init :: Functions (Neural Network Toolbox™)

init

Purpose

Initialize neural network

Syntax

```
net = init(net)
```

To Get Help

Type help network/init.

Description

init(net) returns neural network net with weight and bias values updated according to the network initialization function, indicated by net.initFcn, and the parameter values, indicated by net.initParam.

Examples

Here a perceptron is created with a two-element input (with ranges of 0 to 1 and -2 to 2) and one neuron. Once it is created you can display the neuron's weights and bias.

net = newp([0 1;-2 2],1);
net.iw{1,1}
net.b{1}

Training the perceptron alters its weight and bias values.

P = [0 1 0 1; 0 0 1 1];
T = [0 0 0 1];
net = train(net,P,T);
net.iw{1,1}
net.b{1}

init reinitializes those weight and bias values.

```
net = init(net);
net.iw{1,1}
net.b{1}
```

The weights and biases are zeros again, which are the initial values used by perceptron networks (see newp).

Algorithm

init calls net.initFcn to initialize the weight and bias values according to the parameter values net.initParam.

Typically, net.initFcn is set to 'initlay', which initializes each layer's weights and biases according to its net.layers{i}.initFcn.

Backpropagation networks have net.layers{i}.initFcn set to 'initnw', which calculates the weight and bias values for layer i using the Nguyen-Widrow initialization method.

Other networks have net.layers{i}.initFcn set to 'initwb', which initializes each weight and bias with its own initialization function. The most common weight and bias initialization function is rands, which generates random values between -1 and 1.