Neural Network Toolbox

Custom Networks

The Neural Network Toolbox is designed to allow for many kinds of networks. This makes it possible for many functions to use the same network object data type.

Here are all the standard network creation functions in the toolbox.

New Networks
newc	Create a competitive layer.
newcf	Create a cascade-forward backpropagation network.
newelm	Create an Elman backpropagation network.
newff	Create a feed-forward backpropagation network.
newfftd	Create a feed-forward input-delay backprop network.
newgrnn	Design a generalized regression neural network.
newhop	Create a Hopfield recurrent network.
newlin	Create a linear layer.
newlind	Design a linear layer.
newlvq	Create a learning vector quantization network.
newp	Create a perceptron.
newpnn	Design a probabilistic neural network.
newrb	Design a radial basis network.
newrbe	Design an exact radial basis network.
newsom	Create a self-organizing map.

This flexibility is possible because we have an object-oriented representation for networks. The representation allows various architectures to be defined and allows various algorithms to be assigned to those architectures.

To create custom networks, start with an empty network (obtained with the network function) and set its properties as desired.

• network  - Create a custom neural network.
  

The network object consists of many properties that you can set to specify the structure and behavior of your network. See Network Object Reference for descriptions of all network properties.

The following sections demonstrate how to create a custom network by using these properties.

Advanced Topics

Custom Network

© 1994-2005 The MathWorks, Inc.