def NER_predict(msg):
# 这里指定了模型路径
model_path = '/Users/charlesxu/PycharmProjects/Chatbot_CN/Chatbot_Data/Info_Extraction_save/1562204195/checkpoints'
ckpt_file = tf.train.latest_checkpoint(model_path)
# print('>>>>>>>>>>>',ckpt_file)
paths['model_path'] = ckpt_file
model = BiLSTM_CRF(args,
embeddings,
tag2label,
word2id,
paths,
config=config)
model.build_graph()
saver = tf.train.Saver()
with tf.Session(config=config) as sess:
print('============= demo =============')
saver.restore(sess, ckpt_file)
demo_sent = msg
if demo_sent == '' or demo_sent.isspace():
print('See you next time!')
else:
demo_sent = list(demo_sent.strip())
demo_data = [(demo_sent, ['O'] * len(demo_sent))]
tag = model.demo_one(sess, demo_data)
PER, LOC, ORG = get_entity(tag, demo_sent)
print('PER: {}\nLOC: {}\nORG: {}'.format(PER, LOC, ORG))
paths['model_path'] = ckpt_file
model = BiLSTM_CRF(args, embeddings, tag2label, word2id, paths, config=config)
model.build_graph()
saver = tf.train.Saver()
with tf.Session(config=config) as sess:
print('============= demo =============')
saver.restore(sess, ckpt_file)
try:
for one_text in get_MONGO_data():
addr = one_text['addr'] # 归属地
charge = one_text['charge'] # 犯罪原因
judgementId = one_text['judgementId'] # 判决Id，唯一标示
print(judgementId)
keywords = one_text['keywords'] # 关键词
court = one_text['court'] # 法院信息
text = one_text['judge_text'] # 判决结果，是一个列表，继续循环
proponents = one_text['proponents'] # 原告
opponents = one_text['opponents'] # 被告
if text:
demo_data = [(text, ['O'] * len(text))]
tag = model.demo_one(sess, demo_data)
PER, LOC, ORG, TIM= get_entity(tag, text)
MON = get_MON_entity(text)
print('PER: {}\nLOC: {}\nORG: {}\nMON: {}\n'.format(PER, LOC, ORG, MON))
except Exception as e:
print('Info Error is', e)
def judge_owner():
LOC = get_entity()
pass
# print('Please input your sentence:')
# demo_sent = input()
all_texts = get_datas() # 数据库返回的按“一、二、三、四、”切割返回的文本
for one_text in all_texts:
uuid = one_text[0] # 获取每条数据的uuid
obligors = one_text[1] # 原告
creditors = one_text[2] # 被告
texts = one_text[3]
text_sent = split_sentence.split_sentence_thr(texts) # 分句
for text in text_sent:
print(text)
text_strip = list(text.strip()) #
demo_data = [(text, ['O'] * len(text))]
tag = model.demo_one(sess, demo_data)
PER, LOC, ORG = get_entity(tag, text)
LOC_RE = get_add(text)
# 调用money处理方法，获取金额实体
MON = get_MON_entity(text)
print('PER: {}\nLOC: {}\nORG: {}\nMON: {}\n'.format(
PER, LOC, ORG, MON))
print('LOC_RE :{}'.format(LOC_RE))
# 将数据写入ES
es.index(
index='chatbot',
doc_type='test_type',
body={
'uuid': uuid,
'text': text, # 原文