Neural Network Toolbox™

Divide vectors into three sets using blocks of indices

Syntax

• [trainV,valV,testV,trainInd,valInd,testInd] = 
  divideblock(allV,trainRatio,valRatio,testRatio)
  

Description

divideblock is used to separate input and target vectors into three sets: training, validation, and testing. It takes the following inputs:

allV	R x Q matrix of Q R-element vectors.
trainRatio	Ratio of vectors for training. Default = 0.6.
valRatio	Ratio of vectors for validation. Default = 0.2.
testRatio	Ratio of vectors for testing. Default = 0.2.

and returns

trainV	Training vectors
valV	Validation vectors
testV	Test vectors
trainInd	Training indices
valInd	Validation indices
testInd	Test indices

Examples

• p = rands(3,1000);
  t = [p(1,:).*p(2,:); p(2,:).*p(3,:)];
  [trainP,valP,testV,trainInd,valInd,testInd] =  
    divideblock(p,0.6,0.2,0.2);
  [trainT,valT,testT] = divideind(t,trainInd,valInd,testInd);
  

Network Use

Here are the network properties that define which data division function to use, and what its parameters are, when train is called.

• net.divideFcn
  net.divideParam
  

See Also

divideind, divideint, dividerand

Provide feedback about this page

dist

divideind

© 1984-2008- The MathWorks, Inc.    -   Site Help   -   Patents   -   Trademarks   -   Privacy Policy   -   Preventing Piracy   -   RSS