Skip to main content

Physics Informed Neural Network with JAX

Project description


Physics Informed Neural Networks with JAX. jinns has been developed to estimate solutions to your ODE et PDE problems using neural networks. jinns is built on JAX.

jinns specific points:

  • jinns is coded with JAX as a backend: forward and backward autodiff, vmapping, jitting and more!

  • In jinns, we give the user maximum control on what is happening. We also keep the maths and computations visible and not hidden behind layers of code!

  • In the near future, we want to focus the development on inverse problems and inference in mecanistic-statistical models

  • Separable PINNs are implemented

  • Hyper PINNs are implemented

  • Check out our various notebooks to get started with jinns

For more information, open an issue or contact us!


Install the latest version with pip

pip install jinns


The project's documentation is available at


  • First fork the library on Gitlab.

  • Then clone and install the library in development mode with

pip install -e .
  • Install pre-commit and run it.
pip install pre-commit
pre-commit install
  • Open a merge request once you are done with your changes.

Contributors & references

Active: Hugo Gangloff, Nicolas Jouvin Past: Pierre Gloaguen, Charles Ollion, Achille Thin

Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

jinns-0.8.7.tar.gz (18.2 MB view hashes)

Uploaded Source

Built Distribution

jinns-0.8.7-py3-none-any.whl (87.6 kB view hashes)

Uploaded Python 3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page