#include <NoiseModel.h>

Inheritance diagram for gtsam::noiseModel::Diagonal:

[legend]

Collaboration diagram for gtsam::noiseModel::Diagonal:

Public Types
typedef std::shared_ptr< Diagonal >	shared_ptr

Public Member Functions
	Diagonal ()

void	print (const std::string &name) const override

Vector	sigmas () const override
	Calculate standard deviations.

Vector	whiten (const Vector &v) const override
	Whiten an error vector.

Vector	unwhiten (const Vector &v) const override
	Unwhiten an error vector.

Matrix	Whiten (const Matrix &H) const override

void	WhitenInPlace (Matrix &H) const override

void	WhitenInPlace (Eigen::Block< Matrix > H) const override

double	sigma (size_t i) const

const Vector &	invsigmas () const

double	invsigma (size_t i) const

const Vector &	precisions () const

double	precision (size_t i) const

Matrix	R () const override

bool	equals (const Base &expected, double tol=1e-9) const override

void	WhitenSystem (std::vector< Matrix > &A, Vector &b) const override

void	WhitenSystem (Matrix &A, Vector &b) const override

void	WhitenSystem (Matrix &A1, Matrix &A2, Vector &b) const override

void	WhitenSystem (Matrix &A1, Matrix &A2, Matrix &A3, Vector &b) const override

virtual std::shared_ptr< Diagonal >	QR (Matrix &Ab) const

virtual Matrix	information () const
	Compute information matrix.

virtual Matrix	covariance () const
	Compute covariance matrix.

virtual bool	isConstrained () const
	true if a constrained noise model, saves slow/clumsy dynamic casting

virtual bool	isUnit () const
	true if a unit noise model, saves slow/clumsy dynamic casting

size_t	dim () const
	Dimensionality.

virtual double	squaredMahalanobisDistance (const Vector &v) const
	Squared Mahalanobis distance v'R'Rv = <Rv,R*v>

virtual double	mahalanobisDistance (const Vector &v) const
	Mahalanobis distance.

virtual double	loss (const double squared_distance) const
	loss function, input is Mahalanobis distance

virtual void	whitenInPlace (Vector &v) const

virtual void	whitenInPlace (Eigen::Block< Vector > &v) const

virtual void	unwhitenInPlace (Vector &v) const

virtual void	unwhitenInPlace (Eigen::Block< Vector > &v) const

virtual Vector	unweightedWhiten (const Vector &v) const

virtual double	weight (const Vector &v) const

Static Public Member Functions
static shared_ptr	Sigmas (const Vector &sigmas, bool smart=true)

static shared_ptr	Variances (const Vector &variances, bool smart=true)

static shared_ptr	Precisions (const Vector &precisions, bool smart=true)

static shared_ptr	SqrtInformation (const Matrix &R, bool smart=true)

static shared_ptr	Information (const Matrix &M, bool smart=true)

static shared_ptr	Covariance (const Matrix &covariance, bool smart=true)

Protected Member Functions
	Diagonal (const Vector &sigmas)

Protected Attributes
Vector	sigmas_

Vector	invsigmas_

Vector	precisions_

std::optional< Matrix >	sqrt_information_

size_t	dim_

Detailed Description

A diagonal noise model implements a diagonal covariance matrix, with the elements of the diagonal specified in a Vector. This class has no public constructors, instead, use the static constructor functions Sigmas etc...

Constructor & Destructor Documentation

◆ Diagonal() [1/2]

gtsam::noiseModel::Diagonal::Diagonal ( const Vector & sigmas )

protected

constructor to allow for disabling initialization of invsigmas

◆ Diagonal() [2/2]

gtsam::noiseModel::Diagonal::Diagonal ( )

constructor - no initializations, for serialization

Member Function Documentation

◆ Covariance()

static shared_ptr gtsam::noiseModel::Gaussian::Covariance	(	const Matrix &	covariance,
		bool	smart = `true`
	)

staticinherited

A Gaussian noise model created by specifying a covariance matrix.

Parameters

covariance	The square covariance Matrix
smart	check if can be simplified to derived class

◆ Information()

static shared_ptr gtsam::noiseModel::Gaussian::Information	(	const Matrix &	M,
		bool	smart = `true`
	)

staticinherited

A Gaussian noise model created by specifying an information matrix.

Parameters

M	The information matrix
smart	check if can be simplified to derived class

◆ invsigmas()

const Vector& gtsam::noiseModel::Diagonal::invsigmas ( ) const

inline

Return sqrt precisions

◆ Precisions()

static shared_ptr gtsam::noiseModel::Diagonal::Precisions	(	const Vector &	precisions,
		bool	smart = `true`
	)

static

A diagonal noise model created by specifying a Vector of precisions, i.e. i.e. the diagonal of the information matrix, i.e., weights

◆ precisions()

const Vector& gtsam::noiseModel::Diagonal::precisions ( ) const

inline

Return precisions

◆ QR()

virtual std::shared_ptr<Diagonal> gtsam::noiseModel::Gaussian::QR ( Matrix & Ab ) const

virtualinherited

Apply appropriately weighted QR factorization to the system [A b] Q' * [A b] = [R d] Dimensions: (r*m) * m*(n+1) = r*(n+1), where r = min(m,n). This routine performs an in-place factorization on Ab. Below-diagonal elements are set to zero by this routine.

Parameters

Ab	is the m*(n+1) augmented system matrix [A b]

Returns: Empty SharedDiagonal() noise model: R,d are whitened

Reimplemented in gtsam::noiseModel::Constrained.

◆ R()

Matrix gtsam::noiseModel::Diagonal::R ( ) const

inlineoverridevirtual

Return R itself, but note that Whiten(H) is cheaper than R*H

Reimplemented from gtsam::noiseModel::Gaussian.

◆ sigma()

double gtsam::noiseModel::Diagonal::sigma ( size_t i ) const

inline

Return standard deviations (sqrt of diagonal)

◆ Sigmas()

static shared_ptr gtsam::noiseModel::Diagonal::Sigmas	(	const Vector &	sigmas,
		bool	smart = `true`
	)

static

A diagonal noise model created by specifying a Vector of sigmas, i.e. standard deviations, the diagonal of the square root covariance matrix.

◆ SqrtInformation()

static shared_ptr gtsam::noiseModel::Gaussian::SqrtInformation	(	const Matrix &	R,
		bool	smart = `true`
	)

staticinherited

A Gaussian noise model created by specifying a square root information matrix.

Parameters

R	The (upper-triangular) square root information matrix
smart	check if can be simplified to derived class

◆ unweightedWhiten()

virtual Vector gtsam::noiseModel::Base::unweightedWhiten ( const Vector & v ) const

inlinevirtualinherited

Useful function for robust noise models to get the unweighted but whitened error

Reimplemented in gtsam::noiseModel::Robust.

◆ unwhitenInPlace() [1/2]

virtual void gtsam::noiseModel::Base::unwhitenInPlace ( Vector & v ) const

inlinevirtualinherited

in-place unwhiten, override if can be done more efficiently

Reimplemented in gtsam::noiseModel::Unit.

◆ unwhitenInPlace() [2/2]

virtual void gtsam::noiseModel::Base::unwhitenInPlace ( Eigen::Block< Vector > & v ) const

inlinevirtualinherited

in-place unwhiten, override if can be done more efficiently

Reimplemented in gtsam::noiseModel::Unit.

◆ Variances()

static shared_ptr gtsam::noiseModel::Diagonal::Variances	(	const Vector &	variances,
		bool	smart = `true`
	)

static

A diagonal noise model created by specifying a Vector of variances, i.e. i.e. the diagonal of the covariance matrix.

Parameters

variances	A vector containing the variances of this noise model
smart	check if can be simplified to derived class

◆ weight()

virtual double gtsam::noiseModel::Base::weight ( const Vector & v ) const

inlinevirtualinherited

get the weight from the effective loss function on residual vector v

Reimplemented in gtsam::noiseModel::Robust.

◆ Whiten()

Matrix gtsam::noiseModel::Diagonal::Whiten ( const Matrix & H ) const

overridevirtual

Multiply a derivative with R (derivative of whiten) Equivalent to whitening each column of the input matrix.

Reimplemented from gtsam::noiseModel::Gaussian.

Reimplemented in gtsam::noiseModel::Unit, gtsam::noiseModel::Isotropic, and gtsam::noiseModel::Constrained.

◆ whitenInPlace() [1/2]

virtual void gtsam::noiseModel::Base::whitenInPlace ( Vector & v ) const

inlinevirtualinherited

in-place whiten, override if can be done more efficiently

Reimplemented in gtsam::noiseModel::Unit, and gtsam::noiseModel::Isotropic.

◆ whitenInPlace() [2/2]

virtual void gtsam::noiseModel::Base::whitenInPlace ( Eigen::Block< Vector > & v ) const

inlinevirtualinherited

in-place whiten, override if can be done more efficiently

Reimplemented in gtsam::noiseModel::Unit.

◆ WhitenInPlace() [1/2]

void gtsam::noiseModel::Diagonal::WhitenInPlace ( Matrix & H ) const

overridevirtual

In-place version

Reimplemented from gtsam::noiseModel::Gaussian.

Reimplemented in gtsam::noiseModel::Unit, gtsam::noiseModel::Isotropic, and gtsam::noiseModel::Constrained.

◆ WhitenInPlace() [2/2]

void gtsam::noiseModel::Diagonal::WhitenInPlace ( Eigen::Block< Matrix > H ) const

overridevirtual

In-place version

Reimplemented from gtsam::noiseModel::Gaussian.

Reimplemented in gtsam::noiseModel::Unit, gtsam::noiseModel::Isotropic, and gtsam::noiseModel::Constrained.

◆ WhitenSystem()

void gtsam::noiseModel::Gaussian::WhitenSystem	(	std::vector< Matrix > &	A,
		Vector &	b
	)		const

overridevirtualinherited

Whiten a system, in place as well

Implements gtsam::noiseModel::Base.

Member Data Documentation

◆ sigmas_

Vector gtsam::noiseModel::Diagonal::sigmas_

protected

Standard deviations (sigmas), their inverse and inverse square (weights/precisions) These are all computed at construction: the idea is to use one shared model where computation is done only once, the common use case in many problems.

◆ sqrt_information_

std::optional<Matrix> gtsam::noiseModel::Gaussian::sqrt_information_

protectedinherited

Matrix square root of information matrix (R)

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

/home/docs/checkouts/readthedocs.org/user_builds/gtsam-jlblanco-docs/checkouts/latest/gtsam/linear/NoiseModel.h

Public Types

Public Member Functions

Static Public Member Functions

Protected Member Functions

Protected Attributes

Detailed Description

Constructor & Destructor Documentation

◆ Diagonal() [1/2]

◆ Diagonal() [2/2]

Member Function Documentation

◆ Covariance()

◆ Information()

◆ invsigmas()

◆ Precisions()

◆ precisions()

◆ QR()

◆ R()

◆ sigma()

◆ Sigmas()

◆ SqrtInformation()

◆ unweightedWhiten()

◆ unwhitenInPlace() [1/2]

◆ unwhitenInPlace() [2/2]

◆ Variances()

◆ weight()

◆ Whiten()

◆ whitenInPlace() [1/2]

◆ whitenInPlace() [2/2]

◆ WhitenInPlace() [1/2]

◆ WhitenInPlace() [2/2]

◆ WhitenSystem()

Member Data Documentation

◆ sigmas_

◆ sqrt_information_