gwsnr: A Python package for efficient signal-to-noise ratio calculations of gravitational waves

Abstract

$gwsnr$ is a Python package for efficient signal-to-noise ratio and detectability calculations for compact-binary gravitational-wave sources. It is designed for large simulated populations where repeated evaluation of the optimal signal-to-noise ratio, $\ddot{I}{}_{\mathrm{opt}}$, and the probability of detection, $P_{\det }$, becomes computationally expensive with direct noise-weighted inner-product calculations. The package provides multiple calculation pathways under a unified interface, including multiprocessing-based inner-product evaluation, partial-scaling interpolation for non-spinning and aligned-spin systems, JAX and MLX backends for accelerated array execution, and artificial neural network based detectability estimation for more complex waveform settings. It also supports statistical modelling of the observed signal-to-noise ratio, $\ddot{I}{}_{\mathrm{obs}}$, under stationary Gaussian noise assumptions, threshold estimation from injection catalogues, hybrid recalculation of marginal events near the detection boundary, and horizon-distance calculations. By combining fast numerical methods with configurable detector, waveform, and population settings, $gwsnr$ enables efficient selection-effect modelling, rate estimation, detector-sensitivity studies, and large-scale compact-binary population simulations. The package is publicly available with documentation, validation examples, and reproducible workflows.

6 pages, 0 figures