How can I use "IF" statement to prevent adding same element to an array?

Hello,

So I have a matrix A

A = [ 1 2 3;
      2 3 4;
      4 5 6]

I create another matrix B where B =[] and the element in B will be added from A. Matrix B becomes

  B = [ 1 2;
        2 3;
        3 1;
        2 3;
        3 4;
        4 2]

The forloop technically go through each row of A to pick the elements for each row of B. B is nx2 matrix. As we can see, "2 3" is repeated. I want to make an If statement that prevent adding existing row in B.

Please help, thank you so much.

Answer 1

[the cyclist]

Would it be acceptable to just remove duplicated rows after-the-fact? If so, you could do

B = unique(B,'rows');

It seems likely that this would be more efficient than checking each row against all prior rows, while B is being constructed.

[kiet tran]

Thank you for the advice. But I forgot to say that if there is "1 2" I can't have either "1 2" or "2 1". Unique command only remove "1 2" for existing "1 2" in B. Sorry for not being clear in the question.

[the cyclist]

Would the following be acceptable?

B = unique(sort(B,2),'row');

This would reduce

[1 2;
 2 1]

to just

[1 2]

which seems to be what you want. However, it would also change just

[2 1]

to

[1 2]

even though [1 2] was never present in the first place. Is that OK?

[kiet tran]

Thank you for the advice. So the 'sort' command will organize the rows in B to be somewhat the same to the existing ones, and then 'unique' will remove them, right? I tried it in my code and it works perfectly. I will try more different input to see if it holds the goal consistently. Thank you very much !

[the cyclist]

I don't know what you mean by "somewhat the same to the existing ones". So, let's be specific.

sort will take each row, and put it in ascending numerical order. Therefore the row [2 1] and the row [1 2] will both become [1 2]. That way, unique keeps only one of those rows.

[kiet tran]

Thank you sir, that totally works and very simple and effective.