How to differentiate between Mse(training) and Mse(testing) ??
1 vue (au cours des 30 derniers jours)
Afficher commentaires plus anciens
omar belhaj
le 14 Jan 2016
Réponse apportée : Greg Heath
le 23 Sep 2016
Hello, i would like to know how to differentiate between Mse(training) and Mse(testing) during training phases this is my part of my ANN's code :
net.divideMode = 'sample'; % Divide up every sample
net.divideParam.trainRatio = 70/100;
net.divideParam.valRatio = 20/100;
net.divideParam.testRatio = 10/100;
net.performFcn = 'mse'
MSEt(j,H) = mse(net,targets,outputs);%mean square error
*my question is that MSE calculated is for training error or for testing error ??
Help please*
0 commentaires
Réponse acceptée
Greg Heath
le 23 Sep 2016
In order to obtain the break down of mse into its trn/val/tst parts you have to obtain the training record, tr from training;
[ net tr Y E Xf Af ] = train ( net, X ,T, Xi, Ai );
help train
doc train
Hope this helps.
Greg
0 commentaires
Plus de réponses (0)
Voir également
Catégories
En savoir plus sur Particle Swarm dans Help Center et File Exchange
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!