Dakota  Version 6.4
 All Classes Namespaces Files Functions Variables Typedefs Enumerations Enumerator Friends Pages
Public Member Functions | Private Member Functions | Static Private Member Functions | Private Attributes | List of all members
DDACEDesignCompExp Class Reference

Wrapper class for the DDACE design of experiments library. More...

Inheritance diagram for DDACEDesignCompExp:
PStudyDACE Analyzer Iterator

Public Member Functions

 DDACEDesignCompExp (ProblemDescDB &problem_db, Model &model)
 primary constructor for building a standard DACE iterator More...
 
 DDACEDesignCompExp (Model &model, int samples, int symbols, int seed, unsigned short sampling_method)
 alternate constructor used for building approximations More...
 
 ~DDACEDesignCompExp ()
 destructor
 
bool resize ()
 reinitializes iterator based on new variable size
 
void pre_run ()
 pre-run portion of run (optional); re-implemented by Iterators which can generate all Variables (parameter sets) a priori More...
 
void core_run ()
 core portion of run; implemented by all derived classes and may include pre/post steps in lieu of separate pre/post More...
 
void post_input ()
 read tabular data for post-run mode
 
void post_run (std::ostream &s)
 post-run portion of run (optional); verbose to print results; re-implemented by Iterators that can read all Variables/Responses and perform final analysis phase in a standalone way More...
 
int num_samples () const
 
void sampling_reset (int min_samples, bool all_data_flag, bool stats_flag)
 reset sampling iterator to use at least min_samples
 
unsigned short sampling_scheme () const
 return sampling name
 
void vary_pattern (bool pattern_flag)
 sets varyPattern in derived classes that support it
 
void get_parameter_sets (Model &model)
 Returns one block of samples (ndim * num_samples)
 

Private Member Functions

void compute_main_effects ()
 builds a DDaceMainEffects::OneWayANOVA if mainEffectsFlag is set
 
boost::shared_ptr
< DDaceSamplerBase > 
create_sampler (Model &model)
 create a DDACE sampler
 
void resolve_samples_symbols ()
 convenience function for resolving number of samples and number of symbols from input. More...
 

Static Private Member Functions

static void copy_data (const std::vector< DDaceSamplePoint > &dspa, Real *ptr, const int ptr_len)
 copy DDACE point to RealVector More...
 

Private Attributes

unsigned short daceMethod
 oas, lhs, oa_lhs, random, box_behnken, central_composite, or grid
 
int samplesSpec
 initial specification of number of samples
 
int symbolsSpec
 initial specification of number of symbols
 
int numSamples
 current number of samples to be evaluated
 
int numSymbols
 current number of symbols to be used in generating the sample set (inversely related to number of replications)
 
const int seedSpec
 the user seed specification for the random number generator (allows repeatable results)
 
int randomSeed
 current seed for the random number generator
 
bool allDataFlag
 flag which triggers the update of allVars/allResponses for use by Iterator::all_variables() and Iterator::all_responses()
 
size_t numDACERuns
 counter for number of executions for this object
 
bool varyPattern
 flag for continuing the random number sequence from a previous execution (e.g., for surrogate-based optimization) so that multiple executions are repeatable but not correlated.
 
bool mainEffectsFlag
 flag which specifies main effects
 
std::vector< std::vector< int > > symbolMapping
 mapping of symbols for main effects calculations
 

Additional Inherited Members

- Protected Member Functions inherited from PStudyDACE
 PStudyDACE (ProblemDescDB &problem_db, Model &model)
 constructor
 
 PStudyDACE (unsigned short method_name, Model &model)
 alternate constructor for instantiations "on the fly"
 
 ~PStudyDACE ()
 destructor
 
void print_results (std::ostream &s)
 print the final iterator results More...
 
void volumetric_quality (int ndim, int num_samples, double *sample_points)
 Calculation of volumetric quality measures. More...
 
- Static Protected Member Functions inherited from Iterator
static void gnewton_set_recast (const Variables &recast_vars, const ActiveSet &recast_set, ActiveSet &sub_model_set)
 conversion of request vector values for the Gauss-Newton Hessian approximation More...
 
- Protected Attributes inherited from PStudyDACE
SensAnalysisGlobal pStudyDACESensGlobal
 initialize statistical post processing
 
bool volQualityFlag
 flag which specifies evaluation of volumetric quality measures
 
bool varBasedDecompFlag
 flag which specifies calculating variance based decomposition sensitivity analysis metrics
 

Detailed Description

Wrapper class for the DDACE design of experiments library.

The DDACEDesignCompExp class provides a wrapper for DDACE, a C++ design of experiments library from the Computational Sciences and Mathematics Research (CSMR) department at Sandia's Livermore CA site. This class uses design and analysis of computer experiments (DACE) methods to sample the design space spanned by the bounds of a Model. It returns all generated samples and their corresponding responses as well as the best sample found.

Constructor & Destructor Documentation

DDACEDesignCompExp ( ProblemDescDB problem_db,
Model model 
)

primary constructor for building a standard DACE iterator

This constructor is called for a standard iterator built with data from probDescDB.

References Dakota::abort_handler(), DDACEDesignCompExp::daceMethod, DDACEDesignCompExp::mainEffectsFlag, Iterator::maxEvalConcurrency, Analyzer::numContinuousVars, and DDACEDesignCompExp::numSamples.

DDACEDesignCompExp ( Model model,
int  samples,
int  symbols,
int  seed,
unsigned short  sampling_method 
)

alternate constructor used for building approximations

This alternate constructor is used for instantiations on-the-fly, using only the incoming data. No problem description database queries are used.

References Iterator::maxEvalConcurrency, DDACEDesignCompExp::numSamples, and DDACEDesignCompExp::resolve_samples_symbols().

Member Function Documentation

void pre_run ( )
virtual

pre-run portion of run (optional); re-implemented by Iterators which can generate all Variables (parameter sets) a priori

pre-run phase, which a derived iterator may optionally reimplement; when not present, pre-run is likely integrated into the derived run function. This is a virtual function; when re-implementing, a derived class must call its nearest parent's pre_run(), if implemented, typically before performing its own implementation steps.

Reimplemented from Analyzer.

References DDACEDesignCompExp::get_parameter_sets(), Iterator::iteratedModel, Analyzer::pre_run(), and PStudyDACE::varBasedDecompFlag.

void core_run ( )
virtual

core portion of run; implemented by all derived classes and may include pre/post steps in lieu of separate pre/post

Virtual run function for the iterator class hierarchy. All derived classes need to redefine it.

Reimplemented from Iterator.

References DDACEDesignCompExp::allDataFlag, Analyzer::evaluate_parameter_sets(), Iterator::iteratedModel, DDACEDesignCompExp::mainEffectsFlag, Analyzer::numContinuousVars, Analyzer::numLSqTerms, Analyzer::numObjFns, DDACEDesignCompExp::numSamples, Iterator::subIteratorFlag, PStudyDACE::varBasedDecompFlag, and Analyzer::variance_based_decomp().

void post_run ( std::ostream &  s)
virtual

post-run portion of run (optional); verbose to print results; re-implemented by Iterators that can read all Variables/Responses and perform final analysis phase in a standalone way

Post-run phase, which a derived iterator may optionally reimplement; when not present, post-run is likely integrated into run. This is a virtual function; when re-implementing, a derived class must call its nearest parent's post_run(), typically after performing its own implementation steps.

Reimplemented from Analyzer.

References Dakota::abort_handler(), Analyzer::allResponses, Analyzer::allSamples, SensAnalysisGlobal::compute_correlations(), DDACEDesignCompExp::compute_main_effects(), DDACEDesignCompExp::create_sampler(), Iterator::iteratedModel, DDACEDesignCompExp::mainEffectsFlag, Analyzer::post_run(), PStudyDACE::pStudyDACESensGlobal, DDACEDesignCompExp::seedSpec, Iterator::subIteratorFlag, DDACEDesignCompExp::symbolMapping, and PStudyDACE::varBasedDecompFlag.

int num_samples ( ) const
inlinevirtual

Return current number of evaluation points. Since the calculation of samples, collocation points, etc. might be costly, provide a default implementation here that backs out from the maxEvalConcurrency.

Reimplemented from Analyzer.

References DDACEDesignCompExp::numSamples.

void resolve_samples_symbols ( )
private

convenience function for resolving number of samples and number of symbols from input.

This function must define a combination of samples and symbols that is acceptable for a particular sampling algorithm. Users provide requests for these quantities, but this function must enforce any restrictions imposed by the sampling algorithms.

References Dakota::abort_handler(), DDACEDesignCompExp::daceMethod, Analyzer::numContinuousVars, DDACEDesignCompExp::numSamples, DDACEDesignCompExp::numSymbols, and Iterator::submethod_enum_to_string().

Referenced by DDACEDesignCompExp::DDACEDesignCompExp(), DDACEDesignCompExp::get_parameter_sets(), and DDACEDesignCompExp::post_input().

void copy_data ( const std::vector< DDaceSamplePoint > &  dspa,
Real *  ptr,
const int  ptr_len 
)
staticprivate

copy DDACE point to RealVector

copy DDACE point array to RealVectorArray copy DDACE point array to Real*

References Dakota::abort_handler().

Referenced by DDACEDesignCompExp::get_parameter_sets().


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