Checkout jinns v1.9

Created on April 10, 2026

2026

Changelog: new in 1.9 - Natural Gradient Descent is now available and custom optimizers for second order methods. More incoming soon (e.g. Anagram or self-scaled Broyden)

Many thanks to Hugo Gangloff for his dedication on this, and to the ScimBa team for stimulating discussions. You should go check out their work as well!

jinns is a Python package for physics-informed neural networks (PINNs) we develop together with Hugo Gangloff as a basis for our research. Using the JAX ecosystem, it provides an intuitive and flexible interface for

  • forward problem: learning a PDE solution.
  • inverse problem: learning the parameters of a PDE.
  • meta-modeling: learning a family of PDE indexed by its parameters.

Check out the documentation: https://mia_jinns.gitlab.io/jinns/

Want to use or contribute ? Development happens on Gitlab

image