Hi,

I was looking at the get_modularity function from modularity_maximization.utils. Directed modularity is defined as:
Q = \frac{1}{m}\sum_{i,j} \(A_ij - \frac{k_i^{in} k_j^{out}}{m}\) * \detal_{c_i, c_j}
Strictly speaking A_ij should be A_ji, which is 1 when there is an edge from j to i, and 0 otherwise. This is because of the concept that edges j->i should make larger contributions to Q if k_i^{in} or k_j^{out} is small. See Leicht, E. A., & Newman, M. E. J. (2008). Community Structure in Directed Networks. Physical Review Letters, 100(11), 118703, the paragraph following equ.(3).

However, I also believe this is more of a conceptual point, since using either A_ij or A_ji gives the same Q thanks to the (i,j) symmetry in the definition. Please let me know what you think, as I'm still new to the field and am not 100% sure about my 'beliefs' :)

How do I remove the log message when calling modularity_maximization.utils.get_modularity functions?

It keeps printing Calculating modularity for undirected graph every time I call this function.

In community_newman.py:

nx.set_edge_attributes(G=network, name='weight', values={edge:1 for edge in network.edges})

should be:
nx.set_edge_attributes(G=network, name='weight', values={edge:1 for edge in network.edges()})

Should it not? Otherwise I get "TypeError: 'instancemethod' object is not iterable"

I want to use your implementation of paper titled "Newman, M. E. J. (2006). Modularity and community structure in networks. Proceedings of the National Academy of Sciences of the United States of America, 103(23), 8577–82" but my graph has weight in it's edges.but this implementation ignore graph weight

Hello,

I have a large directed network with 12 000 nodes and 680 000 edges.
Is that possible to use this algorithm in this case ?

Thanks