jsat.classifiers.neuralnetwork.initializers

Interface BiastInitializer

All Superinterfaces:

java.io.Serializable

All Known Implementing Classes:

ConstantInit, GaussianNormalInit, TanhInitializer

public interface BiastInitializer
extends java.io.Serializable

This interface specifies the method of initializing the bias connections in a neural network.

Method Summary

Modifier and Type	Method and Description
BiastInitializer	clone()
void	init(Vec b, int fanIn, java.util.Random rand) Performs the initialization of the given vector of bias values

Method Detail

init

void init(Vec b,
          int fanIn,
          java.util.Random rand)

Performs the initialization of the given vector of bias values

Parameters:

b - the vector to store the biases in

fanIn - the number of connections coming into the layer that these biases are for.

rand - the source of randomness for initialization

clone

BiastInitializer clone()