spyx.m

Same as SPY function but with user defined axes.

Description

Syntax:

        spyx(x,y,S)

This programs uses the SPY function, which is used to plot marks on the non zero matrix elements (even on NaN's), but uses the user-specified axes (x,y) instead of rows and columns indexes (check the screenshot where

S = [
5.6 9.9 1.8 9.1 8.9 0 0 7.5 6.9 0
8.5 5.4 3.6 1.0 0.7 0 0 0 0 0
3.5 7.1 0 7.5 0 0 0.8 0 4.0 8.9
4.5 NaN NaN 7.4 0.5 3.3 7.8 0 0.6 0
0.5 NaN NaN 0 0 0 0 0 7.8 0
1.8 0 1.8 0 0 0.6 0 5.3 3.4 0
6.6 0 2.1 6.0 0 3.0 1.1 7.3 6.1 0
3.3 7.6 9.1 0 NaN 0 NaN 0 7.4 0.9
  0 8.2 6.8 1.3 0 5.1 0 7.8 0 0
1.2 1.0 4.7 0 0 0 0 2.9 0 NaN
];

• Acknowledgements: This submission has inspired the following:; imagescnan.m v2.1 (Aug 2009)
• MATLAB release: MATLAB 7.5 (R2007b)

[Dimitri Shvorob]

Sorry, but I don't quite get the point. The axes in SPY really have to be the way they are, to show row number and column number; what is the use of a different axis scaling?

Your example somewhat complicates matters with the extra mark-NaN-elements functionality.

Let me point you to this file:

http://www.mathworks.com/matlabcentral/fileexchange/loadFile.do?objectId=16888&objectType=FILE

[Carlos Adrian Vargas Aguilera]

Hi Dimitri:
  
Yes, MATLAB's SPY or yours are used to get the sparsity of a matrix X(i,j), but mine is to get the sparsity of a surface z(x,y), i.e., the axes have a PHYSICAL MEANING, but what is most important, one can look the sparsity OVER a held surface (like in the example), which CANNOT be done with the others SPYs functions.
  
Besides, my example in fact are two: one to look for zeros and the other to look for NaN's; so, simple versatility.

Cheers!