I have a dataset of 1000 observed samples of 6 features that form the X and one target variable that forms the Y.

I am using kriging or Gaussian Process Regressor to train my models. I would like to use anisotropic Gaussian and anisotropic exponential correlation functions as kernels. Please see equation 14 and 15 in the attached equation pic for reference.

How do I define these functions in python such that they are compatible with SKlearns's GPRkernel function?


Browse other questions tagged or ask your own question.