Surrogate disciplines

A SurrogateDiscipline is built from the name of a MLRegressionAlgo and its options. These names and options are listed below.

Warning

Some algorithms can require the installation of GEMSEO with all its features and others can depend on plugins.

GaussianProcessRegressor

Module: gemseo.mlearning.regression.gpr

Required parameters
  • data : Dataset

    The learning dataset.

Optional parameters
  • alpha : float | ndarray, optional

    The nugget effect to regularize the model.

    By default it is set to 1e-10.

  • bounds : __Bounds | Mapping[str, __Bounds] | None, optional

    The lower and upper bounds of the parameter length scales when kernel is None. Either a unique lower-upper pair common to all the inputs or lower-upper pairs for some of them. When bounds is None or when an input has no pair, the lower bound is 0.01 and the upper bound is 100.

    By default it is set to None.

  • input_names : Iterable[str] | None, optional

    The names of the input variables. If None, consider all the input variables of the learning dataset.

    By default it is set to None.

  • kernel : Kernel | None, optional

    The kernel specifying the covariance model. If None, use a Matérn(2.5).

    By default it is set to None.

  • n_restarts_optimizer : int, optional

    The number of restarts of the optimizer.

    By default it is set to 10.

  • optimizer : str | Callable, optional

    The optimization algorithm to find the parameter length scales.

    By default it is set to fmin_l_bfgs_b.

  • output_names : Iterable[str] | None, optional

    The names of the output variables. If None, consider all the output variables of the learning dataset.

    By default it is set to None.

  • random_state : int | None, optional

    The seed used to initialize the centers. If None, the random number generator is the RandomState instance used by numpy.random.

    By default it is set to None.

  • transformer : TransformerType, optional

    The strategies to transform the variables. The values are instances of Transformer while the keys are the names of either the variables or the groups of variables, e.g. "inputs" or "outputs" in the case of the regression algorithms. If a group is specified, the Transformer will be applied to all the variables of this group. If IDENTITY, do not transform the variables.

    By default it is set to {}.

LinearRegressor

Module: gemseo.mlearning.regression.linreg

Required parameters
  • data : Dataset

    The learning dataset.

Optional parameters
  • fit_intercept : bool, optional

    Whether to fit the intercept.

    By default it is set to True.

  • input_names : Iterable[str] | None, optional

    The names of the input variables. If None, consider all the input variables of the learning dataset.

    By default it is set to None.

  • l2_penalty_ratio : float, optional

    The penalty ratio related to the l2 regularization. If 1, use the Ridge penalty. If 0, use the Lasso penalty. Between 0 and 1, use the ElasticNet penalty.

    By default it is set to 1.0.

  • output_names : Iterable[str] | None, optional

    The names of the output variables. If None, consider all the output variables of the learning dataset.

    By default it is set to None.

  • penalty_level : float, optional

    The penalty level greater or equal to 0. If 0, there is no penalty.

    By default it is set to 0.0.

  • transformer : TransformerType, optional

    The strategies to transform the variables. The values are instances of Transformer while the keys are the names of either the variables or the groups of variables, e.g. "inputs" or "outputs" in the case of the regression algorithms. If a group is specified, the Transformer will be applied to all the variables of this group. If IDENTITY, do not transform the variables.

    By default it is set to {}.

  • **parameters : float | int | str | bool | None

    The parameters of the machine learning algorithm.

MOERegressor

Module: gemseo.mlearning.regression.moe

Required parameters
  • data : Dataset

    The learning dataset.

Optional parameters
  • hard : bool, optional

    Whether clustering/classification should be hard or soft.

    By default it is set to True.

  • input_names : Iterable[str] | None, optional

    The names of the input variables. If None, consider all the input variables of the learning dataset.

    By default it is set to None.

  • output_names : Iterable[str] | None, optional

    The names of the output variables. If None, consider all the output variables of the learning dataset.

    By default it is set to None.

  • transformer : TransformerType, optional

    The strategies to transform the variables. The values are instances of Transformer while the keys are the names of either the variables or the groups of variables, e.g. "inputs" or "outputs" in the case of the regression algorithms. If a group is specified, the Transformer will be applied to all the variables of this group. If IDENTITY, do not transform the variables.

    By default it is set to {}.

PCERegressor

Module: gemseo.mlearning.regression.pce

Required parameters
  • data : Dataset | None

    The learning dataset required in the case of the least-squares regression or when discipline is None in the case of quadrature.

  • probability_space : ParameterSpace

    The set of random input variables defined by OTDistribution instances.

Optional parameters
  • cleaning_options : CleaningOptions | None, optional

    The options of the `CleaningStrategy`_. If None, use DEFAULT_CLEANING_OPTIONS.

    By default it is set to None.

  • degree : int, optional

    The polynomial degree of the PCE.

    By default it is set to 2.

  • discipline : MDODiscipline | None, optional

    The discipline to be sampled if use_quadrature is True and data is None.

    By default it is set to None.

  • hyperbolic_parameter : float, optional

    The \(q\)-quasi norm parameter of the `hyperbolic and anisotropic enumerate function`_, defined over the interval \(]0,1]\).

    By default it is set to 1.0.

  • input_names : Iterable[str] | None, optional

    The names of the input variables. If None, consider all the input variables of the learning dataset.

    By default it is set to None.

  • n_quadrature_points : int, optional

    The total number of quadrature points used by the quadrature strategy to compute the marginal number of points by input dimension when discipline is not None. If 0, use \((1+P)^d\) points, where \(d\) is the dimension of the input space and \(P\) is the polynomial degree of the PCE.

    By default it is set to 0.

  • output_names : Iterable[str] | None, optional

    The names of the output variables. If None, consider all the output variables of the learning dataset.

    By default it is set to None.

  • transformer : TransformerType, optional

    The strategies to transform the variables. The values are instances of Transformer while the keys are the names of either the variables or the groups of variables, e.g. "inputs" or "outputs" in the case of the regression algorithms. If a group is specified, the Transformer will be applied to all the variables of this group. If IDENTITY, do not transform the variables.

    By default it is set to {}.

  • use_cleaning : bool, optional

    Whether to use the `CleaningStrategy`_ algorithm. Otherwise, use a fixed truncation strategy (`FixedStrategy`_).

    By default it is set to False.

  • use_lars : bool, optional

    Whether to use the `LARS`_ algorithm in the case of the least-squares regression.

    By default it is set to False.

  • use_quadrature : bool, optional

    Whether to estimate the coefficients of the PCE by a quadrature rule; if so, use the quadrature points stored in data or sample discipline. otherwise, estimate the coefficients by least-squares regression.

    By default it is set to False.

PolynomialRegressor

Module: gemseo.mlearning.regression.polyreg

Required parameters
  • data : Dataset

    The learning dataset.

  • degree : int

    The polynomial degree.

Optional parameters
  • fit_intercept : bool, optional

    Whether to fit the intercept.

    By default it is set to True.

  • input_names : Iterable[str] | None, optional

    The names of the input variables. If None, consider all the input variables of the learning dataset.

    By default it is set to None.

  • l2_penalty_ratio : float, optional

    The penalty ratio related to the l2 regularization. If 1, the penalty is the Ridge penalty. If 0, this is the Lasso penalty. Between 0 and 1, the penalty is the ElasticNet penalty.

    By default it is set to 1.0.

  • output_names : Iterable[str] | None, optional

    The names of the output variables. If None, consider all the output variables of the learning dataset.

    By default it is set to None.

  • penalty_level : float, optional

    The penalty level greater or equal to 0. If 0, there is no penalty.

    By default it is set to 0.0.

  • transformer : TransformerType, optional

    The strategies to transform the variables. The values are instances of Transformer while the keys are the names of either the variables or the groups of variables, e.g. "inputs" or "outputs" in the case of the regression algorithms. If a group is specified, the Transformer will be applied to all the variables of this group. If IDENTITY, do not transform the variables.

    By default it is set to {}.

  • **parameters : float | int | str | bool | None

    The parameters of the machine learning algorithm.

RBFRegressor

Module: gemseo.mlearning.regression.rbf

Required parameters
  • data : Dataset

    The learning dataset.

Optional parameters
  • der_function : Callable[[ndarray], ndarray] | None, optional

    The derivative of the radial basis function, only to be provided if function is a callable and if the use of the model with its derivative is required. If None and if function is a callable, an error will be raised. If None and if function is a string, the class will look for its internal implementation and will raise an error if it is missing. The der_function shall take three arguments (input_data, norm_input_data, eps). For an RBF of the form function(\(r\)), der_function(\(x\), \(|x|\), \(\epsilon\)) shall return \(\epsilon^{-1} x/|x| f'(|x|/\epsilon)\).

    By default it is set to None.

  • epsilon : float | None, optional

    An adjustable constant for Gaussian or multiquadric functions. If None, use the average distance between input data.

    By default it is set to None.

  • function : str | Callable[[float, float], float], optional

    The radial basis function taking a radius \(r\) as input, representing a distance between two points. If it is a string, then it must be one of the following:

    • "multiquadric" for \(\sqrt{(r/\epsilon)^2 + 1}\),

    • "inverse" for \(1/\sqrt{(r/\epsilon)^2 + 1}\),

    • "gaussian" for \(\exp(-(r/\epsilon)^2)\),

    • "linear" for \(r\),

    • "cubic" for \(r^3\),

    • "quintic" for \(r^5\),

    • "thin_plate" for \(r^2\log(r)\).

    If it is a callable, then it must take the two arguments self and r as inputs, e.g. lambda self, r: sqrt((r/self.epsilon)**2 + 1) for the multiquadric function. The epsilon parameter will be available as self.epsilon. Other keyword arguments passed in will be available as well.

    By default it is set to multiquadric.

  • input_names : Iterable[str] | None, optional

    The names of the input variables. If None, consider all the input variables of the learning dataset.

    By default it is set to None.

  • norm : str | Callable[[ndarray, ndarray], float], optional

    The distance metric to be used, either a distance function name known by SciPy or a function that computes the distance between two points.

    By default it is set to euclidean.

  • output_names : Iterable[str] | None, optional

    The names of the output variables. If None, consider all the output variables of the learning dataset.

    By default it is set to None.

  • smooth : float, optional

    The degree of smoothness, 0 involving an interpolation of the learning points.

    By default it is set to 0.0.

  • transformer : TransformerType, optional

    The strategies to transform the variables. The values are instances of Transformer while the keys are the names of either the variables or the groups of variables, e.g. "inputs" or "outputs" in the case of the regression algorithms. If a group is specified, the Transformer will be applied to all the variables of this group. If IDENTITY, do not transform the variables.

    By default it is set to {}.

RandomForestRegressor

Module: gemseo.mlearning.regression.random_forest

Required parameters
  • data : Dataset

    The learning dataset.

Optional parameters
  • input_names : Iterable[str] | None, optional

    The names of the input variables. If None, consider all the input variables of the learning dataset.

    By default it is set to None.

  • n_estimators : int, optional

    The number of trees in the forest.

    By default it is set to 100.

  • output_names : Iterable[str] | None, optional

    The names of the output variables. If None, consider all the output variables of the learning dataset.

    By default it is set to None.

  • transformer : TransformerType, optional

    The strategies to transform the variables. The values are instances of Transformer while the keys are the names of either the variables or the groups of variables, e.g. "inputs" or "outputs" in the case of the regression algorithms. If a group is specified, the Transformer will be applied to all the variables of this group. If IDENTITY, do not transform the variables.

    By default it is set to {}.