concepts.pdsketch.regression_rule.RegressionRule

class RegressionRule

Bases: object

Methods

iter_effects()
pddl_str()

Attributes

BodyItemType	alias of AchieveExpression | BindExpression | RuntimeAssignExpression | ListExpansionExpression | RegressionCommitFlag | OperatorApplicationExpression | RegressionRuleApplicationExpression | ConditionalRegressionRuleBodyExpression | LoopRegressionRuleBodyExpression
argument_names	The names of the arguments of the regression rule.
argument_types	The types of the arguments of the regression rule.
nr_arguments	The number of arguments of the regression rule.
serializability
name	The name of the regression rule.
arguments	The arguments to the regression rule.
goal_arguments	The arguments that appear in the goal of the regression rule.
binding_arguments	The arguments that do not appear in the goal of the regression rule.
preconditions	The preconditions of the regression rule.
preconditions_conjunction	The conjunction of the preconditions of the regression rule.
goal_expression	The goal expression of the regression rule.
all_effects	The side effects of the regression rule.
body	The body of the regression rule, including operator applications and achieve expressions.
always	Whether the regression rule is always applicable.
max_reorder_prefix_length	The maximum length of the prefix of the body that might need to be reordered.
max_rule_prefix_length	The maximum length of the prefix of the body that need tracking of all possible refinements.

__call__(*args)

Ground the operator with a list of arguments.

Return type:

RegressionRuleApplier

__init__(name, parameters, preconditions, goal_expression, all_effects, body, always=False)

Initialize a regression rule.

Parameters:

• name (str)

• parameters (Sequence[Variable])

• preconditions (Sequence[Precondition])

• goal_expression (ValueOutputExpression)

• all_effects (Sequence[Effect])

• body (Sequence[AchieveExpression | BindExpression | RuntimeAssignExpression | ListExpansionExpression | RegressionCommitFlag | OperatorApplicationExpression | RegressionRuleApplicationExpression | ConditionalRegressionRuleBodyExpression | LoopRegressionRuleBodyExpression])

• always (bool)

__new__(**kwargs)

iter_effects()

Return type:

Iterable[Effect]

pddl_str()

Return type:

str

BodyItemType

alias of AchieveExpression | BindExpression | RuntimeAssignExpression | ListExpansionExpression | RegressionCommitFlag | OperatorApplicationExpression | RegressionRuleApplicationExpression | ConditionalRegressionRuleBodyExpression | LoopRegressionRuleBodyExpression

all_effects: Tuple[Effect, ...]

The side effects of the regression rule.

always: bool

Whether the regression rule is always applicable.

property argument_names: Tuple[str, ...]

The names of the arguments of the regression rule.

property argument_types: Tuple[ObjectType | NamedTensorValueType, ...]

The types of the arguments of the regression rule.

arguments: Tuple[Variable, ...]

The arguments to the regression rule.

binding_arguments: Tuple[Variable, ...]

The arguments that do not appear in the goal of the regression rule.

body: Tuple[AchieveExpression | BindExpression | RuntimeAssignExpression | ListExpansionExpression | RegressionCommitFlag | OperatorApplicationExpression | RegressionRuleApplicationExpression | ConditionalRegressionRuleBodyExpression | LoopRegressionRuleBodyExpression, ...]

The body of the regression rule, including operator applications and achieve expressions.

goal_arguments: Tuple[Variable, ...]

The arguments that appear in the goal of the regression rule.

goal_expression: ValueOutputExpression

The goal expression of the regression rule.

max_reorder_prefix_length: int

The maximum length of the prefix of the body that might need to be reordered.

max_rule_prefix_length: int

The maximum length of the prefix of the body that need tracking of all possible refinements.

name: str

The name of the regression rule.

property nr_arguments: int

The number of arguments of the regression rule.

preconditions: Tuple[Precondition, ...]

The preconditions of the regression rule.

preconditions_conjunction: AndExpression

The conjunction of the preconditions of the regression rule.

property serializability