To find maximum value of any matrix without using built-in max()

 function [max_value, index] = mymax(x)
  [r,c]=size(x);
max_value_row=x(1,1)
max_value_col=x(1,1)
for i=1:c
    if x(1,i)>max_value_row
        max_value_row=x(i,1);
    end
    for k=1:r
        if x(k,1)>max_value_col
            max_value_col=x(k,1);
        end
    end
end
        if max_value_row > max_value_col
            max_value = max_value_row
        else
            max_value = max_value_col
            index = find(max_value)
        end

THis is what i have tried so far but the value found is wrong, please help me

Answer 1

[Stephen23]

function [val,idx] = mymax(arr)
val = arr(1);
idx = 1;  
for k = 2:numel(arr)
    if arr(k)>val
        val = arr(k);
        idx = k;
    end
end

This return the linear index, which is more useful than the subscript index.

[Kae-Shyang Chen]

Thanks alot for the answer, but i havent learnt numel() so far, what is the function for? Is there any alternative method without using the numel()? THANKS AGAIN

[Stephen23]

numel simply counts the total number of elements in any array:

https://www.mathworks.com/help/matlab/ref/numel.html

Alternative: prod(size(...))

[Kae-Shyang Chen]

Alright i will try to figure out your code ,thanks

[Stephen23]

@Kae-Shyang Chen: you will need to know about linear indexing:

https://www.mathworks.com/help/matlab/math/matrix-indexing.html

https://blogs.mathworks.com/steve/2008/02/08/linear-indexing/

[Kae-Shyang Chen]

I have a clearer picture on linear indexing now. Thank you so much in helping me!

Answer 2

[Guillaume]

Forget about the code for now. What you've written is fundamentally wrong at the algorithmic level for many reasons. You need to spend more time thinking about what the steps of your algorithm should be.

If you can't spot the issues with your code just by reading it, then I suggest you use the debugger to execute your program one line at a time and observe what happens. When you do that, you'll quickly see that your code

• only look at row 1 and column 1 of tha matrix and ignore all other rows/columns
• calculate the maximum of the 1st column c times

Questions that should help you come up with a better algorithm:

• Why do you need to iterate over columns or rows? Can't you just iterate over all the elements just once using linear indexing
• Wouldn't the maximum calculated by row and the maximum calculated by column be the same?

[Kae-Shyang Chen]

hi, thank you for your guidance.. What i was trying to do is to find the maximum value of the entire matrix instead. I tried the following approach: 1. To find max.value of columns and rows 2. To compare both value and decide the max.value for the entire matrix

Answer 3

[Kae-Shyang Chen]

 I tried this and it works! @Guillaume @Stephen Cobeldick
  [r,c]=size(x);
max_value=x(1,1);
for k=1:r
    for i=1:c
    if x(k,i)>max_value
        max_value=x(k,i);
    end
    end
end