for j in range(priority_list.shape[0]):
if pattern_count[j] == 0:
violation_pattern_to_search.append(priority_list[j])
# print(numpy.array(violation_pattern_to_search).shape[0])
weight_dist, sorted_pattern_distance, sorted_pop, distance_ranking, dist_mean = Distance_Ranking(
priority_list, variables, evaluation,
target_value_threshold)
weight_relation, sorted_pattern_relation, relation_ranking, reward = Relation_Ranking(
violation_pattern_to_search, searched_violation_pattern,
priority_list)
weights = [1, 0, 1]
# ratio = numpy.array(evaluation).shape[0] / (population * total_round)
# weights = [ratio, 0, 1 - ratio]
violation_pattern_ranking, overall_rank_list = Ensemble_Ranking(
distance_ranking, relation_ranking,
violation_pattern_to_search, weights)
violation_pattern_ranking_removed = violation_pattern_ranking.copy(
)
for j in range(
numpy.array(violation_pattern_ranking).shape[0]):
for k in range(
numpy.array(searched_violation_pattern).shape[0]):
if (numpy.array(
violation_pattern_ranking[j]) == numpy.array(
searched_violation_pattern[k])).all():
removed_item = violation_pattern_ranking[j]
for ll in range(
numpy.array(
violation_pattern_ranking_removed).
fileList.sort()
for i in range (population * search_round):
textname = vars_file_name + '/' + fileList[i]
pop = numpy.loadtxt(textname)
variables.append(pop)
violation_pattern_to_search = []
for j in range (priority_list.shape[0]):
if pattern_count[j] == 0:
violation_pattern_to_search.append(priority_list[j])
# print(numpy.array(violation_pattern_to_search).shape[0])
weight_dist, sorted_pattern_distance, sorted_pop = Distance_Ranking(violation_pattern_to_search, variables, evaluation)
weight_relation, sorted_pattern_relation = Relation_Ranking (violation_pattern_to_search, searched_violation_pattern, priority_list)
weights = [1, weight_dist, weight_relation]
violation_pattern_ranking = Ensemble_Ranking(sorted_pattern_distance, sorted_pattern_relation, violation_pattern_to_search, weights)
# violation_pattern_ranking = Ensemble_Ranking2(sorted_pattern_distance, violation_pattern_to_search)
# violation_pattern_ranking = sorted_pattern_distance
violation_pattern_ranking_removed = violation_pattern_ranking.copy()
for j in range(numpy.array(violation_pattern_ranking).shape[0]):
for k in range(numpy.array(searched_violation_pattern).shape[0]):
if (numpy.array(violation_pattern_ranking[j]) == numpy.array(searched_violation_pattern[k])).all():
removed_item = violation_pattern_ranking[j]
for ll in range(numpy.array(violation_pattern_ranking_removed).shape[0]):
if (numpy.array(violation_pattern_ranking_removed[ll]) == numpy.array(removed_item)).all():
del violation_pattern_ranking_removed[ll]
break
break
if numpy.array(violation_pattern_ranking_removed).shape[0] == 0:
goal_selection_flag = numpy.ones(7)