Dakota Reference Manual  Version 6.12
Explore and Predict with Confidence
 All Pages
primary_scale_types


How to scale each objective function

Specification

Alias: objective_function_scale_types

Argument(s): STRINGLIST

Default: no scaling

Description

Each string in primary_scale_types indicates the scaling type for each objective function. They only have effect when the associated method specifies scaling.

The options are:

  • 'value' - characteristic value by which response functions will be divided. If this is chosen, then primary_scales must also be specified; 'value' is assumed if scales are given without scale_types
  • 'log' - logarithmic scaling (can be used together with primary_scales, which can be helpfully used to negate values prior to log transformation)

(Automatic scaling is disallowed due to lack of bounds or targets for primary responses.)

If a single string is specified it will apply to all objective_functions. Otherwise the length of primary_scale_types must have length equal to objective_functions. Thus, when scalar and/or field responses are specified, the number of primary_scale_types must equal the number of scalars plus the number of fields, not the total elements in the fields.

Usage Tips:

When scaling is active, it is applied to objective functions prior to any weights and multi-objective sum formation.

See the scaling information under specific methods, e.g., method-*-scaling for details on how to use this keyword.