BayesEoR: Bayesian 21-cm Power Spectrum Estimation from Interferometric Visibilities

BayesEoR is a GPU-accelerated, MPI-compatible Python package for estimating the power spectrum of redshifted 21-cm emission from interferometric observations of the Epoch of Reionization (EoR). Utilizing a Bayesian framework, BayesEoR jointly fits for the 21-cm EoR power spectrum and a "foreground" model, referring to bright, contaminating emission between us and the cosmological signal, and forward models the instrument with which these signals are observed. To perform the sampling, we use MultiNest [arXiv:1402.0004], which calculates the Bayesian evidence as part of the analysis. Thus, BayesEoR can also be used as a tool for model selection [see e.g. arXiv:1701.03384].