Neural Networks, creating and training

Why does a Matlab neural network store example data? That is, if I create a neural network using newpr or newff, it requires representative input and target vectors, why are those stored to net.inputs?

I'm creating a neural network that better fits my workflow (define the architecture at creation and provide training data only at training time) and I want to make sure I completely understand the purpose of providing representative vectors to the newpr and newff functions.

Thanks!

Answer 1

[Ram]

I donno if you know about the nntool option..Its a GUI for developing neural networks.. It requires you to define the architecture first and u can provide the training data during training..

[T.R.]

I do, but thank you very much! I actually need to do this programmatically.