How to delete repeated values in a vector as well as values in its corresponding vector

Hello there!

I am attempting to delete repeated values in a vector and then delete values in a vector of the exact same length at the same index. For instance, say I have the following two vectors:

 n_vector = [1 2 3 5 5 5 5 7 6 6]; 
 t_vector = [1 2 3 4 5 6 7 8 9 10];

The n vector has the repeating values 5 and 6, so I need a bit of code that transforms these vectors into this:

 n_vector = [1 2 3 5 7 6]; 
 t_vector = [1 2 3 4 8 9];

I tried to do this with the following code, however it doesn't work if values repeat more than twice:

 u = 1; 

 while u > 0  

 for i = 1:length(n_vector)

    if length(n_vector(i:end)) > 1

        if n_vector(i) == n_vector(i+1)

            n_vector(i+1) = [];  
            t_vector(i+1) = [];

            u = 1; 

        else

            u = 0; 

        end

    elseif length(n_vector) == 1 

       u = 0; 

    else 

        break
    end

 end

 end

Is there a way that I can fix this code so it can delete values no matter how many times it repeats, or do I need to do try something else all together?

Sincerely, Colin

Answer 1

[Guillaume]

I don't know about your code, but this will be a lot more efficient than what you're trying to do:

repeats = diff(n_vector) == 0; 
n_vector(repeats) = [];
t_vector(repeats) = [];
%all done!

[Colin Lynch]

I am simply amazed you thought of that so quickly! Thank you so much!!!

[OCDER]

Here's a variation of the Answer that's ~3.5 faster. In Matlab, deleting elements from a vector tends to be slower than taking a subset of a vector.

 a = sort(randi(6, 1, 1E7)); 
 b = 1:1E7;

 n_vector = a; t_vector = b; 
 tic
     repeats = [1 diff(n_vector)] == 0; 
     n_vector(repeats) = [];
     t_vector(repeats) = [];
 toc   % 0.4392 s

 n_vector = a; t_vector = b; 
 tic
     nonrepeats = [1 diff(n_vector)] ~= 0; 
     n_vector = n_vector(nonrepeats);
     t_vector = t_vector(nonrepeats);
 toc   % 0.1383 s