Performance reduce when amount nodes more than 300

[ihsansatriawan]

I have (u,v,w) where distinct value from u is more than 300, and time execute so slow,,,, maybe you have any approach to improve this problem ? thx

[canbax]

Algorithm by itself is slow. It is something like O(n^3). Algorithm runs until there are no edges which is really not necessary. You can terminate algorithm early. Actually if you examine graph of modularity, modularity starts to decrease constantly. You can stop algorithm after sometime. You can look my fork