First, you should know that there is more than one mathematical process called "kriging". The formal approach you will most likely encounter at the top of the list in a Google search is rather esoteric and infrequently used. Take, for instance, the very excellent tool, Tecplot... You can easily perform "kriging" on your 6 variables with Tecplot and may get a useful result, but if you read the fine print in the manual, you will find that the process applied inside Tecplot isn't the one that you might have found on the Web. This is a good thing, because you want useful results. Tecplot is a little pricey. TP2 is free and also performs kriging. If you want regression on up to 9 variables, Curvefit is free too. You can get them both free here http://dudleybenton.altervista.org/software/index.html My big question is, "Why would anyone use python?" You can accomplish this task in C and there are free compilers on the Web, like Digital Mars. M$'s C compiler is free too, even without the extremely annoying Visual Studio IDE. There's free source code to perform kriging in this archive http://dudleybenton.altervista.org/software/Plumes.zip (see folder examples/kriging file kriging.c) I discuss kriging in my book, Plumes, which is free on 2/13, 2/21, and 2/29 https://www.amazon.com/dp/B07ZJMW8S4 I'd be glad to help with data or calculations, but not with python.

Dear Dudley J Benton , Thank you for your answer. I will go through these resources carefully.

(P.S. I meant 'Kriging' as 'Stochastic Kriging'. :) Thank you btw)

I wanted to use Python because my previous code for data preprocessing has been done in Python. However, many python pre-built ML libraries make life a bit easier.

Cheers!

check LSTM.