vector_parameter_study

Samples variables along a user-defined vector

Topics

parameter_studies

Specification

• Alias: None

• Arguments: None

Child Keywords:

Required/Optional	Description of Group	Dakota Keyword	Dakota Keyword Description
Required (Choose One)	Step Control	final_point	Final variable values defining vector in vector parameter study
		step_vector	Size of step for each variable
Required		num_steps	Number of sampling steps along the vector in a vector parameter study
Optional		model_pointer	Identifier for model block to be used by a method

Description

Dakota’s vector parameter study computes response data sets at selected intervals along a vector in parameter space. It is often used for single-coordinate parameter studies (to study the effect of a single variable on a response set), but it can be used more generally for multiple coordinate vector studies (to investigate the response variations along some n-dimensional vector such as an optimizer search direction).

Default Behavior

By default, the vector parameter study operates over all types of variables.

Expected Outputs

A vector parameter study produces a set of responses for each parameter set that is generated.

Expected HDF5 Output

If Dakota was built with HDF5 support and run with the environment-results_output-hdf5 keyword, this method writes the following results to HDF5:

• hdf5_results-pstudies

Usage Tips

Group 1 is used to define the vector along which the parameters are varied. Both cases also rely on the variables specification of an initial value, through:

• the variables-continuous_design-initial_point keyword

• the variables-continuous_state-initial_state keyword

• relying on the default initial value, based on the rest of the variables specification

From the initial value, the vector can be defined using one of the two keyword choices.

Once the vector is defined, the samples are then fully specifed by method-vector_parameter_study-num_steps.

Examples

The following example is a good comparison to the examples on method-multidim_parameter_study and method-centered_parameter_study.

# tested on Dakota 6.0 on 140501
environment
  tabular_data
    tabular_data_file = 'rosen_vector.dat'

method
  vector_parameter_study
    num_steps = 10
    final_point =    2.0      2.0
model
  single

variables
  continuous_design = 2
    initial_point =   -2.0     -2.0
    descriptors =     'x1'     "x2"

interface
  analysis_driver = 'rosenbrock'
    fork

responses
  response_functions = 1
  no_gradients
  no_hessians