Install

uv is the preferred tool to install scimba, but you can also use pip. In any case, it is strongly recommended to isolate your installation in a virtual environment (uv does this for you).

uv

If your project is handled by a pyproject.toml:

uv init
uv add "scimba[scimba_jax]"

otherwise:

uv pip install "scimba[scimba_jax]"

pip

pip install "scimba[scimba_jax]"

You might also want to clone the scimba repository to get access to the example suite:

git clone git@gitlab.com:scimba/scimba.git

then in your repo:

uv

uv init uv add “scimba[scimba_jax] @ path/to/local/scimba”

pip

pip install "scimba[scimba_jax] @ path/to/local/scimba"

Tutorials

Each tutorial is a single Jupyter notebook that helps you get started with Scimba.

Basics of scimba_jax

Solve pre-defined stationary and time-dependent models.

Scimba basics I: approximation of the solution of pre-defined physical models

Strong Boundary Conditions

Implementation of strong boundary and initial conditions.

Strong boundary conditions

Mixed Boundary Conditions

How to apply mixed boundary conditions.

Mixed Boundary Conditions with PINNs

Domains and Samplers

How to define domains and samplers.

Domains and Samplers

Save and load trained models.

How to save and load objects.

Save and load PINNs and other scimba_jax objects

Custom Physical models

How to define custom physical models.

Defining a physical model

Neural-operator data

• Neural-operator data

FEM, DG and multigrid

• Linear approximation: FEM, DG and multigrid

Weak forms and parametric functions

• Weak forms and ParamFunction

Examples (JAX)

A complete list of JAX script examples is available in the examples/examples_jax/ directory of the repository.