Python API Reference

This reference is generated from the dakota.study package.

How to use this reference

A dakota.study.Study owns the runtime services used by every object created from it. Build a study from the leaves inward: create variables, responses, and an interface; inject those objects into a model factory; inject the model into a method factory; and finally pass the method to dakota.study.Study.run(). Keep the study alive as long as any of its components are in use, and do not mix components from different studies.

Factory signatures distinguish two kinds of input. Parameters typed as Study API objects are dependencies: pass the actual object rather than an input-file pointer name. The remaining Dakota options form a configuration fragment relative to the selected factory. They may be supplied as keyword arguments, as one dictionary passed through config, or as the corresponding Pydantic configuration model. Do not combine config with configuration keyword arguments.

The generated keyword reference remains the authoritative description of individual configuration options. Each factory below links to the relevant keyword subtree. Pointer fields required by the input-file schema are automatically represented by internal sentinels in API mode when the dependency has already been injected as an object.

API map

Task

Entry point

Result

Notes

Establish runtime ownership

dakota.study.Study

A study with model and method factories

A Study object provides an execution context for a single study. Instantiate only one Study at a time.

Define Model sub-components

Study.variables, Study.responses, Study.interface

dakota.study.Variables, dakota.study.Response, and dakota.study.Interface

Not every Model requires an interface

Assemble a model

dakota.study.ModelFactory

A dakota.study.Model specialization

ModelFactory should not be directly instantiated. The model member variable of Study is a ModelFactory. Some Models have method/iterator or model sub-components

Select an algorithm

dakota.study.MethodFactory

An dakota.study.Iterator specialization

MethodFactory should not be directly instantiated. The method member variable of a Study is a MethodFactory.

Execute the study

dakota.study.Study.run()

Results retained by the returned iterator handle

None

Study and configuration

class dakota.study.Study

Own the runtime services and factories for a library-mode Dakota study.

Keep the Study alive while using components created from it. Construct components from the leaves inward, then pass the top-level method to run().

interface(config: object = None, *, callback: object = None, **kwargs) → dakota.study.Interface

Construct a simulation interface from an interface fragment.

Parameters:
  • config – An InterfaceConfig or equivalent dictionary.

  • callback – For a python interface, a callable that maps a parameters dictionary to a results dictionary; replaces importing the analysis driver module.function.

  • kwargs – Configuration fields used instead of config.

Returns:

An interface owned by this study.

Return type:

dakota.study.Interface

Raises:

TypeError – If both config and configuration keywords are supplied, or callback is given for a non-python interface.

See interface options.

property method

Method factory tied to this Study’s runtime services.

property model

Model factory tied to this Study’s runtime services.

responses(variables: dakota.study.Variables, config: object = None, **kwargs) → dakota.study.Response

Construct a response definition associated with variables.

Parameters:
  • variables (dakota.study.Variables) – Variables that determine active derivative components.

  • config – A ResponsesConfig or equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

A response definition owned by this study.

Return type:

dakota.study.Response

Raises:

TypeError – If both config and configuration keywords are supplied.

See response options.

run(method: dakota.study.Iterator) → None

Execute a method as the top-level iterator.

Parameters:

method (dakota.study.Iterator) – Iterator constructed with this study’s runtime services.

Returns:

None

Raises:

RuntimeError – If the method is null or uses incompatible services.

variables(config: object = None, **kwargs) → dakota.study.Variables

Construct variables from the children of a variables block.

Parameters:
  • config – A VariablesConfig or equivalent dictionary. Omit it when passing configuration fields as keyword arguments.

  • kwargs – Configuration fields used instead of config.

Returns:

Variables owned by this study.

Return type:

dakota.study.Variables

Raises:

TypeError – If both config and configuration keywords are supplied.

See variables options.

class dakota.study.StudyConfig

Top-level typed configuration for a dependency-injection Study.

property output

Output and restart configuration.

property run

Execution-phase configuration.

class dakota.study.StudyOutputConfig

Output, restart, tabular-data, and results settings for a Study.

property error_file

File receiving Dakota error output; empty uses the active stream.

property graphics

Enable Dakota graphics output when supported.

property interface_evals_selection

Selection policy for interface evaluations stored in results output.

property model_evals_selection

Selection policy for model evaluations stored in results output.

property output_file

File receiving Dakota standard output; empty uses the active stream.

property precision

Numeric output precision; zero retains Dakota’s default.

property read_restart

Restart file to read before execution.

property results_output

Enable results-database output.

property results_output_file

Base name for results-database output.

property results_output_format

Dakota results-output format enumeration value.

property stop_restart

Maximum evaluation number to read from the restart file; zero is unlimited.

property tabular_format

Dakota tabular-format enumeration value.

property tabular_graphics_data

Enable tabular graphics-data output.

property tabular_graphics_file

Destination for tabular graphics data.

property write_restart

Restart file to write during execution.

class dakota.study.StudyRunConfig

Controls the pre-run, run, and post-run execution phases of a Study.

property post_run

Enable the post-run phase.

property post_run_input

Input file for the post-run phase.

property post_run_input_format

Dakota format enumeration for post-run input.

property post_run_output

Output file for the post-run phase.

property pre_run

Enable the pre-run phase.

property pre_run_input

Input file for the pre-run phase.

property pre_run_output

Output file for the pre-run phase.

property pre_run_output_format

Dakota format enumeration for pre-run output.

property run

Enable the main run phase.

property run_input

Input file for the main run phase.

property run_output

Output file for the main run phase.

Factories

ModelFactory and MethodFactory support the implementation of dakota.study.Study; users ordinarily should not instantiate them directly. Access the factory instances owned by a study through dakota.study.Study.model and dakota.study.Study.method. This ensures that constructed models and methods use the correct runtime services and ownership context.

Factory calls put object dependencies first, followed by configuration. For example, method factories consistently use factory(model, config=None, **kwargs). Supply Dakota options either as a dictionary or Pydantic model through config, or directly as keyword arguments:

study.method.centered_parameter_study(model, config=method_options)
study.method.centered_parameter_study(
    model, step_vector=[0.1, 0.1], steps_per_variable=[2, 2])

Do not pass both config and configuration keyword arguments. Object dependencies such as model are not configuration fields and replace the corresponding input-file pointer when one exists.

class dakota.study.ModelFactory

Construct models that share the owning Study’s runtime services.

This factory is an implementation detail of Study and ordinarily should not be instantiated directly. Access it through Study.model. Inject variables, responses, interfaces, sub-iterators, and child models as objects rather than resolving input-file pointer strings.

active_subspace(truth_model: dakota.study.Model, variables: dakota.study.Variables, response: dakota.study.Response, config: object = None, **kwargs) → dakota.study.ActiveSubspaceModel

Construct an active subspace model over an injected truth model.

Parameters:
  • truth_model (dakota.study.Model) – Model whose inputs are reduced.

  • variables (dakota.study.Variables) – Variables exposed by the reduced model.

  • response (dakota.study.Response) – Response definition exposed by the reduced model.

  • config – A ActiveSubspaceConfig or equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed model.

Return type:

dakota.study.ActiveSubspaceModel

See active subspace options.

adapted_basis(truth_model: dakota.study.Model, variables: dakota.study.Variables, response: dakota.study.Response, config: object = None, **kwargs) → dakota.study.AdaptedBasisModel

Construct an adapted basis model over an injected truth model.

Parameters:
  • truth_model (dakota.study.Model) – Model whose inputs are reduced.

  • variables (dakota.study.Variables) – Variables exposed by the reduced model.

  • response (dakota.study.Response) – Response definition exposed by the reduced model.

  • config – A AdaptedBasisConfig or equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed model.

Return type:

dakota.study.AdaptedBasisModel

See adapted basis options.

ensemble_surrogate(variables: dakota.study.Variables, response: dakota.study.Response, *, ordered_models: list[dakota.study.Model] | None = None, truth_model: dakota.study.Model = None, approximation_models: list[dakota.study.Model] | None = None, config: object = None, **kwargs) → dakota.study.EnsembleSurrModel

Construct an ensemble surrogate from injected model objects.

Exactly one dependency form is required: provide ordered_models from low to high fidelity, or provide truth_model with optional approximation_models. The final ordered model is the truth model.

Pointer fields omitted from the configuration are filled with API-mode sentinels during validation; the injected model objects replace those input-file dependencies. Pass config or configuration keyword arguments, not both.

Parameters:
  • variables (dakota.study.Variables) – Variables shared by the ensemble model.

  • response (dakota.study.Response) – Response definition shared by the ensemble model.

  • ordered_models (sequence[dakota.study.Model] or None) – Models ordered from low to high fidelity.

  • truth_model (dakota.study.Model or None) – Explicit high-fidelity truth model.

  • approximation_models (sequence[dakota.study.Model] or None) – Unordered approximation models. This may be omitted for a resolution-only hierarchy.

  • config – An EnsembleSurrogateConfig or equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed ensemble surrogate.

Return type:

dakota.study.EnsembleSurrModel

Raises:
  • TypeError – If dependency forms are missing, mixed, null, or otherwise invalid, or if both config and configuration keywords are supplied.

  • ValueError – If ordered_models is empty or configuration validation fails.

See ensemble surrogate options.

global_surrogate(variables: dakota.study.Variables, response: dakota.study.Response, truth_model: dakota.study.Model = None, dace_iterator: dakota.study.Iterator = None, config: object = None, **kwargs) → dakota.study.DataFitSurrModel

Construct a global data-fit surrogate.

Parameters:
  • variables (dakota.study.Variables) – Variables exposed by the surrogate.

  • response (dakota.study.Response) – Response definition exposed by the surrogate.

  • truth_model (dakota.study.Model or None) – Optional model used to generate or check approximation data.

  • dace_iterator (dakota.study.Iterator or None) – Optional iterator used to generate approximation data.

  • config – A GlobalSurrogateConfig or equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed data-fit surrogate.

Return type:

dakota.study.DataFitSurrModel

See global surrogate options.

local_surrogate(truth_model: dakota.study.Model, variables: dakota.study.Variables, response: dakota.study.Response, config: object = None, **kwargs) → dakota.study.DataFitSurrModel

Construct a local surrogate from an injected truth model.

Parameters:
  • truth_model (dakota.study.Model) – Model approximated by the local surrogate.

  • variables (dakota.study.Variables) – Variables exposed by the surrogate.

  • response (dakota.study.Response) – Response definition exposed by the surrogate.

  • config – A LocalSurrogateConfig or equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed data-fit surrogate.

Return type:

dakota.study.DataFitSurrModel

See local surrogate options.

multipoint_surrogate(truth_model: dakota.study.Model, variables: dakota.study.Variables, response: dakota.study.Response, config: object = None, **kwargs) → dakota.study.DataFitSurrModel

Construct a multipoint surrogate from an injected truth model.

Parameters:
  • truth_model (dakota.study.Model) – Model approximated by the multipoint surrogate.

  • variables (dakota.study.Variables) – Variables exposed by the surrogate.

  • response (dakota.study.Response) – Response definition exposed by the surrogate.

  • config – A MultipointSurrogateConfig or equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed data-fit surrogate.

Return type:

dakota.study.DataFitSurrModel

See multipoint surrogate options.

nested(sub_iterator: dakota.study.Iterator, variables: dakota.study.Variables, response: dakota.study.Response, optional_interface: dakota.study.Interface = None, config: object = None, *, optional_interface_responses: dakota.study.Response | None = None, **kwargs) → dakota.study.NestedModel

Construct a nested model from an injected sub-iterator.

Parameters:
  • sub_iterator (dakota.study.Iterator) – Method executed by the nested model.

  • variables (dakota.study.Variables) – Variables exposed by the nested model.

  • response (dakota.study.Response) – Response definition exposed by the nested model.

  • optional_interface (dakota.study.Interface or None) – Optional interface for non-nested responses.

  • optional_interface_responses (dakota.study.Response or None) – Response definition returned by the optional interface; required when optional_interface_responses_pointer is set.

  • config – A NestedConfig or equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed nested model.

Return type:

dakota.study.NestedModel

Raises:

TypeError – If a required dependency is null or both configuration forms are used.

See nested-model options.

random_field(propagation_model: dakota.study.Model, variables: dakota.study.Variables, response: dakota.study.Response, dace_iterator: dakota.study.Iterator = None, config: object = None, **kwargs) → dakota.study.RandomFieldModel

Construct a random field model over an injected propagation model.

Parameters:
  • propagation_model (dakota.study.Model) – Model through which field realizations propagate.

  • variables (dakota.study.Variables) – Variables exposed by the random field model.

  • response (dakota.study.Response) – Response definition exposed by the random field model.

  • dace_iterator (dakota.study.Iterator or None) – Optional iterator that generates field realizations.

  • config – A RandomFieldConfig or equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed model.

Return type:

dakota.study.RandomFieldModel

See random field options.

simulation(variables: dakota.study.Variables, interface: dakota.study.Interface, response: dakota.study.Response, config: object = None, **kwargs) → dakota.study.SimulationModel

Construct a simulation model from injected components.

Parameters:
  • variables (dakota.study.Variables) – Variables evaluated by the model.

  • interface (dakota.study.Interface) – Interface used to perform evaluations.

  • response (dakota.study.Response) – Response definition for evaluations.

  • config – A SingleConfig or equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed simulation model.

Return type:

dakota.study.SimulationModel

Raises:

TypeError – If a dependency is null or both configuration forms are used.

See single-model options.

single(model: object, variables: dakota.study.Variables, interface: dakota.study.Interface, response: dakota.study.Response) → dakota.study.SimulationModel

Construct a simulation model using the legacy config-first calling convention.

class dakota.study.MethodFactory

Construct methods that share the owning Study’s runtime services.

This factory is an implementation detail of Study and ordinarily should not be instantiated directly. Access it through Study.method. Factory availability depends on the capabilities enabled in the Dakota build.

adaptive_sampling(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct adaptive_sampling for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See adaptive_sampling options.

approximate_control_variate(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct approximate_control_variate for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See approximate_control_variate options.

Construct asynch_pattern_search for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See asynch_pattern_search options.

bayes_calibration(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct bayes_calibration for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See bayes_calibration options.

branch_and_bound(model: dakota.study.Model = None, method: dakota.study.Iterator = None, config: object = None, **kwargs) → dakota.study.PebbldMinimizer

Construct a branch-and-bound minimizer.

Parameters:
  • model (dakota.study.Model or None) – Model with relaxed discrete variables to branch on (build its variables with domain={'relaxed': True}; unlike an input file, the domain is not inferred from the method). Required with method_name; with method_pointer it defaults to the model of method and, if given, must be that model.

  • method (dakota.study.Iterator or None) – Sub-problem minimizer; required with method_pointer and omitted with method_name.

  • config – A BranchAndBoundConfig or equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed minimizer.

Return type:

dakota.study.PebbldMinimizer

Raises:
  • TypeError – If both configuration forms are used.

  • RuntimeError – If model or method is missing or unexpected for the selected sub-method, or method iterates on a different model.

This method is present only when Dakota is built with ACRO. See branch_and_bound options.

centered_parameter_study(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.ParamStudy

Construct centered_parameter_study for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See centered_parameter_study options.

coliny_beta(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct coliny_beta for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See coliny_beta options.

coliny_cobyla(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct coliny_cobyla for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See coliny_cobyla options.

coliny_direct(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct coliny_direct for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See coliny_direct options.

coliny_ea(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct coliny_ea for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See coliny_ea options.

Construct coliny_pattern_search for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See coliny_pattern_search options.

coliny_solis_wets(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct coliny_solis_wets for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See coliny_solis_wets options.

conmin_frcg(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct conmin_frcg for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See conmin_frcg options.

conmin_mfd(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct conmin_mfd for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See conmin_mfd options.

dace(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct dace for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See dace options.

dot_bfgs(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.DOTOptimizer

Construct a DOT BFGS method for an injected model.

Parameters:
  • model (dakota.study.Model) – Model optimized by DOT BFGS.

  • config – A DotBfgsConfig or equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed optimizer.

Return type:

dakota.study.DOTOptimizer

Raises:

TypeError – If model is null or both configuration forms are used.

This method is present only when Dakota is built with DOT. See DOT BFGS options.

dot_frcg(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct dot_frcg for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See dot_frcg options.

dot_mmfd(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct dot_mmfd for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See dot_mmfd options.

dot_slp(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct dot_slp for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See dot_slp options.

dot_sqp(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct dot_sqp for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See dot_sqp options.

efficient_global(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.EffGlobalMinimizer

Construct efficient_global for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See efficient_global options.

external_python(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct external_python for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See external_python options.

fsu_cvt(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct fsu_cvt for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See fsu_cvt options.

fsu_quasi_mc(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct fsu_quasi_mc for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See fsu_quasi_mc options.

genie_direct(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct genie_direct for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See genie_direct options.

genie_opt_darts(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct genie_opt_darts for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See genie_opt_darts options.

global_evidence(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct global_evidence for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See global_evidence options.

global_interval_est(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct global_interval_est for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See global_interval_est options.

global_reliability(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct global_reliability for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See global_reliability options.

gpais(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct gpais for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See gpais options.

hybrid_sequential(methods: object = None, models: object = None, config: object = None, **kwargs) → dakota.study.SeqHybridMetaIterator

Construct a sequential hybrid meta-iterator.

Pointer form: pass methods (required, non-None items).

Name form (method_name_list in config): pass models (required) and omit methods. Each sub-iterator is built on-the-fly from the corresponding name and model.

Parameters:
  • methods (list[dakota.study.Iterator] or None) – Ordered list of methods to execute in sequence (pointer form).

  • models (list[dakota.study.Model] or None) – List of models for the named sub-methods (name form). May be a single-element list to share one model across all sub-methods.

  • config – A HybridSequentialConfig or equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed sequential hybrid.

Return type:

dakota.study.SeqHybridMetaIterator

Raises:

TypeError – If neither methods nor models is provided, a method is null, or both configuration forms are used.

See hybrid_sequential options.

import_points(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct import_points for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See import_points options.

importance_sampling(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct importance_sampling for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See importance_sampling options.

list_parameter_study(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.ParamStudy

Construct list_parameter_study for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See list_parameter_study options.

local_evidence(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct local_evidence for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See local_evidence options.

local_interval_est(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.NonDLocalSingleInterval

Construct local_interval_est for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See local_interval_est options.

local_reliability(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct local_reliability for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See local_reliability options.

Construct mesh_adaptive_search for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See mesh_adaptive_search options.

moga(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct moga for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See moga options.

multi_start(sub_iterator: dakota.study.Iterator = None, config: object = None, model: dakota.study.Model = None, **kwargs) → dakota.study.ConcurrentMetaIterator

Construct a multi-start method.

Name form (sub_method_name in config): pass model (required) and omit sub_iterator. The sub-method is built on-the-fly from the named method type using the given model.

Pointer form (sub_method_pointer in config): pass sub_iterator (required) and omit model.

Parameters:
  • sub_iterator (dakota.study.Iterator or None) – Method executed from each starting point (pointer form).

  • model (dakota.study.Model or None) – Model to pass to the named sub-method (name form).

  • config – A MultiStartConfig or equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed meta-iterator.

Return type:

dakota.study.ConcurrentMetaIterator

Raises:

TypeError – If neither sub_iterator nor model is provided, or both configuration forms are used.

See multi-start options.

multidim_parameter_study(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.ParamStudy

Construct multidim_parameter_study for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See multidim_parameter_study options.

multifidelity_polynomial_chaos(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct multifidelity_polynomial_chaos for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See multifidelity_polynomial_chaos options.

multifidelity_sampling(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct multifidelity_sampling for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See multifidelity_sampling options.

multifidelity_stoch_collocation(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct multifidelity_stoch_collocation for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See multifidelity_stoch_collocation options.

multilevel_blue(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct multilevel_blue for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See multilevel_blue options.

multilevel_multifidelity_sampling(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct multilevel_multifidelity_sampling for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See multilevel_multifidelity_sampling options.

multilevel_polynomial_chaos(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct multilevel_polynomial_chaos for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See multilevel_polynomial_chaos options.

multilevel_sampling(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct multilevel_sampling for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See multilevel_sampling options.

ncsu_direct(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct ncsu_direct for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See ncsu_direct options.

nl2sol(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.NL2SOLLeastSq

Construct nl2sol for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See nl2sol options.

nlpql_sqp(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct nlpql_sqp for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See nlpql_sqp options.

nlssol_sqp(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct nlssol_sqp for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See nlssol_sqp options.

nonlinear_cg(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct nonlinear_cg for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See nonlinear_cg options.

npsol_sqp(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.NPSOLOptimizer

Construct npsol_sqp for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See npsol_sqp options.

optpp_cg(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct optpp_cg for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See optpp_cg options.

optpp_fd_newton(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct optpp_fd_newton for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See optpp_fd_newton options.

optpp_g_newton(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct optpp_g_newton for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See optpp_g_newton options.

optpp_newton(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct optpp_newton for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See optpp_newton options.

optpp_pds(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct optpp_pds for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See optpp_pds options.

optpp_q_newton(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct optpp_q_newton for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See optpp_q_newton options.

pareto_set(sub_iterator: dakota.study.Iterator = None, config: object = None, model: dakota.study.Model = None, **kwargs) → dakota.study.ConcurrentMetaIterator

Construct a Pareto-set meta-iterator.

Name form (sub_method_name in config): pass model (required) and omit sub_iterator. The sub-method is built on-the-fly from the named method type using the given model.

Pointer form (sub_method_pointer in config): pass sub_iterator (required) and omit model.

Parameters:
  • sub_iterator (dakota.study.Iterator or None) – Optimizer run for each weight combination (pointer form).

  • model (dakota.study.Model or None) – Model to pass to the named sub-method (name form).

  • config – A ParetoSetConfig or equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed meta-iterator.

Return type:

dakota.study.ConcurrentMetaIterator

Raises:

TypeError – If neither sub_iterator nor model is provided, or both configuration forms are used.

See pareto_set options.

pof_darts(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct pof_darts for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See pof_darts options.

polynomial_chaos(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct polynomial_chaos for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See polynomial_chaos options.

psuade_moat(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct psuade_moat for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See psuade_moat options.

richardson_extrap(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.RichExtrapVerification

Construct richardson_extrap for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See richardson_extrap options.

rkd_darts(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct rkd_darts for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See rkd_darts options.

rol(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct rol for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See rol options.

sampling(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.NonDLHSSampling

Construct a sampling method for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the sampling method.

  • config – A SamplingConfig or equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed sampling iterator.

Return type:

dakota.study.NonDLHSSampling

Raises:

TypeError – If model is null or both configuration forms are used.

See sampling options.

soga(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct soga for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See soga options.

stoch_collocation(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct stoch_collocation for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See stoch_collocation options.

surrogate_based_global(model: dakota.study.Model, method: dakota.study.Iterator = None, config: object = None, **kwargs) → dakota.study.Iterator

Construct a surrogate-based global minimizer.

Parameters:
  • model (dakota.study.Model) – Surrogate model iterated by the outer loop.

  • method (dakota.study.Iterator or None) – Sub-problem optimizer applied to the surrogate. Required with method_pointer; omitted with method_name (built on the fly). If built on a different model, it is redirected to model.

  • config – A SurrogateBasedGlobalConfig or equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed minimizer.

Return type:

dakota.study.Iterator

Raises:

RuntimeError – If method is missing when method_pointer is specified, if method has scaling or calibration data transforms and was built on a different model, or if model is not a surrogate model.

See surrogate_based_global options.

surrogate_based_local(model: dakota.study.Model, method: dakota.study.Iterator = None, config: object = None, **kwargs) → dakota.study.Iterator

Construct a surrogate-based local minimizer.

Parameters:
  • model (dakota.study.Model) – Surrogate model iterated by the outer trust-region loop.

  • method (dakota.study.Iterator or None) – Sub-problem optimizer applied to the approximate sub-problem. Required when sub_method = method_pointer is used; omitted when sub_method_name is used (the optimizer is built on-the-fly).

  • config – A SurrogateBasedLocalConfig or equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed minimizer.

Return type:

dakota.study.Iterator

Raises:

RuntimeError – If method is missing when method_pointer is specified, or if the supplied model is not a surrogate model.

See surrogate_based_local options.

surrogate_based_uq(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.Iterator

Construct surrogate_based_uq for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See surrogate_based_uq options.

vector_parameter_study(model: dakota.study.Model, config: object = None, **kwargs) → dakota.study.ParamStudy

Construct vector_parameter_study for an injected model.

Parameters:
  • model (dakota.study.Model) – Model evaluated by the method.

  • config – The matching dakota.spec.method configuration model or an equivalent dictionary.

  • kwargs – Configuration fields used instead of config.

Returns:

The constructed method/iterator.

Return type:

dakota.study.Iterator

Raises:

TypeError – If model is null or both configuration forms are used.

See vector_parameter_study options.

Component handles

These objects are returned by Study and its factories. They cannot be constructed independently through the public Python API.

class dakota.study.Variables

Variables handle returned by Study.variables().

num_active_cv() → int

Return the number of active continuous variables.

Returns:

Number of active continuous variables.

Return type:

int

class dakota.study.Response

Response definition or evaluated response returned by the Study API.

function_value(i: int) → float

Return one function value from an evaluated response.

Parameters:

i (int) – Zero-based function index.

Returns:

The selected function value.

Return type:

float

class dakota.study.Interface

Simulation-interface handle returned by Study.interface().

class dakota.study.Model

Base model handle returned by ModelFactory.

class dakota.study.Iterator

Base method/iterator handle returned by MethodFactory.

class dakota.study.NonDLHSSampling

Sampling iterator returned by MethodFactory.sampling().

first_response_value() → float

Return the first function value from the first response.

Returns:

The first recorded scalar function value.

Return type:

float

Raises:

RuntimeError – If the iterator has no responses or the first response has no functions.

num_responses() → int

Return the number of responses produced by a completed run.

Returns:

Number of evaluated responses.

Return type:

int