|
|||||||||
PREV CLASS NEXT CLASS | FRAMES NO FRAMES | ||||||||
SUMMARY: NESTED | FIELD | CONSTR | METHOD | DETAIL: FIELD | CONSTR | METHOD |
java.lang.Objectstatistics.Trainer
public abstract class Trainer
Implementation of sequential training algorithms for (mixture) densities. These are: EM, MAP, ML.
Constructor Summary | |
---|---|
Trainer()
|
Method Summary | |
---|---|
static MixtureDensity |
em(MixtureDensity initial,
java.util.List<Sample> data)
Given an initial mixture density, perform a single EM iteration w/ out the use of parallelization. |
static MixtureDensity |
em(MixtureDensity initial,
java.util.List<Sample> data,
int iterations)
Perform a number of EM iterations (single-core, cached posteriors) using the initial density and the given data. |
static MixtureDensity |
map(MixtureDensity initial,
java.util.List<Sample> data,
double r,
int iterations,
java.lang.String update)
Perform a number of MAP iterations, based on the initial estimate |
static MixtureDensity |
map(MixtureDensity initial,
java.util.List<Sample> data,
double r,
java.lang.String update)
Perform a single MAP iteration on the given initial estimate |
static Density |
ml(java.util.List<Sample> data,
boolean diagonalCovariances)
Standard (single-core) maximum likelihood estimation for a single Gaussian density |
Methods inherited from class java.lang.Object |
---|
equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait |
Constructor Detail |
---|
public Trainer()
Method Detail |
---|
public static MixtureDensity em(MixtureDensity initial, java.util.List<Sample> data)
initial
- data
-
TrainingException
public static MixtureDensity em(MixtureDensity initial, java.util.List<Sample> data, int iterations)
initial
- data
- iterations
-
public static MixtureDensity map(MixtureDensity initial, java.util.List<Sample> data, double r, int iterations, java.lang.String update)
initial
- data
- iterations
- update
- String indicating which parameters to update: p for prior, m for mean, c for covariance; if null, "pmc" (i.e. all parameters) is assumed
public static MixtureDensity map(MixtureDensity initial, java.util.List<Sample> data, double r, java.lang.String update)
initial
- r
- Relevance factorupdate
- String indicating which parameters to update: p for prior, m for mean, c for covariance; if null, "pmc" (i.e. all parameters) is assumed
public static Density ml(java.util.List<Sample> data, boolean diagonalCovariances)
data
- diagonalCovariances
-
TrainingException
|
|||||||||
PREV CLASS NEXT CLASS | FRAMES NO FRAMES | ||||||||
SUMMARY: NESTED | FIELD | CONSTR | METHOD | DETAIL: FIELD | CONSTR | METHOD |