ParamStudy Class Reference

Class for vector, list, centered, and multidimensional parameter studies. More...

Inheritance diagram for ParamStudy:

PStudyDACE Analyzer Iterator List of all members.

Public Member Functions

 ParamStudy (Model &model)
 constructor
 ~ParamStudy ()
 destructor
void extract_trends ()
 Redefines the run_iterator virtual function for the PStudy/DACE branch.

Private Member Functions

void compute_vector_steps ()
 and either numSteps or stepLength (pStudyType is 1 or 2)
void sample (const RealVectorArray &list_of_points)
 performs the parameter study by sampling from a list of points
void vector_loop (const RealVector &start, const RealVector &step_vect, const int &num_steps)
 increments of step_vect. Total number of evaluations is num_steps + 1.
void centered_loop (const RealVector &start, const Real &percent_delta, const int &deltas_per_variable)
 centered about start
void multidim_loop (const IntArray &var_partitions)
 performs vector_loops recursively in multiple dimensions

Private Attributes

RealVectorArray listOfPoints
 array of evaluation points for the list_parameter_study
RealVector initialPoint
 the starting point for vector and centered parameter studies
RealVector finalPoint
 the ending point for vector_parameter_study (a specification option)
RealVector stepVector
 the n-dimensional increment in vector_parameter_study
int numSteps
 the number of times stepVector is applied in vector_parameter_study
short pStudyType
 (different vector specifications), 4 (centered), or 5 (multidim)
int deltasPerVariable
 variable in a centered_parameter_study
Real stepLength
 (a specification option)
Real percentDelta
 centered_parameter_study
IntArray variablePartitions
 number of partitions for each variable in a multidim_parameter_study

Detailed Description

Class for vector, list, centered, and multidimensional parameter studies.

The ParamStudy class contains several algorithms for performing parameter studies of different types. It is not a wrapper for an external library, rather its algorithms are self-contained. The vector parameter study steps along an n-dimensional vector from an arbitrary initial point to an arbitrary final point in a specified number of steps. The centered parameter study performs a number of plus and minus offsets in each coordinate direction around a center point. A multidimensional parameter study fills an n-dimensional hypercube based on a specified number of intervals for each dimension. It is a nested study in that it utilizes the vector parameter study internally as it recurses through the variables. And the list parameter study provides for a user specification of a list of points to evaluate, which allows general parameter investigations not fitting the structure of vector, centered, or multidim parameter studies.


The documentation for this class was generated from the following files:
Generated on Wed Nov 5 19:54:07 2008 for DAKOTA by  doxygen 1.5.1