femtoscope

The femtoscope software was written as part of Hugo Lévy's PhD thesis (2021-2024). This Python code can be used to investigate different models of modified gravity within complex geometries on unbounded domains.

Installation

femtoscope can be either installed from PyPI

pip install femtoscope

or from conda-forge

 conda install conda-forge::femtoscope

Generate documentation

The documentation may be generated by running

python -m femtoscope.misc.generate_documentation
cd doc/

Alternatively, one can use the CLI Entry Point

generate-femtoscope-docs
cd doc/

Note that these two options will generate the documentation at the current directory. Additional information about femtoscope can be found in Hugo Lévy's PhD thesis, chapter 4.

Run tests

Note on running tests

femtoscope's test suite uses the pytest framework, which should be installed separately by users. Install it with pip install pytest (PyPI) or conda install conda-forge::pytest (conda-forge).

To check the installation, it is recommended to run the tests bundle with

python -m femtoscope.tests.run_tests

Alternatively, one can use the CLI Entry Point

run-femtoscope-tests

Getting started

Getting started notebooks are available at https://github.com/onera/femtoscope/tree/main/tutorials.

Development mode

Clone the femtoscope repository https://github.com/onera/femtoscope/ which contains the source files. It is recommended to run femtoscope on a dedicated virtual environment. The femtoscope.yml file can be used to automate the creation of a conda environment as follows:

1. Create the environment

   conda env create -f femtoscope.yml

   Note that this process is likely to take several tens of minutes.
2. Activate the new environment

   conda activate femtoscope

3. Verify that the new environment was installed correctly

   conda env list

   and make sure femtoscope appears in the list.

note The project uses Python >= 3.9

Main dependencies

• Python >= 3.9
• meshio == 4.4.6 (read & write mesh files)
• numpy < 2 (sfepy is not yet compatible with numpy 2.x)
• pandas >= 1.5.3
• pyevtk (creation of .vtk files)
• pyvista (visualization of FEM results)
• scipy
• sfepy >= 2024.3 (FEM engine)
• pdoc >= 14.5.1 (patch against polyfill.io malicious CDN)
• gmsh == 4.11.1 (some features of femtoscope are broken in newer versions)

Tree structure

├───doc
│
├───femtoscope
│   ├───core	# weak form representation & solvers
│   ├───data	# I/O files
│   │   ├───mesh
│   │   │   └───geo
│   │   ├───model
│   │   ├───result
│   │   └───tmp
│   ├───display
│   ├───images
│   ├───inout	# mesh & VTK generation
│   ├───misc
│   ├───physics	# Poisson & Chameleon classes
│   ├───tests	# unit & integration tests
│
└───tutorials	# hands-on notebooks

UML Diagram

Citing

If you would like to cite femtoscope in a paper or presentation, please use the following reference:

@article{hlevy:femtoscope,
	doi = {10.1103/PhysRevD.106.124021},
	url = {https://link.aps.org/doi/10.1103/PhysRevD.106.124021},
	author = {Lévy, Hugo and Bergé, Joël and Uzan, Jean-Philippe},
	title = {Solving nonlinear Klein-Gordon equations on unbounded domains via
	the Finite Element Method},
	month = {09},
	year = {2022}
}