| Interface | Description |
|---|---|
| SpectralClusteringConf.ClustererProvider<DATATYPE> |
A function which can represent itself as a string
|
| StoppingCondition |
The stopping condition for a multiview spectral clustering algorithm
|
| Class | Description |
|---|---|
| AbsoluteValueEigenChooser |
Attempts to automatically choose the number of eigen vectors based on the
comparative value of the eigen value with the first eigen value seen.
|
| CachedDoubleSpectralClustering |
DoubleSpectralClustering extention which knows how to write and read its eigenvectors to disk
and therefore not regenerate them when calling the underlying PreparedSpectralClustering |
| ChangeDetectingEigenChooser |
Attempts to automatically choose the number of eigen vectors based on the
relative gap between eigen values.
|
| DoubleFVSimilarityFunction<T> |
Wraps the functionality of a
SimilarityClusterer around a dataset |
| DoubleMultiviewSpectralClustering | |
| DoubleSpectralClustering |
Built from a mixture of this tutorial:
- http://www.kyb.mpg.de/fileadmin/user_upload/files/publications/attachments/Luxburg07_tutorial_4488%5B0%5D.pdf
And this implementation:
- https://github.com/peterklipfel/AutoponicsVision/blob/master/SpectralClustering.java
|
| DummyExtractor | |
| EigenChooser |
Method which makes a decision on how many eigen vectors to select
|
| FBEigenIterator |
A forward or backward iterator of eigen vector/value pairs
|
| GraphLaplacian |
Functions which turn a graph weight adjacency matrix into the Laplacian
matrix.
|
| GraphLaplacian.Normalised |
The inverted symmetric normalised Laplacian is defined as:
L = D^-1/2 A D^-1/2
|
| GraphLaplacian.Unnormalised |
The symmetric normalised Laplacian is defined as:
L = D - W
|
| GraphLaplacian.Warped |
The inverted symmetric normalised Laplacian is defined as:
L = D^-1 .
|
| HardCodedEigenChooser | |
| MultiviewSpectralClusteringConf<DATATYPE> | |
| NormalisedSimilarityDoubleClustererWrapper<T> |
Wraps the functionality of a
SimilarityClusterer around a dataset |
| PreparedSpectralClustering |
For a given set of
Eigenvalues perform the stages of spectral
clustering which involve the selection of the best eigen values and the
calling of an internal clustering algorithm |
| RBFSimilarityDoubleClustererWrapper<T> |
Construct a similarity matrix using a Radial Basis Function
|
| SpectralClusteringConf<DATATYPE> | |
| SpectralClusteringConf.DefaultClustererFunction<DATATYPE> | |
| SpectralIndexedClusters |
IndexClusters which also hold the eigenvector/value pairs which created them |
| StoppingCondition.HardCoded |
Counts the iterations
|
| WineDatasetExperiment |
Perform spectral clustering experiments using the Wine Dataset
|