Classes
class	oevislib_net.DataClassification.Dataset
	Dataset containing features, classes and labels useful for classification purpose. More...
class	oevislib_net.DataClassification.Model
	Class containing a machine learning model. More...

Functions
static Array< int >	oevislib_net.DataClassification.Clustering.KMeans (Array< float > inData, int inClusters, int inMaxIterations=100, int inAttempts=1, KMeansInitialization inClustersInitialization=KMeansInitialization.KMeansPP, Optional< Array< int > > inInitialLabels=null, Optional< Array< float > > outCentroids=null)
	Groups input samples around clusters using K-Means algorithm.
static Array< int >	oevislib_net.DataClassification.Clustering.Dbscan (Array< float > inData, int inMinPoints, float inEpsilon, bool inIgnoreOutliers=true, Optional< Array< float > > outCentroids=null)
	Groups input points around clusters using DBSCAN algorithm.

Detailed Description

Function Documentation

Array< int > oevislib_net.DataClassification.Clustering.Dbscan	(	Array< float >	inData,
		int	inMinPoints,
		float	inEpsilon,
		bool	inIgnoreOutliers = true,
		Optional< Array< float > >	outCentroids = null )

inlinestatic

Groups input points around clusters using DBSCAN algorithm.

Parameters

inData	Data used for clustering (a single feature per sample).
inMinPoints	Minimum number of neighbors to create a cluster from a point. Range: [1, +inf).
inEpsilon	Maximum distance between two points to be considered neighbors. Range: (0, +inf).
inIgnoreOutliers	If true, no errors are thrown if outliers are found.
outCentroids	Clusters' centroids.

Returns: Labels assigned to each sample (indices starting from 0). A label set to -1 indicates an outlier.

Array< int > oevislib_net.DataClassification.Clustering.KMeans	(	Array< float >	inData,
		int	inClusters,
		int	inMaxIterations = 100,
		int	inAttempts = 1,
		KMeansInitialization	inClustersInitialization = KMeansInitialization::KMeansPP,
		Optional< Array< int > >	inInitialLabels = null,
		Optional< Array< float > >	outCentroids = null )

inlinestatic

Groups input samples around clusters using K-Means algorithm.

Parameters

inData	Data used for clustering (a single feature per sample).
inClusters	Number of clusters in which to split the set. Range: [2, +inf).
inMaxIterations	Maximum number of iterations. Range: [1, +inf).
inAttempts	Number of times the algorithm is executed using different initial labels. The algorithm returns at the end the labels that yield the best compactness. Range: [1, +inf).
inClustersInitialization	Method used to initialize clusters' centroids.
inInitialLabels	Labels to use during the first (and possibly the only) attempt, instead of computing them from the initial centers. For the second and further attempts, the method defined by inClustersInitialization is used.
outCentroids	Clusters' centroids.