ai.djl.nn.recurrent

Class RNN.Builder

java.lang.Object

ai.djl.nn.recurrent.RecurrentBlock.BaseBuilder<RNN.Builder>

ai.djl.nn.recurrent.RNN.Builder

Enclosing class:

RNN

public static final class RNN.Builder
extends RecurrentBlock.BaseBuilder<RNN.Builder>

The Builder to construct a RNN type of Block.

Field Summary

Fields inherited from class ai.djl.nn.recurrent.RecurrentBlock.BaseBuilder

activation, clipLstmState, dropRate, lstmStateClipMax, lstmStateClipMin, numStackedLayers, stateOutputs, stateSize, useBidirectional, useSequenceLength

Constructor Summary

Constructors

Constructor and Description
Builder()

Method Summary

Modifier and Type	Method and Description
RNN	build() Builds a RNN block.
protected RNN.Builder	self()
RNN.Builder	setActivation(RNN.Activation activation) Sets the activation for the RNN - ReLu or Tanh.

Methods inherited from class ai.djl.nn.recurrent.RecurrentBlock.BaseBuilder

optBidrectional, optDropRate, optStateOutput, setNumStackedLayers, setSequenceLength, setStateSize

Methods inherited from class java.lang.Object

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

Constructor Detail

Builder

public Builder()

Method Detail

self

protected RNN.Builder self()

Specified by:

self in class RecurrentBlock.BaseBuilder<RNN.Builder>

setActivation

public RNN.Builder setActivation(RNN.Activation activation)

Sets the activation for the RNN - ReLu or Tanh.

Parameters:

activation - the activation

Returns:

this Builder

build

public RNN build()

Builds a RNN block.

Returns:

the RNN block