Vectorizing Arithmetic Operations

Question

I am trying to improve the performance of my code by converting some iterations into matrix operations in Matlab. One of these is the following code and I need to figure out how can I avoid using loop in the operation.

Here gamma_ic & bow are two dimensional matrices. c & z are variables set from outer iterations.

for z=1:maxNumber,
    for c=1:K,
        n = 0;
        for y2=1:number_documents,
            n = n+(gamma_ic(y2,c)*bow(y2,z));
        end
        mu(z,c) = n / 2.3;
    end
end

Appreciate your assistance.

Edit. Added The loop for c and z. The iteration goes on till the maximum indices in gamma_ic & bow. Added mu which is another two dimensional matrix to show usage of n.

Do c and z iterate through all possible indices for gamma_ic and bow respectively? I think it is worth showing how they iterate, and how you are using n. Also any thoughts or understanding you have about vectorisation (it may be easy or hard to vectorise your code, but it would be good to understand what is blocking you and try to fix that) — Neil Slater
– Neil Slater, Commented Oct 27, 2014 at 22:43
If you share the loop codes for c and z, we might be able to help to get rid of those loops too. But for the given code, you could try out - n = sum(gamma_ic(1:number_documents,c).*bow(1:number_documents,z)) — Divakar
– Divakar, Commented Oct 27, 2014 at 22:48
I have updated the code with relevant portions of the outer loops. — Ashwini Khare
– Ashwini Khare, Commented Oct 27, 2014 at 22:50
What exactly or how exactly are you getting maxNumber and K? — Divakar
– Divakar, Commented Oct 27, 2014 at 22:52
maxNumber and K are constants being passed on to the function — Ashwini Khare
– Ashwini Khare, Commented Oct 27, 2014 at 22:53

Divakar · Accepted Answer · 2014-10-27 22:58:11Z

1

This should work for you to get mu, which seems to be the desired output -

mu = bow(1:number_documents,1:maxNumber).'*gamma_ic(1:number_documents,1:K)./2.3

answered Oct 27, 2014 at 22:58

Divakar

222k19 gold badges273 silver badges374 bronze badges

Sign up to request clarification or add additional context in comments.

1 Comment

Ashwini Khare Over a year ago

I'm a bit confused here, will the dimensions agree to perform element wise operation? Note that K may not be equal to maxNumber. Edit: got it, that's why you are taking the transpose I suppose? Ignored it initially

Collectives™ on Stack Overflow

Vectorizing Arithmetic Operations

1 Answer 1

1 Comment

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

1 Comment

Your Answer

Sign up or log in

Post as a guest

Linked

Related