could anyone help me how to check the idx position containing value is not equal to zero

my code generates idx =[2 5 7 12 15]

Before i want to proceed i need to check whether the above mentioned position contain zero or non zero values.

If it contains non zero values i can proceed or else i need to change the idx position till i get all the position containing non zero values.

Could anyone please help me on it.

Answer 1

[Thorsten]

To check if any value of idx is zero, you can use

any(idx == 0)

[jaah navi]

I am having matrix a=[4.1288 5.2574 0 4.9757 3.0069;

0 3.7672 4.7411 3.4989 0;

4.7132 0 4.0136 0 4.9751]

with respect to the matrix the idx= 2 6 8 12 15 gets generated in my code.

But i want to have idx in such a way that all idx should give the non zero position.

But in my code idx=2 6 12 points out zero.

So first i need to check idx points to zero if yes then i need to get another idx which contains only non zero values position.

could you please help me on this.

Answer 2

[Steven Lord]

a(idx) will give you the elements in a whose linear indices are stored in idx. You can check if they are exactly equal to 0 using the == operator.

a = [1 2 0 3 4; 0 5 6 7 0; 8 0 9 0 10];
idx = [2 6 8 12 15];
zeroLocations = a(idx) == 0;
indicesCorrespondingToZero = idx(zeroLocations)

If you need to modify idx, use zeroLocations to index into idx on assignment. Here I'm going to overwrite them with NaN but you could do computations on them, remove them, etc.

idx(zeroLocations) = NaN

[jaah navi]

ok.But i want to have idx in such a way that all idx locations should not be equal to zero.

Could you please help me on this.

[Rik]

Do you want to change idx, or the matrix?

[jaah navi]

i want to chnage idx.

[Rik]

Then you can use this:

a = [1 2 0 3 4; 0 5 6 7 0; 8 0 9 0 10];
idx = [2 6 8 12 15];
new_idx = find(a ~= 0);

[jaah navi]

ok.

But what i actually need is, new_idx should contain only 5 values each referring to one column.

when i used the command line new_idx = find(a ~= 0),it generates all non zero values present in a.

could you please help me on this.

[Rik]

Since you don't care which element is chosen, you can use the second output of the max function.

a = [1 2 0 3 4; 0 5 6 7 0; 8 0 9 0 10];

%if a might contain negative values, use this instead:
%[val,row_idx]=max(abs(a),[],1);

[val,row_idx]=max(a,[],1);
if any(~nnz(val))
    error('some cols have only zeros')
end
idx=sub2ind(size(a),row_idx,1:size(a,2));

[jaah navi]

Thanks a lot.