GriSPy: A Python package for Fixed-Radius Nearest Neighbors Search

MARTÍN CHALELA; EMANUEL SILLERO; LUIS PEREYRA; MARIO ALEJANDRO GARCIA; JUAN B CABRAL; MARCELO LARES; MANUEL MERCHÁN

Astronomy and Computing

Elsevier

Año: 2021 vol. 34

2213-1337

We present a new regular grid search algorithm for quick fixed-radius nearest-neighbor lookup developed in Python. This moduleindexes a set of k-dimensional points in a regular grid, with optional periodic conditions, providing a fast approach for nearestneighbors queries. In this first installment, we provide three types of queries: bubble, shell and the nth-nearest. For these querieswe include three different metrics of interest in astronomy, namely, the euclidean, the haversine and the Vincenty, the last two inspherical coordinates. We also provide the possibility of using a custom distance function. This package results particularly usefulfor large datasets where a brute-force search turns impractical.