org.deeplearning4j.nn.conf.layers

Class GlobalPoolingLayer.Builder

java.lang.Object

org.deeplearning4j.nn.conf.layers.Layer.Builder<GlobalPoolingLayer.Builder>

org.deeplearning4j.nn.conf.layers.GlobalPoolingLayer.Builder

Enclosing class:

GlobalPoolingLayer

public static class GlobalPoolingLayer.Builder
extends Layer.Builder<GlobalPoolingLayer.Builder>

Field Summary

Fields inherited from class org.deeplearning4j.nn.conf.layers.Layer.Builder

activationFn, adamMeanDecay, adamVarDecay, biasInit, biasLearningRate, dist, dropOut, epsilon, gradientNormalization, gradientNormalizationThreshold, l1, l1Bias, l2, l2Bias, layerName, learningRate, learningRatePolicy, learningRateSchedule, momentum, momentumAfter, rho, rmsDecay, updater, weightInit

Constructor Summary

Constructors

Constructor and Description
Builder()
Builder(PoolingType poolingType)

Method Summary

Modifier and Type	Method and Description
GlobalPoolingLayer	build()
GlobalPoolingLayer.Builder	collapseDimensions(boolean collapseDimensions) Whether to collapse dimensions when pooling or not.
GlobalPoolingLayer.Builder	pnorm(int pnorm) P-norm constant.
GlobalPoolingLayer.Builder	poolingDimensions(int... poolingDimensions) Pooling dimensions.
GlobalPoolingLayer.Builder	poolingType(PoolingType poolingType)

Methods inherited from class org.deeplearning4j.nn.conf.layers.Layer.Builder

activation, activation, activation, adamMeanDecay, adamVarDecay, biasInit, biasLearningRate, dist, dropOut, epsilon, gradientNormalization, gradientNormalizationThreshold, l1, l1Bias, l2, l2Bias, learningRate, learningRateDecayPolicy, learningRateSchedule, momentum, momentumAfter, name, rho, rmsDecay, updater, weightInit

Methods inherited from class java.lang.Object

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

Constructor Detail

Builder

public Builder()

Builder

public Builder(PoolingType poolingType)

Method Detail

poolingDimensions

public GlobalPoolingLayer.Builder poolingDimensions(int... poolingDimensions)

Pooling dimensions. Note: most of the time, this doesn't need to be set, and the defaults can be used. Default for RNN data: pooling dimension 2 (time). Default for CNN data: pooling dimensions 2,3 (height and width)

Parameters:

poolingDimensions - Pooling dimensions to use

poolingType

public GlobalPoolingLayer.Builder poolingType(PoolingType poolingType)

Parameters:

poolingType - Pooling type for global pooling

collapseDimensions

public GlobalPoolingLayer.Builder collapseDimensions(boolean collapseDimensions)

Whether to collapse dimensions when pooling or not. Usually you *do* want to do this. Default: true. If true:
- 3d (time series) input with shape [minibatchSize, vectorSize, timeSeriesLength] -> 2d output [minibatchSize, vectorSize]
- 4d (CNN) input with shape [minibatchSize, depth, height, width] -> 2d output [minibatchSize, depth]
If false:
- 3d (time series) input with shape [minibatchSize, vectorSize, timeSeriesLength] -> 3d output [minibatchSize, vectorSize, 1]
- 4d (CNN) input with shape [minibatchSize, depth, height, width] -> 2d output [minibatchSize, depth, 1, 1]

Parameters:

collapseDimensions - Whether to collapse the dimensions or not

pnorm

public GlobalPoolingLayer.Builder pnorm(int pnorm)

P-norm constant. Only used if using PoolingType.PNORM for the pooling type

Parameters:

pnorm - P-norm constant

build

public GlobalPoolingLayer build()

Specified by:

build in class Layer.Builder<GlobalPoolingLayer.Builder>