Package org.hipparchus.optim.nonlinear.vector.leastsquares

This package provides algorithms that minimize the residuals between observations and model values.

See: Description

Interface Summary

Interface	Description
LeastSquaresOptimizer	An algorithm that can be applied to a non-linear least squares problem.
LeastSquaresOptimizer.Optimum	The optimum found by the optimizer.
LeastSquaresProblem	The data necessary to define a non-linear least squares problem.
LeastSquaresProblem.Evaluation	An evaluation of a LeastSquaresProblem at a particular point.
MultivariateJacobianFunction	A interface for functions that compute a vector of values and can compute their derivatives (Jacobian).
ParameterValidator	Interface for validating a set of model parameters.
ValueAndJacobianFunction	A interface for functions that compute a vector of values and can compute their derivatives (Jacobian).

Class Summary

Class	Description
AbstractEvaluation	An implementation of LeastSquaresProblem.Evaluation that is designed for extension.
EvaluationRmsChecker	Check if an optimization has converged based on the change in computed RMS.
GaussNewtonOptimizer	Gauss-Newton least-squares solver.
LeastSquaresAdapter	An adapter that delegates to another implementation of LeastSquaresProblem.
LeastSquaresBuilder	A mutable builder for LeastSquaresProblems.
LeastSquaresFactory	A Factory for creating LeastSquaresProblems.
LevenbergMarquardtOptimizer	This class solves a least-squares problem using the Levenberg-Marquardt algorithm.

Enum Summary

Enum	Description
GaussNewtonOptimizer.Decomposition	Deprecated Use MatrixDecomposer instead.

Package org.hipparchus.optim.nonlinear.vector.leastsquares Description

This package provides algorithms that minimize the residuals between observations and model values. The least-squares optimizers minimize the distance (called cost or χ²) between model and observations.
Algorithms in this category need access to a problem (represented by a LeastSquaresProblem). Such a model predicts a set of values which the algorithm tries to match with a set of given set of observed values.
The problem can be created progressively using a builder or it can be created at once using a factory.