How to extract values from a matrix using a row - column matrix?

I have a matrix like

A =

    0.2180    0.5714    0.2476    0.3942    0.9253    0.5082
    0.3906    0.5881    0.6529    0.3806    0.3819    0.7947
    0.3381    0.0287    0.0652    0.3696    0.3735    0.1351
    0.7459    0.0911    0.0163    0.7531    0.3298    0.2050
    0.3690    0.5010    0.5201    0.6119    0.9038    0.1902
    0.4232    0.9183    0.6523    0.2392    0.4661    0.0019

and a matrix having row and column indices

loc =

     1     2
     2     5
     4     3
     6     6

I want the extract the values of locations (1,2), (2,5), (4,3) and (6,6) from matrix A when I do this

val = A(loc(:,1),loc(:,2));

it gives me

val =

    0.5714    0.9253    0.2476    0.5082
    0.5881    0.3819    0.6529    0.7947
    0.0911    0.3298    0.0163    0.2050
    0.9183    0.4661    0.6523    0.0019

I can use diag(val); to get the locations values but is there any way so that i could get values directly without getting that square matirx

val =

    0.5714
    0.3819
    0.0163
    0.0019

Answer 1

[Stephen23]

Yes! You can use sub2ind to achieve this:

A = [...
    0.2180    0.5714    0.2476    0.3942    0.9253    0.5082
    0.3906    0.5881    0.6529    0.3806    0.3819    0.7947
    0.3381    0.0287    0.0652    0.3696    0.3735    0.1351
    0.7459    0.0911    0.0163    0.7531    0.3298    0.2050
    0.3690    0.5010    0.5201    0.6119    0.9038    0.1902
    0.4232    0.9183    0.6523    0.2392    0.4661    0.0019];
loc = [...
     1     2
     2     5
     4     3
     6     6];
 X = sub2ind(size(A), loc(:,1), loc(:,2));

and then run this in the command window:

>> A(X)
ans =
  0.5714
  0.3819
  0.0163
  0.0019

[VIJENDRA]

Thanks Stephen Cobeldick

Answer 2

[Punit]

[nRows nCols ] = size(A);

A(nRows*(loc(2)-1)+loc(1))

[Guillaume]

This is what sub2ind does and I'd recommend using it as it works regardless of the number of dimensions of A.