org.deeplearning4j.nn.conf.weightnoise

Interface IWeightNoise

All Superinterfaces:

Cloneable, Serializable

All Known Implementing Classes:

DropConnect, WeightNoise

public interface IWeightNoise
extends Serializable, Cloneable

IWeightNoise instances operate on an weight array(s), modifying values at training time or test time, before they are used. Note that the weights are copied before being modified - the original parameters are not changed. However, if the pameters are not changed, the original array is returned. This interface can be used to implement functionality like DropConnect, weight quantization and weight noise.

Author:

Alex Black

Method Summary

Modifier and Type	Method and Description
IWeightNoise	clone()
org.nd4j.linalg.api.ndarray.INDArray	getParameter(Layer layer, String paramKey, int iteration, int epoch, boolean train, LayerWorkspaceMgr workspaceMgr) Get the parameter, after applying weight noise

Method Detail

getParameter

org.nd4j.linalg.api.ndarray.INDArray getParameter(Layer layer,
                                                  String paramKey,
                                                  int iteration,
                                                  int epoch,
                                                  boolean train,
                                                  LayerWorkspaceMgr workspaceMgr)

Get the parameter, after applying weight noise

Parameters:

layer - Layer to get the parameter for

paramKey - Parameter key

iteration - Iteration number

epoch - Epoch number

train - If true: training. False: at test time

Returns:

Parameter, after applying weight noise

clone

IWeightNoise clone()