Error: The variable VAC in a parfor cannot be classified

Dear All,
I read some of the solutions to the error mentioned in the subject of this question but they all seemed vague to me. I have a code as follows:
VAC = zeros(taw_max ,2);
parpool ('local',12)
parfor taw = 0:1:taw_max
taw_matrix = zeros(n-taw , 1);
for t = 0:(n-1-taw)
allmatrix = zeros (nt,1);
for i=1:nt
allmatrix (i,1) = mh * dot (s_t , s_t_taw )/dot(s_t , s_t); %where s_t . s_t_taw and mh are all defined values
end
taw_matrix (t+1 ,1) = sum(allmatrix);
end
VAC (taw +1 ,2) = mean (taw_matrix);
VAC (taw +1 ,1) = taw ;
end
it seems that the parfor loop can not classify the VAC matrix so that I can not write the output of my calculations into the corresponding cell of the VAC matrix. Would you please help me with a solution to this matter?
Thank you so much

 Akzeptierte Antwort

Edric Ellis
Edric Ellis am 14 Mär. 2016
Bearbeitet: Edric Ellis am 15 Mär. 2016
I think you simply need to ensure that you make only a single assignment into VAC, as the code analyzer message suggests. Like this:
VAC(taw + 1, :) = [taw, mean(taw_matrix)];

3 Kommentare

Ced
Ced am 14 Mär. 2016
Bearbeitet: Ced am 14 Mär. 2016
Second that. For correctness, I think the order should be
VAC(taw + 1, :) = [taw, mean(taw_matrix)];
but that's a detail.
On a side-note, you cannot have a space between a function name (here "mean") and the parenthesis of the argument. Matlab will think that mean is a variable name.
PS: Thanks for correcting the indentation, much better ;)
Thank you so much Edric and Ced!
Fixed my answer to put the elements in the correct order - well spotted Ced!

Melden Sie sich an, um zu kommentieren.

Weitere Antworten (1)

mohchali Alichina
mohchali Alichina am 9 Feb. 2022
Bearbeitet: Walter Roberson am 9 Feb. 2022
I have a function and I need to apply this function to an electrician field ( apply waveform to ion motion) . Kindly can anyone guide me to do it! thanks in advance
Fs=25.8e6;
%run swiftfunccycle10ms function
[y,swiftT] = swiftfunccycle10ms(100,10e3,300e3,67e3,80e3,20e3,67e3,80e3);
%------------------------------------
s=real(swiftT);
[cfs,frq] = cwt(s,Fs); %continues wavelet transform
tms = (0:numel(s)-1)/Fs;%length of s
figure
subplot(2,1,1)
plot(tms,s)
axis tight
title('swiftT Scalogram')
xlabel('Time (s)')
ylabel('Amplitude')
subplot(2,1,2)
surface(tms,frq,abs(cfs))
axis tight
shading flat
xlabel('Time (s)')
ylabel('Frequency (Hz)')
%set(gca,'yscale','log')
ylim([10000 500000]);
%---------------------------------------------
%run swiftfunccycle function
[y, signalT]=signalfunccycle(0.1,10000,300000,67000,80000,200000);
%---------------------------------------------
s=real(signalT);
[cfs,frq] = cwt(s,Fs); %continues wavelet transform
tms = (0:numel(s)-1)/Fs;%length of s
figure
subplot(2,1,1)
plot(tms,s)
axis tight
title('signalT Scalogram')
xlabel('Time (s)')
ylabel('Amplitude')
subplot(2,1,2)
surface(tms,frq,abs(cfs))
axis tight
shading flat
xlabel('Time (s)')
ylabel('Frequency (Hz)')
%set(gca,'yscale','log')
ylim([10000 500000])

1 Kommentar

Where you are using parfor that this might apply?
Your signalfunccycle and swiftfunccycle10ms are not found anywhere on the Internet that I can see, so we cannot test your code.

Melden Sie sich an, um zu kommentieren.

Kategorien

Community Treasure Hunt

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

Start Hunting!

Translated by