def Simulate(self, simulation_loop=1, state_value_report=True):
for looptime in range(simulation_loop):
R = 0
is_end = False
next_feature = False
current_feature = -1
current_label = -1
self.Reset()
while True:
if is_end:
Update.MonteCarlo_Update(R, self.state_list,
self.state_action_label_value_map)
break
else:
next_feature = Select.MonteCarlo_Epsilon_Select(
self.feature_remaining, current_feature, current_label,
self.state_action_label_value_map)
Select.Erase_Feature(self.feature_remaining, next_feature)
self.hypo_remaining_set = Observe.Observe_Subset(
self.true_hypothesis, self.hypo_remaining_set,
next_feature)
Observe.Clear_Overlap(self.feature_remaining,
self.hypo_remaining_set)
is_end = Observe.Check_End(self.hypo_remaining_set)
self.state_list.append(
(current_feature, next_feature, current_label))
R += -1
current_label = self.true_hypothesis[next_feature]
current_feature = next_feature
if state_value_report:
Report.Report_State_Value_Map(self.state_action_label_value_map)
def Apply_Policy_To_Random_Hypo(hypo_subset, number_features,
state_action_label_value_map):
R = 0
is_end = False
next_feature = 0
true_hypothesis = Generate.Get_Hypo(hypo_subset)
hypo_remaining_set = hypo_subset
feature_remaining_set = []
feature_trajectory = []
current_feature = -1
current_label = -1
for i in range(number_features):
feature_remaining_set.append(i)
while True:
if is_end:
break
else:
next_feature = Select.MonteCarlo_Select(
feature_remaining_set, current_feature, current_label,
state_action_label_value_map)
Select.Erase_Feature(feature_remaining_set, next_feature)
hypo_remaining_set = Observe.Observe_Subset(
true_hypothesis, hypo_remaining_set, next_feature)
Observe.Clear_Overlap(feature_remaining_set, hypo_remaining_set)
is_end = Observe.Check_End(hypo_remaining_set)
feature_trajectory.append(next_feature)
current_label = true_hypothesis[next_feature]
current_feature = next_feature
return feature_trajectory
