Amua

An open source probabilistic programming language and modeling framework, built around decision trees and Markov models, with discrete event and microsimulation support marked as planned.

Amua is an open source modeling framework and probabilistic programming language, released under GPL-3.0. The name comes from the Swahili word for decide or solve, which sets the tone for what it is meant to do. It is lightweight and runs on any operating system with a Java Virtual Machine installed.

The README lays out three goals. Transparent: a graphical user interface gives an intuitive way to develop and document models, and the open source code is always available. Flexible: models can run in Amua or be exported to other programming languages, so the conceptual model stays separate from any particular implementation. Accessible: it is free software that works as a teaching tool or an integrated environment for researchers.

As a probabilistic programming language, Amua treats model parameters as random variables and fits them to data through model calibration using a variety of techniques. That is the piece that turns a plain modeling tool into something closer to a Bayesian workflow, and it is what separates Amua from a mere diagram editor.

The supported types are decision trees and Markov models. Discrete event simulation is listed as planned, as is microsimulation with multiple methods. The README gives no dates or roadmap beyond the planned labels, so the split between what works today and what is coming is the only timeline on offer.

Amua's pitch is transparency, flexibility, accessibility, and a Bayesian core, with the model type list making it easy to tell what exists today from what is still planned.

• Source 1 · Project repository
• Source 2 · README