Package org.omegahat.Numerics.Optimizers

A collection of interfaces and some basic classes to perform iterations in the context of optimization.

See:
Description

Interface Summary

ConvergenceChecker
ConvergenceListenerInt	Interface to provide tests of numeric convergence for optimization.
CounterListener	Interface to provide counting and limits for optimization.
Derivatives
DerivativeValues
EnumerableOptimizerAlgorithm
Gradient
Hessian	Basic interface defining operations needed by an object representing a second derivative of a function so that it can be used in the generic optimization algorithms.
InformationListener	Classes should implement this interface if they are to be used to provide information to the user about the progress of an optimization.
IterationCounterInt	The interface for IterationCounter objects.
IterationListener	Classes implementing this interface will provide listeners that control the iteration of the optimizer.
LineSearch	An interface for algorithms that implement a search for the optimum value of a function along a specified direction.
LineStep	The interface to store information needed about a candidate parameter set during a line search.
ModelObjective	Interface for objects that define an objective function to be optimized.
ModelPoint	The interface to define a point estimate for a numerical model.
ModelPointNumericInt	An extension to ModelPoint which provides methods for computing numerical quantities such as the gradient and hessian.
Optimizable
OptimizerAlgorithm	An interface for the algorithms that carry out optimization.
OptimizerIterator
ParameterSet

Class Summary

BackgroundTaskManager
BasicDerivatives
BasicDerivativeValues	Represents a particular value of the first and second derivatives of a function, stored together to facilitate computing them in a single to avoid repeating computations for each.
BasicOptimizer
ConvergenceGradient	This listener implements the convergence test that the norm of the gradient is less than the tolerance.
ConvergenceListener	Basic class to provide tests of numeric convergence for optimization.
DefaultListener	A default, general purpose controller for optimization iteration.
IterationCounter	A basic, class for counter listeners.
IterationEvent	An event that signals that something potentially of interest to others has happened in the iteration of an optimization.
IterationEventTask
LineSearchBasic	the basic line search algorithm structure, without any particular algorithm.
LineSearchWolfe	a line search algorithm that finds a point with good enough improvement and slope.
LineStepBasic	the basic representation of a step in a line search.
ListGradient
ModelObjectiveOmegahat	Create an object to evaluate a model function and possible gradients, starting from one or more Omegahat expressions.
ModelPointDatabase	Class to allow ModelPoint objects to associate parameters with names.
ModelPointNumeric	The basic numeric model point class.
NotifyingOptimizer
Optimizer	The standard optimizer class.
OptimizerAlgorithmBasic	A class with some basic methods for any optimizer.
OptimizerAlgorithmBFGS	The BFGS method for unconstrained optimization.
OptimizerBasic	The basic iterative optimizer, using an OptimizerAlgorithm for iteration and a ModelObjective to define the particular model to optmize.
OptimizerNotifying	Optimizer objects that notify listeners during iteration.
PlotListener	Objects to set up and maintain dynamic time-series stack plots to follow optimization.
Utils	Utility routines for standard results.

Exception Summary

InterruptedIterationException
ModelEvaluationException

Package org.omegahat.Numerics.Optimizers Description

A collection of interfaces and some basic classes to perform iterations in the context of optimization. Handles notification of different types of events with both blocking and non-blacking listeners, and convergence decision making.