org.apache.commons.math3.optim.univariate
Class MultiStartUnivariateOptimizer

java.lang.Object
  extended by org.apache.commons.math3.optim.BaseOptimizer<UnivariatePointValuePair>
      extended by org.apache.commons.math3.optim.univariate.UnivariateOptimizer
          extended by org.apache.commons.math3.optim.univariate.MultiStartUnivariateOptimizer

public class MultiStartUnivariateOptimizer
extends UnivariateOptimizer

Special implementation of the UnivariateOptimizer interface adding multi-start features to an existing optimizer.
This class wraps an optimizer in order to use it several times in turn with different starting points (trying to avoid being trapped in a local extremum when looking for a global one).

Since:
3.0
Version:
$Id: MultiStartUnivariateOptimizer.java 1435539 2013-01-19 13:27:24Z tn $

Field Summary
private  RandomGenerator generator
          Random generator for multi-start.
private  int maxEvalIndex
          Location in optimData where the updated maximum number of evaluations will be stored.
private  UnivariatePointValuePair[] optima
          Found optima.
private  OptimizationData[] optimData
          Optimization data.
private  UnivariateOptimizer optimizer
          Underlying classical optimizer.
private  int searchIntervalIndex
          Location in optimData where the updated start value will be stored.
private  int starts
          Number of starts to go.
private  int totalEvaluations
          Number of evaluations already performed for all starts.
 
Fields inherited from class org.apache.commons.math3.optim.BaseOptimizer
evaluations, iterations
 
Constructor Summary
MultiStartUnivariateOptimizer(UnivariateOptimizer optimizer, int starts, RandomGenerator generator)
          Create a multi-start optimizer from a single-start optimizer.
 
Method Summary
protected  UnivariatePointValuePair doOptimize()
          Performs the bulk of the optimization algorithm.
 int getEvaluations()
          Gets the number of evaluations of the objective function.
 UnivariatePointValuePair[] getOptima()
          Gets all the optima found during the last call to optimize.
 UnivariatePointValuePair optimize(OptimizationData... optData)
          Stores data and performs the optimization.
private  void sortPairs(GoalType goal)
          Sort the optima from best to worst, followed by null elements.
 
Methods inherited from class org.apache.commons.math3.optim.univariate.UnivariateOptimizer
computeObjectiveValue, getGoalType, getMax, getMin, getStartValue, parseOptimizationData
 
Methods inherited from class org.apache.commons.math3.optim.BaseOptimizer
getConvergenceChecker, getIterations, getMaxEvaluations, getMaxIterations, incrementEvaluationCount, incrementIterationCount
 
Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait
 

Field Detail

optimizer

private final UnivariateOptimizer optimizer
Underlying classical optimizer.


totalEvaluations

private int totalEvaluations
Number of evaluations already performed for all starts.


starts

private int starts
Number of starts to go.


generator

private RandomGenerator generator
Random generator for multi-start.


optima

private UnivariatePointValuePair[] optima
Found optima.


optimData

private OptimizationData[] optimData
Optimization data.


maxEvalIndex

private int maxEvalIndex
Location in optimData where the updated maximum number of evaluations will be stored.


searchIntervalIndex

private int searchIntervalIndex
Location in optimData where the updated start value will be stored.

Constructor Detail

MultiStartUnivariateOptimizer

public MultiStartUnivariateOptimizer(UnivariateOptimizer optimizer,
                                     int starts,
                                     RandomGenerator generator)
Create a multi-start optimizer from a single-start optimizer.

Parameters:
optimizer - Single-start optimizer to wrap.
starts - Number of starts to perform. If starts == 1, the optimize methods will return the same solution as optimizer would.
generator - Random generator to use for restarts.
Throws:
NotStrictlyPositiveException - if starts < 1.
Method Detail

getEvaluations

public int getEvaluations()
Gets the number of evaluations of the objective function. The number of evaluations corresponds to the last call to the optimize method. It is 0 if the method has not been called yet.

Overrides:
getEvaluations in class BaseOptimizer<UnivariatePointValuePair>
Returns:
the number of evaluations of the objective function.

getOptima

public UnivariatePointValuePair[] getOptima()
Gets all the optima found during the last call to optimize. The optimizer stores all the optima found during a set of restarts. The optimize method returns the best point only. This method returns all the points found at the end of each starts, including the best one already returned by the optimize method.
The returned array as one element for each start as specified in the constructor. It is ordered with the results from the runs that did converge first, sorted from best to worst objective value (i.e in ascending order if minimizing and in descending order if maximizing), followed by null elements corresponding to the runs that did not converge. This means all elements will be null if the optimize method did throw an exception. This also means that if the first element is not null, it is the best point found across all starts.

Returns:
an array containing the optima.
Throws:
MathIllegalStateException - if optimize has not been called.

optimize

public UnivariatePointValuePair optimize(OptimizationData... optData)
Stores data and performs the optimization.
The list of parameters is open-ended so that sub-classes can extend it with arguments specific to their concrete implementations.
When the method is called multiple times, instance data is overwritten only when actually present in the list of arguments: when not specified, data set in a previous call is retained (and thus is optional in subsequent calls).
Important note: Subclasses must override BaseOptimizer.parseOptimizationData(OptimizationData[]) if they need to register their own options; but then, they must also call super.parseOptimizationData(optData) within that method.

Overrides:
optimize in class UnivariateOptimizer
Parameters:
optData - Optimization data. In addition to those documented in BaseOptimizer, this method will register the following data:
Returns:
a point/value pair that satifies the convergence criteria.
Throws:
MathIllegalStateException - if optData does not contain an instance of MaxEval or SearchInterval.

doOptimize

protected UnivariatePointValuePair doOptimize()
Performs the bulk of the optimization algorithm.

Specified by:
doOptimize in class BaseOptimizer<UnivariatePointValuePair>
Returns:
the point/value pair giving the optimal value of the objective function.

sortPairs

private void sortPairs(GoalType goal)
Sort the optima from best to worst, followed by null elements.

Parameters:
goal - Goal type.


Copyright (c) 2003-2013 Apache Software Foundation