org.deeplearning4j.nn.layers.normalization

Class CudnnLocalResponseNormalizationHelper

java.lang.Object

org.deeplearning4j.nn.layers.BaseCudnnHelper

org.deeplearning4j.nn.layers.normalization.CudnnLocalResponseNormalizationHelper

All Implemented Interfaces:

org.deeplearning4j.nn.layers.LayerHelper, org.deeplearning4j.nn.layers.normalization.LocalResponseNormalizationHelper

public class CudnnLocalResponseNormalizationHelper
extends BaseCudnnHelper
implements org.deeplearning4j.nn.layers.normalization.LocalResponseNormalizationHelper

cuDNN-based helper for the local response normalization layer.

Author:

saudet

Nested Class Summary

Nested classes/interfaces inherited from class org.deeplearning4j.nn.layers.BaseCudnnHelper

BaseCudnnHelper.CudnnContext, BaseCudnnHelper.DataCache, BaseCudnnHelper.TensorArray

Field Summary

Fields inherited from class org.deeplearning4j.nn.layers.BaseCudnnHelper

alpha, beta, dataType, dataTypeSize, sizeInBytes, TENSOR_FORMAT

Constructor Summary

Constructors

Constructor and Description
CudnnLocalResponseNormalizationHelper()

Method Summary

Modifier and Type	Method and Description
INDArray	activate(INDArray input, boolean training, double k, double n, double alpha, double beta, org.deeplearning4j.nn.workspace.LayerWorkspaceMgr workspaceMgr)
org.nd4j.linalg.primitives.Pair<org.deeplearning4j.nn.gradient.Gradient,INDArray>	backpropGradient(INDArray input, INDArray epsilon, double k, double n, double alpha, double beta, org.deeplearning4j.nn.workspace.LayerWorkspaceMgr workspaceMgr)
boolean	checkSupported(double k, double n, double alpha, double beta)
Map<String,Long>	helperMemoryUse()

Methods inherited from class org.deeplearning4j.nn.layers.BaseCudnnHelper

adaptForTensorDescr, checkCuda, checkCudnn, checkSupported, toCudnnDataType

Methods inherited from class java.lang.Object

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

Constructor Detail

CudnnLocalResponseNormalizationHelper

public CudnnLocalResponseNormalizationHelper()

Method Detail

checkSupported

public boolean checkSupported(double k,
                              double n,
                              double alpha,
                              double beta)

Specified by:

checkSupported in interface org.deeplearning4j.nn.layers.normalization.LocalResponseNormalizationHelper

backpropGradient

public org.nd4j.linalg.primitives.Pair<org.deeplearning4j.nn.gradient.Gradient,INDArray> backpropGradient(INDArray input,
                                                                                                          INDArray epsilon,
                                                                                                          double k,
                                                                                                          double n,
                                                                                                          double alpha,
                                                                                                          double beta,
                                                                                                          org.deeplearning4j.nn.workspace.LayerWorkspaceMgr workspaceMgr)

Specified by:

backpropGradient in interface org.deeplearning4j.nn.layers.normalization.LocalResponseNormalizationHelper

activate

public INDArray activate(INDArray input,
                         boolean training,
                         double k,
                         double n,
                         double alpha,
                         double beta,
                         org.deeplearning4j.nn.workspace.LayerWorkspaceMgr workspaceMgr)

Specified by:

activate in interface org.deeplearning4j.nn.layers.normalization.LocalResponseNormalizationHelper

helperMemoryUse

public Map<String,Long> helperMemoryUse()

Specified by:

helperMemoryUse in interface org.deeplearning4j.nn.layers.LayerHelper