Checking a matrix for duplicates in specific row, removing the respective columns

So, I try to explain this.

I have a two-row matrix of values, [x;y], f.e.

x=[1 2 2 3 4 5 6 6 7]
y=[1 2 3 4 5 6 7 7 8]

being merged into the matrix:

d= [1 2 2 3 4 5 6 6 7;
    1 2 3 4 5 6 7 7 8]

I then want to check this matrix for repeats in the first row & remove the respective columns, while saving the x values of the columns being removed as a seperate vector. So Output should be something like:

d=[1 3 4 5 7;
   1 3 4 5 8]
x_cut=[2 6]

It should be noted that this would have to scan for multiple repeats of different values, as shown above.

Thank you.

Have a great day & stay safe

Claudius Appel

Answer 1

[Bruno Luong]

d= [1 2 2 3 4 5 6 6 7;
    1 2 3 4 5 6 7 7 8]

dd = diff([nan,d(1,:),nan])==0;
remove = dd(1:end-1) | dd(2:end)
x_cut = unique(d(1,remove),'stable')
d(:,remove) = []

[Claudius Simon Appel]

This works nicely, but I'd have a short follow up question, that is admittetly not within the original question, because I didn't realise it was necessary:

If I, instead of just conserve x_cut, wanted to cut and preserve the pairs, i.e. having as output:

cut=[2 2 6 6;
     2 3 7 7]   

I am having trouble figuring out how excactly that would be done.

I was experimenting with extracting y_cut as is done with x_cut, but that leaves me with the problem that I don't know how to merge those into the right vectors again:

 x_cut = unique(d(1,remove),'stable')
%  yields
 x_cut =

     2     6

 y_cut=unique(d(2,remove),'stable')
%   would yield
 y_cut =

     2     3     7

I wouldn't know how to merge these together to form:

cut=[2 2 6 6;
     2 3 7 7]

[Bruno Luong]

d= [1 2 2 3 4 5 6 6 7;
    1 2 3 4 5 6 7 7 8]

dd = diff([nan,d(1,:),nan])==0;
remove = dd(1:end-1) | dd(2:end);
cut = d(:,remove)
keep = d(:,~remove) % rename keep to d if you like

Or if yoy want to get cut from x_cut

dd = diff([nan,d(1,:),nan])==0;
remove = dd(1:end-1) | dd(2:end)
x_cut = unique(d(1,remove),'stable')
cut = d(:,ismember(d(1,:), x_cut))
d(:,remove) = []

[Claudius Simon Appel]

Thank you.

That works perfectly.

Have a nice day & stay healthy.