MATLAB: How much number of hidden neurons should I take in nprtool

Deep Learning Toolboxnprtoolnumber of hidden neurons

Best Answer

It is always best to

 1. Accept all defaults
 2. Run
 3. Evaluate the non-training validation and test results
 4. If satisfactory, consider reducing H, the number of hidden nodes

If unsatisfactory,

 1. Consider changing to non-default values
 2. Consider changing to a command line approach so that you can use loops to create multiple designs more easily. For example I usually use an outer loop with 10 candidates for H and an inner loop of Ntrials = 10 different weight initializations for each H candidate.
 3. For command line examples search 
     greg Ntrials

Hope this helps.

Thank you for formally accepting my answer

Greg

Related Solutions

MATLAB: Retrain neurons

You could try

[net2 tr Y E] = train(net1,p2,t2); % Train for a new set

% However, performance of net2 on p1 may be unsatisfactory

Therefore use

[net2 tr Y E] = train(net1,[p1 p2],[t1 t2]);

Hope this helps.

Greg

MATLAB: Improve the performance of nprtool

Insufficient information

 Which of the MATLAB classification example datasets are you using?
 help nndatasets
 doc nndatasets
 Number of classes  c =?
 Input vector dimensionality I = 1
 Number of examples  N = ?
 [ I N ] = size(input)
 [ O N ] = size(target)% O = c
 Default 70/15/15 data division? (75/15/15 doesn't add to 100)

Some problems require multiple(e.g., 10) designs for every value of hidden nodes that are tried.

For example, search the NEWSGROUP and ANSWERS using

 greg patternnet Ntrials
 Sorry I can't give you much advice on how to optimize the use of nprtool. However, consulting my command line code should be more than worthwhile.

Hope this helps.

Thank you for formally accepting my answer

Greg.

Best Answer

Related Solutions

MATLAB: Retrain neurons

MATLAB: Improve the performance of nprtool

Related Question