public class NearInverse.M extends Continuous.M
NearInverse
.Model.Defaults
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
Modifier and Type | Field and Description |
---|---|
protected Value.Cts |
delta
The "statistical parameter".
|
double |
logIntegral
The log of the normalising constant.
|
Constructor and Description |
---|
M(double delta_x)
A possible use is for σ's prior in
MML.Normal . |
M(double msg1,
double msg2,
Value delta)
The standard constructor for NearInverse Model;
statistical parameter delta must be a small Value.Cts in
(0, 1) such as 0.1.
|
Modifier and Type | Method and Description |
---|---|
double |
nlLH(Value ss)
Assumes that
statistics
are the data-set itself. |
double |
nlPdf_x(double x)
Nearly - log(1/x), that is + log(x), but not quite,
corresponding to a pdf(x) of ~1/x, but not quite
(as always,
|
asEstimator, asUPModel, m1m2sp, msg, msg1, msg1bits, msg2, msg2bits, msgBits, nl2LH, nl2Pr, pr, random, randomSeries, statParams, stats, stats, sumNlPr, type, zeroTriv
protected final Value.Cts delta
public final double logIntegral
public M(double msg1, double msg2, Value delta)
public M(double delta_x)
MML.Normal
.public double nlPdf_x(double x)
nlPdf_x
in class Continuous.M
public double nlLH(Value ss)
statistics
are the data-set itself.