org.apache.ignite.ml.nn.architecture

Class TransformationLayerArchitecture

java.lang.Object

org.apache.ignite.ml.nn.architecture.LayerArchitecture

org.apache.ignite.ml.nn.architecture.TransformationLayerArchitecture

All Implemented Interfaces:

Serializable

public class TransformationLayerArchitecture
extends LayerArchitecture

Class encapsulation architecture of transformation layer (i.e. non-input layer).

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor and Description
TransformationLayerArchitecture(int neuronsCnt, boolean hasBias, IgniteDifferentiableDoubleToDoubleFunction activationFunction) Construct TransformationLayerArchitecture.

Method Summary

Modifier and Type	Method and Description
IgniteDifferentiableDoubleToDoubleFunction	activationFunction() Get activation function for this layer.
boolean	hasBias() Checks if this layer has a bias.

Methods inherited from class org.apache.ignite.ml.nn.architecture.LayerArchitecture

neuronsCount

Methods inherited from class java.lang.Object

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

Constructor Detail

TransformationLayerArchitecture

public TransformationLayerArchitecture(int neuronsCnt,
                                       boolean hasBias,
                                       IgniteDifferentiableDoubleToDoubleFunction activationFunction)

Construct TransformationLayerArchitecture.

Parameters:

neuronsCnt - Count of neurons in this layer.

hasBias - Flag indicating presence of bias in layer.

activationFunction - Activation function for layer.

Method Detail

hasBias

public boolean hasBias()

Checks if this layer has a bias.

Returns:

Value of predicate "this layer has a bias".

activationFunction

public IgniteDifferentiableDoubleToDoubleFunction activationFunction()

Get activation function for this layer.

Returns:

Activation function for this layer.