How to change number of hidden layers with nprtool

Hi, I am designing a neural network using nprtool, and was wondering if there any way to increase the number of hidden layers?

Thanks,

Nika

Answer 1

[Greg Heath]

I don't think the GUI allows multiple hidden layers.

Typically, that is not important because a single hidden layer with enough nodes yields a universal approximator and therefore, is sufficient.

Multiple hidden layers tend to be used only when there is apriori information about the mathematical or physical significance of the underlying I/O relationship or a single hidden layer would require the estimation of too many weights.

Hope this helps.

Thank you for officially accepting my answer.

Greg

[smriti garg]

Hello Greg sir,

1. I am a starter to Nprtool. I have few doubts. Is there any optimal/ maximum number of hidden layer neurons to be used with nprtool GUI. I have a I/P matrix of 768x33 and o/p matrix of 768x33 of image samples.

2. Can I change the nprtool training function. This feature is available in nntool to vary different functions and layers and see the effect on preformance.

3. When I use nprtool gui, it takes very few time to give me results but when I generate and export the script of it and run the same as matlab M file, the system hangs on. What goes wrong or what is the difference in two?

Kindly help me to better understand the nprtool GUI.

If anybody else can help me, I shall be very very grateful.

Thanx in advance.

[Greg Heath]

Code?

size(input) = [ I N ] for a matix of N I-dimensional input vectors

similarly for target.