feed forward backprob neural network

1 visualización (últimos 30 días)
yo yo
yo yo el 7 de Jul. de 2015
Comentada: Greg Heath el 8 de Jul. de 2015
Hi I am creating a feed forward backprob neural network using nntool i want to use it for character recognition i have a dataset with 10 class and there are 50 samples from each class ( dataset attached you can download ) . my neural net Configuration is below : Training function : Traingdx Adaption learning function : LEARNGDM
Performance function : SSE Number of layer : 2 Number of neurons in Layer1 : 15 Transfer function : LOGSIG
I create network and trained with sample data but when i simulate network with any of sample inputs the result is always [0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 ] ... i do not know whats going wrong please help me thanks ...

Iniciar sesión para responder a esta pregunta.

Respuestas (0)

Categorías

Más información sobre Pattern Recognition and Classification en Help Center y File Exchange.

Etiquetas

Community Treasure Hunt

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

Start Hunting!

Translated by