![]() |
Dakota Reference Manual
Version 6.16
Explore and Predict with Confidence
|
Shows mesh adaptive search's internally held list of all evaluations
Alias: none
Argument(s): none
Default: false
If set, display_all_evaluations will instruct the mesh adaptive direct search method to print out its own record of all evaluations. The information reported may be controlled using display_format.
Default Behavior
By default, mesh adaptive direct search does not report information on all evaluations, only on those for which an improvement in the objective function is found.
Expected Outputs
The information specified by display_format will be reported to the screen for every function evaluation.
Usage Tips
This will most likely only be useful for power users who want to understand and/or report more detailed information on method behavior.
The following example shows the syntax for specifying display_all_evaluations.
method
mesh_adaptive_search
display_all_evaluations
max_function evaluations=20
seed = 1234
Note that the output below reports information (default for display_format) for all function evaluations.
MADS run {
BBE OBJ
1 17.0625000000
2 1.0625000000
3 1297.0625000000
4 257.0625000000
5 81.0625000000
6 151.0625000000
7 1051.0625000000
8 40.0625000000
9 17.0625000000
10 40.0625000000
11 1.0625000000
12 102.0625000000
13 0.0625000000
14 231.0625000000
15 16.0625000000
16 5.0625000000
17 16.0625000000
18 71.0625000000
19 0.0625000000
20 1.0625000000
} end of run (max number of blackbox evaluations)
blackbox evaluations : 20
best feasible solution : ( 1 0.5 1 ) h=0 f=0.0625
That is in contrast with what would be reported by default.
MADS run {
BBE OBJ
1 17.0625000000
2 1.0625000000
13 0.0625000000
20 0.0625000000
} end of run (max number of blackbox evaluations)
blackbox evaluations : 20
best feasible solution : ( 1 0.5 1 ) h=0 f=0.0625
These keywords may also be of interest: