Novacta.Analytics Documentation
Non
Executes a nonmetric multidimensional scaling analysis.
Definition
Namespace: Novacta.Analytics
Assembly: Novacta.Analytics (in Novacta.Analytics.dll) Version: 2.1.0+428f3840cfab98dda567bb0ed350b302533e273a
The result of the nonmetric multidimensional scaling analysis.
Assembly: Novacta.Analytics (in Novacta.Analytics.dll) Version: 2.1.0+428f3840cfab98dda567bb0ed350b302533e273a
C#
public static NonMetricMultidimensionalScaling Analyze(
DoubleMatrix dissimilarities,
int? configurationDimension = null,
double minkowskiDistanceOrder = 2,
int maximumNumberOfIterations = 200,
double terminationTolerance = 0.0001
)VB
Public Shared Function Analyze (
dissimilarities As DoubleMatrix,
Optional configurationDimension As Integer? = Nothing,
Optional minkowskiDistanceOrder As Double = 2,
Optional maximumNumberOfIterations As Integer = 200,
Optional terminationTolerance As Double = 0.0001
) As NonMetricMultidimensionalScalingC++
public:
static NonMetricMultidimensionalScaling^ Analyze(
DoubleMatrix^ dissimilarities,
Nullable<int> configurationDimension = nullptr,
double minkowskiDistanceOrder = 2,
int maximumNumberOfIterations = 200,
double terminationTolerance = 0.0001
)F#
static member Analyze :
dissimilarities : DoubleMatrix *
?configurationDimension : Nullable<int> *
?minkowskiDistanceOrder : float *
?maximumNumberOfIterations : int *
?terminationTolerance : float
(* Defaults:
let _configurationDimension = defaultArg configurationDimension null
let _minkowskiDistanceOrder = defaultArg minkowskiDistanceOrder 2
let _maximumNumberOfIterations = defaultArg maximumNumberOfIterations 200
let _terminationTolerance = defaultArg terminationTolerance 0.0001
*)
-> NonMetricMultidimensionalScaling Parameters
- dissimilarities DoubleMatrix
- The matrix of dissimilarities.
- configurationDimension NullableInt32 (Optional)
- The dimension of the configuration of points in the target space. Defaults to null, meaning that the dimension of the configuration will be selected automatically.
- minkowskiDistanceOrder Double (Optional)
- The order of the Minkowski metric. It must be greater than or equal to 1. Defaults to 2.
- maximumNumberOfIterations Int32 (Optional)
- The maximum number of iterations. Defaults to 200.
- terminationTolerance Double (Optional)
- The termination tolerance. Defaults to 1e-4.
Return Value
NonMetricMultidimensionalScalingThe result of the nonmetric multidimensional scaling analysis.
Remarks
The search for the optimal configuration of points in the target space starts with a classical scaling of the dissimilarities. If the initial configuration can't be computed, an exception is thrown.
Exceptions
| ArgumentNullException | dissimilarities is null. |
| ArgumentException | dissimilarities is not symmetric -or- dissimilarities can't be scaled to an initial configuration via a ClassicalMultidimensionalScaling analysis. |
| ArgumentOutOfRangeException | configurationDimension is less than 1 -or- configurationDimension is greater than the number of rows in dissimilarities -or- configurationDimension is greater than the number of positive eigenvalues of the initial scaling matrix. -or- minkowskiDistanceOrder is less than 1. -or- maximumNumberOfIterations is not positive. -or- terminationTolerance is not positive. |