![]() |
Dakota Reference Manual
Version 6.15
Explore and Predict with Confidence
|
Refine until the estimated covergence order converges
Alias: none
Argument(s): none
The converge_order
option is initialized using the estimate_order
aproach, and additional refinements are performed along the refinement path until the convergence order estimates converge (two-norm of the change in response orders is less than the convergence tolerance).