![]() |
Dakota Reference Manual
Version 6.16
Explore and Predict with Confidence
|
(Experimental) Specify the fitness_metric used to select the next point
Alias: none
Argument(s): none
Default: predicted_variance
Child Keywords:
| Required/Optional | Description of Group | Dakota Keyword | Dakota Keyword Description | |
|---|---|---|---|---|
| Required (Choose One) | Fitness Metric (Group 1) | predicted_variance | Pick points with highest variance | |
| distance | Space filling metric | |||
| gradient | Fill the range space of the surrogate |
adaptive sampling is an experimental capability that is not ready for production use at this time.
The user can specify the fitness_metric used to select the next point (or points) to evaluate and add to the set. The fitness metrics used for scoring candidate points include:
predicted_variance distance gradient