The variable g in a parfor cannot be classified..

1 view (last 30 days)
freebil
freebil on 22 Oct 2014
Edited: Edric Ellis on 23 Oct 2014
Can anyone help me with this?
parfor j=1:N
z1r=normrnd(Z1(1),Z1(2)); %Random sample from distribution of Z1
z2r=normrnd(Z2(1),Z2(2)); %Random sample from distribution of Z2
for i=1:size(T,2)
g(i)=Threshold-M(i,1)-(sqrt(lambda1)*V1(i)*z1r)-(sqrt(lambda2)*V2(i)*z2r);
if g(i)>0 %safe
IND(i,j)=0;
else %failure
IND(i,j)=1;
end
end
end

Answers (1)

Edric Ellis
Edric Ellis on 23 Oct 2014
Edited: Edric Ellis on 23 Oct 2014
It looks like you could simply replace g(i) with g and then hoist the inner for-loop bounds calculation out of the PARFOR loop. (Inner FOR loops must have constant bounds to be able to run correctly).
I.e.
tn = size(T, 2);
parfor j = 1:N
...
for i = 1:tn
...
end
end
  2 Comments
freebil
freebil on 23 Oct 2014
thanks but it says that IND cannot be classified then..
Edric Ellis
Edric Ellis on 23 Oct 2014
Ah, you're right, I'll update my answer.

Sign in to comment.

Categories

Find more on Parallel Computing in Help Center and File Exchange

Community Treasure Hunt

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

Start Hunting!