gov.sandia.cognition.learning.algorithm.regression

Class AbstractParameterCostMinimizer<ResultType extends VectorizableVectorFunction,CostFunctionType extends SupervisedCostFunction<Vector,Vector>>

java.lang.Object

gov.sandia.cognition.util.AbstractCloneableSerializable

gov.sandia.cognition.algorithm.AbstractIterativeAlgorithm

gov.sandia.cognition.algorithm.AbstractAnytimeAlgorithm<ResultType>

gov.sandia.cognition.learning.algorithm.AbstractAnytimeBatchLearner<java.util.Collection<? extends InputOutputPair<? extends InputType,OutputType>>,ResultType>

gov.sandia.cognition.learning.algorithm.AbstractAnytimeSupervisedBatchLearner<Vector,Vector,ResultType>

gov.sandia.cognition.learning.algorithm.regression.AbstractParameterCostMinimizer<ResultType,CostFunctionType>

Type Parameters:

ResultType - Type of result to expect, such as GradientDescendable

CostFunctionType - Type of cost function to use, such as SumSquaredErrorCostFunction

All Implemented Interfaces:

AnytimeAlgorithm<ResultType>, IterativeAlgorithm, MeasurablePerformanceAlgorithm, StoppableAlgorithm, AnytimeBatchLearner<java.util.Collection<? extends InputOutputPair<? extends Vector,Vector>>,ResultType>, BatchCostMinimizationLearner<java.util.Collection<? extends InputOutputPair<? extends Vector,Vector>>,ResultType>, BatchLearner<java.util.Collection<? extends InputOutputPair<? extends Vector,Vector>>,ResultType>, ParameterCostMinimizer<ResultType>, SupervisedBatchLearner<Vector,Vector,ResultType>, CloneableSerializable, java.io.Serializable, java.lang.Cloneable

Direct Known Subclasses:

LeastSquaresEstimator

public abstract class AbstractParameterCostMinimizer<ResultType extends VectorizableVectorFunction,CostFunctionType extends SupervisedCostFunction<Vector,Vector>>
extends AbstractAnytimeSupervisedBatchLearner<Vector,Vector,ResultType>
implements BatchCostMinimizationLearner<java.util.Collection<? extends InputOutputPair<? extends Vector,Vector>>,ResultType>, ParameterCostMinimizer<ResultType>

Partial implementation of ParameterCostMinimizer.

Since:

2.1

Author:

Kevin R. Dixon

See Also:

Serialized Form

Field Summary

Fields

Modifier and Type	Field and Description
static int	DEFAULT_MAX_ITERATIONS Default maximum number of iterations before stopping 1000
static double	DEFAULT_TOLERANCE Default convergence criterion 1.0E-7

Fields inherited from class gov.sandia.cognition.learning.algorithm.AbstractAnytimeBatchLearner

data, keepGoing

Fields inherited from class gov.sandia.cognition.algorithm.AbstractAnytimeAlgorithm

maxIterations

Fields inherited from class gov.sandia.cognition.algorithm.AbstractIterativeAlgorithm

DEFAULT_ITERATION, iteration

Constructor Summary

Constructors

Constructor and Description
AbstractParameterCostMinimizer(CostFunctionType costFunction, int maxIterations, double tolerance) Creates a new instance of AbstractParameterCostMinimizer

Method Summary

Modifier and Type	Method and Description
CostFunctionType	getCostFunction() Gets the cost function that the learner is minimizing.
ResultType	getObjectToOptimize() Getter for objectToOptimize
NamedValue<java.lang.Double>	getPerformance() Gets the name-value pair that describes the current performance of the algorithm.
ResultType	getResult() Gets the current result of the algorithm.
protected java.lang.Double	getResultCost() Getter for resultCost
double	getTolerance() Getter for tolerance
void	setCostFunction(CostFunctionType costFunction) Setter for costFunction
void	setObjectToOptimize(ResultType objectToOptimize) Setter for objectToOptimize
protected void	setResult(ResultType result) Setter for result
protected void	setResultCost(java.lang.Double resultCost) Setter for resultCost
void	setTolerance(double tolerance) Setter for tolerance

Methods inherited from class gov.sandia.cognition.learning.algorithm.AbstractAnytimeBatchLearner

cleanupAlgorithm, clone, getData, getKeepGoing, initializeAlgorithm, learn, setData, setKeepGoing, step, stop

Methods inherited from class gov.sandia.cognition.algorithm.AbstractAnytimeAlgorithm

getMaxIterations, isResultValid, setMaxIterations

Methods inherited from class gov.sandia.cognition.algorithm.AbstractIterativeAlgorithm

addIterativeAlgorithmListener, fireAlgorithmEnded, fireAlgorithmStarted, fireStepEnded, fireStepStarted, getIteration, getListeners, removeIterativeAlgorithmListener, setIteration, setListeners

Methods inherited from class java.lang.Object

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

Methods inherited from interface gov.sandia.cognition.learning.algorithm.BatchCostMinimizationLearner

learn

Methods inherited from interface gov.sandia.cognition.util.CloneableSerializable

clone

Methods inherited from interface gov.sandia.cognition.algorithm.AnytimeAlgorithm

getMaxIterations, setMaxIterations

Methods inherited from interface gov.sandia.cognition.algorithm.IterativeAlgorithm

addIterativeAlgorithmListener, getIteration, removeIterativeAlgorithmListener

Methods inherited from interface gov.sandia.cognition.algorithm.StoppableAlgorithm

isResultValid, stop

Field Detail

DEFAULT_TOLERANCE

public static final double DEFAULT_TOLERANCE

Default convergence criterion 1.0E-7

See Also:

Constant Field Values

DEFAULT_MAX_ITERATIONS

public static final int DEFAULT_MAX_ITERATIONS

Default maximum number of iterations before stopping 1000

See Also:

Constant Field Values

Constructor Detail

AbstractParameterCostMinimizer

public AbstractParameterCostMinimizer(CostFunctionType costFunction,
                                      int maxIterations,
                                      double tolerance)

Creates a new instance of AbstractParameterCostMinimizer

Parameters:

costFunction - Cost function that computes the cost of the object to optimize

maxIterations - Maximum number of iterations before stopping

tolerance - Stopping criterion for the algorithm, typically ~1e-5

Method Detail

getObjectToOptimize

public ResultType getObjectToOptimize()

Getter for objectToOptimize

Specified by:

getObjectToOptimize in interface ParameterCostMinimizer<ResultType extends VectorizableVectorFunction>

Returns:

Vectorizable whose parameters result minimize the cost function

setObjectToOptimize

public void setObjectToOptimize(ResultType objectToOptimize)

Setter for objectToOptimize

Specified by:

setObjectToOptimize in interface ParameterCostMinimizer<ResultType extends VectorizableVectorFunction>

Parameters:

objectToOptimize - Vectorizable whose parameters result minimize the cost function

getResult

public ResultType getResult()

Description copied from interface: AnytimeAlgorithm

Gets the current result of the algorithm.

Specified by:

getResult in interface AnytimeAlgorithm<ResultType extends VectorizableVectorFunction>

Returns:

Current result of the algorithm.

setResult

protected void setResult(ResultType result)

Setter for result

Parameters:

result - Result to return

getTolerance

public double getTolerance()

Getter for tolerance

Returns:

Stopping criterion for the algorithm, typically ~1e-5

setTolerance

public void setTolerance(double tolerance)

Setter for tolerance

Parameters:

tolerance - Stopping criterion for the algorithm, typically ~1e-5

getCostFunction

public CostFunctionType getCostFunction()

Description copied from interface: BatchCostMinimizationLearner

Gets the cost function that the learner is minimizing.

Specified by:

getCostFunction in interface BatchCostMinimizationLearner<java.util.Collection<? extends InputOutputPair<? extends Vector,Vector>>,ResultType extends VectorizableVectorFunction>

Returns:

The CostFunction that the learner's algorithm is minimizing.

setCostFunction

public void setCostFunction(CostFunctionType costFunction)

Setter for costFunction

Parameters:

costFunction - Cost function that computes the cost of the object to optimize

getResultCost

protected java.lang.Double getResultCost()

Getter for resultCost

Returns:

Cost of the result

setResultCost

protected void setResultCost(java.lang.Double resultCost)

Setter for resultCost

Parameters:

resultCost - Cost of the result

getPerformance

public NamedValue<java.lang.Double> getPerformance()

Description copied from interface: MeasurablePerformanceAlgorithm

Gets the name-value pair that describes the current performance of the algorithm. For most algorithms, this is the value that they are attempting to optimize.

Specified by:

getPerformance in interface MeasurablePerformanceAlgorithm

Returns:

The name-value pair that describes the current performance of the algorithm.