understand centrality scores

[z-spider]

Hi,

Thank you for this great package. The average clustering in a package is calculated with the following code

clusters = adata.obs[cluster_key].values
nx.algorithms.cluster.average_clustering(graph, clusters)

or in file:

squidpy/src/squidpy/gr/_nhood.py

Line 281 in e77237c

fun_dict[c.s] = partial(nx.algorithms.cluster.average_clustering, graph)

But if I use subgraph, like nx.algorithms.cluster.average_clustering(graph.subgraph(clusters)) to calculate, the result will be different, and I find the second result easier to interpret.
I'm not an expert on this, should it be modified here? Or maybe I misunderstood something?

Any explanationwould be greatly appreciated!