Mean a matrix columnwise based on another logical matrix
1 view (last 30 days)
Show older comments
I want to mean the matrix a based on the logical matrix b columnwise: In other words this:
mean(a(b(:,1)))
mean(a(b(:,2)))
...
a =[
7 8 5 3
1 1 4 7
9 3 8 7
10 1 8 2
7 1 2 2
8 9 5 5
8 7 5 10
4 4 7 4
7 10 8 6
2 1 8 3];
b =logical([
0 0 1 0
1 0 1 0
1 0 1 1
1 0 1 1
1 1 1 1
1 1 1 1
1 1 1 1
1 1 1 1
1 0 0 1
0 0 0 1]);
1 Comment
Jan
on 18 Jun 2013
This is obviously an interesting question, which allows to shed light on different powers of Matlab. +1
Accepted Answer
Jan
on 16 Jun 2013
Or:
m = sum(a .* b, 1) ./ sum(b, 1);
4 Comments
Wayne King
on 17 Jun 2013
I agree Jan, +1, silly of me not to think of simply using the b matrix to get the correct number of nonzero elements.
More Answers (4)
Andrei Bobrov
on 17 Jun 2013
Edited: Andrei Bobrov
on 18 Jun 2013
b1 = bsxfun(@times,b,1:4);
out = accumarray(b1(b),a(b),[],@mean);
or
[~,jj] = find(b);
out = accumarray(jj,a(b),[],@mean);
and in line with accumarray:
out = accumarray(ceil(find(b)/size(b,1)),a(b),[],@mean);
2 Comments
Jan
on 18 Jun 2013
Although I do not think that accumarray is the fastest approach here, I vote for it a sign of my admiration for all successful accumarray masters.
Wayne King
on 16 Jun 2013
Edited: Wayne King
on 16 Jun 2013
you mean just:
mean(a.*b)
or do you want the mean of just the nonzero entries? In other words, divide by the right number of elements.
C = a.*b;
for nn = 1:size(C,2)
numelements(nn) = nnz(C(:,nn));
end
colsumz = sum(C);
meanz = colsumz./numelements;
0 Comments
Azzi Abdelmalek
on 16 Jun 2013
c=a.*b
c(c==0)=nan;
out=nanmean(c)
2 Comments
John Doe
on 18 Jun 2013
Just learned about nanmean. I've been looking for a function like that for quite some time. I've always used some workaround. Thanks!
See Also
Categories
Find more on Creating and Concatenating Matrices in Help Center and File Exchange
Products
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!