`ontolearn.search`

Node representation.

Module Contents

Classes

`Node`	Simple node.
`OENode`	OENode search tree node.
`EvoLearnerNode`	EvoLearner search tree node.
`NCESNode`	EvoLearner search tree node.
`RL_State`	Abstract search tree node.
`LBLNode`	LBL search tree node.
`LengthOrderedNode`	Abstract base class for generic types.
`HeuristicOrderedNode`	A comparator that orders the Nodes based on Heuristic, then OrderedOWLObject of the concept
`QualityOrderedNode`	QualityOrderedNode search tree node.
`SearchTreePriorityQueue`	Search tree based on priority queue.
`TreeNode`	Simple search tree node.
`DRILLSearchTreePriorityQueue`	Search tree based on priority queue.
`EvaluatedConcept`	Explicitly declare the attributes that should be returned by the evaluate_concept method of a KnowledgeBase.
`SuperProp`	Super wrapper which allows property setting & deletion. Super can't be subclassed with empty __init__ arguments.

class ontolearn.search.Node(concept: owlapy.class_expression.OWLClassExpression, length: int)[source]

Bases: _NodeConcept, _NodeLen, _NodeIndividualsCount, ontolearn.abstracts.AbstractNode

Simple node.

__slots__ = ('_concept', '_len', '_individuals_count')

__str__()[source]: Return str(self).

class ontolearn.search.OENode(concept: owlapy.class_expression.OWLClassExpression, length: int, parent_node: OENode | None = None, is_root: bool = False)[source]

Bases: _NodeConcept, _NodeLen, _NodeIndividualsCount, _NodeQuality, _NodeHeuristic, _NodeParentRef[OENode], ontolearn.abstracts.AbstractNode, ontolearn.abstracts.AbstractConceptNode, ontolearn.abstracts.AbstractOEHeuristicNode

OENode search tree node.

property h_exp: int

Get horizontal expansion.

Returns:: Horizontal expansion.

property refinement_count: int

Get the refinement count for this node.

Returns:: Refinement count.

__slots__ = ('_concept', '_len', '_individuals_count', '_quality', '_heuristic', '_parent_ref',...

renderer: ClassVar[owlapy.owl_object.OWLObjectRenderer]

increment_h_exp()[source]

__str__()[source]: Return str(self).

class ontolearn.search.EvoLearnerNode(concept: owlapy.class_expression.OWLClassExpression, length: int, individuals_count: int, quality: float, tree_length: int, tree_depth: int)[source]

Bases: _NodeConcept, _NodeLen, _NodeIndividualsCount, _NodeQuality, ontolearn.abstracts.AbstractNode, ontolearn.abstracts.AbstractConceptNode

EvoLearner search tree node.

property tree_length

property tree_depth

__slots__ = ('_concept', '_len', '_individuals_count', '_quality', '_tree_length', '_tree_depth')

__str__()[source]: Return str(self).

class ontolearn.search.NCESNode(concept: owlapy.class_expression.OWLClassExpression, length: int, individuals_count: int, quality: float)[source]

Bases: _NodeConcept, _NodeLen, _NodeIndividualsCount, _NodeQuality, ontolearn.abstracts.AbstractNode, ontolearn.abstracts.AbstractConceptNode

EvoLearner search tree node.

__slots__ = ('_concept', '_len', '_individuals_count', '_quality')

__str__()[source]: Return str(self).

class ontolearn.search.RL_State(concept: owlapy.class_expression.OWLClassExpression, parent_node: RL_State | None = None, embeddings=None, is_root: bool = False, length=None)[source]

Bases: _NodeConcept, _NodeQuality, _NodeHeuristic, ontolearn.abstracts.AbstractNode, _NodeParentRef[RL_State]

Abstract search tree node.

renderer: ClassVar[owlapy.owl_object.OWLObjectRenderer]: RL_State node.

__slots__ = ('_concept', 'embeddings', '_quality', '_heuristic', 'length', 'parent_node', 'is_root',...

__str__()[source]: Return str(self).

__lt__(other)[source]: Return self<value.

__gt__(other)[source]: Return self>value.

class ontolearn.search.LBLNode(concept: owlapy.class_expression.OWLClassExpression, length: int, individuals, parent_node: LBLNode | None = None, is_root: bool = False)[source]

Bases: _NodeIndividuals, OENode

LBL search tree node.

property children

property parent_node: LBLNode | None

Get the parent node.

Returns:: Parent node.

__slots__ = ('_children', '_individuals')

add_child(n)[source]

remove_child(n)[source]

class ontolearn.search.LengthOrderedNode(node: _N, length: int)[source]

Bases: Generic[_N]

Abstract base class for generic types.

A generic type is typically declared by inheriting from this class parameterized with one or more type variables. For example, a generic mapping type might be defined as:

class Mapping(Generic[KT, VT]):
    def __getitem__(self, key: KT) -> VT:
        ...
    # Etc.

This class can then be used as follows:

def lookup_name(mapping: Mapping[KT, VT], key: KT, default: VT) -> VT:
    try:
        return mapping[key]
    except KeyError:
        return default

__slots__ = ('node', 'len')

node: Final[_N]

len: Final[int]

__lt__(other)[source]: Return self<value.

__eq__(other)[source]: Return self==value.

class ontolearn.search.HeuristicOrderedNode(node: _N)[source]

Bases: Generic[_N]

A comparator that orders the Nodes based on Heuristic, then OrderedOWLObject of the concept

__slots__ = 'node'

node: Final[_N]

__lt__(other: _N)[source]: Return self<value.

__eq__(other: _N)[source]: Return self==value.

class ontolearn.search.QualityOrderedNode(node: OENode)[source]

QualityOrderedNode search tree node.

__slots__ = 'node'

node: Final[OENode]

__lt__(other)[source]: Return self<value.

__eq__(other)[source]: Return self==value.

class ontolearn.search.SearchTreePriorityQueue(quality_func, heuristic_func)[source]

Bases: ontolearn.abstracts.LBLSearchTree[LBLNode]

Search tree based on priority queue.

Parameters:

quality_func – An instance of a subclass of AbstractScorer that measures the quality of a node.
heuristic_func – An instance of a subclass of AbstractScorer that measures the promise of a node.

quality_func: An instance of a subclass of AbstractScorer that measures the quality of a node.

heuristic_func: An instance of a subclass of AbstractScorer that measures the promise of a node.

items_in_queue: An instance of PriorityQueue Class.

.nodes: A dictionary where keys are string representation of nodes and values are corresponding node objects.

nodes: A property method for ._nodes.

expressionTests: not being used .

str_to_obj_instance_mapping: not being used.

quality_func: AbstractScorer

heuristic_func: AbstractHeuristic

nodes: Dict[owlapy.class_expression.OWLClassExpression, LBLNode]

items_in_queue: PriorityQueue[Tuple[float, HeuristicOrderedNode[LBLNode]]]

add(n: LBLNode)[source]

Append a node into the search tree.

Parameters:: n – A Node object
Returns:: None

add_root(node, kb_learning_problem)[source]

Add the root node to the search tree.

Parameters:

node – Root node to add.
kb_learning_problem – Underlying learning problem to compare the quality to.

add_node(*, node: LBLNode, parent_node: LBLNode, kb_learning_problem: EncodedLearningProblem) → bool | None[source]

Add a node into the search tree after calculating heuristic value given its parent.

Parameters:

node – A Node object
parent_node – A Node object
kb_learning_problem – the encoded learning problem to compare the quality on

Returns:

True if node is a “goal node”, i.e. quality_metric(node)=1.0 False if node is a “weak node”, i.e. quality_metric(node)=0.0 None otherwise

Notes

node is a refinement of refined_node

get_most_promising() → LBLNode[source]

Gets the current most promising node from Queue.

Returns:: A node object
Return type:: node

get_top_n(n: int, key='quality') → List[LBLNode][source]

Gets the top n nodes determined by key from the search tree.

Returns:: A list of node objects
Return type:: top_n_predictions

clean()[source]: Reset the search tree state.

show_search_tree(root_concept: owlapy.class_expression.OWLClassExpression, heading_step: str)[source]

Debugging function to print the search tree to standard output.

Parameters:

root_concept – The tree is printed starting from this search tree node.
heading_step – Message to print at top of the output.

class ontolearn.search.TreeNode(node: _N, parent_tree_node: _TN | None = None, is_root: bool = False)[source]

Bases: Generic[_N]

Simple search tree node.

__slots__ = ('children', 'node')

node: Final[_N]

children: Set[TreeNode[_N]]

class ontolearn.search.DRILLSearchTreePriorityQueue[source]

Bases: ontolearn.abstracts.DRILLAbstractTree

Search tree based on priority queue.

Parameters:

quality_func (An instance of a subclass of AbstractScorer that measures the quality of a node.)
heuristic_func (An instance of a subclass of AbstractScorer that measures the promise of a node.)

quality_func

Type:: An instance of a subclass of AbstractScorer that measures the quality of a node.

heuristic_func

Type:: An instance of a subclass of AbstractScorer that measures the promise of a node.

items_in_queue

Type:: An instance of PriorityQueue Class.

.nodes

Type:: A dictionary where keys are string representation of nodes and values are corresponding node objects.

nodes

Type:: A property method for ._nodes.

expressionTests

Type:: not being used .

str_to_obj_instance_mapping

Type:: not being used.

add(node: RL_State)[source]

Append a node into the search tree. :param node: :type node: A RL_State object

Return type:: None

get_most_promising() → Node[source]

Gets the current most promising node from Queue.

Returns:: node
Return type:: A node object

get_top_n(n: int, key='quality') → List[Node][source]

Gets the top n nodes determined by key from the search tree.

Returns:: top_n_predictions
Return type:: A list of node objects

clean()[source]

class ontolearn.search.EvaluatedConcept[source]

Explicitly declare the attributes that should be returned by the evaluate_concept method of a KnowledgeBase.

This way, Python uses a more efficient way to store the instance attributes, which can significantly reduce the memory usage.

__slots__ = ('q', 'inds', 'ic')

class ontolearn.search.SuperProp(super_obj)[source]

Super wrapper which allows property setting & deletion. Super can’t be subclassed with empty __init__ arguments.

__getattr__(name)[source]

__setattr__(name, value)[source]: Implement setattr(self, name, value).

__delattr__(name)[source]: Implement delattr(self, name).

ontolearn.search

Module Contents

Classes

`ontolearn.search`