Statement of need¶

Building the source population for a gravitational-wave mock data challenge usually means bespoke sampling scripts that are hard to reproduce and hard to reconfigure when priors change. gwmock-pop is the forward counterpart to population-inference tools such as gwpopulation: instead of inferring hyper-parameters from observed events, it draws synthetic catalogues from configurable priors. Its graph-driven sampler lets users declare arbitrary parameter-dependency structures in YAML/TOML — validated without executing arbitrary Python — and ships presets reflecting recent observed populations. It is the population layer of the gwmock mock-data-challenge ecosystem, usable standalone or through the gwmock orchestrator, and scales to the catalogue sizes (of order 10⁵ sources per year) expected from next-generation detectors such as the Einstein Telescope.