Go to the documentation of this file.
68 virtual void evalJacobian(
double& dr_dxi,
double& dr_dxj)
const = 0;
77 const size_t nodeCount);
std::deque< const UnaryFactorVirtualBase * > m_factors_unary
void updateEstimation(mrpt::math::CVectorDouble &solved_x_inc, mrpt::math::CVectorDouble *solved_variances=nullptr)
Simple, scalar (1-dim) constraint (edge) for a GMRF.
A versatile "profiler" that logs the time spent within each pair of calls to enter(X)-leave(X),...
void initialize(const size_t nodeCount)
Initialize the GMRF internal state and copy the prior factors.
virtual double evaluateResidual() const =0
Return the residual/error of this observation.
void addConstraint(const UnaryFactorVirtualBase &listOfConstraints)
Insert constraints into the GMRF problem.
virtual void evalJacobian(double &dr_dx) const =0
Returns the derivative of the residual wrt the node value.
void clearAllConstraintsByType_Unary()
virtual double getInformation() const =0
Return the inverse of the variance of this constraint.
void clear()
Reset state: remove all constraints and nodes.
std::deque< const BinaryFactorVirtualBase * > m_factors_binary
void enableProfiler(bool enable=true)
bool isProfilerEnabled() const
Abstract graph and tree data structures, plus generic graph algorithms.
void clearAllConstraintsByType_Binary()
Versatile class for consistent logging and management of output messages.
mrpt::system::CTimeLogger m_timelogger
virtual void evalJacobian(double &dr_dxi, double &dr_dxj) const =0
Returns the derivative of the residual wrt the node values.
Sparse solver for GMRF (Gaussian Markov Random Fields) graphical models.
bool eraseConstraint(const FactorBase &c)
Removes a constraint.
Simple, scalar (1-dim) constraint (edge) for a GMRF.
size_t m_numNodes
number of nodes in the graph
Page generated by Doxygen 1.8.17 for MRPT 2.0.4 at Fri Jul 17 08:43:33 UTC 2020 | |