DDOLib-Scala
API
Generated with
DDOLib-Scala/org.ddolibscala/org.ddolibscala.tools/org.ddolibscala.tools.ddo/org.ddolibscala.tools.ddo.heuristics/org.ddolibscala.tools.ddo.heuristics.cluster/GHP

GHP

org.ddolibscala.tools.ddo.heuristics.cluster.GHP
object GHP

Generalized Hyperplane Partitioning (GHP) reduction strategy for decision diagram layers.

This class implements ReductionStrategy and clusters nodes in a layer using a distance-based partitioning method inspired by hyperplane separation. It requires a problem-specific StateDistance function to compute distances between states.

The GHP strategy works by:

  1. Selecting two distant pivot nodes from the layer ;
  2. Assigning each remaining node to the cluster of the closer pivot ;
  3. Recursively splitting clusters until the desired number of clusters (maxWidth) is reached

A random number generator is used for tie-breaking and initial shuffling of the layer.

Attributes

See also
Graph
Supertypes
class Object
trait Matchable
class Any
Self type
GHP.type

Members list

Value members

Concrete methods

def apply[T](distance: StateDistance[T], seed: Long = ...): GHP[T]

Constructs a GHP reduction strategy with a specified random seed.

Constructs a GHP reduction strategy with a specified random seed.

Type parameters

T

the type of states in the decision diagram

Value parameters

distance

the distance function used to compare states

seed

the seed used by the random number generator (random by default)

Attributes

Returns

a GHP reduction strategy

Generated with