mml

Class R_D.ForestSearch.M

java.lang.Object

la.la.Value

mml.Model

mml.UPModel.M

mml.ByPdf.M

mml.R_D.M

mml.R_D.ForestSearch.M

All Implemented Interfaces:

java.lang.Comparable<Value>, HasPdf

Enclosing class:

R_D.ForestSearch

public class R_D.ForestSearch.M
extends R_D.M

The fully parameterised R_D.ForestSearch-model. Note stats(Vector,int,int) in particular.

Nested Class Summary

Nested classes/interfaces inherited from class mml.R_D.M

R_D.M.Transform

Nested classes/interfaces inherited from class mml.Model

Model.Defaults

Nested classes/interfaces inherited from class la.la.Value

Value.Atomic, Value.Bool, Value.Char, Value.Chars, Value.Cts, Value.Defer, Value.Discrete, Value.Enum, Value.Inc_Or, Value.Int, Value.Lambda, Value.List, Value.Maybe, Value.Option, Value.Real, Value.Scannable, Value.Structured, Value.Triv, Value.Tuple

Field Summary

Fields

Modifier and Type	Field and Description
Linear1.M[]	Lin1 The models for the with-parent variables.
int	nChildren The number of variables that have a parent.
NormalUPM.M[]	Nrml The models for the parent-less variables.
int[]	parent If parent[i]≥0 then variable i has that parent.

Fields inherited from class mml.Model

msg1, msg2, sp

Fields inherited from class la.la.Value

C, comparator, CR, E, eight, ffalse, five, fiveR, four, fourR, half, negCR, negOne, negOneR, negTenR, nilVal, nine, None, one, oneR, PI, PIby2, point01, point1, point5, point9, quarter, seven, six, ten, tenR, three, threeR, triv, ttrue, two, twoR, zero, zeroR

Constructor Summary

Constructors

Constructor and Description
M(double msg1, double msg2, Value sp) 'sp' contains [p0, p2, ..., p(D-1)] where p_i is the parent of 'i', a negative value indicates that p_i has "no parent", parameters ⟨μ,σ⟩ of the various parent-less N_(μ,σ), and ⟨a,b,s⟩ of the parented Linear1_(a,b,s).

Method Summary

Modifier and Type	Method and Description
double	nlLH(Value ss) Negative log likelihood for data-set ds where ss=stats(ds).
double	nlPdf(Value v) The negative log pdf, nlPdf(d), of a datum d, must be specified to define a ByPdf.
void	random(double[] xs) TODO – not tested!
Value.Tuple	stats(boolean add, Value ss0, Value ss1) Combine (+/-) two statistics, ss0 and ss1.
Value.Tuple	stats(Vector ds, int lo, int hi) Note that these M.stats(...) differs from R_D.ForestSearch.stats(Vector,int,int) because parent[] is known in M.

Methods inherited from class mml.R_D.M

D, random_x, random, transform

Methods inherited from class mml.ByPdf.M

nlPr, nlPr, pdf

Methods inherited from class mml.UPModel.M

asGiven, asGiven, toString

Methods inherited from class mml.Model

asEstimator, asUPModel, m1m2sp, msg, msg1, msg1bits, msg2, msg2bits, msgBits, nl2LH, nl2Pr, pr, random, randomSeries, statParams, stats, stats, sumNlPr, transform, type, zeroTriv

Methods inherited from class la.la.Value

AoM, apply, bOp, compareTo, cons, cts, elt, errMsg, error, force, isTuple, just, main, n, nElts, nlAoM, pair, print, quad, real, RTE, triple, tuple, UOE, uOp, x

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait

Field Detail

parent

public final int[] parent

If parent[i]≥0 then variable i has that parent.

nChildren

public final int nChildren

The number of variables that have a parent.

Nrml

public final NormalUPM.M[] Nrml

The models for the parent-less variables.

Lin1

public final Linear1.M[] Lin1

The models for the with-parent variables.

Constructor Detail

M

public M(double msg1,
         double msg2,
         Value sp)

'sp' contains [p0, p2, ..., p(D-1)] where p_i is the parent of 'i', a negative value indicates that p_i has "no parent", parameters ⟨μ,σ⟩ of the various parent-less N_(μ,σ), and ⟨a,b,s⟩ of the parented Linear1_(a,b,s).

Method Detail

stats

public Value.Tuple stats(Vector ds,
                         int lo,
                         int hi)

Note that these M.stats(...) differs from R_D.ForestSearch.stats(Vector,int,int) because parent[] is known in M. Uses R_D.ds2NorL1stats(la.maths.Vector, int[]).

Overrides:

stats in class UPModel.M

stats

public Value.Tuple stats(boolean add,
                         Value ss0,
                         Value ss1)

Combine (+/-) two statistics, ss0 and ss1.

Overrides:

stats in class UPModel.M

nlPdf

public double nlPdf(Value v)

Description copied from class: ByPdf.M

The negative log pdf, nlPdf(d), of a datum d, must be specified to define a ByPdf. Also see pdf(d).

Specified by:

nlPdf in interface HasPdf

Specified by:

nlPdf in class ByPdf.M

nlLH

public double nlLH(Value ss)

Negative log likelihood for data-set ds where ss=stats(ds).

Specified by:

nlLH in class Model

random

public void random(double[] xs)

TODO – not tested!

Overrides:

random in class R_D.M