ninpol

Library of Nodal Interpolation Techniques for Finite Volume Schemes

Installation

pip install ninpol

Pre-requisites

requires = [
  "setuptools",
  "wheel",
  "Cython",
  "numpy",
  "scipy"
]

Usage

import ninpol

interpolator = ninpol.Interpolator(logging = True)
interpolator.load_mesh(mesh_file)

weights, neumann = interpolator.interpolate(variable, method)

Where:

• mesh_file is the path to the mesh file
• variable is the variable to be interpolated, associated with the elements of the mesh
• method is the interpolation method to be used (can be checked with interpolator.supported_methods)
• weights is a sparse matrix of shape (n_nodes, n_elements) containing the interpolation weights for each node.
• neumann is a numpy array of shape (n_nodes) containing the value of the Neumann boundary condition values for each node. It's filled with 0's if there's no node in a Neumann boundary condition domain.

Tests

On linux, run, on the project root directory:

make test

On Windows or MacOS:

pytest -s --tb=short
python3 ./tests/results/graph.py

The second command will generate the graphs and .csv files with the results of the tests.

OBS: Requires project.optional-dependencies to be installed.