This Python package is based on the Conditioned Latin Hypercube Sampling (cLHS) method of Minasny & McBratney (2006). It follows some of the code from the R package clhs of Roudier et al.
- It attempts to create a Latin Hypercube sample by selecting only from input data.
- It uses simulated annealing to force the sampling to converge more rapidly.
- It allows for setting a stopping criterion on the objective function described in Minasny & McBratney (2006).
You may reproduce the jupyter notebook example on Binder.
Please check online documentation for more information.