VAST: Void Analysis Software Toolkit

The Void Analysis Software Toolkit, or VAST, provides pure Python implementations of two popular classes of void-finding algorithms in galaxy catalogs:

1. Void identification by growing spherical voids.
2. Void identification using watershed algorithms.

Our docs can be found here: https://vast.readthedocs.io/en/latest/

VoidFinder

VoidFinder is an algorithm which utilizes a sphere-growing method on a grid search and a unionization of the sufficiently large spheres. The VoidFinder directory contains the package, which includes an efficient Multi-Process Cythonized version of VoidFinder (from vast.voidfinder import find_voids), as well as an OpenGL based visualization for the output of VoidFinder (the vast.voidfinder.viz package).

See here for 3D OpenGL-based visualization of VoidFinder's voids in SDSS DR7!

V²

V² is a voronoi-tesselation-based algorithm for finding the void regions, based on the ZOBOV algorithm. ZOBOV uses the gradient of the volume of adjacent voronoi cells to flow multiple cells together into large void regions.
The Vsquared directory contains the package, which includes five different
methods for building void regions from voronoi cells, as well as an OpenGL based visualization for the output of V² (the vast.vsquared.viz package).