Services & Resources / Wolfram Forums / MathGroup Archive
-----

MathGroup Archive 2008

[Date Index] [Thread Index] [Author Index]

Search the Archive

Re: Problem with Mesh for MatrixPlot

  • To: mathgroup at smc.vnet.net
  • Subject: [mg89803] Re: Problem with Mesh for MatrixPlot
  • From: "David Park" <djmpark at comcast.net>
  • Date: Sat, 21 Jun 2008 05:29:00 -0400 (EDT)
  • References: <g3fvte$krt$1@smc.vnet.net>

Does using Normal on the SparseArray give what you want?

n1 = 13; n2 = 15;
ntot = n1 n2;
mymat = SparseArray[{Band[{1, 1}] -> 0.87, Band[{1, 2}] -> -0.3,
     Band[{2, 1}] -> 0.3}, {ntot, ntot}] // Normal;
MatrixPlot[mymat,
 Mesh -> {Table[n2*i, {i, 1, n1 - 1}], Table[n2*i, {i, 1, n1 - 1}]}]

-- 
David Park
djmpark at comcast.net
http://home.comcast.net/~djmpark/


"Thibaut Jonckheere" <Thibaut.Jonckheere at cpt.univ-mrs.fr> wrote in message 
news:g3fvte$krt$1 at smc.vnet.net...
>
> Dear group,
>
> I have a large matrix, which has a block structure. In a MatrixPlot of
> the matrix, I want to show the blocks by plotting mesh lines at the
> block boundaries only. However, this seems to work well only for small
> matrices, and not for larger matrices.
>
> Here is a simple example showing the problem:
>
> n1 = 3; n2 = 5;
> ntot = n1 n2;
> mymat = SparseArray[{Band[{1, 1}] -> 0.87, Band[{1, 2}] -> -0.3,
>    Band[{2, 1}] -> 0.3}, {ntot, ntot}];
> MatrixPlot[mymat,
> Mesh -> {Table[n2*i, {i, 1, n1 - 1}], Table[n2*i, {i, 1, n1 - 1}]}]
>
>
> With these small values of n1 and n2, the result is as expected,showing
> the 5x5 blocks. However, when I increase the values of n1 and n2 (for
> example n1=13 and n2 =15), the meshlines are not shown on the whole
> matrice width (they don't go further than approx. 90).
>
> Note that the problem is also visible with the simple command:
> MatrixPlot[mymat, Mesh -> All].
>
>
> Is there something I am doing wrong, or is this a bug in the Mesh option
> of MatrixPlot ?
>
> Thank you for your help.
>
>
> Thibaut Jonckheere
>
> NB: I use Mathematica 6.0.1.0 on Win Xp (32 bit), and also on Linux, and
> the problem is present in both cases.
>
> -- 
> ----------------------------------------------------------------------------
> Thibaut Jonckheere
> Centre de Physique Theorique
> Campus de Luminy, case 907
> 13288 Marseille cedex 9
> tel: (+33).4.91.26.95.36
> email: jonckhee at cpt.univ-mrs.fr
> web: http://www.cpt.univ-mrs.fr/~jonckheere
> ----------------------------------------------------------------------------
>
>
> 



  • Prev by Date: Re: solving for 2 angles
  • Next by Date: Re: WorldPlot and excel data
  • Previous by thread: Re: Re: simple iteration question
  • Next by thread: Re: Problem with Mesh for MatrixPlot