currently, I just have no idea if that is actually the case and if so, how to solve that. So, I will eventually look into it and until then, any hint is appreciated ^^