Public Member Functions | Protected Member Functions | Protected Attributes | Private Member Functions | Private Attributes
Approximation Class Reference

Base class for the approximation class hierarchy. More...

Inheritance diagram for Approximation:
GaussProcApproximation PecosApproximation SurfpackApproximation TANA3Approximation TaylorApproximation

List of all members.

Public Member Functions

 Approximation ()
 default constructor
 Approximation (ProblemDescDB &problem_db, size_t num_vars)
 standard constructor for envelope
 Approximation (const String &approx_type, const UShortArray &approx_order, size_t num_vars, short data_order)
 alternate constructor
 Approximation (const Approximation &approx)
 copy constructor
virtual ~Approximation ()
 destructor
Approximation operator= (const Approximation &approx)
 assignment operator
virtual void build ()
 builds the approximation from scratch
virtual void rebuild ()
 rebuilds the approximation incrementally
virtual void pop (bool save_data)
 removes entries from end of SurrogateData::{vars,resp}Data (last points appended)
virtual void restore ()
 restores state prior to previous append()
virtual bool restore_available ()
 queries availability of restoration for trial set
virtual size_t restoration_index ()
 return index of trial set within restorable bookkeeping sets
virtual void finalize ()
 finalize approximation by applying all remaining trial sets
virtual size_t finalization_index (size_t i)
 return index of i-th trailing trial set within restorable bookkeeping sets
virtual void store ()
 store current approximation for later combination
virtual void combine (short corr_type)
 combine current approximation with previously stored approximation
virtual Real get_value (const RealVector &x)
 retrieve the approximate function value for a given parameter vector
virtual const RealVector & get_gradient (const RealVector &x)
 retrieve the approximate function gradient for a given parameter vector
virtual const RealSymMatrix & get_hessian (const RealVector &x)
 retrieve the approximate function Hessian for a given parameter vector
virtual Real get_prediction_variance (const RealVector &x)
 retrieve the variance of the predicted value for a given parameter vector
virtual Real get_diagnostic (const String &metric_type)
 retrieve the diagnostic metric for the diagnostic type specified
virtual const RealVector & approximation_coefficients () const
 return the coefficient array computed by build()/rebuild()
virtual void approximation_coefficients (const RealVector &approx_coeffs)
 set the coefficient array from external sources, rather than computing with build()/rebuild()
virtual void print_coefficients (std::ostream &s) const
 print the coefficient array computed in build()/rebuild()
virtual int min_coefficients () const
 return the minimum number of samples (unknowns) required to build the derived class approximation type in numVars dimensions
virtual int recommended_coefficients () const
 return the recommended number of samples (unknowns) required to build the derived class approximation type in numVars dimensions
virtual int num_constraints () const
 return the number of constraints to be enforced via an anchor point
virtual void clear_current ()
 clear current build data in preparation for next build
virtual const bool diagnostics_available ()
 check if diagnostics are available for this approximation type
int min_points (bool constraint_flag) const
 return the minimum number of points required to build the approximation type in numVars dimensions. Uses *_coefficients() and num_constraints().
int recommended_points (bool constraint_flag) const
 return the recommended number of samples to build the approximation type in numVars dimensions (default same as min_points)
int num_variables () const
 return the number of variables used in the approximation
const Pecos::SurrogateData & approximation_data () const
 return approxData
void add (const Pecos::SurrogateDataVars &sdv, bool anchor_flag)
 append to SurrogateData::varsData or assign to SurrogateData::anchorVars
void add (const Variables &vars, bool anchor_flag, bool deep_copy)
 extract the relevant RealVector from Variables and invoke add(RealVector&)
void add (const Real *sample_c_vars, bool anchor_flag, bool deep_copy)
 create a RealVector view and invoke add(RealVector&)
void add (const RealVector &sample_c_vars, bool anchor_flag, bool deep_copy)
 shared code among add(Variables&) and add(Real*); adds a new data point by either appending to SurrogateData::varsData or assigning to SurrogateData::anchorVars, as dictated by anchor_flag. Uses add_point() and add_anchor().
void add (const Pecos::SurrogateDataResp &sdr, bool anchor_flag)
 append to SurrogateData::respData or assign to SurrogateData::anchorResp
void add (const Response &response, int fn_index, bool anchor_flag, bool deep_copy)
 adds a new data point by either appending to SurrogateData::respData or assigning to SurrogateData::anchorResp, as dictated by anchor_flag. Uses add_point() and add_anchor().
void pop_count (size_t count)
 sets popCount (number of entries to pop from end of SurrogateData::{vars,resp}Data, based on size of last data set appended)
size_t pop_count () const
 returns popCount (number of entries to pop from end of SurrogateData::{vars,resp}Data, based on size of last data set appended)
void clear_all ()
 clear all build data (current and history) to restore original state
void clear_anchor ()
 clear SurrogateData::anchor{Vars,Resp}
void clear_data ()
 clear SurrogateData::{vars,resp}Data
void set_bounds (const RealVector &lower, const RealVector &upper)
 set approximation lower and upper bounds (currently only used by graphics)
Approximationapprox_rep () const
 returns approxRep for access to derived class member functions that are not mapped to the top Approximation level

Protected Member Functions

 Approximation (BaseConstructor, const ProblemDescDB &problem_db, size_t num_vars)
 constructor initializes the base class part of letter classes (BaseConstructor overloading avoids infinite recursion in the derived class constructors - Coplien, p. 139)
 Approximation (BaseConstructor, const String &approx_type, size_t num_vars, short data_order)
 constructor initializes the base class part of letter classes (BaseConstructor overloading avoids infinite recursion in the derived class constructors - Coplien, p. 139)

Protected Attributes

short outputLevel
 output verbosity level: {SILENT,QUIET,NORMAL,VERBOSE,DEBUG}_OUTPUT
int numVars
 number of variables in the approximation
String approxType
 approximation type identifier
short buildDataOrder
 order of the data used for surrogate construction, in ActiveSet request vector 3-bit format.
RealVector approxGradient
 gradient of the approximation returned by get_gradient()
RealSymMatrix approxHessian
 Hessian of the approximation returned by get_hessian()
Pecos::SurrogateData approxData
 contains the variables/response data for constructing a single approximation model (one response function)
RealVector approxLowerBounds
 approximation lower bounds (used by 3D graphics and Surfpack KrigingModel)
RealVector approxUpperBounds
 approximation upper bounds (used by 3D graphics and Surfpack KrigingModel)

Private Member Functions

Approximationget_approx (ProblemDescDB &problem_db, size_t num_vars)
 Used only by the standard envelope constructor to initialize approxRep to the appropriate derived type.
Approximationget_approx (const String &approx_type, const UShortArray &approx_order, size_t num_vars, short data_order)
 Used only by the alternate envelope constructor to initialize approxRep to the appropriate derived type.

Private Attributes

size_t popCount
 number of points previously added by append() to be removed by pop()
ApproximationapproxRep
 pointer to the letter (initialized only for the envelope)
int referenceCount
 number of objects sharing approxRep

Detailed Description

Base class for the approximation class hierarchy.

The Approximation class is the base class for the response data fit approximation class hierarchy in DAKOTA. One instance of an Approximation must be created for each function to be approximated (a vector of Approximations is contained in ApproximationInterface). For memory efficiency and enhanced polymorphism, the approximation hierarchy employs the "letter/envelope idiom" (see Coplien "Advanced C++", p. 133), for which the base class (Approximation) serves as the envelope and one of the derived classes (selected in Approximation::get_approx()) serves as the letter.


Constructor & Destructor Documentation

default constructor

The default constructor is used in Array<Approximation> instantiations and by the alternate envelope constructor. approxRep is NULL in this case (problem_db is needed to build a meaningful Approximation object). This makes it necessary to check for NULL in the copy constructor, assignment operator, and destructor.

Approximation ( ProblemDescDB problem_db,
size_t  num_vars 
)

standard constructor for envelope

Envelope constructor only needs to extract enough data to properly execute get_approx, since Approximation(BaseConstructor, problem_db) builds the actual base class data for the derived approximations.

References Dakota::abort_handler(), Approximation::approxRep, and Approximation::get_approx().

Approximation ( const String approx_type,
const UShortArray &  approx_order,
size_t  num_vars,
short  data_order 
)

alternate constructor

This is the alternate envelope constructor for instantiations on the fly. Since it does not have access to problem_db, the letter class is not fully populated. This constructor executes get_approx(type), which invokes the default constructor of the derived letter class, which in turn invokes the default constructor of the base class.

References Dakota::abort_handler(), Approximation::approxRep, and Approximation::get_approx().

Approximation ( const Approximation approx)

copy constructor

Copy constructor manages sharing of approxRep and incrementing of referenceCount.

References Approximation::approxRep, and Approximation::referenceCount.

~Approximation ( ) [virtual]

destructor

Destructor decrements referenceCount and only deletes approxRep when referenceCount reaches zero.

References Approximation::approxRep, and Approximation::referenceCount.

Approximation ( BaseConstructor  ,
const ProblemDescDB problem_db,
size_t  num_vars 
) [protected]

constructor initializes the base class part of letter classes (BaseConstructor overloading avoids infinite recursion in the derived class constructors - Coplien, p. 139)

This constructor is the one which must build the base class data for all derived classes. get_approx() instantiates a derived class letter and the derived constructor selects this base class constructor in its initialization list (to avoid recursion in the base class constructor calling get_approx() again). Since the letter IS the representation, its rep pointer is set to NULL (an uninitialized pointer causes problems in ~Approximation).

approxType.ends("_interpolation_polynomial") && approxType.ends("_orthogonal_polynomial"))

References Approximation::approxType, and ProblemDescDB::get_bool().

Approximation ( BaseConstructor  ,
const String approx_type,
size_t  num_vars,
short  data_order 
) [protected]

constructor initializes the base class part of letter classes (BaseConstructor overloading avoids infinite recursion in the derived class constructors - Coplien, p. 139)

This constructor is the one which must build the base class data for all derived classes. get_approx() instantiates a derived class letter and the derived constructor selects this base class constructor in its initialization list (to avoid recursion in the base class constructor calling get_approx() again). Since the letter IS the representation, its rep pointer is set to NULL (an uninitialized pointer causes problems in ~Approximation).


Member Function Documentation

Approximation operator= ( const Approximation approx)

assignment operator

Assignment operator decrements referenceCount for old approxRep, assigns new approxRep, and increments referenceCount for new approxRep.

References Approximation::approxRep, and Approximation::referenceCount.

void build ( ) [virtual]

builds the approximation from scratch

This is the common base class portion of the virtual fn and is insufficient on its own; derived implementations should explicitly invoke (or reimplement) this base class contribution.

Reimplemented in PecosApproximation, GaussProcApproximation, SurfpackApproximation, TaylorApproximation, and TANA3Approximation.

References Dakota::abort_handler(), Approximation::approxData, Approximation::approxRep, Approximation::build(), Approximation::min_points(), and Approximation::numVars.

Referenced by Approximation::build(), and Approximation::rebuild().

void rebuild ( ) [virtual]

rebuilds the approximation incrementally

This is the common base class portion of the virtual fn and is insufficient on its own; derived implementations should explicitly invoke (or reimplement) this base class contribution.

Reimplemented in PecosApproximation.

References Approximation::approxRep, Approximation::build(), and Approximation::rebuild().

Referenced by Approximation::rebuild().

void pop ( bool  save_data) [virtual]

removes entries from end of SurrogateData::{vars,resp}Data (last points appended)

This is the common base class portion of the virtual fn and is insufficient on its own; derived implementations should explicitly invoke (or reimplement) this base class contribution.

Reimplemented in PecosApproximation.

References Approximation::approxData, Approximation::approxRep, Approximation::pop(), and Approximation::popCount.

Referenced by Approximation::pop().

void restore ( ) [virtual]

restores state prior to previous append()

This is the common base class portion of the virtual fn and is insufficient on its own; derived implementations should explicitly invoke (or reimplement) this base class contribution.

Reimplemented in PecosApproximation.

References Approximation::approxData, Approximation::approxRep, Approximation::popCount, Approximation::restoration_index(), and Approximation::restore().

Referenced by Approximation::restore().

void finalize ( ) [virtual]

finalize approximation by applying all remaining trial sets

This is the common base class portion of the virtual fn and is insufficient on its own; derived implementations should explicitly invoke (or reimplement) this base class contribution.

Reimplemented in PecosApproximation.

References Approximation::approxData, Approximation::approxRep, Approximation::finalization_index(), and Approximation::finalize().

Referenced by Approximation::finalize().

void clear_current ( ) [inline, virtual]

clear current build data in preparation for next build

Redefined by TANA3Approximation to clear current data but preserve history.

Reimplemented in TANA3Approximation.

References Approximation::approxRep, Approximation::clear_all(), and Approximation::clear_current().

Referenced by Approximation::clear_current().

void clear_all ( ) [inline]

clear all build data (current and history) to restore original state

Clears out any history (e.g., TANA3Approximation use for a different response function in NonDReliability).

References Approximation::approxData, Approximation::approxRep, and Approximation::clear_all().

Referenced by Approximation::clear_all(), and Approximation::clear_current().

Approximation * get_approx ( ProblemDescDB problem_db,
size_t  num_vars 
) [private]

Used only by the standard envelope constructor to initialize approxRep to the appropriate derived type.

Used only by the envelope constructor to initialize approxRep to the appropriate derived type.

References String::ends(), and ProblemDescDB::get_string().

Referenced by Approximation::Approximation().

Approximation * get_approx ( const String approx_type,
const UShortArray &  approx_order,
size_t  num_vars,
short  data_order 
) [private]

Used only by the alternate envelope constructor to initialize approxRep to the appropriate derived type.

Used only by the envelope constructor to initialize approxRep to the appropriate derived type.

References String::ends().


Member Data Documentation

short buildDataOrder [protected]

order of the data used for surrogate construction, in ActiveSet request vector 3-bit format.

This setting distinguishes derivative data intended for use in construction (includes derivatives w.r.t. the build variables) from derivative data that may be approximated separately (excludes derivatives w.r.t. auxilliary variables). This setting should also not be inferred directly from the responses specification, since we may need gradient support for evaluating gradients at a single point (e.g., the center of a trust region), but not require gradient evaluations at every point.

Referenced by SurfpackApproximation::add_sdp_to_surfdata(), TANA3Approximation::build(), TaylorApproximation::build(), GaussProcApproximation::GaussProcApproximation(), TaylorApproximation::get_gradient(), TaylorApproximation::get_hessian(), TaylorApproximation::get_value(), TaylorApproximation::min_coefficients(), Approximation::min_points(), PecosApproximation::PecosApproximation(), Approximation::recommended_points(), SurfpackApproximation::SurfpackApproximation(), SurfpackApproximation::surrogates_to_surf_data(), TANA3Approximation::TANA3Approximation(), and TaylorApproximation::TaylorApproximation().


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