org.encog.neural.networks.training.propagation.resilient
public class ResilientPropagation extends Propagation
Modifier and Type | Field and Description |
---|---|
static String |
LAST_GRADIENTS
Continuation tag for the last gradients.
|
static String |
UPDATE_VALUES
Continuation tag for the last values.
|
gradients, network
Constructor and Description |
---|
ResilientPropagation(ContainsFlat network,
MLDataSet training)
Construct an RPROP trainer, allows an OpenCL device to be specified.
|
ResilientPropagation(ContainsFlat network,
MLDataSet training,
double initialUpdate,
double maxStep)
Construct a resilient training object, allow the training parameters to
be specified.
|
Modifier and Type | Method and Description |
---|---|
boolean |
canContinue() |
RPROPType |
getRPROPType() |
double[] |
getUpdateValues() |
void |
initOthers()
Perform training method specific init.
|
boolean |
isValidResume(TrainingContinuation state)
Determine if the specified continuation object is valid to resume with.
|
TrainingContinuation |
pause()
Pause the training.
|
void |
postIteration()
Call the strategies after an iteration.
|
void |
resume(TrainingContinuation state)
Resume training.
|
void |
setRPROPType(RPROPType t)
Set the type of RPROP to use.
|
double |
updateiWeightMinus(double[] gradients,
double[] lastGradient,
int index) |
double |
updateiWeightPlus(double[] gradients,
double[] lastGradient,
int index) |
double |
updateWeight(double[] gradients,
double[] lastGradient,
int index)
Calculate the amount to change the weight by.
|
double |
updateWeightMinus(double[] gradients,
double[] lastGradient,
int index) |
double |
updateWeightPlus(double[] gradients,
double[] lastGradient,
int index) |
calculateGradients, finishTraining, fixFlatSpot, getBatchSize, getCurrentFlatNetwork, getLastGradient, getMethod, getThreadCount, iteration, iteration, learn, learnLimited, report, rollIteration, setBatchSize, setErrorFunction, setThreadCount
addStrategy, getError, getImplementationType, getIteration, getStrategies, getTraining, isTrainingDone, preIteration, setError, setIteration, setTraining
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait
addStrategy, getError, getImplementationType, getIteration, getStrategies, getTraining, isTrainingDone, setError, setIteration
public static final String LAST_GRADIENTS
public static final String UPDATE_VALUES
public ResilientPropagation(ContainsFlat network, MLDataSet training)
network
- The network to train.training
- The training data to use.public ResilientPropagation(ContainsFlat network, MLDataSet training, double initialUpdate, double maxStep)
network
- The network to train.training
- The training set to use.initialUpdate
- The initial update values, this is the amount that the deltas
are all initially set to.maxStep
- The maximum that a delta can reach.public boolean canContinue()
public boolean isValidResume(TrainingContinuation state)
state
- The continuation object to check.public TrainingContinuation pause()
public void resume(TrainingContinuation state)
state
- The training state to return to.public void setRPROPType(RPROPType t)
t
- The type.public RPROPType getRPROPType()
public void initOthers()
initOthers
in class Propagation
public double updateWeight(double[] gradients, double[] lastGradient, int index)
updateWeight
in class Propagation
gradients
- The gradients.lastGradient
- The last gradients.index
- The index to update.public double updateWeightPlus(double[] gradients, double[] lastGradient, int index)
public double updateWeightMinus(double[] gradients, double[] lastGradient, int index)
public double updateiWeightPlus(double[] gradients, double[] lastGradient, int index)
public double updateiWeightMinus(double[] gradients, double[] lastGradient, int index)
public void postIteration()
postIteration
in class BasicTraining
public double[] getUpdateValues()
Copyright © 2014. All Rights Reserved.