Weighted Degree Distribution in Python's graph_tool

Question

In Pythons graph_tool, is there an easy way to calculate the weighted degree distribution (sum of the weights on out-going, in-coming, or all edges)?

In the Stats package, vertex_hist gives the un-weighted in-degree, out-degree, and total-degree histograms, but there doesn't seem to be a way to get the weighted version of these.

Note: I am working with 31,000 vertices and >10 million edges. I am looking for a way to do this utilizing graph_tool as as much as possible.

You ask, "Is there an easy way" but then say "I know there are easy ways of doing this"? Then what are they and what's lacking in their efficiency? — Lack, Jan 15 '15 at 04:05
You are right, poor wording on my part. I re-worded it to make it more clear. — nobillygreen, Jan 15 '15 at 18:28

score 5 · Accepted Answer · answered Jan 15 '15 at 09:25

Yes, this is easy. You have to obtain a property map of the weighted degrees, and then do a histogram:

d = g.degree_property_map("out", weight)   # weight is an edge property map
bins = linspace(d.a.min(), d.a.max(), 40)  # linear bins
h = vertex_hist(g, d, bins)

Weighted Degree Distribution in Python's graph_tool

1 Answers1