multiprod implementation question for matrix multiplication optimization
3 Ansichten (letzte 30 Tage)
Ältere Kommentare anzeigen
Hi all. I'm trying to optimize matrix multiplication. What I am trying to do is take a stack of 3 x 3 matrices and multiply each matrix by one vector from an array of vectors .
I did find an implementation not using multiprod that solves this issue without for loops, but makes extensive use of repmat and performs worse than if I were to just use a for loop.
So for the case below: multiprod(A,B) would return [1 4 7] and [12 15 18].
A = [1 2 3; 4 5 6; 7 8 9; 10 11 12; 13 14 15; 16 17 18;]
B = [1 0; 0 0; 0 1]
Currently, multiprod assumes that you want to multiply each 3x3 by the 3x2 and as a result I get extraneous data. Adding the extra parameters about the inner dimensions to the function call did not work either.
So I guess my question is...can multiprod solve this issue and I'm just not implementing it correctly or am I better off using for loops?
0 Kommentare
Akzeptierte Antwort
Matt J
am 11 Feb. 2016
Bearbeitet: Matt J
am 11 Feb. 2016
>> A = cat(3,[1 2 3; 4 5 6; 7 8 9],[ 10 11 12; 13 14 15; 16 17 18])
A(:,:,1) =
1 2 3
4 5 6
7 8 9
A(:,:,2) =
10 11 12
13 14 15
16 17 18
>> B=reshape(B,3,1,[])
B(:,:,1) =
1
0
0
B(:,:,2) =
0
0
1
>> mtimesx(A,B)
ans(:,:,1) =
1
4
7
ans(:,:,2) =
12
15
18
Weitere Antworten (0)
Siehe auch
Kategorien
Mehr zu Operating on Diagonal Matrices finden Sie in Help Center und File Exchange
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!