def test_text():
cg = CorpusGraph()
# 从json文件读取语料库模型
# cg.load_from_json()
# 连接mongodb建立语料库模型
cg.build_corpus()
# 保存为json文件
cg.save_as_json()
tg = TextGraph()
# 从mongodb读取句子，以便分词
# sentences = tg.get_sentences(isRandom=False)
sentences = ["准许原告肖振明撤回起诉"]
# 对句子数组建立图模型
tg.build(sentences)
# 填入边的权重
tg.fill_edge(cg)
# 输出语句图需要的json文件, path如果为None则返回json，而不保存在硬盘
tg.make_json(cg, path='./data/text.json')
def test_text():
cg = CorpusGraph()
cg.build_corpus()
cg.get_sorted_neighbour('一')
# print("###############")
# for cge in cg.corpus.edges:
# print(cge)
# break
# print('###', cg.corpus['朝'])
tg = TextGraph()
sentences = tg.get_sentences(isRandom=False)
tg.build(sentences)
tg.fill_edge(cg)
tg.make_json(cg)
#!/usr/bin/env python
# -*- coding: utf-8 -*-
from Network import CorpusGraph
from Network import TextGraph
from ResultReference import JiebaChecker
from ResultReference import ThulacChecker
from IO import DisIO
cg = CorpusGraph()
# cg.build_corpus()
#cg.save_as_json('./data/ten.json')
cg.load_from_json('./data/corpus_50k.json')
jieba_checker = JiebaChecker()
thulac_checker = ThulacChecker()
def tokenize(sentence):
tg = TextGraph()
tg.build([sentence])
tg.fill_edge(cg)
# 暂时只对单句分词
result = tg.cut()[0]
jieba_check = jieba_checker.check(sentence, result)
thulac_check = thulac_checker.check(sentence, result)
jieba_result = jieba_check["jieba_result"]
jieba_overlap = jieba_check["overlap"]
thulac_result = thulac_check["thulac_result"]
thulac_overlap = thulac_check["overlap"]
from flask import Flask
from flask import request
from flask import send_from_directory
from flask import send_file
from IO import RemoteIO
from Network import CorpusGraph
from Network import TextGraph
from ResultReference import JiebaChecker, ThulacChecker
from utl import count as time_count
import os
import json
# 从json文件建立语料库图模型
cg = CorpusGraph()
cg.load_from_json()
# 分词结果校对
jieba_checker = JiebaChecker()
thulac_checker = ThulacChecker()
rio = RemoteIO()
app = Flask(__name__, template_folder='./presentation', static_folder='./presentation')
@app.route('/')
def hello_world():
return send_file('./presentation/WordLink.html')
def make_corpus():
cg = CorpusGraph()
cg.build_corpus()
# cg.save_as_json()
cg.load_from_json()
import sys
from Network import CorpusGraph
cmds = sys.argv
cg = CorpusGraph()
if "build" in cmds and "toJson" in cmds:
cg.build_corpus()
cg.save_as_json()