DOE algorithms

Warning

Some capabilities may require the installation of GEMSEO with all its features and some others may depend on plugins.

Note

All the features of the wrapped libraries may not be exposed through GEMSEO.

See also

You can find more information about this family of algorithms in the user guide.

CustomDOE

Module: gemseo.algos.doe.lib_custom

This samples are provided either as a file in text or csv format or as a sequence of sequences of numbers.

Required parameters
  • doe_file : str | Path | None

    The path to the file containing the input samples. If None, use samples.

  • samples : RealArray | dict[str, RealArray] | list[dict[str, RealArray]] | None

    The input samples. They must be at least a 2D-array, a dictionary of 2D-arrays or a list of dictionaries of 1D-arrays. If None, use doe_file.

Optional parameters
  • callbacks : Iterable[CallbackType], optional

    The functions to be evaluated after each call to OptimizationProblem.evaluate_functions(); to be called as callback(index, (output, jacobian)).

    By default it is set to ().

  • comments : str | Sequence[str] | None, optional

    The characters or list of characters used to indicate the start of a comment. None implies no comments.

    By default it is set to #.

  • delimiter : str | None, optional

    The character used to separate values. If None, use whitespace.

    By default it is set to ,.

  • eval_jac : bool, optional

    Whether to evaluate the jacobian.

    By default it is set to False.

  • max_time : float, optional

    The maximum runtime in seconds, disabled if 0.

    By default it is set to 0.

  • n_processes : int, optional

    The maximum simultaneous number of processes used to parallelize the execution.

    By default it is set to 1.

  • skiprows : int, optional

    The number of first lines to skip.

    By default it is set to 0.

  • wait_time_between_samples : float, optional

    The waiting time between two samples.

    By default it is set to 0.0.

DiagonalDOE

Module: gemseo.algos.doe.lib_scalable

Diagonal design of experiments

Required parameters
  • n_samples : int

    The number of samples. The number of samples must be greater than or equal to 2.

Optional parameters
  • callbacks : Iterable[CallbackType], optional

    The functions to be evaluated after each call to OptimizationProblem.evaluate_functions(); to be called as callback(index, (output, jacobian)).

    By default it is set to ().

  • eval_jac : bool, optional

    Whether to evaluate the Jacobian.

    By default it is set to False.

  • max_time : float, optional

    The maximum runtime in seconds. If 0, no maximum runtime is set.

    By default it is set to 0.

  • n_processes : int, optional

    The maximum simultaneous number of processes used to parallelize the execution.

    By default it is set to 1.

  • reverse : Container[str] | None, optional

    The dimensions or variables to sample from their upper bounds to their lower bounds. If None, every dimension will be sampled from its lower bound to its upper bound.

    By default it is set to None.

  • wait_time_between_samples : float, optional

    The waiting time between two samples.

    By default it is set to 0.0.

Halton

Module: gemseo.algos.doe.lib_scipy

Halton sequence

Required parameters
  • n_samples : int

    The number of samples.

  • optimization : Optimizer

    The name of an optimization scheme to improve the quality of the DOE. If None, use the DOE as is. New in SciPy 1.10.0.

Optional parameters
  • callbacks : Iterable[CallbackType], optional

    The functions to be evaluated after each call to OptimizationProblem.evaluate_functions(); to be called as callback(index, (output, jacobian)).

    By default it is set to ().

  • eval_jac : bool, optional

    Whether to evaluate the jacobian.

    By default it is set to False.

  • max_time : float, optional

    The maximum runtime in seconds, disabled if 0.

    By default it is set to 0.

  • n_processes : int, optional

    The maximum simultaneous number of processes used to parallelize the execution.

    By default it is set to 1.

  • scramble : bool, optional

    Whether to use scrambling (Owen type). Only available with SciPy >= 1.10.0.

    By default it is set to True.

  • seed : int | None, optional

    The seed used for reproducibility reasons. If None, use seed.

    By default it is set to None.

  • wait_time_between_samples : float, optional

    The waiting time between two samples.

    By default it is set to 0.0.

LHS

Module: gemseo.algos.doe.lib_scipy

Latin hypercube sampling (LHS)

Required parameters
  • n_samples : int

    The number of samples.

  • optimization : Optimizer

    The name of an optimization scheme to improve the quality of the DOE. If None, use the DOE as is. New in SciPy 1.10.0.

Optional parameters
  • callbacks : Iterable[CallbackType], optional

    The functions to be evaluated after each call to OptimizationProblem.evaluate_functions(); to be called as callback(index, (output, jacobian)).

    By default it is set to ().

  • centered : bool, optional

    Whether to center the samples within the cells of a multi-dimensional grid. If SciPy >= 1.10.0, use scramble instead.

    By default it is set to False.

  • eval_jac : bool, optional

    Whether to evaluate the jacobian.

    By default it is set to False.

  • max_time : float, optional

    The maximum runtime in seconds, disabled if 0.

    By default it is set to 0.

  • n_processes : int, optional

    The maximum simultaneous number of processes used to parallelize the execution.

    By default it is set to 1.

  • scramble : bool, optional

    Whether to use scrambling (Owen type). Only available with SciPy >= 1.10.0.

    By default it is set to True.

  • seed : int | None, optional

    The seed used for reproducibility reasons. If None, use seed.

    By default it is set to None.

  • strength : Literal[1, 2], optional

    The strength of the LHS.

    By default it is set to 1.

  • wait_time_between_samples : float, optional

    The waiting time between two samples.

    By default it is set to 0.0.

MC

Module: gemseo.algos.doe.lib_scipy

Monte Carlo sampling

Required parameters
  • n_samples : int

    The number of samples.

Optional parameters
  • callbacks : Iterable[CallbackType], optional

    The functions to be evaluated after each call to OptimizationProblem.evaluate_functions(); to be called as callback(index, (output, jacobian)).

    By default it is set to ().

  • eval_jac : bool, optional

    Whether to evaluate the jacobian.

    By default it is set to False.

  • max_time : float, optional

    The maximum runtime in seconds, disabled if 0.

    By default it is set to 0.

  • n_processes : int, optional

    The maximum simultaneous number of processes used to parallelize the execution.

    By default it is set to 1.

  • seed : int | None, optional

    The seed used for reproducibility reasons. If None, use seed.

    By default it is set to None.

  • wait_time_between_samples : float, optional

    The waiting time between two samples.

    By default it is set to 0.0.

OT_AXIAL

Module: gemseo.algos.doe.lib_openturns

Axial design

More details about the algorithm and its options on http://openturns.github.io/openturns/latest/user_manual/_generated/openturns.Axial.html.

Required parameters
  • levels : float | Sequence[float] | None

    1) In the case of axial, composite and factorial DOEs, the positions of the levels relative to the center; the levels will be equispaced and symmetrical relative to the center; e.g. [0.2, 0.8] in dimension 1 will generate the samples [0.15, 0.6, 0.75, 0.8, 0.95, 1] for an axial DOE; the values must be in \(]0,1]\). 2) In the case of a full-factorial DOE, the number of levels per input direction; if scalar, this value is applied to each input direction.

  • n_samples : int | None

    The maximum number of samples required by the user; for axial, composite and factorial DOEs, a minimum number of samples is required and depends on the dimension of the space to sample; if None in the case of for axial, composite, factorial and full-factorial DOEs the effective number of samples is computed from this dimension and the number of levels.

Optional parameters
  • callbacks : Iterable[CallbackType], optional

    The functions to be evaluated after each call to OptimizationProblem.evaluate_functions(); to be called as callback(index, (output, jacobian)).

    By default it is set to ().

  • centers : Sequence[float] | float, optional

    The center of the unit hypercube that the axial, composite or factorial DOE algorithm will sample; if scalar, this value is applied to each direction of the hypercube; the values must be in \(]0,1[\).

    By default it is set to 0.5.

  • eval_jac : bool, optional

    Whether to evaluate the jacobian.

    By default it is set to False.

  • max_time : float, optional

    The maximum runtime in seconds, disabled if 0.

    By default it is set to 0.

  • n_processes : int, optional

    The maximum simultaneous number of processes used to parallelize the execution.

    By default it is set to 1.

  • seed : int | None, optional

    The seed used for reproducibility reasons. If None, use seed.

    By default it is set to None.

  • wait_time_between_samples : float, optional

    The waiting time between two samples.

    By default it is set to 0.0.

OT_COMPOSITE

Module: gemseo.algos.doe.lib_openturns

Composite design

More details about the algorithm and its options on http://openturns.github.io/openturns/latest/user_manual/_generated/openturns.Composite.html.

Required parameters
  • levels : float | Sequence[float] | None

    1) In the case of axial, composite and factorial DOEs, the positions of the levels relative to the center; the levels will be equispaced and symmetrical relative to the center; e.g. [0.2, 0.8] in dimension 1 will generate the samples [0.15, 0.6, 0.75, 0.8, 0.95, 1] for an axial DOE; the values must be in \(]0,1]\). 2) In the case of a full-factorial DOE, the number of levels per input direction; if scalar, this value is applied to each input direction.

  • n_samples : int | None

    The maximum number of samples required by the user; for axial, composite and factorial DOEs, a minimum number of samples is required and depends on the dimension of the space to sample; if None in the case of for axial, composite, factorial and full-factorial DOEs the effective number of samples is computed from this dimension and the number of levels.

Optional parameters
  • callbacks : Iterable[CallbackType], optional

    The functions to be evaluated after each call to OptimizationProblem.evaluate_functions(); to be called as callback(index, (output, jacobian)).

    By default it is set to ().

  • centers : Sequence[float] | float, optional

    The center of the unit hypercube that the axial, composite or factorial DOE algorithm will sample; if scalar, this value is applied to each direction of the hypercube; the values must be in \(]0,1[\).

    By default it is set to 0.5.

  • eval_jac : bool, optional

    Whether to evaluate the jacobian.

    By default it is set to False.

  • max_time : float, optional

    The maximum runtime in seconds, disabled if 0.

    By default it is set to 0.

  • n_processes : int, optional

    The maximum simultaneous number of processes used to parallelize the execution.

    By default it is set to 1.

  • seed : int | None, optional

    The seed used for reproducibility reasons. If None, use seed.

    By default it is set to None.

  • wait_time_between_samples : float, optional

    The waiting time between two samples.

    By default it is set to 0.0.

OT_FACTORIAL

Module: gemseo.algos.doe.lib_openturns

Factorial design

More details about the algorithm and its options on http://openturns.github.io/openturns/latest/user_manual/_generated/openturns.Factorial.html.

Required parameters
  • levels : float | Sequence[float] | None

    1) In the case of axial, composite and factorial DOEs, the positions of the levels relative to the center; the levels will be equispaced and symmetrical relative to the center; e.g. [0.2, 0.8] in dimension 1 will generate the samples [0.15, 0.6, 0.75, 0.8, 0.95, 1] for an axial DOE; the values must be in \(]0,1]\). 2) In the case of a full-factorial DOE, the number of levels per input direction; if scalar, this value is applied to each input direction.

  • n_samples : int | None

    The maximum number of samples required by the user; for axial, composite and factorial DOEs, a minimum number of samples is required and depends on the dimension of the space to sample; if None in the case of for axial, composite, factorial and full-factorial DOEs the effective number of samples is computed from this dimension and the number of levels.

Optional parameters
  • callbacks : Iterable[CallbackType], optional

    The functions to be evaluated after each call to OptimizationProblem.evaluate_functions(); to be called as callback(index, (output, jacobian)).

    By default it is set to ().

  • centers : Sequence[float] | float, optional

    The center of the unit hypercube that the axial, composite or factorial DOE algorithm will sample; if scalar, this value is applied to each direction of the hypercube; the values must be in \(]0,1[\).

    By default it is set to 0.5.

  • eval_jac : bool, optional

    Whether to evaluate the jacobian.

    By default it is set to False.

  • max_time : float, optional

    The maximum runtime in seconds, disabled if 0.

    By default it is set to 0.

  • n_processes : int, optional

    The maximum simultaneous number of processes used to parallelize the execution.

    By default it is set to 1.

  • seed : int | None, optional

    The seed used for reproducibility reasons. If None, use seed.

    By default it is set to None.

  • wait_time_between_samples : float, optional

    The waiting time between two samples.

    By default it is set to 0.0.

OT_FAURE

Module: gemseo.algos.doe.lib_openturns

Faure sequence

More details about the algorithm and its options on http://openturns.github.io/openturns/latest/user_manual/_generated/openturns.FaureSequence.html.

Required parameters
  • n_samples : int | None

    The maximum number of samples required by the user; for axial, composite and factorial DOEs, a minimum number of samples is required and depends on the dimension of the space to sample; if None in the case of for axial, composite, factorial and full-factorial DOEs the effective number of samples is computed from this dimension and the number of levels.

Optional parameters
  • callbacks : Iterable[CallbackType], optional

    The functions to be evaluated after each call to OptimizationProblem.evaluate_functions(); to be called as callback(index, (output, jacobian)).

    By default it is set to ().

  • eval_jac : bool, optional

    Whether to evaluate the jacobian.

    By default it is set to False.

  • max_time : float, optional

    The maximum runtime in seconds, disabled if 0.

    By default it is set to 0.

  • n_processes : int, optional

    The maximum simultaneous number of processes used to parallelize the execution.

    By default it is set to 1.

  • seed : int | None, optional

    The seed used for reproducibility reasons. If None, use seed.

    By default it is set to None.

  • wait_time_between_samples : float, optional

    The waiting time between two samples.

    By default it is set to 0.0.

OT_FULLFACT

Module: gemseo.algos.doe.lib_openturns

Full factorial design

More details about the algorithm and its options on http://openturns.github.io/openturns/latest/user_manual/_generated/openturns.Box.html.

Required parameters
  • levels : float | Sequence[float] | None

    1) In the case of axial, composite and factorial DOEs, the positions of the levels relative to the center; the levels will be equispaced and symmetrical relative to the center; e.g. [0.2, 0.8] in dimension 1 will generate the samples [0.15, 0.6, 0.75, 0.8, 0.95, 1] for an axial DOE; the values must be in \(]0,1]\). 2) In the case of a full-factorial DOE, the number of levels per input direction; if scalar, this value is applied to each input direction.

  • n_samples : int | None

    The maximum number of samples required by the user; for axial, composite and factorial DOEs, a minimum number of samples is required and depends on the dimension of the space to sample; if None in the case of for axial, composite, factorial and full-factorial DOEs the effective number of samples is computed from this dimension and the number of levels.

Optional parameters
  • callbacks : Iterable[CallbackType], optional

    The functions to be evaluated after each call to OptimizationProblem.evaluate_functions(); to be called as callback(index, (output, jacobian)).

    By default it is set to ().

  • eval_jac : bool, optional

    Whether to evaluate the jacobian.

    By default it is set to False.

  • max_time : float, optional

    The maximum runtime in seconds, disabled if 0.

    By default it is set to 0.

  • n_processes : int, optional

    The maximum simultaneous number of processes used to parallelize the execution.

    By default it is set to 1.

  • seed : int | None, optional

    The seed used for reproducibility reasons. If None, use seed.

    By default it is set to None.

  • wait_time_between_samples : float, optional

    The waiting time between two samples.

    By default it is set to 0.0.

OT_HALTON

Module: gemseo.algos.doe.lib_openturns

Halton sequence

More details about the algorithm and its options on http://openturns.github.io/openturns/latest/user_manual/_generated/openturns.HaltonSequence.html.

Required parameters
  • n_samples : int | None

    The maximum number of samples required by the user; for axial, composite and factorial DOEs, a minimum number of samples is required and depends on the dimension of the space to sample; if None in the case of for axial, composite, factorial and full-factorial DOEs the effective number of samples is computed from this dimension and the number of levels.

Optional parameters
  • callbacks : Iterable[CallbackType], optional

    The functions to be evaluated after each call to OptimizationProblem.evaluate_functions(); to be called as callback(index, (output, jacobian)).

    By default it is set to ().

  • eval_jac : bool, optional

    Whether to evaluate the jacobian.

    By default it is set to False.

  • max_time : float, optional

    The maximum runtime in seconds, disabled if 0.

    By default it is set to 0.

  • n_processes : int, optional

    The maximum simultaneous number of processes used to parallelize the execution.

    By default it is set to 1.

  • seed : int | None, optional

    The seed used for reproducibility reasons. If None, use seed.

    By default it is set to None.

  • wait_time_between_samples : float, optional

    The waiting time between two samples.

    By default it is set to 0.0.

OT_HASELGROVE

Module: gemseo.algos.doe.lib_openturns

Haselgrove sequence

More details about the algorithm and its options on http://openturns.github.io/openturns/latest/user_manual/_generated/openturns.HaselgroveSequence.html.

Required parameters
  • n_samples : int | None

    The maximum number of samples required by the user; for axial, composite and factorial DOEs, a minimum number of samples is required and depends on the dimension of the space to sample; if None in the case of for axial, composite, factorial and full-factorial DOEs the effective number of samples is computed from this dimension and the number of levels.

Optional parameters
  • callbacks : Iterable[CallbackType], optional

    The functions to be evaluated after each call to OptimizationProblem.evaluate_functions(); to be called as callback(index, (output, jacobian)).

    By default it is set to ().

  • eval_jac : bool, optional

    Whether to evaluate the jacobian.

    By default it is set to False.

  • max_time : float, optional

    The maximum runtime in seconds, disabled if 0.

    By default it is set to 0.

  • n_processes : int, optional

    The maximum simultaneous number of processes used to parallelize the execution.

    By default it is set to 1.

  • seed : int | None, optional

    The seed used for reproducibility reasons. If None, use seed.

    By default it is set to None.

  • wait_time_between_samples : float, optional

    The waiting time between two samples.

    By default it is set to 0.0.

OT_LHS

Module: gemseo.algos.doe.lib_openturns

Latin Hypercube Sampling

More details about the algorithm and its options on http://openturns.github.io/openturns/latest/user_manual/_generated/openturns.LHS.html.

Required parameters
  • n_samples : int | None

    The maximum number of samples required by the user; for axial, composite and factorial DOEs, a minimum number of samples is required and depends on the dimension of the space to sample; if None in the case of for axial, composite, factorial and full-factorial DOEs the effective number of samples is computed from this dimension and the number of levels.

Optional parameters
  • callbacks : Iterable[CallbackType], optional

    The functions to be evaluated after each call to OptimizationProblem.evaluate_functions(); to be called as callback(index, (output, jacobian)).

    By default it is set to ().

  • eval_jac : bool, optional

    Whether to evaluate the jacobian.

    By default it is set to False.

  • max_time : float, optional

    The maximum runtime in seconds, disabled if 0.

    By default it is set to 0.

  • n_processes : int, optional

    The maximum simultaneous number of processes used to parallelize the execution.

    By default it is set to 1.

  • seed : int | None, optional

    The seed used for reproducibility reasons. If None, use seed.

    By default it is set to None.

  • wait_time_between_samples : float, optional

    The waiting time between two samples.

    By default it is set to 0.0.

OT_LHSC

Module: gemseo.algos.doe.lib_openturns

Centered Latin Hypercube Sampling

More details about the algorithm and its options on http://openturns.github.io/openturns/latest/user_manual/_generated/openturns.LHS.html.

Required parameters
  • n_samples : int | None

    The maximum number of samples required by the user; for axial, composite and factorial DOEs, a minimum number of samples is required and depends on the dimension of the space to sample; if None in the case of for axial, composite, factorial and full-factorial DOEs the effective number of samples is computed from this dimension and the number of levels.

Optional parameters
  • callbacks : Iterable[CallbackType], optional

    The functions to be evaluated after each call to OptimizationProblem.evaluate_functions(); to be called as callback(index, (output, jacobian)).

    By default it is set to ().

  • eval_jac : bool, optional

    Whether to evaluate the jacobian.

    By default it is set to False.

  • max_time : float, optional

    The maximum runtime in seconds, disabled if 0.

    By default it is set to 0.

  • n_processes : int, optional

    The maximum simultaneous number of processes used to parallelize the execution.

    By default it is set to 1.

  • seed : int | None, optional

    The seed used for reproducibility reasons. If None, use seed.

    By default it is set to None.

  • wait_time_between_samples : float, optional

    The waiting time between two samples.

    By default it is set to 0.0.

OT_MONTE_CARLO

Module: gemseo.algos.doe.lib_openturns

Monte Carlo sequence

More details about the algorithm and its options on http://openturns.github.io/openturns/latest/user_manual/_generated/openturns.Uniform.html.

Required parameters
  • n_samples : int | None

    The maximum number of samples required by the user; for axial, composite and factorial DOEs, a minimum number of samples is required and depends on the dimension of the space to sample; if None in the case of for axial, composite, factorial and full-factorial DOEs the effective number of samples is computed from this dimension and the number of levels.

Optional parameters
  • callbacks : Iterable[CallbackType], optional

    The functions to be evaluated after each call to OptimizationProblem.evaluate_functions(); to be called as callback(index, (output, jacobian)).

    By default it is set to ().

  • eval_jac : bool, optional

    Whether to evaluate the jacobian.

    By default it is set to False.

  • max_time : float, optional

    The maximum runtime in seconds, disabled if 0.

    By default it is set to 0.

  • n_processes : int, optional

    The maximum simultaneous number of processes used to parallelize the execution.

    By default it is set to 1.

  • seed : int | None, optional

    The seed used for reproducibility reasons. If None, use seed.

    By default it is set to None.

  • wait_time_between_samples : float, optional

    The waiting time between two samples.

    By default it is set to 0.0.

OT_OPT_LHS

Module: gemseo.algos.doe.lib_openturns

Optimal Latin Hypercube Sampling

More details about the algorithm and its options on http://openturns.github.io/openturns/latest/user_manual/_generated/openturns.SimulatedAnnealingLHS.html.

Required parameters
  • n_samples : int | None

    The maximum number of samples required by the user; for axial, composite and factorial DOEs, a minimum number of samples is required and depends on the dimension of the space to sample; if None in the case of for axial, composite, factorial and full-factorial DOEs the effective number of samples is computed from this dimension and the number of levels.

Optional parameters
  • annealing : bool, optional

    If True, use simulated annealing to optimize LHS. Otherwise, use crude Monte Carlo.

    By default it is set to True.

  • callbacks : Iterable[CallbackType], optional

    The functions to be evaluated after each call to OptimizationProblem.evaluate_functions(); to be called as callback(index, (output, jacobian)).

    By default it is set to ().

  • criterion : OTOptimalLHS.SpaceFillingCriterion, optional

    The space-filling criterion, either “C2”, “PhiP” or “MinDist”.

    By default it is set to C2.

  • eval_jac : bool, optional

    Whether to evaluate the jacobian.

    By default it is set to False.

  • levels : float | Sequence[float] | None, optional

    1) In the case of axial, composite and factorial DOEs, the positions of the levels relative to the center; the levels will be equispaced and symmetrical relative to the center; e.g. [0.2, 0.8] in dimension 1 will generate the samples [0.15, 0.6, 0.75, 0.8, 0.95, 1] for an axial DOE; the values must be in \(]0,1]\). 2) In the case of a full-factorial DOE, the number of levels per input direction; if scalar, this value is applied to each input direction.

    By default it is set to None.

  • max_time : float, optional

    The maximum runtime in seconds, disabled if 0.

    By default it is set to 0.

  • n_processes : int, optional

    The maximum simultaneous number of processes used to parallelize the execution.

    By default it is set to 1.

  • n_replicates : int, optional

    The number of Monte Carlo replicates to optimize LHS.

    By default it is set to 1000.

  • seed : int | None, optional

    The seed used for reproducibility reasons. If None, use seed.

    By default it is set to None.

  • temperature : OTOptimalLHS.TemperatureProfile, optional

    The temperature profile for simulated annealing, either “Geometric” or “Linear”.

    By default it is set to Geometric.

  • wait_time_between_samples : float, optional

    The waiting time between two samples.

    By default it is set to 0.0.

OT_RANDOM

Module: gemseo.algos.doe.lib_openturns

Random sampling

More details about the algorithm and its options on http://openturns.github.io/openturns/latest/user_manual/_generated/openturns.Uniform.html.

Required parameters
  • n_samples : int | None

    The maximum number of samples required by the user; for axial, composite and factorial DOEs, a minimum number of samples is required and depends on the dimension of the space to sample; if None in the case of for axial, composite, factorial and full-factorial DOEs the effective number of samples is computed from this dimension and the number of levels.

Optional parameters
  • callbacks : Iterable[CallbackType], optional

    The functions to be evaluated after each call to OptimizationProblem.evaluate_functions(); to be called as callback(index, (output, jacobian)).

    By default it is set to ().

  • eval_jac : bool, optional

    Whether to evaluate the jacobian.

    By default it is set to False.

  • max_time : float, optional

    The maximum runtime in seconds, disabled if 0.

    By default it is set to 0.

  • n_processes : int, optional

    The maximum simultaneous number of processes used to parallelize the execution.

    By default it is set to 1.

  • seed : int | None, optional

    The seed used for reproducibility reasons. If None, use seed.

    By default it is set to None.

  • wait_time_between_samples : float, optional

    The waiting time between two samples.

    By default it is set to 0.0.

OT_REVERSE_HALTON

Module: gemseo.algos.doe.lib_openturns

Reverse Halton

More details about the algorithm and its options on http://openturns.github.io/openturns/latest/user_manual/_generated/openturns.ReverseHaltonSequence.html.

Required parameters
  • n_samples : int | None

    The maximum number of samples required by the user; for axial, composite and factorial DOEs, a minimum number of samples is required and depends on the dimension of the space to sample; if None in the case of for axial, composite, factorial and full-factorial DOEs the effective number of samples is computed from this dimension and the number of levels.

Optional parameters
  • callbacks : Iterable[CallbackType], optional

    The functions to be evaluated after each call to OptimizationProblem.evaluate_functions(); to be called as callback(index, (output, jacobian)).

    By default it is set to ().

  • eval_jac : bool, optional

    Whether to evaluate the jacobian.

    By default it is set to False.

  • max_time : float, optional

    The maximum runtime in seconds, disabled if 0.

    By default it is set to 0.

  • n_processes : int, optional

    The maximum simultaneous number of processes used to parallelize the execution.

    By default it is set to 1.

  • seed : int | None, optional

    The seed used for reproducibility reasons. If None, use seed.

    By default it is set to None.

  • wait_time_between_samples : float, optional

    The waiting time between two samples.

    By default it is set to 0.0.

OT_SOBOL

Module: gemseo.algos.doe.lib_openturns

Sobol sequence

More details about the algorithm and its options on http://openturns.github.io/openturns/latest/user_manual/_generated/openturns.SobolSequence.html.

Required parameters
  • n_samples : int | None

    The maximum number of samples required by the user; for axial, composite and factorial DOEs, a minimum number of samples is required and depends on the dimension of the space to sample; if None in the case of for axial, composite, factorial and full-factorial DOEs the effective number of samples is computed from this dimension and the number of levels.

Optional parameters
  • callbacks : Iterable[CallbackType], optional

    The functions to be evaluated after each call to OptimizationProblem.evaluate_functions(); to be called as callback(index, (output, jacobian)).

    By default it is set to ().

  • eval_jac : bool, optional

    Whether to evaluate the jacobian.

    By default it is set to False.

  • max_time : float, optional

    The maximum runtime in seconds, disabled if 0.

    By default it is set to 0.

  • n_processes : int, optional

    The maximum simultaneous number of processes used to parallelize the execution.

    By default it is set to 1.

  • seed : int | None, optional

    The seed used for reproducibility reasons. If None, use seed.

    By default it is set to None.

  • wait_time_between_samples : float, optional

    The waiting time between two samples.

    By default it is set to 0.0.

OT_SOBOL_INDICES

Module: gemseo.algos.doe.lib_openturns

DOE for Sobol ‘indices

More details about the algorithm and its options on http://openturns.github.io/openturns/latest/user_manual/_generated/openturns.SobolIndicesAlgorithm.html.

Required parameters
  • n_samples : int | None

    The maximum number of samples required by the user; for axial, composite and factorial DOEs, a minimum number of samples is required and depends on the dimension of the space to sample; if None in the case of for axial, composite, factorial and full-factorial DOEs the effective number of samples is computed from this dimension and the number of levels.

Optional parameters
  • callbacks : Iterable[CallbackType], optional

    The functions to be evaluated after each call to OptimizationProblem.evaluate_functions(); to be called as callback(index, (output, jacobian)).

    By default it is set to ().

  • eval_jac : bool, optional

    Whether to evaluate the jacobian.

    By default it is set to False.

  • eval_second_order : bool, optional

    Whether to build a DOE to evaluate also the second-order indices; otherwise, the DOE is designed for first- and total-order indices only.

    By default it is set to True.

  • max_time : float, optional

    The maximum runtime in seconds, disabled if 0.

    By default it is set to 0.

  • n_processes : int, optional

    The maximum simultaneous number of processes used to parallelize the execution.

    By default it is set to 1.

  • seed : int | None, optional

    The seed used for reproducibility reasons. If None, use seed.

    By default it is set to None.

  • wait_time_between_samples : float, optional

    The waiting time between two samples.

    By default it is set to 0.0.

PoissonDisk

Module: gemseo.algos.doe.lib_scipy

Poisson disk sampling

Required parameters
  • n_samples : int

    The number of samples.

Optional parameters
  • callbacks : Iterable[CallbackType], optional

    The functions to be evaluated after each call to OptimizationProblem.evaluate_functions(); to be called as callback(index, (output, jacobian)).

    By default it is set to ().

  • eval_jac : bool, optional

    Whether to evaluate the jacobian.

    By default it is set to False.

  • hypersphere : Hypersphere, optional

    The sampling strategy to generate potential candidates to be added in the final sample.

    By default it is set to volume.

  • max_time : float, optional

    The maximum runtime in seconds, disabled if 0.

    By default it is set to 0.

  • n_processes : int, optional

    The maximum simultaneous number of processes used to parallelize the execution.

    By default it is set to 1.

  • ncandidates : int, optional

    The number of candidates to sample per iteration.

    By default it is set to 30.

  • radius : float, optional

    The minimal distance to keep between points when sampling new candidates.

    By default it is set to 0.05.

  • seed : int | None, optional

    The seed used for reproducibility reasons. If None, use seed.

    By default it is set to None.

  • wait_time_between_samples : float, optional

    The waiting time between two samples.

    By default it is set to 0.0.

Sobol

Module: gemseo.algos.doe.lib_scipy

Engine for generating (scrambled) Sobol’ sequences

Required parameters
  • n_samples : int

    The number of samples.

  • optimization : Optimizer

    The name of an optimization scheme to improve the quality of the DOE. If None, use the DOE as is. New in SciPy 1.10.0.

Optional parameters
  • bits : int | None, optional

    The number of bits of the generator. New in SciPy 1.9.0.

    By default it is set to None.

  • callbacks : Iterable[CallbackType], optional

    The functions to be evaluated after each call to OptimizationProblem.evaluate_functions(); to be called as callback(index, (output, jacobian)).

    By default it is set to ().

  • eval_jac : bool, optional

    Whether to evaluate the jacobian.

    By default it is set to False.

  • max_time : float, optional

    The maximum runtime in seconds, disabled if 0.

    By default it is set to 0.

  • n_processes : int, optional

    The maximum simultaneous number of processes used to parallelize the execution.

    By default it is set to 1.

  • scramble : bool, optional

    Whether to use scrambling (Owen type). Only available with SciPy >= 1.10.0.

    By default it is set to True.

  • seed : int | None, optional

    The seed used for reproducibility reasons. If None, use seed.

    By default it is set to None.

  • wait_time_between_samples : float, optional

    The waiting time between two samples.

    By default it is set to 0.0.

bbdesign

Module: gemseo.algos.doe.lib_pydoe

Box-Behnken design implemented in pyDOE

More details about the algorithm and its options on https://pythonhosted.org/pyDOE/rsm.html#box-behnken.

Optional parameters
  • callbacks : Iterable[CallbackType], optional

    The functions to be evaluated after each call to OptimizationProblem.evaluate_functions(); to be called as callback(index, (output, jacobian)).

    By default it is set to ().

  • center_bb : int | None, optional

    The number of center points for the Box-Behnken design. If None, use a pre-determined number of points.

    By default it is set to None.

  • eval_jac : bool, optional

    Whether to evaluate the jacobian.

    By default it is set to False.

  • max_time : float, optional

    The maximum runtime in seconds, disabled if 0.

    By default it is set to 0.

  • n_processes : int, optional

    The maximum simultaneous number of processes used to parallelize the execution.

    By default it is set to 1.

  • wait_time_between_samples : float, optional

    The waiting time between two samples.

    By default it is set to 0.0.

ccdesign

Module: gemseo.algos.doe.lib_pydoe

Central Composite implemented in pyDOE

More details about the algorithm and its options on https://pythonhosted.org/pyDOE/rsm.html#central-composite.

Optional parameters
  • alpha : str, optional

    A parameter to describe how the variance is distributed. Either “orthogonal” or “rotatable”.

    By default it is set to orthogonal.

  • callbacks : Iterable[CallbackType], optional

    The functions to be evaluated after each call to OptimizationProblem.evaluate_functions(); to be called as callback(index, (output, jacobian)).

    By default it is set to ().

  • center_cc : tuple[int, int] | None, optional

    The 2-tuple of center points for the central composite design. If None, use (4, 4).

    By default it is set to None.

  • eval_jac : bool, optional

    Whether to evaluate the jacobian.

    By default it is set to False.

  • face : str, optional

    The relation between the start points and the corner (factorial) points. Either “circumscribed”, “inscribed” or “faced”.

    By default it is set to faced.

  • max_time : float, optional

    The maximum runtime in seconds, disabled if 0.

    By default it is set to 0.

  • n_processes : int, optional

    The maximum simultaneous number of processes used to parallelize the execution.

    By default it is set to 1.

  • wait_time_between_samples : float, optional

    The waiting time between two samples.

    By default it is set to 0.0.

ff2n

Module: gemseo.algos.doe.lib_pydoe

2-Level Full-Factorial implemented in pyDOE

More details about the algorithm and its options on https://pythonhosted.org/pyDOE/factorial.html#level-full-factorial.

Optional parameters
  • callbacks : Iterable[CallbackType], optional

    The functions to be evaluated after each call to OptimizationProblem.evaluate_functions(); to be called as callback(index, (output, jacobian)).

    By default it is set to ().

  • eval_jac : bool, optional

    Whether to evaluate the jacobian.

    By default it is set to False.

  • max_time : float, optional

    The maximum runtime in seconds, disabled if 0.

    By default it is set to 0.

  • n_processes : int, optional

    The maximum simultaneous number of processes used to parallelize the execution.

    By default it is set to 1.

  • wait_time_between_samples : float, optional

    The waiting time between two samples.

    By default it is set to 0.0.

fullfact

Module: gemseo.algos.doe.lib_pydoe

Full-Factorial implemented in pyDOE

More details about the algorithm and its options on https://pythonhosted.org/pyDOE/factorial.html#general-full-factorial.

Required parameters
  • levels : Sequence[int] | None

    The levels. If there is a parameter n_samples, the latter can be specified and the former set to its default value None.

  • n_samples : int | None

    The number of samples. If there is a parameter levels, the latter can be specified and the former set to its default value None.

Optional parameters
  • callbacks : Iterable[CallbackType], optional

    The functions to be evaluated after each call to OptimizationProblem.evaluate_functions(); to be called as callback(index, (output, jacobian)).

    By default it is set to ().

  • eval_jac : bool, optional

    Whether to evaluate the jacobian.

    By default it is set to False.

  • max_time : float, optional

    The maximum runtime in seconds, disabled if 0.

    By default it is set to 0.

  • n_processes : int, optional

    The maximum simultaneous number of processes used to parallelize the execution.

    By default it is set to 1.

  • wait_time_between_samples : float, optional

    The waiting time between two samples.

    By default it is set to 0.0.

lhs

Module: gemseo.algos.doe.lib_pydoe

Latin Hypercube Sampling implemented in pyDOE

More details about the algorithm and its options on https://pythonhosted.org/pyDOE/randomized.html#latin-hypercube.

Required parameters
  • n_samples : int | None

    The number of samples. If there is a parameter levels, the latter can be specified and the former set to its default value None.

Optional parameters
  • callbacks : Iterable[CallbackType], optional

    The functions to be evaluated after each call to OptimizationProblem.evaluate_functions(); to be called as callback(index, (output, jacobian)).

    By default it is set to ().

  • criterion : str | None, optional

    The criterion to use when sampling the points. If None, randomize the points within the intervals.

    By default it is set to None.

  • eval_jac : bool, optional

    Whether to evaluate the jacobian.

    By default it is set to False.

  • iterations : int, optional

    The number of iterations in the correlation and maximin algorithms.

    By default it is set to 5.

  • max_time : float, optional

    The maximum runtime in seconds, disabled if 0.

    By default it is set to 0.

  • n_processes : int, optional

    The maximum simultaneous number of processes used to parallelize the execution.

    By default it is set to 1.

  • seed : int | None, optional

    The seed used for reproducibility reasons. If None, use seed.

    By default it is set to None.

  • wait_time_between_samples : float, optional

    The waiting time between two samples.

    By default it is set to 0.0.

pbdesign

Module: gemseo.algos.doe.lib_pydoe

Plackett-Burman design implemented in pyDOE

More details about the algorithm and its options on https://pythonhosted.org/pyDOE/factorial.html#plackett-burman.

Optional parameters
  • callbacks : Iterable[CallbackType], optional

    The functions to be evaluated after each call to OptimizationProblem.evaluate_functions(); to be called as callback(index, (output, jacobian)).

    By default it is set to ().

  • eval_jac : bool, optional

    Whether to evaluate the jacobian.

    By default it is set to False.

  • max_time : float, optional

    The maximum runtime in seconds, disabled if 0.

    By default it is set to 0.

  • n_processes : int, optional

    The maximum simultaneous number of processes used to parallelize the execution.

    By default it is set to 1.

  • wait_time_between_samples : float, optional

    The waiting time between two samples.

    By default it is set to 0.0.