how to sum all row values in a matrix except the ith row?

i have a matrix X=rand([6,3,6]);

I want all the row value sum=1 except the ith row? this ith can be 4/5 according to user's choice? all the matrices row sum will be 1 except the ith row in every matrices? Here i dnt want to exclude the ith row. the elements of the ith row will be there as it is. could someone help me in solving this problem

Answer 1

[DGM]

There are probably other ways, but here's a way (if I understand the question correctly).

A = rand(6,3,6);
dnn = 5; % do not normalize this row

Adn = A(dnn,:,:); % save this row
A = A./sum(A,2); % normalize
A(dnn,:,:) = Adn; % replace the row

sum(A,2) % show that the row sums are 1 except where specified

ans = 
ans(:,:,1) =

    1.0000
    1.0000
    1.0000
    1.0000
    1.3225
    1.0000


ans(:,:,2) =

    1.0000
    1.0000
    1.0000
    1.0000
    1.3834
    1.0000


ans(:,:,3) =

    1.0000
    1.0000
    1.0000
    1.0000
    1.9078
    1.0000


ans(:,:,4) =

    1.0000
    1.0000
    1.0000
    1.0000
    1.4412
    1.0000


ans(:,:,5) =

    1.0000
    1.0000
    1.0000
    1.0000
    1.3090
    1.0000


ans(:,:,6) =

    1.0000
    1.0000
    1.0000
    1.0000
    1.3265
    1.0000

That will normalize all rows except the specified row. I should point out that while this method discards the normalized results for one row, it's faster than indexing

[chan]

Thank you it helps me after modifying the code. when i run your code some error occur in this step

A = A./sum(A,2);

[DGM]

Not knowing what was modified or what the error is, I'm going to guess that it's an error about mismatched array dimensions. If that's the case, i'm going to have to guess maybe you're running an older version prior to the introduction to generalized implicit array expansion. If that's the case, you can do this:

%A = A./sum(A,2); % normalize (R2016b or newer)
A = bsxfun(@rdivide,A,sum(A,2)); % normalize (pre-R2016b)

If that's not the case, you'll have to let me know what's been changed and what the specific error message is.