Implements generic message-passing algorithms on factor graphs. See derived classes for specific algorithms. More...

Inheritance diagram for drwnMessagePassingInference:

Public Member Functions
	drwnMessagePassingInference (const drwnFactorGraph &graph)

void	clear ()
	clear internally cached data (e.g., computation graph)

bool	inference ()
	run inference (or resume for iterative algorithms) and return true if converged

Public Member Functions inherited from drwnInference
	drwnInference (const drwnFactorGraph &graph)
	reference to initial clique potentials

	drwnInference (const drwnInference &inf)

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`

Static Public Attributes
static unsigned	MAX_ITERATIONS = 1000
	maximum number of iterations

Protected Attributes
vector< drwnTableFactor * >	_forwardMessages

vector< drwnTableFactor * >	_backwardMessages

vector< drwnTableFactor * >	_oldForwardMessages

vector< drwnTableFactor * >	_oldBackwardMessages

vector< drwnTableFactor * >	_intermediateFactors

vector< drwnFactorOperation * >	_computations

vector< drwnTableFactorStorage * >	_sharedStorage

Protected Attributes inherited from drwnInference
const drwnFactorGraph &	_graph

Detailed Description

Implements generic message-passing algorithms on factor graphs. See derived classes for specific algorithms.

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

Public Member Functions