rbmatlab  1.16.09
 All Classes Namespaces Files Functions Variables Modules Pages
List of all members | Public Member Functions | Public Attributes
ParameterSampling.Uniform Class Reference

Detailed Description

Parameter sampling class with uniformly distributed parameters in the parameter space.

Definition at line 18 of file Uniform.m.

Inheritance diagram for ParameterSampling.Uniform:
Inheritance graph
[legend]
Collaboration diagram for ParameterSampling.Uniform:
Collaboration graph
[legend]

Public Member Functions

 Uniform (init_numintervals, log_distribution)
 constructor More...
 
function  init_sample (IDetailedModel dmodel)
 initializes the parameter sampling object More...
 
function levels = get_refinement_levels ()
 returns the number of refinement steps for each sample element More...
 
function size = size ()
 returns the number of parameter vectors in this sampling More...
 
function this = refine (elems)
 refines the object at the the given elements More...
 
function cogs = cogs ()
 returns the center of gravity of sample cells (in case the sample can be viewed as a grid) More...
 
function skips = skipped_refinements ()
 determines for every sample element how often it was skipped during the last refinement steps. More...
 
function
max_vertex = 
elementwise_maximum_of_vertex_values (values)
 returns \(\max_{v \in V(e)} \text{val}_v\) for each sample cell \(e\) (in case the sample can be viewed as a grid with vertices as sample vectors) More...
 
- Public Member Functions inherited from ParameterSampling.IRefineable
function sampleget_sample_at_ref_level (ref_level)
 helper function returning the sample at a given refinement level More...
 
- Public Member Functions inherited from ParameterSampling.Interface
function
please_init = 
init_required ()
 returns a boolean indicating whether the object is fully functional, or must be initialized by a call to init_sample(). More...
 

Public Attributes

 sample
 matrix storing the parameter vectors in the parameter samples as row vectors. More...
 
 log_distribution = false
 boolean value determining whether a logarithmic function shall be applied on the uniform distribution of the parameter samples. More...
 
 init_numintervals = 2
 vector or scalar determining how many equally distant vectors shall be added to the parameter sample in each direction. More...
 
::cubegrid pgrid
 underlying grid
 
- Public Attributes inherited from ParameterSampling.IRefineable
 sample_history = {""}
 cell array of samples that are generated during refinement cycle More...
 
 refined_elements = {""}
 cell array of elements which are refined at each refinement step More...
 
- Public Attributes inherited from ParameterSampling.Interface
 sample
 matrix storing the parameter vectors in the parameter samples as row vectors. More...
 

Additional Inherited Members

- Protected Attributes inherited from ParameterSampling.Interface
 init_done = false
 boolean indicating whether the object has already been initialized by a call of init_sample(). More...
 

Constructor & Destructor Documentation

ParameterSampling.Uniform.Uniform (   init_numintervals,
  log_distribution 
)

constructor

Parameters
init_numintervalsvector or scalar determining how many equally distant vectors shall be added to the parameter sample in each direction.
log_distributionboolean indicator determining whether a logarithmic function shall be applied on the uniform distribution of the parameter samples.

Definition at line 78 of file Uniform.m.

Member Function Documentation

function cogs = ParameterSampling.Uniform.cogs ( )
virtual

returns the center of gravity of sample cells (in case the sample can be viewed as a grid)

Usually the center of gravity of grid cells is a new sample vector in case the element is refined.

Todo:
Maybe this should be renamed to refinement_candidates, because otherwise it is restricted to grids.
Return values
cogsa matrix with sample vectors as row vectors

Implements ParameterSampling.IRefineable.

Definition at line 180 of file Uniform.m.

function max_vertex = ParameterSampling.Uniform.elementwise_maximum_of_vertex_values (   values)
virtual

returns \(\max_{v \in V(e)} \text{val}_v\) for each sample cell \(e\) (in case the sample can be viewed as a grid with vertices as sample vectors)

Parameters
valuesvalues
Return values
max_vertexvector of maxima as described above

Implements ParameterSampling.IRefineable.

Definition at line 203 of file Uniform.m.

Here is the call graph for this function:

function levels = ParameterSampling.Uniform.get_refinement_levels ( )
virtual

returns the number of refinement steps for each sample element

Return values
levela vector with number of refinement steps for each sample element

Implements ParameterSampling.IRefineable.

Definition at line 142 of file Uniform.m.

function ParameterSampling.Uniform.init_sample ( IDetailedModel  dmodel)
virtual

initializes the parameter sampling object

Parameters
dmodelobject specifying how the high dimensional data can be computed.

Implements ParameterSampling.Interface.

Definition at line 99 of file Uniform.m.

function this = ParameterSampling.Uniform.refine (   elems)
virtual

refines the object at the the given elements

At each refinement, the refined elements and the previous sample is added to the cell arrays sample_history and refined_elements

Parameters
elemsA vector of element indices to be refined
Return values
thisthe updated object

Implements ParameterSampling.IRefineable.

Definition at line 164 of file Uniform.m.

Here is the call graph for this function:

function size = ParameterSampling.Uniform.size ( )
virtual

returns the number of parameter vectors in this sampling

Return values
sizesample size

Implements ParameterSampling.Interface.

Definition at line 153 of file Uniform.m.

Here is the caller graph for this function:

function skips = ParameterSampling.Uniform.skipped_refinements ( )
virtual

determines for every sample element how often it was skipped during the last refinement steps.

Return values
skipsa vector of the number of skips for each sample element.

Implements ParameterSampling.IRefineable.

Definition at line 191 of file Uniform.m.

Member Data Documentation

ParameterSampling.Uniform.init_numintervals = 2

vector or scalar determining how many equally distant vectors shall be added to the parameter sample in each direction.

If it is a vector, for each direction of the parameter space a different number can be given.


Default: 2

Definition at line 53 of file Uniform.m.

ParameterSampling.Uniform.log_distribution = false

boolean value determining whether a logarithmic function shall be applied on the uniform distribution of the parameter samples.


Default: false

Definition at line 43 of file Uniform.m.

ParameterSampling.Uniform.sample

matrix storing the parameter vectors in the parameter samples as row vectors.

Note
This property has the MATLAB attribute Dependent set to true.
This property has non-standard access specifiers: SetAccess = Private, GetAccess = Public
Matlab documentation of property attributes.
[readonly]

Definition at line 29 of file Uniform.m.


The documentation for this class was generated from the following file: