The SUMO toolbox  2018a
 All Classes Namespaces Files Functions Variables Typedefs Enumerations Enumerator Pages
Public Member Functions | Public Attributes | List of all members
CandidateRanker Class Reference

An interface that allows the object to score a set of candidates according to its own system. More...

Inheritance diagram for CandidateRanker:
Inheritance graph
[legend]

Public Member Functions

function scoreCandidates (var candidates, var state)
 
function getDesignType ()
 
function initNewSamples (var state)
 Initialized when new samples arrive. More...
 
function CandidateRanker (var ip, var varargin)
 This constructor will look for an option specifying a generator function to call. More...
 
function score (var candidates, var state)
 Scores a set of candidate samples. More...
 
function scoreMinimize (var candidates, var state)
 Simply calls -CandidateRanker.score(candidates, state) More...
 
function setOrder (var order)
 
function getType ()
 
function getDimension ()
 
function instantiate (var inDim, var varargin)
 
function plotRanker (var state, var newsamples)
 

Public Attributes

var inDim
 
var scalingFunction
 
var sortOrder
 
var debug
 
var debugSave
 
var debugPlot
 

Detailed Description

An interface that allows the object to score a set of candidates according to its own system.

This is used in the PipelineSequentialDesign to score a set of externally generated candidate samples according to some criterion. Note: when a candidate ranker gives a HIGHER score to a candidate, the sample selector will prefer this candidate above one with a LOWER score.

Todo:
dscores (derivatives) are not scaled in CandidateRanker.score

Constructor & Destructor Documentation

function CandidateRanker ( var  ip,
var  varargin 
)
inline

This constructor will look for an option specifying a generator function to call.

If such an option is found, the samples are generated from this function. Otherwise it is left to the subclass.

Return values
instanceof the class.

Member Function Documentation

function getDesignType ( )
function getDimension ( )
inline
function getType ( )
inline
function initNewSamples ( var  state)
inline

Initialized when new samples arrive.

function instantiate ( var  inDim,
var  varargin 
)
inline
function plotRanker ( var  state,
var  newsamples 
)
inline
function score ( var  candidates,
var  state 
)
inline

Scores a set of candidate samples.

Parameters
candidatesmatrix of candidate samples
statecurrent state
Return values
scoresvector of scores (priorites)
dscoresderivatives of scores w.r.t. candidates
function scoreCandidates ( var  candidates,
var  state 
)
function scoreMinimize ( var  candidates,
var  state 
)
inline

Simply calls -CandidateRanker.score(candidates, state)

This is needed primarily for OptimizeCriterion, as multiple output arguments don't work with: func = -this.score(...)

Parameters
candidatesmatrix of candidate samples
statecurrent state
Return values
scoresvector of -scores (lower is better)
dscoresderivatives of -scores w.r.t. candidates
function setOrder ( var  order)
inline

Member Data Documentation

var debug
var debugPlot
var debugSave
var inDim
var scalingFunction
var sortOrder

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