MATLAB: In different execution different performance comes in the Neural Network same program

MATLABneural network

Why in different execution different performance comes in the Neural Network same program ?

Best Answer

The state of the RNG is different. Therefore the random initial weights and random trn/val/tst data divisions are different.

For examples of initializing the RNG so that runs can be duplicated try searching on

 greg rng(4151941)
 greg rng(0)
 greg rng('default')

Related Solutions

MATLAB: Different result neural network classification

Yes. The default trn/val/tst data-division and weight initializations are all random.

For reproducibility, initialize the RNG to the same state before TRAIN is called.

Over the years I have used

 rng(0) and rng(4151941) (a famous birth date).

Now I tend to use

 rng('default')

To search for a good design, I usually use a double loop search over number of hidden nodes and Ntrials random generator states per hidden node choice:

 rng('default')
 j=0
 for h = Hmin:dH:Hmax
     j=j+1
     net = fitnet(h);
     blah; blah; blah;
     for i = 1: Ntrials
        net = configure(net,x,t);
        blah; blah; blah;
        NMSE(i,j) = mse(t-y)/mean(var(t',1));
      end
end

Over the years I have posted zillions of examples in BOTH the NEWSGROUP and ANSWERS.

Hope this helps.

Thank you for formally accepting my answer

Greg

MATLAB: Neural Network Output is different every time??

Initial weights and data division are random unless you initialize the random number generator to the same state before training.

Best Answer

Related Solutions

MATLAB: Different result neural network classification

MATLAB: Neural Network Output is different every time??

Related Question