sortrows based on previous sort results

Here is the thing

I have a large matrix and I need to sort that based on some columns:

B is the large matrix

A=sortrows(B,[1 4 9 10]);%Just a sample input vector

Its OK so far.But then in next step (I'm iterating this function) I will need this:

A=sortrows(B,[1 4 9 10 16]);%Notice I just add a new column to input vector

And same story in next iterations

So my question is how can I use previous sort results in each iteration?

Please consider that the sequence of input vector doesn't matter for me.And if it is possible please give me a note on this sortrows algorithm.

Edit------------ Can someone please explain sortrows algorithm?

Thanks in advance.

[Image Analyst]

Why do you even need to do the first sort? The second sort will give you the same results regardless of whether you had a sorted or unsorted array to begin with. So doing the first sort is useless. Is there some reason why you needed/wanted to do the first sort?

[amir]

Yes this is going to be a fitness function .I am evaluating the subset[1 4 9 10] in iteration t based on some fitness function and then in iteration t+1 I need to evaluate the subset[1 4 9 10 16] .In evaluation phase I am using sortrows as a preprocess then the rest.So if I can use sorted matrix in previous iteration it will save me god knows how much time!!

Answer 1

[amir]

This is the function that i came up with might be useful for someone in future :

    function [ndx]=sort2(x,cols,oldIndex)

    x=x(:,cols);

    [m,n] = size(x);

    if nargin<3

    ndx = (1:m)';

    else

    ndx=oldIndex;

    end

    for k = n:-1:1

        [ignore,ind] = sort(x(ndx,k));

        ndx = ndx(ind);

    end

result--------

    a=sort2(IS,[1 4 9 10]);
    a=sort2(IS,16,a);
    b=sort2(IS,[16 1 4 9 10]);

    isequal(a,b)=1

[Azzi Abdelmalek]

Useful for what?

Answer 2

[Azzi Abdelmalek]

B=sortrows(B,[1 4 9 10])
B=sortrows(B,[1 4 9 10 16])

[amir]

Yes thanks but you didn't get the question I want to use first sort result in next one not repeat sorting all those columns.Its a matter of time efficiency.

[Azzi Abdelmalek]

This is not obvious to find faster

[amir]

as far as i releaized about sortrows algorithm this two must have the same result :

%C is original matrix

A=sortrows(C,[16 1 4 9 10]);
B=sortrows(C,[1 4 9 10]);
B=sortrows(B,16);

But its not what is your opinion?

[Azzi Abdelmalek]

In this case it's true, but it's not what you are asking. Check this case

C=randi(5,20) % Example
A=sortrows(C,[1 4 9 10 16]);
B=sortrows(C,[1 4 9 10]);
B=sortrows(B,16);
isequal(A,B)

A and B are different

[amir]

Please consider that sequence does not matter for me so [1 4 9 10 16] , [ 10 16 9 14] , ... any permutation is same thing.I just want to achieve a matrix sorted by any of these vectors based on a matrix that has already been sorted in last iteration based on [1 4 9 10] .I can not explain it better please excuse me for my poor english.