neureka.backend.main.implementations.fun

Class ScalarSoftplus

java.lang.Object

neureka.backend.main.implementations.fun.ScalarSoftplus

All Implemented Interfaces:

ScalarFun

public final class ScalarSoftplus
extends java.lang.Object
implements ScalarFun

SoftPlus is a smooth approximation to the ReLU function and can be used to constrain the output of a machine to always be positive.

Field Summary

Fields inherited from interface neureka.backend.main.implementations.fun.api.ScalarFun

ABSOLUTE, CBRT, COSINUS, EXP, GASU, GATU, GAUSSIAN, GAUSSIAN_FAST, GELU, IDENTITY, LOG10, LOGARITHM, QUADRATIC, RELU, SELU, SIGMOID, SILU, SINUS, SOFTPLUS, SOFTSIGN, SQRT, TANH, TANH_FAST

Constructor Summary

Constructors

Constructor and Description
ScalarSoftplus()

Method Summary

Modifier and Type	Method and Description
java.lang.String	activationCode()
java.lang.String	derivationCode()
CPUFun	getActivation()
CPUFun	getDerivative()
java.lang.String	id()

Methods inherited from class java.lang.Object

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

Methods inherited from interface neureka.backend.main.implementations.fun.api.ScalarFun

calculate

Constructor Detail

ScalarSoftplus

public ScalarSoftplus()

Method Detail

id

public java.lang.String id()

Specified by:

id in interface ScalarFun

activationCode

public java.lang.String activationCode()

Specified by:

activationCode in interface ScalarFun

derivationCode

public java.lang.String derivationCode()

Specified by:

derivationCode in interface ScalarFun

getActivation

public CPUFun getActivation()

Specified by:

getActivation in interface ScalarFun

getDerivative

public CPUFun getDerivative()

Specified by:

getDerivative in interface ScalarFun