Implementation of a matrix-free isogeometric Galerkin method for Karhunen-Loeve approximation of random fields using tensor product splines, tensor contraction and interpolation based quadrature.

Preprint: https://arxiv.org/abs/2011.13861
Publication: https://doi.org/10.1016/j.cma.2021.113730

This package will not be registered in the public Python Package Index, I recommend the following procedure

1. Clone the repository to your local machine and navigate to the main directory
2. Run sudo pip3 install -e ./ in order to install the tensiga package locally on your machine.

The installation references to the cloned repository. You can work on this repository and changes will be immediately applied to the installation.

Scikit-sparse requires libsuitesparse to build. Ubuntu users can install this library as follows

# apt update && apt install libsuitesparse-dev

Archlinux users might need to provide the suitesparse package

# pacman -Suy suitesparse

Many modules and routines have sample code in the __main__ function, which can be run from the terminal, i.e.

$ python3 -i tensiga/iga/bfun.py

Besides that, under bin/ you can find some full examples to run, i.e.

$ python3 -i bin/ibq_mf_galerkin3d.py

which produces

or a more computationally involved example, like

$ python3 -i bin/spherical_shell.py

which produces

The sample scripts are also available system-wide and directly from the terminal, i.e.

$ ibq_mf_galerkin3d.py