swift-emulator: A Python package for emulation of simulated scaling relations

The Journal of Open Source Software, vol. 7, pp. 4240

Abstract

swift-emulator is a Python toolkit for using Gaussian processes machine learning to emulate scaling relations from cosmological simulations.swift-emulator focusses on implementing a clear, easy to use design and API to remove the barrier to entry for using emulator techniques.swift-emulator provides tools for every step: the design of the parameter sampling, the training of the Gaussian process model, and validating and anaylsing the trained emulators.By making these techniques easier to use, in particular in combination with the SWIFT code (Borrow & Borrisov, 2020;Schaller et al., 2018), it will be possible use fitting methods (like MCMC) to calibrate and better understand theoretical simulation models.

Authors 2

1. Roi Kugel Aachen Leiden Observatory

   Leiden University · Leiden Observatory

   Affiliation as printed

   Leiden Observatory , Leiden University , PO Box 9513 , NL-2300 RA Leiden , The Netherlands

   Leiden Observatory, Leiden University, PO Box 9513, NL-2300 RA Leiden, The Netherlands

2. Josh Borrow

   Massachusetts Institute of Technology

   Affiliation as printed

   Department of Physics , Kavli Institute for Astrophysics and Space Research , Massachusetts Institute of Technology , Cambridge , MA 02139 , USA

   Department of Physics, Kavli Institute for Astrophysics and Space Research, Massachusetts Institute of Technology, Cambridge, MA 02139, USA