GTSAM
4.0.2
C++ library for smoothing and mapping (SAM)
|
#include <GaussianBayesNet.h>
Public Types | |
typedef BayesNet< GaussianConditional > | Base |
typedef GaussianBayesNet | This |
typedef GaussianConditional | ConditionalType |
typedef std::shared_ptr< This > | shared_ptr |
typedef std::shared_ptr< ConditionalType > | sharedConditional |
typedef GaussianConditional | FactorType |
factor type | |
typedef std::shared_ptr< GaussianConditional > | sharedFactor |
Shared pointer to a factor. | |
typedef sharedFactor | value_type |
typedef FastVector< sharedFactor >::iterator | iterator |
typedef FastVector< sharedFactor >::const_iterator | const_iterator |
Public Member Functions | |
Standard Constructors | |
GaussianBayesNet () | |
template<typename ITERATOR > | |
GaussianBayesNet (ITERATOR firstConditional, ITERATOR lastConditional) | |
template<class CONTAINER > | |
GaussianBayesNet (const CONTAINER &conditionals) | |
template<class DERIVEDCONDITIONAL > | |
GaussianBayesNet (const FactorGraph< DERIVEDCONDITIONAL > &graph) | |
template<class DERIVEDCONDITIONAL > | |
GaussianBayesNet (std::initializer_list< std::shared_ptr< DERIVEDCONDITIONAL > > conditionals) | |
Testable | |
bool | equals (const This &bn, double tol=1e-9) const |
void | print (const std::string &s="", const KeyFormatter &formatter=DefaultKeyFormatter) const override |
print graph | |
Standard Interface | |
double | error (const VectorValues &x) const |
Sum error over all variables. | |
double | logProbability (const VectorValues &x) const |
Sum logProbability over all variables. | |
double | evaluate (const VectorValues &x) const |
double | operator() (const VectorValues &x) const |
Evaluate probability density, sugar. | |
VectorValues | optimize () const |
VectorValues | optimize (const VectorValues &given) const |
Version of optimize for incomplete BayesNet, given missing variables. | |
VectorValues | sample (std::mt19937_64 *rng) const |
VectorValues | sample (const VectorValues &given, std::mt19937_64 *rng) const |
VectorValues | sample () const |
Sample using ancestral sampling, use default rng. | |
VectorValues | sample (const VectorValues &given) const |
Sample from an incomplete BayesNet, use default rng. | |
Ordering | ordering () const |
Linear Algebra | |
std::pair< Matrix, Vector > | matrix (const Ordering &ordering) const |
std::pair< Matrix, Vector > | matrix () const |
VectorValues | optimizeGradientSearch () const |
VectorValues | gradient (const VectorValues &x0) const |
VectorValues | gradientAtZero () const |
double | determinant () const |
double | logDeterminant () const |
VectorValues | backSubstitute (const VectorValues &gx) const |
VectorValues | backSubstituteTranspose (const VectorValues &gx) const |
Graph Display | |
void | dot (std::ostream &os, const KeyFormatter &keyFormatter=DefaultKeyFormatter, const DotWriter &writer=DotWriter()) const |
Output to graphviz format, stream version. | |
std::string | dot (const KeyFormatter &keyFormatter=DefaultKeyFormatter, const DotWriter &writer=DotWriter()) const |
Output to graphviz format string. | |
void | saveGraph (const std::string &filename, const KeyFormatter &keyFormatter=DefaultKeyFormatter, const DotWriter &writer=DotWriter()) const |
output to file with graphviz format. | |
HybridValues methods | |
double | logProbability (const HybridValues &x) const |
double | evaluate (const HybridValues &c) const |
Adding Single Factors | |
void | reserve (size_t size) |
IsDerived< DERIVEDFACTOR > | push_back (std::shared_ptr< DERIVEDFACTOR > factor) |
Add a factor directly using a shared_ptr. | |
IsDerived< DERIVEDFACTOR > | push_back (const DERIVEDFACTOR &factor) |
IsDerived< DERIVEDFACTOR > | emplace_shared (Args &&... args) |
Emplace a shared pointer to factor of given type. | |
IsDerived< DERIVEDFACTOR > | add (std::shared_ptr< DERIVEDFACTOR > factor) |
add is a synonym for push_back. | |
Adding via iterators | |
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) | |
Adding via container | |
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). | |
void | add (const FACTOR_OR_CONTAINER &factorOrContainer) |
Specialized versions | |
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) |
Testable | |
bool | equals (const This &fg, double tol=1e-9) const |
Check equality up to tolerance. | |
Standard Interface | |
size_t | size () const |
bool | empty () const |
const sharedFactor | at (size_t i) const |
sharedFactor & | at (size_t i) |
const sharedFactor | operator[] (size_t i) const |
sharedFactor & | operator[] (size_t i) |
const_iterator | begin () const |
const_iterator | end () const |
sharedFactor | front () const |
sharedFactor | back () const |
double | error (const HybridValues &values) const |
Modifying Factor Graphs (imperative, discouraged) | |
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) |
Advanced Interface | |
size_t | nrFactors () const |
KeySet | keys () const |
KeyVector | keyVector () const |
bool | exists (size_t idx) const |
Protected Member Functions | |
bool | isEqual (const FactorGraph &other) const |
Check exact equality of the factor pointers. Useful for derived ==. | |
Protected Attributes | |
FastVector< sharedFactor > | factors_ |
GaussianBayesNet is a Bayes net made from linear-Gaussian conditionals.
|
inline |
Construct empty bayes net
|
inline |
Construct from iterator over conditionals
|
inlineexplicit |
Construct from container of factors (shared_ptr or plain objects)
|
inlineexplicit |
Implicit copy/downcast constructor to override explicit template container constructor
|
inline |
Constructor that takes an initializer list of shared pointers. BayesNet bn = {make_shared<Conditional>(), ...};
|
inlineinherited |
Add a factor or container of factors, including STL collections, BayesTrees, etc.
|
inherited |
Add new factors to a factor graph and returns a list of new factor indices, optionally finding and reusing empty factor slots.
|
inlineinherited |
Get a specific factor by index (this checks array bounds and may throw an exception, as opposed to operator[] which does not).
|
inlineinherited |
Get a specific factor by index (this checks array bounds and may throw an exception, as opposed to operator[] which does not).
|
inlineinherited |
Get the last factor
VectorValues gtsam::GaussianBayesNet::backSubstitute | ( | const VectorValues & | gx | ) | const |
Backsubstitute with a different RHS vector than the one stored in this BayesNet. gy=inv(R*inv(Sigma))*gx
VectorValues gtsam::GaussianBayesNet::backSubstituteTranspose | ( | const VectorValues & | gx | ) | const |
Transpose backsubstitute with a different RHS vector than the one stored in this BayesNet. gy=inv(L)*gx by solving L*gy=gx. gy=inv(R'*inv(Sigma))*gx gz'*R'=gx', gy = gz.*sigmas
|
inlineinherited |
Iterator to beginning of factors.
|
inlineinherited |
non-const STL-style begin()
double gtsam::GaussianBayesNet::determinant | ( | ) | const |
Computes the determinant of a GassianBayesNet. A GaussianBayesNet is an upper triangular matrix and for an upper triangular matrix determinant is the product of the diagonal elements. Instead of actually multiplying we add the logarithms of the diagonal elements and take the exponent at the end because this is more numerically stable.
bayesNet | The input GaussianBayesNet |
|
inlineinherited |
Check if the graph is empty (null factors set by remove() will cause this to return false).
|
inlineinherited |
Iterator to end of factors.
|
inlineinherited |
non-const STL-style end()
bool gtsam::GaussianBayesNet::equals | ( | const This & | bn, |
double | tol = 1e-9 |
||
) | const |
Check equality
|
inlineinherited |
Erase factor and rearrange other factors to take up the empty space
|
inlineinherited |
Erase factors and rearrange other factors to take up the empty space
|
inherited |
Add error for all factors.
double gtsam::GaussianBayesNet::evaluate | ( | const VectorValues & | x | ) | const |
Calculate probability density for given values x
: exp(logProbability) where x is the vector of values.
|
inlineinherited |
MATLAB interface utility: Checks whether a factor index idx exists in the graph and is a live pointer
|
inlineinherited |
Get the first factor
VectorValues gtsam::GaussianBayesNet::gradient | ( | const VectorValues & | x0 | ) | const |
Compute the gradient of the energy function, \( \nabla_{x=x_0} \left\Vert \Sigma^{-1} R x - d \right\Vert^2 \), centered around \( x = x_0 \). The gradient is \( R^T(Rx-d) \).
x0 | The center about which to compute the gradient |
VectorValues gtsam::GaussianBayesNet::gradientAtZero | ( | ) | const |
Compute the gradient of the energy function, \( \nabla_{x=0} \left\Vert \Sigma^{-1} R x - d \right\Vert^2 \), centered around zero. The gradient about zero is \( -R^T d \). See also gradient(const GaussianBayesNet&, const VectorValues&).
[output] | g A VectorValues to store the gradient, which must be preallocated, see allocateVectorValues |
|
inherited |
Potentially slow function to return all keys involved, sorted, as a set
|
inherited |
Potentially slow function to return all keys involved, sorted, as a vector
double gtsam::GaussianBayesNet::logDeterminant | ( | ) | const |
Computes the log of the determinant of a GassianBayesNet. A GaussianBayesNet is an upper triangular matrix and for an upper triangular matrix determinant is the product of the diagonal elements.
bayesNet | The input GaussianBayesNet |
std::pair<Matrix, Vector> gtsam::GaussianBayesNet::matrix | ( | const Ordering & | ordering | ) | const |
Return (dense) upper-triangular matrix representation Will return upper-triangular matrix only when using 'ordering' above. In case Bayes net is incomplete zero columns are added to the end.
std::pair<Matrix, Vector> gtsam::GaussianBayesNet::matrix | ( | ) | const |
Return (dense) upper-triangular matrix representation Will return upper-triangular matrix only when using 'ordering' above. In case Bayes net is incomplete zero columns are added to the end.
|
inherited |
return the number of non-null factors
|
inlineinherited |
Get a specific factor by index (this does not check array bounds, as opposed to at() which does).
|
inlineinherited |
Get a specific factor by index (this does not check array bounds, as opposed to at() which does).
VectorValues gtsam::GaussianBayesNet::optimize | ( | ) | const |
Solve the GaussianBayesNet, i.e. return \( x = R^{-1}*d \), by back-substitution
VectorValues gtsam::GaussianBayesNet::optimizeGradientSearch | ( | ) | const |
Optimize along the gradient direction, with a closed-form computation to perform the line search. The gradient is computed about \( \delta x=0 \).
This function returns \( \delta x \) that minimizes a reparametrized problem. The error function of a GaussianBayesNet is
\[ f(\delta x) = \frac{1}{2} |R \delta x - d|^2 = \frac{1}{2}d^T d - d^T R \delta x + \frac{1}{2} \delta x^T R^T R \delta x \]
with gradient and Hessian
\[ g(\delta x) = R^T(R\delta x - d), \qquad G(\delta x) = R^T R. \]
This function performs the line search in the direction of the gradient evaluated at \( g = g(\delta x = 0) \) with step size \( \alpha \) that minimizes \( f(\delta x = \alpha g) \):
\[ f(\alpha) = \frac{1}{2} d^T d + g^T \delta x + \frac{1}{2} \alpha^2 g^T G g \]
Optimizing by setting the derivative to zero yields \( \hat \alpha = (-g^T g) / (g^T G g) \). For efficiency, this function evaluates the denominator without computing the Hessian \( G \), returning
\[ \delta x = \hat\alpha g = \frac{-g^T g}{(R g)^T(R g)} \]
Ordering gtsam::GaussianBayesNet::ordering | ( | ) | const |
Return ordering corresponding to a topological sort. There are many topological sorts of a Bayes net. This one corresponds to the one that makes 'matrix' below upper-triangular. In case Bayes net is incomplete any non-frontal are added to the end.
|
inlineinherited |
Add a factor by value, will be copy-constructed (use push_back with a shared_ptr to avoid the copy).
|
inlineinherited |
Push back many factors with an iterator over shared_ptr (factors are not copied)
|
inlineinherited |
Push back many factors as shared_ptr's in a container (factors are not copied)
|
inlineinherited |
Push back a BayesTree as a collection of factors. NOTE: This should be hidden in derived classes in favor of a type-specialized version that calls this templated function.
|
inlineinherited |
delete factor without re-arranging indexes by inserting a nullptr pointer
|
inlineinherited |
replace a factor by index
|
inlineinherited |
Reserve space for the specified number of factors if you know in advance how many there will be (works like FastVector::reserve).
|
inlinevirtualinherited |
Directly resize the number of factors in the graph. If the new size is less than the original, factors at the end will be removed. If the new size is larger than the original, null factors will be appended.
VectorValues gtsam::GaussianBayesNet::sample | ( | std::mt19937_64 * | rng | ) | const |
Sample using ancestral sampling Example: std::mt19937_64 rng(42); auto sample = gbn.sample(&rng);
VectorValues gtsam::GaussianBayesNet::sample | ( | const VectorValues & | given, |
std::mt19937_64 * | rng | ||
) | const |
Sample from an incomplete BayesNet, given missing variables Example: std::mt19937_64 rng(42); VectorValues given = ...; auto sample = gbn.sample(given, &rng);
|
inlineinherited |
return the number of factors (including any null factors set by remove() ).
|
protectedinherited |
concept check, makes sure FACTOR defines print and equals Collection of factors