def test_search_goal_sequence_1(self):
# given
gs = GoalSequenceSearcher()
start = Perception("11111111")
goal = Perception("11111110")
empty_list = ClassifiersList()
# when
result = gs.search_goal_sequence(empty_list, start=start, goal=goal)
# then
assert result == []
def test_search_goal_sequence_4(self, cfg):
# given
gs = GoalSequenceSearcher()
start = "01111111"
goal = "00111111"
reliable_classifiers = ClassifiersList(
Classifier(condition="#1######",
action=1,
effect="#0######",
cfg=cfg))
# when
result = gs.search_goal_sequence(reliable_classifiers,
start=start,
goal=goal)
# then
assert result == [1]
def search_goal_sequence(classifiers: ClassifiersList, p0: Perception,
p1: Perception) -> List:
"""
Searches a path from start to goal using a bidirectional method in the
environmental model (i.e. the list of reliable classifiers).
Parameters
----------
classifiers: ClassifiersList
p0: Perception
start state
p1: Perception
destination state
Returns
-------
list
sequence of actions
"""
reliable = [cl for cl in classifiers if cl.is_reliable()]
gs = GoalSequenceSearcher()
return gs.search_goal_sequence(ClassifiersList(*reliable), p0, p1)