step_vector

Size of step for each variable

Specification

  • Alias: None

  • Arguments: None

  • JSON argument key: None

Description

The step_vector keyword specifies how much each variable will be incremented in a single step.

step_vector works in conjunction with num_steps, which determines the number of steps taken during the vector_parameter_study. If instead of step_vector, final_point is specified with num_steps, Dakota will infer the step sizes.

Entries in the step_vector are the actual amounts by which continuous and range variables are incremented. For set variables, step_vector entries are interpreted as indexes into the underlying set.

Default Behavior

The user is required to specify either final_point or step_vector. There is no default definition for the vector.

Examples

variables
  continuous_design 1
    initial_point 1.0
    descriptors 'x1'
  discrete_design_set
    string 1
     elements 'bar' 'baz' 'foo' 'fuzz'
     initial_point 'bar'
     descriptors 's1'

method
  vector_parameter_study
    num_steps = 3
    # Add 2.0 to x1 and increment s1 by 1 element in each step
    step_vector =    2.0      1
{
  "variables": [
    {
      "continuous_design": {
        "initial_point": [
          1.0
        ],
        "descriptors": [
          "x1"
        ],
        "count": 1
      },
      "discrete_design_set": {
        "string": {
          "count": 1,
          "elements": [
            "bar",
            "baz",
            "foo",
            "fuzz"
          ],
          "initial_point": [
            "bar"
          ],
          "descriptors": [
            "s1"
          ]
        }
      }
    }
  ],
  "method": [
    {
      "vector_parameter_study": {
        "num_steps": 3,
        "step_control": {
          "step_vector": [
            2.0,
            1
          ]
        }
      }
    }
  ]
}