MATLAB Answers

FIR
0

Error using newrb and newgrnn

Asked by FIR
on 7 Jan 2013
I have a code for svm for classifying
load fisheriris
xdata = meas(51:end,3:4);
group = species(51:end);
svmStruct = svmtrain(xdata,group,'showplot',true);
same way i tried using newrb and newgrnn
net = newgrnn(xdata,group)
??? Error using ==> cat
CAT arguments dimensions are not consistent.
Error in ==> cell2mat at 89
m{n} = cat(1,c{:,n});
Error in ==> newgrnn at 70
if isa(t,'cell'), t = cell2mat(t); end
please tell how to fix it
syntax for those says, data and target class,but why i get error

  0 Comments

Sign in to comment.

1 Answer

Answer by Greg Heath
on 7 Jan 2013
 Accepted Answer

help grnn
doc grnn
explain the required type and orientation of the input and target data
whos
shows both the type and orientation of your data
xdata
group
without the semicolons allows you to "see" the data.
With the above info you should be able to readily debug your code.
Hope this helps.
Thank you for formally accepting my answer.
Greg

  3 Comments

FIR
on 7 Jan 2013
I tried with this small code ,
P=magic(5);
T=[0 0 1 1 1];
net = newgrnn(P,T)
K=P(:,4);
y = sim(net,K);
y gives 1 as output,plz tell what it indicates
Greg Heath
on 7 Jan 2013
You created the net to output 0 when either of the first two P columns are used as input and to output 1 when one of the last three columns of P are used as input.
Have you folloed up on my previous advice?
If so are there anymore questions?
G
FIR
on 7 Jan 2013
final question how to find the accuracy

Sign in to comment.