Hi, I think the best way to fix your issue is to fix https://trac.sagemath.org/ticket/21313 This will allow is_isomorphic to use bliss or nauty. (In fact, IIRC checking for isomorphism is in practice a bit faster than canonical labelling).
Also, do I get it right that you have a new "generic" construction of strongly regular graphs? As you know, we are collecting them - would you be interested in adding it into sage/graphs/generators/ ? Dima On Tuesday, August 23, 2016 at 2:53:34 PM UTC+1, Paul Leopardi wrote: > > Hi all, > I am currently trying to use Sage to classify bent functions by their > Cayley graphs. > > I have attached an example where I have two (256, 120, 56, 56) strongly > regular graphs, g and h, which are also canonical labels, such that g does > not equal h, and so g and h are not isomorphic. > It takes a relatively short time to produce the canonical labels, 38.8s to > run h.is_isomorphic(g), and g.is_isomorphic(h) is still running. > > The example includes the Sage code that I used to generate the graphs, as > well as a transcript of my Sage session, including the graph6_string() of > each of the two graphs. > > I would like some advice on how I should proceed with the classification, > given that I want to compare about 2^(16) such graphs to a small number of > representative graphs (about 6). > > I have done other tests with graphs of this size that show that usually > g.is_isomorphic(h) is faster than comparing g.automorphism_group().order() > with h.automorphism_group().order(), which is faster than comparing > g.canonical_label() with h.canonical_label(). > It's just that occasionally I have found cases where g.is_isomorphic(h) is > extremely slow. > > I'm thinking of using threading and a timeout of (e.g.) 5 seconds, so that > my search first tries to use g.is_isomorphic(h), and if this times out, it > then compares the automorphism group orders, and only if they are equal, > does it then compute the canonical label. > > What are other people doing? > All the best, Paul > > -- You received this message because you are subscribed to the Google Groups "sage-support" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at https://groups.google.com/group/sage-support. For more options, visit https://groups.google.com/d/optout.
