[graph-tool] Fast initialization of graph
jonas at ifany.org
Wed Mar 20 08:34:28 CET 2013
To follow up on your answer in case others will encounter it someday: Using
random_graph is indeed much faster (0.01s as opposed to 7s for a fully
connected graph with 500 vertices)
An important pitfall when using random_graph to generate a graph is that
the edges of the graph aren't iterated over in the order of their index.
This means that assigning values to the underlying array of a propertyMap
is tricky. A simple fix is to call
after generating the graph. This operation also runs in about 0.01s on a
fully connected graph of 500 nodes on my machine.
On 19 March 2013 21:35, Jonas Arnfred <jonas at ifany.org> wrote:
> Thanks a lot, I'll try it out!
> On 19 March 2013 18:51, Tiago de Paula Peixoto <tiago at skewed.de> wrote:
>> On 03/19/2013 09:39 AM, arnfred wrote:
>> > I'm currently trying to instantiate a fully connected graph with some
>> > vertices, but I find that adding all the edges usually takes around 10
>> > seconds on my system. The fastest way of doing it that I have come up
>> > so far is to write:
>> > from itertools import combinations
>> > edges = [g.add_edge(v1,v2) for (v1,v2) in combinations(g.vertices(),2)]
>> > But I'm wondering if there is a faster method?
>> You can create a "random" graph with all degrees equal to N - 1:
>> g = random_graph(600, lambda: 600 - 1, directed=False, random=False)
>> This should be much faster. Note the option 'random=False' which avoids
>> the random placement of the edges, which would be completely pointless
>> in this case.
>> I'm planning to include a complete graph generator, as well as some
>> other simple generators, which would make this more straightforward.
>> Tiago de Paula Peixoto <tiago at skewed.de>
>> graph-tool mailing list
>> graph-tool at skewed.de
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the graph-tool