org.nd4j.linalg.learning

Interface GradientUpdater

All Superinterfaces:

Serializable

All Known Implementing Classes:

AdaDelta, AdaGrad, Adam, Nesterovs, RmsProp, Sgd

public interface GradientUpdater
extends Serializable

Gradient modifications: Calculates an update and tracks related information for gradient changes over time for handling updates.

Author:

Adam Gibson

Method Summary

Modifier and Type	Method and Description
GradientUpdaterAggregator	getAggregator(boolean addThis) Get a GradientUpdaterAggregator.
INDArray	getGradient(INDArray gradient, int iteration) Modify the gradient to be an update
void	update(Object... args) update(learningRate,momentum)

Method Detail

update

void update(Object... args)

update(learningRate,momentum)

Parameters:

args -

getGradient

INDArray getGradient(INDArray gradient,
                     int iteration)

Modify the gradient to be an update

Parameters:

gradient - the gradient to modify

iteration -

Returns:

the modified gradient

getAggregator

GradientUpdaterAggregator getAggregator(boolean addThis)

Get a GradientUpdaterAggregator. The aggregator is used (typically in distributed learning scenarios) to combine separate GradientUpdater instances for different networks (usually by averaging).

Parameters:

addThis - If true: return a GradientUpdaterAggregator with the GradientUpdater already added. If false: return an empty (uninitialized) GradientUpdaterAggregator