Delete all values in a row after the first value

Hi everyone

I have a matrix with 3000 rows and 4 columns. the values in the matrix are just one and zero. The sum of each row can just be one. In some rows the sum is 2. What I want now is to delete all 1 after the first 1. So in my matrix in each row I just want the first vaue 1 and all the following values in this matrix should be 0.

For example: row number 345: 0001000 1 0000 change to 0001000 0 0000

thank you for your help.

Answer 1

[KSSV]

m = 10 ; n = 5 ;
A = randi([0 1],m,n) ;
B = zeros(size(A)) ;
for i = 1:m
    idx = find(A(i,:)==1) ;
    B(i,idx(1))= 1;    
end

There might be more elegant solution.

[FC93]

Thank you for your help. When I replace A by my original Matrix and I create a new matrix B and then I do the code I get an error "Index exceeds matrix dimensions". Now I don't really know why?

[KSSV]

What you have taken m as? Put your code here.

[FC93]

m=5540 , n=4

FOMC_X_try = zeros(size(FOMC_X));

for i = 1:m

    idx = find(FOMC_X(i,:)==1) ;
    FOMC_X_try(i,idx(1))= 1;

end

That is now my code. FOMC_X is my real matrix and FOMC_X_try is what I genereated.

[KSSV]

m=5540 , n=4

FOMC_X_try = zeros(size(FOMC_X));

for i = 1:m

      idx = find(FOMC_X(i,:)==1) ;
      if ~isempty(idx)
          FOMC_X_try(i,idx(1))= 1;    
      end
  end

If there are no ones in a particular row, idx will be empty. So is the error.

[FC93]

I replaced the old code with this one. Now it works. thank you very much for your help and your time.

Answer 2

[Walter Roberson]

B = cumsum(~cumprod(~A,2), 2)  <= 1;

If I worked it out properly as I fall asleep...