Parax v0.7: JAX Parametric Modeling

·2026.05.10 18:31

Key point

Parax, a JAX-based parametric modeling library, has been updated to version v0.7.

Details

Parax is a Parametric modeling library designed to bridge the gap between JAX's pure PyTree approach and object-oriented modeling approaches like Equinox.

This v0.7 update comes with a more refined API and detailed documented examples. Key features include:

Support for Derived/constrained parameters with metadata
Computed PyTrees and Callable parameterizations
Abstract interfaces for fixed, bounded, and probabilistic PyTrees and parameters

Also, new usage examples such as Bounded optimization using JAXopt and Bayesian sampling using BlackJAX can be found in the documentation.

machine-learning optimization open-source jax parametric-modeling

Read originalShare on Threads

This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.

Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.