hi i am new on ANN. i can't understand how to use nn for forecasting. After training process i will get weigths and what is role of these weigths for prediction or how can i use these weigths for prediction? please explain to me anyone

 Réponse acceptée

Greg Heath
Greg Heath le 23 Avr 2015
After training you will have a net with weights. Even if the number of weights is small, it is often difficult to understand what role each weight plays in obtaining the correct answer.
Furthermore, the number of weights, typically, far exceeds the number of inputs. In particular,
for a typical I-H-O net,
Nw = (I+1)*H+(H+1)*O >> I.
Consequently, most practitioners focus on the resulting output changes when certain inputs change.
Finally, when a net is to be used further, the whole net as a structure is usually transported rather than a collection of weights.
Hope this helps.
Greg

Plus de réponses (0)

Catégories

En savoir plus sur Deep Learning Toolbox dans Centre d'aide et File Exchange

Community Treasure Hunt

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

Start Hunting!

Translated by