Neural Networks implementation In Matlab

The nntool GUI can be used to create and train different types of neural network available under MATLAB.

The GUI can be invoked by typing nntool at the command window.

First, define the training inputs by clicking “Import…”, select group from the list of variables. Assign a name to the inputs and indicate that this variable should be imported as inputs.

User can provide inputs directly by specifying them in the value field if no imports are available.

Similarly target data to the inputs can be assigned.

A new perceptron network is created by clicking “New Network…”, a new window appears where network architecutre can be defined. The following network parameters are to be set before creating the network.

Network type – type of neural network to be used for solving the problem at hand (feed-forward backprop, hopfield, LVQ, etc).

Input data – input to the neural network.

Target data – desired output of the neural network regarding the input data.

Training function – function to be used in neural network for training (batch training and incremental training).

Adaptation learning function – function to be used in neural network for traing (incremental training).

Performance function – function that will improve the performance of the neural network.

No. of layers – total number of layers in the neural network (min. 2).

No. of neurons – total number of neurons I the neural network (input layer + output layer + hidden layer).

Transfer function – transfer function to determine the firing strength of the neuron in the neural network.

A new network is created by clicking “Create” button.