How to do meshfree 3D krige interpolation?

[Reybor]

I have about 100,000 meshfree points in 3D coordinate. The examples just show the codes to generate a structured mesh filed to interpolation. So I wanna know how to interpolate in meshfree format or unstructured mesh format. And what tools are need to get unstructured mesh.
In the paper "Influence of Subsurface Critical Zone Structure onHydrological Partitioning in Mountainous HeadwaterCatchments", Chen said One advantage of using GSTools is that the topography can be kept during the Kriging. It's what I need

[MuellerSeb]

Hey there, to have an unstructured target grid, just call the unstructured method or pass mesh_type="unstructured" to the kriging call (actually not needed, since this is the default mesh type):

import numpy as np
import gstools as gs

# conditioning data
cond_x = [0.3, 1.9, 1.1, 3.3, 4.7]
cond_y = [1.2, 0.6, 3.2, 4.4, 3.8]
cond_val = [0.47, 0.56, 0.74, 1.47, 1.74]
cond_pos = [cond_x, cond_y]

# unstructued points of target grid in (0, 6) x (0, 6)
x = np.random.RandomState(19970221).rand(1000) * 6.
y = np.random.RandomState(20011012).rand(1000) * 6.

model = gs.Gaussian(
    dim=2, len_scale=1, anis=0.2, angles=-0.5, var=0.5, nugget=0.1
)
krig = gs.krige.Ordinary(model, cond_pos=cond_pos, cond_val=cond_val)
field, var = krig.unstructured(pos=(x, y))  # or krig(pos=(x, y), mesh_type="unstructured")

Hope this answers your question!

Cheers,
Sebastian