For loop, exceeded dimension error.

Hi, this is most certainly an easy fix on which I seem to got myself stuck.
The loop should compare elements of an array and if it finds that they are equal it shall erase that row, next it shall recompute the size of the array and redo the comparison for the next element:
for ii=1:N
if (ii>1) && (ii<N)
i3=ii+1;
for jj=i3:N
if DD(ii,1)==DD(jj,1)
DD(jj,:)=[];
[N,M]=size(DD);
end
end
end
end
I get the following error:
??? Index exceeds matrix dimensions.
Error in ==> VB_CompDin2Darray at 29
if DD(ii,1)==DD(jj,1)
A pointer would be great. Cheers.

 Respuesta aceptada

Sean de Wolski
Sean de Wolski el 20 de Jul. de 2011
Simply run the loop backwards so row 4 doesn't become row 3 when row 3 is deleted.
e.g:
for jj = N:-1:i3
%delete stuff
end
Now of course you're still best off (for speed) building an index vector and doing the whole delete once at the end.
idx = false(N-i3,1);
for jj = i3:N
if stuff
idx(jj) = true;
end
Mat(idx,:) = [];
end

4 comentarios

Victor Buza
Victor Buza el 20 de Jul. de 2011
Great! Thanks. What was the problem with row 4 becoming 3? I mean, it was still resizing and comparing the element, wasn't it?
Jan
Jan el 20 de Jul. de 2011
@Victor: The loop "for jj=i3:N" runs until the inital value of N, even if N is modified inside the loop. In most other programming languages modifying the limit inside the loop effect the number of iterations.
Sean de Wolski
Sean de Wolski el 20 de Jul. de 2011
If you delete row 3; what happens to row 4? It becomes row 3! Now your matrix size is smaller, but the for-loop still wants to go to the ORIGINAL N.
Sean de Wolski
Sean de Wolski el 20 de Jul. de 2011
Use unique, as Jan has suggested, anyway.

Iniciar sesión para comentar.

Más respuestas (1)

Jan
Jan el 20 de Jul. de 2011
Simple and efficient for large matrices:
[dummy, index] = unique(DD(:, 1));
DD = DD(index, :);

Categorías

Más información sobre Loops and Conditional Statements en Centro de ayuda y File Exchange.

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by