Package blog

Interface Summary
BayesNetVar A random variable that serves as a node in a Bayes net (directed graphical model).
CondProbDistrib An interface which all user-defined conditional probability distributions (CPDs), as well as the EqualsCPD class, are expected to implement.
EvalContext Interface for objects that encapsulate a partial world and an assignment of values to logical variables.
FunctionInterp Interface for classes that define the interpretation of a non-random function symbol.
ObjectIterator Iterator over an ObjectSet.
ObjectSet Interface for sets of objects in a partial world.
PartialWorld A partial description of a possible world in a BLOG model.
Proposer Interface for Metropolis-Hastings proposal distributions.
Query Interface for queries that can be answered by an inference engine.
SetupExtender Interface for objects that can extend the setup of a BLOG inference problem: that is, the model, evidence, and query list.
WorldDiffListener Interface for objects that can be notified when a PartialWorldDiff object is saved or reverted.
WorldListener Interface for objects that can be notified of changes to the values of random variables in a PartialWorld.
 

Class Summary
AbstractCondProbDistrib Abstract implementation of the CondProbDistrib interface.
AbstractFunctionInterp Abstract implementation of FunctionInterp including a getInverseTuples method that always returns null.
AbstractObjectIterator Abstract implementation of the ObjectIterator interface.
AbstractObjectSet Abstract implementation of the ObjectSet interface.
AbstractPartialWorld An implementation of the PartialWorld interface that just requires concrete subclasses to initialize some protected variables.
AbstractQuery An implementation of Query that defines default implementations to some methods.
ArgSpec Represents an abstract argument specification.
ArgSpecQuery  
AtomicFormula A Formula consisting of a single boolean-valued term.
BasicVar A basic random variable in a BLOG model.
BLOGLexer  
BLOGParser CUP v0.10k generated parser.
BLOGParser.CPDSpec Nested class that specifies a CPD with parameters and arguments.
BLOGParser.GenObjSpec Nested class that specifies a origin function and a variable for the generating object.
BLOGParser.ModelEvidenceQueries  
BLOGParser.PotentialSpec Nested class that specifies a potential with parameters and arguments.
BLOGParser.TypedVar Nested class that specifies a variable name and a type.
BLOGParser.VarsWithConstraint Nested class that specifies a list of logical variables and a formula that constrains them.
BLOGTokenConstants CUP generated class containing symbol constants.
BuiltInFunctions Class with static methods and constants for built-in non-random functions (including built-in constants).
BuiltInTypes Class with static methods and variables for dealing with the built-in types that exist in every BLOG model.
CardinalitySpec Argument specifier that refers to the size of an implicitly defined set.
Clause Represents a clause in dependency statements and number statements.
CompiledSetSpec Data structure that facilitates iterating over the set of objects x such that a given formula phi(x) is true.
ConjFormula Represents a conjuction of expressions, each of which is a Formula.
ConstantInterp Implementation of FunctionInterp for constants (zero-ary functions).
DecayedProposer A DecayedProposer as defined by Bhaskara Marthi, Hanna Pasula, Stuart Russell, Yuval Peres, "Decayed MCMC Filtering" (2002).
DefaultEvalContext Basic implementation of the EvalContext interface.
DefaultObjectIterator ObjectIterator implementation that wraps around an ordinary Iterator.
DefaultObjectSet ObjectSet implementation that can be constructed using an ordinary Collection.
DefaultPartialWorld Straightforward implementation of AbstractPartialWorld.
DependencyModel Represents dependency statements for functions and number statements for potential object patterns.
DependencyModel.Distrib Nested class representing a distribution over child values, in the form of a CPD and a list of values for the CPD's arguments.
DerivedVar A random variable whose value is a deterministic function of some basic variables.
DetCondProbDistrib Abstract class for deterministic conditional probability distributions.
DisjFormula Represents a disjunction of expressions, each of which is a Formula.
DMHSampler Implements a Metropolis-Hastings sampler with the DecayedProposer.
EnumeratedObject An EnumeratedObject represents a guaranteed object that is explicitly defined and given a name in the model file, using a statement like: guaranteed typename name1, name2, ..., nameN A "name" is really a nonrandom constant symbol that refers to this object.
EqualityFormula Represents an equality test on 2 expressions, each of which is a non-boolean-valued Term.
EqualsCPD EqualsCPD class is a convenience hack to represent the situation when the value of the right-hand side of a Clause depends on a function application rather than on a CPD.
Evidence Stores the information extracted from evidence file.
ExistentialFormula Represents an existential instantiation for one variable satisfying an expression of type Formula.
ExplicitSetSpec Represents an argument - set with explicit listing of its elements.
Formula Formula is an abstract class from which all classes denoting particular kinds of formulas inherit.
FormulaQuery  
FuncAppTerm Represents a function invocation.
Function Represents a BLOG function.
Function.Sig Nested class representing a function signature, that is, a function name and a tuple of argument types.
GenericObject Represents a generic object of some type, possibly with constraints on the values of its origin functions.
GenericPOPAppSatisfier Represents a generic satisfier of some POP application.
GenericProposer Implements a generic MCMC proposal algorithm.
ImplicFormula An implication formula alpha -> beta.
ImplicitSetSpec Represents an argument - set with implicit specification of its elements.
InferenceEngine Abstract class representing a BLOG inference engine.
InstantiatingEvalContext Implementation of the EvalContext interface that instantiates new basic random variables as necessary in response to method calls.
LogicalVar A logical variable.
LWSampler A likelihood weighting sampler.
Main Main program for the BLOG (Bayesian Logic) inference engine.
MatrixType Represents a matrix or vector type.
MHSampler Generates samples from a Markov chain over possible worlds using a Metropolis-Hastings algorithm.
Model This class contains all the information available about the structure of the model.
NegFormula Represents a logical negation of an expression of type Formula.
NonGuaranteedObject  
NonRandomFunction Represents the symbol for a non-random function, whose value for the given tuple of arguments is constant over worlds.
NumberVar A number random variable.
ObjectIdentifier An identifier for a non-guaranteed object in a PartialWorld.
ObjGenGraph An object generation graph contains nodes that represent sets of objects with certain generation histories.
ObjGenGraph.GuaranteedNode  
ObjGenGraph.IntegerNode  
ObjGenGraph.Node  
ObjGenGraph.OrNode  
ObjGenGraph.POPNode Node satisfied by the set of objects that satisfy a given potential object pattern (POP).
ObjGenGraph.TermNode Node satisfied only by the denotation of the given term.
ObjGenGraph.TimestepNode  
OriginFunction Represents a BLOG generating function.
OriginVar A variable whose value is the POP application satisfied by a given object identifier.
ParentRecEvalContext Implementation of the EvalContext interface that records the basic random variables used by calls to the access methods.
ParserUtilities  
PartialWorld.EmptyPartialWorld  
PartialWorldDiff Represents a PartialWorld as a set of differences relative to an underlying "saved" PartialWorld.
Particle A Particle for Particle Filtering.
ParticleFilter A Particle Filter.
ParticleFilterTest Unit testing for the ParticleFilter.
POP Represents a potential object pattern (POP), which includes the type of object to be generated and a tuple of origin functions.
POPAppBasedSet ObjectSet implementation that is composed of individual objects, plus all the objects that satisfy certain potential object pattern (POP) applications (number variables), with certain exceptional objects excluded.
RandFuncAppVar A random variable whose value is the value of a certain random function on a certain tuple of arguments.
RandomFunction Represents random functions, whose value for the given tuple of arguments changes from world to world.
RejectionSampler Generates a partial world by sampling basic random variables in an order that does not depend on the evidence or queries, but only on which basic RVs are supported by the instantiation created so far.
Sampler An object that stochastically generates a sequence of samples, which are possible worlds of a given BLOG model.
SamplingEngine Inference engine that answers queries by sampling partial worlds (possibly with associated weights) from some sampling distribution, and collecting statistics over these samples.
SimpleInstEvalContext Evaluation context that instantiates random variables as needed, using some very simple rules.
SkolemConstant Represents a skolem constant -- a constant introduced by symbol evidence statements.
Substitution A mapping from logical variables to terms.
SymbolEvidenceStatement Represents a symbol evidence statement, which is of the form type_name var_name:cond_formula = { symbol_list }.
SymbolTerm A term consisting of a single symbol.
Term Represents a Bayesian atom.
Timestep Represents a time step, and is used for DBLOG (Dynamic BLOG) to indicate temporal random variables.
TrueFormula A Formula consisting of just the zero-ary logical operator "true", which is true in every world under every assignment.
TupleSetSpec Represents an argument - set consisting of implicitly specified tuples.
Type Represents a type in a BLOG model.
UniversalFormula Represents a universal instantiation for one variable using one expression of type Formula.
UrnBallsSplitMerge Split-merge proposal distribution for the urn-and-balls model.
UrnBallsSplitMergeNoIds Split-merge proposal distribution for the urn and balls scenario that doesn't use identifiers.
ValueEvidenceStatement Represents a statement that a certain term has a certain value.
VarWithDistrib Abstract class for variables that have a distribution specified by the BLOG model.
WorldInProgress PartialWorld implementation that maintains a list of uninstantiated VarWithDistrib objects, and allows variables to be instantiated using a special method of an iterator over this list.