casa::NonLinearFitLM< T > Class Template Reference

Solve non-linear fit with Levenberg-Marquardt method.

More...

#include <NonLinearFitLM.h>

Inheritance diagram for casa::NonLinearFitLM< T >:

List of all members.

Public Member Functions
	NonLinearFitLM (Bool svd=False)
	Create a fitter: the normal way to generate a fitter object.
	NonLinearFitLM (const NonLinearFitLM &other)
	Copy constructor (deep copy)
NonLinearFitLM &	operator= (const NonLinearFitLM &other)
	Assignment (deep copy)
virtual	~NonLinearFitLM ()
	Destructor.
Protected Member Functions
virtual Bool	fitIt (Vector< typename FunctionTraits< T >::BaseType > &sol, const Array< typename FunctionTraits< T >::BaseType > &x, const Vector< typename FunctionTraits< T >::BaseType > &y, const Vector< typename FunctionTraits< T >::BaseType > *const sigma, const Vector< Bool > *const mask=0)
	Generalised fitter.
Private Attributes
Double	lamda_p
	The parameter that makes this the Levenberg-Marquardt method.
Double	fitit_p
	The current fit state.

---

Detailed Description

template<class T>
class casa::NonLinearFitLM< T >

Solve non-linear fit with Levenberg-Marquardt method.

Review Status

Reviewed By:

wbrouw

Date Reviewed:

2004/06/15

Test programs:

tNonLinearFitLM

Prerequisite

• NonLinearFit
• Fitting

Etymology

This class uses the Levenberg-Marquardt method to solve the non-linear least-squares fit problem hence NonLinearFitLM

Synopsis

NOTE: Constraints added. Documentation out of date at moment, check the tLinearFitSVD and tNonLinearFirLM programs for examples.

See the NonLinearFit class for a general description.

This class is derived from the general NonLinearFit class. It does a non-linear least-squares fit using the Levenberg-Marquardt method.

See Numerical Recipes for more information on the Levenberg-Marquardt method.

Template Type Argument Requirements (T)

• Float
• Double
• Complex
• DComplex

Motivation

Levenberg-Marquardt method is a standard method for non-linear least-squares fits. It works well in practice over a wide range of problems.

Example

Definition at line 87 of file NonLinearFitLM.h.

---

Constructor & Destructor Documentation

template<class T>

casa::NonLinearFitLM< T >::NonLinearFitLM

(

Bool

svd = False

)

[explicit]

Create a fitter: the normal way to generate a fitter object.

Necessary data will be deduced from the Functional provided with setFunction(). Optionally, a fitter with SVD behaviour

template<class T>

casa::NonLinearFitLM< T >::NonLinearFitLM

(

const NonLinearFitLM< T > &

other

)

Copy constructor (deep copy)

template<class T>

virtual casa::NonLinearFitLM< T >::~NonLinearFitLM

(

)

[virtual]

Destructor.

---

Member Function Documentation

template<class T>

virtual Bool casa::NonLinearFitLM< T >::fitIt	(	Vector< typename FunctionTraits< T >::BaseType > &	sol,
		const Array< typename FunctionTraits< T >::BaseType > &	x,
		const Vector< typename FunctionTraits< T >::BaseType > &	y,
		const Vector< typename FunctionTraits< T >::BaseType > *const	sigma,
		const Vector< Bool > *const	mask = 0
	)		[protected, virtual]

Generalised fitter.

Implements casa::NonLinearFit< T >.

template<class T>

NonLinearFitLM& casa::NonLinearFitLM< T >::operator=

(

const NonLinearFitLM< T > &

other

)

Assignment (deep copy)

---

Member Data Documentation

template<class T>

Double casa::NonLinearFitLM< T >::fitit_p [private]

The current fit state.

Definition at line 119 of file NonLinearFitLM.h.

template<class T>

Double casa::NonLinearFitLM< T >::lamda_p [private]

The parameter that makes this the Levenberg-Marquardt method.

Definition at line 117 of file NonLinearFitLM.h.

---

The documentation for this class was generated from the following file:

• casacore/scimath/Fitting/NonLinearFitLM.h