How to remove repeating elements from an array

Hi Guys,

I have two matrices X and Y of the same size. Each element in Y has a corresponding element in X. There are some repeating values in X (and correspondingly in Y as well). Is there any way to remove these elements from the matrices.

Thanks, NS

[Hesham Hendy]

If I want to keep one value of repeated values without sorting, what can I do ???

[Image Analyst]

Start a new question and give a small example.

Answer 1

[UJJWAL]

Hi NS,

Ok. Now I understand what you want. ok. There is a way to do this. Take a look at the code below :

a = randi(10,[1,20]);

[b,m1,n1] = unique(a,'first');

[c1,d1] =sort(m1);

b = b(d1);

In this way your problem is solved. similarly you can also do so for the other matrix. Hope this helps...

Happy to help

UJJWAL

[Andrei Bobrov]

[a b] = unique([X;Y]','rows','first');

xyb = sortrows([a b],3)';

X = xyb(1,:);

Y = xyb(2,:);

[NS]

Thanks. I figured it that day itself. :)

Answer 2

[UJJWAL]

Use the unique function. Take the following example :-

a = randi(10,[1,20]) will display a 1 by 20 matrix of pseudorandom integers with uniform distribution in the range 1:10 . Obviously there will be repeated elements in the matrix. Suppose now you create a second matrix p = sin(a). So there is a mapping defined. Now type unique(a). It will give you only the unique values in a that is it will not display repeated values of the same element. Now type unique(p) : It will similarly display the unique values in p.

Hope it resolves your doubt

Best Wishes

Ujjwal

[Jan]

If you want to remove elements on the same positions in both matrices, you can use the 2nd and 3th output of unique. See "help unique".

[NS]

Hi Ujjwal,

I dont want the order of the elements to be rearranged. When I use unique, the elements are arranged in an ascending order.

Eg. if X=[0 13 18 22 22] and Y=[0 1 -1 3 3]

I want my output to be as X=[0 13 18 22] and Y=[0 1 -1 3]

How do I do that

Answer 3

[Fangjun Jiang]

For duplicated x-y pairs, if you want to keep the last one:

X=[0 13 22 18 22 22 10];
Y=[0 1 3 -1 3 3 -1];
A=[X;Y]';

[UniXY,Index]=unique(A,'rows');

DupIndex=setdiff(1:size(A,1),Index);
A(DupIndex,:)=[];
X=A(:,1)'
Y=A(:,2)'

X =

       0    13    18    22    10

Y =

       0     1    -1     3    -1

If you want to keep the first one, use:

[UniXY,Index]=unique(A,'rows','first');

[Aidin Golrokh]

For a big matrix, this code will produce repeating numbers.

Answer 4

[SIVAKUMAR KARURNKARAN]

function B=adjacentmat(n,e) R=combntns(1:n,2); k=1:e; r=nchoosek(n,2); B=zeros(n,n,r,r);

for i=1:r for j=1:r R([i j],:)=R([j i],:); F=R(k,:); A=zeros(n,n); for m =1:e; A(F(m,1),F(m,2))=1; A(F(m,2),F(m,1))=1;

        end
        B(:,:,i,j)=A;

    end
end

return in this output has more duplicate matrices i want only distinct matrices. i need the code of comparing the matrices and find which is distinct

Answer 5

[Ismet Ozturk]

    function [ a ] = deleteRepetation( a )
    leng=size(a);tempa=zeros(1,2);
    for i=1:leng(1)
        tempa=a(i,:);
        tempa=sort(tempa);
        a(i,:)=tempa;
    end
    deletionList=[1];
    while size(deletionList)>0
        deletionList=[];leng=size(a);
        for i=1:leng(1)
            tempa=a(i,:);
            for j=i+1:leng(1)
                if tempa==a(j,:)
                    deletionList=[deletionList j];
                end
            end
        end

        a(deletionList,:)=[];   
    end

[Life is Wonderful]

Couple of things noted and suggestion added for the above code

Code point-1

tempa=sort(tempa);

Replace with

tempa=sortrows(tempa);

Code point-2

if tempa==a(j,:)          %might NOT always work for Tables

Replace with

if isequal(tempa,a(j,:))  % better implementation