THis is the initial commit, not release version. any code shared AS IS
This is a python wrapper around C++ implementation (for Modularity maximization) of the community detection algorithm called "Combo" described in the paper "General optimization technique for high-quality community detection in complex networks" by Stanislav Sobolevsky, Riccardo Campari, Alexander Belyi and Carlo Ratti.
"Combo" algorithm for community detection described in the paper:
Sobolevsky S., Campari R., Belyi A., and Ratti C. "General optimization technique for high-quality community detection in complex networks" Phys. Rev. E 90, 012811
Combo achieves hight quality of partitioning, while being less greedy in terms of computation, than other algorithms.
pip install -e git+git@github.com:Casyfill/pyCombo.git@master#egg=pyCombo
pyCombo does not have any dependency, yet it was created having NetworkX module in mind.
All copiryghts and licensing is same as covered in the initial package
- Initial release ASIS
- Tests, Travis Ci, Coveralls
- Package delivery, setup.py
- Installation via pip
- unweighted graph
- switched to Github Actions
- switch to poetry
- check & test modularity
- assert reproductibility
- understand 2 partitions issue/reason
- directed graph
- Exceptions
- pyCombo does not work on graph of N=1
- pyCombo for some reason always split the network into 2, even for the complete graph
- binaries are now pulled from https://github.com/Express50/Combo/blob/master/Main.cpp, which fixes a few bugs and passes