photomancy 0.0.1

Divination from light: Bayesian inference and value-of-information for HWO direct imaging

photomancy

Divination from light.

A JAX-native Bayesian inference and value-of-information engine for the Habitable Worlds Observatory direct-imaging simulation suite. orbix builds the geometry, skyscapes builds the scene, and photomancy divines the scene back from the data: posteriors, evidence, and the next-best observation, over orbits, disks, and (later) atmospheres and images.

The engine is forward-model agnostic. A fit is a logdensity over a partitioned scene PyTree, assembled from three plug-ins -- a forward model, a likelihood, and a prior -- and run through a uniform Backend (Laplace mixture, NUTS, adaptive tempered SMC, MCLMC, Pathfinder) that returns one Posterior exposing .sample, .log_prob, and .evidence.

Status: early development. Orbit fitting is implemented; disk, atmosphere, and image-domain fitting are planned.