gemseo.mlearning.regression.algos.ot_gpr_settings module

Settings of the Gaussian process regressor from OpenTURNS.

class CovarianceModel(value)

Bases: StrEnum

The name of a covariance model.

ABSOLUTE_EXPONENTIAL = 'AbsoluteExponential'

The absolute exponential kernel.

EXPONENTIAL = 'Exponential'

The exponential kernel.

MATERN12 = 'Matern12'

The Matérn 1/2 kernel.

MATERN32 = 'Matern32'

The Matérn 3/2 kernel.

MATERN52 = 'Matern52'

The Matérn 5/2 kernel.

SQUARED_EXPONENTIAL = 'SquaredExponential'

The squared exponential kernel.

class DOEAlgorithmName(value)

Bases: StrEnum

The name of a DOE algorithm.

CustomDOE = 'CustomDOE'

DiagonalDOE = 'DiagonalDOE'

Halton = 'Halton'

LHS = 'LHS'

MC = 'MC'

MorrisDOE = 'MorrisDOE'

OATDOE = 'OATDOE'

OT_AXIAL = 'OT_AXIAL'

OT_COMPOSITE = 'OT_COMPOSITE'

OT_FACTORIAL = 'OT_FACTORIAL'

OT_FAURE = 'OT_FAURE'

OT_FULLFACT = 'OT_FULLFACT'

OT_HALTON = 'OT_HALTON'

OT_HASELGROVE = 'OT_HASELGROVE'

OT_LHS = 'OT_LHS'

OT_LHSC = 'OT_LHSC'

OT_MONTE_CARLO = 'OT_MONTE_CARLO'

OT_OPT_LHS = 'OT_OPT_LHS'

OT_RANDOM = 'OT_RANDOM'

OT_REVERSE_HALTON = 'OT_REVERSE_HALTON'

OT_SOBOL = 'OT_SOBOL'

OT_SOBOL_INDICES = 'OT_SOBOL_INDICES'

PYDOE_BBDESIGN = 'PYDOE_BBDESIGN'

PYDOE_CCDESIGN = 'PYDOE_CCDESIGN'

PYDOE_FF2N = 'PYDOE_FF2N'

PYDOE_FULLFACT = 'PYDOE_FULLFACT'

PYDOE_LHS = 'PYDOE_LHS'

PYDOE_PBDESIGN = 'PYDOE_PBDESIGN'

PoissonDisk = 'PoissonDisk'

Sobol = 'Sobol'

Settings OTGaussianProcessRegressor_Settings(*, transformer=None, parameters=None, input_names=(), output_names=(), use_hmat=None, trend=Trend.CONSTANT, optimizer=class=TNC class=OptimizationAlgorithmImplementation problem=class=OptimizationProblem implementation=class=OptimizationProblemImplementation objective=class=Function name=Unnamed implementation=class=FunctionImplementation name=Unnamed description=[] evaluationImplementation=class=NoEvaluation name=Unnamed gradientImplementation=class=NoGradient name=Unnamed hessianImplementation=class=NoHessian name=Unnamed equality constraint=none inequality constraint=none bounds=none minimization=true dimension=0 startingPoint=class=Point name=Unnamed dimension=0 values=[] maximumIterationNumber=100 maximumCallsNumber=1000 maximumAbsoluteError=1e-05 maximumRelativeError=1e-05 maximumResidualError=1e-05 maximumConstraintError=1e-05 scale=class=Point name=Unnamed dimension=0 values=[] offset=class=Point name=Unnamed dimension=0 values=[] maxCGit=50 eta=0.25 stepmx=10 accuracy=0.0001 fmin=1 rescale=1.3, optimization_space=None, covariance_model=CovarianceModel.MATERN52, multi_start_n_samples=10, multi_start_algo_name=DOEAlgorithmName.OT_OPT_LHS, multi_start_algo_settings=None)

Bases: BaseRegressorSettings

The settings of the Gaussian process regressor from OpenTURNS.

Create a new model by parsing and validating input data from keyword arguments.

Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.

self is explicitly positional-only to allow self as a field name.

Parameters:

• transformer (Mapping[str, Any])

• parameters (Mapping[str, Any])

• input_names (Sequence[str]) --

  By default it is set to ().

• output_names (Sequence[str]) --

  By default it is set to ().

• use_hmat (bool | None)

• trend (Trend) --

  By default it is set to "constant".

• optimizer (OptimizationAlgorithmImplementation) --

  By default it is set to class=TNC class=OptimizationAlgorithmImplementation problem=class=OptimizationProblem implementation=class=OptimizationProblemImplementation objective=class=Function name=Unnamed implementation=class=FunctionImplementation name=Unnamed description=[] evaluationImplementation=class=NoEvaluation name=Unnamed gradientImplementation=class=NoGradient name=Unnamed hessianImplementation=class=NoHessian name=Unnamed equality constraint=none inequality constraint=none bounds=none minimization=true dimension=0 startingPoint=class=Point name=Unnamed dimension=0 values=[] maximumIterationNumber=100 maximumCallsNumber=1000 maximumAbsoluteError=1e-05 maximumRelativeError=1e-05 maximumResidualError=1e-05 maximumConstraintError=1e-05 scale=class=Point name=Unnamed dimension=0 values=[] offset=class=Point name=Unnamed dimension=0 values=[] maxCGit=50 eta=0.25 stepmx=10 accuracy=0.0001 fmin=1 rescale=1.3.

• optimization_space (DesignSpace | None)

• covariance_model (Sequence[CovarianceModelImplementation | type[CovarianceModelImplementation] | CovarianceModel] | CovarianceModelImplementation | type[CovarianceModelImplementation] | CovarianceModel) --

  By default it is set to "Matern52".

• multi_start_n_samples (Annotated[int, Ge(ge=0)]) --

  By default it is set to 10.

• multi_start_algo_name (DOEAlgorithmName) --

  By default it is set to "OT_OPT_LHS".

• multi_start_algo_settings (Mapping[str, Any])

Return type:

None

covariance_model: Sequence[CovarianceModelType] | CovarianceModelType = CovarianceModel.MATERN52

The covariance model of the Gaussian process.

Either an OpenTURNS covariance model class, an OpenTURNS covariance model class instance, a name of covariance model, or a list of OpenTURNS covariance model classes, OpenTURNS class instances and covariance model names, whose size is equal to the output dimension.

multi_start_algo_name: DOEAlgorithmName = DOEAlgorithmName.OT_OPT_LHS

The name of the DOE algorithm.

This DOE is used for the multi-start optimization of the covariance model parameters.

multi_start_algo_settings: StrKeyMapping [Optional]

The settings of the DOE algorithm.

multi_start_n_samples: NonNegativeInt = 10

The number of starting points of the multi-start optimizer.

This optimizer is used for the covariance model parameters.

Constraints:

• ge = 0

optimization_space: DesignSpace | None = None

The covariance model parameter space.

The size of a variable must take into account the size of the output space. If None, the algorithm will use a design space with bounds defined by OpenTURNS.

optimizer: OptimizationAlgorithmImplementation = class=TNC class=OptimizationAlgorithmImplementation problem=class=OptimizationProblem implementation=class=OptimizationProblemImplementation objective=class=Function name=Unnamed implementation=class=FunctionImplementation name=Unnamed description=[] evaluationImplementation=class=NoEvaluation name=Unnamed gradientImplementation=class=NoGradient name=Unnamed hessianImplementation=class=NoHessian name=Unnamed equality constraint=none inequality constraint=none bounds=none minimization=true dimension=0 startingPoint=class=Point name=Unnamed dimension=0 values=[] maximumIterationNumber=100 maximumCallsNumber=1000 maximumAbsoluteError=1e-05 maximumRelativeError=1e-05 maximumResidualError=1e-05 maximumConstraintError=1e-05 scale=class=Point name=Unnamed dimension=0 values=[] offset=class=Point name=Unnamed dimension=0 values=[] maxCGit=50 eta=0.25 stepmx=10 accuracy=0.0001 fmin=1 rescale=1.3

The solver used to optimize the covariance model parameters.

trend: Trend = Trend.CONSTANT

The name of the trend.

use_hmat: bool | None = None

Whether to use the HMAT or LAPACK as linear algebra method.

If None, use HMAT when the learning size is greater than MAX_SIZE_FOR_LAPACK.

class Trend(value)

Bases: StrEnum

The name of a trend.

CONSTANT = 'constant'

LINEAR = 'linear'

QUADRATIC = 'quadratic'

TNC: Final[TNC] = class=TNC class=OptimizationAlgorithmImplementation problem=class=OptimizationProblem implementation=class=OptimizationProblemImplementation objective=class=Function name=Unnamed implementation=class=FunctionImplementation name=Unnamed description=[] evaluationImplementation=class=NoEvaluation name=Unnamed gradientImplementation=class=NoGradient name=Unnamed hessianImplementation=class=NoHessian name=Unnamed equality constraint=none inequality constraint=none bounds=none minimization=true dimension=0 startingPoint=class=Point name=Unnamed dimension=0 values=[] maximumIterationNumber=100 maximumCallsNumber=1000 maximumAbsoluteError=1e-05 maximumRelativeError=1e-05 maximumResidualError=1e-05 maximumConstraintError=1e-05 scale=class=Point name=Unnamed dimension=0 values=[] offset=class=Point name=Unnamed dimension=0 values=[] maxCGit=50 eta=0.25 stepmx=10 accuracy=0.0001 fmin=1 rescale=1.3

The TNC algorithm.