- It is not used for training
- .If it were used it would yield a random mixture of trn, val and tst data
Hi, I am working on a set of code implying the ANN. My end purpose is to plot a comparison graph between actual output and simulated output. However, there isn't any points displayed in my plot. What is the correct syntax?
1 view (last 30 days)
Show older comments
A=data(:,2)
R1=data(:,5)
TA=A' % Transpose the Input and Output data from column to row matrix
TR1=R1'
[NA,PS]=mapminmax(TA) %Normalize the transposed input and output data
[NR1,TS]=mapminmax(TR1)
for i=1:20;
net = feedforwardnet(i,'trainlm');
net.trainParam.show = 50;
net.trainParam.lr = 0.05;
net.trainParam.epochs = 300;
net.trainParam.goal = 1e-5;
[net,tr] = train(net,NA,NR1);
[trainInd,valInd,testInd] = dividerand(20,0.7,0.15,0.15); %Divide data randomly
Y=sim(net,NA)
TY=mapminmax('reverse',Y,TS); %Denormalize simulated data
plot(i,TR1,'red',i,TY,'blue') % Plot comparison graph of experimental output data and simulated data
title('Graph of Comparison between Experimental Output Data and Simulated Data')
xlabel('Experimental Input Data')
ylabel('Experimental Output Data and Simulated Data')
legend('Experimental Output Data','Simulated Data')
end
1 Comment
Greg Heath
on 21 Dec 2018
The dividerand statement makes no sense.
OR am I missing something ?
Greg
Accepted Answer
madhan ravi
on 18 Dec 2018
Edited: madhan ravi
on 18 Dec 2018
However, there isn't any points displayed in my plot. What is the correct syntax?
plot(i,TR1,'-or',i,TY,'-ob') % add 'o' as marker to view the data points
2 Comments
madhan ravi
on 21 Dec 2018
save every variable in each iteration and plot them outside the loop ,
P.S - I don't have neural network toolbox so I cannot test.
More Answers (0)
See Also
Categories
Find more on Sequence and Numeric Feature Data Workflows 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!