Re: [igraph] A problem about the theory of the layout function

[Gábor Csárdi]

That function was ported from the sna R package, and we used the same defaults.

In the next version of igraph a completely new implementation of this
function is included, and the new version does exactly the same as the
original Fruchterman-Reingold paper (unlike the ported sna
implementation).

Best,
Gabor

On Mon, Jul 14, 2014 at 10:51 PM, 李甜 <address@hidden> wrote:
> Hi,
> The version I was using is igraph-0.6.5-msvc. I used function
> igraph_layout_fruchterman_reingold from layout.c,and I want to ask a
> question about this function.When I read the source code ,I found the
> cooling function was set "t=maxdelta*pow(i/(double)niter,coolexp)".So I want
> to know why did you set it like this and what is the theory about it?And
> also why the default value of coolexp is 1.5 and the default value of
> maxdelta is the number of vertices?
> Hope for your reply.
> Thank you.
> Best regards!
>
> Candy
>
> _______________________________________________
> igraph-help mailing list
> address@hidden
> https://lists.nongnu.org/mailman/listinfo/igraph-help
>