Benjamin Murphy

Sorry for the slow response... that is indeed something that's been requested before and is on my long-term radar, but @rth and I have a bunch of other stuff to...

Thanks for getting on this @rth, and sorry it's taken me a little while to take a look at this. So, the way I wrote this originally, I believe `external_drift`...

What exactly are you trying to do? What is the external drift that you're trying to use? The idea with the external scalar drift is that you have some known...

Also, you may have already figured this out, but the external grid drift should still work if the point spacing isn't uniform. As long as the grid is rectangular and...

Yes, your first sentence is correct. If you have any rectangular grid of points (not necessarily with uniformly spaced x's and y's, although you need one grid value for every...

I think this is a really good idea, especially as it can cut down a lot on redundancies in the code. You're right, all of the kriging classes rely on...

Currently, the statistics do rely purely on the variogram/distances, as in the case of OK. That's not technically correct tho, so it's an area for improvement...

Here are some more thoughts... I do like the ```sklearn.pipeline``` idea, as it allows for a great deal of flexibility. However, I also think we should support a "simple" interface...

That is a great question... unfortunately right now there aren't any methods to return those quantities, but this would be easy enough to implement. I'm going to add this to...

For all the datasets I've worked with, the statistics calculation has been really fast, although I guess I don't actually know how it scales... regardless, if you think a flag...