Discover MakerZone

MATLAB and Simulink resources for Arduino, LEGO, and Raspberry Pi

Learn more

Discover what MATLAB® can do for your career.

Opportunities for recent engineering grads.

Apply Today

New to MATLAB?

How to set target vector for training of neural network?

Asked by Abbas

Abbas (view profile)

on 21 May 2013
Accepted Answer by Jing

Jing (view profile)

I have six data values which I have to classify in two classes of three data values each.I am using a feedforward neural network for this purpose.The values are: 8.8950 9.9709 10.4990 15.0184 11.9559 and 12.5560.I want to classify first three values in one class and remaining three values in the other class.I have set my input vector like below:


Now please help me to set my target vector.Moreover if input vector needs some improvements then please highlight them as well.



Abbas (view profile)

2 Answers

Answer by Jing

Jing (view profile)

on 22 May 2013
Accepted answer

Then the target vector can be [1 1 1 0 0 0;0 0 0 1 1 1]

1 Comment


Abbas (view profile)

on 24 May 2013

Dear Jing,

My input matrix has two rows and three columns whereas target vector suggested by you contains six columns which generates error.I have transformed my input matrix into input vector of six values(one row and six columns) and it is working.But what will be target vector if I have input matrix of two rows and three columns?

Thanks for your help.


Jing (view profile)

Answer by Greg Heath

Greg Heath (view profile)

on 24 May 2013

The input matrices for N I-dimensional inputs and the corresponding O-dimensional targets must have the dimensions

[ I N ] = size(inputs)

[ O N ] = size(targets)

Therefore, your syntax is incorrect.

However, the syntax of Jing is correct.

Hope this helps.

Thank you for formally accepting my answer



Greg Heath

Greg Heath (view profile)

Contact us