Different matrix dimention of input and target in neural networks?

1 view (last 30 days)
Amalia
Amalia on 6 Oct 2014
Commented: Amalia on 8 Oct 2014
Why couldn't I get the network to work with a 1x250 input and a 2x250 target? When the input and target are both a 2x250 matrix the program worked just fine. My inputs and targets are taken from excel. I didn't know how to fix this problem.
This is the code: Ptr=xlsread('liaa(10s).xlsx','sheet1','I3:I252');
T=xlsread('liaa(10s).xlsx','sheet1','G3:H252');
net=newff(Ptr,T,4,{'logsig','purelin'},'trainlm','learngdm');
net.trainParam.lr=0.5; net.trainParam.epochs=25; net.trainParam.goal=1e-10;
[net,tr]=train(net,Ptr,T);
and the error said: ??? Error using ==> network.train at 146 Targets are incorrectly sized for network. Matrix must have 1 columns.
Error in ==> training_lia at 11 [net,tr]=train(net,Ptr,T);
  1 Comment
Geoff Hayes
Geoff Hayes on 6 Oct 2014
Amalia - if you are observing an error, then please include the error message and any text that is relevant. Including a portion of your code that, when executed, generates the error would be helpful too.

Sign in to comment.

Sign in to answer this question.

Accepted Answer

Greg Heath
Greg Heath on 7 Oct 2014
You may have to transpose your data
Check dimensions
[ I N ] = size(input) % [ 1 250]
[ O N ] = size(target)% [ 2 250 ]
P.S. To simplify your code, use default parameters whenever possible. See the examples in
help newff
doc newff
Hope this helps.
Thank you for formally accepting my answer
Greg

More Answers (0)

Categories

Find more on Image Data Workflows in Help Center and File Exchange

Tags

Community Treasure Hunt

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

Start Hunting!