See https://doi.org/10.48550/arXiv.2308.00083 for the theoretical background.
This repository contains two algorithms for enumerating subnetworks of multilayer networks, both implemented in Python and in C++. For the Python implementation, the pymnet library is used. C++ version uses Boost Graph Library.
Compile C++ with:
g++ -DN_ASPECTS=n_aspects -std=c++17 -O3 mesu.cpp -o mesu_n_aspects.out
where n_aspects should be replaced with the desired number of aspects.
Or:
make
will compile versions from 1 to 6 aspects.
Use C++ with:
mesu_d.out inputfile outputfile 'subnet_size_in_aspect_0,subnet_size_in_aspect_1,...,subnet_size_in_aspect_d'
To run C++ benchmarks with run_benchmark_models_cpp (in benchmarks.py) there need to be compiled versions of mesu.cpp with names mesu_n_aspects.out.