Finding the row and column number in a matrix

for a matrix [1 2 3 4 6 10 7 5 9] mXn matrix-- how is it that i can find the min or max element and then find the row number and the column number for further use in calculations

[Meryem]

You can have an answer with a few lines of code which is:

    %you have ndata matrix
    [r,c] = size(ndata);    %get row and column values of data matrix
    fprintf('\nRow of data matrix is: %d' ,r);      %print number of row
    fprintf('\nColumn of data matrix is: %d ' ,c);  %print number of column

Answer 1

[Jan]

data = rand(5, 3);
[maxNum, maxIndex] = max(data(:));
[row, col] = ind2sub(size(data), maxIndex);

Another less compact approach finds the max values for each column at first:

data = rand(5, 3);
[maxNumCol, maxIndexCol] = max(data);
[maxNum, col] = max(maxNumCol);
row = maxIndexCol(col);

Please read "help max" also.

[Paulo Silva]

+1 vote

[Matt Fig]

This is the preferred method.

[Raghuram]

thank you :)

[Raghuram]

the method is not working for large sized matrices like i have to deal with a 30X600 matrix

[Matt Fig]

What do you mean "not working?" Are you saying that this approach gives the wrong result? Is there an error? If so, what error?

[Jan]

@Raghuram: The method works properly for matrices of any size (tested with 40000x40000). If you get an error, it is caused by another problem. Please post the error message or explain the difference bewteen the results and your expectations.

[Raghuram]

This was the code i used, F is a matrix that keeps increasing its size through columns that is 30X1 to 30 X 600

[minNumCol minIndexCol] = min(f(:));

[minNum col] = min(minNumCol);

row = minIndexCol(col);

the program is stopping at a point where the values are

row=32

col=1

minIndexCol=32

I have to find the least value and its row and column numbers

[Raghuram]

?? Index exceeds matrix dimensions.

Error in ==> PSO1stage5double at 156

pggbest(i)=pg(row,col);

this is the error

[Matt Fig]

Raghuram, look carefully at the two different codes Jan gave you, then look at the code you are using. Your code is not the same as either method! You have made a hybrid of the two different methods, but that doesn't work as you found out.

[Raghuram]

I used the first method only this time

the code was

[minNum1, minIndex1] = min(f(:));

[row1, col1] = ind2sub(size(f(:)), minIndex1);

error:

??? Attempted to access pg(32,1); index out of bounds because size(pg)=[30,2].

Error in ==> PSO1stage5double at 155

pggbest(i)=pg(row1,col1);

is there any mistake in the code, the notations i mean

[Raghuram]

I used the first method only this time

the code was

[minNum1, minIndex1] = min(f(:));

[row1, col1] = ind2sub(size(f(:)), minIndex1);

error:

??? Attempted to access pg(32,1); index out of bounds because size(pg)=[30,2].

Error in ==> PSO1stage5double at 155

pggbest(i)=pg(row1,col1);

is there any mistake in the code, the notations i mean

[Walter Roberson]

Don't use size(f(:)), use size(f)

[Raghuram]

yea it works now, thank you :)

Answer 2

[Rajashree Jain]

[val row]=max(A(:));

[val col]=max(A(row,:));

[val row col];

[Jan]

This will not work. In the first line you can get e.g. the last element of A as maximum, then "row==numel(A)". Then "A(row, :)" will fail.

Answer 3

[Amey]

The first answer given by Jan Simon is absolutely right and the most efficient way to do it.