InitialDesign - Revision history

Admin at 14:41, 2 October 2009

2009-10-02T14:41:48Z

← Older revision		Revision as of 14:41, 2 October 2009
Line 1:		Line 1:
	== Latin Hypercube Design ==		== Latin Hypercube Design ==
	Choose an initial sampleset in such a way that they form a latin hypercube.		Choose an initial sampleset in such a way that they form a latin hypercube. This initial design will first try to download a pre-optimized maximin Latin hypercube from the website [http://www.spacefillingdesigns.nl]. If such a design is not available, or no proper connection to the website can be made, the Latin hypercube is generated by the toolbox. Note that this Latin hypercube may be suboptimal, and may leave considerable gaps in the design space.
	{{OptionsHeader}}		{{OptionsHeader}}
	{{Option		{{Option

Kcrombec at 09:55, 16 January 2009

2009-01-16T09:55:03Z

← Older revision		Revision as of 09:55, 16 January 2009
Line 11:		Line 11:
	== Dataset Design ==		== Dataset Design ==
	Reads an initial design from a dataset. Each row should contain one sample. The inputvalues come first. The outputvalues, if there are any, come second.		Reads an initial design from a dataset. Each row should contain one sample. The inputvalues come first. The outputvalues, if there are any, come second.

			If only input values are supplied (hasOutputs=false), each value has to be in the fixed [-1,1] range (model space). The samples are later automatically transformed to their original range before they are passed to the sample evaluator. This is done to easily support different ranges for variables without having to change the initial dataset design.

			If, however, output values are also provided (hasOutputs=true), the samples are instantly fed to the toolbox for modelling, since they don't have to be evaluated anymore. Because of this property, the inputs have to be in the original range defined in the simulator xml file, instead of the fixed [-1,1] range.

	{{OptionsHeader}}		{{OptionsHeader}}

Icouckuy: Config:InitialDesign backup moved to InitialDesign: page generated by script should link to this custom page

2008-07-17T13:31:54Z

Config:InitialDesign backup moved to InitialDesign: page generated by script should link to this custom page

← Older revision	Revision as of 13:31, 17 July 2008
(No difference)

Icouckuy: Config:InitialDesign moved to Config:InitialDesign backup: initialdesign is out of sumo tag, so scripts generate a page for it

2008-07-17T13:30:22Z

Config:InitialDesign moved to Config:InitialDesign backup: initialdesign is out of sumo tag, so scripts generate a page for it

← Older revision	Revision as of 13:30, 17 July 2008
(No difference)

Dgorissen at 11:46, 9 July 2008

2008-07-09T11:46:06Z

@@ Line 7: / Line 7: @@
 |default     = 30
 |description = The amount of samples that will be taken.
 }}
@@ Line 47: / Line 37: @@
 }}
 == Factorial Design ==

Icouckuy: Initial Designs moved to Config:InitialDesign: tag links to this now

2008-02-08T09:07:50Z

Initial Designs moved to Config:InitialDesign: <InitialDesign> tag links to this now

New page

== Latin Hypercube Design ==
Choose an initial sampleset in such a way that they form a latin hypercube.
{{OptionsHeader}}
{{Option
|name = points
|values = positive integer
|default = 30
|description = The amount of samples that will be taken.
}}

== Central Composite Design ==

{{OptionsHeader}}
{{Option
|name = type
|values = [inscribed,factoral]
|default = factoral
|description = Where are the '''''star points''''' placed? See http://www.itl.nist.gov/div898/handbook/pri/section3/pri3361.htm
}}

== Dataset Design ==
Reads an initial design from a dataset. Each row should contain one sample. The inputvalues come first. The outputvalues, if there are any, come second.

{{OptionsHeader}}
{{Option
|name = filename
|values = path
|default = none
|description = Path of the file that contains the dataset. '''This option is required!'''
}}
{{Option
|name = hasOutputs
|values = boolean
|default = true
|description = Does the datasetfile contain evaluated samples or just sample locations?
}}

== Combined Design ==

Combines two other initial designs.

{{NodesHeader}}
{{Node
|name = InitialDesign
|required = one or more
|description = Configuration for each InitialDesign Method.
}}

== Screening Designs ==

Create a design from information retrieved by parameter screening.

{{NodesHeader}}
{{Node
|name = InitialDesign
|required = yes
|description = Configuration for the requested InitialDesign Method.
}}
{{Node
|name = ScreeningMethod
|required = yes
|description = Configuration for the requested Screening Method. See below for further instructions.
}}

=== ScreeningMethod: OneD Screening ===
Build one-dimensional metamodels (minimetamodel) for each factor. Then choose the initial design based upon a factor importance metric.
{{OptionsHeader}}
{{Option
|name = importanceMetric
|values = [samples,sensitivity,minmax]
|default = samples
|description = What defines the importance of a factor? '''(1) samples:''' amount of samples required to reach a certain accuray. '''(2) sensitivity:''' order of the metamodel, only works with polynomial models. '''(3) minmax:''' greatest difference in the metamodel
}}
{{Option
|name = configFile
|values = path
|default = '/config/defaultMini.xml'
|description = Path of the file that contains the configuration to create each minimetalmodel. Best to leave this default.
}}

== Factorial Design ==
Choose an initial sampleset in such a way that they form an uniform grid
{{OptionsHeader}}
{{Option
|name = levels
|values = vector of positive integers > 1
|default = [5,...,5]
|description = vector specifying for each dimension the number of samples. Sampling level1*level2*...*leveln points. If a scalar s is specified, The toolbox assumes [s s ... s].
}}