.. _model-active_subspace-truncation_method-cross_validation:

""""""""""""""""
cross_validation
""""""""""""""""


Truncate the subspace to minimize surrogate cross-validation error


.. toctree::
   :hidden:
   :maxdepth: 1

   model-active_subspace-truncation_method-cross_validation-minimum
   model-active_subspace-truncation_method-cross_validation-relative
   model-active_subspace-truncation_method-cross_validation-decrease
   model-active_subspace-truncation_method-cross_validation-relative_tolerance
   model-active_subspace-truncation_method-cross_validation-decrease_tolerance
   model-active_subspace-truncation_method-cross_validation-max_rank
   model-active_subspace-truncation_method-cross_validation-exhaustive


**Specification**

- *Alias:* None

- *Arguments:* None

- *Default:* relative


**Child Keywords:**

+-------------------------+--------------------+------------------------+---------------------------------------------+
| Required/Optional       | Description of     | Dakota Keyword         | Dakota Keyword Description                  |
|                         | Group              |                        |                                             |
+=========================+====================+========================+=============================================+
| Optional (Choose One)   | CV Selection       | `minimum`__            | Select subspace to minimize                 |
|                         | Criterion          |                        | cross-validation error                      |
|                         |                    +------------------------+---------------------------------------------+
|                         |                    | `relative`__           | Choose subspace with cross-validation error |
|                         |                    |                        | less than tolerance                         |
|                         |                    +------------------------+---------------------------------------------+
|                         |                    | `decrease`__           | Choose subspace where cross-validation      |
|                         |                    |                        | error stabilizes                            |
+-------------------------+--------------------+------------------------+---------------------------------------------+
| Optional                                     | `relative_tolerance`__ | Tolerance for cross-validation error value  |
+----------------------------------------------+------------------------+---------------------------------------------+
| Optional                                     | `decrease_tolerance`__ | Tolerance for cross-validation error        |
|                                              |                        | stabilization                               |
+----------------------------------------------+------------------------+---------------------------------------------+
| Optional                                     | `max_rank`__           | Maximum subspace dimension to consider      |
+----------------------------------------------+------------------------+---------------------------------------------+
| Optional                                     | `exhaustive`__         | Assess all admissible subspace dimensions   |
+----------------------------------------------+------------------------+---------------------------------------------+

.. __: model-active_subspace-truncation_method-cross_validation-minimum.html
__ model-active_subspace-truncation_method-cross_validation-relative.html
__ model-active_subspace-truncation_method-cross_validation-decrease.html
__ model-active_subspace-truncation_method-cross_validation-relative_tolerance.html
__ model-active_subspace-truncation_method-cross_validation-decrease_tolerance.html
__ model-active_subspace-truncation_method-cross_validation-max_rank.html
__ model-active_subspace-truncation_method-cross_validation-exhaustive.html



**Description**


Select the subspace dimension that minimizes the 10-fold
cross-validation error when constructing a moving least squares
surrogate model.