concepts.dm.pdsketch.regression_rule.gen_all_grounded_regression_rules#

gen_all_grounded_regression_rules(executor, state, continuous_values=None, regression_rule_names=None, regression_rule_filter=None, filter_static=True)[source]#

Generate all grounded regression rules applicable in an environment, given the initial state. Note that this function does not check if the action is applicable at the current state.

Parameters:

executor (PDSketchExecutor) – a PDSketchExecutor object.
state (State) – the current state.
continuous_values (Mapping[str, Iterable[TensorValue]] | None) – a dictionary mapping the typename of continuous types to a list of possible values.
regression_rule_names (Sequence[str] | None) – a list of regression rule names to generate. If None, all regression rules will be generated.
regression_rule_filter (Callable[[RegressionRuleApplier], bool] | None) – a function that takes an RegressionRuleApplier object and returns a boolean value indicating whether the action should be included in the result.
filter_static (bool) – whether to use the filter_static_actions() function to filter out static regression rules. The function will check all static predicates in the precondition list.

Return type:

List[RegressionRuleApplier]