org.cognity.neural.init

Class UniformInitializer

java.lang.Object

org.cognity.neural.init.UniformInitializer

All Implemented Interfaces:

NeuralNetInitializer

public final class UniformInitializer
extends Object
implements NeuralNetInitializer

The uniform initializer sets every weight and bias to a random number from range [min, max). Every number in this range has equal probability of being chosen.

Author:

Artur Markiewicz

Constructor Summary

Constructors

Constructor and Description
UniformInitializer()
UniformInitializer(double min, double max) Creates a new initializer using specified range.

Method Summary

Methods

Modifier and Type	Method and Description
void	init(NeuralNet net) Initializes specified network.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

UniformInitializer

public UniformInitializer()

UniformInitializer

public UniformInitializer(double min,
                  double max)

Creates a new initializer using specified range.

Parameters:

min - the lower bound of the range

max - the upper bound of the range

Method Detail

init

public void init(NeuralNet net)

Description copied from interface: NeuralNetInitializer

Initializes specified network.

Specified by:

init in interface NeuralNetInitializer

Parameters:

net - the neural network to initialize