def test_rule_and_params(self):
domain = XMLDomainReader.extract_domain(TestRuleAndParams.domain_file)
system = DialogueSystem(domain)
system.get_settings().show_gui = False
system.start_system()
assert system.get_content("theta_moves").to_continuous().get_function().get_mean()[0] == pytest.approx(0.2, abs=0.02)
assert system.get_content("a_u^p").get_prob("I want left") == pytest.approx(0.12, abs=0.03)
assert len(system.get_state().get_chance_node("theta_moves").get_output_node_ids()) == 1
assert system.get_state().has_chance_node("movements")
assert isinstance(system.get_state().get_chance_node("movements").get_distrib(), AnchoredRule)
t = CategoricalTableBuilder("a_u")
t.add_row("I want left", 0.8)
t.add_row("I want forward", 0.1)
system.add_content(t.build())
assert len(system.get_state().get_chance_node("theta_moves").get_output_node_ids()) == 0
assert not system.get_state().has_chance_node("movements")
assert system.get_content("theta_moves").to_continuous().get_function().get_mean()[0] == pytest.approx(2.0 / 6.0, abs=0.07)
system.add_content("a_m", "turning left")
assert system.get_content("a_u^p").get_prob("I want left") == pytest.approx(0.23, abs=0.04)
assert len(system.get_state().get_chance_node("theta_moves").get_output_node_ids()) == 1
def test_IS2013(self):
domain = XMLDomainReader.extract_domain(TestLearning.domain_file)
params = XMLStateReader.extract_bayesian_network(TestLearning.parameters_file, "parameters")
domain.set_parameters(params)
system = DialogueSystem(domain)
system.get_settings().show_gui = False
system.detach_module(ForwardPlanner)
Settings.nr_samples = Settings.nr_samples * 3
Settings.max_sampling_time = Settings.max_sampling_time * 10
system.start_system()
init_mean = system.get_content("theta_1").to_continuous().get_function().get_mean()
builder = CategoricalTableBuilder("a_u")
builder.add_row("Move(Left)", 1.0)
builder.add_row("Move(Right)", 0.0)
builder.add_row("None", 0.0)
system.add_content(builder.build())
system.get_state().remove_nodes(system.get_state().get_utility_node_ids())
system.get_state().remove_nodes(system.get_state().get_action_node_ids())
after_mean = system.get_content("theta_1").to_continuous().get_function().get_mean()
assert after_mean[0] - init_mean[0] > 0.04
assert after_mean[1] - init_mean[1] < 0.04
assert after_mean[2] - init_mean[2] < 0.04
assert after_mean[3] - init_mean[3] < 0.04
assert after_mean[4] - init_mean[4] < 0.04
assert after_mean[5] - init_mean[5] < 0.04
assert after_mean[6] - init_mean[6] < 0.04
assert after_mean[7] - init_mean[7] < 0.04
Settings.nr_samples = int(Settings.nr_samples / 3)
Settings.max_sampling_time = Settings.max_sampling_time / 10
def test_5(self):
domain2 = XMLDomainReader.extract_domain(TestRule2.domain_file3)
system2 = DialogueSystem(domain2)
system2.detach_module(ForwardPlanner)
system2.get_settings().show_gui = False
system2.start_system()
TestRule2.inference.check_util(
system2.get_state(), ["a_ml'", "a_mg'", "a_md'"],
Assignment([
Assignment("a_ml'", "SayYes"),
Assignment("a_mg'", "Nod"),
Assignment("a_md'", "None")
]), 2.4)
TestRule2.inference.check_util(
system2.get_state(), ["a_ml'", "a_mg'", "a_md'"],
Assignment([
Assignment("a_ml'", "SayYes"),
Assignment("a_mg'", "Nod"),
Assignment("a_md'", "DanceAround")
]), -0.6)
TestRule2.inference.check_util(
system2.get_state(), ["a_ml'", "a_mg'", "a_md'"],
Assignment([
Assignment("a_ml'", "SayYes"),
Assignment("a_mg'", "None"),
Assignment("a_md'", "None")
]), 1.6)
def __init__(self, arg1=None, arg2=None):
from dialogue_system import DialogueSystem
if not isinstance(arg1, DialogueSystem):
raise NotImplementedError("UNDEFINED PARAMETERS")
system = arg1
if isinstance(arg2, str):
domain = arg2
"""
Creates a new user/environment simulator.
:param system: the main dialogue system to which the simulator should connect
:param domain: the dialogue domain for the simulator simulator could
not be created
"""
domain = XMLDomainReader.extract_domain(domain)
elif isinstance(arg2, Domain):
domain = arg2
else:
raise NotImplementedError("UNDEFINED PARAMETERS")
"""
Creates a new user/environment simulator.
:param system: the main dialogue system to which the simulator should connect
:param domain: the dialogue domain for the simulator not be created
"""
self.system = system
self.domain = domain
self.simulator_state = copy(domain.get_initial_state())
self.simulator_state.set_parameters(domain.get_parameters())
self.system.change_settings(domain.get_settings())
self._lock = threading.RLock()
def test_priority(self):
system = DialogueSystem(XMLDomainReader.extract_domain(TestRule3.domain_file))
system.get_settings().show_gui = False
system.start_system()
assert system.get_content("a_u").get_prob("Opening") == pytest.approx(0.8, abs=0.01)
assert system.get_content("a_u").get_prob("Nothing") == pytest.approx(0.1, abs=0.01)
assert system.get_content("a_u").get_prob("start") == pytest.approx(0.0, abs=0.01)
assert not system.get_content("a_u").to_discrete().has_prob(ValueFactory.create("start"))
def test_template_quick(self):
domain = XMLDomainReader.extract_domain("test/data/quicktest.xml")
system = DialogueSystem(domain)
system.get_settings().show_gui = False
system.start_system()
assert system.get_content("caught").get_prob(False) == pytest.approx(1.0, abs=0.01)
assert system.get_content("caught2").get_prob(True) == pytest.approx(1.0, abs=0.01)
def test_param_6(self):
system = DialogueSystem(XMLDomainReader.extract_domain("test/data/testparams3.xml"))
system.get_settings().show_gui = False
system.start_system()
table = system.get_content("b").to_discrete()
assert len(table) == 6
assert table.get_prob("something else") == pytest.approx(0.45, abs=0.05)
assert table.get_prob("value: first with type 1") == pytest.approx(0.175, abs=0.05)
assert table.get_prob("value: second with type 2") == pytest.approx(0.05, abs=0.05)
def test_6(self):
domain2 = XMLDomainReader.extract_domain(TestRule2.domain_file4)
system2 = DialogueSystem(domain2)
system2.detach_module(ForwardPlanner)
system2.get_settings().show_gui = False
system2.start_system()
TestRule2.inference.check_prob(system2.get_state(), "A",
ValueFactory.create("[a1,a2]"), 1.0)
TestRule2.inference.check_prob(system2.get_state(), "a_u",
"Request(ball)", 0.5)
def test_incondition(self):
domain = XMLDomainReader.extract_domain(TestRule3.incondition_file)
system = DialogueSystem(domain)
system.get_settings().show_gui = False
system.start_system()
assert system.get_content("out").get_prob("val1 is in [val1, val2]") + system.get_content("out").get_prob("val1 is in [val2, val1]") == pytest.approx(0.56, abs=0.01)
assert system.get_content("out2").get_prob("this is a string is matched") == pytest.approx(0.5, abs=0.01)
def test_underspec(self):
domain = XMLDomainReader.extract_domain("test/data/underspectest.xml")
system = DialogueSystem(domain)
system.get_settings().show_gui = False
StatePruner.enable_reduction = False
system.start_system()
assert system.get_content("match").get_prob("obj_1") == pytest.approx(0.66, abs=0.05)
assert system.get_content("match").get_prob("obj_3") == pytest.approx(0.307, abs=0.05)
assert len(system.get_state().get_chance_node_ids()) == 14
StatePruner.enable_reduction = True
class TestDialogueState:
domain_file = "test/data/domain1.xml"
domain = XMLDomainReader.extract_domain(domain_file)
inference = InferenceChecks()
def test_state_copy(self):
system = DialogueSystem(TestDialogueState.domain)
system.detach_module(ForwardPlanner)
StatePruner.enable_reduction = False
system.get_settings().show_gui = False
system.start_system()
initial_state = copy(system.get_state())
rule_id = ""
for id in system.get_state().get_node("u_u2").get_output_node_ids():
if str(system.get_content(id)).find("+=HowAreYou") != -1:
rule_id = id
TestDialogueState.inference.check_prob(
initial_state, rule_id, Effect.parse_effect("a_u2+=HowAreYou"),
0.9)
TestDialogueState.inference.check_prob(initial_state, rule_id,
Effect.parse_effect("Void"),
0.1)
TestDialogueState.inference.check_prob(initial_state, "a_u2",
"[HowAreYou]", 0.2)
TestDialogueState.inference.check_prob(initial_state, "a_u2",
"[Greet, HowAreYou]", 0.7)
TestDialogueState.inference.check_prob(initial_state, "a_u2", "[]",
0.1)
StatePruner.enable_reduction = True
def test_state_copy2(self):
InferenceChecks.exact_threshold = 0.08
system = DialogueSystem(TestDialogueState.domain)
system.get_settings().show_gui = False
system.detach_module(ForwardPlanner)
system.start_system()
initial_state = copy(system.get_state())
TestDialogueState.inference.check_prob(initial_state, "a_u2",
"[HowAreYou]", 0.2)
TestDialogueState.inference.check_prob(initial_state, "a_u2",
"[Greet, HowAreYou]", 0.7)
TestDialogueState.inference.check_prob(initial_state, "a_u2", "[]",
0.1)
def test_1(self):
domain = XMLDomainReader.extract_domain(TestRule3.test1_domain_file)
inference = InferenceChecks()
inference.exact_threshold = 0.06
system = DialogueSystem(domain)
system.get_settings().show_gui = False
system.detach_module(ForwardPlanner)
StatePruner.enable_reduction = False
system.start_system()
inference.check_prob(system.get_state(), "found", "A", 0.7)
inference.check_prob(system.get_state(), "found2", "D", 0.3)
inference.check_prob(system.get_state(), "found2", "C", 0.5)
StatePruner.enable_reduction = True
def test_2(self):
inference = InferenceChecks()
domain = XMLDomainReader.extract_domain(TestRule3.test2_domain_file)
system = DialogueSystem(domain)
system.get_settings().show_gui = False
system.detach_module(ForwardPlanner)
StatePruner.enable_reduction = False
system.start_system()
inference.check_prob(system.get_state(), "graspable(obj1)", "True", 0.81)
inference.check_prob(system.get_state(), "graspable(obj2)", "True", 0.16)
inference.check_util(system.get_state(), "a_m'", "grasp(obj1)", 0.592)
StatePruner.enable_reduction = True
def __init__(self, arg1=None):
if arg1 is None:
"""
Creates a new dialogue system with an empty dialogue system
"""
self._settings = Settings() # the system setting
self._cur_state = DialogueState() # the dialogue state
self._domain = Domain() # the dialogue domain
self._paused = True # whether the system is paused or active
self._modules = [] # the set of modules attached to the system
# Inserting standard modules
system = self
self._modules.append(GUIFrame(system))
self._modules.append(DialogueRecorder(self))
if self._settings.planner == 'forward':
self.log.info("Forward planner will be used.")
self._modules.append(ForwardPlanner(self))
elif self._settings.planner == 'mcts':
self.log.info("MCTS planner will be used.")
self._modules.append(MCTSPlanner(self))
else:
raise ValueError("Not supported planner: %s" %
self._settings.planner)
self._init_lock()
elif isinstance(arg1, Domain):
domain = arg1
"""
Creates a new dialogue system with the provided dialogue domain
:param domain: the dialogue domain to employ
"""
self.__init__()
self.change_domain(domain)
elif isinstance(arg1, str):
domain_file = arg1
"""
Creates a new dialogue system with the provided dialogue domain
:param domain_file: the dialogue domain to employ
"""
self.__init__()
self.change_domain(XMLDomainReader.extract_domain(domain_file))
def refresh_domain(self):
"""
Refreshes the dialogue domain by rereading its source file (in case it has been changed by the user).
"""
if self._domain.is_empty():
return
src_file = self._domain.get_source_file().get_path()
try:
self._domain = XMLDomainReader.extract_domain(src_file)
self.change_settings(self._domain.get_settings())
self.display_comment("Dialogue domain successfully updated")
except Exception as e:
self.log.critical("Cannot refresh domain %s" % e)
self.display_comment("Syntax error: %s" % e)
self._domain = Domain()
self._domain.set_source_file(src_file)
def open_domain(self, domain=False):
if domain is False:
fname = QFileDialog.getOpenFileName(self, 'Open File')
domain = XMLDomainReader.extract_domain(fname[0])
domain_name = str(domain).split('.')[0]
self.chatlog.clear()
self.chatlog.append("[%s domain successfully attached]\n" %
domain_name)
self.chatlog.setAlignment(Qt.AlignLeft)
self._system.change_domain(domain)
self.save_as.setEnabled(True)
self.reset.setEnabled(True)
self.pause.setEnabled(True)
self.export_act1.setEnabled(True)
else:
self._system.change_domain(domain)
def test_param1(self):
domain = XMLDomainReader.extract_domain(TestDemo.domain_file)
params = XMLStateReader.extract_bayesian_network(
TestDemo.param_file, "parameters")
domain.set_parameters(params)
system = DialogueSystem(domain)
system.get_settings().show_gui = False
system.detach_module(ForwardPlanner)
system.get_settings().show_gui = False
system.start_system()
system.add_content("a_m", "AskRepeat")
t = CategoricalTableBuilder("a_u")
t.add_row("DoA", 0.7)
t.add_row("a_u", 0.2)
t.add_row("a_u", 0.1)
system.add_content(t.build())
for i in range(3000):
print((system.get_state().get_chance_node("theta").sample()
).get_array()[0])
def test_4(self):
domain2 = XMLDomainReader.extract_domain(TestRule2.domain_file2)
system2 = DialogueSystem(domain2)
system2.get_settings().show_gui = False
system2.detach_module(ForwardPlanner)
system2.start_system()
TestRule2.inference.check_util(
system2.get_state(), ["a_m3'", "obj(a_m3)'"],
Assignment(
[Assignment("a_m3'", "Do"),
Assignment("obj(a_m3)'", "A")]), 0.3)
TestRule2.inference.check_util(
system2.get_state(), ["a_m3'", "obj(a_m3)'"],
Assignment(
[Assignment("a_m3'", "Do"),
Assignment("obj(a_m3)'", "B")]), -1.7)
TestRule2.inference.check_util(
system2.get_state(), ["a_m3'", "obj(a_m3)'"],
Assignment([
Assignment("a_m3'", "SayHi"),
Assignment("obj(a_m3)'", "None")
]), -0.9)
def test1(self):
domain = XMLDomainReader.extract_domain(TestIncremental.domain_file)
system = DialogueSystem(domain)
# NEED GUI & Recording
system.get_settings().show_gui = False
# system.get_settings().recording = Settings.Recording.ALL
system.start_system()
system.add_content(system.get_settings().user_speech, "busy")
system.add_incremental_content(SingleValueDistribution("u_u", "go"),
False)
sleep(0.1)
assert ValueFactory.create("go") in system.get_content(
"u_u").get_values()
t = CategoricalTableBuilder("u_u")
t.add_row("forward", 0.7)
t.add_row("backward", 0.2)
system.add_incremental_content(t.build(), True)
sleep(0.1)
assert ValueFactory.create("go forward") in system.get_content(
"u_u").get_values()
assert system.get_content("u_u").get_prob(
"go backward") == pytest.approx(0.2, abs=0.001)
assert system.get_state().has_chance_node("nlu")
system.add_content(system.get_settings().user_speech, "None")
assert len(system.get_state().get_chance_nodes()) == 7
system.add_incremental_content(
SingleValueDistribution("u_u", "please"), True)
assert system.get_state().get_evidence().contains_pair(
"=_a_u", ValueFactory.create(True))
assert system.get_content("u_u").get_prob(
"go please") == pytest.approx(0.1, abs=0.001)
assert system.get_state().has_chance_node("nlu")
system.get_state().set_as_committed("u_u")
assert not system.get_state().has_chance_node("nlu")
t2 = CategoricalTableBuilder("u_u")
t2.add_row("I said go backward", 0.3)
system.add_incremental_content(t2.build(), True)
assert system.get_content("a_u").get_prob(
"Request(Backward)") == pytest.approx(0.82, abs=0.05)
assert ValueFactory.create("I said go backward") in system.get_content(
"u_u").get_values()
assert system.get_state().has_chance_node("nlu")
system.get_state().set_as_committed("u_u")
assert not system.get_state().has_chance_node("nlu")
system.add_incremental_content(
SingleValueDistribution("u_u", "yes that is right"), False)
assert ValueFactory.create("yes that is right") in system.get_content(
"u_u").get_values()
class TestPlanning:
domain_file = "test/data/domain3.xml"
domain_file2 = "test/data/basicplanning.xml"
domain_file3 = "test/data/planning2.xml"
settings_file = "test/data/settings_test2.xml"
inference = InferenceChecks()
domain = XMLDomainReader.extract_domain(domain_file)
domain2 = XMLDomainReader.extract_domain(domain_file2)
domain3 = XMLDomainReader.extract_domain(domain_file3)
def test_planning(self):
system = DialogueSystem(TestPlanning.domain)
system.get_settings().show_gui = False
system.start_system()
assert len(system.get_state().get_nodes()) == 3
assert len(system.get_state().get_chance_nodes()) == 3
assert len(system.get_state().get_evidence().get_variables()) == 0
TestPlanning.inference.check_prob(system.get_state(), "a_m3", "Do",
1.0)
TestPlanning.inference.check_prob(system.get_state(), "obj(a_m3)", "A",
1.0)
def test_planning2(self):
system = DialogueSystem(TestPlanning.domain2)
system.get_settings().show_gui = False
system.start_system()
assert len(system.get_state().get_node_ids()) == 2
assert not system.get_state().has_chance_node("a_m")
def test_planning3(self):
system = DialogueSystem(TestPlanning.domain2)
system.get_settings().show_gui = False
system.get_settings().horizon = 2
system.start_system()
TestPlanning.inference.check_prob(system.get_state(), "a_m",
"AskRepeat", 1.0)
def test_planning4(self):
system = DialogueSystem(TestPlanning.domain3)
system.get_settings().show_gui = False
system.get_settings().horizon = 3
system.start_system()
t1 = CategoricalTableBuilder("a_u")
t1.add_row("Ask(Coffee)", 0.95)
t1.add_row("Ask(Tea)", 0.02)
system.add_content(t1.build())
TestPlanning.inference.check_prob(system.get_state(), "a_m",
"Do(Coffee)", 1.0)
def test_planning5(self):
system = DialogueSystem(TestPlanning.domain3)
system.get_settings().show_gui = False
system.get_settings().horizon = 3
system.start_system()
t1 = CategoricalTableBuilder("a_u")
t1.add_row("Ask(Coffee)", 0.3)
t1.add_row("Ask(Tea)", 0.3)
system.add_content(t1.build())
TestPlanning.inference.check_prob(system.get_state(), "a_m",
"AskRepeat", 1.0)
class TestFlightBooking:
domain = XMLDomainReader.extract_domain("test/data/example-flightbooking.xml")
def test_dialogue(self):
system = DialogueSystem(TestFlightBooking.domain)
system.get_settings().show_gui = False
system.start_system()
assert str(system.get_content("u_m").get_best()).find("your destination?") != -1
u_u = dict()
u_u["to Bergen"] = 0.4
u_u["to Bethleem"] = 0.2
system.add_user_input(u_u)
assert system.get_content("a_u").get_prob("[Inform(Airport,Bergen)]") == pytest.approx(0.833, abs=0.01)
assert len(system.get_content("a_u").get_values()) == pytest.approx(3, abs=0.01)
assert system.get_state().query_prob("a_u", False).get_prob("[Other]") == pytest.approx(0.055, abs=0.01)
assert str(system.get_content("a_m").to_discrete().get_best()) == "Confirm(Destination,Bergen)"
u_u.clear()
u_u["yes exactly"] = 0.8
system.add_user_input(u_u)
assert system.get_content("a_u").get_prob("[Confirm]") == pytest.approx(0.98, abs=0.01)
assert system.get_content("Destination").get_prob("Bergen") == pytest.approx(1.0, abs=0.01)
assert str(system.get_content("a_m").to_discrete().get_best()) == "Ground(Destination,Bergen)"
assert str(system.get_content("u_m").get_best()).find("your departure?") != -1
u_u.clear()
u_u["to Stockholm"] = 0.8
system.add_user_input(u_u)
assert system.get_content("a_u").get_prob("[Other]") == pytest.approx(0.8, abs=0.01)
assert str(system.get_content("a_m-prev").to_discrete().get_best()) == "Ground(Destination,Bergen)"
assert len(system.get_content("Destination").to_discrete().get_values()) == 1
assert system.get_content("Destination").get_prob("Bergen") == pytest.approx(1.0, abs=0.01)
assert not system.get_state().has_chance_node("Departure")
assert str(system.get_content("a_m").get_best()) == "AskRepeat"
assert str(system.get_content("u_m").get_best()).find("you repeat?") != -1
u_u.clear()
u_u["to Sandefjord then"] = 0.6
system.add_user_input(u_u)
assert system.get_content("a_u").get_prob("None") == pytest.approx(0.149, abs=0.05)
assert system.get_content("Departure").get_prob("Sandefjord") == pytest.approx(0.88, abs=0.05)
assert str(system.get_content("a_m").to_discrete().get_best()) == "Confirm(Departure,Sandefjord)"
assert str(system.get_content("u_m").get_best()).find("that correct?") != -1
u_u.clear()
u_u["no to Trondheim sorry"] = 0.08
system.add_user_input(u_u)
assert system.get_content("a_u").get_prob("[Inform(Airport,Trondheim),Disconfirm]") == pytest.approx(0.51, abs=0.01)
assert system.get_content("Departure").get_prob("Trondheim") == pytest.approx(0.51, abs=0.05)
assert str(system.get_content("a_m").to_discrete().get_best()) == "AskRepeat"
assert str(system.get_content("u_m").get_best()).find("repeat?") != -1
u_u.clear()
u_u["to Trondheim"] = 0.3
u_u["Sandefjord"] = 0.1
system.add_user_input(u_u)
assert system.get_content("a_u").get_prob("[Inform(Airport,Trondheim)]") == pytest.approx(0.667, abs=0.01)
assert system.get_content("Destination").get_prob("Bergen") == pytest.approx(1.0, abs=0.01)
assert system.get_content("Departure").get_prob("Trondheim") == pytest.approx(0.89, abs=0.01)
assert str(system.get_content("a_m").to_discrete().get_best()) == "Confirm(Departure,Trondheim)"
u_u.clear()
u_u["yes exactly that's it"] = 0.8
system.add_user_input(u_u)
assert str(system.get_content("a_m").to_discrete().get_best()) == "Ground(Departure,Trondheim)"
assert str(system.get_content("u_m").get_best()).find("which date") != -1
u_u.clear()
u_u["that will be on May 26"] = 0.4
u_u["this will be on May 24"] = 0.2
system.add_user_input(u_u)
assert system.get_content("a_u").get_prob("[Inform(Date,May,24)]") == pytest.approx(0.2, abs=0.01)
assert system.get_content("a_u").get_prob("[Inform(Date,May,26)]") == pytest.approx(0.4, abs=0.01)
assert system.get_content("Destination").get_prob("Bergen") == pytest.approx(1.0, abs=0.01)
assert system.get_content("Departure").get_prob("Trondheim") == pytest.approx(1.0, abs=0.01)
assert system.get_content("Date").get_prob("May 26") == pytest.approx(0.4, abs=0.01)
assert system.get_content("Date").get_prob("May 24") == pytest.approx(0.2, abs=0.01)
assert str(system.get_content("a_m").to_discrete().get_best()) == "AskRepeat"
assert str(system.get_content("a_m-prev").to_discrete().get_best()) == "Ground(Departure,Trondheim)"
u_u.clear()
u_u["May 24"] = 0.5
u_u["Mayday four"] = 0.5
system.add_user_input(u_u)
assert system.get_content("a_u").get_prob("[Inform(Date,May,24)]") == pytest.approx(0.82, abs=0.05)
assert system.get_content("a_u").get_prob("[Inform(Number,4)]") == pytest.approx(0.176, abs=0.01)
assert system.get_content("Date").get_prob("May 26") == pytest.approx(0.02, abs=0.01)
assert system.get_content("Date").get_prob("May 24") == pytest.approx(0.94, abs=0.01)
assert system.get_state().has_chance_node("a_m")
assert str(system.get_content("a_m-prev").to_discrete().get_best()) == "AskRepeat"
assert str(system.get_content("a_m").to_discrete().get_best()) == "Ground(Date,May 24)"
assert str(system.get_content("u_m").get_best()).find("return trip") != -1
u_u.clear()
u_u["no thanks"] = 0.9
system.add_user_input(u_u)
assert str(system.get_content("u_m").get_best()).find("to order tickets?") != -1
assert system.get_content("ReturnDate").get_prob("NoReturn") == pytest.approx(1.0, abs=0.01)
assert system.get_content("current_step").get_prob("MakeOffer") == pytest.approx(1.0, abs=0.01)
assert str(system.get_content("a_m").get_best()) == "MakeOffer(179)"
u_u.clear()
u_u["yes"] = 0.02
system.add_user_input(u_u)
assert system.get_content("a_u").get_prob("[Confirm]") == pytest.approx(0.177, abs=0.01)
assert system.get_content("current_step").get_prob("MakeOffer") == pytest.approx(1.0, abs=0.01)
assert not system.get_state().has_chance_node("a_m")
u_u.clear()
u_u["yes"] = 0.8
system.add_user_input(u_u)
assert system.get_content("a_u").get_prob("[Confirm]") == pytest.approx(0.978, abs=0.01)
assert system.get_content("current_step").get_prob("NbTickets") == pytest.approx(1.0, abs=0.01)
assert str(system.get_content("u_m").get_best()).find("many tickets") != -1
u_u.clear()
u_u["uh I don't know me"] = 0.6
system.add_user_input(u_u)
assert system.get_content("a_u").get_prob("[Other]") == pytest.approx(0.6, abs=0.01)
assert system.get_content("current_step").get_prob("NbTickets") == pytest.approx(1.0, abs=0.01)
assert str(system.get_content("a_m-prev").to_discrete().get_best()) == "Ground(MakeOffer)"
assert str(system.get_content("a_m").to_discrete().get_best()) == "AskRepeat"
u_u.clear()
u_u["three tickets please"] = 0.9
system.add_user_input(u_u)
assert system.get_content("a_u").get_prob("[Inform(Number,3)]") == pytest.approx(0.9, abs=0.01)
assert system.get_content("current_step").get_prob("NbTickets") == pytest.approx(1.0, abs=0.01)
assert str(system.get_content("a_m-prev").to_discrete().get_best()) == "AskRepeat"
assert str(system.get_content("a_m").to_discrete().get_best()) == "Confirm(NbTickets,3)"
u_u.clear()
u_u["no sorry two tickets"] = 0.4
u_u["sorry to tickets"] = 0.3
system.add_user_input(u_u)
assert len(system.get_content("a_u").get_values()) == pytest.approx(3, abs=0.01)
assert system.get_content("a_u").get_prob("[Disconfirm,Inform(Number,2)]") == pytest.approx(0.86, abs=0.05)
assert system.get_content("NbTickets").get_prob(2) == pytest.approx(0.86, abs=0.05)
assert system.get_content("NbTickets").get_prob(3) == pytest.approx(0.125, abs=0.05)
assert str(system.get_content("a_m-prev").to_discrete().get_best()) == "Confirm(NbTickets,3)"
assert str(system.get_content("a_m").to_discrete().get_best()) == "Confirm(NbTickets,2)"
assert system.get_content("current_step").get_prob("NbTickets") == pytest.approx(1.0, abs=0.01)
u_u.clear()
u_u["yes thank you"] = 0.75
u_u["yes mind you"] = 0.15
system.add_user_input(u_u)
assert len(system.get_content("a_u").get_values()) == pytest.approx(2, abs=0.01)
assert system.get_content("a_u").get_prob("[Confirm]") == pytest.approx(1.0, abs=0.05)
assert system.get_content("NbTickets").get_prob(2) == pytest.approx(1.0, abs=0.05)
assert system.get_content("NbTickets").get_prob(3) == pytest.approx(0.0, abs=0.05)
assert str(system.get_content("a_m-prev").to_discrete().get_best()) == "Confirm(NbTickets,2)"
assert str(system.get_content("a_m").to_discrete().get_best()) == "Ground(NbTickets,2)"
assert system.get_content("current_step").get_prob("LastConfirm") == pytest.approx(1.0, abs=0.01)
assert str(system.get_content("u_m").get_best()).find("Shall I confirm") != -1
assert str(system.get_content("u_m").get_best()).find("358 EUR") != -1
u_u.clear()
u_u["err yes"] = 0.2
system.add_user_input(u_u)
assert system.get_content("a_u").get_prob("[Confirm]") == pytest.approx(0.726, abs=0.01)
assert system.get_content("current_step").get_prob("LastConfirm") == pytest.approx(1.0, abs=0.01)
u_u.clear()
u_u["yes please confirm"] = 0.5
system.add_user_input(u_u)
assert system.get_content("a_u").get_prob("[Confirm]") == pytest.approx(0.934, abs=0.01)
assert system.get_content("current_step").get_prob("Final") == pytest.approx(1.0, abs=0.01)
assert str(system.get_content("a_m").to_discrete().get_best()) == "Book"
assert str(system.get_content("u_m").get_best()).find("additional tickets?") != -1
u_u.clear()
u_u["thanks but no thanks"] = 0.7
system.add_user_input(u_u)
assert system.get_content("a_u").get_prob("[Disconfirm]") == pytest.approx(0.97, abs=0.01)
assert system.get_content("current_step").get_prob("Close") == pytest.approx(1.0, abs=0.01)
assert str(system.get_content("u_m").get_best()).find("welcome back!") != -1
assert sorted(system.get_state().get_chance_node_ids()) == ["Date", "Departure", "Destination", "NbTickets", "ReturnDate", "TotalCost", "a_m", "a_m-prev", "a_u", "current_step", "u_m", "u_u"]
assert len(system.get_content(["Date", "Departure", "Destination", "NbTickets", "ReturnDate", "TotalCost"]).to_discrete().get_values()) == 1
assert system.get_content(["Date", "Departure", "Destination", "NbTickets", "ReturnDate", "TotalCost"]).to_discrete().get_prob(
Assignment([
Assignment("Date", "May 24"),
Assignment("Departure", "Trondheim"),
Assignment("Destination", "Bergen"),
Assignment("NbTickets", 2.),
Assignment("ReturnDate", "NoReturn"),
Assignment("TotalCost", 358.)
])
) == pytest.approx(1.0, abs=0.01)
def test_dialogue2(self):
system = DialogueSystem(TestFlightBooking.domain)
system.get_settings().show_gui = False
system.start_system()
u_u = dict()
u_u["err I don't know, where can I go?"] = 0.8
system.add_user_input(u_u)
assert system.get_content("a_u").to_discrete().get_prob("[Other]") == pytest.approx(0.8, abs=0.01)
assert str(system.get_content("a_m").get_best()) == "AskRepeat"
u_u.clear()
u_u["ah ok well I want to go to Tromsø please"] = 0.8
system.add_user_input(u_u)
assert system.get_content("a_u").to_discrete().get_prob("[Inform(Airport,Tromsø)]") == pytest.approx(0.91, abs=0.01)
assert system.get_content("Destination").to_discrete().get_prob("Tromsø") == pytest.approx(0.91, abs=0.01)
assert str(system.get_content("a_m").get_best()) == "Confirm(Destination,Tromsø)"
u_u.clear()
u_u["that's right"] = 0.6
system.add_user_input(u_u)
assert str(system.get_content("a_m").get_best()) == "Ground(Destination,Tromsø)"
assert str(system.get_content("u_m").get_best()).find("departure?") != -1
u_u.clear()
u_u["I'll be leaving from Moss"] = 0.1
system.add_user_input(u_u)
assert system.get_content("a_u").to_discrete().get_prob("[Inform(Airport,Moss)]") == pytest.approx(0.357, abs=0.01)
assert system.get_content("Destination").to_discrete().get_prob("Tromsø") == pytest.approx(1.0, abs=0.01)
assert system.get_content("Departure").to_discrete().get_prob("Moss") == pytest.approx(0.357, abs=0.01)
assert str(system.get_content("a_m").get_best()) == "AskRepeat"
u_u.clear()
u_u["I am leaving from Moss, did you get that right?"] = 0.2
u_u["Bodø, did you get that right?"] = 0.4
system.add_user_input(u_u)
assert system.get_content("a_u").to_discrete().get_prob("[Confirm,Inform(Airport,Moss)]") == pytest.approx(0.72, abs=0.01)
assert system.get_content("Departure").to_discrete().get_prob("Moss") == pytest.approx(0.88, abs=0.01)
assert system.get_content("Departure").to_discrete().get_prob("Bodø") == pytest.approx(0.10, abs=0.01)
assert str(system.get_content("a_m").get_best()) == "Confirm(Departure,Moss)"
u_u.clear()
u_u["yes"] = 0.6
system.add_user_input(u_u)
assert str(system.get_content("a_m").get_best()) == "Ground(Departure,Moss)"
assert str(system.get_content("u_m").get_best()).find("which date") != -1
u_u.clear()
u_u["March 16"] = 0.7
u_u["March 60"] = 0.2
system.add_user_input(u_u)
assert system.get_content("a_u").to_discrete().get_prob("[Inform(Date,March,16)]") == pytest.approx(0.7, abs=0.01)
assert system.get_content("a_u").to_discrete().get_prob("[Other]") == pytest.approx(0.2, abs=0.01)
assert str(system.get_content("a_m").get_best()) == "AskRepeat"
u_u.clear()
u_u["March 16"] = 0.05
u_u["March 60"] = 0.3
system.add_user_input(u_u)
assert str(system.get_content("a_m").get_best()) == "Confirm(Date,March 16)"
u_u.clear()
u_u["yes"] = 0.6
system.add_user_input(u_u)
assert str(system.get_content("a_m").get_best()) == "Ground(Date,March 16)"
assert str(system.get_content("u_m").get_best()).find("return trip?") != -1
u_u.clear()
u_u["err"] = 0.1
system.add_user_input(u_u)
assert not system.get_state().has_chance_node("a_m")
u_u.clear()
u_u["yes"] = 0.3
system.add_user_input(u_u)
assert str(system.get_content("a_m").get_best()) == "AskRepeat"
u_u.clear()
u_u["yes"] = 0.5
system.add_user_input(u_u)
assert system.get_content("current_step").get_prob("ReturnDate") == pytest.approx(1.0, abs=0.01)
assert str(system.get_content("u_m").get_best()).find("travel back") != -1
u_u.clear()
u_u["on the 20th of March"] = 0.7
system.add_user_input(u_u)
assert str(system.get_content("a_m").get_best()) == "Confirm(ReturnDate,March 20)"
assert system.get_content("ReturnDate").to_discrete().get_prob("March 20") == pytest.approx(0.7, abs=0.01)
u_u.clear()
u_u["yes"] = 0.6
system.add_user_input(u_u)
assert str(system.get_content("u_m").get_best()).find("299 EUR") != -1
assert str(system.get_content("u_m").get_best()).find("to order tickets?") != -1
copy_state1 = copy(system.get_state())
u_u.clear()
u_u["no"] = 0.7
system.add_user_input(u_u)
assert str(system.get_content("a_m").get_best()) == "Ground(Cancel)"
assert str(system.get_content("current_step").get_best()) == "Final"
assert str(system.get_content("u_m").get_best()).find("additional tickets?") != -1
copy_state2 = copy(system.get_state())
assert str(copy_state2.query_prob("current_step").get_best()) == "Final"
u_u.clear()
u_u["no"] = 0.7
system.add_user_input(u_u)
assert str(copy_state2.query_prob("current_step").get_best()) == "Final"
assert str(system.get_content("a_m").get_best()) == "Ground(Close)"
assert str(system.get_content("current_step").get_best()) == "Close"
assert str(system.get_content("u_m").get_best()).find("welcome back") != -1
system.get_state().remove_nodes(system.get_state().get_chance_node_ids())
system.get_state().add_network(copy_state2)
assert str(copy_state2.query_prob("current_step").get_best()) == "Final"
u_u.clear()
u_u["yes"] = 0.7
system.add_user_input(u_u)
assert not system.get_state().has_chance_node("Destination")
assert str(system.get_content("u_m").get_best()).find("destination?") != -1
assert sorted(system.get_state().get_chance_node_ids()) == ["Destination^p", "a_m-prev", "current_step", "u_m", "u_u"]
system.add_user_input("Oslo")
assert str(system.get_content("a_m").get_best()) == "Ground(Destination,Oslo)"
system.get_state().remove_nodes(system.get_state().get_chance_node_ids())
system.get_state().add_network(copy_state1)
u_u.clear()
u_u["yes"] = 0.8
system.add_user_input(u_u)
assert system.get_content("current_step").get_prob("NbTickets") == pytest.approx(1.0, abs=0.01)
assert str(system.get_content("a_m").to_discrete().get_best()) == "Ground(MakeOffer)"
u_u.clear()
u_u["one single ticket"] = 0.9
system.add_user_input(u_u)
assert system.get_content("a_u").get_prob("[Inform(Number,1)]") == pytest.approx(0.9, abs=0.01)
assert system.get_content("current_step").get_prob("NbTickets") == pytest.approx(1.0, abs=0.01)
assert str(system.get_content("a_m-prev").to_discrete().get_best()) == "Ground(MakeOffer)"
assert str(system.get_content("a_m").to_discrete().get_best()) == "Confirm(NbTickets,1)"
u_u.clear()
u_u["yes thank you"] = 1.0
system.add_user_input(u_u)
assert len(system.get_content("a_u").get_values()) == pytest.approx(1, abs=0.01)
assert system.get_content("a_u").get_prob("[Confirm]") == pytest.approx(1.0, abs=0.05)
assert system.get_content("NbTickets").get_prob(1.) == pytest.approx(1.0, abs=0.05)
assert str(system.get_content("a_m-prev").to_discrete().get_best()) == "Confirm(NbTickets,1)"
assert str(system.get_content("a_m").to_discrete().get_best()) == "Ground(NbTickets,1)"
assert system.get_content("current_step").get_prob("LastConfirm") == pytest.approx(1.0, abs=0.01)
assert str(system.get_content("u_m").get_best()).find("Shall I confirm") != -1
assert str(system.get_content("u_m").get_best()).find("299 EUR") != -1
u_u.clear()
u_u["yes please"] = 0.5
u_u["yellow"] = 0.4
system.add_user_input(u_u)
assert system.get_content("a_u").get_prob("[Confirm]") == pytest.approx(0.9397, abs=0.01)
assert system.get_content("current_step").get_prob("Final") == pytest.approx(1.0, abs=0.01)
assert str(system.get_content("a_m").to_discrete().get_best()) == "Book"
assert str(system.get_content("u_m").get_best()).find("additional tickets?") != -1
copystate3 = copy(system.get_state())
u_u.clear()
u_u["thanks but no thanks"] = 0.7
system.add_user_input(u_u)
assert system.get_content("a_u").get_prob("[Disconfirm]") == pytest.approx(0.97, abs=0.01)
assert system.get_content("current_step").get_prob("Close") == pytest.approx(1.0, abs=0.01)
assert str(system.get_content("u_m").get_best()).find("welcome back!") != -1
assert sorted(system.get_state().get_chance_node_ids()) == ["Date", "Departure", "Destination", "NbTickets", "ReturnDate", "TotalCost", "a_m", "a_m-prev", "a_u", "current_step", "u_m", "u_u"]
assert len(system.get_content(["Date", "Departure", "Destination", "NbTickets", "ReturnDate", "TotalCost"]).to_discrete().get_values()) == 1
assert system.get_content(["Date", "Departure", "Destination", "NbTickets", "ReturnDate", "TotalCost"]).to_discrete().get_prob(
Assignment([
Assignment("Date", "March 16"),
Assignment("Departure", "Moss"),
Assignment("Destination", "Tromsø"),
Assignment("NbTickets", 1.),
Assignment("ReturnDate", "March 20"),
Assignment("TotalCost", 299.)
])
) == pytest.approx(1.0, abs=0.01)
system.get_state().remove_nodes(system.get_state().get_chance_node_ids()) ##
system.get_state().add_network(copystate3)
u_u.clear()
u_u["yes"] = 0.7
system.add_user_input(u_u)
assert not system.get_state().has_chance_node("Destination")
assert str(system.get_content("u_m").get_best()).find("destination?") != -1
assert sorted(system.get_state().get_chance_node_ids()) == ["Destination^p", "a_m-prev", "current_step", "u_m", "u_u"]
system.add_user_input("Oslo")
assert str(system.get_content("a_m").get_best()) == "Ground(Destination,Oslo)"
class TestPruning:
domainFile = "test/data/domain1.xml"
domain = XMLDomainReader.extract_domain(domainFile)
inference = InferenceChecks()
InferenceChecks.exact_threshold = 0.1
InferenceChecks.sampling_threshold = 0.1
system = DialogueSystem(domain)
system.get_settings().show_gui = False
system.start_system()
def test_pruning0(self):
assert len(TestPruning.system.get_state().get_node_ids()) == 15
assert len(TestPruning.system.get_state().get_evidence().get_variables()) == 0
def test_pruning1(self):
TestPruning.inference.check_prob(TestPruning.system.get_state(), "a_u", "Greeting", 0.8)
TestPruning.inference.check_prob(TestPruning.system.get_state(), "a_u", "None", 0.2)
def test_pruning2(self):
TestPruning.inference.check_prob(TestPruning.system.get_state(), "i_u", "Inform", 0.7 * 0.8)
TestPruning.inference.check_prob(TestPruning.system.get_state(), "i_u", "None", 1 - 0.7 * 0.8)
def test_pruning3(self):
TestPruning.inference.check_prob(TestPruning.system.get_state(), "direction", "straight", 0.79)
TestPruning.inference.check_prob(TestPruning.system.get_state(), "direction", "left", 0.20)
TestPruning.inference.check_prob(TestPruning.system.get_state(), "direction", "right", 0.01)
def test_pruning4(self):
TestPruning.inference.check_prob(TestPruning.system.get_state(), "o", "and we have var1=value2", 0.3)
TestPruning.inference.check_prob(TestPruning.system.get_state(), "o", "and we have localvar=value1", 0.2)
TestPruning.inference.check_prob(TestPruning.system.get_state(), "o", "and we have localvar=value3", 0.31)
def test_pruning5(self):
TestPruning.inference.check_prob(TestPruning.system.get_state(), "o2", "here is value1", 0.35)
TestPruning.inference.check_prob(TestPruning.system.get_state(), "o2", "and value2 is over there", 0.07)
TestPruning.inference.check_prob(TestPruning.system.get_state(), "o2", "value3, finally", 0.28)
def test_pruning6(self):
initial_state = copy(TestPruning.system.get_state())
builder = CategoricalTableBuilder("var1")
builder.add_row("value2", 0.9)
TestPruning.system.get_state().add_to_state(builder.build())
TestPruning.inference.check_prob(TestPruning.system.get_state(), "o", "and we have var1=value2", 0.3)
TestPruning.inference.check_prob(TestPruning.system.get_state(), "o", "and we have localvar=value1", 0.2)
TestPruning.inference.check_prob(TestPruning.system.get_state(), "o", "and we have localvar=value3", 0.31)
TestPruning.system.get_state().reset(initial_state)
def test_pruning7(self):
TestPruning.inference.check_prob(TestPruning.system.get_state(), "a_u2", "[Greet, HowAreYou]", 0.7)
TestPruning.inference.check_prob(TestPruning.system.get_state(), "a_u2", "none", 0.1)
TestPruning.inference.check_prob(TestPruning.system.get_state(), "a_u2", "[HowAreYou]", 0.2)
def test_pruning8(self):
initial_state = copy(TestPruning.system.get_state())
created_nodes = SortedSet()
for node_id in TestPruning.system.get_state().get_node_ids():
if node_id.find("a_u3^") != -1:
created_nodes.add(node_id)
assert len(created_nodes) == 2
values = TestPruning.system.get_state().get_node(created_nodes[0] + "").get_values()
if ValueFactory.create("Greet") in values:
greet_node = created_nodes[0] # created_nodes.first()
howareyou_node = created_nodes[-1] # created_nodes.last()
else:
greet_node = created_nodes[-1] # created_nodes.last()
howareyou_node = created_nodes[0] # created_nodes.first()
TestPruning.inference.check_prob(TestPruning.system.get_state(), "a_u3", "[" + howareyou_node + "," + greet_node + "]", 0.7)
TestPruning.inference.check_prob(TestPruning.system.get_state(), "a_u3", "none", 0.1)
TestPruning.inference.check_prob(TestPruning.system.get_state(), "a_u3", "[" + howareyou_node + "]", 0.2)
TestPruning.inference.check_prob(TestPruning.system.get_state(), greet_node + "", "Greet", 0.7)
TestPruning.inference.check_prob(TestPruning.system.get_state(), howareyou_node + "", "HowAreYou", 0.9)
TestPruning.system.get_state().reset(initial_state)
class TestParameters:
domain_file = "test/data/testwithparams.xml"
domain_file2 = "test/data/testwithparams2.xml"
param_file = "test/data/params.xml"
params = XMLStateReader.extract_bayesian_network(param_file, "parameters")
domain1 = XMLDomainReader.extract_domain(domain_file)
domain1.set_parameters(params)
domain2 = XMLDomainReader.extract_domain(domain_file2)
domain2.set_parameters(params)
inference = InferenceChecks()
def test_param_1(self):
InferenceChecks.exact_threshold = 0.1
system = DialogueSystem(TestParameters.domain1)
system.detach_module(ForwardPlanner)
system.get_settings().show_gui = False
assert system.get_state().has_chance_node("theta_1")
TestParameters.inference.check_cdf(system.get_state(), "theta_1", 0.5, 0.5)
TestParameters.inference.check_cdf(system.get_state(), "theta_1", 5., 0.99)
TestParameters.inference.check_cdf(system.get_state(), "theta_2", 1., 0.07)
TestParameters.inference.check_cdf(system.get_state(), "theta_2", 2., 0.5)
system.start_system()
system.add_content("u_u", "hello there")
utils = ((SamplingAlgorithm()).query_util(system.get_state(), "u_m'"))
assert utils.get_util(Assignment("u_m'", "yeah yeah talk to my hand")) > 0
assert utils.get_util(Assignment("u_m'", "so interesting!")) > 1.7
assert utils.get_util(Assignment("u_m'", "yeah yeah talk to my hand")) < utils.get_util(Assignment("u_m'", "so interesting!"))
assert len(system.get_state().get_node_ids()) == 11
def test_param_2(self):
InferenceChecks.exact_threshold = 0.1
system = DialogueSystem(TestParameters.domain1)
system.detach_module(ForwardPlanner)
system.get_settings().show_gui = False
assert system.get_state().has_chance_node("theta_3")
TestParameters.inference.check_cdf(system.get_state(), "theta_3", 0.6, 0.0)
TestParameters.inference.check_cdf(system.get_state(), "theta_3", 0.8, 0.5)
TestParameters.inference.check_cdf(system.get_state(), "theta_3", 0.95, 1.0)
system.start_system()
system.add_content("u_u", "brilliant")
distrib = system.get_content("a_u")
assert distrib.get_prob("approval") == pytest.approx(0.8, abs=0.05)
def test_param_3(self):
system = DialogueSystem(TestParameters.domain1)
system.detach_module(ForwardPlanner)
system.get_settings().show_gui = False
system.start_system()
rules = TestParameters.domain1.get_models()[0].get_rules()
outputs = rules[1].get_output(Assignment("u_u", "no no"))
o = Effect(BasicEffect("a_u", "disapproval"))
assert isinstance(outputs.get_parameter(o), SingleParameter)
input = Assignment("theta_4", ValueFactory.create("[0.36, 0.64]"))
assert outputs.get_parameter(o).get_value(input) == pytest.approx(0.64, abs=0.01)
system.get_state().remove_nodes(system.get_state().get_action_node_ids())
system.get_state().remove_nodes(system.get_state().get_utility_node_ids())
system.add_content("u_u", "no no")
assert system.get_state().query_prob("theta_4").to_continuous().get_function().get_mean()[0] == pytest.approx(0.36, abs=0.1)
assert system.get_content("a_u").get_prob("disapproval") == pytest.approx(0.64, abs=0.1)
def test_param_4(self):
system = DialogueSystem(TestParameters.domain1)
system.detach_module(ForwardPlanner)
system.get_settings().show_gui = False
system.start_system()
rules = TestParameters.domain1.get_models()[1].get_rules()
outputs = rules[0].get_output(Assignment("u_u", "my name is"))
o = Effect(BasicEffect("u_u^p", "Pierre"))
assert isinstance(outputs.get_parameter(o), SingleParameter)
input = Assignment("theta_5", ValueFactory.create("[0.36, 0.24, 0.40]"))
assert outputs.get_parameter(o).get_value(input) == pytest.approx(0.36, abs=0.01)
system.get_state().remove_nodes(system.get_state().get_action_node_ids())
system.get_state().remove_nodes(system.get_state().get_utility_node_ids())
system.add_content("u_u", "my name is")
system.get_state().remove_nodes(system.get_state().get_action_node_ids())
system.get_state().remove_nodes(system.get_state().get_utility_node_ids())
system.add_content("u_u", "Pierre")
system.get_state().remove_nodes(system.get_state().get_action_node_ids())
system.get_state().remove_nodes(system.get_state().get_utility_node_ids())
system.add_content("u_u", "my name is")
system.get_state().remove_nodes(system.get_state().get_action_node_ids())
system.get_state().remove_nodes(system.get_state().get_utility_node_ids())
system.add_content("u_u", "Pierre")
assert system.get_state().query_prob("theta_5").to_continuous().get_function().get_mean()[0] == pytest.approx(0.3, abs=0.12)
def test_param_5(self):
system = DialogueSystem(TestParameters.domain2)
system.detach_module(ForwardPlanner)
system.get_settings().show_gui = False
system.start_system()
rules = TestParameters.domain2.get_models()[0].get_rules()
outputs = rules[0].get_output(Assignment("u_u", "brilliant"))
o = Effect(BasicEffect("a_u", "approval"))
assert isinstance(outputs.get_parameter(o), ComplexParameter)
input = Assignment([Assignment("theta_6", 2.1), Assignment("theta_7", 1.3)])
assert outputs.get_parameter(o).get_value(input) == pytest.approx(0.74, abs=0.01)
system.get_state().remove_nodes(system.get_state().get_action_node_ids())
system.get_state().remove_nodes(system.get_state().get_utility_node_ids())
system.add_content("u_u", "brilliant")
assert system.get_state().query_prob("theta_6").to_continuous().get_function().get_mean()[0] == pytest.approx(1.0, abs=0.08)
assert system.get_content("a_u").get_prob("approval") == pytest.approx(0.63, abs=0.08)
assert system.get_content("a_u").get_prob("irony") == pytest.approx(0.3, abs=0.08)
def test_param_6(self):
system = DialogueSystem(XMLDomainReader.extract_domain("test/data/testparams3.xml"))
system.get_settings().show_gui = False
system.start_system()
table = system.get_content("b").to_discrete()
assert len(table) == 6
assert table.get_prob("something else") == pytest.approx(0.45, abs=0.05)
assert table.get_prob("value: first with type 1") == pytest.approx(0.175, abs=0.05)
assert table.get_prob("value: second with type 2") == pytest.approx(0.05, abs=0.05)
def test_function(self):
d = XMLDomainReader.extract_domain("test/data/relationaltest.xml")
system = DialogueSystem(d)
system.get_settings().show_gui = False
system.start_system()
def test_demo(self):
domain = XMLDomainReader.extract_domain(TestDemo.domain_file2)
system = DialogueSystem(domain)
system.get_settings().show_gui = False
system.start_system()
assert len(system.get_state().get_chance_nodes()) == 5
t = CategoricalTableBuilder("u_u")
t.add_row("hello there", 0.7)
t.add_row("hello", 0.2)
updates = system.add_content(t.build())
assert updates.issuperset({"a_u", "a_m", "u_m"})
assert str(system.get_content("u_m").get_best()) == "Hi there"
t2 = dict()
t2["move forward"] = 0.06
system.add_user_input(t2)
assert not system.get_state().has_chance_node("u_m")
t2 = dict()
t2["move forward"] = 0.45
system.add_user_input(t2)
assert str(
system.get_content("u_m").get_best()) == "OK, moving Forward"
t = CategoricalTableBuilder("u_u")
t.add_row("now do that again", 0.3)
t.add_row("move backward", 0.22)
t.add_row("move a bit to the left", 0.22)
system.add_content(t.build())
assert str(
system.get_content("u_m").get_best()) == "Sorry, could you repeat?"
t = CategoricalTableBuilder("u_u")
t.add_row("do that one more time", 0.65)
system.add_content(t.build())
assert str(
system.get_content("u_m").get_best()) == "OK, moving Forward"
system.add_content(SingleValueDistribution("perceived", "[BlueObj]"))
t = CategoricalTableBuilder("u_u")
t.add_row("what do you see", 0.6)
t.add_row("do you see it", 0.3)
system.add_content(t.build())
assert str(
system.get_content("u_m").get_best()) == "I see a blue cylinder"
t = CategoricalTableBuilder("u_u")
t.add_row("pick up the blue object", 0.75)
t.add_row("turn left", 0.12)
system.add_content(t.build())
assert str(system.get_content(
"u_m").get_best()) == "OK, picking up the blue object"
system.add_content(SingleValueDistribution("perceived", "[]"))
system.add_content(SingleValueDistribution("carried", "[BlueObj]"))
t = CategoricalTableBuilder("u_u")
t.add_row("now please move a bit forward", 0.21)
t.add_row("move backward a little bit", 0.13)
system.add_content(t.build())
assert str(system.get_content(
"u_m").get_best()) == "Should I move a bit forward?"
t = CategoricalTableBuilder("u_u")
t.add_row("yes", 0.8)
t.add_row("move backward", 0.1)
system.add_content(t.build())
assert str(system.get_content(
"u_m").get_best()) == "OK, moving Forward a little bit"
t = CategoricalTableBuilder("u_u")
t.add_row("and now move forward", 0.21)
t.add_row("move backward", 0.09)
system.add_content(t.build())
assert str(
system.get_content("u_m").get_best()) == "Should I move forward?"
t = CategoricalTableBuilder("u_u")
t.add_row("no", 0.6)
system.add_content(t.build())
assert str(
system.get_content("u_m").get_best()) == "Should I move backward?"
t = CategoricalTableBuilder("u_u")
t.add_row("yes", 0.5)
system.add_content(t.build())
assert str(
system.get_content("u_m").get_best()) == "OK, moving Backward"
t = CategoricalTableBuilder("u_u")
t.add_row("now what can you see now?", 0.7)
system.add_content(t.build())
assert str(
system.get_content("u_m").get_best()) == "I do not see anything"
t = CategoricalTableBuilder("u_u")
t.add_row("please release the object", 0.5)
system.add_content(t.build())
assert str(system.get_content(
"u_m").get_best()) == "OK, putting down the object"
t = CategoricalTableBuilder("u_u")
t.add_row("something unexpected", 0.7)
system.add_content(t.build())
assert not system.get_state().has_chance_node("u_m")
t = CategoricalTableBuilder("u_u")
t.add_row("goodbye", 0.7)
system.add_content(t.build())
assert str(
system.get_content("u_m").get_best()) == "Bye, see you next time"
class TestRule1:
domain = XMLDomainReader.extract_domain('test/data/domain1.xml')
inference = InferenceChecks()
def test_1(self):
system = DialogueSystem(TestRule1.domain)
system.detach_module(ForwardPlanner)
StatePruner.enable_reduction = False
system.get_settings().show_gui = False
system.start_system()
TestRule1.inference.check_prob(system.get_state(), "a_u", "Greeting",
0.8)
TestRule1.inference.check_prob(system.get_state(), "a_u", "None", 0.2)
StatePruner.enable_reduction = True
def test_2(self):
system = DialogueSystem(TestRule1.domain)
system.detach_module(ForwardPlanner)
StatePruner.enable_reduction = False
system.get_settings().show_gui = False
system.start_system()
TestRule1.inference.check_prob(system.get_state(), "i_u", "Inform",
0.7 * 0.8)
TestRule1.inference.check_prob(system.get_state(), "i_u", "None",
1. - 0.7 * 0.8)
StatePruner.enable_reduction = True
def test_3(self):
InferenceChecks.exact_threshold = 0.06
system = DialogueSystem(TestRule1.domain)
system.detach_module(ForwardPlanner)
StatePruner.enable_reduction = False
system.get_settings().show_gui = False
system.start_system()
TestRule1.inference.check_prob(system.get_state(), "direction",
"straight", 0.79)
TestRule1.inference.check_prob(system.get_state(), "direction", "left",
0.20)
TestRule1.inference.check_prob(system.get_state(), "direction",
"right", 0.01)
StatePruner.enable_reduction = True
def test_4(self):
system = DialogueSystem(TestRule1.domain)
system.detach_module(ForwardPlanner)
StatePruner.enable_reduction = False
system.get_settings().show_gui = False
system.start_system()
TestRule1.inference.check_prob(system.get_state(), "o",
"and we have var1=value2", 0.3)
TestRule1.inference.check_prob(system.get_state(), "o",
"and we have localvar=value1", 0.2)
TestRule1.inference.check_prob(system.get_state(), "o",
"and we have localvar=value3", 0.28)
StatePruner.enable_reduction = True
def test_5(self):
system = DialogueSystem(TestRule1.domain)
system.detach_module(ForwardPlanner)
StatePruner.enable_reduction = False
system.get_settings().show_gui = False
system.start_system()
TestRule1.inference.check_prob(system.get_state(), "o2",
"here is value1", 0.35)
TestRule1.inference.check_prob(system.get_state(), "o2",
"and value2 is over there", 0.07)
TestRule1.inference.check_prob(system.get_state(), "o2",
"value3, finally", 0.28)
StatePruner.enable_reduction = True
def test_6(self):
InferenceChecks.exact_threshold = 0.06
system = DialogueSystem(TestRule1.domain)
system.detach_module(ForwardPlanner)
system.get_settings().show_gui = False
system.start_system()
builder = CategoricalTableBuilder("var1")
builder.add_row("value2", 0.9)
system.add_content(builder.build())
TestRule1.inference.check_prob(system.get_state(), "o",
"and we have var1=value2", 0.9)
TestRule1.inference.check_prob(system.get_state(), "o",
"and we have localvar=value1", 0.05)
TestRule1.inference.check_prob(system.get_state(), "o",
"and we have localvar=value3", 0.04)
StatePruner.enable_reduction = True
def test_7(self):
system = DialogueSystem(TestRule1.domain)
system.detach_module(ForwardPlanner)
StatePruner.enable_reduction = False
system.get_settings().show_gui = False
system.start_system()
TestRule1.inference.check_prob(system.get_state(), "a_u2",
"[Greet, HowAreYou]", 0.7)
TestRule1.inference.check_prob(system.get_state(), "a_u2", "[]", 0.1)
TestRule1.inference.check_prob(system.get_state(), "a_u2",
"[HowAreYou]", 0.2)
StatePruner.enable_reduction = True
class TestRule2:
domain_file = "test/data/domain2.xml"
domain_file2 = "test/data/domain3.xml"
domain_file3 = "test/data/domain4.xml"
domain_file4 = "test/data/thesistest2.xml"
domain = XMLDomainReader.extract_domain(domain_file)
inference = InferenceChecks()
def test_1(self):
system = DialogueSystem(TestRule2.domain)
eq_factor = EquivalenceDistribution.none_prob
EquivalenceDistribution.none_prob = 0.1
old_prune_threshold = StatePruner.value_pruning_threshold
StatePruner.value_pruning_threshold = 0.0
system.get_settings().show_gui = False
system.detach_module(ForwardPlanner)
system.start_system()
TestRule2.inference.check_prob(system.get_state(), "a_u^p", "Ask(A)",
0.63)
TestRule2.inference.check_prob(system.get_state(), "a_u^p", "Ask(B)",
0.27)
TestRule2.inference.check_prob(system.get_state(), "a_u^p", "None",
0.1)
builder = CategoricalTableBuilder("a_u")
builder.add_row("Ask(B)", 0.8)
builder.add_row("None", 0.2)
system.get_state().remove_nodes(
system.get_state().get_action_node_ids())
system.get_state().remove_nodes(
system.get_state().get_utility_node_ids())
system.add_content(builder.build())
TestRule2.inference.check_prob(system.get_state(), "i_u", "Want(A)",
0.090)
TestRule2.inference.check_prob(system.get_state(), "i_u", "Want(B)",
0.91)
TestRule2.inference.check_prob(system.get_state(), "a_u^p", "Ask(B)",
0.91 * 0.9)
TestRule2.inference.check_prob(system.get_state(), "a_u^p", "Ask(A)",
0.09 * 0.9)
TestRule2.inference.check_prob(system.get_state(), "a_u^p", "None",
0.1)
TestRule2.inference.check_prob(system.get_state(), "a_u", "Ask(B)",
0.918)
TestRule2.inference.check_prob(system.get_state(), "a_u", "None",
0.081)
EquivalenceDistribution.none_prob = eq_factor
StatePruner.value_pruning_threshold = old_prune_threshold
def test_2(self):
system = DialogueSystem(TestRule2.domain)
system.get_settings().show_gui = False
system.detach_module(ForwardPlanner)
eq_factor = EquivalenceDistribution.none_prob
EquivalenceDistribution.none_prob = 0.1
old_prune_threshold = StatePruner.value_pruning_threshold
StatePruner.value_pruning_threshold = 0.0
system.start_system()
TestRule2.inference.check_prob(system.get_state(), "u_u2^p", "Do A",
0.216)
TestRule2.inference.check_prob(system.get_state(), "u_u2^p",
"Please do C", 0.027)
TestRule2.inference.check_prob(system.get_state(), "u_u2^p",
"Could you do B?", 0.054)
TestRule2.inference.check_prob(system.get_state(), "u_u2^p",
"Could you do A?", 0.162)
TestRule2.inference.check_prob(system.get_state(), "u_u2^p", "none",
0.19)
builder = CategoricalTableBuilder("u_u2")
builder.add_row("Please do B", 0.4)
builder.add_row("Do B", 0.4)
system.get_state().remove_nodes(
system.get_state().get_action_node_ids())
system.get_state().remove_nodes(
system.get_state().get_utility_node_ids())
system.add_content(builder.build())
TestRule2.inference.check_prob(system.get_state(), "i_u2", "Want(B)",
0.654)
TestRule2.inference.check_prob(system.get_state(), "i_u2", "Want(A)",
0.1963)
TestRule2.inference.check_prob(system.get_state(), "i_u2", "Want(C)",
0.0327)
TestRule2.inference.check_prob(system.get_state(), "i_u2", "none",
0.1168)
EquivalenceDistribution.none_prob = eq_factor
StatePruner.value_pruning_threshold = old_prune_threshold
def test_3(self):
system = DialogueSystem(TestRule2.domain)
system.get_settings().show_gui = False
system.detach_module(ForwardPlanner)
eq_factor = EquivalenceDistribution.none_prob
EquivalenceDistribution.none_prob = 0.1
old_prune_threshold = StatePruner.value_pruning_threshold
StatePruner.value_pruning_threshold = 0.0
system.start_system()
TestRule2.inference.check_util(system.get_state(), "a_m'", "Do(A)",
0.6)
TestRule2.inference.check_util(system.get_state(), "a_m'", "Do(B)",
-2.6)
builder = CategoricalTableBuilder("a_u")
builder.add_row("Ask(B)", 0.8)
builder.add_row("None", 0.2)
system.get_state().remove_nodes(
system.get_state().get_action_node_ids())
system.get_state().remove_nodes(
system.get_state().get_utility_node_ids())
system.add_content(builder.build())
TestRule2.inference.check_util(system.get_state(), "a_m'", "Do(A)",
-4.35)
TestRule2.inference.check_util(system.get_state(), "a_m'", "Do(B)",
2.357)
EquivalenceDistribution.none_prob = eq_factor
StatePruner.value_pruning_threshold = old_prune_threshold
def test_4(self):
domain2 = XMLDomainReader.extract_domain(TestRule2.domain_file2)
system2 = DialogueSystem(domain2)
system2.get_settings().show_gui = False
system2.detach_module(ForwardPlanner)
system2.start_system()
TestRule2.inference.check_util(
system2.get_state(), ["a_m3'", "obj(a_m3)'"],
Assignment(
[Assignment("a_m3'", "Do"),
Assignment("obj(a_m3)'", "A")]), 0.3)
TestRule2.inference.check_util(
system2.get_state(), ["a_m3'", "obj(a_m3)'"],
Assignment(
[Assignment("a_m3'", "Do"),
Assignment("obj(a_m3)'", "B")]), -1.7)
TestRule2.inference.check_util(
system2.get_state(), ["a_m3'", "obj(a_m3)'"],
Assignment([
Assignment("a_m3'", "SayHi"),
Assignment("obj(a_m3)'", "None")
]), -0.9)
def test_5(self):
domain2 = XMLDomainReader.extract_domain(TestRule2.domain_file3)
system2 = DialogueSystem(domain2)
system2.detach_module(ForwardPlanner)
system2.get_settings().show_gui = False
system2.start_system()
TestRule2.inference.check_util(
system2.get_state(), ["a_ml'", "a_mg'", "a_md'"],
Assignment([
Assignment("a_ml'", "SayYes"),
Assignment("a_mg'", "Nod"),
Assignment("a_md'", "None")
]), 2.4)
TestRule2.inference.check_util(
system2.get_state(), ["a_ml'", "a_mg'", "a_md'"],
Assignment([
Assignment("a_ml'", "SayYes"),
Assignment("a_mg'", "Nod"),
Assignment("a_md'", "DanceAround")
]), -0.6)
TestRule2.inference.check_util(
system2.get_state(), ["a_ml'", "a_mg'", "a_md'"],
Assignment([
Assignment("a_ml'", "SayYes"),
Assignment("a_mg'", "None"),
Assignment("a_md'", "None")
]), 1.6)
def test_6(self):
domain2 = XMLDomainReader.extract_domain(TestRule2.domain_file4)
system2 = DialogueSystem(domain2)
system2.detach_module(ForwardPlanner)
system2.get_settings().show_gui = False
system2.start_system()
TestRule2.inference.check_prob(system2.get_state(), "A",
ValueFactory.create("[a1,a2]"), 1.0)
TestRule2.inference.check_prob(system2.get_state(), "a_u",
"Request(ball)", 0.5)
parser.add_argument('--simulator', type=str, help='simulator file path')
args = parser.parse_args()
# Set logger
logger = logging.getLogger('PyOpenDial')
logger.setLevel(logging.DEBUG)
logger.addHandler(logging.StreamHandler())
if args.domain is not None:
system = DialogueSystem(args.domain)
else:
system = DialogueSystem()
if args.domain:
try:
domain = XMLDomainReader.extract_domain(args.domain)
system.log.info("Domain from %s successfully extracted" % args.domain)
except Exception as e:
system.display_comment("Cannot load domain: %s" % e)
domain = XMLDomainReader.extract_empty_domain(args.domain)
system.change_domain(domain)
if args.simulator:
simulator = Simulator(system,
XMLDomainReader.extract_domain(args.simulator))
system.log.info("Simulator with domain %s successfully extracted" %
args.simulator)
system.attach_module(simulator)
settings = system.get_settings()
system.change_settings(settings)
def test_domain(self):
system = DialogueSystem(
XMLDomainReader.extract_domain(
"test/data/example-step-by-step_params.xml"))
system.detach_module(ForwardPlanner)
system.get_settings().show_gui = False
system.start_system()
o1 = CategoricalTableBuilder("u_u")
o1.add_row("move a little bit left", 0.4)
o1.add_row("please move a little right", 0.3)
system.add_content(o1.build())
assert system.get_state().query_util(["a_m'"]).get_util(
Assignment("a_m'", "AskRepeat")) == pytest.approx(0.0, abs=0.3)
assert system.get_state().query_util(["a_m'"]).get_util(
Assignment("a_m'", "Move(Left)")) == pytest.approx(-0.1, abs=0.15)
system.get_state().remove_nodes(
system.get_state().get_action_node_ids())
system.get_state().remove_nodes(
system.get_state().get_utility_node_ids())
o1 = CategoricalTableBuilder("u_u")
o1.add_row("move a little bit left", 0.5)
o1.add_row("please move a little left", 0.2)
system.add_content(o1.build())
assert system.get_state().query_util(["a_m'"]).get_util(
Assignment("a_m'", "AskRepeat")) == pytest.approx(0.0, abs=0.3)
assert system.get_state().query_util(["a_m'"]).get_util(
Assignment("a_m'", "Move(Left)")) == pytest.approx(0.2, abs=0.15)
system.get_state().remove_nodes(
system.get_state().get_action_node_ids())
system.get_state().remove_nodes(
system.get_state().get_utility_node_ids())
o1 = CategoricalTableBuilder("u_u")
o1.add_row("now move right please", 0.8)
system.add_content(o1.build())
assert system.get_state().query_util(["a_m'"]).get_util(
Assignment("a_m'", "AskRepeat")) == pytest.approx(0.0, abs=0.3)
assert system.get_state().query_util(["a_m'"]).get_util(
Assignment("a_m'", "Move(Right)")) == pytest.approx(0.3, abs=0.15)
system.get_state().remove_nodes(
system.get_state().get_action_node_ids())
system.get_state().remove_nodes(
system.get_state().get_utility_node_ids())
o1 = CategoricalTableBuilder("u_u")
o1.add_row("move left", 0.7)
system.add_content(o1.build())
assert system.get_state().query_util(["a_m'"]).get_util(
Assignment("a_m'", "AskRepeat")) == pytest.approx(0.0, abs=0.3)
assert system.get_state().query_util(["a_m'"]).get_util(
Assignment("a_m'", "Move(Left)")) == pytest.approx(0.2, abs=0.15)
system.get_state().remove_nodes(
system.get_state().get_action_node_ids())
system.get_state().remove_nodes(
system.get_state().get_utility_node_ids())
o1 = CategoricalTableBuilder("u_u")
o1.add_row("turn left", 0.32)
o1.add_row("move left again", 0.3)
system.add_content(o1.build())
assert system.get_state().query_util(["a_m'"]).get_util(
Assignment("a_m'", "AskRepeat")) == pytest.approx(0.0, abs=0.3)
assert system.get_state().query_util(["a_m'"]).get_util(
Assignment("a_m'", "Move(Left)")) == pytest.approx(0.1, abs=0.15)
system.get_state().remove_nodes(
system.get_state().get_action_node_ids())
system.get_state().remove_nodes(
system.get_state().get_utility_node_ids())
o1 = CategoricalTableBuilder("u_u")
o1.add_row("and do that again", 0.0)
system.add_content(o1.build())
system.get_state().remove_nodes(
system.get_state().get_action_node_ids())
system.get_state().remove_nodes(
system.get_state().get_utility_node_ids())
o1 = CategoricalTableBuilder("u_u")
o1.add_row("turn left", 1.0)
system.add_content(o1.build())
assert system.get_state().query_util(["a_m'"]).get_util(
Assignment("a_m'", "AskRepeat")) == pytest.approx(0.0, abs=0.3)
assert system.get_state().query_util(["a_m'"]).get_util(
Assignment("a_m'", "Move(Left)")) == pytest.approx(0.5, abs=0.15)
system.get_state().remove_nodes(
system.get_state().get_action_node_ids())
system.get_state().remove_nodes(
system.get_state().get_utility_node_ids())
o1 = CategoricalTableBuilder("u_u")
o1.add_row("turn right", 0.4)
system.add_content(o1.build())
assert system.get_state().query_util(["a_m'"]).get_util(
Assignment("a_m'", "AskRepeat")) == pytest.approx(0.0, abs=0.3)
assert system.get_state().query_util(["a_m'"]).get_util(
Assignment("a_m'", "Move(Right)")) == pytest.approx(-0.1, abs=0.15)
system.get_state().remove_nodes(
system.get_state().get_action_node_ids())
system.get_state().remove_nodes(
system.get_state().get_utility_node_ids())
o1 = CategoricalTableBuilder("u_u")
o1.add_row("please turn right", 0.8)
system.add_content(o1.build())
assert system.get_state().query_util(["a_m'"]).get_util(
Assignment("a_m'", "AskRepeat")) == pytest.approx(0.0, abs=0.3)
assert system.get_state().query_util(["a_m'"]).get_util(
Assignment("a_m'", "Move(Right)")) == pytest.approx(0.3, abs=0.15)
system.get_state().remove_nodes(
system.get_state().get_action_node_ids())
system.get_state().remove_nodes(
system.get_state().get_utility_node_ids())
o1 = CategoricalTableBuilder("u_u")
o1.add_row("and turn a bit left", 0.3)
o1.add_row("move a bit left", 0.3)
system.add_content(o1.build())
assert system.get_state().query_util(["a_m'"]).get_util(
Assignment("a_m'", "AskRepeat")) == pytest.approx(0.0, abs=0.3)
assert system.get_state().query_util(["a_m'"]).get_util(
Assignment("a_m'", "Move(Left)")) == pytest.approx(0.1, abs=0.15)
system.get_state().remove_nodes(
system.get_state().get_action_node_ids())
system.get_state().remove_nodes(
system.get_state().get_utility_node_ids())
class TestRefResolution:
domain = XMLDomainReader.extract_domain("test/data/refres.xml")
def test_nlu(self):
system = DialogueSystem(TestRefResolution.domain)
system.get_settings().show_gui = False
system.start_system()
system.add_user_input("take the red box")
assert system.get_content("properties(ref_main)").get_best() == ValueFactory.create("[type=box, def=def, nb=sg, attr=red]")
system.add_user_input("take the big yellow box")
assert system.get_content("properties(ref_main)").get_best() == ValueFactory.create("[type=box, def=def, nb=sg, attr=big, attr=yellow]")
system.add_user_input("take the big and yellow box")
assert system.get_content("properties(ref_main)").get_best() == ValueFactory.create("[type=box, def=def, nb=sg, attr=big, attr=yellow]")
system.add_user_input("take the big box on your left")
assert system.get_content("properties(ref_main)").get_best() == ValueFactory.create("[rel=left(agent), type=box, def=def, nb=sg, attr=big]")
system.add_user_input("take the big box on the left")
assert system.get_content("properties(ref_main)").to_discrete().get_prob("[rel=left(agent), type=box, def=def, nb=sg, attr=big]") == pytest.approx(0.5, abs=0.01)
assert system.get_content("properties(ref_main)").to_discrete().get_prob("[rel=left(spk), type=box, def=def, nb=sg, attr=big]") == pytest.approx(0.5, abs=0.01)
system.add_user_input("take one box now")
assert system.get_content("properties(ref_main)").get_best() == ValueFactory.create("[def=indef, nb=sg, type=box]")
system.add_user_input("take the small and ugly box ")
assert system.get_content("properties(ref_main)").get_best() == ValueFactory.create("[type=box, def=def, nb=sg, attr=small, attr=ugly]")
system.add_user_input("now please pick up the book that is behind you")
assert system.get_content("properties(ref_main)").get_best() == ValueFactory.create("[type=book, def=def, nb=sg, rel=behind(ref_behind)]")
assert system.get_content("ref_behind").get_best() == ValueFactory.create("you")
assert system.get_content("ref_main").get_best() == ValueFactory.create("the book")
system.add_user_input("could you take the red ball on the desk")
assert system.get_content("properties(ref_main)").get_best() == ValueFactory.create("[type=ball, attr=red, rel=on(ref_on), def=def, nb=sg]")
assert system.get_content("ref_main").get_best() == ValueFactory.create("the red ball")
assert system.get_content("ref_on").get_best() == ValueFactory.create("the desk")
system.add_user_input("could you take the red ball next to the window")
assert system.get_content("properties(ref_main)").get_best() == ValueFactory.create("[type=ball, attr=red, rel=next to(ref_next to), def=def, nb=sg]")
assert system.get_content("ref_main").get_best() == ValueFactory.create("the red ball")
assert system.get_content("ref_next to").get_best() == ValueFactory.create("the window")
system.add_user_input("could you take the big red ball near the window to your left")
assert system.get_content("properties(ref_main)").get_best() == ValueFactory.create("[type=ball, attr=red, attr=big, rel=near(ref_near), def=def, nb=sg]")
assert system.get_content("ref_main").get_best() == ValueFactory.create("the big red ball")
assert system.get_content("properties(ref_near)").get_best() == ValueFactory.create("[type=window, rel=left(agent), def=def, nb=sg]")
assert system.get_content("ref_near").get_best() == ValueFactory.create("the window")
system.add_user_input("could you take the big red ball near the window and to your left")
assert system.get_content("properties(ref_main)").get_best() == ValueFactory.create("[type=ball, attr=red, attr=big, rel=left(agent), rel=near(ref_near), def=def, nb=sg]")
assert system.get_content("ref_main").get_best() == ValueFactory.create("the big red ball")
assert system.get_content("properties(ref_near)").get_best() == ValueFactory.create("[type=window, def=def, nb=sg]")
assert system.get_content("ref_near").get_best() == ValueFactory.create("the window")
system.add_user_input("and now pick up the books that are on top of the shelf")
assert system.get_content("properties(ref_main)").get_best() == ValueFactory.create("[type=book, rel=top(ref_top), def=def, nb=pl]")
assert system.get_content("properties(ref_top)").get_best() == ValueFactory.create("[type=shelf,def=def, nb=sg]")
system.add_user_input("and now pick up one book which is big")
assert system.get_content("properties(ref_main)").get_best() == ValueFactory.create("[type=book,def=indef, attr=big, nb=sg]")
nbest = dict()
nbest["and take the red book"] = 0.5
nbest["and take the wred hook"] = 0.1
system.add_user_input(nbest)
assert system.get_content("properties(ref_main)").get_prob("[type=book,attr=red,def=def,nb=sg]") == pytest.approx(0.5, abs=0.01)
assert system.get_content("properties(ref_main)").get_prob("[type=hook,attr=wred,def=def,nb=sg]") == pytest.approx(0.1, abs=0.01)
def test_resolution(self):
system = DialogueSystem(TestRefResolution.domain)
system.get_settings().show_gui = False
system.start_system()
system.add_user_input("take the red ball")
assert str(system.get_content("a_m").get_best()) == "Select(object_1)"
assert system.get_content("matches(ref_main)").get_prob("[object_1]") == pytest.approx(0.94, abs=0.05)
system.add_user_input("take the red object")
assert str(system.get_content("a_m").get_best()) == "AskConfirm(object_1)"
assert system.get_content("matches(ref_main)").get_prob("[object_1,object_3]") == pytest.approx(0.39, abs=0.05)
assert system.get_content("matches(ref_main)").get_prob("[object_1]") == pytest.approx(0.108, abs=0.05)
system.add_user_input("take the box")
assert str(system.get_content("a_m").get_best()) == "Select(object_2)"
assert system.get_content("matches(ref_main)").get_prob("[object_2]") == pytest.approx(0.7, abs=0.05)
assert system.get_content("matches(ref_main)").get_prob("[]") == pytest.approx(0.3, abs=0.05)
nbest = dict()
nbest["and take the ball now"] = 0.3
system.add_user_input(nbest)
assert str(system.get_content("a_m").get_best()) == "AskConfirm(object_1)"
assert system.get_content("matches(ref_main)").get_prob("[object_1]") == pytest.approx(0.27, abs=0.005)
system.add_user_input("yes")
assert str(system.get_content("a_m").get_best()) == "Select(object_1)"
system.add_user_input("pick up the newspaper")
assert str(system.get_content("a_m").get_best()) == "Failed(the newspaper)"
system.add_user_input("pick up an object")
assert str(system.get_content("a_m").get_best()).startswith("AskConfirm(object_")
assert system.get_content("matches(ref_main)").get_prob("[object_1,object_2,object_3]") == pytest.approx(1.0, abs=0.005)
system.add_user_input("no")
assert str(system.get_content("a_m").get_best()).startswith("AskConfirm(object_")
system.add_user_input("yes")
assert str(system.get_content("a_m").get_best()).startswith("Select(object_")
system.add_user_input("pick up the ball to the left of the box")
assert str(system.get_content("a_m").get_best()) == "Select(object_1)"
assert system.get_content("matches(ref_main)").get_prob("[object_1]") == pytest.approx(0.75, abs=0.05)
system.add_user_input("pick up the box to the left of the ball")
assert str(system.get_content("a_m").get_best()) == "Failed(the box)"
assert system.get_content("matches(ref_main)").get_prob("[object_2]") == pytest.approx(0.34, abs=0.05)
def test_underspec(self):
domain = XMLDomainReader.extract_domain("test/data/underspectest.xml")
system = DialogueSystem(domain)
system.get_settings().show_gui = False
StatePruner.enable_reduction = False
system.start_system()
assert system.get_content("match").get_prob("obj_1") == pytest.approx(0.66, abs=0.05)
assert system.get_content("match").get_prob("obj_3") == pytest.approx(0.307, abs=0.05)
assert len(system.get_state().get_chance_node_ids()) == 14
StatePruner.enable_reduction = True