Darwin  1.10(beta)
 All Classes Namespaces Files Functions Variables Typedefs Enumerations Enumerator Friends Groups Pages
Public Member Functions | Protected Attributes | List of all members
drwnInference Class Referenceabstract

Interface for various (marginal) inference algorithms. More...

Inheritance diagram for drwnInference:
drwnMessagePassingInference drwnSumProdInference drwnAsyncSumProdInference

Public Member Functions

 drwnInference (const drwnFactorGraph &graph)
 reference to initial clique potentials
 drwnInference (const drwnInference &inf)
virtual void clear ()
 clear internally cached data (e.g., computation graph)
virtual bool inference ()=0
 run inference (or resume for iterative algorithms) and return true if converged
virtual void marginal (drwnTableFactor &belief) const =0
 return the belief over the variables in the given factor, which must be one of the cliques in the original factor graph
virtual drwnFactorGraph varMarginals () const
 returns marginals for each variable in the factor graph's universe
drwnTableFactor operator[] (int varIndx) const
 return the marginal distribution over variable varIndx
drwnTableFactor operator[] (const char *varName) const
 return the marginal distribution over variable varName

Protected Attributes

const drwnFactorGraph_graph

Detailed Description

Interface for various (marginal) inference algorithms.

Algorithms operate on a factor graph (see drwnFactorGraph) that is assumed to be provided as unnormalized probability distributions, i.e., $P(x) \propto \prod_c \Psi_c(x_c)$ and the task is to find the marginal distibution over each variable. A constant reference is maintained to the factor graph, so it is important that the drwnFactorGraph object not be destoryed before destroying the drwnMapInference object.

derive from drwnStdObjectIface to allow factory creation

The documentation for this class was generated from the following files: