skysegmentor 1.0.0

SkySegmentor: Splits a binary map into several regions using the binary space partition algorithm.

Introduction

SkySegmentor is a python package for dividing points or maps (in HEALPix format) on the celestial sphere into equal-sized segments. It employs a sequential binary space partitioning scheme -- a generalization of the k-d tree algorithm -- that supports segmentation of arbitrarily shaped sky regions. By design, all partitions are approximately equal in area, with discrepancies no larger than the HEALPix pixel scale.

Dependencies

• numpy>=1.22,<1.27
• healpy>=1.15.0

Installation

Pip installation:

pip install skysegmentor

From the source:

Clone the repository

git clone https://github.com/knaidoo29/SkySegmentor.git
cd SkySegmentor

and install by running

pip install . [--user]

Documentation

Documentation, including tutorials and explanation of API, can be found here https://skysegmentor.readthedocs.io/. Alternatively a PDF version of the documentation is located in the docs/ folder called skysegmentor.pdf. Offline documentation can be generating by running make html in the docs/ folder which will generate html documentation in the docs/build/html folder that can be accessed by opening the index.html file in a browser.

Tutorial

Basic Usage

Segmenting Healpix Maps

import healpy
import skysegmentor

# Healpix mask, where zeros are regions outside of the mask and ones inside the
# mask. You can also input a weighted map, where instead of 1s you give weights.
mask = # define mask values

Npartitions = 100 # Number of partitions
partitionmap = skysegmentor.segmentmapN(mask, Npartitions)

Segmenting Points on the Sphere

import skysegmentor

# Define points on the sphere to be segmented.
phi = # longitude defined in radians from [0, 2*pi]
the = # latitude defined in radians from [0, pi], where 0 = North Pole.

Npartitions = 100 # Number of partitions
partitionIDs = skysegmentor.segmentpointsN(phi, the, Npartitions)

if using RA and Dec in degrees you can convert to phi and the using

phi = np.deg2rad(ra)
the = np.deg2rad(90. - dec)

if not all points are equal, you can specify a weight

weights = # define point weights
partitionIDs = skysegmentor.segmentpointsN(phi, the, Npartitions, weights=weights)

Citing

SkySegmentor was developed as part of the Euclid angular power spectra internal covariance pipeline. If you use SkySegmentor please cite this paper, which can be found here:

--THESE ARE PLACEHOLDERS TO BE UPDATED LATER--

• NASA ADS:
• ArXiv:
• BibTex:

  @ARTICLE{Naidoo2025,
          author = {{Euclid Collaboration} and {Naidoo}, K. and 
          {Ruiz-Zapatero}, J. and {Tessore}, N. and 
          {Joachimi}, B. and {Loureiro}, A. and
          others ...}
          title = "{Euclid preparation: TBD. Accurate and precise data-driven angular power spectrum covariances}
  }
  

and include a link to the SkySegmentor documentation page:

https://skysegmentor.readthedocs.io/

Support

If you have any issues with the code or want to suggest ways to improve it please open a new issue (here) or (if you don't have a github account) email [e-mail].