Community detection with Louvain

// Load network from file
Louvain louvain("from_to.network");

// Partition communities
louvain.partition();
auto communities = louvain.get_communities();
auto reverse_communities = louvain.get_reverse_communities();

// Save communities to file
louvain.save_to_from("to_from.network"); // For CheiRank
louvain.save_reverse_communities("example.rcomm");
Vector<std::string> filenames{"1.comm", "2.comm", ...};
louvain.save_communities(filenames);

Saving paritions (THESE NEEDS TO BE DONE SEPERATELY, THE CODE FOR THIS IS DISGUSTING AND FOR NOW ONLY WORKS IT ONLY WORKS LIKE THIS)

Louvain louvain("from_to.network");
louvain.set_communities("example.rcomm");
Vector<std::string> filenames{"1_ft.network", "2_ft.network", ...};
louvain.save_partitions(filenames);

Computing PageRank of overall network

GoogleMatrix google("from_to.network");
auto pagerank = google.pagerank(0.85);

Computing CheiRank of overall network

GoogleMatrix google("to_from.network");
auto cheirank = google.pagerank(0.85);

Computing PageRank within community

GoogleMatrix google("1_ft.network");
auto pagerank = google.pagerank(0.85);

Computing CheiRank within community
First get to_from

SparseNetwork network("1_ft.network");
network.save_to_from("1_tf.network");

Then calculate

GoogleMatrix google("1_tf.network");
auto cheirank = google.pagerank(0.85);