I'd like to delete collums that start with a value that already has been in the matrix

This is the matrix I'm facing;

     1     1     1     1     4     5     5     7    14    14    14    14    14    15    15    15    17   

     4     3     2     1     1     2     1     1     5     4     3     2     1     3     2     1     2     

     1     1     1     1     0     1     1     1     1     1     1     1     1     3     2     1     2

And this is the matrix I want to become:

     1    4     5     7    14     15    17    

     4    1     2     1     5     3     2     

     1    0     1     1     1     3     2

So I want to delete the collums, that have a value on the first row that already has passed in the first row... No idea how

It would be great to keep it as simple as possible

Answer 1

[Titus Edelhofer]

Hi Benoit,

what about this solution? Assuming your matrix is named A,

[~,idx] = unique(A(1,:));
Anew = A(:, idx);

Titus

Answer 2

[Walter Roberson]

row1 = A(1,:);
[urow1, row1order] = unique(row1);
[urow1indices, order_reorder] = sort(row1order);
selected_subset = A(:,row1order(order_reorder));

This can be simplified if it is guaranteed that the elements in the first row are in non-decreasing order.

[Benoit Cuyvers]

How would that simplifcied version look like?

[Titus Edelhofer]

Note, that in recent versions of MATLAB unique has the parameter setOrder which is sorted per default. No need to sort the order of indices, simply indexing should do.