Proposals:Refactoring Statistics Framework 2007 New Statistics Framework: Difference between revisions

From KitwarePublic
Jump to navigationJump to search
 
(11 intermediate revisions by the same user not shown)
Line 126: Line 126:
Object [URL="http://public.kitware.com/Insight/Doxygen/html/classitk_1_1Object.html"];
Object [URL="http://public.kitware.com/Insight/Doxygen/html/classitk_1_1Object.html"];
FunctionBase [URL="http://public.kitware.com/Insight/Doxygen/html/classitk_1_1FunctionBase.html"];
FunctionBase [URL="http://public.kitware.com/Insight/Doxygen/html/classitk_1_1FunctionBase.html"];
CostFunction [URL="http://public.kitware.com/Insight/Doxygen/html/classitk_1_1CostFunction.html"];
SingleValuedCostFunction [URL="http://public.kitware.com/Insight/Doxygen/html/classitk_1_1SingleValuedCostFunction.html"];
MembershipFunctionBase [shape=box, URL="http://public.kitware.com/Insight/Doxygen/html/classitk_1_1Statistics_1_1MembershipFunctionBase.html"];
MembershipFunctionBase [shape=box, URL="http://public.kitware.com/Insight/Doxygen/html/classitk_1_1Statistics_1_1MembershipFunctionBase.html"];
DensityFunction [shape=box, URL="http://public.kitware.com/Insight/Doxygen/html/classitk_1_1Statistics_1_1DensityFunction.html"];
GaussianDensityFunction [shape=box, URL="http://public.kitware.com/Insight/Doxygen/html/classitk_1_1Statistics_1_1GaussianDensityFunction.html"];
DistanceToCentroidMembershipFunction [shape=box, URL="http://public.kitware.com/Insight/Doxygen/html/classitk_1_1Statistics_1_1DistanceToCentroidMembershipFunction.html"];
DistanceToCentroidMembershipFunction [shape=box, URL="http://public.kitware.com/Insight/Doxygen/html/classitk_1_1Statistics_1_1DistanceToCentroidMembershipFunction.html"];
MahalanobisDistanceMembershipFunction [shape=box, URL="http://public.kitware.com/Insight/Doxygen/html/classitk_1_1Statistics_1_1DistanceToCentroidMembershipFunction.html"];
DistanceMetric  [shape=box, URL="http://public.kitware.com/Insight/Doxygen/html/classitk_1_1Statistics_1_1DistanceMetric.html"];
GoodnessOfFitFunctionBase [shape=box, URL="http://public.kitware.com/Insight/Doxygen/html/classitk_1_1Statistics_1_1GoodnessOfFitFunctionBase.html"];
EuclideanDistanceMetric  [shape=box, URL="http://public.kitware.com/Insight/Doxygen/html/classitk_1_1Statistics_1_1EuclideanDistanceMetric.html"];
GoodnessOfFitComponentBase [shape=box, URL="http://public.kitware.com/Insight/Doxygen/html/classitk_1_1Statistics_1_1GoodnessOfFitComponentBase.html"];
MahalanobisDistanceMetric  [shape=box, URL="http://public.kitware.com/Insight/Doxygen/html/classitk_1_1Statistics_1_1MahalanobisDistanceMetric.html"];
GoodnessOfFitMixtureModelCostFunction [shape=box, URL="http://public.kitware.com/Insight/Doxygen/html/classitk_1_1Statistics_1_1GoodnessOfFitMixtureModelCostFunction.html"];
SampleAlgorithmBase [shape=box, URL="http://public.kitware.com/Insight/Doxygen/html/classitk_1_1Statistics_1_1SampleAlgorithmBase.html"];
LogLikelihoodGoodnessOfFitFunction [shape=box, URL="http://public.kitware.com/Insight/Doxygen/html/classitk_1_1Statistics_1_1LogLikelihoodGoodnessOfFitFunction.html"];
DistanceMetrics  [label="Distance Metrics\n see below", URL="http://www.itk.org/Wiki/Proposals:Refactoring_Statistics_Framework_2007_Class_Manifesto#Distance_Metrics_2"];
Object -> FunctionBase
Object -> FunctionBase
Object -> GoodnessOfFitComponentBase
Object -> CostFunction
Object -> SampleAlgorithmBase
SampleAlgorithmBase -> GoodnessOfFitFunctionBase
FunctionBase -> MembershipFunctionBase
FunctionBase -> MembershipFunctionBase
MembershipFunctionBase -> DistanceMetrics
FunctionBase -> DistanceMetric
MembershipFunctionBase -> DensityFunction
DistanceMetric -> MahalanobisDistanceMetric
DensityFunction -> GaussianDensityFunction
DistanceMetric -> EuclideanDistanceMetric
MembershipFunctionBase -> MahalanobisDistanceMembershipFunction
DistanceMetric -> EuclideanSquaredDistanceMetric
DistanceMetric -> ManhattanDistanceMetric
MembershipFunctionBase -> DistanceToCentroidMembershipFunction
MembershipFunctionBase -> DistanceToCentroidMembershipFunction
GoodnessOfFitFunctionBase -> LogLikelihoodGoodnessOfFitFunction
SingleValuedCostFunction -> GoodnessOfFitMixtureModelCostFunction
CostFunction -> SingleValuedCostFunction
}
}
</graphviz>
</graphviz>
=== Distance notation ===
* Manhattan (L1) = sum of absolute values
* Euclidean = square root of ( sum of squares )
* Euclidean Squared  (L2) = sum of squares
* Mahalanobis = square root of ( V . M . VT )
=== API ===
* DistanceToCentroidMembershipFunction
** SetDistanceMetric( const DistanceMetric * ) (new)
** const GetDistanceMetric()  (new)
** Evaluate( Measurement vector ) (already there)
** SetCentroid( )  (already there)

Latest revision as of 20:57, 17 July 2008

Class Manifesto of New Statistics Framework

Summary Table

The classes that integrate the new statistics framework are categorized in the following table


Conceptual Class Number
Traits 1
Data Objects 4
Filters 11
Total 16

List of Classes per Category

Traits



  • MeasurementVectorTraits

Data Objects



  • Sample
  • ListSample
  • Histogram
  • Subsample

Filters

  • SampleToHistogramFilter
  • MeanFilter
  • WeightedMeanFilter
  • CovarianceFilter
  • WeightedCovarianceFilter
  • HistogramToTextureFeaturesFilter
  • ImageToListSampleFilter
  • ScalarImageToCooccurrenceMatrixFilter
  • SampleToSubsampleFilter
  • SampleClassifierFilter
  • NeighborhoodSubsampler

Classifiers (Suggested Design)

Elements

  • MembershipFunctionBase
    • DistanceToCentroidMembershipFunction (plugs in a DistanceMetric)
  • DistanceMetrics
    • Euclidean
    • Mahalanobis
    • 1_1

Filters

  • Sample, Array of Membership Functions --> MembershipSample(sample,labels) == SampleClassifierFilter
  • Sample, Array of Membership Functions --> GoodnessOfFitComponent (sample,weights) == SampleGoodnessOfFitFilter

Class Diagrams

Traits

Error writing graphviz file to disk.

Data Objects

Error writing graphviz file to disk.

Filters

Error writing graphviz file to disk.

Classifiers (Suggested Design)

Error writing graphviz file to disk.


Distance notation

  • Manhattan (L1) = sum of absolute values
  • Euclidean = square root of ( sum of squares )
  • Euclidean Squared (L2) = sum of squares
  • Mahalanobis = square root of ( V . M . VT )

API

  • DistanceToCentroidMembershipFunction
    • SetDistanceMetric( const DistanceMetric * ) (new)
    • const GetDistanceMetric() (new)
    • Evaluate( Measurement vector ) (already there)
    • SetCentroid( ) (already there)
Retrieved from "https://public.kitware.com/Wiki/index.php?title=Proposals:Refactoring_Statistics_Framework_2007_New_Statistics_Framework&oldid=12702"