Population inference
Numpyro/jax version. GPU-ready hierarchical inference for compact-binary populations, with support for current mass models and high-dimensional population analyses.
Public repository →Selected open-source tools developed for gravitational-wave population inference, cosmology, detector forecasting, and statistical analysis.
Current focus
pymcpop-gw is my main current software project for hierarchical inference of compact-binary populations and cosmology with jax, numpyro and pymc and Hamiltonian Monte Carlo. Two active implementations target complementary analysis problems.
Numpyro/jax version. GPU-ready hierarchical inference for compact-binary populations, with support for current mass models and high-dimensional population analyses.
Public repository →Pymc version. Joint gravitational-wave and galaxy-catalogue inference, including luminosity weighting and host-galaxy ranking.
Repository link coming soon →Current development
Tools to identify and characterize morphological features in two-dimensional probability densities, including ensembles of posterior draws and uncertainty on the detected structures.
Main collaborative software
Hierarchical dark-siren inference combining gravitational-wave observations with galaxy catalogues, including population and selection effects. Main developer Nicola Borghi. Maintained by the University of Bologna.
Vectorised Python/JAX forecasts for signal-to-noise ratios and Fisher-matrix parameter estimation in networks of gravitational-wave detectors.
A dedicated GWFAST module for forecasting population-level constraints with future gravitational-wave observatories, with a scientific scope distinct from event-level Fisher forecasts.
Other research software
Earlier software
Retained for previous analyses and reproducibility; newer cosmology workflows are developed primarily in CHIMERA and pymcpop-gw.
Repositories