@Reference(title="Database-friendly random projections: Johnson-Lindenstrauss with binary coins", authors="D. Achlioptas", booktitle="Proc. 20th ACM SIGMOD-SIGACT-SIGART symposium on Principles of database systems", url="https://doi.org/10.1145/375551.375608", bibkey="DBLP:conf/pods/Achlioptas01") @Alias(value="de.lmu.ifi.dbs.elki.math.linearalgebra.randomprojections.AchlioptasRandomProjectionFamily") public class AchlioptasRandomProjectionFamily extends AbstractRandomProjectionFamily
Reference:
D. Achlioptas
Database-friendly random projections: Johnson-Lindenstrauss with binary coins
Proc. 20th ACM SIGMOD-SIGACT-SIGART Symp. on Principles of database systems
TODO: faster implementation exploiting sparsity.
| Modifier and Type | Class and Description |
|---|---|
static class |
AchlioptasRandomProjectionFamily.Parameterizer
Parameterization class.
|
AbstractRandomProjectionFamily.MatrixProjectionRandomProjectionFamily.Projection| Modifier and Type | Field and Description |
|---|---|
private double |
sparsity
Projection sparsity.
|
random| Constructor and Description |
|---|
AchlioptasRandomProjectionFamily(double sparsity,
RandomFactory random)
Constructor.
|
| Modifier and Type | Method and Description |
|---|---|
RandomProjectionFamily.Projection |
generateProjection(int idim,
int odim)
Generate a projection matrix for the given dimensionalities.
|
public AchlioptasRandomProjectionFamily(double sparsity,
RandomFactory random)
sparsity - Projection sparsityrandom - Random number generator.public RandomProjectionFamily.Projection generateProjection(int idim, int odim)
RandomProjectionFamilyidim - Input Dimensionalityodim - Output DimensionalityCopyright © 2019 ELKI Development Team. License information.