add mpi example?

[basaks]

I have already implemented an MPI Krige pipeline in another package, using which I can krige large geotifs.

I have so far used Kriging on 20000x12000 pixels in 10 minutes using ordinary krige + linear + n_esitmators_points=50 using 32 cores.

Let me know if this will be desirable, and I can put together another PR with an MPI example.

[rth]

That is definitely interesting, my only concern against including an MPI example in PyKrige is that once it's in the code base it would have to be supported (i.e. if it doesn't work for a user due to his MPI setup, he might open issues here, which is a bit off topic ).

Are you using MPI for the parallelism (with mpi4py etc), not multiprocessing or joblib.Parallel (the former is build-in and the later is included in sklearn.externals)?

If you are interested, maybe you could put your MPI example in a separate Github gist, and we will link to it from the documentation (once it is added in #14 )? Not sure what's the best way of handling this..

[basaks]

Sorry did not see this earlier.
Yes, I was talking about mpi4py.

Let's do what you suggest.

[basaks]

What would be the equivalent of n_closest_points for UniversalKriging?

[rth]

Kriging with a moving window is not yet implemented for UniversalKriging (see #5 (comment)). Feel free to make a new PR with it .

[MuellerSeb]

Closing due to inactivity. Feel free to re-open.