bskit is a Python package for measuring density bispectra from snapshots of cosmological N-body or hydrodynamical simulations. It can measure auto or cross bispectra in a user-specified set of triangle bins (that is, triplets of 3-vector wavenumbers (k_1,k_2,k_3) such that k_1, k_2, and k_3 fall into separately specified vector magnitude bins and k_1+k_2+k_3=0). Several common sets of bins are also implemented:
- all triangle bins with min(|k_1|,|k_2|,|k_3|) > k_min and max(|k_1|,|k_2|,|k_3|) < k_max for specified k_min and k_max;
- equilateral triangles between specified k_min and k_max;
- isosceles triangles defined by m*|k_1| ~ |k_2| ~ |k_3| for specified multiplier m;
- squeezed isosceles triangles with |k_1| fixed and |k_2| ~ |k_3|.
bskit is built upon the nbodykit simulation analysis package, and users should familiarize themselves with the central nbodykit concepts in the documentation before getting started with bskit.
This package uses the FFT-based bispectrum measurement algorithm presented e.g. in Sec. 2 of
Tomlinson et al., Efficient parallel algorithm for estimating higher-order polyspectra, Astrophys. J., 158, 116 (2019), arXiv:1904.11055
(see below for further references), and was first used for the bispectrum measurements in the following paper:
Foreman, Coulton, Villaescusa-Navarro, and Barreira, Baryonic effects on the matter bispectrum, 2019, arXiv:1910.03597
bskit requires Python 3; otherwise, its main dependency is on nbodykit. After following the nbodykit installation instructions (preferably via conda), all bskit dependencies will also be installed. At this point, simply clone this repository and ensure that the root bskit directory is in your PYTHONPATH, e.g. via
export PYTHONPATH=/path/to/bskit:$PYTHONPATH
Usage instructions and a guide to the included examples can be found here.
If bskit is used in original research, please cite the associated paper:
Foreman, Coulton, Villaescusa-Navarro, and Barreira, Baryonic effects on the matter bispectrum, 2019, arXiv:1910.03597
In addition, please cite the nbodykit paper,
Hand et al., nbodykit: an open-source, massively parallel toolkit for large-scale structure, Astron. J., 156, 160 (2018), arXiv:1712.05834
and the following standard references for the FFT-based bispectrum estimator that the package implements:
Scoccimarro, The Bispectrum: From Theory to Observations, Astrophys. J., 544, 597 (2000), arXiv:astro-ph/0004086
Sefusatti et al., Accurate estimators of correlation functions in Fourier space, Mon. Not. Roy. Astron. Soc., 460, 3624 (2016), arXiv:1512.07295
If you have any questions or would like to contribute to this code, please open an issue or email Simon directly.