A class used to perform Adaptive Importance Sampling using a Markov Chain Monte Carlo algorithm. More...

#include <AdaptiveImportanceSampler.h>

Inheritance diagram for AdaptiveImportanceSampler:

Public Types
enum	SampleMode

typedef DataFileName	DataFileParameterType

Public Member Functions
	AdaptiveImportanceSampler (const InputParameters &parameters)

const std::vector< Real > &	getInitialValues () const

const int &	getNumSamplesTrain () const

const bool &	getUseAbsoluteValue () const

const Real &	getOutputLimit () const

const std::vector< Real > &	getImportanceVectorMean () const

const std::vector< Real > &	getImportanceVectorStd () const

const std::vector< const Distribution * > &	getDistributionNames () const

const Real &	getStdFactor () const

virtual bool	isAdaptiveSamplingCompleted () const override
	Returns true if the adaptive sampling is completed.

std::vector< Real >	getNextLocalRow ()

dof_id_type	getNumberOfRows () const

dof_id_type	getNumberOfCols () const

dof_id_type	getNumberOfLocalRows () const

const LocalRankConfig &	getRankConfig (bool batch_mode) const

libMesh::Parallel::Communicator &	getLocalComm ()

virtual bool	enabled () const

std::shared_ptr< MooseObject >	getSharedPtr ()

std::shared_ptr< const MooseObject >	getSharedPtr () const

bool	isKokkosObject () const

MooseApp &	getMooseApp () const

const std::string &	type () const

const std::string &	name () const

std::string	typeAndName () const

MooseObjectParameterName	uniqueParameterName (const std::string &parameter_name) const

MooseObjectName	uniqueName () const

const InputParameters &	parameters () const

const hit::Node *	getHitNode () const

bool	hasBase () const

const std::string &	getBase () const

const T &	getParam (const std::string &name) const

std::vector< std::pair< T1, T2 > >	getParam (const std::string &param1, const std::string &param2) const

const T *	queryParam (const std::string &name) const

const T &	getRenamedParam (const std::string &old_name, const std::string &new_name) const

T	getCheckedPointerParam (const std::string &name, const std::string &error_string="") const

bool	isParamValid (const std::string &name) const

bool	isParamSetByUser (const std::string &name) const

void	connectControllableParams (const std::string &parameter, const std::string &object_type, const std::string &object_name, const std::string &object_parameter) const

void	paramError (const std::string &param, Args... args) const

void	paramWarning (const std::string &param, Args... args) const

void	paramWarning (const std::string &param, Args... args) const

void	paramInfo (const std::string &param, Args... args) const

std::string	messagePrefix (const bool hit_prefix=true) const

std::string	errorPrefix (const std::string &) const

void	mooseError (Args &&... args) const

void	mooseDocumentedError (const std::string &repo_name, const unsigned int issue_num, Args &&... args) const

void	mooseErrorNonPrefixed (Args &&... args) const

void	mooseWarning (Args &&... args) const

void	mooseWarning (Args &&... args) const

void	mooseWarningNonPrefixed (Args &&... args) const

void	mooseWarningNonPrefixed (Args &&... args) const

void	mooseDeprecated (Args &&... args) const

void	mooseDeprecated (Args &&... args) const

void	mooseDeprecatedNoTrace (Args &&... args) const

void	mooseInfo (Args &&... args) const

void	callMooseError (std::string msg, const bool with_prefix, const hit::Node *node=nullptr, const bool show_trace=true) const

std::string	getDataFileName (const std::string &param) const

std::string	getDataFileNameByName (const std::string &relative_path) const

std::string	getDataFilePath (const std::string &relative_path) const

virtual void	initialSetup ()

virtual void	timestepSetup ()

virtual void	jacobianSetup ()

virtual void	residualSetup ()

virtual void	subdomainSetup ()

virtual void	customSetup (const ExecFlagType &)

const ExecFlagEnum &	getExecuteOnEnum () const

PerfGraph &	perfGraph ()

T &	getSampler (const std::string &name)

Sampler &	getSampler (const std::string &name)

T &	getSamplerByName (const SamplerName &name)

Sampler &	getSamplerByName (const SamplerName &name)

const VectorPostprocessorValue &	getVectorPostprocessorValue (const std::string &param_name, const std::string &vector_name) const

const VectorPostprocessorValue &	getVectorPostprocessorValue (const std::string &param_name, const std::string &vector_name, bool needs_broadcast) const

const VectorPostprocessorValue &	getVectorPostprocessorValueByName (const VectorPostprocessorName &name, const std::string &vector_name) const

const VectorPostprocessorValue &	getVectorPostprocessorValueByName (const VectorPostprocessorName &name, const std::string &vector_name, bool needs_broadcast) const

const VectorPostprocessorValue &	getVectorPostprocessorValueOld (const std::string &param_name, const std::string &vector_name) const

const VectorPostprocessorValue &	getVectorPostprocessorValueOld (const std::string &param_name, const std::string &vector_name, bool needs_broadcast) const

const VectorPostprocessorValue &	getVectorPostprocessorValueOldByName (const VectorPostprocessorName &name, const std::string &vector_name) const

const VectorPostprocessorValue &	getVectorPostprocessorValueOldByName (const VectorPostprocessorName &name, const std::string &vector_name, bool needs_broadcast) const

const ScatterVectorPostprocessorValue &	getScatterVectorPostprocessorValue (const std::string &param_name, const std::string &vector_name) const

const ScatterVectorPostprocessorValue &	getScatterVectorPostprocessorValueByName (const VectorPostprocessorName &name, const std::string &vector_name) const

const ScatterVectorPostprocessorValue &	getScatterVectorPostprocessorValueOld (const std::string &param_name, const std::string &vector_name) const

const ScatterVectorPostprocessorValue &	getScatterVectorPostprocessorValueOldByName (const VectorPostprocessorName &name, const std::string &vector_name) const

bool	hasVectorPostprocessor (const std::string &param_name, const std::string &vector_name) const

bool	hasVectorPostprocessor (const std::string &param_name) const

bool	hasVectorPostprocessorByName (const VectorPostprocessorName &name, const std::string &vector_name) const

bool	hasVectorPostprocessorByName (const VectorPostprocessorName &name) const

const VectorPostprocessorName &	getVectorPostprocessorName (const std::string &param_name) const

DenseMatrix< Real >	getGlobalSamples ()

DenseMatrix< Real >	getGlobalSamples ()

DenseMatrix< Real >	getLocalSamples ()

DenseMatrix< Real >	getLocalSamples ()

dof_id_type	getLocalRowBegin () const

dof_id_type	getLocalRowBegin () const

dof_id_type	getLocalRowEnd () const

dof_id_type	getLocalRowEnd () const

const Distribution &	getDistribution (const std::string &name) const

const T &	getDistribution (const std::string &name) const

const Distribution &	getDistribution (const std::string &name) const

const T &	getDistribution (const std::string &name) const

const Distribution &	getDistributionByName (const DistributionName &name) const

const T &	getDistributionByName (const std::string &name) const

const Distribution &	getDistributionByName (const DistributionName &name) const

const T &	getDistributionByName (const std::string &name) const

bool	isVectorPostprocessorDistributed (const std::string &param_name) const

bool	isVectorPostprocessorDistributed (const std::string &param_name) const

bool	isVectorPostprocessorDistributedByName (const VectorPostprocessorName &name) const

bool	isVectorPostprocessorDistributedByName (const VectorPostprocessorName &name) const

const Parallel::Communicator &	comm () const

processor_id_type	n_processors () const

processor_id_type	processor_id () const

bool	isImplicit ()

Moose::StateArg	determineState () const

Static Public Member Functions
static InputParameters	validParams ()

static void	callMooseError (MooseApp const app, const InputParameters &params, std::string msg, const bool with_prefix, const hit::Node node, const bool show_trace=true)

Public Attributes
	usingCombinedWarningSolutionWarnings

const ConsoleStream	_console

Static Public Attributes
static const std::string	type_param

static const std::string	name_param

static const std::string	unique_name_param

static const std::string	app_param

static const std::string	moose_base_param

static const std::string	kokkos_object_param

Protected Member Functions
virtual void	executeSetUp () override

virtual Real	computeSample (dof_id_type row_index, dof_id_type col_index) override
	Return the sample for the given row (the sample index) and column (the parameter index)

void	setNumberOfRandomSeeds (std::size_t number)

Real	getRand (std::size_t n, unsigned int index=0) const

unsigned int	getRandl (std::size_t n, unsigned int lower, unsigned int upper, unsigned int index=0) const

virtual LocalRankConfig	constructRankConfig (bool batch_mode) const

void	flagInvalidSolutionInternal (const InvalidSolutionID invalid_solution_id) const

InvalidSolutionID	registerInvalidSolutionInternal (const std::string &message, const bool warning) const

PerfID	registerTimedSection (const std::string &section_name, const unsigned int level) const

PerfID	registerTimedSection (const std::string &section_name, const unsigned int level, const std::string &live_message, const bool print_dots=true) const

std::string	timedSectionName (const std::string &section_name) const

virtual void	addVectorPostprocessorDependencyHelper (const VectorPostprocessorName &) const

const ReporterContextBase &	getReporterContextBaseByName (const ReporterName &reporter_name) const

const ReporterName &	getReporterName (const std::string &param_name) const

virtual void	addReporterDependencyHelper (const ReporterName &)

void	setNumberOfRows (dof_id_type n_rows)

void	setNumberOfRows (dof_id_type n_rows)

void	setNumberOfCols (dof_id_type n_cols)

void	setNumberOfCols (dof_id_type n_cols)

virtual void	computeSampleMatrix (DenseMatrix< Real > &matrix)

virtual void	computeSampleMatrix (DenseMatrix< Real > &matrix)

virtual void	computeLocalSampleMatrix (DenseMatrix< Real > &matrix)

virtual void	computeLocalSampleMatrix (DenseMatrix< Real > &matrix)

virtual void	computeSampleRow (dof_id_type i, std::vector< Real > &data)

virtual void	computeSampleRow (dof_id_type i, std::vector< Real > &data)

virtual void	advanceGenerators (const dof_id_type count)

virtual void	advanceGenerators (const dof_id_type count)

virtual void	advanceGenerator (const unsigned int seed_index, const dof_id_type count)

virtual void	advanceGenerator (const unsigned int seed_index, const dof_id_type count)

void	setAutoAdvanceGenerators (const bool state)

void	setAutoAdvanceGenerators (const bool state)

virtual void	executeTearDown ()

virtual void	executeTearDown ()

const T &	getReporterValue (const std::string &param_name, const std::size_t time_index=0)

const T &	getReporterValue (const std::string &param_name, ReporterMode mode, const std::size_t time_index=0)

const T &	getReporterValue (const std::string &param_name, const std::size_t time_index=0)

const T &	getReporterValue (const std::string &param_name, ReporterMode mode, const std::size_t time_index=0)

const T &	getReporterValueByName (const ReporterName &reporter_name, const std::size_t time_index=0)

const T &	getReporterValueByName (const ReporterName &reporter_name, ReporterMode mode, const std::size_t time_index=0)

const T &	getReporterValueByName (const ReporterName &reporter_name, const std::size_t time_index=0)

const T &	getReporterValueByName (const ReporterName &reporter_name, ReporterMode mode, const std::size_t time_index=0)

bool	hasReporterValue (const std::string &param_name) const

bool	hasReporterValue (const std::string &param_name) const

bool	hasReporterValue (const std::string &param_name) const

bool	hasReporterValue (const std::string &param_name) const

bool	hasReporterValueByName (const ReporterName &reporter_name) const

bool	hasReporterValueByName (const ReporterName &reporter_name) const

bool	hasReporterValueByName (const ReporterName &reporter_name) const

bool	hasReporterValueByName (const ReporterName &reporter_name) const

Protected Attributes
std::vector< const Distribution * >	_distributions
	Storage for distribution objects to be utilized.

const std::vector< Real > &	_proposal_std
	The proposal distribution standard deviations.

const std::vector< Real > &	_initial_values
	Initial values values vector to start the importance sampler.

const Real &	_output_limit
	The output limit, exceedance of which indicates failure.

const int &	_num_samples_train
	Number of samples to train the importance sampler.

const int &	_num_importance_sampling_steps
	Number of importance sampling steps (after the importance distribution has been trained)

const Real &	_std_factor
	Factor to be multiplied to the standard deviation of the proposal distribution.

const bool &	_use_absolute_value
	Absolute value of the model result. Use this when failure is defined as a non-exceedance rather than an exceedance.

const unsigned int &	_num_random_seeds
	Initialize a certain number of random seeds. Change from the default only if you have to.

bool	_is_sampling_completed
	True if the sampling is completed.

const dof_id_type	_min_procs_per_row

const dof_id_type	_max_procs_per_row

libMesh::Parallel::Communicator	_local_comm

const bool &	_enabled

MooseApp &	_app

Factory &	_factory

ActionFactory &	_action_factory

const std::string &	_type

const std::string &	_name

const InputParameters &	_pars

const ExecFlagEnum &	_execute_enum

const ExecFlagType &	_current_execute_flag

MooseApp &	_pg_moose_app

const std::string	_prefix

const Parallel::Communicator &	_communicator

const InputParameters &	_ti_params

FEProblemBase &	_ti_feproblem

bool	_is_implicit

Real &	_t

const Real &	_t_old

int &	_t_step

Real &	_dt

Real &	_dt_old

bool	_is_transient

Private Member Functions
void	execute ()

void	checkReinitStatus () const

void	advanceGeneratorsInternal (const dof_id_type count)

const VectorPostprocessorValue &	getVectorPostprocessorByNameHelper (const VectorPostprocessorName &name, const std::string &vector_name, bool broadcast, std::size_t t_index) const

const VectorPostprocessorContext< VectorPostprocessorValue > &	getVectorPostprocessorContextByNameHelper (const VectorPostprocessorName &name, const std::string &vector_name) const

bool	vectorPostprocessorsAdded () const

bool	reportersAdded () const

void	possiblyCheckHasReporter (const ReporterName &reporter_name, const std::string &param_name="") const

void	init ()

void	init ()

void	reinit ()

void	reinit ()

void	possiblyCheckHasVectorPostprocessor (const std::string &param_name, const std::string &vector_name) const

void	possiblyCheckHasVectorPostprocessor (const std::string &param_name, const std::string &vector_name) const

void	possiblyCheckHasVectorPostprocessorByName (const VectorPostprocessorName &name, const std::string &vector_name) const

void	possiblyCheckHasVectorPostprocessorByName (const VectorPostprocessorName &name, const std::string &vector_name) const

Static Private Member Functions
static const hit::Node *	getHitNode (const InputParameters &params)

static std::string	messagePrefix (const InputParameters &params, const bool hit_prefix)

Private Attributes
const std::vector< std::vector< Real > > &	_inputs
	Storage for the inputs vector obtained from the reporter.

std::vector< Real >	_prev_value
	For proposing the next sample in the MCMC algorithm.

std::vector< Real >	_mean_sto
	Storage for means of input values for proposing the next sample.

std::vector< Real >	_std_sto
	Storage for standard deviations of input values for proposing the next sample.

std::vector< std::vector< Real > >	_inputs_sto
	Storage for previously accepted samples by the decision reporter system.

int	_retraining_steps
	Number of retraining performed.

const std::vector< bool > *const	_gp_flag
	Indicate whether GP prediction is good or bad to influence next proposed sample.

std::vector< std::unique_ptr< MooseRandomStateless > >	_generators

dof_id_type	_n_local_rows

dof_id_type	_local_row_begin

dof_id_type	_local_row_end

dof_id_type	_n_rows

dof_id_type	_n_cols

std::size_t	_n_seeds

dof_id_type	_next_local_row

bool	_next_local_row_requires_state_restore

bool	_initialized

bool	_needs_reinit

bool	_has_executed

const dof_id_type	_limit_get_global_samples

const dof_id_type	_limit_get_local_samples

const dof_id_type	_limit_get_next_local_row

std::pair< LocalRankConfig, LocalRankConfig >	_rank_config

bool	_auto_advance_generators

const ParallelParamObject &	_parent

const MooseBase &	_si_moose_base

const FEProblemBase *	_si_problem

ExecFlagEnum	_empty_execute_enum

const InputParameters &	_dni_params

FEProblemBase &	_dni_feproblem

const MooseObject *const	_dni_moose_object_ptr

const InputParameters &	_si_params

FEProblemBase &	_si_feproblem

THREAD_ID	_si_tid

const bool	_broadcast_by_default

const MooseObject &	_vpi_moose_object

const FEProblemBase &	_vpi_feproblem

const THREAD_ID	_vpi_tid

const InputParameters &	_ri_params

FEProblemBase &	_ri_fe_problem_base

const ReporterData &	_ri_reporter_data

const MooseObject &	_ri_moose_object

const std::string	_ti_name

Detailed Description

A class used to perform Adaptive Importance Sampling using a Markov Chain Monte Carlo algorithm.

Definition at line 18 of file AdaptiveImportanceSampler.h.

Constructor & Destructor Documentation

◆ AdaptiveImportanceSampler()

AdaptiveImportanceSampler::AdaptiveImportanceSampler ( const InputParameters & parameters )

Definition at line 53 of file AdaptiveImportanceSampler.C.

  : Sampler(parameters),
    TransientInterface(this),
    _proposal_std(getParam<std::vector<Real>>("proposal_std")),
    _initial_values(getParam<std::vector<Real>>("initial_values")),
    _output_limit(getParam<Real>("output_limit")),
    _num_samples_train(getParam<int>("num_samples_train")),
    _num_importance_sampling_steps(getParam<int>("num_importance_sampling_steps")),
    _std_factor(getParam<Real>("std_factor")),
    _use_absolute_value(getParam<bool>("use_absolute_value")),
    _num_random_seeds(getParam<unsigned int>("num_random_seeds")),
    _is_sampling_completed(false),
    _inputs(getReporterValue<std::vector<std::vector<Real>>>("inputs_reporter")),
    _retraining_steps(0),
    _gp_flag(isParamValid("flag_sample") ? &getReporterValue<std::vector<bool>>("flag_sample")
                                         : nullptr)
{
  // Filling the `distributions` vector with the user-provided distributions.
  for (const DistributionName & name : getParam<std::vector<DistributionName>>("distributions"))
    _distributions.push_back(&getDistributionByName(name));
 
  /* Adaptive Importance Sampling (AdaptiveImportanceSampler) relies on a Markov Chain Monte Carlo
     (MCMC) algorithm. As such, in MOOSE, any use of MCMC algorithms requires that the `num_steps`
     parameter in the main App's executioner would control the total number of samples. Therefore,
     the `num_rows` parameter typically used by exisiting non-MCMC samplers to set the total number
     of samples has no use here and is fixed to 1.*/
  setNumberOfRows(1);
 
  // Setting the number of columns in the sampler matrix (equal to the number of distributions).
  setNumberOfCols(_distributions.size());
 
  /* `inputs_sto` is a member variable that aids in forming the importance distribution.
     One dimension of this variable is equal to the number of distributions. The other dimension
     of the variable, at the last step, is equal to the number of samples the user desires.*/
  _inputs_sto.resize(_distributions.size());
 
  // Mapping all the input distributions to a standard normal space
  for (unsigned int i = 0; i < _distributions.size(); ++i)
    _inputs_sto[i].push_back(Normal::quantile(_distributions[i]->cdf(_initial_values[i]), 0, 1));
 
  /* `prev_value` is a member variable for tracking the previously accepted samples in the
     MCMC algorithm and proposing the next sample.*/
  _prev_value.resize(_distributions.size());
 
  // Storage for means of input values for proposing the next sample
  _mean_sto.resize(_distributions.size());
 
  // Storage for standard deviations of input values for proposing the next sample
  _std_sto.resize(_distributions.size());
 
  setNumberOfRandomSeeds(_num_random_seeds);
  setAutoAdvanceGenerators(false);
}

Member Function Documentation

◆ computeSample()

Real AdaptiveImportanceSampler::computeSample	(	dof_id_type	row_index,
		dof_id_type	col_index
	)

overrideprotectedvirtual

Return the sample for the given row (the sample index) and column (the parameter index)

Implements Sampler.

Definition at line 177 of file AdaptiveImportanceSampler.C.

{
  return _distributions[col_index]->quantile(Normal::cdf(_prev_value[col_index], 0.0, 1.0));
}

◆ executeSetUp()

void AdaptiveImportanceSampler::executeSetUp ( )

overrideprotectedvirtual

Reimplemented from Sampler.

Definition at line 108 of file AdaptiveImportanceSampler.C.

{
  const bool sample = _t_step > 1;
  const bool gp_flag = _gp_flag ? (*_gp_flag)[0] : false;
 
  if (sample && _is_sampling_completed)
    mooseError("Internal bug: the adaptive sampling is supposed to be completed but another sample "
               "has been requested.");
 
  if (_t_step <= _num_samples_train)
  {
    /* This is the importance distribution training step. Markov Chains are set up
      to sample from the importance region or the failure region using the Metropolis
      algorithm. Given that the previous sample resulted in a model failure, the next
      sample is proposed such that it is very likely to result in a model failure as well.
      The `initial_values` and `proposal_std` parameters provided by the user affects the
      formation of the importance distribution. */
    if (sample && !gp_flag)
    {
      for (dof_id_type j = 0; j < _distributions.size(); ++j)
        _prev_value[j] = Normal::quantile(_distributions[j]->cdf(_inputs[j][0]), 0.0, 1.0);
      Real acceptance_ratio = 0.0;
      for (dof_id_type i = 0; i < _distributions.size(); ++i)
        acceptance_ratio += std::log(Normal::pdf(_prev_value[i], 0.0, 1.0)) -
                            std::log(Normal::pdf(_inputs_sto[i].back(), 0.0, 1.0));
      if (acceptance_ratio > std::log(getRand(0, _t_step)))
      {
        for (dof_id_type i = 0; i < _distributions.size(); ++i)
          _inputs_sto[i].push_back(_prev_value[i]);
      }
      else
      {
        for (dof_id_type i = 0; i < _distributions.size(); ++i)
          _inputs_sto[i].push_back(_inputs_sto[i].back());
      }
      for (dof_id_type i = 0; i < _distributions.size(); ++i)
        _prev_value[i] =
            Normal::quantile(getRand(i + 1, _t_step), _inputs_sto[i].back(), _proposal_std[i]);
    }
  }
  else if (sample && !gp_flag)
  {
    /* This is the importance sampling step using the importance distribution created
      in the previous step. Once the importance distribution is known, sampling from
      it is similar to a regular Monte Carlo sampling. */
    for (dof_id_type i = 0; i < _distributions.size(); ++i)
    {
      if (_t_step == _num_samples_train + 1)
      {
        _mean_sto[i] = AdaptiveMonteCarloUtils::computeMean(_inputs_sto[i], 1);
        _std_sto[i] = AdaptiveMonteCarloUtils::computeSTD(_inputs_sto[i], 1);
      }
      _prev_value[i] =
          (Normal::quantile(getRand(i, _t_step), _mean_sto[i], _std_factor * _std_sto[i]));
    }
 
    // check if we have performed all the importance sampling steps
    if (_t_step >= _num_samples_train + _num_importance_sampling_steps + _retraining_steps)
      _is_sampling_completed = true;
  }
 
  // When the GP fails, the current time step is 'wasted' and the retraining step doesn't
  // happen until the next time step. Therefore, keep track of the number of retraining steps
  // to increase the total number of steps taken.
  if (sample && gp_flag && _t_step > _num_samples_train)
    ++_retraining_steps;
}

◆ getDistributionNames()

const std::vector< const Distribution * > & AdaptiveImportanceSampler::getDistributionNames ( ) const

inline

Definition at line 44 of file AdaptiveImportanceSampler.h.

44{ return _distributions; }

Referenced by AdaptiveImportanceStats::AdaptiveImportanceStats().

◆ getImportanceVectorMean()

const std::vector< Real > & AdaptiveImportanceSampler::getImportanceVectorMean ( ) const

inline

Definition at line 38 of file AdaptiveImportanceSampler.h.

38{ return _mean_sto; }

Referenced by AdaptiveImportanceStats::execute().

◆ getImportanceVectorStd()

const std::vector< Real > & AdaptiveImportanceSampler::getImportanceVectorStd ( ) const

inline

Definition at line 41 of file AdaptiveImportanceSampler.h.

41{ return _std_sto; }

Referenced by AdaptiveImportanceStats::execute().

◆ getInitialValues()

const std::vector< Real > & AdaptiveImportanceSampler::getInitialValues ( ) const

inline

Definition at line 26 of file AdaptiveImportanceSampler.h.

26{ return _initial_values; }

Referenced by AdaptiveMonteCarloDecision::AdaptiveMonteCarloDecision().

◆ getNumSamplesTrain()

const int & AdaptiveImportanceSampler::getNumSamplesTrain ( ) const

inline

Definition at line 29 of file AdaptiveImportanceSampler.h.

29{ return _num_samples_train; }

Referenced by AdaptiveImportanceStats::execute(), and AdaptiveMonteCarloDecision::execute().

◆ getOutputLimit()

const Real & AdaptiveImportanceSampler::getOutputLimit ( ) const

inline

Definition at line 35 of file AdaptiveImportanceSampler.h.

35{ return _output_limit; }

Referenced by AdaptiveMonteCarloDecision::AdaptiveMonteCarloDecision(), and AdaptiveImportanceStats::execute().

◆ getStdFactor()

const Real & AdaptiveImportanceSampler::getStdFactor ( ) const

inline

Definition at line 47 of file AdaptiveImportanceSampler.h.

47{ return _std_factor; }

Referenced by AdaptiveImportanceStats::AdaptiveImportanceStats().

◆ getUseAbsoluteValue()

const bool & AdaptiveImportanceSampler::getUseAbsoluteValue ( ) const

inline

Definition at line 32 of file AdaptiveImportanceSampler.h.

32{ return _use_absolute_value; }

Referenced by AdaptiveImportanceStats::execute(), and AdaptiveMonteCarloDecision::execute().

◆ isAdaptiveSamplingCompleted()

virtual bool AdaptiveImportanceSampler::isAdaptiveSamplingCompleted ( ) const

inlineoverridevirtual

Returns true if the adaptive sampling is completed.

Reimplemented from Sampler.

Definition at line 52 of file AdaptiveImportanceSampler.h.

52{ return _is_sampling_completed; }

◆ validParams()

InputParameters AdaptiveImportanceSampler::validParams ( )

static

Definition at line 19 of file AdaptiveImportanceSampler.C.

{
  InputParameters params = Sampler::validParams();
  params.addClassDescription("Adaptive Importance Sampler.");
  params.addRequiredParam<std::vector<DistributionName>>(
      "distributions",
      "The distribution names to be sampled, the number of distributions provided defines the "
      "number of columns per matrix.");
  params.addRequiredParam<ReporterName>("inputs_reporter", "Reporter with input parameters.");
  params.addRequiredParam<std::vector<Real>>("proposal_std",
                                             "Standard deviations of the proposal distributions");
  params.addRequiredParam<Real>("output_limit", "Limiting values of the VPPs");
  params.addRequiredParam<std::vector<Real>>(
      "initial_values", "Initial input values to get the importance sampler started");
  params.addRequiredRangeCheckedParam<int>(
      "num_samples_train",
      "num_samples_train>0",
      "Number of samples to learn the importance distribution");
  params.addRequiredRangeCheckedParam<int>(
      "num_importance_sampling_steps",
      "num_importance_sampling_steps>0",
      "Number of importance sampling steps (after the importance distribution has been trained)");
  params.addRequiredParam<Real>(
      "std_factor", "Factor to be multiplied to the standard deviation of the importance samples");
  params.addParam<bool>("use_absolute_value", false, "Use absolute value of the sub app output");
  params.addParam<unsigned int>(
      "num_random_seeds",
      100000,
      "Initialize a certain number of random seeds. Change from the default only if you have to.");
  params.addParam<ReporterName>("flag_sample",
                                "Flag samples if the surrogate prediction was inadequate.");
  return params;
}

Referenced by AISActiveLearning::validParams().

Member Data Documentation

◆ _distributions

std::vector<const Distribution *> AdaptiveImportanceSampler::_distributions

protected

Storage for distribution objects to be utilized.

Definition at line 60 of file AdaptiveImportanceSampler.h.

Referenced by AdaptiveImportanceSampler(), computeSample(), executeSetUp(), and getDistributionNames().

◆ _gp_flag

const std::vector<bool>* const AdaptiveImportanceSampler::_gp_flag

private

Indicate whether GP prediction is good or bad to influence next proposed sample.

Definition at line 109 of file AdaptiveImportanceSampler.h.

Referenced by executeSetUp().

◆ _initial_values

const std::vector<Real>& AdaptiveImportanceSampler::_initial_values

protected

Initial values values vector to start the importance sampler.

Definition at line 66 of file AdaptiveImportanceSampler.h.

Referenced by AdaptiveImportanceSampler(), and getInitialValues().

◆ _inputs

const std::vector<std::vector<Real> >& AdaptiveImportanceSampler::_inputs

private

Storage for the inputs vector obtained from the reporter.

Definition at line 91 of file AdaptiveImportanceSampler.h.

Referenced by executeSetUp().

◆ _inputs_sto

std::vector<std::vector<Real> > AdaptiveImportanceSampler::_inputs_sto

private

Storage for previously accepted samples by the decision reporter system.

Definition at line 103 of file AdaptiveImportanceSampler.h.

Referenced by AdaptiveImportanceSampler(), and executeSetUp().

◆ _is_sampling_completed

bool AdaptiveImportanceSampler::_is_sampling_completed

protected

True if the sampling is completed.

Definition at line 87 of file AdaptiveImportanceSampler.h.

Referenced by executeSetUp(), and isAdaptiveSamplingCompleted().

◆ _mean_sto

std::vector<Real> AdaptiveImportanceSampler::_mean_sto

private

Storage for means of input values for proposing the next sample.

Definition at line 97 of file AdaptiveImportanceSampler.h.

Referenced by AdaptiveImportanceSampler(), executeSetUp(), and getImportanceVectorMean().

◆ _num_importance_sampling_steps

const int& AdaptiveImportanceSampler::_num_importance_sampling_steps

protected

Number of importance sampling steps (after the importance distribution has been trained)

Definition at line 75 of file AdaptiveImportanceSampler.h.

Referenced by executeSetUp().

◆ _num_random_seeds

const unsigned int& AdaptiveImportanceSampler::_num_random_seeds

protected

Initialize a certain number of random seeds. Change from the default only if you have to.

Definition at line 84 of file AdaptiveImportanceSampler.h.

Referenced by AdaptiveImportanceSampler().

◆ _num_samples_train

const int& AdaptiveImportanceSampler::_num_samples_train

protected

Number of samples to train the importance sampler.

Definition at line 72 of file AdaptiveImportanceSampler.h.

Referenced by executeSetUp(), and getNumSamplesTrain().

◆ _output_limit

const Real& AdaptiveImportanceSampler::_output_limit

protected

The output limit, exceedance of which indicates failure.

Definition at line 69 of file AdaptiveImportanceSampler.h.

Referenced by getOutputLimit().

◆ _prev_value

std::vector<Real> AdaptiveImportanceSampler::_prev_value

private

For proposing the next sample in the MCMC algorithm.

Definition at line 94 of file AdaptiveImportanceSampler.h.

Referenced by AdaptiveImportanceSampler(), computeSample(), and executeSetUp().

◆ _proposal_std

const std::vector<Real>& AdaptiveImportanceSampler::_proposal_std

protected

The proposal distribution standard deviations.

Definition at line 63 of file AdaptiveImportanceSampler.h.

Referenced by executeSetUp().

◆ _retraining_steps

int AdaptiveImportanceSampler::_retraining_steps

private

Number of retraining performed.

Definition at line 106 of file AdaptiveImportanceSampler.h.

Referenced by executeSetUp().

◆ _std_factor

const Real& AdaptiveImportanceSampler::_std_factor

protected

Factor to be multiplied to the standard deviation of the proposal distribution.

Definition at line 78 of file AdaptiveImportanceSampler.h.

Referenced by executeSetUp(), and getStdFactor().

◆ _std_sto

std::vector<Real> AdaptiveImportanceSampler::_std_sto

private

Storage for standard deviations of input values for proposing the next sample.

Definition at line 100 of file AdaptiveImportanceSampler.h.

Referenced by AdaptiveImportanceSampler(), executeSetUp(), and getImportanceVectorStd().

◆ _use_absolute_value

const bool& AdaptiveImportanceSampler::_use_absolute_value

protected

Absolute value of the model result. Use this when failure is defined as a non-exceedance rather than an exceedance.

Definition at line 81 of file AdaptiveImportanceSampler.h.

Referenced by getUseAbsoluteValue().

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

stochastic_tools/include/samplers/AdaptiveImportanceSampler.h
stochastic_tools/src/samplers/AdaptiveImportanceSampler.C

Public Types

Public Member Functions

Static Public Member Functions

Public Attributes

Static Public Attributes

Protected Member Functions

Protected Attributes

Private Member Functions

Static Private Member Functions

Private Attributes

Detailed Description

Constructor & Destructor Documentation

◆ AdaptiveImportanceSampler()

Member Function Documentation

◆ computeSample()

◆ executeSetUp()

◆ getDistributionNames()

◆ getImportanceVectorMean()

◆ getImportanceVectorStd()

◆ getInitialValues()

◆ getNumSamplesTrain()

◆ getOutputLimit()

◆ getStdFactor()

◆ getUseAbsoluteValue()

◆ isAdaptiveSamplingCompleted()

◆ validParams()

Member Data Documentation

◆ _distributions

◆ _gp_flag

◆ _initial_values

◆ _inputs

◆ _inputs_sto

◆ _is_sampling_completed

◆ _mean_sto

◆ _num_importance_sampling_steps

◆ _num_random_seeds

◆ _num_samples_train

◆ _output_limit

◆ _prev_value

◆ _proposal_std

◆ _retraining_steps

◆ _std_factor

◆ _std_sto

◆ _use_absolute_value