Vectorize the following loop

Hi all,
I'm trying to vectorize the following loop to speed it up
c = cumsum(weights);
A = ones(1,n);
x = rand(1,n);
for i = 1:n
j = find(c > x(i) ,1,'first');
A(i) = j;
end
where weights is an array of doubles which sum to 1 and n <= size(weights).
Any help would be grand!
B

 Réponse acceptée

Oleg Komarov
Oleg Komarov le 13 Avr 2011

3 votes

n = 3e4;
weights = rand(n,1);
c = cumsum(weights/sum(weights));
A = zeros(1,n);
x = rand(1,n);
tic
for i = 1:n
j = find(c > x(i) ,1,'first');
A(i) = j;
end
toc
% WARNING: only if weights are monotonically increasing
tic
[B1,B1] = histc(x,c);
toc
tic
B2 = n-sum(bsxfun(@gt,c,x))+1; % Goes fast in out of memory
toc
isequal(A,B1+1,B2) % Don't forget to add 1 to histc result
LOOP : Elapsed time is 2.247998 seconds.
HISTC : Elapsed time is 0.005381 seconds.
BSXFUN: Elapsed time is 1.770875 seconds.

4 commentaires

Sean de Wolski
Sean de Wolski le 13 Avr 2011
Nice!
Andrew Newell
Andrew Newell le 13 Avr 2011
In recent versions of MATLAB you can use
[~,B1] = histc(x,c)
(but it doesn't make it any faster).
Matt Fig
Matt Fig le 13 Avr 2011
Well-played, Oleg!
Jan
Jan le 28 Juin 2012
accepted by JSimon

Connectez-vous pour commenter.

Plus de réponses (1)

Sean de Wolski
Sean de Wolski le 13 Avr 2011

0 votes

Note sure if it'll be faster but:
[row col] = find(bsxfun(@gt,c(:)',x(:)));
A2 = accumarray(row,col,[],@min)';

2 commentaires

Matt Fig
Matt Fig le 13 Avr 2011
Your intuition is correct. On my machine this is 50 times slower than the simple loop, using:
A = magic(1000);
weights = A(1,:)/sum(A(1,:));
n = 1000;
Sean de Wolski
Sean de Wolski le 13 Avr 2011
It would only get slower as the matrices get bigger. I think Ben's elementary, but properly constructed, FOR-loop is probably optimal.
I just realized and am kind of surprised FIND doesn't have a dimensional argument.

Connectez-vous pour commenter.

Catégories

En savoir plus sur Loops and Conditional Statements dans Centre d'aide et File Exchange

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by