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

Implements asynchronous max-product (min-sum) inference. More...

Inheritance diagram for drwnAsyncMaxProdInference:
drwnMessagePassingMAPInference drwnMAPInference

Public Member Functions

 drwnAsyncMaxProdInference (const drwnFactorGraph &graph)
- Public Member Functions inherited from drwnMessagePassingMAPInference
 drwnMessagePassingMAPInference (const drwnFactorGraph &graph)
void clear ()
 Clear internally cached data (e.g., computation graph)
std::pair< double, double > inference (drwnFullAssignment &mapAssignment)
 Run inference (or resume for iterative algorithms). Algorithms may initialize from mapAssignment if not empty. Returns an upper and lower bound (if available) of the minimum energy. The upper bound is the same as the energy of the best solution found (i.e., same as graph.getEnergy(mapAssignment)).
- Public Member Functions inherited from drwnMAPInference
 drwnMAPInference (const drwnFactorGraph &graph)
 drwnMAPInference (const drwnMAPInference &inf)

Protected Member Functions

void buildComputationGraph ()
void decodeBeliefs (drwnFullAssignment &mapAssignment)
- Protected Member Functions inherited from drwnMessagePassingMAPInference
virtual void initializeMessages ()

Additional Inherited Members

- Static Public Attributes inherited from drwnMessagePassingMAPInference
static unsigned MAX_ITERATIONS = 1000
 maximum number of iterations
static double DAMPING_FACTOR = 0.0
 damping factor for updating messages
- Protected Attributes inherited from drwnMessagePassingMAPInference
std::vector< drwnTableFactor * > _forwardMessages
std::vector< drwnTableFactor * > _backwardMessages
std::vector< drwnTableFactor * > _oldForwardMessages
std::vector< drwnTableFactor * > _oldBackwardMessages
std::vector< drwnTableFactor * > _intermediateFactors
< drwnFactorOperation * > 
< drwnTableFactorStorage * > 
- Protected Attributes inherited from drwnMAPInference
const drwnFactorGraph_graph
 reference to initial clique potentials

Detailed Description

Implements asynchronous max-product (min-sum) inference.

Since the factor graph is expected to be in energy form (i.e., negative log-probability), this algorithm is equivalent to min-sum.

Member Function Documentation

void drwnAsyncMaxProdInference::decodeBeliefs ( drwnFullAssignment mapAssignment)
improve this? consisted decoding?

Reimplemented from drwnMessagePassingMAPInference.

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