![]() |
Dakota Reference Manual
Version 6.16
Explore and Predict with Confidence
|
Truncate the subspace to minimize surrogate cross-validation error
Alias: none
Argument(s): none
Default: relative
Child Keywords:
| Required/Optional | Description of Group | Dakota Keyword | Dakota Keyword Description | |
|---|---|---|---|---|
| Optional (Choose One) | CV Selection Criterion (Group 1) | minimum | Select subspace to minimize 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 | ||
Select the subspace dimension that minimizes the 10-fold cross-validation error when constructing a moving least squares surrogate model.