13.1 Change the Parameter Values of an Existing Network

This section describes how you can change a neural network by changing the values of the numerical parameters. You will need to do this if, for example, you want to implement your own training algorithms and then insert the parameters into an existing network.

The following subsections explain how to change each type of network available in Neural Networks, except for the perceptron and the Hopfield network. Changing the parameters of the perceptron and Hopfield network is straightforward, considering their storage format described in Section 4.1.1, InitializePerceptron and Section 9.1.1, HopfieldFit.