size ANN training input
|
View:
New views
5 Messages
—
Rating Filter:
Alert me
|
 |
size ANN training input
by Andrea_Viano
::
Rate this Message:
Some parts of this message have been removed.
Learn more about Nabble's security policy.
Hi, Is there anyone that knows why training inputs of an ANN have to be less than 1 to have a convergence? I tried with my regular input( 10000,or 8 etc), and I had a very high and constant error. If I divided training inputs to reach numbers close to the unity I could find a solution of my ANN. Can you help me to solve this problem? Thanks Andrea Ing. Andrea Viano Industry Division
CFD Engineering S.r.l. Headquarters Piazza della Vittoria, 7/2A - 16123 Genova (Italy) Phone +39-010-8540050 - Fax +39-010-8631199 ------------------------------------------------------------------------------ RSA® Conference 2012 Save $700 by Nov 18 Register now! http://p.sf.net/sfu/rsa-sfdev2dev1 _______________________________________________ Fann-general mailing list Fann-general@... https://lists.sourceforge.net/lists/listinfo/fann-general |
|
|
Re: size ANN training input
by D. Cooper Stevenson-2
::
Rate this Message:
Hi Andrea,
I have an algorithm I can provide you that will scale between 1 and 0. It's written in shell but you're welcome to it if you like. Very Best, -Cooper
On Wed, Nov 2, 2011 at 9:21 AM, Andrea_Viano <viano@...> wrote:
-- D. Cooper Stevenson ph: 541.971.0366 em: cooper@... www: http://cooper.stevenson.name ------------------------------------------------------------------------------ RSA® Conference 2012 Save $700 by Nov 18 Register now! http://p.sf.net/sfu/rsa-sfdev2dev1 _______________________________________________ Fann-general mailing list Fann-general@... https://lists.sourceforge.net/lists/listinfo/fann-general |
|
|
Re: size ANN training input
by Joan Maspons-2
::
Rate this Message:
Hello,
Some activation functions outputs such as sigmoid ones give and output between 0 and 1 or -1 and 1. If you try to get values around 10000 as a pondered summation of the inputs and the outputs range is between 0 and 1 it's impossible to get such a big values. You can take a look to the fann*scale* functions at http://leenissen.dk/fann/html/files/fann_train-h.html. J. Maspons 2011/11/2 Andrea_Viano <viano@...>
-- Joan Maspons CREAF (Centre de Recerca Ecològica i Aplicacions Forestals) Universitat Autònoma de Barcelona, 08193 Bellaterra (Barcelona), Catalonia Tel +34 93 581 2915 j.maspons@... http://www.creaf.uab.cat ------------------------------------------------------------------------------ RSA® Conference 2012 Save $700 by Nov 18 Register now! http://p.sf.net/sfu/rsa-sfdev2dev1 _______________________________________________ Fann-general mailing list Fann-general@... https://lists.sourceforge.net/lists/listinfo/fann-general |
|
|
Re: size ANN training input
by Fábio Blessa
::
Rate this Message:
Normalization you mean? On Nov 2, 2011 2:49 PM, "Andrea_Viano" <viano@...> wrote:
------------------------------------------------------------------------------ RSA® Conference 2012 Save $700 by Nov 18 Register now! http://p.sf.net/sfu/rsa-sfdev2dev1 _______________________________________________ Fann-general mailing list Fann-general@... https://lists.sourceforge.net/lists/listinfo/fann-general |
|
|
R: size ANN training input
by Andrea_Viano
::
Rate this Message:
Some parts of this message have been removed.
Learn more about Nabble's security policy.
Thank you Fabio, I used a backpropagation method and I used activation sigmoid function. Is it maybe for that I have to normalize? Andrea Da: Fábio Blessa [mailto:fabioblessa@...] Normalization you mean? On Nov 2, 2011 2:49 PM, "Andrea_Viano" <viano@...> wrote: Hi, Is there anyone that knows why training inputs of an ANN have to be less than 1 to have a convergence? I tried with my regular input( 10000,or 8 etc), and I had a very high and constant error. If I divided training inputs to reach numbers close to the unity I could find a solution of my ANN. Can you help me to solve this problem? Thanks Andrea Ing. Andrea Viano Industry Division
CFD Engineering S.r.l. Headquarters Piazza della Vittoria, 7/2A - 16123 Genova (Italy) Phone +39-010-8540050 - Fax +39-010-8631199
------------------------------------------------------------------------------ RSA(R) Conference 2012 Save $700 by Nov 18 Register now http://p.sf.net/sfu/rsa-sfdev2dev1 _______________________________________________ Fann-general mailing list Fann-general@... https://lists.sourceforge.net/lists/listinfo/fann-general |
| Free embeddable forum powered by Nabble | Forum Help |
