Seldon is a performant code for performing various types of opinion dynamics simulations (under development), written primarily in C++.
If you want to use micromamba as the package manager, create and activate the environment.
micromamba create -f environment.yml
micromamba activate seldonenvWe use meson to compile and build Seldon.
meson setup build
meson compile -C buildTo install seldon to your conda environment, run the following:
meson setup build --prefix $CONDA_PREFIX
meson install -C buildRun the executable, and provide the input configuration TOML file (as the first positional argument), and an optional output directory location. If the output location is not specified, it is set to the parent directory in which the config file is present
cd build
./seldon /path/to/config -o /path/to/output/dirIf you've installed it, you can simply run seldon anywhere.
seldon /path/to/config -o /path/to/output/dirThe file network.txt contains information about the network.
First column is the index of the agent, then the next column is the number of incoming agent connections including the agent itself. Subsequent columns are the neighbouring incoming agent indices and weights. In addition, every iteration produces a double opinion value for each agent. These are outputted to files named opinions_i.txt.
To run the tests, go into the build directory and run the following:
meson testAll contributions are welcome!!
- We follow a modified NumPy commit guidelines.
- Each commit is to be prefixed by the area of code it affects
- Please run our linter and tests locally to ensure no errors exist
- Co-author commits generously
GNU GPL v3 or later.