def make_hmd_model(self, temp_list, model_name): model = hmd_pb2.HmdModel() model.lang = lang_pb2.kor model.model = model_name rules = list() for item in temp_list: category_list = item[0] dtc_rule = item[1] rule = hmd_pb2.HmdRule() rule.rule = dtc_rule rule.categories.extend(category_list) rules.append(rule) model.rules.extend(rules) self.stub.SetModel(model) model_key = hmd_pb2.ModelKey() model_key.lang = lang_pb2.kor model_key.model = model_name self.stub.GetModel(model_key)
def set_model(self, model_name, target_list): model = hmd_pb2.HmdModel() model.lang = lang_pb2.kor model.model = model_name rules_list = list() for item_dict in target_list: category = item_dict['category'] rule = item_dict['rule'] category_list = category.split('!@#$') hmd_client = hmd_pb2.HmdRule() hmd_client.rule = rule hmd_client.categories.extend(category_list) rules_list.append(hmd_client) model.rules.extend(rules_list) self.stub.SetModel(model) model_key = hmd_pb2.ModelKey() model_key.lang = lang_pb2.kor model_key.model = model_name
def set_model(self, model_name, target_file_path): model = hmd_pb2.HmdModel() model.lang = lang_pb2.kor model.model = model_name rules_list = list() with open(target_file_path) as target_file: for line in target_file: line = line.strip() line_list = line.split(self.args.file_delimiter) if len(line_list) < 2: print '[ERROR] Line field count at least two [{0}]'.format( line) continue hmd_client = hmd_pb2.HmdRule() hmd_client.rule = line_list[-1] hmd_client.categories.extend(line_list[:-1]) rules_list.append(hmd_client) model.rules.extend(rules_list) self.stub.SetModel(model) model_key = hmd_pb2.ModelKey() model_key.lang = lang_pb2.kor model_key.model = model_name
def set_model(self): model = hmd_pb2.HmdModel() model.lang = lang_pb2.kor model.model = 'news' rules = list() rule1 = hmd_pb2.HmdRule() # 형태소 분석 결과를 바탕으로 원형 단어를 이용 # ex) 안녕하세요. -> 안녕하/pa 시/ep 어/ec 요/jx ./s (ETRI 기준) # 안녕하, 시어, 요 를 이용 rule을 제작 rule1.rule = '(안녕하)' rule1.categories.extend(['level1', 'level2']) rules.append(rule1) rule2 = hmd_pb2.HmdRule() rule2.rule = '(자연)' rule2.categories.extend(['level1', 'level3']) rules.append(rule2) model.rules.extend(rules) self.stub.SetModel(model) model_key = hmd_pb2.ModelKey() model_key.lang = lang_pb2.kor model_key.model = 'news' ret_model = self.stub.GetModel(model_key) print unicode(ret_model)
def set_model2(self, file_name, model_name): model = hmd_pb2.HmdModel() model.lang = lang_pb2.kor model.model = model_name rules = list() f = open(file_name) for line in f.readlines(): tokens = line.strip().split('\t') # level1, level2, level3... levels = tokens[:-1] # Last element of tokens keyword = tokens[-1] print levels, keyword rule = hmd_pb2.HmdRule() rule.rule = keyword rule.categories.extend(levels) rules.append(rule) model.rules.extend(rules) self.stub.SetModel(model) model_key = hmd_pb2.ModelKey() model_key.lang = lang_pb2.kor model_key.model = model_name ret_model = self.stub.GetModel(model_key) print unicode(ret_model)