About Edmot

[Liuxx21]

Hello,
thanks for the awesome work!!

I have two questions about this code.

The first one is :

in the code below,
"""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""
def _fill_blocks(self):
"""
Filling the dense blocks of the adjacency matrix.
"""
new_edges = [(n_1, n_2) for nodes in self.blocks for n_1 in nodes for n_2 in nodes if n_1!= n_2]
new_graph = nx.from_edgelist(new_edges)
self.graph = nx.disjoint_union(self.graph, new_graph)
"""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""
when using the nx.disjoint_union() function to enhance the graph，this function will lead the total count of node in graph rise from 2708 to 2946. I wonder if this will cause some problem.

The second one is :

The proposed model in the original paper(Edmot by Pei-Zhen Li) noticed that there are 4 steps. But I found that your code only derive a clique from the largest connected component without partitioning the component into modules, which is different from the paper.

wait for your reply:)

Hope it helps :)

[Benedek Rozemberczki]

1. Was resolved.
2. If you take a closer look You will see that the number of components is a parameter.

[粽子]

good project!