concepts.dm.pdsketch.operator.OperatorApplier#

class OperatorApplier[source]#

Bases: object

An operator applier is essentially a grounded operator, composed of an operator and its arguments.

Methods

iter_sub_operator_appliers()

pddl_str()

replace_arguments(argument_map)

Replace the arguments of the operator applier with a map from argument names to new arguments.

Attributes

is_macro

Whether this operator is a macro operator.

name

The name of the operator.

operator

The operator.

arguments

The arguments of the grounded operator.

regression_rule

The regression rule that generates this operator applier.

__init__(operator, arguments, regression_rule=None)[source]#

Initialize an operator applier.

Parameters:
__new__(**kwargs)#
iter_sub_operator_appliers()[source]#
Return type:

Iterator[OperatorApplier]

pddl_str()[source]#
Return type:

str

replace_arguments(argument_map)[source]#

Replace the arguments of the operator applier with a map from argument names to new arguments.

Parameters:

argument_map (Mapping[str, Variable | UnnamedPlaceholder | TensorValue])

Return type:

OperatorApplier

arguments: Tuple[str | Variable | UnnamedPlaceholder | TensorValue, ...]#

The arguments of the grounded operator.

property is_macro: bool#

Whether this operator is a macro operator.

property name: str#

The name of the operator.

operator: Operator | MacroOperator#

The operator.

regression_rule: RegressionRuleApplier | None#

The regression rule that generates this operator applier. It is only used in the context of goal regression search.