Dakota  Version 6.20
Explore and Predict with Confidence
Public Member Functions | Protected Attributes | List of all members
MFSolutionData Class Reference

Container class for the numerical solution for a given configuration (e.g. ensemble + DAG) More...

Public Member Functions

 MFSolutionData ()
 default constructor
 
 MFSolutionData (const RealVector &soln_vars, Real avg_est_var, Real avg_est_var_ratio, Real equiv_hf)
 
 MFSolutionData (const MFSolutionData &sd)
 copy constructor
 
 ~MFSolutionData ()
 destructor
 
MFSolutionDataoperator= (const MFSolutionData &)
 assignment operator
 
const RealVector & solution_variables () const
 
Real solution_variable (size_t i) const
 
void solution_variables (const RealVector &soln_vars)
 
void solution_variables (const SizetArray &samples)
 
std::pair< RealVector, Real > anchored_solution_ratios () const
 
void anchored_solution_ratios (const RealVector &soln_ratios, Real soln_ref)
 
RealVector solution_ratios () const
 
Real solution_reference () const
 
Real average_estimator_variance () const
 
void average_estimator_variance (Real avg_estvar)
 
Real average_estimator_variance_ratio () const
 
void average_estimator_variance_ratio (Real avg_estvar_ratio)
 
Real equivalent_hf_allocation () const
 
void equivalent_hf_allocation (Real equiv_hf_alloc)
 

Protected Attributes

RealVector solutionVars
 
Real avgEstVar
 average estimator variance for model graph
 
Real avgEstVarRatio
 average estimator variance ratio (1 - R^2) for model graph: the ratio of final estimator variance (optimizer result) and final MC estimator variance using final N_H samples (not equivHF)
 
Real equivHFAlloc
 

Detailed Description

Container class for the numerical solution for a given configuration (e.g. ensemble + DAG)

Used for caching the optimization solution data that is associated with one configuration among multiple possibilities, e.g. ensemble membership and directed acyclic graph (DAG) of paired control variates.


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