Symmetric Latin hypercube design.
More...
#include <experimental_design.h>
Symmetric Latin hypercube design.
Symmetric Latin hypercube sampling are popular for generating near-random samples of parameter values from a multidimensional distribution. The Symmetric Latin hypercube does better than the original Latin hypercube when it comes to entropy and maximin and is the experimental design of choice for surrogate optimization. Due to rank-deficiencies it's recommended to use 2*dim points to ensure that the sample has rank dim.
- Author
- David Eriksson, dme65.nosp@m.@cor.nosp@m.nell..nosp@m.edu
| sot::SymmetricLatinHypercube::SymmetricLatinHypercube |
( |
int |
numPoints, |
|
|
int |
dim |
|
) |
| |
|
inline |
Constructor.
- Parameters
-
| numPoints | Number of points in the experimental design |
| dim | Number of dimensions |
| int sot::SymmetricLatinHypercube::dim |
( |
| ) |
const |
|
inlinevirtual |
Method for getting the number of dimensions.
- Returns
- The number of dimensions
Implements sot::ExpDesign.
| mat sot::SymmetricLatinHypercube::generatePoints |
( |
| ) |
const |
|
inlinevirtual |
Method that generates a symmetric Latin hypercube design.
- Returns
- A symmetric Latin hypercube design
Implements sot::ExpDesign.
| int sot::SymmetricLatinHypercube::numPoints |
( |
| ) |
const |
|
inlinevirtual |
Method for getting the number of points in the experimental design.
- Returns
- The number of points
Implements sot::ExpDesign.
| int sot::SymmetricLatinHypercube::mDim |
|
protected |
| int sot::SymmetricLatinHypercube::mNumPoints |
|
protected |
Number of points in the experimental design
The documentation for this class was generated from the following file:
1.8.11
