igraph-help
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [igraph] Recommended values for Kamada-Kawai layout


From: Matthew Walker
Subject: Re: [igraph] Recommended values for Kamada-Kawai layout
Date: Thu, 22 Oct 2009 12:24:04 -0400
User-agent: Thunderbird 2.0.0.23 (X11/20090812)

Hi Gábor,

I've tried to follow your instructions.  My patch (to layout.c) is attached but I have a couple of questions:
(i) is this correctly done?
(ii) is this the best way to send such patches?

Cheers,

Matthew

Gábor Csárdi wrote:
Hi Matthew,

good point. Actually, for Python and R we have default values for all
layout generators, please see e.g.
http://igraph.sourceforge.net/doc/R/layout.html and
http://igraph.sourceforge.net/doc/R/layout.drl.html

But you are right, for C the default values should be documented. If
you wish to do this, the best way is downloading the latest igraph
version from Launchpad, or in your case a couple of files, such as
src/layout.c should be enough, updating the files, and sending a patch
with the changes.
You can download the individual files here:
http://bazaar.launchpad.net/~igraph/igraph/0.6-main/files/head:/src/

Please tell me if anything is unclear. Thanks, best,
Gabor

On Thu, Oct 22, 2009 at 5:20 PM, Matthew Walker
<address@hidden> wrote:
  
Hi,

I do not know if this is the appropriate place to say this, but I found it
useful to know some recommended values [1] for the Kamada-Kawai layout
algorithm.  I noticed that the graphopt algorithm's documentation included
such values and I found them very useful.  Perhaps we could add some
recommended values to the documentation for igraph_layout_kamada_kawai()?

I would be happy to contribute such a (minor) modification to the code's
documentation, but I do not know if you would want this, nor the best way to
offer it to you.

[1] http://svitsrv25.epfl.ch/R-doc/library/sna/html/gplot3d.layout.html

Cheers,

Matthew

_______________________________________________
igraph-help mailing list
address@hidden
http://lists.nongnu.org/mailman/listinfo/igraph-help


    



  
*** layout.c    2009-10-22 11:53:00.000000000 -0400
--- layout-modified.c   2009-10-22 12:22:57.000000000 -0400
***************
*** 536,551 ****
   * This function was ported from the SNA R package.
   * \param graph A graph object.
   * \param res Pointer to an initialized matrix object. This will
!  *        contain the result and will be resized if needed.
!  * \param niter The number of iterations to perform.
   * \param sigma Sets the base standard deviation of position
!  *        change proposals. 
   * \param initemp Sets the initial temperature for the annealing.
!  * \param coolexp The cooling exponent of the annealing.
   * \param kkconst The Kamada-Kawai vertex attraction constant.
!  * \param use_seed Boolean, whether to use the values cupplied in the \p res 
!  *     argument as the initial configuration. If zero then a random initial 
!  *     configuration is used.
   * \return Error code.
   * 
   * Time complexity: O(|V|^2) for each
--- 536,557 ----
   * This function was ported from the SNA R package.
   * \param graph A graph object.
   * \param res Pointer to an initialized matrix object. This will
!  *        contain the result (x-positions in column zero and
!  *        y-positions in column one) and will be resized if needed.
!  * \param niter The number of iterations to perform.  
!  *        Typical value: 1000.  
   * \param sigma Sets the base standard deviation of position
!  *        change proposals.  
!  *        Typical value: (number of vertices)/4.
   * \param initemp Sets the initial temperature for the annealing.
!  *        Typical value: 10.
!  * \param coolexp The cooling exponent of the annealing.  
!  *        Typical value: 0.99.
   * \param kkconst The Kamada-Kawai vertex attraction constant.
!  *        Typical value: (number of vertices)^2
!  * \param use_seed Boolean, whether to use the values supplied in the
!  *        \p res argument as the initial configuration. If zero then a
!  *        random initial configuration is used.
   * \return Error code.
   * 
   * Time complexity: O(|V|^2) for each

reply via email to

[Prev in Thread] Current Thread [Next in Thread]