org.apache.commons.math3.fitting.leastsquares

Interface LeastSquaresProblem

All Superinterfaces:

OptimizationProblem<LeastSquaresProblem.Evaluation>

All Known Implementing Classes:

LeastSquaresAdapter

public interface LeastSquaresProblem
extends OptimizationProblem<LeastSquaresProblem.Evaluation>

The data necessary to define a non-linear least squares problem.

Includes the observed values, computed model function, and convergence/divergence criteria. Weights are implicit in LeastSquaresProblem.Evaluation.getResiduals() and LeastSquaresProblem.Evaluation.getJacobian().

Instances are typically either created progressively using a builder or created at once using a factory.

Since:

3.3

See Also:

LeastSquaresBuilder, LeastSquaresFactory, LeastSquaresAdapter

Nested Class Summary

Nested Classes

Modifier and Type	Interface and Description
static interface	LeastSquaresProblem.Evaluation An evaluation of a LeastSquaresProblem at a particular point.

Method Summary

Modifier and Type	Method and Description
LeastSquaresProblem.Evaluation	evaluate(RealVector point) Evaluate the model at the specified point.
int	getObservationSize() Get the number of observations (rows in the Jacobian) in this problem.
int	getParameterSize() Get the number of parameters (columns in the Jacobian) in this problem.
RealVector	getStart() Gets the initial guess.

Methods inherited from interface org.apache.commons.math3.optim.OptimizationProblem

getConvergenceChecker, getEvaluationCounter, getIterationCounter

Method Detail

getStart

RealVector getStart()

Gets the initial guess.

Returns:

the initial guess values.

getObservationSize

int getObservationSize()

Get the number of observations (rows in the Jacobian) in this problem.

Returns:

the number of scalar observations

getParameterSize

int getParameterSize()

Get the number of parameters (columns in the Jacobian) in this problem.

Returns:

the number of scalar parameters

evaluate

LeastSquaresProblem.Evaluation evaluate(RealVector point)

Evaluate the model at the specified point.

Parameters:

point - the parameter values.

Returns:

the model's value and derivative at the given point.

Throws:

TooManyEvaluationsException - if the maximal number of evaluations (of the model vector function) is exceeded.