jhplot

Class HBsom

java.lang.Object

jhplot.HBsom

public class HBsom
extends java.lang.Object

The Bayesian self-organizing map (BSOM). This is a method for estimating a probability distribution generating data points on the basis of a Bayesian stochastic model. It is also regarded as a learning method for a kind of neural network. The black dots in the below figure denote artificially generated data points. Based on work of: Akio Utsugi.

This class is based on: A. Utsugi (1996) ``Topology selection for self-organizing maps", Network: Computation in Neural Systems, vol. 7, no. 4, 727-740.

A. Utsugi (1997) ``Hyperparameter selection for self-organizing maps", Neural Computation, vol. 9, no. 3, pp. 623-635.

Field Summary

Fields

Modifier and Type	Field and Description
Bsom	bsom

Constructor Summary

Constructors

Constructor and Description
HBsom() Initialize BSOM.

Method Summary

Modifier and Type	Method and Description
void	doc() Show documentation
double	getAlpha() Get alpha.
double	getBeta() Get beta.
double	getDelta() Get calculation precision.
int	getNiterations() Get number of iterations used for fitting
P1D	getResult() Get results of training.
void	run() Run the algorithm
void	setAlphaBeta(double alpha, double beta) Set initial alpha and beta parameters.
void	setData(H1D h) Load histogram data to BSOM
void	setData(P1D p1d) Load data to BSOM
void	setDelta(double delta) Set calculation precision.
void	setNPoints(int units) Set number of points for fit
void	visible() Set visible frame
void	visible(boolean vis) Set frame visible or not

Methods inherited from class java.lang.Object

equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Detail

bsom

public Bsom bsom

Constructor Detail

HBsom

public HBsom()

Initialize BSOM.

Method Detail

setNPoints

public void setNPoints(int units)

Set number of points for fit

Parameters:

units - set number of points for fit.

setData

public void setData(P1D p1d)

Load data to BSOM

Parameters:

p1d - input data

setData

public void setData(H1D h)

Load histogram data to BSOM

Parameters:

p1d - input histogram data

doc

public void doc()

Show documentation

setAlphaBeta

public void setAlphaBeta(double alpha,
                         double beta)

Set initial alpha and beta parameters. The BSOM model has a pair of hyperparameters: alpha and beta, which represent `the strength of topological constraint' and `the estimate of noise level in data' respectively. You can vary them using the sliders. Observe the variation of the centroid configuration according to the values of the hyperparameters and grasp their meaning. Then try to find the optimal values of the hyperparameters giving the best centroid configuration. Remark that the configuration depends on not only the present values of hyperparameters but also their history. Poor moving of the hyperparameters will lead to a poor local optimal configuration.

Parameters:

alpha - alpha value

beta - beta value

getResult

public P1D getResult()

Get results of training.

Returns:

P1D with results.

run

public void run()

Run the algorithm

visible

public void visible(boolean vis)

Set frame visible or not

Parameters:

vis - true if visible

setDelta

public void setDelta(double delta)

Set calculation precision. Iterations stop if (current(alpha)-previous(alpha) .lt. delta) The value should be very small for best results

Parameters:

delta - precision

getDelta

public double getDelta()

Get calculation precision.

getNiterations

public int getNiterations()

Get number of iterations used for fitting

getAlpha

public double getAlpha()

Get alpha. The strength of topological constraint

getBeta

public double getBeta()

Get beta. This is the estimate of noise level in data.

visible

public void visible()

Set visible frame