Class ActivationFunction

java.lang.Object

smile.deep.activation.ActivationFunction

All Implemented Interfaces:

Serializable, Function<Tensor,Tensor>, Layer

Direct Known Subclasses:

GELU, GLU, HardShrink, LeakyReLU, LogSigmoid, LogSoftmax, ReLU, Sigmoid, SiLU, Softmax, SoftShrink, Tanh, TanhShrink

public abstract class ActivationFunction
extends Object
implements Layer, Serializable

The activation function. It also implements the layer interface so that it can be treated added into a network as a layer.

See Also:

• Serialized Form

Constructor Summary

Constructors

Constructor	Description
ActivationFunction(String name, boolean inplace)	Constructor.

Method Summary

Modifier and Type	Method	Description
org.bytedeco.pytorch.Module	asTorch()	Returns the PyTorch Module object.
boolean	isInplace()	Returns true if the operation executes in-place.
String	name()	Returns the name of activation function.

Methods inherited from class Object

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

Methods inherited from interface Function

andThen, compose

Methods inherited from interface Layer

apply, forward, to, to

Constructor Details

ActivationFunction

public ActivationFunction(String name,
 boolean inplace)

Constructor.

Parameters:

name - the function name.

inplace - true if the operation executes in-place.

Method Details

name

public String name()

Returns the name of activation function.

Returns:

the name of activation function.

isInplace

public boolean isInplace()

Returns true if the operation executes in-place.

Returns:

true if the operation executes in-place.

asTorch

public org.bytedeco.pytorch.Module asTorch()

Description copied from interface: Layer

Returns the PyTorch Module object.

Specified by:

asTorch in interface Layer

Returns:

the PyTorch Module object.