Public Types | List of all members
gtsam::HybridBayesNet Class Reference

#include <HybridBayesNet.h>

Inheritance diagram for gtsam::HybridBayesNet:
Inheritance graph
[legend]

Public Types

using Base = BayesNet< HybridConditional >
 
using ConditionalType = HybridConditional
 
using shared_ptr = std::shared_ptr< HybridBayesNet >
 
using sharedConditional = std::shared_ptr< ConditionalType >
 
using This = HybridBayesNet
 
- Public Types inherited from gtsam::BayesNet< HybridConditional >
typedef std::shared_ptr< HybridConditionalsharedConditional
 A shared pointer to a conditional. More...
 
- Public Types inherited from gtsam::FactorGraph< HybridConditional >
typedef FastVector< sharedFactor >::const_iterator const_iterator
 
typedef HybridConditional FactorType
 factor type More...
 
typedef FastVector< sharedFactor >::iterator iterator
 
typedef std::shared_ptr< HybridConditionalsharedFactor
 Shared pointer to a factor. More...
 
typedef sharedFactor value_type
 

Public Member Functions

Standard Constructors
 HybridBayesNet ()=default
 Construct empty Bayes net. More...
 
 HybridBayesNet (std::initializer_list< HybridConditional::shared_ptr > conditionals)
 Constructor that takes an initializer list of shared pointers. More...
 
Testable
void print (const std::string &s="", const KeyFormatter &formatter=DefaultKeyFormatter) const override
 GTSAM-style printing. More...
 
bool equals (const This &fg, double tol=1e-9) const
 GTSAM-style equals. More...
 
Standard Interface
void push_back (std::shared_ptr< HybridConditional > conditional)
 Add a hybrid conditional using a shared_ptr. More...
 
void push_back (HybridConditional &&conditional)
 
template<class CONDITIONAL >
void push_back (const std::shared_ptr< CONDITIONAL > &conditional)
 Add a conditional to the Bayes net. Implicitly convert to a HybridConditional. More...
 
template<class CONDITIONAL , class... Args>
void emplace_shared (Args &&...args)
 
DiscreteBayesNet discreteMarginal () const
 Get the discrete Bayes Net P(M). As the hybrid Bayes net defines P(X,M) = P(X|M) P(M), this method returns the marginal distribution on the discrete variables. More...
 
GaussianBayesNet choose (const DiscreteValues &assignment) const
 Get the Gaussian Bayes net P(X|M=m) corresponding to a specific assignment m for the discrete variables M. As the hybrid Bayes net defines P(X,M) = P(X|M) P(M), this method returns the posterior p(X|M=m). More...
 
double evaluate (const HybridValues &values) const
 Evaluate hybrid probability density for given HybridValues. More...
 
double operator() (const HybridValues &values) const
 Evaluate hybrid probability density for given HybridValues, sugar. More...
 
HybridValues optimize () const
 Solve the HybridBayesNet by first computing the MPE of all the discrete variables and then optimizing the continuous variables based on the MPE assignment. More...
 
VectorValues optimize (const DiscreteValues &assignment) const
 Given the discrete assignment, return the optimized estimate for the selected Gaussian BayesNet. More...
 
HybridValues sample (const HybridValues &given, std::mt19937_64 *rng) const
 Sample from an incomplete BayesNet, given missing variables. More...
 
HybridValues sample (std::mt19937_64 *rng) const
 Sample using ancestral sampling. More...
 
HybridValues sample (const HybridValues &given) const
 Sample from an incomplete BayesNet, use default rng. More...
 
HybridValues sample () const
 Sample using ancestral sampling, use default rng. More...
 
HybridBayesNet prune (size_t maxNrLeaves) const
 Prune the Bayes Net such that we have at most maxNrLeaves leaves. More...
 
AlgebraicDecisionTree< KeyerrorTree (const VectorValues &continuousValues) const
 Compute the negative log posterior log P'(M|x) of all assignments up to a constant, returning the result as an algebraic decision tree. More...
 
AlgebraicDecisionTree< KeydiscretePosterior (const VectorValues &continuousValues) const
 Compute normalized posterior P(M|X=x) and return as a tree. More...
 
HybridGaussianFactorGraph toFactorGraph (const VectorValues &measurements) const
 
double logProbability (const HybridValues &x) const
 
- Public Member Functions inherited from gtsam::BayesNet< HybridConditional >
void print (const std::string &s="BayesNet", const KeyFormatter &formatter=DefaultKeyFormatter) const override
 
void dot (std::ostream &os, const KeyFormatter &keyFormatter=DefaultKeyFormatter, const DotWriter &writer=DotWriter()) const
 Output to graphviz format, stream version. More...
 
std::string dot (const KeyFormatter &keyFormatter=DefaultKeyFormatter, const DotWriter &writer=DotWriter()) const
 Output to graphviz format string. More...
 
void saveGraph (const std::string &filename, const KeyFormatter &keyFormatter=DefaultKeyFormatter, const DotWriter &writer=DotWriter()) const
 output to file with graphviz format. More...
 
double logProbability (const HybridValues &x) const
 
double evaluate (const HybridValues &c) const
 
- Public Member Functions inherited from gtsam::FactorGraph< HybridConditional >
 FactorGraph (std::initializer_list< std::shared_ptr< DERIVEDFACTOR >> sharedFactors)
 
virtual ~FactorGraph ()=default
 
void reserve (size_t size)
 
IsDerived< DERIVEDFACTOR > push_back (std::shared_ptr< DERIVEDFACTOR > factor)
 Add a factor directly using a shared_ptr. More...
 
IsDerived< DERIVEDFACTOR > push_back (const DERIVEDFACTOR &factor)
 
IsDerived< DERIVEDFACTOR > emplace_shared (Args &&... args)
 Emplace a shared pointer to factor of given type. More...
 
IsDerived< DERIVEDFACTOR > add (std::shared_ptr< DERIVEDFACTOR > factor)
 add is a synonym for push_back. More...
 
std::enable_if< std::is_base_of< FactorType, DERIVEDFACTOR >::value, This >::typeoperator+= (std::shared_ptr< DERIVEDFACTOR > factor)
 Append factor to factor graph. More...
 
std::enable_if< std::is_base_of< FactorType, DERIVEDFACTOR >::value, This >::typeoperator, (std::shared_ptr< DERIVEDFACTOR > factor)
 Overload comma operator to allow for append chaining. More...
 
HasDerivedElementType< ITERATOR > push_back (ITERATOR firstFactor, ITERATOR lastFactor)
 
HasDerivedValueType< ITERATOR > push_back (ITERATOR firstFactor, ITERATOR lastFactor)
 Push back many factors with an iterator (factors are copied) More...
 
HasDerivedElementType< CONTAINER > push_back (const CONTAINER &container)
 
HasDerivedValueType< CONTAINER > push_back (const CONTAINER &container)
 Push back non-pointer objects in a container (factors are copied). More...
 
void add (const FACTOR_OR_CONTAINER &factorOrContainer)
 
Thisoperator+= (const FACTOR_OR_CONTAINER &factorOrContainer)
 
std::enable_if< std::is_base_of< This, typename CLIQUE::FactorGraphType >::value >::type push_back (const BayesTree< CLIQUE > &bayesTree)
 
FactorIndices add_factors (const CONTAINER &factors, bool useEmptySlots=false)
 
bool equals (const This &fg, double tol=1e-9) const
 Check equality up to tolerance. More...
 
size_t size () const
 
bool empty () const
 
const sharedFactor at (size_t i) const
 
sharedFactorat (size_t i)
 
std::shared_ptr< F > at (size_t i)
 
const std::shared_ptr< F > at (size_t i) const
 Const version of templated at method. More...
 
const sharedFactor operator[] (size_t i) const
 
sharedFactoroperator[] (size_t i)
 
const_iterator begin () const
 
const_iterator end () const
 
sharedFactor front () const
 
sharedFactor back () const
 
double error (const HybridValues &values) const
 
iterator begin ()
 
iterator end ()
 
virtual void resize (size_t size)
 
void remove (size_t i)
 
void replace (size_t index, sharedFactor factor)
 
iterator erase (iterator item)
 
iterator erase (iterator first, iterator last)
 
void dot (std::ostream &os, const KeyFormatter &keyFormatter=DefaultKeyFormatter, const DotWriter &writer=DotWriter()) const
 Output to graphviz format, stream version. More...
 
std::string dot (const KeyFormatter &keyFormatter=DefaultKeyFormatter, const DotWriter &writer=DotWriter()) const
 Output to graphviz format string. More...
 
void saveGraph (const std::string &filename, const KeyFormatter &keyFormatter=DefaultKeyFormatter, const DotWriter &writer=DotWriter()) const
 output to file with graphviz format. More...
 
size_t nrFactors () const
 
KeySet keys () const
 
KeyVector keyVector () const
 
bool exists (size_t idx) const
 

Additional Inherited Members

- Protected Member Functions inherited from gtsam::BayesNet< HybridConditional >
 BayesNet ()
 
 BayesNet (ITERATOR firstConditional, ITERATOR lastConditional)
 
 BayesNet (std::initializer_list< sharedConditional > conditionals)
 
- Protected Member Functions inherited from gtsam::FactorGraph< HybridConditional >
bool isEqual (const FactorGraph &other) const
 Check exact equality of the factor pointers. Useful for derived ==. More...
 
 FactorGraph ()
 
 FactorGraph (ITERATOR firstFactor, ITERATOR lastFactor)
 
 FactorGraph (const CONTAINER &factors)
 
- Protected Attributes inherited from gtsam::FactorGraph< HybridConditional >
FastVector< sharedFactorfactors_
 

Detailed Description

A hybrid Bayes net is a collection of HybridConditionals, which can have discrete conditionals, hybrid Gaussian conditionals, or pure Gaussian conditionals.

Definition at line 37 of file HybridBayesNet.h.

Member Typedef Documentation

◆ Base

Definition at line 39 of file HybridBayesNet.h.

◆ ConditionalType

Definition at line 41 of file HybridBayesNet.h.

◆ shared_ptr

Definition at line 42 of file HybridBayesNet.h.

◆ sharedConditional

Definition at line 43 of file HybridBayesNet.h.

◆ This

Definition at line 40 of file HybridBayesNet.h.

Constructor & Destructor Documentation

◆ HybridBayesNet() [1/2]

gtsam::HybridBayesNet::HybridBayesNet ( )
default

Construct empty Bayes net.

◆ HybridBayesNet() [2/2]

gtsam::HybridBayesNet::HybridBayesNet ( std::initializer_list< HybridConditional::shared_ptr conditionals)
inline

Constructor that takes an initializer list of shared pointers.

Definition at line 52 of file HybridBayesNet.h.

Member Function Documentation

◆ choose()

GaussianBayesNet gtsam::HybridBayesNet::choose ( const DiscreteValues assignment) const

Get the Gaussian Bayes net P(X|M=m) corresponding to a specific assignment m for the discrete variables M. As the hybrid Bayes net defines P(X,M) = P(X|M) P(M), this method returns the posterior p(X|M=m).

Parameters
assignmentThe discrete value assignment for the discrete keys.
Returns
Gaussian posterior P(X|M=m) as a GaussianBayesNet.

Definition at line 103 of file HybridBayesNet.cpp.

◆ discreteMarginal()

DiscreteBayesNet gtsam::HybridBayesNet::discreteMarginal ( ) const

Get the discrete Bayes Net P(M). As the hybrid Bayes net defines P(X,M) = P(X|M) P(M), this method returns the marginal distribution on the discrete variables.

Returns
discrete marginal as a DiscreteBayesNet.

Definition at line 92 of file HybridBayesNet.cpp.

◆ discretePosterior()

AlgebraicDecisionTree< Key > gtsam::HybridBayesNet::discretePosterior ( const VectorValues continuousValues) const

Compute normalized posterior P(M|X=x) and return as a tree.

Note
Not a DiscreteConditional as the cardinalities of the DiscreteKeys, which we would need, are hard to recover.
Parameters
continuousValuesContinuous values x to condition P(M|X=x) on.
Returns
AlgebraicDecisionTree<Key>

Definition at line 201 of file HybridBayesNet.cpp.

◆ emplace_shared()

template<class CONDITIONAL , class... Args>
void gtsam::HybridBayesNet::emplace_shared ( Args &&...  args)
inline

Preferred: Emplace a conditional directly using arguments.

Examples: hbn.emplace_shared<HybridGaussianConditional>(...))); hbn.emplace_shared<GaussianConditional>(...))); hbn.emplace_shared<DiscreteConditional>(...)));

Definition at line 116 of file HybridBayesNet.h.

◆ equals()

bool gtsam::HybridBayesNet::equals ( const This fg,
double  tol = 1e-9 
) const

GTSAM-style equals.

Definition at line 39 of file HybridBayesNet.cpp.

◆ errorTree()

AlgebraicDecisionTree< Key > gtsam::HybridBayesNet::errorTree ( const VectorValues continuousValues) const

Compute the negative log posterior log P'(M|x) of all assignments up to a constant, returning the result as an algebraic decision tree.

Note
The joint P(X,M) is p(X|M) P(M) Then the posterior on M given X=x is is P(M|x) = p(x|M) P(M) / p(x). Ideally we want log P(M|x) = log p(x|M) + log P(M) - log p(x), but unfortunately log p(x) is expensive, so we compute the log of the unnormalized posterior log P'(M|x) = log p(x|M) + log P(M)
Parameters
continuousValuesContinuous values x at which to compute log P'(M|x)
Returns
AlgebraicDecisionTree<Key>

Definition at line 188 of file HybridBayesNet.cpp.

◆ evaluate()

double gtsam::HybridBayesNet::evaluate ( const HybridValues values) const

Evaluate hybrid probability density for given HybridValues.

Definition at line 210 of file HybridBayesNet.cpp.

◆ logProbability()

double gtsam::BayesNet< CONDITIONAL >::logProbability

Definition at line 94 of file BayesNet-inst.h.

◆ operator()()

double gtsam::HybridBayesNet::operator() ( const HybridValues values) const
inline

Evaluate hybrid probability density for given HybridValues, sugar.

Definition at line 145 of file HybridBayesNet.h.

◆ optimize() [1/2]

HybridValues gtsam::HybridBayesNet::optimize ( ) const

Solve the HybridBayesNet by first computing the MPE of all the discrete variables and then optimizing the continuous variables based on the MPE assignment.

Returns
HybridValues

Definition at line 123 of file HybridBayesNet.cpp.

◆ optimize() [2/2]

VectorValues gtsam::HybridBayesNet::optimize ( const DiscreteValues assignment) const

Given the discrete assignment, return the optimized estimate for the selected Gaussian BayesNet.

Parameters
assignmentAn assignment of discrete values.
Returns
Values

Definition at line 141 of file HybridBayesNet.cpp.

◆ print()

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

GTSAM-style printing.

Reimplemented from gtsam::FactorGraph< HybridConditional >.

Definition at line 33 of file HybridBayesNet.cpp.

◆ prune()

HybridBayesNet gtsam::HybridBayesNet::prune ( size_t  maxNrLeaves) const

Prune the Bayes Net such that we have at most maxNrLeaves leaves.

Parameters
maxNrLeavesContinuous values at which to compute the error.
Returns
A pruned HybridBayesNet

Definition at line 48 of file HybridBayesNet.cpp.

◆ push_back() [1/3]

template<class CONDITIONAL >
void gtsam::HybridBayesNet::push_back ( const std::shared_ptr< CONDITIONAL > &  conditional)
inline

Add a conditional to the Bayes net. Implicitly convert to a HybridConditional.

E.g. hbn.push_back(std::make_shared<DiscreteConditional>(m, "1/1"));

Template Parameters
CONDITIONALType of conditional. This is shared_ptr version.
Parameters
conditionalThe conditional as a shared pointer.

Definition at line 103 of file HybridBayesNet.h.

◆ push_back() [2/3]

void gtsam::HybridBayesNet::push_back ( HybridConditional &&  conditional)
inline

Move a HybridConditional into a shared pointer and add.

Example: HybridGaussianConditional conditional(...); hbn.push_back(conditional); // loses the original conditional

Definition at line 87 of file HybridBayesNet.h.

◆ push_back() [3/3]

void gtsam::HybridBayesNet::push_back ( std::shared_ptr< HybridConditional conditional)
inline

Add a hybrid conditional using a shared_ptr.

This is the "native" push back, as this class stores hybrid conditionals.

Definition at line 76 of file HybridBayesNet.h.

◆ sample() [1/4]

HybridValues gtsam::HybridBayesNet::sample ( ) const

Sample using ancestral sampling, use default rng.

Returns
HybridValues

Definition at line 183 of file HybridBayesNet.cpp.

◆ sample() [2/4]

HybridValues gtsam::HybridBayesNet::sample ( const HybridValues given) const

Sample from an incomplete BayesNet, use default rng.

Parameters
givenValues of missing variables.
Returns
HybridValues

Definition at line 178 of file HybridBayesNet.cpp.

◆ sample() [3/4]

HybridValues gtsam::HybridBayesNet::sample ( const HybridValues given,
std::mt19937_64 *  rng 
) const

Sample from an incomplete BayesNet, given missing variables.

Example: std::mt19937_64 rng(42); VectorValues given = ...; auto sample = bn.sample(given, &rng);

Parameters
givenValues of missing variables.
rngThe pseudo-random number generator.
Returns
HybridValues

Definition at line 153 of file HybridBayesNet.cpp.

◆ sample() [4/4]

HybridValues gtsam::HybridBayesNet::sample ( std::mt19937_64 *  rng) const

Sample using ancestral sampling.

Example: std::mt19937_64 rng(42); auto sample = bn.sample(&rng);

Parameters
rngThe pseudo-random number generator.
Returns
HybridValues

Definition at line 172 of file HybridBayesNet.cpp.

◆ toFactorGraph()

HybridGaussianFactorGraph gtsam::HybridBayesNet::toFactorGraph ( const VectorValues measurements) const

Convert a hybrid Bayes net to a hybrid Gaussian factor graph by converting all conditionals with instantiated measurements into likelihood factors.

Definition at line 215 of file HybridBayesNet.cpp.


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


gtsam
Author(s):
autogenerated on Fri Nov 1 2024 03:51:33