#include <BayesianPosteriorTargeted.h>

Inheritance diagram for BayesianPosteriorTargeted:

Public Types
typedef DataFileName	DataFileParameterType

Public Member Functions
	BayesianPosteriorTargeted (const InputParameters &parameters)

void	computeAcquisition (std::vector< Real > &acq, const std::vector< Real > &gp_mean, const std::vector< Real > &gp_std, const std::vector< std::vector< Real >> &test_inputs, const std::vector< std::vector< Real >> &train_inputs, const std::vector< Real > &generic) const
	Compute the acquisition function values. More...

void	penalizeAcquisition (std::vector< Real > &modified_acq, std::vector< unsigned int > &sorted_indices, const std::vector< Real > &acq, const std::vector< Real > &length_scales, const std::vector< std::vector< Real >> &inputs)
	Return the modified acquisition function values and sorted indices considering local penalization (inspired from Zhan et al. More...

void	computeCorrelation (Real &corr, const std::vector< Real > &input1, const std::vector< Real > &input2, const std::vector< Real > &length_scales)
	Compute the correlation between two inputs using the length scales. More...

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	haveParameter (const std::string &name) 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

const Parallel::Communicator &	comm () const

processor_id_type	n_processors () const

processor_id_type	processor_id () 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
void	computeAcquisitionInternal (std::vector< Real > &acq, const std::vector< Real > &gp_mean, const std::vector< Real > &gp_std, const std::vector< std::vector< Real >> &test_inputs, const std::vector< std::vector< Real >> &train_inputs, const std::vector< Real > &generic) const override
	Implementation hook for derived classes (no size checks here). More...

void	flagInvalidSolutionInternal (const InvalidSolutionID invalid_solution_id) const

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

Protected Attributes
const bool &	_enabled

MooseApp &	_app

Factory &	_factory

ActionFactory &	_action_factory

const std::string &	_type

const std::string &	_name

const InputParameters &	_pars

const Parallel::Communicator &	_communicator

Detailed Description

Definition at line 14 of file BayesianPosteriorTargeted.h.

Constructor & Destructor Documentation

◆ BayesianPosteriorTargeted()

BayesianPosteriorTargeted::BayesianPosteriorTargeted ( const InputParameters & parameters )

Definition at line 23 of file BayesianPosteriorTargeted.C.

   : ParallelAcquisitionFunctionBase(parameters)
 {
 }

Member Function Documentation

◆ computeAcquisition()

void ParallelAcquisitionFunctionBase::computeAcquisition	(	std::vector< Real > &	acq,
		const std::vector< Real > &	gp_mean,
		const std::vector< Real > &	gp_std,
		const std::vector< std::vector< Real >> &	test_inputs,
		const std::vector< std::vector< Real >> &	train_inputs,
		const std::vector< Real > &	generic
	)		const

inherited

Compute the acquisition function values.

Performs all argument/size checks, then dispatches to computeAcquisitionInternal implemented by derived classes.

Parameters

acq	The computed acquisition function values
gp_mean	The provided GP mean values
gp_std	The provided GP standard deviation values
test_inputs	All the input values under which the GP has to be tested
train_inputs	All the input values under which the GP has been trained
generic	A generic parameter (can be output values under which the GP has been trained or threshold parameter under the U-Function etc.)

Definition at line 28 of file ParallelAcquisitionFunctionBase.C.

 {
   // Size checks
   if (test_inputs.size() == 0)
     mooseError("computeAcquisition: test_inputs must be non-empty.");
 
   if (gp_mean.size() != test_inputs.size() || gp_std.size() != test_inputs.size())
     mooseError("computeAcquisition: gp_mean, gp_std, and test_inputs must have the same length.");
 
   if (acq.size() != test_inputs.size())
     mooseError("computeAcquisition: output 'acq' must be pre-sized to test_inputs.size().");
 
   // Dimensionality checks (all rows same dim; train dim matches test dim)
   if (test_inputs.front().size() == 0)
     mooseError("computeAcquisition: test_inputs row dimension must be > 0.");
 
   if (train_inputs[0].size() != test_inputs.front().size())
     mooseError("computeAcquisition: train_inputs rows must match test_inputs dimension.");
 
   // Hand off to the derived implementation
   computeAcquisitionInternal(acq, gp_mean, gp_std, test_inputs, train_inputs, generic);
 }

◆ computeAcquisitionInternal()

void BayesianPosteriorTargeted::computeAcquisitionInternal	(	std::vector< Real > &	acq,
		const std::vector< Real > &	gp_mean,
		const std::vector< Real > &	gp_std,
		const std::vector< std::vector< Real >> &	test_inputs,
		const std::vector< std::vector< Real >> &	train_inputs,
		const std::vector< Real > &	generic
	)		const

overrideprotectedvirtual

Implementation hook for derived classes (no size checks here).

Parameters

acq	The computed acquisition function values
gp_mean	The provided GP mean values
gp_std	The provided GP standard deviation values
test_inputs	All the input values under which the GP has to be tested
train_inputs	All the input values under which the GP has been trained
generic	A generic parameter (can be output values under which the GP has been trained or threshold parameter under the U-Function etc.)

Implements ParallelAcquisitionFunctionBase.

Definition at line 29 of file BayesianPosteriorTargeted.C.

 {
   Real psi = std::pow(test_inputs[0].size(), -0.85);
   for (unsigned int i = 0; i < test_inputs.size(); ++i)
   {
     acq[i] = std::exp(2.0 * psi * gp_mean[i]) * (std::exp(gp_std[i]) - 1.0);
     if (std::isinf(acq[i]))
       acq[i] = std::numeric_limits<Real>::max();
   }
 }

◆ computeCorrelation()

void ParallelAcquisitionFunctionBase::computeCorrelation	(	Real &	corr,
		const std::vector< Real > &	input1,
		const std::vector< Real > &	input2,
		const std::vector< Real > &	length_scales
	)

inherited

Compute the correlation between two inputs using the length scales.

Parameters

corr	The computed correlation value
input1	The first input
input2	The second input
length_scales	The length scales to compute the correlation between inputs

Definition at line 98 of file ParallelAcquisitionFunctionBase.C.

Referenced by ParallelAcquisitionFunctionBase::penalizeAcquisition().

 {
   if (input1.size() != input2.size() || input1.size() != length_scales.size())
     mooseError("computeCorrelation: input1, input2, and length_scales must be the same size.");
   corr = 0.0;
   for (unsigned int i = 0; i < input1.size(); ++i)
     corr -= Utility::pow<2>(input1[i] - input2[i]) / (2.0 * Utility::pow<2>(length_scales[i]));
   corr = 1.0 - std::exp(corr);
 }

◆ penalizeAcquisition()

void ParallelAcquisitionFunctionBase::penalizeAcquisition	(	std::vector< Real > &	modified_acq,
		std::vector< unsigned int > &	sorted_indices,
		const std::vector< Real > &	acq,
		const std::vector< Real > &	length_scales,
		const std::vector< std::vector< Real >> &	inputs
	)

inherited

Return the modified acquisition function values and sorted indices considering local penalization (inspired from Zhan et al.

2017)

Parameters

modified_acq	The modified acquisition function values
sorted_indices	The sorted indices modified acquisition function values
acq	The originally acquisition function values
length_scales	The length scales to compute the correlation between inputs
inputs	All the input values under which acquisition needs to be computed
penalize	Bool to indicate whether to compute the correlations or not

Definition at line 58 of file ParallelAcquisitionFunctionBase.C.

 {
   if (inputs.size() == 0)
     mooseError("penalizeAcquisition: 'inputs' must be non-empty.");
   if (modified_acq.size() != inputs.size() || sorted_indices.size() != inputs.size() ||
       acq.size() != inputs.size())
     mooseError(
         "penalizeAcquisition: modified_acq, sorted_indices, and acq must match inputs.size().");
 
   if (inputs.front().size() != inputs.front().size())
     mooseError("penalizeAcquisition: all input rows must have the same dimension.");
   if (length_scales.size() != inputs.front().size())
     mooseError("penalizeAcquisition: length_scales must match input dimension.");
 
   std::vector<Real> negate_acq = acq;
   std::transform(negate_acq.cbegin(), negate_acq.cend(), negate_acq.begin(), std::negate<double>());
   std::vector<size_t> ind;
   Moose::indirectSort(negate_acq.begin(), negate_acq.end(), ind);
   modified_acq[0] = -negate_acq[ind[0]];
   sorted_indices[0] = ind[0];
 
   Real correlation = 0.0;
   for (unsigned int i = 0; i < inputs.size() - 1; ++i)
   {
     for (unsigned int j = 0; j < inputs.size(); ++j)
     {
       computeCorrelation(correlation, inputs[j], inputs[ind[0]], length_scales);
       negate_acq[j] = negate_acq[j] * correlation;
     }
     Moose::indirectSort(negate_acq.begin(), negate_acq.end(), ind);
     modified_acq[i + 1] = -negate_acq[ind[0]];
     sorted_indices[i + 1] = ind[0];
   }
 }

◆ validParams()

InputParameters BayesianPosteriorTargeted::validParams ( )

static

Definition at line 16 of file BayesianPosteriorTargeted.C.

 {
   InputParameters params = ParallelAcquisitionFunctionBase::validParams();
   params.addClassDescription("Bayesian posterior targeted from El Gammal et al. 2023.");
   return params;
 }

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

stochastic_tools/include/acquisitions/BayesianPosteriorTargeted.h
stochastic_tools/src/acquisitions/BayesianPosteriorTargeted.C

Public Types

Public Member Functions

Static Public Member Functions

Public Attributes

Static Public Attributes

Protected Member Functions

Protected Attributes

Detailed Description

Constructor & Destructor Documentation

◆ BayesianPosteriorTargeted()

Member Function Documentation

◆ computeAcquisition()

◆ computeAcquisitionInternal()

◆ computeCorrelation()

◆ penalizeAcquisition()

◆ validParams()