The PolynomialRegression class constructs a polynomial regressor using ordinary least squares. More...

Inheritance diagram for PolynomialRegression:

Public Member Functions
	PolynomialRegression ()
	Constructor that uses defaultConfigOptions and does not build.

	PolynomialRegression (const ParameterList &options)
	Constructor that sets configOptions and does not build. More...

	PolynomialRegression (const std::string &param_list_yaml_filename)
	Constructor for the PolynomialRegression class that sets configOptions but does not build the surrogate. More...

	PolynomialRegression (const MatrixXd &samples, const MatrixXd &response, const ParameterList &options)
	Constructor sets configOptions and builds the Polynomial Regression surrogate. More...

	PolynomialRegression (const MatrixXd &samples, const MatrixXd &response, const std::string &param_list_yaml_filename)
	Constructor for the PolynomialRegression class that sets configOptions and builds the surrogate. More...

	~PolynomialRegression ()
	Default destructor.

void	compute_basis_matrix (const MatrixXd &samples, MatrixXd &basis_matrix) const
	Constructs a basis matrix for a set of samples according to the member variable basisIndices. More...

void	build (const MatrixXd &samples, const MatrixXd &response) override
	Build the polynomial surrogate using specified build data. More...

VectorXd	value (const MatrixXd &eval_points, const int qoi) override
	Evaluate the polynomial surrogate at a set of prediction points for a single QoI. More...

VectorXd	value (const MatrixXd &eval_points)
	Evaluate the polynomial surrogate at a set of prediction points for QoI index 0. More...

MatrixXd	gradient (const MatrixXd &eval_points, const int qoi) override
	Evaluate the gradient of the polynomial surrogate at a set of prediction points for a single QoI. More...

MatrixXd	gradient (const MatrixXd &eval_points)
	Evaluate the gradient of the polynomial surrogate at a set of prediction points for QoI index 0. More...

MatrixXd	hessian (const MatrixXd &eval_point, const int qoi) override
	Evaluate the Hessian of the polynomial surrogate at a single point for a single QoI. More...

MatrixXd	hessian (const MatrixXd &eval_point)
	Evaluate the Hessian of the polynomial surrogate at a single point for QoI index 0. More...

const MatrixXd &	get_polynomial_coeffs () const
	Get the polynomial surrogate's coefficients.

double	get_polynomial_intercept () const
	Get the polynomial surrogate's intercept/offset.

int	get_num_terms () const
	Get the number of terms in the polynomial surrogate.

void	set_polynomial_coeffs (const MatrixXd &coeffs)
	Set the polynomial surrogate's coefficients.

std::shared_ptr< Surrogate >	clone () const override
	clone derived Surrogate class for use in cross-validation

Public Member Functions inherited from Surrogate
	Surrogate ()
	Constructor that uses defaultConfigOptions and does not build.

	Surrogate (const ParameterList &param_list)
	Constructor that sets configOptions but does not build. More...

	Surrogate (const MatrixXd &samples, const MatrixXd &response, const ParameterList &param_list)
	Constructor for the Surrogate that sets configOptions and builds the surrogate (does nothing in the base class). More...

virtual	~Surrogate ()
	Default destructor.

VectorXd	value (const MatrixXd &eval_points)
	Evaluate the Surrogate at a set of prediction points for QoI index 0. More...

MatrixXd	gradient (const MatrixXd &eval_points)
	Evaluate the gradient of the Surrogate at a set of prediction points for QoI index 0. More...

MatrixXd	hessian (const MatrixXd &eval_point)
	Evaluate the Hessian of the Surrogate at a single point for QoI index 0. More...

void	variable_labels (const std::vector< std::string > &var_labels)
	Set the variable/feature names. More...

const std::vector< std::string > &	variable_labels () const
	Get the (possibly empty) variable/feature names. More...

void	response_labels (const std::vector< std::string > &resp_labels)
	Set the response/QoI names. More...

const std::vector< std::string > &	response_labels () const
	Get the (possibly empty) response/QoI names. More...

void	set_options (const ParameterList &options)
	Set the Surrogate's configOptions. More...

void	get_options (ParameterList &options)
	Get the Surrogate's configOptions. More...

void	print_options ()
	Print the Surrogate's configOptions.

VectorXd	evaluate_metrics (const StringArray &mnames, const MatrixXd &points, const MatrixXd &ref_values)
	Evalute metrics at specified points (within surrogates)

VectorXd	cross_validate (const MatrixXd &samples, const MatrixXd &response, const StringArray &mnames, const int num_folds=5, const int seed=20)
	Perform K-folds cross-validation (within surrogates)

template<typename DerivedSurr >
void	save (const DerivedSurr &surr_out, const std::string &outfile, const bool binary)
	Serialize a derived (i.e. non-base) surrogate model. More...

template<typename DerivedSurr >
void	load (const std::string &infile, const bool binary, DerivedSurr &surr_in)
	Load a derived (i.e. non-base) surrogate model. More...

Private Member Functions
void	default_options () override
	Construct and populate the defaultConfigOptions.

template<class Archive >
void	serialize (Archive &archive, const unsigned int version)
	Serializer for save/load.

Private Attributes
MatrixXi	basisIndices
	Matrix that specifies the powers of each variable for each term in the polynomial - (numVariables by numTerms).

std::shared_ptr< util::LinearSolverBase >	linearSolver
	Linear solver for the ordinary least squares problem.

int	numTerms
	Number of terms in the polynomial basis.

MatrixXd	polynomialCoeffs
	Vector of coefficients for the polynomial surrogate.

double	polynomialIntercept
	Offset/intercept term for the polynomial surrogate.

int	verbosity
	Verbosity level.

Friends
class	boost::serialization::access
	Allow serializers access to private class data.

Additional Inherited Members
Static Public Member Functions inherited from Surrogate
template<typename SurrHandle >
static void	save (const SurrHandle &surr_out, const std::string &outfile, const bool binary)
	serialize Surrogate to file (typically through shared_ptr<Surrogate>, but Derived& or Derived* okay too)

template<typename SurrHandle >
static void	load (const std::string &infile, const bool binary, SurrHandle &surr_in)
	serialize Surrogate from file (typically through shared_ptr<Surrogate>, but Derived& or Derived* okay too)

static std::shared_ptr< Surrogate >	load (const std::string &infile, const bool binary)
	serialize Surrogate from file through pointer to base class (must have been saved via same data type)

Public Attributes inherited from Surrogate
util::DataScaler	dataScaler
	DataScaler class for a Surrogate's build samples.

double	responseOffset = 0.
	Response offset.

double	responseScaleFactor = 1.
	Response scale factor.

Protected Attributes inherited from Surrogate
int	numSamples
	Number of samples in the Surrogate's build samples.

int	numVariables
	Number of features/variables in the Surrogate's build samples.

std::vector< std::string >	variableLabels
	Names of the variables/features; need not be populated.

int	numQOI
	Number of quantities of interest predicted by the surrogate. For scalar-valued surrogates numQOI = 1.

std::vector< std::string >	responseLabels
	Names of the responses/QoIs; need not be populated.

ParameterList	defaultConfigOptions
	Default Key/value options to configure the surrogate.

ParameterList	configOptions
	Key/value options to configure the surrogate - will override defaultConfigOptions.

Detailed Description

The PolynomialRegression class constructs a polynomial regressor using ordinary least squares.

Users may specify the max degree and p-norm for a hyperbolic cross scheme to specify the terms in the polynomial basis. A p-norm = 1 results in a total order specification of max degree.

The DataScaler class provides the option of scaling the basis matrix.

Constructor & Destructor Documentation

◆ PolynomialRegression() [1/4]

PolynomialRegression ( const ParameterList & options )

Constructor that sets configOptions and does not build.

Parameters

[in] options List that overrides entries in defaultConfigOptions.

References Surrogate::configOptions, PolynomialRegression::default_options(), and Surrogate::defaultConfigOptions.

◆ PolynomialRegression() [2/4]

PolynomialRegression ( const std::string & param_list_yaml_filename )

Constructor for the PolynomialRegression class that sets configOptions but does not build the surrogate.

Parameters

[in] param_list_yaml_filename A ParameterList file (relative to the location of the Dakota input file) that overrides entries in defaultConfigOptions.

References Surrogate::configOptions, PolynomialRegression::default_options(), and Surrogate::defaultConfigOptions.

◆ PolynomialRegression() [3/4]

PolynomialRegression	(	const MatrixXd &	samples,
		const MatrixXd &	response,
		const ParameterList &	options
	)

Constructor sets configOptions and builds the Polynomial Regression surrogate.

Parameters

[in]	samples	Matrix of data for surrogate construction - (num_samples by num_features)
[in]	response	Vector of targets for surrogate construction - (num_samples by num_qoi = 1; only 1 response is supported currently).
[in]	options	List that overrides entries in defaultConfigOptions

References PolynomialRegression::build(), Surrogate::configOptions, and PolynomialRegression::default_options().

◆ PolynomialRegression() [4/4]

PolynomialRegression	(	const MatrixXd &	samples,
		const MatrixXd &	response,
		const std::string &	param_list_yaml_filename
	)

Constructor for the PolynomialRegression class that sets configOptions and builds the surrogate.

Parameters

[in]	samples	Matrix of data for surrogate construction - (num_samples by num_features)
[in]	response	Vector of targets for surrogate construction - (num_samples by num_qoi = 1; only 1 response is supported currently).
[in]	param_list_yaml_filename	A ParameterList file (relative to the location of the Dakota input file) that overrides entries in defaultConfigOptions.

References PolynomialRegression::build(), Surrogate::configOptions, and PolynomialRegression::default_options().

Member Function Documentation

◆ compute_basis_matrix()

void compute_basis_matrix	(	const MatrixXd &	samples,
		MatrixXd &	basis_matrix
	)		const

Constructs a basis matrix for a set of samples according to the member variable basisIndices.

Parameters

[in]	samples	Matrix of sample points - (num_points by num_features).
[out]	basis_matrix	Matrix that contains polynomial basis function evaluations in its rows for each sample point - (num_points by numTerms), numTerms being the number of terms in the polynomial basis.

References PolynomialRegression::basisIndices, PolynomialRegression::numTerms, and Surrogate::numVariables.

Referenced by PolynomialRegression::build(), PolynomialRegression::gradient(), PolynomialRegression::hessian(), and PolynomialRegression::value().

◆ build()

void build	(	const MatrixXd &	samples,
		const MatrixXd &	response
	)

overridevirtual

Build the polynomial surrogate using specified build data.

Parameters

[in]	samples	Matrix of data for surrogate construction - (num_samples by num_features)
[in]	response	Vector of targets for surrogate construction - (num_samples by num_qoi = 1; only 1 response is supported currently).

Implements Surrogate.

References PolynomialRegression::basisIndices, PolynomialRegression::compute_basis_matrix(), dakota::surrogates::compute_hyperbolic_indices(), dakota::surrogates::compute_reduced_indices(), Surrogate::configOptions, Surrogate::dataScaler, Surrogate::defaultConfigOptions, PolynomialRegression::linearSolver, Surrogate::numQOI, Surrogate::numSamples, PolynomialRegression::numTerms, Surrogate::numVariables, PolynomialRegression::polynomialCoeffs, PolynomialRegression::polynomialIntercept, Surrogate::responseOffset, Surrogate::responseScaleFactor, DataScaler::scale_samples(), dakota::util::scaler_factory(), DataScaler::scaler_type(), dakota::util::solver_factory(), LinearSolverBase::solver_type(), and PolynomialRegression::verbosity.

Referenced by PolynomialRegression::PolynomialRegression().

◆ value() [1/2]

VectorXd value	(	const MatrixXd &	eval_points,
		const int	qoi
	)

overridevirtual

Evaluate the polynomial surrogate at a set of prediction points for a single QoI.

Parameters

[in]	eval_points	Matrix of prediction points - (num_pts by num_features).
[in]	qoi	Index for surrogate QoI.

Returns: Values of the polynomial surrogate at the prediction points - (num_pts)

Implements Surrogate.

References PolynomialRegression::compute_basis_matrix(), Surrogate::dataScaler, PolynomialRegression::polynomialCoeffs, PolynomialRegression::polynomialIntercept, Surrogate::responseOffset, Surrogate::responseScaleFactor, DataScaler::scale_samples(), and dakota::silence_unused_args().

Referenced by Dakota::compute_regression_coeffs().

◆ value() [2/2]

VectorXd value ( const MatrixXd & eval_points )

inline

Evaluate the polynomial surrogate at a set of prediction points for QoI index 0.

Parameters

[in] eval_points Matrix of prediction points - (num_pts by num_features).

Returns: Values of the polynomial surrogate at the prediction points - (num_pts)

References Surrogate::value().

◆ gradient() [1/2]

MatrixXd gradient	(	const MatrixXd &	eval_points,
		const int	qoi
	)

overridevirtual

Evaluate the gradient of the polynomial surrogate at a set of prediction points for a single QoI.

Parameters

[in]	eval_points	Coordinates of the prediction points - (num_pts by num_features).
[in]	qoi	Index of response/QOI for which to compute derivatives.

Returns: Matrix of gradient vectors at the prediction points - (num_pts by num_features).

Reimplemented from Surrogate.

References PolynomialRegression::basisIndices, PolynomialRegression::compute_basis_matrix(), Surrogate::dataScaler, PolynomialRegression::numTerms, Surrogate::numVariables, PolynomialRegression::polynomialCoeffs, Surrogate::responseScaleFactor, DataScaler::scale_samples(), and dakota::silence_unused_args().

◆ gradient() [2/2]

MatrixXd gradient ( const MatrixXd & eval_points )

inline

Evaluate the gradient of the polynomial surrogate at a set of prediction points for QoI index 0.

Parameters

[in] eval_points Coordinates of the prediction points - (num_pts by num_features).

Returns: Matrix of gradient vectors at the prediction points - (num_pts by num_features).

References Surrogate::gradient().

◆ hessian() [1/2]

MatrixXd hessian	(	const MatrixXd &	eval_point,
		const int	qoi
	)

overridevirtual

Evaluate the Hessian of the polynomial surrogate at a single point for a single QoI.

Parameters

[in]	eval_point	Coordinates of the prediction point - (1 by num_features).
[in]	qoi	Index of response/QOI for which to compute derivatives.

Returns: Hessian matrix at the prediction point - (num_features by num_features).

Reimplemented from Surrogate.

References PolynomialRegression::basisIndices, PolynomialRegression::compute_basis_matrix(), Surrogate::dataScaler, PolynomialRegression::numTerms, Surrogate::numVariables, PolynomialRegression::polynomialCoeffs, Surrogate::responseScaleFactor, DataScaler::scale_samples(), and dakota::silence_unused_args().

◆ hessian() [2/2]

MatrixXd hessian ( const MatrixXd & eval_point )

inline

Evaluate the Hessian of the polynomial surrogate at a single point for QoI index 0.

Parameters

[in] eval_point Coordinates of the prediction point - (1 by num_features).

Returns: Hessian matrix at the prediction point - (num_features by num_features).

References Surrogate::hessian().

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

SurrogatesPolynomialRegression.hpp
SurrogatesPolynomialRegression.cpp

Public Member Functions

Private Member Functions

Private Attributes

Friends

Additional Inherited Members

Detailed Description

Constructor & Destructor Documentation

◆ PolynomialRegression() [1/4]

◆ PolynomialRegression() [2/4]

◆ PolynomialRegression() [3/4]

◆ PolynomialRegression() [4/4]

Member Function Documentation

◆ compute_basis_matrix()

◆ build()

◆ value() [1/2]

◆ value() [2/2]

◆ gradient() [1/2]

◆ gradient() [2/2]

◆ hessian() [1/2]

◆ hessian() [2/2]