Dear Sir/Madam, I am using the ANN toolbox (nnstart) to create and train my ANN. There are some questions I would really appreciate if any of you guys could explain to me. I read some comments in this community that we do not need to normalise the inseted data before inputing them in to our network. May I ask whether it is correct? Sometimes, I'd like to use other training function such as the Gradient Descent. But this option does not appear in the toolbox. Whenever I 'd like to se it, I have to call it by command lines? If I'd like to change the transfer function to other types, for example, I'd like to use sigmoid function for the output layer as well, how can I do it? I'd like to extract the biases and weights of the trained ANN to derive the correlated function between my inputs and the output. How could I do it? Thank you so much, Kind Regards, Kim,

Artificial Neural Network questions

1 visualización (últimos 30 días)

Hoang Trinh el 10 de Dic. de 2019

0
Enlazar

Enlace directo a esta pregunta

https://la.mathworks.com/matlabcentral/answers/495857-artificial-neural-network-questions

Dear Sir/Madam,

I am using the ANN toolbox (nnstart) to create and train my ANN. There are some questions I would really appreciate if any of you guys could explain to me.

I read some comments in this community that we do not need to normalise the inseted data before inputing them in to our network. May I ask whether it is correct?
Sometimes, I'd like to use other training function such as the Gradient Descent. But this option does not appear in the toolbox. Whenever I 'd like to se it, I have to call it by command lines?
If I'd like to change the transfer function to other types, for example, I'd like to use sigmoid function for the output layer as well, how can I do it?
I'd like to extract the biases and weights of the trained ANN to derive the correlated function between my inputs and the output. How could I do it?

Thank you so much,

Kind Regards,

Kim,