Thread Subject: "Performance goal met" error?

Subject: "Performance goal met" error?

From: Ruben Saborido Infantes

Date: 13 Oct, 2009 08:15:21

Message: 1 of 2

Hello.

I have a Multilayer Perceptron. I train this neural network using cross validation with dividerand function. Whe I train the neural network, the training stop early. If goal params is 0.0025, why TRAINLM say that "Performance goal met" and finish training.


This is the output in MATLAB:

Training with TRAINLM.
Epoch 0/35000, Time 1.885, Performance 0.49407/0.0025, Gradient 2.1001/1e-006, Mu 0.001/10000000000, Validation Checks 0/15
Epoch 10/35000, Time 49.973, Performance 0.045913/0.0025, Gradient 0.23065/1e-006, Mu 1/10000000000, Validation Checks 0/15
Epoch 20/35000, Time 93.108, Performance 0.019561/0.0025, Gradient 0.082148/1e-006, Mu 0.1/10000000000, Validation Checks 4/15
Epoch 23/35000, Time 106.502, Performance 0.019561/0.0025, Gradient 0.030073/1e-006, Mu 0.1/10000000000, Validation Checks 7/15
Training with TRAINLM completed: Performance goal met.


I hope an answer.

Thanks for all ;-)

Subject: "Performance goal met" error?

From: Greg

Date: 14 Oct, 2009 00:07:22

Message: 2 of 2

On Oct 13, 4:15 am, "Ruben Saborido Infantes"
<ruben.saborido.infan...@gmail.com> wrote:
> Hello.
>
> I have a Multilayer Perceptron. I train this neural network using cross validation with dividerand function. Whe I train the neural network, the training stop early. If goal params is 0.0025, why TRAINLM say that "Performance goal met" and finish training.
>
> This is the output in MATLAB:
>
> Training with TRAINLM.
> Epoch 0/35000, Time 1.885, Performance 0.49407/0.0025, Gradient 2.1001/1e-006, Mu 0.001/10000000000, Validation Checks 0/15
> Epoch 10/35000, Time 49.973, Performance 0.045913/0.0025, Gradient 0.23065/1e-006, Mu 1/10000000000, Validation Checks 0/15
> Epoch 20/35000, Time 93.108, Performance 0.019561/0.0025, Gradient 0.082148/1e-006, Mu 0.1/10000000000, Validation Checks 4/15
> Epoch 23/35000, Time 106.502, Performance 0.019561/0.0025, Gradient 0.030073/1e-006, Mu 0.1/10000000000, Validation Checks 7/15
> Training with TRAINLM completed: Performance goal met.
>
> I hope an answer.
>
> Thanks for all ;-)

Check the plot to see if the MSE of the validation set has passed a
minimum.

Hope this helps.

Greg

Tags for this Thread

Everyone's Tags:

Add a New Tag:

Separated by commas
Ex.: root locus, bode

What are tags?

A tag is like a keyword or category label associated with each thread. Tags make it easier for you to find threads of interest.

Anyone can tag a thread. Tags are public and visible to everyone.

Tag Activity for This Thread
Tag Applied By Date/Time
cross validation Ruben Saborido Infantes 13 Oct, 2009 04:19:04
backpropagation Ruben Saborido Infantes 13 Oct, 2009 04:19:04
mlp Ruben Saborido Infantes 13 Oct, 2009 04:19:04
goal Ruben Saborido Infantes 13 Oct, 2009 04:19:04
nn Ruben Saborido Infantes 13 Oct, 2009 04:19:03
rssFeed for this Thread

Contact us at files@mathworks.com