Problem with unique command

Hello guys. I have a problem with unique command in the following simplified code. The problems are in the 4th iteration as the values of G aren't correct, IC values aren't the repetitions of IA and the resulted Good_path1 isn't a 16*4 matrix. However, the problems are resolved if the position of loops of alpha and beta is changed with the loop of s. Sorry for a lenghy code. I really appreciated if someone help me to solve the problem.

n_c_1=1;
thr_Wght=1e-50;
M=2;
TGponts_1=4;
% &&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&
K1_1 =[0.0552 - 0.1423i,  -0.0145 - 0.1465i,  -0.0652 - 0.1285i];
K01_1 = 0.0245 - 0.1175i;
% &&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&
Good_path=[0     0     0     0; 0     0     1     0; 1     0     0     0; 1     0     1     0];
R2=[0.9983 + 0.0000i   0.0047 + 0.0399i   0.0047 - 0.0400i   0.0017 + 0.0000i];

n_c1_1=n_c_1+1;  n_c2_1=2*n_c_1;
L_1=2*(n_c1_1);
for n=3:TGponts_1
    if n>n_c1_1
        Good_path(:,n_c1_1)=[];
        Good_path(:,L_1-1)=[];
%         Good_path
        [Dis_Good_path,IA,IC]=unique(Good_path,'rows','stable');
        Dis_R2=zeros(length(IA),1);
        for i=1:length(IA)
            G=find(IC==IA(i));
            G
            Dis_R2(i)=sum(R2(G));
        end
        Good_path=Dis_Good_path
        R2=Dis_R2
    end
    h=size(Good_path);
    h1=h(1);
    q=0; 
    nn_1=n-1; nn_2=2*n-2;
    s1=M.^2*h1;
    if n<=n_c1_1                 
        s2=2*n;
        Good_path1=zeros(s1,s2);
    else
        Good_path1=zeros(s1,L_1);  
    end
    R=zeros(1,s1);    
                
    for alpha=0:1
        for beta=0:1          

            for s=1:h1             
                           
                if (alpha==0 && beta==0) || (alpha==1 && beta==1)
                    T3=1;
                else
                    Self_int=-(alpha-beta).*(K01_1.*alpha-conj(K01_1).*beta);
                    %%&&&&&&&&&&&&&&&&&&&&&&&
                    if n<=n_c1_1
                        Sum_nlocl_int=0;
                        for m=n-1:1
                            a_prim=Good_path(s,m);
                            b_prim=Good_path(s,n-1+m);
                            nlocl_int=(K1_1(m).*a_prim-conj(K1_1(m)).*b_prim);
                            Sum_nlocl_int=Sum_nlocl_int+nlocl_int;
                        end
                    else
                        Sum_nlocl_int=0;
                        for mm=n_c_1:1
                            a_prim=Good_path(s,mm);
                            b_prim=Good_path(s,n_c_1+mm);
                            nlocl_int=(K1_1(mm).*a_prim-conj(K1_1(mm)).*b_prim);
                            Sum_nlocl_int=Sum_nlocl_int+nlocl_int;
                        end
                    end
                    Sum_nlocl_int=-(alpha-beta).*Sum_nlocl_int;
                    T3=exp(Self_int+Sum_nlocl_int);
                end
                R3=T3*R2(s);
                %&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&
                p=abs(R3);
                if p>thr_Wght
                    q=q+1;
                    if n<=n_c1_1
                        Good_path1(q,:)=[alpha,Good_path(s,1:nn_1),beta,Good_path(s,n:nn_2)];
                    else
                        Good_path1(q,:)=[alpha,Good_path(s,1:n_c_1),beta,Good_path(s,n_c1_1:n_c2_1)];
                    end
                    R(q)=R3;
                end
            end          
        end
    end
    %&&&&&&&&&&&&&&&&&&&&&&&&&&&&
    Good_path1(q+1:s1,:)=[];
    R(q+1:s1)=[];    
    Good_path=Good_path1
    R2=R;     
end 

[Geoff Hayes]

Davoud - please clarify what you mean by in the 4th iteration as the values of G aren't correct? What are you expecting the values to be?

[Davoud]

Thank you Geoff. I mean at n=TGponts_1=4, there are four Gs which their values should be 1 2 3 4, 5 6 7 8, 9 10 11 12, 13 14 15 16, because after removing columns of 2 and 4 of Good_path matrix, the first four rows are the same, the fifth to eighth, the ninth to twelfth, and thirteenth to sixteenth rows are the same, respectively.. Here is the shown result

Good_path =

0 0 0 0

0 0 0 1

0 1 0 0

0 1 0 1

0 0 1 0

0 0 1 1

0 1 1 0

0 1 1 1

1 0 0 0

1 0 0 1

1 1 0 0

1 1 0 1

1 0 1 0

1 0 1 1

1 1 1 0

1 1 1 1

G =

1

2

3

4

G =

0×1 empty double column vector

G =

0×1 empty double column vector

G =

0×1 empty double column vector

In the present case, Dis_R2 is correct only for first four rows but not for the 2nd, third, and fourth four rows. In this case

IA =

1

5

9

13

Therefore, IC should contains 1, 5, 9, and 13 but it is

IC =

1

1

1

1

2

2

2

2

3

3

3

3

4

4

4

4

Answer 1

[Raghunandan V]

Hi,

I think there is a confusion between the functions of IA and IC.

lets consider an example

A = [9 2 9 5];
[C, ia, ic] = unique(A,'stable')

the solution is

C = 1×3

9 2 5 (this is the matrix output expected)

ia = 3×1

1 (here 1 refers to the index of the matrix A which is placed in matrix C in index 1)

2 (here 2 refers to the index of the matrix A which is placed in matrix C in index 2)

4 (here 4 refers to the index of the matrix A which is placed in matrix C in index 3)

ic = 4×1

1 (here 1 says that the 1st element of the matrix A is the first element of Matrix C)

2 (here 2 says that the 2nd element of the matrix A is the second element of Matrix C)

1 (here 1 says that the 3rd element of the matrix A is the first element of Matrix C)

3 (here 3 says that the 4th element of the matrix A is the third element of Matrix C)

I am not able to understand what you are doing with the code. The code is very confusing. Please add some comments such that other people can understand.

[Davoud]

Thank you about the clarification. How can I find the index of repeated rows in the Good_path matrix? I have used the following loop.

for i=1:length(IA)

G=find(IC==IA(i));

G

Dis_R2(i)=sum(R2(G));

end

which seems not working here. Thank you.

[Raghunandan V]

This is what IC gives you. You can find the index of repeated numbers.

for i= 1:max(IC)
        Index = find(IC== i)
end

Try this. :)

[Davoud]

In fact, max(IC) and length(IA) are the same. right?

[Raghunandan V]

Yes