Azzera filtri
Azzera filtri

How do Size input and Target correctly during training of Feed Forward Neural Network

1 visualizzazione (ultimi 30 giorni)
I have a input data set consisting of 12 columns and 108 rows. The target data set consist of 1 column and 108 rows. While training my Neural Network the error message "Targets are incorrectly sized for network. Matrix must have 12 columns" is flagged. Please how do I solve this problem.
Note I am using R2008b version.
Here is my code
load RainData_Cal_Inputs.dat load RData_Cal_target yinputs= RainData_Cal_Inputs ytarget= RData_Cal_target;
[y11,PS] = mapminmax(yinputs)
[y22,PS] = mapminmax(ytarget);
y22=y22'
%Network traning and validation
net.divideParam.trainRatio = 70/100;
net.divideParam.valRatio = 15/100;
net.divideParam.testRatio = 15/100;
net.trainParam.epochs = 200;
net = newff(y11,y22,12); net =init(net); net.trainParam.showCommandLine=true view(net) YY = sim(net,y11);
net = train(net,y11,y22)

Risposta accettata

Greg Heath
Greg Heath il 1 Giu 2014
Transpose your matrices so that
[ I N ] = size(input) %[ 12 108 ]
[ O N ] = size(target) %[ 1 108 ]
Then search
greg newff
Hope this helps.
Thank you for formally accepting my answer
Greg

Più risposte (0)

Categorie

Scopri di più su Sequence and Numeric Feature Data Workflows in Help Center e File Exchange

Community Treasure Hunt

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

Start Hunting!

Translated by