gtsam::Conditional< FACTOR, DERIVEDCONDITIONAL > Class Template Reference

Serialization
class	boost::serialization::access
	Serialization function.

Advanced Interface
FACTOR::const_iterator	beginFrontals () const
	Iterator pointing to first frontal key.
FACTOR::const_iterator	endFrontals () const
	Iterator pointing past the last frontal key.
FACTOR::const_iterator	beginParents () const
	Iterator pointing to the first parent key.
FACTOR::const_iterator	endParents () const
	Iterator pointing past the last parent key.
size_t &	nrFrontals ()
	Mutable version of nrFrontals.
FACTOR::iterator	beginFrontals ()
	Mutable iterator pointing to first frontal key.
FACTOR::iterator	endFrontals ()
	Mutable iterator pointing past the last frontal key.
FACTOR::iterator	beginParents ()
	Mutable iterator pointing to the first parent key.
FACTOR::iterator	endParents ()
	Mutable iterator pointing past the last parent key.
template<class VALUES>
static bool	CheckInvariants (const DERIVEDCONDITIONAL &conditional, const VALUES &x)
	Check invariants of this conditional, given the values x.

Public Member Functions
Testable
void	print (const std::string &s="Conditional", const KeyFormatter &formatter=DefaultKeyFormatter) const
	print with optional formatter
bool	equals (const This &c, double tol=1e-9) const
	check equality
Standard Interface
size_t	nrFrontals () const
	return the number of frontals
size_t	nrParents () const
	return the number of parents
Key	firstFrontalKey () const
	Convenience function to get the first frontal key.
Frontals	frontals () const
	return a view of the frontal keys
Parents	parents () const
	return a view of the parent keys
virtual double	logProbability (const HybridValues &c) const
	All conditional types need to implement a logProbability function, for which exp(logProbability(x)) = evaluate(x).
virtual double	evaluate (const HybridValues &c) const
	All conditional types need to implement an evaluate function, that yields a true probability.
double	operator() (const HybridValues &x) const
	Evaluate probability density, sugar.
virtual double	logNormalizationConstant () const
	All conditional types need to implement a log normalization constant to make it such that error>=0.
double	normalizationConstant () const
	Non-virtual, exponentiate logNormalizationConstant.

Public Types
typedef boost::iterator_range< typename FACTOR::const_iterator >	Frontals
	View of the frontal keys (call frontals()).
typedef boost::iterator_range< typename FACTOR::const_iterator >	Parents
	View of the separator keys (call parents()).

Protected Member Functions
Standard Constructors
	Conditional ()
	Empty Constructor to make serialization possible.
	Conditional (size_t nrFrontals)
	Constructor.

Protected Attributes
size_t	nrFrontals_
	The first nrFrontal variables are frontal and the rest are parents.

Member Function Documentation

CheckInvariants()

template<class FACTOR, class DERIVEDCONDITIONAL>

template<class VALUES>

bool gtsam::Conditional< FACTOR, DERIVEDCONDITIONAL >::CheckInvariants ( const DERIVEDCONDITIONAL & conditional, const VALUES & x )

static

Check invariants of this conditional, given the values x.

It tests:

• evaluate >= 0.0
• evaluate(x) == conditional(x)
• exp(logProbability(x)) == evaluate(x)
• logNormalizationConstant() = log(normalizationConstant())
• error >= 0.0
• logProbability(x) == logNormalizationConstant() - error(x)

Parameters

conditional

The conditional to test, as a reference to the derived type.

Template Parameters

VALUES

HybridValues, or a more narrow type like DiscreteValues.

evaluate()

template<class FACTOR, class DERIVEDCONDITIONAL>

double gtsam::Conditional< FACTOR, DERIVEDCONDITIONAL >::evaluate ( const HybridValues & c ) const

virtual

All conditional types need to implement an evaluate function, that yields a true probability.

The default implementation just exponentiates logProbability.

Reimplemented in gtsam::DiscreteConditional, gtsam::GaussianConditional, gtsam::GaussianMixture, gtsam::HybridConditional, and gtsam::SymbolicConditional.

logNormalizationConstant()

template<class FACTOR, class DERIVEDCONDITIONAL>

double gtsam::Conditional< FACTOR, DERIVEDCONDITIONAL >::logNormalizationConstant ( ) const

virtual

All conditional types need to implement a log normalization constant to make it such that error>=0.

Reimplemented in gtsam::DiscreteConditional, gtsam::GaussianConditional, gtsam::GaussianMixture, and gtsam::HybridConditional.

logProbability()

template<class FACTOR, class DERIVEDCONDITIONAL>

double gtsam::Conditional< FACTOR, DERIVEDCONDITIONAL >::logProbability ( const HybridValues & c ) const

virtual

All conditional types need to implement a logProbability function, for which exp(logProbability(x)) = evaluate(x).

Reimplemented in gtsam::DiscreteConditional, gtsam::GaussianConditional, gtsam::GaussianMixture, gtsam::HybridConditional, and gtsam::SymbolicConditional.

---

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

• /tmp/gtsam-4.2-docs.H5EUbA/src/gtsam/inference/Conditional.h
• /tmp/gtsam-4.2-docs.H5EUbA/src/gtsam/inference/Conditional-inst.h