Difference between revisions of "Config:SequentialDesign"
From SUMOwiki
Jump to navigationJump to searchm |
m |
||
Line 1: | Line 1: | ||
== SampleSelector == | == SampleSelector == | ||
− | + | ||
=== empty === | === empty === | ||
Dont select any new samples, useful when modeling multiple outputs in paralel | Dont select any new samples, useful when modeling multiple outputs in paralel | ||
− | <source lang="xml"> | + | <source xmlns:saxon="http://icl.com/saxon" lang="xml"> |
<[[Config:SampleSelector|SampleSelector]] type="[[SampleSelector#EmptySampleSelector|EmptySampleSelector]]" combineOutputs="false"/> | <[[Config:SampleSelector|SampleSelector]] type="[[SampleSelector#EmptySampleSelector|EmptySampleSelector]]" combineOutputs="false"/> | ||
</source> | </source> | ||
=== random === | === random === | ||
Each sampling iterations new samples are selected randomly | Each sampling iterations new samples are selected randomly | ||
− | <source lang="xml"> | + | <source xmlns:saxon="http://icl.com/saxon" lang="xml"> |
<[[Config:SampleSelector|SampleSelector]] type="[[SampleSelector#RandomSampleSelector|RandomSampleSelector]]" combineOutputs="false"/> | <[[Config:SampleSelector|SampleSelector]] type="[[SampleSelector#RandomSampleSelector|RandomSampleSelector]]" combineOutputs="false"/> | ||
</source> | </source> | ||
=== combo === | === combo === | ||
Allows you combine multiple sample selector algorithms | Allows you combine multiple sample selector algorithms | ||
− | <source lang="xml"> | + | <source xmlns:saxon="http://icl.com/saxon" lang="xml"> |
<[[Config:SampleSelector|SampleSelector]] type="[[SampleSelector#CombinedSampleSelector|CombinedSampleSelector]]" combineOutputs="false"> | <[[Config:SampleSelector|SampleSelector]] type="[[SampleSelector#CombinedSampleSelector|CombinedSampleSelector]]" combineOutputs="false"> | ||
− | <[[Config:SampleSelector|SampleSelector]] type="[[SampleSelector#RationalPoleSuppressionSampleSelector|RationalPoleSuppressionSampleSelector]]" combineOutputs="false"><!-- Currently no options are available, if the model is a rational model, and a | + | <[[Config:SampleSelector|SampleSelector]] type="[[SampleSelector#RationalPoleSuppressionSampleSelector|RationalPoleSuppressionSampleSelector]]" combineOutputs="false"> |
− | + | <!-- Currently no options are available, if the model is a rational model, and a | |
− | <[[Config:SampleSelector|SampleSelector]] type="[[SampleSelector#DelaunaySampleSelector|DelaunaySampleSelector]]" combineOutputs="false"><!-- One of all, data --> | + | - - pole is found (using a simple search strategy), the pole is returned --> |
− | <Option key="sampleSelect" value="all"/><!-- Integer between 2 and 20 --> | + | </[[Config:SampleSelector|SampleSelector]]> |
− | <Option key="nLastModels" value="2"/><!-- One of densityBased, differenceBased, weightedLinear, weightedGeometric --> | + | |
− | <Option key="scoreFunction" value="weightedLinear"/><!-- Weighting for weightedLinear --> | + | <[[Config:SampleSelector|SampleSelector]] type="[[SampleSelector#DelaunaySampleSelector|DelaunaySampleSelector]]" combineOutputs="false"> |
− | <Option key="lambda" value="0.5"/><!-- Weighting for weightedGeometric --> | + | <!-- One of all, data --> |
− | <Option key="mu" value="0.5"/><!-- One of none, max, cap, capmax --> | + | <Option key="sampleSelect" value="all"/> |
+ | <!-- Integer between 2 and 20 --> | ||
+ | <Option key="nLastModels" value="2"/> | ||
+ | <!-- One of densityBased, differenceBased, weightedLinear, weightedGeometric --> | ||
+ | <Option key="scoreFunction" value="weightedLinear"/> | ||
+ | <!-- Weighting for weightedLinear --> | ||
+ | <Option key="lambda" value="0.5"/> | ||
+ | <!-- Weighting for weightedGeometric --> | ||
+ | <Option key="mu" value="0.5"/> | ||
+ | <!-- One of none, max, cap, capmax --> | ||
<Option key="volumeScaling" value="max"/> | <Option key="volumeScaling" value="max"/> | ||
− | <Option key="differenceScaling" value="capmax"/><!-- Boolean flag, if set all points closer than snapThreshold to the boundary of | + | <Option key="differenceScaling" value="capmax"/> |
− | + | <!-- Boolean flag, if set all points closer than snapThreshold to the boundary of | |
+ | - - the domain are clipped to the boundary --> | ||
<Option key="snapToEdge" value="enable"/> | <Option key="snapToEdge" value="enable"/> | ||
<Option key="snapThreshold" value=".05"/> | <Option key="snapThreshold" value=".05"/> | ||
Line 33: | Line 43: | ||
=== delaunay === | === delaunay === | ||
An adaptive sample selection algorithm that does a trade-off between error and density | An adaptive sample selection algorithm that does a trade-off between error and density | ||
− | <source lang="xml"> | + | <source xmlns:saxon="http://icl.com/saxon" lang="xml"> |
− | <[[Config:SampleSelector|SampleSelector]] type="[[SampleSelector#DelaunaySampleSelector|DelaunaySampleSelector]]" combineOutputs="false"><!-- One of all, data --> | + | <[[Config:SampleSelector|SampleSelector]] type="[[SampleSelector#DelaunaySampleSelector|DelaunaySampleSelector]]" combineOutputs="false"> |
− | <Option key="sampleSelect" value="all"/><!-- Integer between 2 and 20 --> | + | <!-- One of all, data --> |
− | <Option key="nLastModels" value="2"/><!-- One of densityBased, differenceBased, weightedLinear, weightedGeometric --> | + | <Option key="sampleSelect" value="all"/> |
+ | <!-- Integer between 2 and 20 --> | ||
+ | <Option key="nLastModels" value="2"/> | ||
+ | <!-- One of densityBased, differenceBased, weightedLinear, weightedGeometric --> | ||
<Option key="scoreFunction" value="weightedLinear"/> | <Option key="scoreFunction" value="weightedLinear"/> | ||
<Option key="lambda" value="0.5"/> | <Option key="lambda" value="0.5"/> | ||
− | <Option key="mu" value="0.5"/><!-- One of none, max, cap, capmax --> | + | <Option key="mu" value="0.5"/> |
+ | <!-- One of none, max, cap, capmax --> | ||
<Option key="volumeScaling" value="max"/> | <Option key="volumeScaling" value="max"/> | ||
− | <Option key="differenceScaling" value="capmax"/><!-- Boolean flag --> | + | <Option key="differenceScaling" value="capmax"/> |
+ | <!-- Boolean flag --> | ||
<Option key="snapToEdge" value="enable"/> | <Option key="snapToEdge" value="enable"/> | ||
<Option key="snapThreshold" value=".2"/> | <Option key="snapThreshold" value=".2"/> | ||
Line 48: | Line 63: | ||
=== density === | === density === | ||
A simple density based sample selection algorithm | A simple density based sample selection algorithm | ||
− | <source lang="xml"> | + | <source xmlns:saxon="http://icl.com/saxon" lang="xml"> |
<[[Config:SampleSelector|SampleSelector]] type="[[SampleSelector#DensitySampleSelector|DensitySampleSelector]]" combineOutputs="false"/> | <[[Config:SampleSelector|SampleSelector]] type="[[SampleSelector#DensitySampleSelector|DensitySampleSelector]]" combineOutputs="false"/> | ||
</source> | </source> | ||
=== error === | === error === | ||
An adaptive sample selection algorithm (error based), driven by the evaluation of your model on a dense grid | An adaptive sample selection algorithm (error based), driven by the evaluation of your model on a dense grid | ||
− | <source lang="xml"> | + | <source xmlns:saxon="http://icl.com/saxon" lang="xml"> |
− | <[[Config:SampleSelector|SampleSelector]] type="[[SampleSelector#ErrorSampleSelector|ErrorSampleSelector]]" combineOutputs="false"><!-- Integer between 2 and 20 --> | + | <[[Config:SampleSelector|SampleSelector]] type="[[SampleSelector#ErrorSampleSelector|ErrorSampleSelector]]" combineOutputs="false"> |
− | <Option key="nLastModels" value="4"/><!-- One of none, max, cap, capmax --> | + | <!-- Integer between 2 and 20 --> |
− | <Option key="differenceScaling" value="none"/><!-- Gridsize to evaluate on, one of int or array of dimension length --> | + | <Option key="nLastModels" value="4"/> |
− | <Option key="gridSize" value="50"/><!-- Maximum total points to evaluate, distributed over dimensions --> | + | <!-- One of none, max, cap, capmax --> |
− | <Option key="maxGridSize" value="100000"/><!-- Closeness threshold, Double --> | + | <Option key="differenceScaling" value="none"/> |
− | <Option key="closenessThreshold" value="0.2"/><!-- Set a % of the maximumSamples to randomly chosen --> | + | <!-- Gridsize to evaluate on, one of int or array of dimension length --> |
+ | <Option key="gridSize" value="50"/> | ||
+ | <!-- Maximum total points to evaluate, distributed over dimensions --> | ||
+ | <Option key="maxGridSize" value="100000"/> | ||
+ | <!-- Closeness threshold, Double --> | ||
+ | <Option key="closenessThreshold" value="0.2"/> | ||
+ | <!-- Set a % of the maximumSamples to randomly chosen --> | ||
<Option key="randomPercentage" value="20"/> | <Option key="randomPercentage" value="20"/> | ||
</[[Config:SampleSelector|SampleSelector]]> | </[[Config:SampleSelector|SampleSelector]]> | ||
Line 65: | Line 86: | ||
=== gradient === | === gradient === | ||
A highly adaptive sampling algorithm, error and density based | A highly adaptive sampling algorithm, error and density based | ||
− | <source lang="xml"> | + | <source xmlns:saxon="http://icl.com/saxon" lang="xml"> |
− | <[[Config:SampleSelector|SampleSelector]] type="[[SampleSelector#GradientSampleSelector|GradientSampleSelector]]" combineOutputs="false"><!-- Integer between 2 and 20 --> | + | <[[Config:SampleSelector|SampleSelector]] type="[[SampleSelector#GradientSampleSelector|GradientSampleSelector]]" combineOutputs="false"> |
+ | <!-- Integer between 2 and 20 --> | ||
<Option key="neighbourhoodSize" value="2"/> | <Option key="neighbourhoodSize" value="2"/> | ||
</[[Config:SampleSelector|SampleSelector]]> | </[[Config:SampleSelector|SampleSelector]]> | ||
Line 72: | Line 94: | ||
=== isc === | === isc === | ||
A sampling algorithm aimed at optimization problems | A sampling algorithm aimed at optimization problems | ||
− | <source lang="xml"> | + | <source xmlns:saxon="http://icl.com/saxon" lang="xml"> |
− | <[[Config:SampleSelector|SampleSelector]] type="[[SampleSelector#InfillSamplingCriterion|InfillSamplingCriterion]]" combineOutputs="false"><!-- A criterion determines the interesting regions to sample --><!-- Choose 1 from the following: --> | + | <[[Config:SampleSelector|SampleSelector]] type="[[SampleSelector#InfillSamplingCriterion|InfillSamplingCriterion]]" combineOutputs="false"> |
− | <Option key="criterion" value="gei"/><!-- generalized expected improvement --> | + | <!-- A criterion determines the interesting regions to sample --> |
− | <Option key="g" value="1"/><!-- balanced local-global search --><!--<Option key="criterion" value="wei" />--><!-- weighted expected improvement --><!--<Option key="w" value="0.5" />--><!-- weight, 0 is global search, 1 is local search --><!--<Option key="criterion" value="ei" />--><!-- expected improvement --><!--<Option key="criterion" value="kushner" />--><!-- kushner --><!--<Option key="eps" value="0.001" />--><!--<Option key="criterion" value="lcb" />--><!-- lower confidence bound function --><!--<Option key="lcb" value="2.0" />--><!--<Option key="criterion" value="maxvar" />--><!-- maximizes variation --><!-- Watson and Barnes criterions --><!--<Option key="criterion" value="wb1" />--><!-- threshold-bounded extreme --><!--<Option key="criterion" value="wb2" />--><!-- regional extreme --><!--<Option key="criterion" value="crowdedness" />--><!-- crowdedness function --><!-- This criterion has to be solved to choose new samples, one can choose the optimizer used here --> | + | <!-- Choose 1 from the following: --> |
+ | <Option key="criterion" value="gei"/> <!-- generalized expected improvement --> | ||
+ | <Option key="g" value="1"/> <!-- balanced local-global search --> | ||
+ | |||
+ | <!--<Option key="criterion" value="wei" />--> <!-- weighted expected improvement --> | ||
+ | <!--<Option key="w" value="0.5" />--> <!-- weight, 0 is global search, 1 is local search --> | ||
+ | |||
+ | <!--<Option key="criterion" value="ei" />--> <!-- expected improvement --> | ||
+ | |||
+ | <!--<Option key="criterion" value="kushner" />--> <!-- kushner --> | ||
+ | <!--<Option key="eps" value="0.001" />--> | ||
+ | |||
+ | <!--<Option key="criterion" value="lcb" />--> <!-- lower confidence bound function --> | ||
+ | <!--<Option key="lcb" value="2.0" />--> | ||
+ | <!--<Option key="criterion" value="maxvar" />--> <!-- maximizes variation --> | ||
+ | |||
+ | <!-- Watson and Barnes criterions --> | ||
+ | <!--<Option key="criterion" value="wb1" />--> <!-- threshold-bounded extreme --> | ||
+ | <!--<Option key="criterion" value="wb2" />--> <!-- regional extreme --> | ||
+ | |||
+ | <!--<Option key="criterion" value="crowdedness" />--> <!-- crowdedness function --> | ||
+ | |||
+ | <!-- This criterion has to be solved to choose new samples, one can choose the optimizer used here --> | ||
<[[Config:Optimizer|Optimizer]] type="[[Optimizer#DirectOptimizer|DirectOptimizer]]"> | <[[Config:Optimizer|Optimizer]] type="[[Optimizer#DirectOptimizer|DirectOptimizer]]"> | ||
<Option key="maxevals" value="1000"/> | <Option key="maxevals" value="1000"/> | ||
<Option key="maxits" value="300"/> | <Option key="maxits" value="300"/> | ||
− | </[[Config:Optimizer|Optimizer]]><!-- | + | </[[Config:Optimizer|Optimizer]]> |
− | + | ||
− | + | <!-- | |
− | + | <[[Config:Optimizer|Optimizer]] type="[[Optimizer#MatlabGA|MatlabGA]]"> | |
− | <Option key="debug" value=" | + | </[[Config:Optimizer|Optimizer]]> |
+ | --> | ||
+ | |||
+ | <!-- | ||
+ | when debug is 'on' a contour plot of the ISC function is drawn every iteration. | ||
+ | Together with the current samples and the chosen samples | ||
+ | --> | ||
+ | <Option key="debug" value="on"/> | ||
</[[Config:SampleSelector|SampleSelector]]> | </[[Config:SampleSelector|SampleSelector]]> | ||
</source> | </source> |
Revision as of 13:56, 13 February 2008
SampleSelector
empty
Dont select any new samples, useful when modeling multiple outputs in paralel
<[[Config:SampleSelector|SampleSelector]] type="[[SampleSelector#EmptySampleSelector|EmptySampleSelector]]" combineOutputs="false"/>
random
Each sampling iterations new samples are selected randomly
<[[Config:SampleSelector|SampleSelector]] type="[[SampleSelector#RandomSampleSelector|RandomSampleSelector]]" combineOutputs="false"/>
combo
Allows you combine multiple sample selector algorithms
<[[Config:SampleSelector|SampleSelector]] type="[[SampleSelector#CombinedSampleSelector|CombinedSampleSelector]]" combineOutputs="false">
<[[Config:SampleSelector|SampleSelector]] type="[[SampleSelector#RationalPoleSuppressionSampleSelector|RationalPoleSuppressionSampleSelector]]" combineOutputs="false">
<!-- Currently no options are available, if the model is a rational model, and a
- - pole is found (using a simple search strategy), the pole is returned -->
</[[Config:SampleSelector|SampleSelector]]>
<[[Config:SampleSelector|SampleSelector]] type="[[SampleSelector#DelaunaySampleSelector|DelaunaySampleSelector]]" combineOutputs="false">
<!-- One of all, data -->
<Option key="sampleSelect" value="all"/>
<!-- Integer between 2 and 20 -->
<Option key="nLastModels" value="2"/>
<!-- One of densityBased, differenceBased, weightedLinear, weightedGeometric -->
<Option key="scoreFunction" value="weightedLinear"/>
<!-- Weighting for weightedLinear -->
<Option key="lambda" value="0.5"/>
<!-- Weighting for weightedGeometric -->
<Option key="mu" value="0.5"/>
<!-- One of none, max, cap, capmax -->
<Option key="volumeScaling" value="max"/>
<Option key="differenceScaling" value="capmax"/>
<!-- Boolean flag, if set all points closer than snapThreshold to the boundary of
- - the domain are clipped to the boundary -->
<Option key="snapToEdge" value="enable"/>
<Option key="snapThreshold" value=".05"/>
</[[Config:SampleSelector|SampleSelector]]>
</[[Config:SampleSelector|SampleSelector]]>
delaunay
An adaptive sample selection algorithm that does a trade-off between error and density
<[[Config:SampleSelector|SampleSelector]] type="[[SampleSelector#DelaunaySampleSelector|DelaunaySampleSelector]]" combineOutputs="false">
<!-- One of all, data -->
<Option key="sampleSelect" value="all"/>
<!-- Integer between 2 and 20 -->
<Option key="nLastModels" value="2"/>
<!-- One of densityBased, differenceBased, weightedLinear, weightedGeometric -->
<Option key="scoreFunction" value="weightedLinear"/>
<Option key="lambda" value="0.5"/>
<Option key="mu" value="0.5"/>
<!-- One of none, max, cap, capmax -->
<Option key="volumeScaling" value="max"/>
<Option key="differenceScaling" value="capmax"/>
<!-- Boolean flag -->
<Option key="snapToEdge" value="enable"/>
<Option key="snapThreshold" value=".2"/>
</[[Config:SampleSelector|SampleSelector]]>
density
A simple density based sample selection algorithm
<[[Config:SampleSelector|SampleSelector]] type="[[SampleSelector#DensitySampleSelector|DensitySampleSelector]]" combineOutputs="false"/>
error
An adaptive sample selection algorithm (error based), driven by the evaluation of your model on a dense grid
<[[Config:SampleSelector|SampleSelector]] type="[[SampleSelector#ErrorSampleSelector|ErrorSampleSelector]]" combineOutputs="false">
<!-- Integer between 2 and 20 -->
<Option key="nLastModels" value="4"/>
<!-- One of none, max, cap, capmax -->
<Option key="differenceScaling" value="none"/>
<!-- Gridsize to evaluate on, one of int or array of dimension length -->
<Option key="gridSize" value="50"/>
<!-- Maximum total points to evaluate, distributed over dimensions -->
<Option key="maxGridSize" value="100000"/>
<!-- Closeness threshold, Double -->
<Option key="closenessThreshold" value="0.2"/>
<!-- Set a % of the maximumSamples to randomly chosen -->
<Option key="randomPercentage" value="20"/>
</[[Config:SampleSelector|SampleSelector]]>
gradient
A highly adaptive sampling algorithm, error and density based
<[[Config:SampleSelector|SampleSelector]] type="[[SampleSelector#GradientSampleSelector|GradientSampleSelector]]" combineOutputs="false">
<!-- Integer between 2 and 20 -->
<Option key="neighbourhoodSize" value="2"/>
</[[Config:SampleSelector|SampleSelector]]>
isc
A sampling algorithm aimed at optimization problems
<[[Config:SampleSelector|SampleSelector]] type="[[SampleSelector#InfillSamplingCriterion|InfillSamplingCriterion]]" combineOutputs="false">
<!-- A criterion determines the interesting regions to sample -->
<!-- Choose 1 from the following: -->
<Option key="criterion" value="gei"/> <!-- generalized expected improvement -->
<Option key="g" value="1"/> <!-- balanced local-global search -->
<!--<Option key="criterion" value="wei" />--> <!-- weighted expected improvement -->
<!--<Option key="w" value="0.5" />--> <!-- weight, 0 is global search, 1 is local search -->
<!--<Option key="criterion" value="ei" />--> <!-- expected improvement -->
<!--<Option key="criterion" value="kushner" />--> <!-- kushner -->
<!--<Option key="eps" value="0.001" />-->
<!--<Option key="criterion" value="lcb" />--> <!-- lower confidence bound function -->
<!--<Option key="lcb" value="2.0" />-->
<!--<Option key="criterion" value="maxvar" />--> <!-- maximizes variation -->
<!-- Watson and Barnes criterions -->
<!--<Option key="criterion" value="wb1" />--> <!-- threshold-bounded extreme -->
<!--<Option key="criterion" value="wb2" />--> <!-- regional extreme -->
<!--<Option key="criterion" value="crowdedness" />--> <!-- crowdedness function -->
<!-- This criterion has to be solved to choose new samples, one can choose the optimizer used here -->
<[[Config:Optimizer|Optimizer]] type="[[Optimizer#DirectOptimizer|DirectOptimizer]]">
<Option key="maxevals" value="1000"/>
<Option key="maxits" value="300"/>
</[[Config:Optimizer|Optimizer]]>
<!--
<[[Config:Optimizer|Optimizer]] type="[[Optimizer#MatlabGA|MatlabGA]]">
</[[Config:Optimizer|Optimizer]]>
-->
<!--
when debug is 'on' a contour plot of the ISC function is drawn every iteration.
Together with the current samples and the chosen samples
-->
<Option key="debug" value="on"/>
</[[Config:SampleSelector|SampleSelector]]>