Re: Degree sequence game segfault, was: Re: [igraph] problems with parameters in barabasi.game()

[Mario Lavezzi]

I did many simulations today and, following Gabor's suggestions (many 
thanks, again), everything went fine.

Only little problem: the degree.sequence.game() function actually does 
not generate the graph quite often as I increase the power-law exponent. 
With the  exponent set to 2 I found no particular problems, with the 
exponent set to 2.6 I find quite difficult to obtain a graph.

For the moment, I am circumventing this by running simulations with 
different seeds (using the R instruction set.seed()), until I obtain a 
graph.

I am simulating networks with 1000 agents, and I am not quite sure that, 
by augmenting the number of agents, the difficulty for the code to 
generate a graph with exponents higher than two will disappear.

Mario


Csardi Gabor wrote:
> Mario, please read this for a workaround:
>
> http://lists.gnu.org/archive/html/igraph-help/2008-09/msg00083.html
>
> Gabor
>
> On Wed, Sep 10, 2008 at 06:12:37PM +0200, Csardi Gabor wrote:
>   
> > Mario, no, it is not supposed to "work" like this, this is 
> > a bug, possibly in igraph, possibly in the C++ code we use 
> > for the graph generation. Sorry about this, I'll try to correct it.
> >
> > G.
> >
> > On Wed, Sep 10, 2008 at 05:59:10PM +0200, Mario Lavezzi wrote:
> >     
> > > Gabor, thanks a lot. It seems to work but, most of the times, after 
> > > following your suggestions, I get a message like this:
> > >
> > > --------------------------------------------------------------------
> > > *** caught segfault ***
> > > address (nil), cause 'memory not mapped'
> > >
> > > Traceback:
> > > 1: .Call("R_igraph_degree_sequence_game", as.numeric(out.deg), 
> > > as.numeric(in.deg),     as.numeric(method), PACKAGE = "igraph")
> > > 2: degree.sequence.game(degreeSeq, method = "vl")
> > >
> > > Possible actions:
> > > 1: abort (with core dump, if enabled)
> > > 2: normal R exit
> > > 3: exit R without saving workspace
> > > 4: exit R saving workspace
> > >
> > > -----------------------------------------------------------
> > >
> > > ...and have to stop..
> > >
> > > I understand this has to do with the way R works, but I have not found 
> > > any useful help on this in the web
> > > (by the way, I use R 2.6.2 on Linux Ubuntu 8.04)
> > >
> > > Mario
> > >
> > >       
> > > > Mario, I would try 'sample' first, e.g. 
> > > >
> > > > degs <- sample(1:1000, 1000, replace=TRUE, prob=(1:1000)^-2)
> > > > g <- degree.sequence.game(degs, method="vl")
> > > >
> > > > Note that it is not always possible to realize a given degree sequence,
> > > > e.g. if the sum of degrees is odd, then it is trivially impossible.
> > > >
> > > > G.
> > > >  
> > > >         
> > > --
> > > Andrea Mario Lavezzi
> > > Dipartimento di Studi su Politica, Diritto e Società
> > > Università di Palermo
> > > Piazza Bologni 8
> > > 90134 Palermo, Italy
> > > tel. ++39 091 6625600
> > > fax ++39 091 6112023
> > > skype: lavezzimario
> > > email: lavezzi (at) unipa.it
> > > web: http://www.unipa.it/~lavezzi
> > >
> > >
> > >
> > > _______________________________________________
> > > igraph-help mailing list
> > > address@hidden
> > > http://lists.nongnu.org/mailman/listinfo/igraph-help
> > >       
> > --
> > Csardi Gabor <address@hidden>    MTA RMKI, ELTE TTK
> >
> >
> > _______________________________________________
> > igraph-help mailing list
> > address@hidden
> > http://lists.nongnu.org/mailman/listinfo/igraph-help
> >     
>
>   

--
Andrea Mario Lavezzi
Dipartimento di Studi su Politica, Diritto e Società
Università di Palermo
Piazza Bologni 8
90134 Palermo, Italy
tel. ++39 091 6625600
fax ++39 091 6112023
skype: lavezzimario
email: lavezzi (at) unipa.it
web: http://www.unipa.it/~lavezzi