How to take training data from different indexes in a matrix in nprtool ?

Hi,

I am trying to do classification using nprtool, for a problem I need to divide my data from specific indexes for training and testing. For example - From 1372 samples, I need 1:272 and 546:1372 to be my training set and rest as validation and test sets. I am using 'divideInd', how can I accomplish such a division of data ?

net.divideFcn = 'divideind';  
net.divideMode = 'sample';  

net.divideParam.trainInd = 1:272 + 546:1372;

net.divideParam.valInd = 273:409;
net.divideParam.testInd = 410:545; 
net.trainFcn = 'trainscg';

Answer 1

[Greg Heath]

net.divideParam.trainInd = [1:272, 546:1372];

Hope this helps

Thank you for formally accepting my answer

Greg

[Rohit Chandra]

Thanks Greg.