.. _model-global_surrogate-experimental_gaussian_process: """"""""""""""""""""""""""""" experimental_gaussian_process """"""""""""""""""""""""""""" Use the Gaussian process regression surrogate from the surrogates module .. toctree:: :hidden: :maxdepth: 1 model-global_surrogate-experimental_gaussian_process-import_model model-global_surrogate-experimental_gaussian_process-export_model model-global_surrogate-experimental_gaussian_process-export_approx_variance_file model-global_surrogate-experimental_gaussian_process-trend model-global_surrogate-experimental_gaussian_process-num_restarts model-global_surrogate-experimental_gaussian_process-nugget model-global_surrogate-experimental_gaussian_process-find_nugget model-global_surrogate-experimental_gaussian_process-options_file **Specification** - *Alias:* None - *Arguments:* None - *JSON argument key:* None **Child Keywords:** +-------------------------+------------------------+--------------------+---------------------------------+------------------------------------------------------------+ | Required/Optional | Description of Group | JSON Group Key | Dakota Keyword | Dakota Keyword Description | +=========================+========================+====================+=================================+============================================================+ | Optional | | | `import_model`__ | Import surrogate model from archive file | +-------------------------+------------------------+--------------------+---------------------------------+------------------------------------------------------------+ | Optional | | | `export_model`__ | Exports surrogate model in user-specified format(s) | +-------------------------+------------------------+--------------------+---------------------------------+------------------------------------------------------------+ | Optional | | | `export_approx_variance_file`__ | Output file for surrogate model variance evaluations | +-------------------------+------------------------+--------------------+---------------------------------+------------------------------------------------------------+ | Optional | | | `trend`__ | This keyword enables the use of deterministic polynomial | | | | | | trend function | +-------------------------+------------------------+--------------------+---------------------------------+------------------------------------------------------------+ | Optional | | | `num_restarts`__ | Number of optimization restarts for L-BFGS-B | +-------------------------+------------------------+--------------------+---------------------------------+------------------------------------------------------------+ | Optional (Choose One) | Nugget | nugget | `nugget`__ | Value for the fixed nugget parameter | + + + +---------------------------------+------------------------------------------------------------+ | | | | `find_nugget`__ | Use regression to estimate the nugget. | +-------------------------+------------------------+--------------------+---------------------------------+------------------------------------------------------------+ | Optional | | | `options_file`__ | Filename for a YAML file that specifies Gaussian process | | | | | | options | +-------------------------+------------------------+--------------------+---------------------------------+------------------------------------------------------------+ .. __: model-global_surrogate-experimental_gaussian_process-import_model.html __ model-global_surrogate-experimental_gaussian_process-export_model.html __ model-global_surrogate-experimental_gaussian_process-export_approx_variance_file.html __ model-global_surrogate-experimental_gaussian_process-trend.html __ model-global_surrogate-experimental_gaussian_process-num_restarts.html __ model-global_surrogate-experimental_gaussian_process-nugget.html __ model-global_surrogate-experimental_gaussian_process-find_nugget.html __ model-global_surrogate-experimental_gaussian_process-options_file.html **Description** This Gaussian process implementation is contained in Dakota's surrogates module and is considered experimental. It uses gradient-based optimization with restarts to determine hyperparmeters and trend coefficients. Nugget and trend estimation are optional.