gtsam::HybridGaussianProductFactor Class Reference

Detailed Description

Alias for DecisionTree of GaussianFactorGraphs and their scalar sums.

Inheritance diagram for gtsam::HybridGaussianProductFactor:

Public Member Functions
Constructors
	HybridGaussianProductFactor ()=default
	Default constructor.
template<class FACTOR>
	HybridGaussianProductFactor (const std::shared_ptr< FACTOR > &factor)
	Construct from a single factor.
	HybridGaussianProductFactor (Base &&tree)
	Construct from DecisionTree.
Operators
HybridGaussianProductFactor	operator+ (const GaussianFactor::shared_ptr &factor) const
	Add GaussianFactor into HybridGaussianProductFactor.
HybridGaussianProductFactor	operator+ (const HybridGaussianFactor &factor) const
	Add HybridGaussianFactor into HybridGaussianProductFactor.
HybridGaussianProductFactor &	operator+= (const GaussianFactor::shared_ptr &factor)
	Add-assign operator for GaussianFactor.
HybridGaussianProductFactor &	operator+= (const HybridGaussianFactor &factor)
	Add-assign operator for HybridGaussianFactor.
Testable
void	print (const std::string &s="", const KeyFormatter &formatter=DefaultKeyFormatter) const
	Print the HybridGaussianProductFactor.
bool	equals (const HybridGaussianProductFactor &other, double tol=1e-9) const
	Check if this HybridGaussianProductFactor is equal to another.
Other methods
HybridGaussianProductFactor	removeEmpty () const
	Remove empty GaussianFactorGraphs from the decision tree.
Public Member Functions inherited from gtsam::DecisionTree< Key, GaussianFactorGraphValuePair >
	DecisionTree ()
	Default constructor (for serialization).
	DecisionTree (const GaussianFactorGraphValuePair &y)
	Create a constant.
	DecisionTree (const Key &label, const GaussianFactorGraphValuePair &y1, const GaussianFactorGraphValuePair &y2)
	Create tree with 2 assignments y1, y2, splitting on variable label.
	DecisionTree (const LabelC &label, const GaussianFactorGraphValuePair &y1, const GaussianFactorGraphValuePair &y2)
	Allow Label+Cardinality for convenience.
	DecisionTree (const std::vector< LabelC > &labelCs, const std::vector< GaussianFactorGraphValuePair > &ys)
	Create from keys and a corresponding vector of values.
	DecisionTree (const std::vector< LabelC > &labelCs, const std::string &table)
	Create from keys and string table.
	DecisionTree (Iterator begin, Iterator end, const Key &label)
	Create DecisionTree from others.
	DecisionTree (const Key &label, const DecisionTree &f0, const DecisionTree &f1)
	Create DecisionTree from two others.
	DecisionTree (const Unary &op, DecisionTree &&other) noexcept
	Move constructor for DecisionTree.
	DecisionTree (const DecisionTree< Key, X > &other, Func Y_of_X)
	Convert from a different value type.
	DecisionTree (const DecisionTree< M, X > &other, const std::map< M, Key > &map, Func Y_of_X)
	Convert from a different value type X to value type Y, also translate labels via map from type M to L.
void	print (const std::string &s, const LabelFormatter &labelFormatter, const ValueFormatter &valueFormatter) const
	GTSAM-style print.
bool	equals (const DecisionTree &other, const CompareFunc &compare=&DefaultCompare) const
virtual	~DecisionTree ()=default
	Make virtual.
bool	empty () const
	Check if tree is empty.
bool	operator== (const DecisionTree &q) const
	equality
const GaussianFactorGraphValuePair &	operator() (const Assignment< Key > &x) const
	evaluate
void	visit (Func f) const
	Visit all leaves in depth-first fashion.
void	visitLeaf (Func f) const
	Visit all leaves in depth-first fashion.
void	visitWith (Func f) const
	Visit all leaves in depth-first fashion.
size_t	nrLeaves () const
	Return the number of leaves in the tree.
X	fold (Func f, X x0) const
	Fold a binary function over the tree, returning accumulator.
std::set< Key >	labels () const
	Retrieve all unique labels as a set.
DecisionTree	apply (const Unary &op) const
	apply Unary operation "op" to f
DecisionTree	apply (const UnaryAssignment &op) const
	Apply Unary operation "op" to f while also providing the corresponding assignment.
DecisionTree	apply (const DecisionTree &g, const Binary &op) const
	apply binary operation "op" to f and g
DecisionTree	choose (const Key &label, size_t index) const
	create a new function where value(label)==index It's like "restrict" in Darwiche09book pg329, 330?
DecisionTree	restrict (const Assignment< Key > &assignment) const
	Choose multiple values.
DecisionTree	combine (const Key &label, size_t cardinality, const Binary &op) const
	combine subtrees on key with binary operation "op"
DecisionTree	combine (const LabelC &labelC, const Binary &op) const
	combine with LabelC for convenience
void	dot (std::ostream &os, const LabelFormatter &labelFormatter, const ValueFormatter &valueFormatter, bool showZero=true) const
	output to graphviz format, stream version
void	dot (const std::string &name, const LabelFormatter &labelFormatter, const ValueFormatter &valueFormatter, bool showZero=true) const
	output to graphviz format, open a file
std::string	dot (const LabelFormatter &labelFormatter, const ValueFormatter &valueFormatter, bool showZero=true) const
	output to graphviz format string
std::pair< DecisionTree< Key, A >, DecisionTree< Key, B > >	split (std::function< std::pair< A, B >(const GaussianFactorGraphValuePair &)> AB_of_Y) const
	Convert into two trees with value types A and B.
	DecisionTree (const NodePtr &root)

Public Types
using	Base = DecisionTree<Key, GaussianFactorGraphValuePair>
Public Types inherited from gtsam::DecisionTree< Key, GaussianFactorGraphValuePair >
using	LabelFormatter
using	ValueFormatter
using	CompareFunc
using	Unary
	Handy typedefs for unary and binary function types.
using	UnaryAssignment
using	Binary
using	LabelC
	A label annotated with cardinality.
using	NodePtr
	---------------------— Node base class ------------------------—

Additional Inherited Members
static NodePtr	compose (Iterator begin, Iterator end, const Key &label)
Public Attributes inherited from gtsam::DecisionTree< Key, GaussianFactorGraphValuePair >
NodePtr	root_
	A DecisionTree just contains the root. TODO(dellaert): make protected.
Static Protected Member Functions inherited from gtsam::DecisionTree< Key, GaussianFactorGraphValuePair >
static bool	DefaultCompare (const GaussianFactorGraphValuePair &a, const GaussianFactorGraphValuePair &b)
	Default method for comparison of two objects of type Y.
static NodePtr	build (It begin, It end, ValueIt beginY, ValueIt endY)
	Internal recursive function to create from keys, cardinalities, and Y values.
static NodePtr	create (It begin, It end, ValueIt beginY, ValueIt endY)
	Internal helper function to create a tree from keys, cardinalities, and Y values.
static NodePtr	convertFrom (const typename DecisionTree< Key, X >::NodePtr &f, std::function< GaussianFactorGraphValuePair(const X &)> Y_of_X)
	Convert from a DecisionTree<L, X> to DecisionTree<L, Y>.
static NodePtr	convertFrom (const typename DecisionTree< M, X >::NodePtr &f, std::function< Key(const M &)> L_of_M, std::function< GaussianFactorGraphValuePair(const X &)> Y_of_X)
	Convert from a DecisionTree<M, X> to DecisionTree<L, Y>.

Constructor & Destructor Documentation

HybridGaussianProductFactor() [1/2]

template<class FACTOR>

gtsam::HybridGaussianProductFactor::HybridGaussianProductFactor ( const std::shared_ptr< FACTOR > & factor )

inline

Construct from a single factor.

Template Parameters

FACTOR

Factor type

Parameters

factor

Shared pointer to the factor

HybridGaussianProductFactor() [2/2]

gtsam::HybridGaussianProductFactor::HybridGaussianProductFactor ( Base && tree )

inline

Construct from DecisionTree.

Parameters

tree	Decision tree to construct from

Member Function Documentation

equals()

bool gtsam::HybridGaussianProductFactor::equals	(	const HybridGaussianProductFactor &	other,
		double	tol = 1e-9 ) const

Check if this HybridGaussianProductFactor is equal to another.

Parameters

other	The other HybridGaussianProductFactor to compare with
tol	Tolerance for floating point comparisons

Returns

true if equal, false otherwise

print()

void gtsam::HybridGaussianProductFactor::print	(	const std::string &	s = "",
		const KeyFormatter &	formatter = DefaultKeyFormatter ) const

Print the HybridGaussianProductFactor.

Parameters

s	Optional string to prepend
formatter	Optional key formatter

removeEmpty()

HybridGaussianProductFactor gtsam::HybridGaussianProductFactor::removeEmpty

(

)

const

Remove empty GaussianFactorGraphs from the decision tree.

Returns

A new HybridGaussianProductFactor with empty GaussianFactorGraphs removed

If any GaussianFactorGraph in the decision tree contains a nullptr, convert that leaf to an empty GaussianFactorGraph with zero scalar sum. This is needed because the DecisionTree will otherwise create a GaussianFactorGraph with a single (null) factor, which doesn't register as null.

---

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

• /tmp/gtsam-4.3.0-doxygen.rsXPUS/source/gtsam/hybrid/HybridGaussianProductFactor.h
• /tmp/gtsam-4.3.0-doxygen.rsXPUS/source/gtsam/hybrid/HybridGaussianProductFactor.cpp