Package de.lmu.ifi.dbs.elki.algorithm.projection

Data projections (see also preprocessing filters for basic projections).

See: Description

Interface Summary

Interface	Description
AffinityMatrix	Abstraction interface for an affinity matrix.
AffinityMatrixBuilder<O>	Interface for computing an affinity matrix.

Class Summary

Class	Description
AbstractProjectionAlgorithm<R extends Result>	Abstract base class for projection algorithms.
BarnesHutTSNE<O>	tSNE using Barnes-Hut-Approximation.
BarnesHutTSNE.Parameterizer<O>	Parameterization class.
BarnesHutTSNE.QuadTree	Quad Tree for use in a Barnes-Hut approximation.
DenseAffinityMatrix	Dense affinity matrix storage.
GaussianAffinityMatrixBuilder<O>	Compute the affinity matrix for SNE and tSNE using a Gaussian distribution with a constant sigma.
GaussianAffinityMatrixBuilder.Parameterizer<O>	Parameterization class.
IntrinsicNearestNeighborAffinityMatrixBuilder<O>	Build sparse affinity matrix using the nearest neighbors only, adjusting for intrinsic dimensionality.
IntrinsicNearestNeighborAffinityMatrixBuilder.Parameterizer<O>	Parameterization class.
NearestNeighborAffinityMatrixBuilder<O>	Build sparse affinity matrix using the nearest neighbors only.
NearestNeighborAffinityMatrixBuilder.Parameterizer<O>	Parameterization class.
PerplexityAffinityMatrixBuilder<O>	Compute the affinity matrix for SNE and tSNE.
PerplexityAffinityMatrixBuilder.Parameterizer<O>	Parameterization class.
SNE<O>	Stochastic Neighbor Embedding is a projection technique designed for visualization that tries to preserve the nearest neighbor structure.
SNE.Parameterizer<O>	Parameterization class.
SparseAffinityMatrix	Dense affinity matrix storage.
TSNE<O>	t-Stochastic Neighbor Embedding is a projection technique designed for visualization that tries to preserve the nearest neighbor structure.
TSNE.Parameterizer<O>	Parameterization class.

Package de.lmu.ifi.dbs.elki.algorithm.projection Description

Data projections (see also preprocessing filters for basic projections).