gwinferno.pipeline.analysis

a module that stores the meat of the calculations for hierarchical population inference

Functions

construct_hierarchical_model
detection_efficiency	Calculates the detection efficiency -- the expected fraction of sources detected from a population parameterized by \(\Lambda\) -- estimated by importance sampling over the found injections from a fiducial population parameterized by \(\Lambda_\emptyset\):
find_map	Find the MAP estimate for a given NumPyro model using SVI with Adam optimizing the ELBO
hierarchical_likelihood	Performs the hierarchical likelihood calculation using importance sampling over injections and PE samples from each event's posterior samples assuming a fiducial prior density.
per_event_log_bayes_factors	Calculates per-event log Bayes factors via importance sampling