Z coordinates and custom study areas?

[ViriatoII]

Hey,

Your package is awesome, I'm just missing the feature of using z coordinates. Is this planned for the future?

Here is an implementation of Ripley's K that uses xyz coordinates, I wish we could join the best of both packages!

Also, how do your functions deal with study areas for Monte Carlo Simulations? Do they assume a size and shape? Is it possible to give a custom one?

Cheers,
Ricardo

[ljwolf]

Hi! Sorry that we've been so delayed in responding here. Yes, definitely would be good to have centralized implementations of Ripley functions into a performant and clear package (scikit-spatstat?). We don't plan to support z axes any time soon, but are happy to incorporate/integrate changes to do so. It should not be too difficult to extend the distance/nearest neighbor computations.

For the Monte Carlo simulations, we simulate reference point patterns inside of a given hull. By default, we simulate within the bounding box of the input pattern. But, we support options using arbitrary Shapley polygons, or computing the convex hull (or concave hull) on the fly & simulating from within that.