Effacer les filtres
Effacer les filtres

How to get validation and Test data of Neural network in MATLAB

8 vues (au cours des 30 derniers jours)
ghazale
ghazale le 8 Oct 2013
Commenté : Jan le 21 Jan 2019
Hi
I have a question. I have trained a neural network and for example I had 1000 Inputs, 70 are used for training. 15% for validation and 15 % for testing and now outputs are shown and the network is trained. I want to understand how an I extract the validation and test data in command window of MATLAB. For example if I have an input matrix 3*1000. and my out put matrix is 2*1000. so the validation matrix will be 3*150. How can I show this matrix and finally its related out put in command window of MATLAB ?
Thanks a lot
  1 commentaire
Jan
Jan le 21 Jan 2019
@Amine Bendali: Please post a comment to ask for clarifications. Flags are thought to inform admins aboud messages, which might collide with the terms of use, e.g. by spam or rudeness.

Connectez-vous pour commenter.

Réponses (2)

Greg Heath
Greg Heath le 8 Oct 2013
The input matrix contains train, val and test data. Similarly for the output data.
Indices are available in the training record tr
[ net tr Y E Xf Af ] = train(net,input,target,Xi,Ai).
Type, without a semicolon
tr = tr
for other useful information.
Hope this helps.
Thank you for formally accepting my answer
Greg
  1 commentaire
ghazale
ghazale le 8 Oct 2013
Hi, thanks for response, but I trained my network with fitting tool, I mean nnstart and then fitting tool, this tool choose 15 of data as validation and 15 as test data, how can I extract them?

Connectez-vous pour commenter.


ghazale
ghazale le 8 Oct 2013
Hi, thanks for response, but I trained my network with fitting tool, I mean nnstart and then fitting tool, this tool choose 15 of data as validation and 15 as test data, how can I extract them?
  2 commentaires
Greg Heath
Greg Heath le 8 Oct 2013
PLEASE DO NOT USE THE ANSWER BOX FOR COMMENTS OR QUESTIONS!
Greg Heath
Greg Heath le 8 Oct 2013
Don't know how to do it by GUI.

Connectez-vous pour commenter.

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by