public class GaussianVertex extends DoubleVertex implements ProbabilisticDouble
Constructor and Description |
---|
GaussianVertex(double mu,
double sigma) |
GaussianVertex(double mu,
DoubleVertex sigma) |
GaussianVertex(DoubleVertex mu,
double sigma) |
GaussianVertex(DoubleVertex mu,
DoubleVertex sigma) |
GaussianVertex(int[] tensorShape,
double mu,
double sigma) |
GaussianVertex(int[] tensorShape,
double mu,
DoubleVertex sigma) |
GaussianVertex(int[] tensorShape,
DoubleVertex mu,
double sigma) |
GaussianVertex(int[] tensorShape,
DoubleVertex mu,
DoubleVertex sigma)
One mu or sigma or both that match a proposed tensor shape of Gaussian
|
Modifier and Type | Method and Description |
---|---|
DualNumber |
calculateDualNumber(java.util.Map<Vertex,DualNumber> dualNumbers) |
java.util.Map<Vertex,DoubleTensor> |
dLogProb(DoubleTensor value,
java.util.Set<? extends Vertex> withRespectTo)
The partial derivatives of the natural log prob.
|
DoubleVertex |
getMu() |
DoubleVertex |
getSigma() |
double |
logProb(DoubleTensor value)
This is the natural log of the probability at the supplied value.
|
DoubleTensor |
sample(KeanuRandom random) |
abs, acos, asin, atan, atan2, ceil, concat, cos, div, div, divideBy, divideBy, equalTo, exp, floor, getValue, greaterThan, greaterThanOrEqualTo, lambda, lambda, lessThan, lessThanOrEqualTo, log, matrixInverse, matrixMultiply, minus, minus, multiply, multiply, notEqualTo, observe, observe, plus, plus, pow, pow, reshape, reverseModeAutoDifferentiation, round, setAndCascade, setAndCascade, setValue, setValue, sigmoid, sin, slice, sum, take, tan, times, times, unaryMinus
addChild, addParent, addParents, equals, eval, getChildren, getConnectedGraph, getId, getIndentation, getLabel, getObservedValue, getParents, getRawValue, getShape, getValue, hashCode, hasValue, isObserved, isProbabilistic, labeledAs, lazyEval, observe, observeOwnValue, sample, setAndCascade, setLabel, setParents, setParents, setValue, toString, unobserve
clone, finalize, getClass, notify, notifyAll, wait, wait, wait
dLogPdf, dLogPdf, dLogPdf, dLogPdf, dLogPdf, dLogPdf, logPdf, logPdf, logPdf
dLogProb, dLogProbAtValue, dLogProbAtValue, getValue, keepOnlyProbabilisticVertices, logProbAtValue, setValue
getObservedValue, isObserved, observableTypeFor, observe, unobserve
getDualNumber
public GaussianVertex(int[] tensorShape, DoubleVertex mu, DoubleVertex sigma)
If all provided parameters are scalar then the proposed shape determines the shape
tensorShape
- the desired shape of the tensor in this vertexmu
- the mu of the Gaussian with either the same tensorShape as specified for this vertex or a scalarsigma
- the sigma of the Gaussian with either the same tensorShape as specified for this vertex or a scalarpublic GaussianVertex(DoubleVertex mu, DoubleVertex sigma)
public GaussianVertex(DoubleVertex mu, double sigma)
public GaussianVertex(double mu, DoubleVertex sigma)
public GaussianVertex(double mu, double sigma)
public GaussianVertex(int[] tensorShape, DoubleVertex mu, double sigma)
public GaussianVertex(int[] tensorShape, double mu, DoubleVertex sigma)
public GaussianVertex(int[] tensorShape, double mu, double sigma)
public DoubleVertex getMu()
public DoubleVertex getSigma()
public double logProb(DoubleTensor value)
Probabilistic
logProb
in interface Probabilistic<DoubleTensor>
value
- The supplied value.public java.util.Map<Vertex,DoubleTensor> dLogProb(DoubleTensor value, java.util.Set<? extends Vertex> withRespectTo)
Probabilistic
dLogProb
in interface Probabilistic<DoubleTensor>
value
- at a given valuewithRespectTo
- list of parents to differentiate with respect topublic DoubleTensor sample(KeanuRandom random)
sample
in class Vertex<DoubleTensor>
random
- source of randomnesspublic DualNumber calculateDualNumber(java.util.Map<Vertex,DualNumber> dualNumbers)
calculateDualNumber
in interface Differentiable
calculateDualNumber
in class DoubleVertex