A conditional of gaussian conditionals indexed by discrete variables, as part of a Bayes Network. This is the result of the elimination of a continuous variable in a hybrid scheme, such that the remaining variables are discrete+continuous. More...

#include <HybridGaussianConditional.h>

Inheritance diagram for gtsam::HybridGaussianConditional:

[legend]

Public Types
using	BaseConditional = Conditional< BaseFactor, HybridGaussianConditional >

using	BaseFactor = HybridGaussianFactor

using	Conditionals = DecisionTree< Key, GaussianConditional::shared_ptr >
	typedef for Decision Tree of Gaussian Conditionals More...

using	shared_ptr = std::shared_ptr< This >

using	This = HybridGaussianConditional

Public Types inherited from gtsam::HybridGaussianFactor
using	Base = HybridFactor

using	Factors = DecisionTree< Key, sharedFactor >
	typedef for Decision Tree of Gaussian factors. More...

using	FactorValuePairs = DecisionTree< Key, GaussianFactorValuePair >
	typedef for Decision Tree of Gaussian factors and arbitrary value. More...

using	shared_ptr = std::shared_ptr< This >

using	sharedFactor = std::shared_ptr< GaussianFactor >

using	This = HybridGaussianFactor

Public Types inherited from gtsam::HybridFactor
typedef Factor	Base
	Our base class. More...

enum	Category { Category::None, Category::Discrete, Category::Continuous, Category::Hybrid }
	Enum to help with categorizing hybrid factors. More...

typedef std::shared_ptr< HybridFactor >	shared_ptr
	shared_ptr to this class More...

typedef HybridFactor	This
	This class. More...

Public Types inherited from gtsam::Factor
typedef KeyVector::const_iterator	const_iterator
	Const iterator over keys. More...

typedef KeyVector::iterator	iterator
	Iterator over keys. More...

Public Types inherited from gtsam::Conditional< HybridGaussianFactor, HybridGaussianConditional >
typedef std::pair< typename HybridGaussianFactor ::const_iterator, typename HybridGaussianFactor ::const_iterator >	ConstFactorRange

typedef ConstFactorRangeIterator	Frontals

typedef ConstFactorRangeIterator	Parents

Public Member Functions
Constructors
	HybridGaussianConditional ()=default
	Default constructor, mainly for serialization. More...

	HybridGaussianConditional (const KeyVector &continuousFrontals, const KeyVector &continuousParents, const DiscreteKeys &discreteParents, const Conditionals &conditionals)
	Construct a new HybridGaussianConditional object. More...

	HybridGaussianConditional (const KeyVector &continuousFrontals, const KeyVector &continuousParents, const DiscreteKey &discreteParent, const std::vector< GaussianConditional::shared_ptr > &conditionals)
	Make a Hybrid Gaussian Conditional from a vector of Gaussian conditionals. The DecisionTree-based constructor is preferred over this one. More...

Testable
bool	equals (const HybridFactor &lf, double tol=1e-9) const override
	Test equality with base HybridFactor. More...

void	print (const std::string &s="HybridGaussianConditional\n", const KeyFormatter &formatter=DefaultKeyFormatter) const override
	Print utility. More...

Standard API
GaussianConditional::shared_ptr	operator() (const DiscreteValues &discreteValues) const
	Return the conditional Gaussian for the given discrete assignment. More...

size_t	nrComponents () const
	Returns the total number of continuous components. More...

KeyVector	continuousParents () const
	Returns the continuous keys among the parents. More...

double	negLogConstant () const override
	Return log normalization constant in negative log space. More...

std::shared_ptr< HybridGaussianFactor >	likelihood (const VectorValues &given) const

const Conditionals &	conditionals () const
	Getter for the underlying Conditionals DecisionTree. More...

AlgebraicDecisionTree< Key >	logProbability (const VectorValues &continuousValues) const
	Compute logProbability of the HybridGaussianConditional as a tree. More...

double	logProbability (const HybridValues &values) const override
	Compute the logProbability of this hybrid Gaussian conditional. More...

double	evaluate (const HybridValues &values) const override
	Calculate probability density for given `values`. More...

double	operator() (const HybridValues &values) const
	Evaluate probability density, sugar. More...

void	prune (const DecisionTreeFactor &discreteProbs)
	Prune the decision tree of Gaussian factors as per the discrete `discreteProbs`. More...

Public Member Functions inherited from gtsam::HybridGaussianFactor
	HybridGaussianFactor ()=default
	Default constructor, mainly for serialization. More...

	HybridGaussianFactor (const KeyVector &continuousKeys, const DiscreteKey &discreteKey, const std::vector< GaussianFactor::shared_ptr > &factors)
	Construct a new HybridGaussianFactor on a single discrete key, providing the factors for each mode m as a vector of factors ϕ_m(x). The value ϕ(x,m) for the factor is simply ϕ_m(x). More...

	HybridGaussianFactor (const KeyVector &continuousKeys, const DiscreteKey &discreteKey, const std::vector< GaussianFactorValuePair > &factors)
	Construct a new HybridGaussianFactor on a single discrete key, including a scalar error value for each mode m. The factors and scalars are provided as a vector of pairs (ϕ_m(x), E_m). The value ϕ(x,m) for the factor is now ϕ_m(x) + E_m. More...

	HybridGaussianFactor (const KeyVector &continuousKeys, const DiscreteKeys &discreteKeys, const FactorValuePairs &factors)
	Construct a new HybridGaussianFactor on a several discrete keys M, including a scalar error value for each assignment m. The factors and scalars are provided as a DecisionTree<Key> of pairs (ϕ_M(x), E_M). The value ϕ(x,M) for the factor is again ϕ_m(x) + E_m. More...

bool	equals (const HybridFactor &lf, double tol=1e-9) const override
	equals More...

void	print (const std::string &s="", const KeyFormatter &formatter=DefaultKeyFormatter) const override
	print More...

sharedFactor	operator() (const DiscreteValues &assignment) const
	Get factor at a given discrete assignment. More...

GaussianFactorGraphTree	add (const GaussianFactorGraphTree &sum) const
	Combine the Gaussian Factor Graphs in `sum` and `this` while maintaining the original tree structure. More...

AlgebraicDecisionTree< Key >	errorTree (const VectorValues &continuousValues) const override
	Compute error of the HybridGaussianFactor as a tree. More...

double	error (const HybridValues &values) const override
	Compute the log-likelihood, including the log-normalizing constant. More...

const Factors &	factors () const
	Getter for GaussianFactor decision tree. More...

Public Member Functions inherited from gtsam::HybridFactor
	HybridFactor ()=default

	HybridFactor (const KeyVector &keys)
	Construct hybrid factor from continuous keys. More...

	HybridFactor (const DiscreteKeys &discreteKeys)
	Construct hybrid factor from discrete keys. More...

	HybridFactor (const KeyVector &continuousKeys, const DiscreteKeys &discreteKeys)
	Construct a new Hybrid Factor object. More...

bool	isDiscrete () const
	True if this is a factor of discrete variables only. More...

bool	isContinuous () const
	True if this is a factor of continuous variables only. More...

bool	isHybrid () const
	True is this is a Discrete-Continuous factor. More...

size_t	nrContinuous () const
	Return the number of continuous variables in this factor. More...

const DiscreteKeys &	discreteKeys () const
	Return the discrete keys for this factor. More...

const KeyVector &	continuousKeys () const
	Return only the continuous keys for this factor. More...

Public Member Functions inherited from gtsam::Factor
virtual	~Factor ()=default
	Default destructor. More...

bool	empty () const
	Whether the factor is empty (involves zero variables). More...

Key	front () const
	First key. More...

Key	back () const
	Last key. More...

const_iterator	find (Key key) const
	find More...

const KeyVector &	keys () const
	Access the factor's involved variable keys. More...

const_iterator	begin () const

const_iterator	end () const

size_t	size () const

virtual void	printKeys (const std::string &s="Factor", const KeyFormatter &formatter=DefaultKeyFormatter) const
	print only keys More...

bool	equals (const This &other, double tol=1e-9) const
	check equality More...

KeyVector &	keys ()

iterator	begin ()

iterator	end ()

Public Member Functions inherited from gtsam::Conditional< HybridGaussianFactor, HybridGaussianConditional >
void	print (const std::string &s="Conditional", const KeyFormatter &formatter=DefaultKeyFormatter) const

bool	equals (const This &c, double tol=1e-9) const

virtual	~Conditional ()

size_t	nrFrontals () const

size_t	nrParents () const

Key	firstFrontalKey () const

Frontals	frontals () const

Parents	parents () const

double	operator() (const HybridValues &x) const
	Evaluate probability density, sugar. More...

virtual double	negLogConstant () const
	All conditional types need to implement this as the negative log of the normalization constant to make it such that error>=0. More...

size_t &	nrFrontals ()

HybridGaussianFactor ::const_iterator	beginFrontals () const

HybridGaussianFactor ::iterator	beginFrontals ()

HybridGaussianFactor ::const_iterator	endFrontals () const

HybridGaussianFactor ::iterator	endFrontals ()

HybridGaussianFactor ::const_iterator	beginParents () const

HybridGaussianFactor ::iterator	beginParents ()

HybridGaussianFactor ::const_iterator	endParents () const

HybridGaussianFactor ::iterator	endParents ()

Private Member Functions
bool	allFrontalsGiven (const VectorValues &given) const
	Check whether `given` has values for all frontal keys. More...

GaussianFactorGraphTree	asGaussianFactorGraphTree () const
	Convert a HybridGaussianConditional of conditionals into a DecisionTree of Gaussian factor graphs. More...

std::function< GaussianConditional::shared_ptr(const Assignment< Key > &, const GaussianConditional::shared_ptr &)>	prunerFunc (const DecisionTreeFactor &discreteProbs)
	Helper function to get the pruner functor. More...

Private Attributes
Conditionals	conditionals_

double	negLogConstant_

Additional Inherited Members
Static Public Member Functions inherited from gtsam::Conditional< HybridGaussianFactor, HybridGaussianConditional >
static bool	CheckInvariants (const HybridGaussianConditional &conditional, const VALUES &x)

Protected Member Functions inherited from gtsam::Factor
	Factor ()

template<typename CONTAINER >
	Factor (const CONTAINER &keys)

template<typename ITERATOR >
	Factor (ITERATOR first, ITERATOR last)

Protected Member Functions inherited from gtsam::Conditional< HybridGaussianFactor, HybridGaussianConditional >
	Conditional ()

	Conditional (size_t nrFrontals)

Static Protected Member Functions inherited from gtsam::Factor
template<typename CONTAINER >
static Factor	FromKeys (const CONTAINER &keys)

template<typename ITERATOR >
static Factor	FromIterators (ITERATOR first, ITERATOR last)

Protected Attributes inherited from gtsam::HybridFactor
KeyVector	continuousKeys_
	Record continuous keys for book-keeping. More...

DiscreteKeys	discreteKeys_

Protected Attributes inherited from gtsam::Factor
KeyVector	keys_
	The keys involved in this factor. More...

Protected Attributes inherited from gtsam::Conditional< HybridGaussianFactor, HybridGaussianConditional >
size_t	nrFrontals_

Detailed Description

A conditional of gaussian conditionals indexed by discrete variables, as part of a Bayes Network. This is the result of the elimination of a continuous variable in a hybrid scheme, such that the remaining variables are discrete+continuous.

Represents the conditional density P(X | M, Z) where X is the set of continuous random variables, M is the selection of discrete variables corresponding to a subset of the Gaussian variables and Z is parent of this node .

The probability P(x|y,z,...) is proportional to $\sum_i k_i \exp - \frac{1}{2} |R_i x - (d_i - S_i y - T_i z - ...)|^2$ where i indexes the components and k_i is a component-wise normalization constant.

a density over continuous variables given discrete/continuous parents.

Symbolic factors, used to represent a graph structure, implemented in

Definition at line 53 of file HybridGaussianConditional.h.

Member Typedef Documentation

◆ BaseConditional

using gtsam::HybridGaussianConditional::BaseConditional = Conditional<BaseFactor, HybridGaussianConditional>

Definition at line 60 of file HybridGaussianConditional.h.

◆ BaseFactor

using gtsam::HybridGaussianConditional::BaseFactor = HybridGaussianFactor

Definition at line 59 of file HybridGaussianConditional.h.

◆ Conditionals

using gtsam::HybridGaussianConditional::Conditionals = DecisionTree<Key, GaussianConditional::shared_ptr>

typedef for Decision Tree of Gaussian Conditionals

Definition at line 63 of file HybridGaussianConditional.h.

◆ shared_ptr

using gtsam::HybridGaussianConditional::shared_ptr = std::shared_ptr<This>

Definition at line 58 of file HybridGaussianConditional.h.

◆ This

using gtsam::HybridGaussianConditional::This = HybridGaussianConditional

Definition at line 57 of file HybridGaussianConditional.h.

Constructor & Destructor Documentation

◆ HybridGaussianConditional() [1/3]

gtsam::HybridGaussianConditional::HybridGaussianConditional ( )

default

Default constructor, mainly for serialization.

◆ HybridGaussianConditional() [2/3]

gtsam::HybridGaussianConditional::HybridGaussianConditional	(	const KeyVector &	continuousFrontals,
		const KeyVector &	continuousParents,
		const DiscreteKeys &	discreteParents,
		const Conditionals &	conditionals
	)

Construct a new HybridGaussianConditional object.

Parameters

continuousFrontals	the continuous frontals.
continuousParents	the continuous parents.
discreteParents	the discrete parents. Will be placed last.
conditionals	a decision tree of GaussianConditionals. The number of conditionals should be C^(number of discrete parents), where C is the cardinality of the DiscreteKeys in discreteParents, since the discreteParents will be used as the labels in the decision tree.

Definition at line 46 of file HybridGaussianConditional.cpp.

◆ HybridGaussianConditional() [3/3]

gtsam::HybridGaussianConditional::HybridGaussianConditional	(	const KeyVector &	continuousFrontals,
		const KeyVector &	continuousParents,
		const DiscreteKey &	discreteParent,
		const std::vector< GaussianConditional::shared_ptr > &	conditionals
	)

Make a Hybrid Gaussian Conditional from a vector of Gaussian conditionals. The DecisionTree-based constructor is preferred over this one.

Parameters

continuousFrontals	The continuous frontal variables
continuousParents	The continuous parent variables
discreteParent	Single discrete parent variable
conditionals	Vector of conditionals with the same size as the cardinality of the discrete parent.

Definition at line 73 of file HybridGaussianConditional.cpp.

Member Function Documentation

◆ allFrontalsGiven()

bool gtsam::HybridGaussianConditional::allFrontalsGiven ( const VectorValues & given ) const

private

Check whether given has values for all frontal keys.

Definition at line 190 of file HybridGaussianConditional.cpp.

◆ asGaussianFactorGraphTree()

GaussianFactorGraphTree gtsam::HybridGaussianConditional::asGaussianFactorGraphTree ( ) const

private

Convert a HybridGaussianConditional of conditionals into a DecisionTree of Gaussian factor graphs.

Definition at line 82 of file HybridGaussianConditional.cpp.

◆ conditionals()

const HybridGaussianConditional::Conditionals & gtsam::HybridGaussianConditional::conditionals ( ) const

Getter for the underlying Conditionals DecisionTree.

Definition at line 68 of file HybridGaussianConditional.cpp.

◆ continuousParents()

KeyVector gtsam::HybridGaussianConditional::continuousParents ( ) const

Returns the continuous keys among the parents.

Definition at line 174 of file HybridGaussianConditional.cpp.

◆ equals()

bool gtsam::HybridGaussianConditional::equals	(	const HybridFactor &	lf,
		double	tol = `1e-9`
	)		const

overridevirtual

Test equality with base HybridFactor.

Reimplemented from gtsam::HybridFactor.

Definition at line 127 of file HybridGaussianConditional.cpp.

◆ evaluate()

double gtsam::HybridGaussianConditional::evaluate ( const HybridValues & values ) const

overridevirtual

Calculate probability density for given values.

Reimplemented from gtsam::Conditional< HybridGaussianFactor, HybridGaussianConditional >.

Definition at line 331 of file HybridGaussianConditional.cpp.

◆ likelihood()

std::shared_ptr< HybridGaussianFactor > gtsam::HybridGaussianConditional::likelihood ( const VectorValues & given ) const

Create a likelihood factor for a hybrid Gaussian conditional, return a hybrid Gaussian factor on the parents.

Definition at line 201 of file HybridGaussianConditional.cpp.

◆ logProbability() [1/2]

double gtsam::HybridGaussianConditional::logProbability ( const HybridValues & values ) const

overridevirtual

Compute the logProbability of this hybrid Gaussian conditional.

Parameters

values Continuous values and discrete assignment.

Returns: double

Reimplemented from gtsam::Conditional< HybridGaussianFactor, HybridGaussianConditional >.

Definition at line 324 of file HybridGaussianConditional.cpp.

◆ logProbability() [2/2]

AlgebraicDecisionTree< Key > gtsam::HybridGaussianConditional::logProbability ( const VectorValues & continuousValues ) const

Compute logProbability of the HybridGaussianConditional as a tree.

Parameters

continuousValues The continuous VectorValues.

Returns: AlgebraicDecisionTree<Key> A decision tree with the same keys as the conditionals, and leaf values as the logProbability.

Definition at line 306 of file HybridGaussianConditional.cpp.

◆ negLogConstant()

double gtsam::HybridGaussianConditional::negLogConstant ( ) const

inlineoverride

Return log normalization constant in negative log space.

The log normalization constant is the min of the individual log-normalization constants.

Returns: double

Definition at line 161 of file HybridGaussianConditional.h.

◆ nrComponents()

size_t gtsam::HybridGaussianConditional::nrComponents ( ) const

Returns the total number of continuous components.

Definition at line 105 of file HybridGaussianConditional.cpp.

◆ operator()() [1/2]

GaussianConditional::shared_ptr gtsam::HybridGaussianConditional::operator() ( const DiscreteValues & discreteValues ) const

Return the conditional Gaussian for the given discrete assignment.

Definition at line 114 of file HybridGaussianConditional.cpp.

◆ operator()() [2/2]

double gtsam::HybridGaussianConditional::operator() ( const HybridValues & values ) const

inline

Evaluate probability density, sugar.

Definition at line 195 of file HybridGaussianConditional.h.

◆ print()

void gtsam::HybridGaussianConditional::print	(	const std::string &	s = `"HybridGaussianConditional\n"`,
		const KeyFormatter &	formatter = `DefaultKeyFormatter`
	)		const

overridevirtual

Print utility.

Reimplemented from gtsam::HybridFactor.

Definition at line 146 of file HybridGaussianConditional.cpp.

◆ prune()

void gtsam::HybridGaussianConditional::prune ( const DecisionTreeFactor & discreteProbs )

Prune the decision tree of Gaussian factors as per the discrete discreteProbs.

Parameters

discreteProbs A pruned set of probabilities for the discrete keys.

Definition at line 296 of file HybridGaussianConditional.cpp.

◆ prunerFunc()

std::function< GaussianConditional::shared_ptr(const Assignment< Key > &, const GaussianConditional::shared_ptr &)> gtsam::HybridGaussianConditional::prunerFunc ( const DecisionTreeFactor & discreteProbs )

private

Helper function to get the pruner functor.

Helper function to get the pruner functional.

Parameters

discreteProbs The pruned discrete probabilities.

Returns: std::function<GaussianConditional::shared_ptr( const Assignment<Key> &, const GaussianConditional::shared_ptr &)>

Parameters

discreteProbs The probabilities of only discrete keys.

Returns: std::function<GaussianConditional::shared_ptr( const Assignment<Key> &, const GaussianConditional::shared_ptr &)>

Definition at line 245 of file HybridGaussianConditional.cpp.

Member Data Documentation

◆ conditionals_

Conditionals gtsam::HybridGaussianConditional::conditionals_

private

a decision tree of Gaussian conditionals. Negative-log of the normalization constant (log(\sqrt(|2πΣ|))). Take advantage of the neg-log space so everything is a minimization

Definition at line 66 of file HybridGaussianConditional.h.

◆ negLogConstant_

double gtsam::HybridGaussianConditional::negLogConstant_

private

Definition at line 69 of file HybridGaussianConditional.h.

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

Public Types

Public Member Functions

Private Member Functions

Private Attributes

Additional Inherited Members

Detailed Description

Member Typedef Documentation

◆ BaseConditional

◆ BaseFactor

◆ Conditionals

◆ shared_ptr

◆ This

Constructor & Destructor Documentation

◆ HybridGaussianConditional() [1/3]

◆ HybridGaussianConditional() [2/3]

◆ HybridGaussianConditional() [3/3]

Member Function Documentation

◆ allFrontalsGiven()

◆ asGaussianFactorGraphTree()

◆ conditionals()

◆ continuousParents()

◆ equals()

◆ evaluate()

◆ likelihood()

◆ logProbability() [1/2]

◆ logProbability() [2/2]

◆ negLogConstant()

◆ nrComponents()

◆ operator()() [1/2]

◆ operator()() [2/2]

◆ print()

◆ prune()

◆ prunerFunc()

Member Data Documentation

◆ conditionals_

◆ negLogConstant_