文本分类.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "文本挖掘是一个以半结构（如 WEB 网页）或者无结构（如纯文本）的自然语言文本为对象的数据挖掘，是从大规模文本数据集中发现隐藏的、重要的、新颖的、潜在的有用的规律的过程。，直观的说，当数据挖掘的对象完全由文本这种数据类型组成时，这个过程就称为文本挖掘。文本挖掘也称为文本数据挖掘。\n",
    "<img src=a.png height=50% width=60%>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 文本预处理技术"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "## 正则表达式"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "hidden": true,
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "import re\n",
    "string = 'The small cake smell good. But it is too small.'\n",
    "\n",
    "print(re.match(pattern='small', string=string))  # 从第一个字母开始\n",
    "print(re.match(pattern='The small', string=string))  # 从第一个字母开始\n",
    "print(re.search(pattern='small', string=string))  # 查找一个\n",
    "print(re.findall(pattern='small', string=string))  # 查找所有\n",
    "print(re.sub(pattern='small', repl='big', string=string))  # 替换"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "string = 'small smell sm.ll smll smaall sm3ll smAll smaaaall'\n",
    "\n",
    "re.findall('small', string)  # 匹配small\n",
    "re.findall('small|smell', string)  # 匹配small或者smell\n",
    "re.findall('sm.ll', string)  # .表示任意一个字符\n",
    "re.findall('sm[aesd]ll', string)  # []里的字符可以被匹配一次\n",
    "re.findall('sm[a-z]ll', string)  # [a-z]所有小写字母可以被匹配一次\n",
    "re.findall('sm[a-zA-Z0-9]ll', string)  # [a-zA-Z0-9]所有大小写字母和数字可以被匹配一次\n",
    "re.findall('sm\\.ll', string)  # 匹配.本身需要转义\n",
    "\n",
    "# 量化符号\n",
    "print(re.findall('sm[a-z]{1,2}ll', string))  # {1,2}表示前面的元素匹配一次或者两次\n",
    "print(re.findall('sm[a-z]{2}ll', string))  # {2}表示前面的元素匹配两次\n",
    "print(re.findall('sm[a-z]{2,}ll', string))  # {2}表示前面的元素匹配两次以上(贪婪)\n",
    "print(re.findall('sm[a-z]?ll', string))  # 前面的内容匹配0次或1次，同{0,1}\n",
    "print(re.findall('sm[a-z]+ll', string))  # 前面的内容匹配1次或以上，同{1,}\n",
    "print(re.findall('sm[a-z]*ll', string))  # 前面的内容匹配0次或以上，同{0,}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "string = 'My name is YangHui.'\n",
    "\n",
    "s = re.findall('My name is ([a-zA-Z]+)\\.', string)\n",
    "\n",
    "print('\\nHi!'+ 'YangHui')\n",
    "print('\\nHi!'+ s[0])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "练习"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "hidden": true,
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "rawdata = '555-1239Moe Szyslak(636) 555-0113Burns, C.Montgomery555-6542Rev. Timothy Lovejoy555 8904Ned Flanders636-555-3226Simpson,Homer5553642Dr. Julius Hibbert'\n",
    "tel = re.findall('\\(?[0-9]{0,3}\\)?[ -]?[0-9]{3}[- ]?[0-9]{4}', rawdata)\n",
    "name = re.findall('[A-Z][A-Z .,a-z]+', rawdata)\n",
    "import pandas as pd\n",
    "pd.DataFrame({'TelPhone': tel, 'Name': name})"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "作业：<br>\n",
    "1、找出以下信息中的电话号码及人名。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "rd = '李磊: 248-555-1234；韩梅梅: (313) 555-1234  迪丽热巴: (810)555-1234阿里巴巴: 734.555.9999'\n",
    "\n",
    "print(re.findall('[^0-9 ；()\\-:]{2,}', rd))\n",
    "print(re.findall('[\\u4E00-\\u9FD5]+', rd))\n",
    "\n",
    "print(re.findall('[0-9(][0-9\\- ).]+', rd))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "2、将句子中年份的大写数字改为阿拉伯数字"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "import re\n",
    "m0 = \"在一九四九年新中国成立\"\n",
    "m1 = \"比一九九零年低百分之五点二\"\n",
    "m2 = '人一九九六年击败俄军,取得实质独立'\n",
    "data = [m0, m1, m2]\n",
    "a = list('一二三四五六七八九零')\n",
    "b = list('1234567890')\n",
    "my_dict = dict(zip(a, b))\n",
    "data2 = []\n",
    "for m in data:\n",
    "    tem = re.findall('(.{4})年', m)\n",
    "    c = ''.join([my_dict[i] for i in list(tem[0])])\n",
    "    data2.append(re.sub('(.{4})年', c + '年', m))\n",
    "print(data2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "## 分词"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "### 最大正向匹配法"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "1、读取词典"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "with open('../data/字典.txt', 'r', encoding='utf-8') as f:\n",
    "    my_dict = f.readlines()\n",
    "my_dict = [i.split()[0] for i in my_dict]\n",
    "my_dict[:5]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "2、最大正向匹配法"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "sentence = '今天我来到北京清华大学'\n",
    "max_len = 5\n",
    "\n",
    "words = []\n",
    "while len(sentence) != 0:\n",
    "    tmp = sentence[:max_len]\n",
    "    while tmp not in my_dict and len(tmp) > 1:\n",
    "        tmp = tmp[:-1]\n",
    "    words.append(tmp)\n",
    "    sentence = sentence[len(tmp):]\n",
    "    \n",
    "print(words)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "### 最大逆向匹配法（练习）"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "sentence = '今天我来到北京清华大学'\n",
    "max_len = 5\n",
    "\n",
    "words = []\n",
    "while len(sentence) != 0:\n",
    "    tmp = sentence[-max_len:]\n",
    "    while tmp not in my_dict and len(tmp) > 1:\n",
    "        tmp = tmp[1:]\n",
    "    words = [tmp] + words\n",
    "    sentence = sentence[:-len(tmp)]\n",
    "print(words)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "### HMM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "p_start = {'good':0.63, 'normal':0.17, 'bad':0.2}  # 初始概率矩阵\n",
    "p_emit = {\n",
    "    'good':{'working':0.05, 'travel':0.35, 'shopping':0.35, 'running':0.25},\n",
    "    'normal':{'working':0.25, 'travel':0.25, 'shopping':0.25, 'running':0.25},\n",
    "    'bad':{'working':0.6, 'travel':0.2, 'shopping':0.05, 'running':0.15}\n",
    "}  # 发射概率矩阵\n",
    "p_trans = {\n",
    "    'good':{'good':0.5, 'normal':0.375, 'bad':0.125},\n",
    "    'normal':{'good':0.25, 'normal':0.125, 'bad':0.625},\n",
    "    'bad':{'good':0.25, 'normal':0.375, 'bad':0.375} }  # 转移概率矩阵"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "1、穷举法"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "obs = ['working', 'shopping', 'travel']\n",
    "states = ['good', 'normal', 'bad']\n",
    "V = [{}]\n",
    "\n",
    "# 初始化\n",
    "for y in states:\n",
    "    V[0][y] = p_start[y] * p_emit[y][obs[0]]\n",
    "print(V)\n",
    "\n",
    "for t in range(1, len(obs)):\n",
    "    V.append({})\n",
    "    for y in states:\n",
    "        for i, j in V[t-1].items():\n",
    "            pre_state = i.split('-')[-1]\n",
    "            V[t][i+'-'+y] = j * p_trans[pre_state][y] * p_emit[y][obs[t]]\n",
    "print(sum([len(i) for i in V]))\n",
    "\n",
    "(prob, path) = max((j, i) for i,j in V[-1].items())\n",
    "print(f'观察状态连续为{obs}, 心情可能是{path}, 概率为{prob}')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "2、维特比算法"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "obs = ['working', 'shopping', 'travel']\n",
    "states = ['good', 'normal', 'bad']\n",
    "V = [{}]\n",
    "path = [{}]\n",
    "\n",
    "# 初始化\n",
    "for y in states:\n",
    "    V[0][y] = p_start[y] * p_emit[y][obs[0]]\n",
    "    path[0][y] = [y]\n",
    "\n",
    "for t in range(1, len(obs)):\n",
    "    V.append({})\n",
    "    path.append({})\n",
    "    for y in states:\n",
    "        em_p = p_emit[y][obs[t]]\n",
    "        (prob, state) = max((V[t-1][y0]*p_trans[y0][y]*em_p, y0) for y0 in states)\n",
    "        V[t][y] = prob\n",
    "        path[t][y] = path[t-1][state] + [y]\n",
    "(prob, last_state) = max(((j, i) for i, j in V[-1].items()))\n",
    "\n",
    "print(f'观察状态连续为{obs}, 心情可能是{path[-1][last_state]}, 概率为{prob}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "print(sum([len(i) for i in V]))b"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "3、HMM应用：分词（维特比算法）"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "import sys\n",
    "sys.path.append(r'..\\data')\n",
    "from hmm.prob_start import P as p_start\n",
    "from hmm.prob_trans import P as p_trans\n",
    "from hmm.prob_emit import P as p_emit"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "hidden": true,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[{'B': -6.8985414338214115, 'M': -3.14e+100, 'E': -3.14e+100, 'S': -8.598245859008907}] \n",
      " [{'B': ['B'], 'M': ['M'], 'E': ['E'], 'S': ['S']}]\n",
      "\n",
      "\"今天我来到北京清华大学\"的标注为：\n",
      "['B', 'E', 'S', 'S', 'S', 'B', 'E', 'B', 'M', 'M', 'E']\n",
      "分词结果为：\n",
      "['今天', '我', '来', '到', '北京', '清华大学']\n"
     ]
    }
   ],
   "source": [
    "obs = '今天我来到北京清华大学'\n",
    "states = list('BMES')\n",
    "\n",
    "V = [{}]\n",
    "path = [{}]\n",
    "\n",
    "# t时刻状态：t-1时刻可能的状态\n",
    "prevStates = {'M': list('MB'),\n",
    "              'B': list('ES'),\n",
    "              'S': list('SE'),\n",
    "              'E': list('BM')}\n",
    "\n",
    "\n",
    "# 初始化\n",
    "for y in states:\n",
    "    V[0][y] = p_start[y] + p_emit[y][obs[0]]\n",
    "    path[0][y] = [y]\n",
    "print(V, '\\n', path)    \n",
    "\n",
    "for t in range(1, len(obs)):\n",
    "    V.append({})\n",
    "    path.append({})\n",
    "    for y in states:\n",
    "        emit_p = p_emit[y][obs[t]]\n",
    "        (prob, state) = max((V[t-1][y0] + p_trans[y0][y] + emit_p, y0) for y0 in prevStates[y])\n",
    "        V[t][y] = prob\n",
    "        path[t][y] = path[t-1][state] + [y]\n",
    "(prob, last_state) = max((V[-1][y], y) for y in list('ES'))\n",
    "sentence_label = path[-1][last_state]\n",
    "\n",
    "words = []\n",
    "for i, char in enumerate(obs):\n",
    "    sign = sentence_label[i]\n",
    "    if sign == 'B':\n",
    "        begin = i\n",
    "    elif sign == 'E':\n",
    "        words.append(obs[begin: i+1])\n",
    "    elif sign == 'S':\n",
    "        words.append(obs[i])\n",
    "\n",
    "print(f'\\n\"{obs}\"的标注为：\\n{sentence_label}\\n分词结果为：\\n{words}')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "### jieba分词"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['今天', '我', '来到', '北京', '清华大学']\n",
      "['今天', '我', '来到', '北京', '清华', '华大', '大学', '清华大学']\n"
     ]
    }
   ],
   "source": [
    "import jieba\n",
    "print(jieba.lcut(obs))\n",
    "print(jieba.lcut_for_search(obs))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['广州', '泰迪', '智能', '科技', '有限公司', '欢迎您', '！']\n",
      "['广州泰迪智能科技有限公司', '欢迎您', '！']\n"
     ]
    }
   ],
   "source": [
    "sentence = '广州泰迪智能科技有限公司欢迎您！'\n",
    "print(jieba.lcut(sentence))\n",
    "jieba.add_word('广州泰迪智能科技有限公司')\n",
    "print(jieba.lcut(sentence))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "《鹿鼎记》分词"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "collapsed": true,
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['《鹿鼎记》',\n",
       " '作者：金庸',\n",
       " '正文',\n",
       " '第一回',\n",
       " '纵横钩党清流祸',\n",
       " '峭茜风期月旦评',\n",
       " '北风如刀，满地冰霜。江南近海滨的一条大路上，一队清兵手执刀枪，押着七辆囚车，冲风冒寒，向北而行。前面三辆囚车中分别监禁的是三个男子，都作书生打扮，一个是白发老者，两个是中年人。后面四辆囚车中坐的是女子，最后一辆囚车中是个少*妇，怀中抱着个女婴，女婴啼哭不休。她母亲温言相呵，女婴只是大哭。囚车旁一清兵恼了，伸腿在车上踢了一脚，喝道：“再哭，再哭，老子踢死你！”那女婴一惊，哭得更加响了。离开道路数十丈处有座大屋，屋檐下站着一个中年文士，一个十一二岁的小孩。那文士见到这等情景，不禁长叹一声，眼眶也红了，说道：“可怜，可怜！”农小孩问道：“爹爹，他们犯了什么罪？”那文士道：“又犯了什么罪？昨日和今朝已逮去了三十几人，都是我们浙江有名的读书人，个个都是无辜株连。”他说到“无辜株连”四子，声音压得甚低，生怕给押囚车的官兵听见了。那小孩道：“哪个小女孩还在吃奶，难道也犯了罪么？真没道理。”那文士道：“你懂得官兵没道理，真是好孩子。咳，人为刀俎，我为鱼肉，人为鼎锅，我为麋鹿！”那小孩道：“爹，你前几天教过我。人为刀俎，我为鱼肉，就是给人家斩割屠杀的意思。人家是切菜刀，是铁板，我们就是鱼和肉。“人为鼎锅，我为麋鹿”这两句话，意思也差不多么？”那文士道：“正是！”眼见官兵和囚车已经去远，拉着小孩的手道：“外面风大，我们回屋里去。”当下父子二人走进书房。那文士提笔醮上了墨，在纸上写了个“鹿”字，说道：“鹿这种野兽，虽是庞然大物，性子却极为平和，只吃青草和树叶，从来不伤害别的野兽。凶猛的野兽要伤它吃它，它只有逃跑倘若逃不了，那只有给人家吃力。”又写了“逐鹿”两字，说道：“因此古人常常拿鹿来比喻天下。世上百姓都温顺善良，只有给人欺压残害的份儿。《汉书》上说：“秦失其鹿，天下共逐之。”那就是说，秦朝失了天下，群雄并起，大家争夺，最后汉高祖打败了楚霸王，就得了这只又肥又大的鹿。那文士提笔醮上了墨，在纸上写了个“鹿”字，说道：“鹿这种野兽，虽是庞然大物，性子却极为平和，只吃青草和树叶，从来不伤害别的野兽。凶猛的野兽要伤它吃它，它只有逃跑倘若逃不了，那只有给人家吃力。”又写了“逐鹿”两字，说道：“因此古人常常拿鹿来比喻天下。世上百姓都温顺善良，只有给人欺压残害的份儿。《汉书》上说：“秦失其鹿，天下共逐之。”那就是说，秦朝失了天下，群雄并起，大家争夺，最后汉高祖打败了楚霸王，就得了这只又肥又大的鹿。”那小孩点头道：“我明白了。小说书上说“逐鹿中原”，就是大家争着要作皇帝的意思。”那文士甚是喜欢，点了点头，在纸上画了一只鼎的图形，道：“古人煮食，不用灶头锅子，用这样三只脚的鼎，下面烧柴，捉到了鹿，就在鼎里煮来吃。皇帝和大官都很残忍，心里不喜欢谁，就说他犯了罪，把他放在鼎里活活煮熟。《史记》中记载蔺相如对秦王说：“臣知欺大王之罪当诛也，臣请就鼎锅。”就是说：“我该死，将我在鼎里烧死了罢！”那小孩道：“小说书上又常说‘问鼎中原’，这跟‘逐鹿中原’好象意思差不多”。那文士道：“不错。夏禹王收九州之金，铸了九大鼎。当时的所谓“金”其实是铜。每一口鼎上铸了九州的名字和山川图形，后世为天下之主的，便保有九鼎。《左传》上说：“楚子观兵于周疆。定王使王孙满劳楚子。楚子只是楚国的诸侯，他问鼎的轻重大小，便是心存不轨，想取周王之位而代之。”那小孩道：“所以”问鼎“，”逐鹿“便是想做皇帝。”未知鹿死谁手，就是不知那一个做成了皇帝。”那文士道：“正是。到得后来，问鼎，逐鹿，这四个字，也可借用于别处，但原来的出典，是专指做皇帝而言。“说道这里，叹了口气，道：“咱们做百姓的，总是死路一条。未知鹿死谁手，只不过未知是谁来杀了这头鹿，这头鹿，却是死定了的。”他说着走到窗边，向窗外望去。只见天色沉沉地。似要下雪，叹道：“老天爷何其不仁，数百个无辜之人。在这冰霜遍地的道上行走。下起雪来，可又多受一番折磨了。”忽见南边大道上两个人戴着斗笠，并肩而来，走到近处，认出了面貌。那文士大喜，道：“是你黄伯伯，顾伯伯来了！”快步迎将出去，叫道：“梨洲兄，亭林兄，那一阵好风，吹得二位光临？”右首一人身形微胖，额下一部黑须，姓黄名宗羲，字梨洲，浙江余姚人士。左首一人又高又瘦，面目黝黑，姓顾名炎武，字亭林，江苏昆山人士。黄顾两人都是当世大儒，明亡之后，心伤国变，隐居不仕，这日连袂来到崇德。顾炎武走上几步，说道：“晚村兄，有一件要紧的事，特来和你商议。”这文士辛吕名留良，号晚村，世居浙江府崇德县，也是明末，清初一位极有名的隐士他眼见黄顾二人脸色凝重，又知顾炎武向来极富机变，临事镇定，即说是要紧事，自然非同小可，拱手道：“两位请进去先喝三杯，解解寒气。”当下请二人进屋，吩咐那小孩道：“葆中，去跟娘说，黄伯伯，顾伯伯到了，先切两盘羊膏来下酒。”不多时，那小孩女葆中和兄弟毅中搬出三副杯筷，布在书房桌上。一名老仆奉上酒菜。吕留良待三人退出，关上了书房门，说道：“黄兄，顾兄，先喝三杯！”黄宗羲神色惨淡，摇了摇头。顾炎武却自斟自饮，一口气连干了六七杯。吕留良道：“二位此来，可是和《明史》一案有关吗？”黄宗羲道：“正是。”顾炎武提起酒杯，高声呤道：“清风虽细难吹我，明月何尝不照人？晚村兄，你这两句诗，真是绝唱！我每逢饮酒，必诵此诗，必浮大白。”吕留良心怀故国，不肯在清朝做官。当地大吏仰慕他声名，保荐他为“山林隐士”，应徵赴朝为官，吕留良誓死相拒，大吏不敢在逼。后来又有一名大官保荐他为“博学鸿儒”，吕留良眼见若再相拒，显是轻辱朝廷，不免有杀身之祸，于是削发为僧，做了假和尚。地方官员见他意坚，从此不再劝他出山。“清风，明月”两句，意在讽刺清廷，怀念前明，虽然不敢刊行，但在志同道合的朋辈之间传诵已遍，此刻顾炎武又读了出来。黄宗羲道：“真是好诗！”举起酒杯，也喝了一杯。吕留良道：“两位谬赞了。”顾炎武一抬头，见到壁上挂着一幅高约五尺，宽约丈许的大画，绘的是一大片山水，笔势纵横，气象雄伟，不禁喝了声采，画上只题了四个大字：“如此江山”，说道：“看这笔路，当是二瞻先生的丹青了。”留良道：“正是。那‘二瞻’先生姓查，名士标，是明末清初的一位大画家，也和顾黄吕诸人交好。黄宗羲道：“这等好画，如何却无题跋？”吕留良叹道：“二瞻先生此画，颇有深意。只是他为人稳重谨慎，即不落款，亦无题跋。他上个月在舍间盘亘，一时兴到，画送了我，两位便题上几句如何？”顾黄二人站起身来，走到画前仔细观看，只见大江浩浩东流，两岸峰峦无数，点缀着奇树怪石，只是画中云气弥漫，山川虽美，却令人一见之下，胸臆间顿生郁积之气。顾炎武道：“如此江山，沦于夷狄。我辈忍气吞声。偷生其间，实令人悲愤填膺。晚村兄何不便提诗一首。将二瞻先生之意，表而出之？”吕留良道：“好！”当即取下画来，平铺于桌。黄宗羲研起了墨。吕留良提笔沉吟半晌，便在画上振笔直书。顷刻诗成，诗云：“其为宋之南渡耶？如此江山真可耻。其为崖山以后耶？如此江山不忍视。吾今始悟作画意，痛哭流涕有若是。以今视昔昔犹今，吞声不用枚衔嘴。画将皋羽西台泪，研入丹青提笔呲。所以有画无诗文，诗文尽在四字里。尝谓生逢洪武初，如瞽忽瞳跛可履。山川开霁故壁完，何处登临不狂喜？”书完，掷笔于地，不禁泪下。顾炎武道：“痛快淋漓，真是绝妙好辞。”吕留良道：“这诗殊无含蓄，算不得好，也只是将二瞻先生之原意写了出来，好教观画之人得知。”黄宗羲道：“何日故国重光，那时山川开霁故壁完，纵然穷山恶水，也令人观之大畅胸怀，真所谓何处登临不狂喜了！”顾炎武道：“此诗结得甚妙！终有一日驱除胡虏，还我大汉河山，比之徒抒悲愤，更加令人气壮。”黄宗羲慢慢将画卷了起来，说道：“这画是挂不得了，晚村兄得须妥为收藏才是。倘若给吴之荣之类的奸人见到，官府查究起来，晚村兄固然麻烦，还牵连了二瞻先生。”顾炎武拍桌骂道：“吴之荣这狗贼，我真恨不得生食其肉。”吕留良道：“二位枉顾说道有件要紧事。我辈书生积习，作诗题画，却搁下了正事。不知究竟如何？”黄宗羲道：我二人来止，乃是为了二瞻先生的那位本家伊璜先生小弟和顾兄前日得到讯息，原来这场‘明史’大案，竟将伊璜先生也牵连在内。”吕留良道：“伊璜兄也受了牵连？”黄宗羲道：“是啊。我二人前日晚上匆匆赶到海宁袁华镇，伊璜先生并不在家，说是出外访友去了。炎武兄眼见事势紧急，忙瞩伊璜先生家人连夜躲避，想起伊璜先生和晚村兄交好，特来探访。”吕留良道：“他……他却没有来。不知到了何处。”顾炎武道：“他如在府上，这会儿自己出来相见。我已在他的书房的墙壁上提诗一首，他若归家，自然明白，知所趋避，怕的是不知音讯，在外露面，给公人拿了，那可糟了。”黄宗羲道：“这‘明史’一案，令我浙江名士几乎尽遭毒手。清廷之意甚恶，晚村兄名头太大，亭林兄和小弟之意，要劝晚村兄离家远游，避一避风头。”吕留良气愤道：“清廷皇帝倘若将我捉到北京，拼着千刀万剐，好歹也要痛骂他一场，出了胸中这口恶气，才痛痛快快的就死。”顾炎武道：“恶臭兄豪气干云，令人好生敬佩。怕的是见不到清廷皇帝，却死于一般的下贱的奴才手里。再说，清廷皇帝只是个小孩子，什么也不懂，朝政大权，尽操纵于权臣鳌拜之手。兄弟和梨洲兄推想，这次‘名士’一案所以如此大张旗鼓，雷厉风行，当是鳌拜意欲挫折我江南士人之气。”吕留良道：“两位所见甚是。清兵入关以来，在江北横行无阻，一到江南，却处处遇到反抗，尤其读书人知道华夷之防，不断根他们捣乱。鳌拜乘此机会，对我江南士子大加镇压。哼，野火烧不尽，春风吹又生，除非他把咱们江南读书人杀得干干净净。”黄宗羲道：“是啊，因此咱们要留着有用之身，和清廷周旋到底，倘若逞了一时血气之勇，反是堕入他们的算中了。”吕留良登时省悟，黄顾二人冒寒枉顾，一来固是寻觅查伊璜，二来是劝自己一时按奈不住，枉自送了性命，良友苦心，实深感激，说道：“二位金石良言，兄弟那敢不尊？明日一早，兄弟全家便出去避一避。“顾黄二人大喜，齐声道：“自该如此。”吕留良沉呤道：“却不知避向何处才好？“只觉天涯茫茫，到处是敌人的天下，真无一片干净土地，沉呤道：“桃源何处，可避暴秦？桃源何处，可避暴秦？“顾炎武道：“当今之世，便真有桃源乐土，咱们也不能独善其身，去躲了起来……“吕留良不等他辞毕，拍案而起，大声道：“亭林兄此言责备的是。国家兴亡，匹夫有责，暂时避祸则可，但若去躲在桃花源里，逍遥自在，忍令亿万百姓在清兵铁蹄下受苦，于心何安？兄弟失言了。~”顾炎武微笑道：“兄弟近年浪迹江湖，着实结交了不少好朋友。大江南北，见闻所及，不但读书人反对清廷，而贩夫走卒，屠沽市井之中，也到处有热血满腔的豪杰。晚村兄要是有意，咱三人结伴同去扬州，兄弟给你引见几位同道中人如何？”吕留良大喜，道：“妙极~，妙极！咱们明日便去扬州，二位少坐，兄弟去告知拙荆，让她收拾收拾。”说着匆匆入内。不多时吕留良回到书房，说道：“‘明史’一案，外间虽传说纷纷，但一来传闻未必确实，二来说话之人顾忌甚多，不敢尽言。兄弟独处蜗居，未知其详，到底是何起因？”顾炎武叹了口气，道：“这部明史，咱们大家都是看过的了，其中对清廷不大恭敬，那也是有的。此书本是出于我大明朱国桢相国之手，说到关外建洲卫之事，又如何会对他们客气？”吕留良点头道：“听说湖洲庄家花了几千两银子，从朱相国后人手中将明史原稿买了来，以己名刊行，不想竟然酿此大祸。”浙西杭州，嘉兴，湖洲三府，处于太湖之滨，地势平坦，土质肥沃，盛产稻米蚕丝。湖洲府的首县今日称为吴兴县，清时分为乌程，归安二县。自来文风甚盛，历代才士辈出，梁时将汉字分为平上去入四深的沈约，元代书画皆至极品的赵孟业，都是湖洲人氏。当地又以产笔著名，湖洲之笔，徽洲之墨，宣城之纸，肇怯谒溪之砚，文房四宝，天下驰名。湖洲府有一南浔镇，虽是一个镇，却比寻常州县还大，镇上富负极多，著名的富室大族之中有一家姓庄。其实庄家的富户名叫庄允城，生有数子，长子名叫廷珑，自幼爱好诗书~，和江南名士才子多所结交。到得顺治年间，庄廷珑因读书过于勤，忽然眼盲，寻遍名医，无法治愈，自是郁郁不欢。忽有一日，邻里有一朱姓少年携来一部手稿，说是祖父朱相国的遗稿，向庄家抵押，求借数百两银子。庄家素来慷慨，对朱相国的后人一直照顾着，既来求借，当即允若，也不要他用什么遗稿抵押。但那朱姓少年说道借得银子之后，要出门远游，这部祖先的遗稿带在身边，恐有遗失，存在家里又不放心，要寄存在庄家。庄允城便达因了。那朱姓少年去后，庄允城为替儿子解闷，叫家中清客读给他听。朱国桢这部明史稿，大部份已经刊行，流传于世，这次他孙子携来向庄家抵押的，是最后的许多篇列传。庄廷珑听清客读了数日，很感兴味，忽然想起：“昔时左丘明也是盲眼之人，却因一部史书《左传》得享大名于千载之后。我今日眼盲，闲居无聊，何不也撰述一部史书出来，流传后世？”大富之家，办事容易，他即兴了此念，当即聘请了好几位士人，将那部明史稿从头至尾的他认为何处当增，何处当删，便口述出来，由宾客笔录。但想自己眼盲，无法博览群籍，这部明史修撰出来，如内容谬误甚多，不但大名难享，反而被人讥笑，于是又花了大批银两，延请许多通士文儒，再加修订，务求尽美。有些大有学问之人非钱财所能请到，便辗转托人，埤辞相邀。太湖之滨向来文士甚多，受到庄家邀请的，一来怜其眼盲，感其意诚，二来又觉得修撰明史乃是一件美事，大都到庄家来作客十天半月，对稿本或修正其误，或加润饰，或撰写一两篇文字。因此这部明史确是集了不少大手笔之力。书成不久，庄廷珑便去世。庄允城心伤爱子之逝，即行刊书。清代刊印一部书，着实不易，要招请工匠，雕成一块块木版，这才印刷成书。这部明史卷轶浩繁，雕工印工，费用甚巨。好在庄家有的是钱，拨出几件大屋作为工场，多请工匠，数年间便将书刊成了，书名叫作《明书辑略》，撰书人列名为庄廷珑，请名士李令皙作序。所有曾经襄助其事的学者也都列名其上，有茅元铭，吴之铭，吴之蓉，李祁涛，茅次莱，吴楚，唐元楼，严云起，蒋麟徽，韦金佑，韦一园，张契，董二西，吴炎，潘圣章等，共十八人。书中又提到此书是根据朱氏的原稿增删而成，不过朱国桢是明朝相国，名头太大，不便直书其名，因此含含糊糊的只说是“朱氏原稿”。“明书辑略”经过这许多文人学士撰改修订，是以体例精备，叙述详明，文字又华瞻雅致，书出后大获士林赞誉。庄家又是志在扬名，书价取得极廉。原稿中涉及满洲之时，本有不少攻柜指责的言语，修史诸人早已一一删去，但赞扬明朝的文字却也在所不免。当时明亡未久，读书人心怀故国，书一刊行，立刻就大大畅销。庄廷珑之名噪江北江南。庄允城虽有丧子之痛，但见儿子成名于身后，自是老怀弥慰。也是乱世之时，该当小人得志，君子遭祸。湖洲归安县的知县姓吴名之荣，在任贪赃枉法，百姓恨之切齿，终于为人告发，朝廷下令革职。吴之荣做了一任归安县知县，虽然搜刮了上万两银子，但革职的廷令一下，他东贿西赂，到处打点，才免得抄家查办的处分，这上万两赃款却也已荡然无存，连随身家人也走得不知去向。他官财两失，只得向各家富室一处处去打秋风，说道为官清苦，此番丢官，连回家也没有盘缠，无法成行。有些富人为免麻烦~，便送他十量八两银子。待得来到富室朱家，主人朱佑明却是个嫉恶如仇的正人君子，非但不送仪程，反而狠狠讥讽，说道搁下在湖洲做官，百姓给你害得好苦，我朱某就算有钱，也宁可去周济给搁下害苦了的贫民。吴之荣虽然恼怒，却也无法可施，他即已被革职，无权无势~，有怎能奈何得了富家巨室？当下又来拜访庄允城。庄允城平素结交清流名士，对这赃官很瞧不起，见他到来求索，冷笑一声，封了一两银子给他，说道：“依搁下的为人，这两银子本是不该送的，只是湖洲百姓盼望阁下早去一刻也好，多一两银子，能早去片刻也是好的。”吴之荣心下怒极，一瞥眼见到大厅桌上放得有一部《明书辑略》，心想：“这姓庄的爱听奉承，人家只要一赞这部明史修得如何如何好，白花花的银子双手捧给人家，再也不皱一皱眉头。”便笑道：“庄翁厚赐之，却不恭。兄弟今日离别湖洲，最遗憾的便是无法将‘湖洲之宝’带一部回家，好让敝乡孤陋寡闻之辈大开眼界。”庄允城问道：“什么叫着‘湖洲之宝’？”吴之荣笑道：“庄翁这可太谦了。士林之中，纷纷都说，令郎廷珑公子亲笔所撰的那部《明书辑略》，史才，史识，史笔，无一不是旷古罕有，左马班庄，乃是古今良史四大家。这‘湖洲之宝’，自然便是令郎亲笔所撰的明史了。~”吴之荣前一句“令郎所撰”，后一句“令郎亲笔所撰”把庄允城听的心花怒放。他明知此书并非儿子所作，内心不免遗憾，吴之荣如此说，正好大投所好，心想：“人家都说此人贪赃，是个龌龊小人，但他毕竟是个读书人，眼光到是有的。原来外间说珑儿此书是‘湖洲之宝’，这话倒是第一次听见。”不由得笑容满面，说道：“荣翁说什么左马班庄，古今四大良史，兄弟可不大明白，还请指教。”吴之荣见他脸色顿和，知道马屁已经拍上，心下暗暗欢喜，说道：“庄翁未免太谦了。左丘明作《左传》，司马迁作《史记》，班固作《汉书》，都是传诵千载的名作。自班固而后，大史家就没有了。欧阳修作《五代史》，司马光作《资治通鉴》，文章虽佳，才识终究差了。直到我大清盛世，令郎亲笔所撰这部煌煌巨作《明史辑略》出来，方始有人能和左丘明，司马迁，班固三位前辈齐驱，‘四大良史，左马班庄’，这句话便是由此而生。”庄允城笑容满面，连连拱手，说道：“谬赞，谬赞！不过‘湖洲之宝’这句话，毕竟当不起。”吴之荣正色道：“怎么当不起？外间大家都说：‘湖洲之宝史丝笔，还是庄史居第一！”蚕丝和毛笔是湖洲两大名产，吴之荣品格卑下，却有三分才情，出口成章，将“庄史”和湖洲丝，湖笔并称。庄允城听得更是喜欢。吴之荣又道：“兄弟来到贵处做官，两袖清风，一无所得。今日老着脸皮，要向庄翁求一部明史，作为我家传家之宝。日后我吴家子孙日夕诵读，自必才思大进，光宗耀祖，全仗庄文之赐了。”庄允城笑道：“自当奉赠。”吴之荣又谈了几句，不见庄允城有何举动，当下又将这部明史大大恭维了一阵，其实这部书他一页也未读过，只是史才如何如何了得，史识又如何如何超卓，不着边际的瞎说。庄允城道：“荣翁且请宽坐。”回进内堂。过了良久，一名家丁捧了一个包裹出来，放在桌上。吴之荣见庄允城尚未出来，幔将包裹掂了掂，那包裹虽大，却是清飘飘地，内中显然并无银两，心下好生失望。过得片刻，庄允城回到厅上，捧起包裹，笑道：“荣翁瞧得起敝处的土产，谨以相赠。”吴之荣谢了，告辞出来，没回到客店，便伸手到包裹中一阵掏摸，摸到的竟是一部书，一束蚕丝，几十管毛笔。他费了许多唇舌，本想庄允城在一部明史之外，另有几百两银子相赠，可是赠送的是他信口胡诌的‘湖洲三宝’心下暗骂：“***，南浔这些财主，都如此小气！也是我说错了话，倘若我说‘湖州三宝’乃是金子和银子和明史，岂不是大有所获？”气愤愤的回到客店，将包裹往桌上一丢，倒头便睡，一觉醒来，天已大黑，客店中吃饭的时候已过，他又舍不得令叫饭菜，愁肠饥火，两相煎熬，再也睡不着觉，当下解开包裹，翻开那部《明史》阅看。看得几页，眼前金光一闪，赫然出现一张金叶。吴之荣一颗心怦怦乱跳，揉了揉眼细看，却不是金叶是什么？当下一阵乱抖，从书中抖了十张金叶出来，每一张少说也有五钱，十张金叶便有五两黄金，五两黄金抵得四百两银子。吴之荣喜不自胜，寻思：“这姓庄的果然狡猾，他怕我讨得这部书去，随手抛弃，翻也不翻，因此将金叶子夹在书中，看是谁读他儿子的这部书，谁便有福气得此金叶。是了，我便多读几篇，明天再上门去，一面谢他赠金之惠，一面将书中文章背诵几段，大赞而特赞。他心中一喜，说不定另有几两黄金相送。”当下剔亮油灯，翻书诵读，读到明万历四十四年，后金太祖努儿哈赤即位，国号金，建元“天命”突然间心中一凛：“我太祖于丙辰建元，从这年起，就不该用明朝万历年号，该用大金天命元年才是。”一路翻阅下去，只见丁卯年后金太宗即位，书中仍用“明天启七年”，不作“大金天聪元年”。丙子年后金改国号为清，改元崇德，这部书仍作“崇祯九年”，不书“大清崇德元年”，甲申年书作“崇祯十七年”不书“清顺治元年”。又看入关之后，书中于乙西年书作“隆武元年”，丁亥年书作“永历永历”，那隆武，永历，乃明朝唐王，桂王的年号，作书之人明明白白是仍奉明朝正朔，不将清朝放在眼里。他看到这里，不由得拍案大叫：“反了，反了，这还了得！”一拍之下，桌子震动，油灯登时跌翻，溅得他手上襟上都是灯油。黑暗之中，突然间灵机一动，不由得大喜若狂：“这不是老天爷赐给我的一注横才？生官发财，皆由于此。”想到开心处，不由得大声叫唤起来。忽然听得店伴拍门叫道：“客官，客官，什么事？”吴之荣笑道：“没什么！”点燃油灯，重新翻阅。这一晚直看到雄鸡啼鸣，这才和衣上床，却又在书中找了七八十出忌讳犯禁的文字出来，便在睡梦中，也是不住的嬉笑。换朝改代之际，当政者于这年号正朔，最是着意。最犯忌这，莫过于文字言语之中，引人思念前朝。《明书辑略》记叙的是明代之事，以明代年号纪年，原无不合，担当文字禁网极密之际，却是极大的祸端。参与修史的学者文士，大都只助修数卷，未能通阅全书，而修撰最后数卷之人，偏是对前朝痛恨入骨，决不肯在书中用大清年号。庄廷珑是富室公子，双眼有盲，未免粗疏，终予小人可乘之隙。次日中午，吴之荣便即乘船东行，到了杭州，在客店中写了一张禀帖，连同这部明史，送入将军松魁府中。他料想松魁收到禀帖后，便会召见。其时满清于检举叛逆，赏赐极厚，自己立此大功，开复原官顾是意料之事，说不定还会连升三级。不料在客店中左等右等，一连等上大半年，日日道将军府去打探消息，却如石沉大海一般，后来那门房竟厉声斥责，不许他再上门罗唣。吴之荣心焦已极，庄允城所赠金叶兑换的银子即将用尽，这场告发却没半点结果，又是烦恼，又是诧异。这日在杭州城中闲逛，走过文通堂书局门口，踱进去想看看白书，以消永日，只见书架上陈列着三部《明书辑略》，心想：“难道我所找出的岔子，还不足以告倒庄允城？且再找几处大逆不道的文字出来，明日再写一张禀帖，递进将军府去。~”浙江巡抚是汉人，将军则是满洲人，他生怕巡抚不肯行此文字大狱，是以定要向满洲将军告发。他打开书来，只看得几页，不由得吓了一跳，全身犹如堕入冰窟，一时宛如涨二和尚，摸不着头脑，只见书中犯忌的文字竟已全然无踪，自大清太祖开国以后，也都改用了大金大清的年号纪年，至于功旰建州卫都督，以及大书隆武，永历等年号的文字，更是一字不见。但文字前后贯串，书页上干干净净，更无丝毫涂改痕迹，这戏法如何变来，实是奇哉怪也~。他双手捧书，在书铺中呆呆出神，过得半响，大叫一声：“是了！”眼见此书书页封函，洁白崭新，向店倌一问之下，果然是湖洲贩书客人新近送来，送货还不过七八天。他心道：这庄允城好厉害！”当真是钱可通神收回旧书，重新镌版，另刊新书，将原书中所有干犯禁忌之处，尽行删削干净。哼，难道就此罢了不成？”吴之荣所料果然不错。原来杭州将军松魁不识汉字，幕府师爷见到吴之荣的禀帖，登时全身吓出了一身冷汗，知道此事牵连重大之极，拿着禀帖的双手竟不由自主的颤抖不已。这幕客姓程，名维藩，浙江绍兴人氏。明清两朝，官府的幕僚十之八九是绍兴人，所以“师爷”二字之上，往往冠以“绍兴”，称为“绍兴”师爷“。这些师爷先跟同乡先辈学到一套秘诀，此后办理刑名钱谷，处事便十分老到。官府中所有公文，钧由师爷手拟”大家既是同乡，下级官员的公文呈到上级衙门去，也就不易遇到挑剔批驳。所以大小新官上任，最要紧的便是重金聘请一位绍兴师爷。明清两朝，绍兴人做大官的人并不多，却操纵了中国庶政大数百年之久，也是中国政治史上的一项奇迹。那程维藩宅心忠厚，信奉“公门之中好修行”这句名言。那是说官府手操百姓生杀大权，师爷拟稿之中略重，便能令百姓家破人亡，稍加开脱，便可使之死里逃生，因之在公门中救人，比之在寺庙中修行效力更大。他见明史一案倘若酿成大狱，苏南浙西不知将有多少人丧生破家，当即向将军告几天假，星夜坐船，来到湖洲南浔镇上，将此事告诉庄允城。庄允城陡然大祸临头，自是魂飞天外，登时吓得全身瘫软，口诞直流，不知如何是好，过了良久，这才站起身来，双膝跪地，向程维藩叩谢大恩，然后现他问计。程维藩从杭州坐船到南浔之时，反覆推考，已思得良策，心想这部《明书辑略》流传已久，隐瞒是瞒不了的，唯有施一个釜底抽薪之计，一面派人前赴各地书铺，将这部书尽数收购回来销毁，一面赶开夜工，另镌新版，删除所有讳忌之处，重印新书，行销于外。官府追究之时，将新版明史拿来一查，发觉吴之荣所告不实，便可消一场横祸了。成维藩又教了他不少关节，某某官府处应送礼若干，某某衙门处应如何疏通，庄允城一一受教。程维藩回到杭州，隔了半个多月，才将原书及吴之荣的禀帖移送浙江巡抚朱昌柞，轻描淡写的批了几个字，说道投禀者是因赃已革知县，似有挟怨吹求之嫌，请府台大人详查。吴之荣在杭州客店中苦候消息之时，庄允城的银子却如流水价将出去。其时庄允城的重贿，已经送到将军衙门，巡抚衙门和学政衙门。朱昌柞接到公事，这等刊书之事，属学政该管，压了十多天后，才移牒学政胡尚衡。学政衙门的师爷先搁上大半个月，又告了一个月的病假，这才慢吞吞的拟稿发文，将公事送到湖洲府去。湖洲府学官又耽搁了二十几天，才移文归安县和乌程县的学官，要他二人申覆。那两个学官也早得到庄允城的大笔贿赂，其时新版明史也已印就，二人将二部新版书缴了上去，回说道：“该书平庸粗疏，无裨世道人心，然细查全书，尚无讳禁犯例之处。”层层申覆，就此不了了之。吴之荣直到在书铺中发现了新版明史，方知就里，心想唯有弄到一部原版明史，才能重揭此案。杭州各家书铺之中，原版书早给庄家买清，当下前赴浙东偏僻洲县收购，岂知仍是一部也觅不到。他穷乡潦倒，只好废然还乡。也是事有凑巧，旅途之中，却在一家客店中见到店主人正在摇头晃脑的读书，一看之下，所读的便是这部《明书辑略》，借来一翻，竟是原版。这一下大喜过望，心想若向店主人求购，一来他未必肯售，二来自己也无银子，买不起，只好偷。深夜之中悄悄起床，偷了书便即溜出店门，心想浙江全省有关官员都已受了庄允城之贿，一不作，二不休，索性告倒北京城去。吴之荣来到北京，便写了禀帖，告倒礼部，都察院，通政司三处衙门，说明庄家如何贿赂官员，改镌新版。不料在京中等不到一个月，三处衙门先后驳覆下来，都称细查庄廷珑所著《明书辑略》一书，无违禁犯例，该革职知县吴之荣所告，并非实情，显系挟嫌诬告，至于贿赂官员云云，更系扑风捉影之通政司的批驳更是严厉，说道：“该吴之荣以贪墨被革，遂以天下清官，皆如彼之贪。”原来庄允城受了教，早将新版明史送到了礼部，都察院，通政司三处衙门，有关官吏师爷，也早送了厚礼打点。吴之荣又碰了一鼻子灰，眼见回家已无盘缠，势将流落异乡。其时清廷对待汉人文士极为严峻，稍有犯禁，便即处死，吴之荣所告的若是寻常文人，早已得手，偏生遇着的对手是富豪之家，这才阻难即无退路，心想拼着坐牢，也要将这件案子干到底，当下又写了四张禀帖，分呈四位顾命大臣，同时又中写了数百张招纸，揭露其事，在北京城中到处张贴。他这一着却大是行险，倘若官府追究起来，说他危言耸听，扰乱人心，不免有杀头的重罪。那四个顾命大臣，名叫索尼，苏克萨哈，遏必隆，鳌拜，均是满洲的开国功臣。顺治皇帝逝世之时，遗诏命这四大臣铺政。其中鳌拜最为凶横，朝中党羽极众，清廷大权，几乎尽操于他一人之手。他生怕敌党他不利，是以派出无数探子，在京城内外打探动静。这日得到密报，说道北京城中出现许多招贴，揭发浙江姓庄百姓著书谋叛，大逆不道，浙江官员受贿，置之不理等情。鳌拜得悉之下，立即查究，登时雷厉风行的办了起来。便在此时，吴之荣的禀帖也已递入鳌拜府中。他当即召见吴之荣，详问其事，再命手下汉人细阅吴之荣所呈缴的那部原版明史，所言果是实情。鳌拜以军功而封公爵，做大官，向来歧视汉人和读书人，掌握大权后便想办几件大案，镇慑人心，不但使汉人不敢兴反叛之念，也令朝中敌党不敢有甚矣诏，当即派出钦差，赴浙江查究。这一来，庄家全家固然逮入京中，连杭州将军松魁，浙江巡抚朱昌柞以下所有大小官员，也都革职查办。在明史上列名的文学之士，无一不锒铛入狱。顾炎武，黄宗羲二人在吕流落家中，将此案的来龙去脉，详细道来，吕来龙听得只是叹惜。当晚三人联榻长谈。议论世事，说道明末魏忠贤等太监陷害忠良，把持朝政，种种倒行逆施众至明室覆亡，入清后汉人惨遭屠戮，祸难方深，无不扼腕切齿。次日一早，吕来龙全家和顾黄二人登舟东行。江南中常以上人家，家中都自备有船，江南水乡，河道四通八达，密如蛛网，一般人出行都是坐船，所谓“北人乘马，南人乘舟”，自古已然。到得杭州以后，自运河折而向北，这晚在杭州听到消息，清廷已因此案处决了不少百姓官员，庄廷珑已死，开棺戳尸，庄允城在狱中不堪虐待而死，庄家全家数十口，十五岁以上的尽数处斩，妻女发配沈阳，给满洲骑兵为奴。前礼部侍郎李令皙为该书作序，凌迟处死，四子处斩。李令皙的幼子刚十六岁，法司见杀得人多，心肠软了，命他减供一岁，按照清律，十五岁以下者得免死充军。那少年道：“我爹爹哥哥都死了，我也不愿独生。”终于不肯易供，一并处斩。松魁，朱昌柞入狱候审，幕客程维藩凌迟弃市。归安，乌程的两名学官处斩。因此案牵连，冤枉而死的人亦死不计其数。湖洲知府谭希闵到任还只半月，朝廷说他知情不报，受贿隐匿，和推官李焕，训导王兆祯同处绞刑。吴之荣对南浔富人朱佑明心下怀恨最深，那日去打秋风，给他抢白了一场，逐出门来，当下向办理此案的法司声称，该书注明依据“朱氏原稿增删润饰而成”，这朱氏便是朱佑明了，又说他的名字”朱佑明“，显是心存前明，诅咒本朝。这样一来，朱佑明和他的五个儿子同处斩首，朱家的十余万财产，清廷下令都赏给吴之荣。最惭的是，所有雕版的刻工，印书的印工，装订的钉工，以及书贾，书铺的主人，卖书的店员，买书的读者，查明后尽皆处斩。据史书记载，其时苏州浒墅关有一个榷货主事李尚白，喜读史书，听说苏州阊门书坊中有一部新刊的明史，内容很好，派一个工役去买。工役到时，书店主人外出，那工役便在书铺隔壁一家姓朱的老者家中坐着等候，等到店主回来，将书买回。李尚白读了几卷，也不以为意。过了几个月，案子发作，一直查究到各处贩书买书之人。其时李尚白在北京公干，以购逆书之罪，在北京立即斩决。书店主人和奉命买书的工役斩首。连那隔壁姓朱老者也受牵连，说他即知那人来购逆书，何以不即举报，还让他在家中闲坐？本因斩首，姑念年逾七十，免死，和妻子充军边远之处。至于江南名士，因庄廷珑慕其大名，在书中列名参校者，同日凌迟处死，计有茅元锡等十四人。所谓凌迟处死，乃是一刀一刀，将其全身肢体肌肉慢慢切割下来，直到犯人受尽痛苦，方才处死。因这一部书而家破人亡的，当真难以计数。吕留良等三人得到消息，愤恨难当，切齿痛骂。黄宗羲道：“伊璜先生列名参校，这一会也怕难逃此劫。”他三人和查伊璜向来交好，都十分挂念。这一日舟至嘉兴，顾炎武在城中买了一份邸报，上面详列明史一案中获罪诸人的姓名。却见上谕中有一句说：“查继佐，范骧，陆坼三人，虽列名参校，然事先未见其书，免罪不究。”顾炎武将邸报拿到舟中，和黄宗羲，吕留良三人同阅，啧啧称奇。黄宗羲道：“此事必是大力将军所为。“吕留良道：“大力将军是谁？到要请教。“黄宗羲道：“两年之前，兄弟到伊璜先生家中作客，但见他府第焕然一新，庭院宽大，陈设富丽，与先前大不相同。府中更养了一班昆曲戏班子，声色曲艺，江南少见。兄弟和伊璜先生向来交好，说得上互托肝胆，便问起情由。伊璜先生说出一段话来，确是风尘中的奇遇，“当下便将这段故事转述了出来。查继佐，字伊璜。这一天家居岁暮，命酒独酌，不久下起雪来，约下越大。查伊璜独饮无聊，走到门外观赏雪景，见有个乞丐站在屋檐下避雪，这丐者身形魁梧，骨格雄奇，只穿一件破单衫，在寒风中却丝毫不以为意，只是脸上颇有郁怒悲愤之色。查伊璜心下奇怪，便道：“这雪非是一时能止。进来喝一杯如何？“那乞丐道：“甚好查伊璜便邀请他进屋，命书僮取出杯筷，斟了杯酒，说道：“请！”那乞丐举杯便干，赞得：“好酒！”查伊璜给他连斟了三杯，那丐者饮得极是爽快。查伊璜最喜的是爽快人，心下喜欢，说道：“兄台酒量极好，不知能饮多少？”那丐者道：“酒逢知己千杯少，话不投机半句多？”这两句虽是熟套语，但在一个乞丐口中说出来，却令查伊璜暗暗称奇，当即命书僮捧出一大坛绍兴女儿红来，笑道：“在下酒量有限，适才又已饮过，不能陪兄畅饮。老兄喝一大碗，我陪一小杯如何？”那丐者道：“这也使得。”当下书僮将酒烫热，分斟在碗中杯内。查伊璜喝一杯，那乞丐便喝一大碗。待那乞丐喝到二十余碗时，脸上日无酒意，查伊璜却已颓然醉倒。要知那绍兴女儿红酒入口温和，酒性却颇厉害。绍兴人家生下儿子女儿，便酿数坛至数十坛不等，埋入地下，待女儿长大嫁人，将酒取出宴客，那酒其时作琥珀色，称为女儿红。想那酒埋藏十七八年以至二十余年，自然醇厚之极。至于生儿子人家所藏之酒，称为“状元红”，盼望儿子日后中状元时取出宴客。状元非人人可中，多半是在儿子娶媳妇时用以飨客了。酒坊中酿酒用以贩卖的，也袭用了状元红，女儿红之名。书僮将查伊璜扶入内堂安睡，那乞丐自行又到屋檐之下。次晨查伊璜醒转，忙去瞧那乞丐时，只见他负手而立，正在欣赏雪景。一阵北风吹来，查伊璜只觉寒入骨髓，那乞丐却是泰然自若。查伊璜道：“天寒地冻，兄台衣衫未免过于单薄，”当即解下身上的羊疲袍子，披在他肩头，又取了十两银子，双手捧上，说道：“这些买酒之资，兄台勿却。何时有兴，请再来喝酒。昨晚兄弟醉倒，未能扫塌留宾，简慢勿怪。”那乞丐接过了银子，说道：“好说。”也不道谢，扬长而去。第二年春天，查伊璜到杭州游玩，一日在一座破庙之中，见到有口极大的古钟，少说也有四百来斤，他正在鉴赏钟上所刻的文字花纹，忽有一名乞丐大踏步走进佛殿，左手抓住钟钮，向上一提，一口大钟竟然离地数尺。那乞丐在钟下取出一大完肉，一大钵酒来，放在一旁，再将古钟置于原处。查伊璜见他如此神力，不禁赫然，仔细看时，竟然便是去冬一起喝酒的那乞丐，笑问：“兄台还认得我吗？”那乞丐向他望了一眼，笑道：“啊，原来是你。今日我来作东，大家再喝个痛快，来来来，喝酒。”说着将土钵递了过去。查伊璜接过土钵，喝了一大口，笑道：“这酒挺不错啊。”那乞丐从破碗中抓起一大块肉，道：“这是狗肉，吃不吃？”查伊璜虽觉肮脏，但想：“我即当他是酒友，倘若推辞，未免瞧他不起了。”道谢伸手接过，咬了一口，咀嚼之下，倒也甘美可口。两人便在破庙中席地而坐，将土钵递来递去，你喝一口，我喝一口，吃肉时便伸手到碗中去抓，不多时酒肉俱尽。那乞丐哈哈大笑，说道：“只可惜酒少了，醉不到孝廉公。”查伊璜道：“去年冬天在敝处邂逅，今日又再无意中相遇，实是有缘。兄台神力惊人，原来是一位海内男子，得能结交你这位朋友，小弟好生喜欢，兄台有兴，咱们到酒楼去再饮如何？”那乞丐道：“甚妙！甚妙！”两人到西湖边的楼外楼，呼酒又饮，不久查伊璜又即醉倒。待得酒醒，那乞丐已不知去向。那是明朝崇祯末年之事，过得数年，清兵入关，明朝覆亡。查伊璜绝意进取，只在家中闲居，一日忽有一名军官，领兵四名，来到查府。查伊璜吃了一惊，只道是祸事上门，岂知那军官执礼甚恭，说道：“奉广东吴军门之命，有薄礼奉赠。”查伊璜道：“我和贵上素不相识，只怕是弄错了。”那军官取出拜盒，拿出一张大红泥金名帖，上写“拜上查先生伊璜，讳继佐”，下面写的是“眷晚生吴六奇顿首百拜”。查伊璜心想：“我连吴六奇的名字也没听见过，为何送礼于我？”当下沉呤不语。那军官道：“敝上说道，这些薄礼，请查先生不要见笑。”说着将两只朱漆烫金的圆盒放在桌上，俯身请安，便即别去。查圆伊璜打开礼盒，赫然是五十两黄金，另一盒却是六瓶洋酒，酒瓶上缀以明珠翡翠，华贵非凡。查伊璜一惊更甚，追出去要那军官收回礼品，武人步快，早已去得远了。查伊璜心下纳闷，寻思：“飞来横财，非祸是福，莫非有人陷害于我？”当下将两只礼盒用封条封起，藏于密室。查氏家境小康，黄金倒也不必动用，只是久闻洋酒之名，不敢开瓶品尝，未免心痒。过了数月，亦无他异。这一日，却有一名身穿华贵的贵介公子到来。那公子不过十七八岁，精神饱满，气宇轩昂，带着八名从人，一见查一盒，便即跪下磕头，口称：“查世伯，侄子吴宝宇拜见。”查伊璜忙即扶起，道：“世伯之称，可不敢当，不知尊大人是谁？”那吴宝宇道：“家严名讳，上六下奇，现居广东通省水陆提督之职，特命小侄造府，恭请世伯到广东盘亘数月。”查伊璜道：“前承令尊大人厚赐，心下好生不安，说来惭愧，兄弟生性蔬阔，记不起何时和令尊大人相识，兄弟一介书生，素来不结交贵官。公子请少坐。”说着走进内室，将那两只礼盒捧了出来，道：“还请公子携回，实在不敢受此厚礼。”他心想恶吴六奇在广东做提督，必是慕己之名，欲以重金聘去做幕客。这人官居高位，为满洲人做鹰犬，欺压汉人，倘若受了他金银，污了自己的清白，当下脸色之间颇为不豫。吴宝宇道：“家严吩咐，务必请到世伯。世伯若是忘了家严，有一件信物在此，世伯请看。”在从人手中接过一个包裹，打了开来，却是一件十分敝旧的羊皮袍子。查伊璜见到袍子，记得是昔年赠给雪中奇丐的，这才恍然，原来这吴六奇将军，便是当年共醉的酒友，心中一动：“清兵占我天下，若有手握兵符之人先建义旗，四方嫌诏，说不定便能将清兵逐出关外。这奇丐居然还记得我昔日一饭一袍之惠，不是没有良心之人，我若动以大义，未始没有指望。男儿建功报国，正在此时，至不济他将我杀了，却又如何？”当下欣然就道，来到广州。吴六奇将军接入府中，神态极是恭谨，说道：“六奇流落江南，得蒙查先生不弃，当我是个朋友。请我喝酒，送我皮袍，倒是小事，在那破庙中肯和我同钵喝酒，手抓狗肉，那才是真正瞧得起我了。六奇其时穷途潦倒，到处遭人冷眼，查先生如此热肠相待，登时令六奇大为振奋。得有今日，都是出于查先生之赐。”查一盒淡淡的道：“在晚生看来，今日的吴将军，也不见得就比当年的雪中奇丐高明了。”吴六奇一怔，也不再问，只道：“是，是！”当晚大开筵席，遍邀广州城中的文武官员与宴，推查伊璜坐了首席，自己在下首相陪。广东省自巡抚以下的文武百官，见提督大人对查伊璜如此恭敬，无不暗暗称异。那巡抚还道查伊璜是皇帝派出来微服查访的钦差大臣，否则吴六奇平素对人十分倨傲，何以对这个江南书生却这等必恭必敬？酒散之后，那巡抚悄悄向吴六奇探问，这位贵客是否朝中红员。吴六奇微微一笑，说道：“老兄当真聪明，鉴貌辨色，十有九中。“这句话本来意存讥讽，说他这第十次却猜错了。岂知那巡抚竟会错了意，只道查伊璜真是钦差，心想这位查大人在吴提督府中居住，已给他巴结上了，吴提督向来和自己不甚投机，倘若钦差大人回京之后。奏本中对我不利，那可糟糕，回去后备了一份重礼，次日清晨，便送到提督府来。吴六奇出来见客，说道查先生昨晚大醉未醒，府台的礼物一定代为交到，一切放心，不必多所挂怀。巡抚一听大喜，连连称谢而去。消息传出，众官员都知巡抚大人送了份厚礼给查先生。这位查先生是何来头，不得而知，但连巡抚都送厚礼，自己岂可不送？数日之间，提督府中礼物有如山积。吴六奇命帐房一一照收，却不令查先生得知。他每日除了赴军府办理公事外，总是陪着查伊璜喝酒。这一日傍晚时分，两人又在华亭凉台中对坐饮酒。酒过数巡，查伊璜道：“在府上叨扰多日，已感盛情，晚生明日便要北归了。“吴六奇道：“先生说那里话来？先生南来不易，若不住上一年半载，决计不放先生回去。明日陪先生到五层楼去玩玩。广东风景名胜甚众，几个月内，游览不尽。”查伊璜乘着酒意，大胆说道：“山河虽好，已沦夷狄之手，观之徒增伤心。”吴六奇脸色微变，道：“先生醉理，早些休息罢。”查伊璜道：“初遇之时，我敬你是个风尘豪杰，足堪为友，岂知竟是失眼了。”吴六奇问道：“如何失眼？”查伊璜朗声道：“你具大好身手~，不为国民出力，却助纣为虐，作朝廷的鹰犬，欺压我大汉的百姓，此刻兀自洋洋得意，不以为耻。查某未免羞以为友。“说着霍地站起身来。吴六奇道：“先生噤声，这等话给人听见了，可是一场大祸。“查伊璜道：“我今日还当你是朋友，有一番良言相劝。你如不听，不妨便将我杀了。查某手缚鸡之力，反正难以相抗。“吴六奇道：“在下洗耳恭听。“查伊璜道：“将军手绾广东全省兵符，正事起义反正的良机。登高一呼，天下响应，纵然大事不成，也教清廷破胆，轰轰烈烈的干它一场，才不负你天生神勇，大好头颅。”吴六奇斟酒于碗，一口干了，说道：“先生说得好痛快！”双手一伸，嗤的一声响，撕破了自己袍子衣襟，露出黑髦髦的胸膛，拨开胸毛，却见肌肤上刺着八个小字：“天地父母，反清复明。”查伊璜又惊又喜，问道：“这……这是什么？”吴六奇掩好衣襟，说得：“适才听得先生一番宏论，可敬可佩。先生不顾殒身灭族的大祸，披肝沥胆，向在下指点，在下何干再行隐瞒。在下本在丐帮，此刻是天地会的洪顺堂红旗香主，誓以满腔热血，反清复明。”查伊璜见了吴六奇的胸口刺字，更无怀疑，说得：“来将军身在曹营心在汉，适才言语冒犯，多有得罪。”六奇大喜，心想这“身在曹营心在汉”，那是将自己比作关云长了，道：“这等比喻，可不敢当。”查伊璜道：“不知何谓丐帮，何谓天地会，倒要请教。”吴六奇道：“生请再喝一杯，待在下慢慢说来。”当下二人各饮了一杯。吴六奇道：“由来已久，自宋朝以来，便是江湖上的一个大帮。帮中兄弟均是以行乞为生，就算是家财豪富之人，入了丐帮，也须散尽家资，过叫化子的生活。帮中帮主以下是四大长老，其下是前后左右中五方护法。在左护法，在帮中算是八袋弟子，位份已颇不低。后来因和一位姓孙的长老不和，打起架来，在下其时酒醉，失手将重伤。不敬尊长已是大犯帮规，殴伤长老更是大罪，帮主和四长老集议之后，将在下斥革出帮。那日在府上相遇，先生请我饮酒，其时在下初遭斥逐，心中好生郁闷，承蒙先生不弃，胸怀登时舒畅了不少。”查伊璜道：“原来如此。”吴六奇道：“第二年春，在西湖边上再度相逢，先生折节下交，誉我是海内奇男子。在下苦思数日，心想我不容于丐帮，江湖上朋友都瞧我不起，每日里烂醉如泥，自暴自弃，眼见数年之间，就会醉死。这位查先生却说我是位奇男子，难道就此一蹶不振，再无出头之日？过不多时，清兵南下，我心下愤怒，不明是非，竟去投效清军，立了不少军功，残杀同胞，思之好生惭愧。”。查伊璜正色道：“这就不对了。兄台不容于丐帮，独来独往也好，自树门户也好，何苦出此下策，前去投效清军？“吴六奇道：“在下愚鲁，当时未得先生教诲，干了不少错事，当真该死之极。“查伊璜点头道：“将军既然知错，将功赎罪，也还不迟。”吴六奇道：“后来清兵席卷南北，我也官封提督。两年之前，半夜里忽然有人闯入我卧室行刺。这刺客武功不是我的对手，给我拿住了，点灯一看，竟然便是昔年给我打伤的那位丐帮孙长老。他破口大骂，说我卑鄙无耻，甘为异族鹰犬。他越骂越凶，每一句话都打中了我心坎。这些话有时我也想到了，明知自己的所作所为很是不对，深夜扪心自问，好生惭愧，只是自己所想，远不如他所骂得那么痛快明白。我叹了口气，解开他被我封住的穴道，说道：‘孙长老，你骂得很对，你这就去罢！他颇为诧异，便即越窗而去。”查伊璜道：“这件事做对了！”吴六奇道：“其时提督衙门的牢狱之中，关得有不少反清的好汉子。第二天的清早，我寻些藉口，一个个将他们放了，有的说是捉错了人，有的说不是主犯，从轻发落。过了一个多月，那位孙长老半夜又来见我，开门见山的问我，是否已有了悔悟之心，原意反清立功。我拔出刀来，一刀斩去左手两根手指，说：“吴六奇决心痛改前非，今后听从孙长老号令。伸出左手，果然无名指和小指已然不见，只剩三根手指。查伊璜大拇指一竖，赞道：“好汉子！”吴六奇继续说道：“孙长老见我意诚，又知我虽然生性鲁莽，说过的话倒是从未失言，便道：“很好，待我回覆帮主，请帮主的示下。“十天之后，孙长老又来见我，说帮主和四长老会商，决定收我回帮，重新由一袋弟子做起。又说丐帮已和天地会结盟，同心协力，反清复明。那天地会是台湾国姓爷郑大帅手下谋主陈永华陈先生所创，近年来在福建，浙江。广东一带，好生兴旺。孙长老替我引见会中广东洪顺堂香主，投入天地会。天地会查了我一年，交我办了几件要事，见我确是忠心不贰，最近陈先生从台湾传讯来，封我为洪顺堂香主之职。”查伊璜索然不明白天地会的来历，但台湾国姓爷延平郡王郑成功孤军抗清，精忠英勇，天下无不知闻。这天地会既是他手下谋主陈永华所创，自然是同道中人，当下不住点头。吴六奇又道：“国姓爷昔年率领大军，围攻金陵，可惜寡不敌众，退回台湾，但留在江浙闽三省不及退回的旧部官兵却着实不少。陈先生暗中联络老兄弟，组成了这个天地会，会里的口号是‘天地父母，反清复明’，那便是在下胸口所刺的八个字。寻常会中兄弟，身上也不刺字，在下所以自行刺字，是学一学当年岳武穆尽忠报国的意思。”查伊璜心下甚喜，连喝理两杯酒，说道：“兄台如此行为，才真正不愧为海内奇男子之称了吴六奇道：“‘海内奇男子’，在下愧不敢当，只要查先生认我是个朋友，姓吴的已快活不已了。我们天地会总舵主陈永华陈先生，又有一个名字叫作陈近南，那才着实响当当的英雄好汉，江湖上说起来无人不敬，有两句话说的好：‘平生不识陈近南，就称英雄也枉然。在下尚未见过陈总舵主之面，算不了什么人物。”查伊璜想象陈近南的英雄气概，不禁神往。斟了两杯酒，说道：“来，咱们为陈总舵主干一杯！”两人一口饮干。查伊璜道：“查某一介书生，于国于民，全无裨益。只须将军那一日乘机而动，奋起抗清，查某必当投效军前，稍尽微劳。”自这日起，查伊璜在吴六奇府中，与他日夜密谈，商讨抗清的策略。吴六奇说道：“天地会的势力已逐步扩展到北方诸省，各个大省之中都已开了香堂。查伊璜在吴六奇幕中直耽了六七月之久，这才回乡。回到家里，却大吃一惊，旧宅旁竟起了好大一片新屋，原来吴六奇派人携了广东大小官员所送的礼金，来到浙江查伊璜府上大兴土木，营建楼台。查伊璜素知黄宗羲和顾炎武志切兴复，奔走四方，聚合天下英雄豪杰，共图反清，因此将这件事毫不隐瞒的跟他说了。黄宗羲在舟中将这件事源源本本的告知了吕留良，说道：“此事若有泄漏，给清廷先下手为强，伊璜先生和吴将军固是灭族之祸，而反清的大业是折了一条栋梁。“吕留良道：“除了你我三人之外，此事自是决不能吐露只字，纵然见到伊璜先生，也绝不能提到广东吴将军的名字。“黄宗羲道：“伊璜先生和吴将军有这样一段渊源，朝中大臣对吴将军倚畀正殷，吴将军出面给伊璜先生说项疏通，朝廷非卖他这个面子不可。”吕留良道：“黄兄所见甚是，只不知陆，范二人，如何也和伊璜先生一般，说是未见其书，免罪不究？难道他二人也有朝中有力者代为疏通吗？”黄宗羲道：“吴将军替伊璜先生疏通，倘若单提一人，只怕惹起疑心，拉上两个人来陪衬一下，也未可知。”吕留良笑道：“这等说来，范陆二人只怕直到此刻，还不知这条命是如何拾来的。”顾炎武点头道：“江南名士能多保全一位，也就多保留一份元气。”他三人所谈，乃当世最隐秘之事，其时身在运河舟中，后舱中只有吕室母子三人，黄宗羲又压低了嗓子而说，自不虞为旁人窃听，舟既无墙，也不怕隔墙有耳了。不料顾炎武一句话刚说完，忽听得头顶喋喋一声怪笑。三人大吃一惊，齐喝：“什么人？”却更无半点声息。三人面面相觑，均想：“难道真有鬼怪不成？”三人中顾炎武最为大胆，也学过一点粗浅的防身武艺，一凝神间，伸手入怀，摸出一把匕首，推开窗门，走向船头，凝目向船篷顶瞧去，突然船篷窜起一条非黑影，扑将下来。顾炎武喝道：“是谁？”举匕首向那黑影刺去。但觉手腕一痛，已给人抓住，跟着后心酸麻，已给人点中了穴道，匕首脱手，人也给推进船舱之中。黄走向和吕留良见顾炎武给人推进舱来，后面站着一个黑衣汉子，心中大惊，见那汉子身材魁梧，满面狞笑。吕留良道：“阁下黑夜之中擅自闯入，是何用意？”那人冷笑道：“多谢你们三个挑老子发财哪。吴六奇要造反，查运河要造反，鳌少保得知密报，还不重重有赏？嘿嘿，三位这就跟我上北京去作个见证。”吕顾黄三人暗暗心惊，均深自悔恨：“我们深宵在舟中私语，还是给他听见了，我们行事鲁莽，死不足惜，这一下累了吴将军，可坏了大事。”吕留良道：“阁下说什么话，我们可半点不懂。你要诬陷好人，尽管自己去干，要想拉扯上旁人，那可不行。”他决意以死相拼，如给他杀了，那便死无对证。那大汉冷笑一声，突然欺身向前，在吕留良和黄宗羲胸口各点一点，吕黄二人登时也动弹不得。那大汉哈哈一声，说道：“众位兄弟，都进舱来罢，这一次咱们前锋营立的功劳可大着啦。”后梢几个人齐声答应，进来了四人，都是船家打扮，一齐哈哈大笑。顾黄吕三人面面相觑，知道前锋营是皇帝的亲兵，不知如何，这几人竟会早跟上自己，扮着船夫，一直在船篷外窃听。黄宗羲发吕留良也还罢了，顾炎武这十几年来足迹遍神州，到处结识英雄豪杰，眼光可谓不弱，对这几名船夫竟没留神。只听一名亲兵叫道：“船家调过船头，回杭州去，有什么古怪，小心你的狗命。”后梢上那掌舵的梢公应道：“是！”掌舵梢公是个六七十岁的老头儿，顾炎武雇船时曾跟他说过话，这梢公满脸皱纹，弯腰如弓，确是年长摇橹拉纤的模样，当时见了便毫不起疑。没想到这老梢公虽是货真价实，他手下的船夫都掉了包，自是众亲兵威逼之下，无可奈何，只怪自己但顾得和黄吕二人高谈阔论，陷身危局而不自知。那黑衣大汉笑道：“顾先生，黄先生，吕先生，你们三位名头太大，连京里大老爷们也知道了，否则我们也不会跟上了你们，哈哈！”转头向四位属下道：“咱们得了广东吴提督谋反的真凭实据，这就赶紧去海宁把那姓查的抓了去来。这三个反贼倔强的紧，逃是逃不了的，得提防他们服毒跳河。你们一个钉住一个，有什么岔子，干系可不小。”那四人应道：“是，谨遵瓜管带吩咐。”瓜管带道：“回京后见了鳌少保，人人不愁生官发财。”一名亲兵笑道：“那都是瓜管带提拔栽培，单凭我们四个，那有这等福分？”船头忽然有人嘿嘿一笑，说道：“凭你们四人，原也没这等福分。”船舱门呼的一声，向两旁飞开，一个三十来岁的书生现身舱口，负手背后，脸露微笑。瓜管带道：“官老爷们在这里办案，你是谁？”那书生微笑不答，迈步踏进船舱。刀光闪动，两柄单刀分从左右劈落。那书生闪身避过，随即欺向瓜管带，挥掌拍向他头顶。瓜管带忙伸左臂挡格，右手成拳，猛力击出。那书生左脚反踢，踹中了一名亲兵胸口，那亲兵大叫一声，登时鲜血狂喷。另外三名亲兵举刀或削或剁。船舱中地形狭窄，那书生施展擒拿功夫，劈击勾打，咯的一声响，一名亲兵给他掌缘劈断了颈骨。瓜管带右掌拍出，击向那书生的后脑。那书生反过左掌，砰的一声，双掌相交，瓜管带背心重重撞上船舱，船舱登时塌了一片。那书生连出两掌，拍在余下两名亲兵的胸口，咯咯声响，二人肋骨齐断。瓜管带纵身从船舱缺口中跳将出去。那书生喝到：“那里走？”左掌急拍而出，眼见便将击到他背心，不料瓜管带正在此时左脚反踢，这一掌恰好击在他的足底，一股掌力反而推着他向前飞去。瓜管带急跃窜出，见岸边有一株垂柳挂向河中，当即抓住柳枝，一个倒翻筋斗，飞过了柳树。那书生奔到船头，提起竹篙，挥手掷出。月光之下，竹篙犹似飞蛇，急射而前。但听得瓜管带“啊“的一声长叫，斥革已插入他后心，将他钉在地上，篙身兀自不住晃动。那书生走进船舱，解开顾黄吕三人的穴道，将四名亲兵的尸体抛入运河，重点灯烛。顾黄吕三人不住道谢，问起姓名。那书生笑道：“贱名适才承蒙黄先生齿及，在下姓陈，草字近南。']"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 读取文本文件\n",
    "with open('../data/鹿鼎记.txt', 'r', encoding='utf-8') as f:\n",
    "    txt = f.read()\n",
    "txt = txt.replace('&#39;', '')\n",
    "txt = txt.replace('\\ufeff', '')\n",
    "txt = txt.split()\n",
    "txt[:7]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "collapsed": true,
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['没什麽',\n",
       " '阿碧',\n",
       " '阿二',\n",
       " '阿尔尼',\n",
       " '阿凡提',\n",
       " '阿凡提妻',\n",
       " '阿根',\n",
       " '阿洪',\n",
       " '阿济赤',\n",
       " '阿九',\n",
       " '阿里',\n",
       " '阿曼',\n",
       " '阿三',\n",
       " '阿胜',\n",
       " '阿朱',\n",
       " '阿紫',\n",
       " '哀牢山三十六剑',\n",
       " '矮老者',\n",
       " '安大娘',\n",
       " '安奉日',\n",
       " '安健刚',\n",
       " '安剑清',\n",
       " '黯然销魂掌',\n",
       " '安提督',\n",
       " '暗香疏影',\n",
       " '安小慧',\n",
       " '鳌拜',\n",
       " '敖彪',\n",
       " '獒口夺杖',\n",
       " '巴豆汁',\n",
       " '八方归元',\n",
       " '八卦刀',\n",
       " '八卦金刀',\n",
       " '八卦门',\n",
       " '八卦游龙掌',\n",
       " '八卦游身掌',\n",
       " '八卦掌',\n",
       " '八卦阵',\n",
       " '八荒六合唯我独尊功',\n",
       " '八极门',\n",
       " '八极拳',\n",
       " '八口山铭',\n",
       " '巴郎星',\n",
       " '八十一路三节棍',\n",
       " '八十一路五虎刀',\n",
       " '巴泰',\n",
       " '巴天石',\n",
       " '八仙剑',\n",
       " '八仙剑法',\n",
       " '八仙汤',\n",
       " '巴颜法师',\n",
       " '八阵八卦掌',\n",
       " '白阿绣',\n",
       " '百变千幻衡山云雾十三式',\n",
       " '百草仙',\n",
       " '白二',\n",
       " '白龟寿',\n",
       " '白寒枫',\n",
       " '白寒松',\n",
       " '白鹤亮翅',\n",
       " '百合散',\n",
       " '白虹贯日',\n",
       " '白虎夺命丹',\n",
       " '白虎跳涧',\n",
       " '百花错拳',\n",
       " '百花腹蛇膏',\n",
       " '百花消魂散',\n",
       " '白脸人',\n",
       " '白马啸西风',\n",
       " '白蟒鞭',\n",
       " '百年炼狱',\n",
       " '百胜软鞭',\n",
       " '白世镜',\n",
       " '白驼雪山掌',\n",
       " '白万剑',\n",
       " '百涎丸',\n",
       " '白熊',\n",
       " '白云苍狗',\n",
       " '白云出岫',\n",
       " '白云熊胆丸',\n",
       " '百丈飞瀑',\n",
       " '白振',\n",
       " '白自在',\n",
       " '般若掌',\n",
       " '班淑娴',\n",
       " '棒打狗头',\n",
       " '棒打双犬',\n",
       " '傍花拂柳',\n",
       " '包不同',\n",
       " '抱残式',\n",
       " '鲍承先',\n",
       " '鲍大楚',\n",
       " '抱虎归山',\n",
       " '宝灵拳',\n",
       " '鲍千灵',\n",
       " '宝树',\n",
       " '豹胎易筋丸',\n",
       " '豹尾脚',\n",
       " '包惜弱',\n",
       " '宝象和尚',\n",
       " '鲍鱼之肆',\n",
       " '北斗大阵',\n",
       " '贝海石',\n",
       " '贝锦仪',\n",
       " '北冥神功',\n",
       " '贝人龙',\n",
       " '悲酥清风',\n",
       " '北渔拓跋氏',\n",
       " '本参',\n",
       " '本观',\n",
       " '本相',\n",
       " '本因',\n",
       " '彼得',\n",
       " '笔贴式',\n",
       " '辟邪剑法',\n",
       " '碧血剑',\n",
       " '碧渊腾蛟',\n",
       " '碧针清掌',\n",
       " '冰河倒泻',\n",
       " '冰河开冻',\n",
       " '冰魄银针解药',\n",
       " '拨草寻蛇',\n",
       " '博尔忽',\n",
       " '博尔术',\n",
       " '拨狗朝天',\n",
       " '渤海派',\n",
       " '波罗密手',\n",
       " '波罗星',\n",
       " '拨速台',\n",
       " '拨云见日',\n",
       " '卜沉',\n",
       " '捕风式',\n",
       " '补锅匠',\n",
       " '不戒和尚',\n",
       " '不平道人',\n",
       " '卜泰',\n",
       " '卜垣',\n",
       " '蔡德忠',\n",
       " '蔡威',\n",
       " '采燕功',\n",
       " '参合指',\n",
       " '参阳玉酒',\n",
       " '藏边大丑',\n",
       " '藏边二丑',\n",
       " '藏边三丑',\n",
       " '藏边四丑',\n",
       " '藏边五丑',\n",
       " '藏刀式',\n",
       " '苍松迎客',\n",
       " '曹化淳',\n",
       " '曹令割鼻',\n",
       " '曹猛',\n",
       " '曹能',\n",
       " '曹司朋',\n",
       " '曹云奇',\n",
       " '岑其斯',\n",
       " '曾铁鸥',\n",
       " '曾图南',\n",
       " '察尔珠',\n",
       " '察罕特穆尔',\n",
       " '察合台',\n",
       " '查继左',\n",
       " '查拳',\n",
       " '柴刀十八路',\n",
       " '蟾蜍毒砂',\n",
       " '缠丝擒拿手',\n",
       " '长安一片月',\n",
       " '常伯志',\n",
       " '常长风',\n",
       " '常赫志',\n",
       " '常金鹏',\n",
       " '常敬之',\n",
       " '昌齐',\n",
       " '长青子',\n",
       " '长拳十段锦',\n",
       " '常山蛇势',\n",
       " '常胜宝树王',\n",
       " '长苇渡江',\n",
       " '长啸雁落',\n",
       " '常遇春',\n",
       " '朝阳峰',\n",
       " '车尔库',\n",
       " '车夫',\n",
       " '陈冲之',\n",
       " '陈大方',\n",
       " '陈达海',\n",
       " '陈高波',\n",
       " '陈孤雁',\n",
       " '陈家洛',\n",
       " '陈近南',\n",
       " '陈珂',\n",
       " '陈老丐',\n",
       " '陈七',\n",
       " '陈歪嘴',\n",
       " '陈玄风',\n",
       " '陈友谅',\n",
       " '陈禹',\n",
       " '陈圆圆',\n",
       " '陈正德',\n",
       " '成不忧',\n",
       " '成高道人',\n",
       " '澄观',\n",
       " '澄光',\n",
       " '成璜',\n",
       " '成昆',\n",
       " '程灵素',\n",
       " '程青霜',\n",
       " '程青竹',\n",
       " '澄识',\n",
       " '程坛主',\n",
       " '澄通',\n",
       " '程维藩',\n",
       " '澄心',\n",
       " '程瑶迦',\n",
       " '程遥迦',\n",
       " '程瑛',\n",
       " '城字十八破',\n",
       " '成自学',\n",
       " '迟百诚',\n",
       " '池边调鹤',\n",
       " '池潢不敢顾',\n",
       " '赤尻连拳',\n",
       " '赤老温',\n",
       " '赤日炎炎',\n",
       " '赤蝎粉',\n",
       " '迟玄',\n",
       " '炽翼千羽',\n",
       " '赤月血祭',\n",
       " '赤月血魔剑',\n",
       " '冲灵剑法',\n",
       " '冲虚',\n",
       " '冲虚道长',\n",
       " '冲阵斩将',\n",
       " '丑脸汉子',\n",
       " '仇松年',\n",
       " '出尘子',\n",
       " '处处鸟衔飞',\n",
       " '褚轰',\n",
       " '褚红柏',\n",
       " '褚红柳',\n",
       " '褚万春',\n",
       " '褚万里',\n",
       " '褚圆',\n",
       " '传功长老',\n",
       " '穿针引线',\n",
       " '闯少林',\n",
       " '春蚕掌法',\n",
       " '纯阳无极功',\n",
       " '春云乍展',\n",
       " '慈悲刀',\n",
       " '磁蛋',\n",
       " '丛不弃',\n",
       " '崔百泉',\n",
       " '崔百胜',\n",
       " '崔镖头',\n",
       " '崔绿华',\n",
       " '催命鬼',\n",
       " '崔秋山',\n",
       " '崔希敏',\n",
       " '崔瞎子',\n",
       " '摧心掌',\n",
       " '崔志方',\n",
       " '大悲老人',\n",
       " '大痴',\n",
       " '大癫',\n",
       " '达尔巴',\n",
       " '大风云飞掌',\n",
       " '大伏魔拳法',\n",
       " '打狗棒法',\n",
       " '打狗入洞',\n",
       " '大海沉沙',\n",
       " '大汉',\n",
       " '大金刚拳',\n",
       " '大金刚掌',\n",
       " '大苦',\n",
       " '大理段氏',\n",
       " '大力鹰爪功',\n",
       " '大漠飞沙',\n",
       " '大漠孤烟',\n",
       " '达摩剑',\n",
       " '达魔剑',\n",
       " '达摩剑术',\n",
       " '大漠平沙',\n",
       " '达摩掌',\n",
       " '大擒拿手',\n",
       " '大摔碑手',\n",
       " '大嵩阳神掌',\n",
       " '大头鬼',\n",
       " '大韦陀杵',\n",
       " '大夜叉棍',\n",
       " '大智禅师',\n",
       " '大智无定指',\n",
       " '呆若木鸡',\n",
       " '戴永明',\n",
       " '岱宗如何',\n",
       " '单鞭',\n",
       " '单伯山',\n",
       " '单季山',\n",
       " '丹青生',\n",
       " '单叔山',\n",
       " '单铁生',\n",
       " '弹腿',\n",
       " '单小山',\n",
       " '单正',\n",
       " '当头棒喝',\n",
       " '刀白凤',\n",
       " '刀剑如梦',\n",
       " '到江吴地尽',\n",
       " '倒提金钟',\n",
       " '倒行逆施',\n",
       " '捣虚式',\n",
       " '德布',\n",
       " '德成',\n",
       " '德鄂',\n",
       " '邓八公',\n",
       " '邓百川',\n",
       " '邓炳春',\n",
       " '登嵩望月',\n",
       " '邓愈',\n",
       " '狄镖头',\n",
       " '地火功',\n",
       " '狄青降龙',\n",
       " '地堂刀法',\n",
       " '狄修',\n",
       " '羝羊触蕃',\n",
       " '狄云',\n",
       " '地藏颂经',\n",
       " '店伴',\n",
       " '点苍剑法',\n",
       " '点苍派',\n",
       " '点苍渔隐',\n",
       " '电母照镜',\n",
       " '貂蝉拜月',\n",
       " '吊客临门',\n",
       " '吊死鬼',\n",
       " '叠翠浮青',\n",
       " '丁不三',\n",
       " '丁不四',\n",
       " '丁春秋',\n",
       " '丁大全',\n",
       " '丁珰',\n",
       " '丁典',\n",
       " '丁坚',\n",
       " '定静师太',\n",
       " '丁勉',\n",
       " '丁敏尹',\n",
       " '定判生死',\n",
       " '丁同',\n",
       " '定阳针',\n",
       " '定逸',\n",
       " '丁游',\n",
       " '东方不败',\n",
       " '董开山',\n",
       " '洞玄',\n",
       " '都大锦',\n",
       " '都史',\n",
       " '斗转星移',\n",
       " '杜百当',\n",
       " '独臂刀法',\n",
       " '毒钉解药',\n",
       " '渡厄',\n",
       " '独孤九剑',\n",
       " '独孤求败',\n",
       " '毒蒺藜解药',\n",
       " '毒箭解药',\n",
       " '渡劫',\n",
       " '杜立德',\n",
       " '毒菱解药',\n",
       " '毒龙鞭法',\n",
       " '渡难',\n",
       " '毒雾',\n",
       " '杜希孟',\n",
       " '毒盐',\n",
       " '杜长老',\n",
       " '毒针环解药',\n",
       " '毒针解药',\n",
       " '断肠散',\n",
       " '断筋腐骨丸',\n",
       " '端木氏赤焰洞',\n",
       " '端木元',\n",
       " '段天德',\n",
       " '段延庆',\n",
       " '段誉',\n",
       " '段正淳',\n",
       " '段正明',\n",
       " '对喀纳',\n",
       " '多尔衮',\n",
       " '多隆',\n",
       " '多罗叶指',\n",
       " '夺命连环三仙剑',\n",
       " '夺魄指',\n",
       " '鄂北鬼见愁锺家',\n",
       " '鄂尔多',\n",
       " '恶虎拦路',\n",
       " '峨嵋剑法',\n",
       " '峨嵋九阳功',\n",
       " '峨嵋派',\n",
       " '恶魔火神剑',\n",
       " '恶犬拦路',\n",
       " '二十八招石鼓打穴笔法',\n",
       " '范百龄',\n",
       " '范帮主',\n",
       " '樊纲',\n",
       " '梵光初现',\n",
       " '范骅',\n",
       " '反截狗臀',\n",
       " '饭来伸手',\n",
       " '反两仪刀法',\n",
       " '范文程',\n",
       " '范遥',\n",
       " '范一飞',\n",
       " '樊一翁',\n",
       " '范禹',\n",
       " '范中恩',\n",
       " '方大洪',\n",
       " '方东白',\n",
       " '方人智',\n",
       " '方生',\n",
       " '方天劳',\n",
       " '方怡',\n",
       " '方有德',\n",
       " '方证',\n",
       " '费彬',\n",
       " '飞凤手',\n",
       " '飞花落叶',\n",
       " '飞龙在天',\n",
       " '废寝忘食',\n",
       " '飞沙走石',\n",
       " '飞天神行',\n",
       " '飞絮劲',\n",
       " '飞燕穿柳',\n",
       " '飞燕回翔',\n",
       " '费要多罗',\n",
       " '分花拂柳',\n",
       " '分筋错骨手',\n",
       " '冯阿三',\n",
       " '风波恶',\n",
       " '冯不摧',\n",
       " '封不平',\n",
       " '冯不破',\n",
       " '凤点头',\n",
       " '冯衡',\n",
       " '冯辉',\n",
       " '峰回路转',\n",
       " '风际中',\n",
       " '风卷长草',\n",
       " '风良',\n",
       " '冯默风',\n",
       " '疯魔杖法',\n",
       " '冯难敌',\n",
       " '凤南天',\n",
       " '凤七',\n",
       " '风清扬',\n",
       " '风轻云淡',\n",
       " '凤曲长鸣',\n",
       " '风沙莽莽',\n",
       " '封豕长蛇',\n",
       " '冯坦',\n",
       " '封坛主',\n",
       " '冯同知',\n",
       " '封万里',\n",
       " '冯锡范',\n",
       " '凤一鸣',\n",
       " '风云三使',\n",
       " '冯振武',\n",
       " '拂尘拂法',\n",
       " '富春',\n",
       " '腐骨穿心膏',\n",
       " '伏虎拳',\n",
       " '伏虎掌',\n",
       " '福康安',\n",
       " '茯苓首乌丸',\n",
       " '茯苓燕窝汤',\n",
       " '符敏仪',\n",
       " '伏魔剑',\n",
       " '伏魔杖法',\n",
       " '伏牛派',\n",
       " '夫妻刀法',\n",
       " '抚琴按箫',\n",
       " '抚琴式',\n",
       " '浮丘挹袖',\n",
       " '芙蓉剑法',\n",
       " '芙蓉金针',\n",
       " '伏狮胜象',\n",
       " '傅思归',\n",
       " '俯听闻惊风',\n",
       " '负子渡河',\n",
       " '丐帮',\n",
       " '盖一鸣',\n",
       " '盖运聪',\n",
       " '甘宝宝',\n",
       " '罡风扫叶',\n",
       " '刚相',\n",
       " '高管家',\n",
       " '高克新',\n",
       " '高老者',\n",
       " '高里津',\n",
       " '高明根',\n",
       " '高三娘子',\n",
       " '高升泰',\n",
       " '高师弟',\n",
       " '高探马',\n",
       " '高颜超',\n",
       " '高则成',\n",
       " '隔岸越山多',\n",
       " '葛尔丹',\n",
       " '葛光佩',\n",
       " '葛通',\n",
       " '葛长老',\n",
       " '耿天霜',\n",
       " '耿万钟',\n",
       " '龚光杰',\n",
       " '宫九佳',\n",
       " '公孙绿萼',\n",
       " '公孙止',\n",
       " '公冶乾',\n",
       " '苟读',\n",
       " '狗头陀',\n",
       " '钩弋握拳',\n",
       " '古笃诚',\n",
       " '孤鸿海上来',\n",
       " '顾金标',\n",
       " '古木丛青霭',\n",
       " '古墓派',\n",
       " '孤萍对月',\n",
       " '贾人达',\n",
       " '古若般',\n",
       " '鼓瑟式',\n",
       " '姑苏慕容',\n",
       " '孤形只影',\n",
       " '顾炎武',\n",
       " '关安基',\n",
       " '关冲剑',\n",
       " '关明梅',\n",
       " '广西黎山洞黎老汉',\n",
       " '归二娘',\n",
       " '贵妃回眸',\n",
       " '贵妃醉酒',\n",
       " '归辛树',\n",
       " '归钟',\n",
       " '郭芙',\n",
       " '郭靖',\n",
       " '郭破虏',\n",
       " '过三拳',\n",
       " '郭襄',\n",
       " '郭啸天',\n",
       " '过眼云烟',\n",
       " '过彦之',\n",
       " '郭玉堂',\n",
       " '哈赤大师',\n",
       " '哈大霸',\n",
       " '哈合台',\n",
       " '蛤蟆功',\n",
       " '哈总管',\n",
       " '海大富',\n",
       " '海兰弼',\n",
       " '海市蜃楼',\n",
       " '韩宝驹',\n",
       " '寒冰绵掌',\n",
       " '寒冰神掌',\n",
       " '韩春霖',\n",
       " '韩老丐',\n",
       " '韩林儿',\n",
       " '寒梅剑阵',\n",
       " '韩千叶',\n",
       " '罕贴摩',\n",
       " '韩文冲',\n",
       " '韩无垢',\n",
       " '韩小莹',\n",
       " '寒玉冰蟾膏',\n",
       " '汉子',\n",
       " '郝大通',\n",
       " '郝密',\n",
       " '郝太监',\n",
       " '河北沧州郑家',\n",
       " '鹤笔翁',\n",
       " '赫大通',\n",
       " '和尔大',\n",
       " '鹤飞九天',\n",
       " '何红药',\n",
       " '贺老三',\n",
       " '和里布',\n",
       " '赫连铁树',\n",
       " '何三七',\n",
       " '鹤蛇八打',\n",
       " '和珅',\n",
       " '河朔立威',\n",
       " '何思豪',\n",
       " '何太冲',\n",
       " '何惕守',\n",
       " '何铁手',\n",
       " '何望海',\n",
       " '鹤翔紫盖',\n",
       " '何足道',\n",
       " '黑暗诱惑',\n",
       " '黑白子',\n",
       " '黑风双煞',\n",
       " '黑砂掌',\n",
       " '黑熊',\n",
       " '黑玉断续膏',\n",
       " '横拜观音',\n",
       " '恒河入海',\n",
       " '横江飞渡',\n",
       " '恒山剑法',\n",
       " '恒山剑阵',\n",
       " '恒山派',\n",
       " '衡山派',\n",
       " '衡山五神剑',\n",
       " '横云断峰',\n",
       " '恒云悯天',\n",
       " '洪安通',\n",
       " '洪朝',\n",
       " '鸿飞冥冥',\n",
       " '红拂夜奔',\n",
       " '鸿渐于陆',\n",
       " '洪凌波',\n",
       " '虹魔嗜血剑',\n",
       " '红娘子',\n",
       " '洪七公',\n",
       " '洪人雄',\n",
       " '洪胜海',\n",
       " '红线盗盒',\n",
       " '洪崖拍肩',\n",
       " '红玉击鼓',\n",
       " '烘云托月',\n",
       " '侯飞文',\n",
       " '侯人英',\n",
       " '後山三绝招',\n",
       " '侯通海',\n",
       " '呼巴音',\n",
       " '湖北阮家',\n",
       " '忽必烈',\n",
       " '胡大哥',\n",
       " '胡德第',\n",
       " '忽都虎',\n",
       " '胡斐',\n",
       " '胡夫人',\n",
       " '胡桂南',\n",
       " '胡家刀法',\n",
       " '虎筋虎骨膏',\n",
       " '胡老三',\n",
       " '胡老爷',\n",
       " '忽伦大虎',\n",
       " '忽伦二虎',\n",
       " '忽伦三虎',\n",
       " '忽伦四虎',\n",
       " '胡马越岭',\n",
       " '胡青牛',\n",
       " '胡青羊',\n",
       " '呼延十八鞭',\n",
       " '呼延万善',\n",
       " '胡一刀',\n",
       " '胡逸之',\n",
       " '呼音克',\n",
       " '虎爪绝户手',\n",
       " '虎爪手',\n",
       " '花斑毒蝎解药',\n",
       " '华伯斯基',\n",
       " '化功大法',\n",
       " '化骨绵掌',\n",
       " '花海飘香',\n",
       " '华赫艮',\n",
       " '花剑影',\n",
       " '花开见佛',\n",
       " '花前月下',\n",
       " '华拳门',\n",
       " '花容云鬓',\n",
       " '华山剑法',\n",
       " '华山派',\n",
       " '化石丹',\n",
       " '化尸粉',\n",
       " '花铁干',\n",
       " '花万紫',\n",
       " '化血腐骨粉',\n",
       " '花云',\n",
       " '华筝',\n",
       " '怀素自叙帖',\n",
       " '幻剑式',\n",
       " '幻阴指',\n",
       " '黄伯流',\n",
       " '黄二毛子',\n",
       " '黄甫',\n",
       " '皇甫阁',\n",
       " '黄国柏',\n",
       " '黄金魁',\n",
       " '黄眉和尚',\n",
       " '黄面道人',\n",
       " '黄泉逆流',\n",
       " '黄蓉',\n",
       " '黄沙万里鞭法',\n",
       " '黄裳',\n",
       " '皇太后',\n",
       " '皇太极',\n",
       " '黄希节',\n",
       " '黄须人',\n",
       " '皇扬震首',\n",
       " '黄药师',\n",
       " '黄真',\n",
       " '黄钟公',\n",
       " '黄锺公',\n",
       " '黄宗羲',\n",
       " '晦聪禅师',\n",
       " '回打软鞭十三式',\n",
       " '慧方',\n",
       " '回风拂柳刀',\n",
       " '回风拂柳拳',\n",
       " '回风落雁剑',\n",
       " '回风落雁剑法',\n",
       " '回风掌',\n",
       " '慧观',\n",
       " '慧净',\n",
       " '回头是岸',\n",
       " '彗星飞堕',\n",
       " '回雁三决',\n",
       " '回阳五龙膏',\n",
       " '辉月使',\n",
       " '慧真',\n",
       " '混沌一破',\n",
       " '混天功',\n",
       " '混元功',\n",
       " '混元剑',\n",
       " '混元气',\n",
       " '混元掌',\n",
       " '霍阿伊',\n",
       " '霍都',\n",
       " '霍青桐',\n",
       " '火焰刀',\n",
       " '霍元龙',\n",
       " '季镖头',\n",
       " '寂灭爪',\n",
       " '姬清玄',\n",
       " '吉人通',\n",
       " '计无施',\n",
       " '姬晓峰',\n",
       " '纪晓芙',\n",
       " '急雨打萍',\n",
       " '季长老',\n",
       " '挤字诀',\n",
       " '贾布',\n",
       " '家传剑法',\n",
       " '假东方不败',\n",
       " '贾老六',\n",
       " '贾老者',\n",
       " '袈裟伏魔功',\n",
       " '剪补自如',\n",
       " '建除',\n",
       " '简管家',\n",
       " '简捷',\n",
       " '见龙在田',\n",
       " '剑落双雁',\n",
       " '建宁公主',\n",
       " '见人伸手',\n",
       " '简长老',\n",
       " '江百胜',\n",
       " '江南史家',\n",
       " '江上弄笛',\n",
       " '僵尸拳',\n",
       " '蒋士铨',\n",
       " '姜师叔',\n",
       " '蒋四根',\n",
       " '蒋涛',\n",
       " '蒋天寿',\n",
       " '蒋调侯',\n",
       " '姜铁山',\n",
       " '姜文',\n",
       " '姜小铁',\n",
       " '脚夫',\n",
       " '焦公礼',\n",
       " '焦木和尚',\n",
       " '教士',\n",
       " '教书先生',\n",
       " '焦宛儿',\n",
       " '焦文期',\n",
       " '教长',\n",
       " '解风',\n",
       " '截剑式',\n",
       " '截手九式',\n",
       " '杰书',\n",
       " '解文豹',\n",
       " '解穴秘诀',\n",
       " '锦笔生花',\n",
       " '进步搬拦锤',\n",
       " '金顶佛光',\n",
       " '金顶绵掌',\n",
       " '金刚般若掌',\n",
       " '金刚禅狮子吼',\n",
       " '金刚掣尾',\n",
       " '金刚伏虎',\n",
       " '金刚伏虎刀法',\n",
       " '金刚护体神功',\n",
       " '金刚门',\n",
       " '金刚指',\n",
       " '金花婆婆',\n",
       " '金龙鞭法',\n",
       " '金轮度劫',\n",
       " '金轮法王',\n",
       " '紧罗那王棍',\n",
       " '金蛇缠丝式',\n",
       " '金蛇剑法',\n",
       " '金蛇郎君',\n",
       " '金蛇盘身式',\n",
       " '金蛇擒鹤拳',\n",
       " '金蛇升天式',\n",
       " '金蛇吐信式',\n",
       " '金蛇游身式',\n",
       " '金蛇游身掌',\n",
       " '金蛇锥',\n",
       " '金声玉振',\n",
       " '金乌刀法',\n",
       " '金乌派',\n",
       " '金雁功',\n",
       " '金雁横空',\n",
       " '金玉满堂',\n",
       " '金针渡劫',\n",
       " '静慧',\n",
       " '净济',\n",
       " '静迦',\n",
       " '静空',\n",
       " '敬捧宝经',\n",
       " '净清',\n",
       " '静虚师太',\n",
       " '静玄师太',\n",
       " '静照',\n",
       " '静智大师',\n",
       " '九花玉露丸',\n",
       " '九九丸',\n",
       " '九龙鞭法',\n",
       " '九龙派',\n",
       " '鸠摩智',\n",
       " '九难',\n",
       " '鸠砒丸',\n",
       " '九品连台',\n",
       " '九死生',\n",
       " '九翼道人',\n",
       " '九阴白骨爪',\n",
       " '九阴白骨爪解药',\n",
       " '九阴真经',\n",
       " '救赵挥金锤',\n",
       " '九转回春汤',\n",
       " '九转灵宝丸',\n",
       " '九转熊蛇丸',\n",
       " '菊芳',\n",
       " '举火撩天',\n",
       " '举火燎天',\n",
       " '菊剑',\n",
       " '俱明宝树王',\n",
       " '菊友',\n",
       " '绝剑',\n",
       " '绝门毒针解药',\n",
       " '绝情丹',\n",
       " '觉远',\n",
       " '觉远大师',\n",
       " '觉月',\n",
       " '峻岭横空',\n",
       " '喀丝丽',\n",
       " '凯别兴',\n",
       " '开门揖盗',\n",
       " '开山掌',\n",
       " '康广陵',\n",
       " '亢龙有悔',\n",
       " '柯万钧',\n",
       " '柯镇恶',\n",
       " '控鹤功',\n",
       " '空击苍鹰',\n",
       " '空见',\n",
       " '空明拳',\n",
       " '孔雀开屏',\n",
       " '崆峒派',\n",
       " '空闻',\n",
       " '空心菜',\n",
       " '空性',\n",
       " '空智',\n",
       " '枯草岭',\n",
       " '苦茶斋',\n",
       " '枯木',\n",
       " '苦恼拳',\n",
       " '枯荣禅功',\n",
       " '枯荣长老',\n",
       " '快活三十掌',\n",
       " '邝宝官',\n",
       " '狂风快剑',\n",
       " '邝天雄',\n",
       " '葵花宝典',\n",
       " '昆仑剑法',\n",
       " '昆仑派',\n",
       " '坤宁宫',\n",
       " '腊八粥',\n",
       " '来福儿',\n",
       " '来鹤清泉',\n",
       " '蓝凤凰',\n",
       " '兰花拂穴手',\n",
       " '兰剑',\n",
       " '烂柯山',\n",
       " '蓝秦',\n",
       " '揽雀尾',\n",
       " '蓝天和',\n",
       " '琅環玉洞',\n",
       " '浪迹天涯',\n",
       " '郎师傅',\n",
       " '朗月无云',\n",
       " '浪子回头',\n",
       " '老不死',\n",
       " '劳德诺',\n",
       " '老李',\n",
       " '劳太监',\n",
       " '老头子',\n",
       " '老王',\n",
       " '老吴',\n",
       " '捞月式',\n",
       " '老枝横斜',\n",
       " '乐厚',\n",
       " '乐寿堂',\n",
       " '雷动於九天之上剑法',\n",
       " '擂鼓山',\n",
       " '雷猛',\n",
       " '雷一啸',\n",
       " '冷谦',\n",
       " '力不从心',\n",
       " '李春来',\n",
       " '丽春院',\n",
       " '李大元',\n",
       " '黎夫人',\n",
       " '黎刚',\n",
       " '厉鬼峰',\n",
       " '李可秀',\n",
       " '李傀儡',\n",
       " '李力世',\n",
       " '李莫愁',\n",
       " '力劈三关',\n",
       " '李萍',\n",
       " '李秋水',\n",
       " '李三',\n",
       " '利涉大川',\n",
       " '黎生',\n",
       " '李式开',\n",
       " '李四',\n",
       " '李四摧',\n",
       " '李天恒',\n",
       " '李廷豹',\n",
       " '李万山',\n",
       " '李文秀',\n",
       " '李西华',\n",
       " '梨香院',\n",
       " '李岩',\n",
       " '李沅芷',\n",
       " '李志常',\n",
       " '李自成',\n",
       " '连诚诀',\n",
       " '莲花山庄',\n",
       " '怜花研姿',\n",
       " '莲花掌',\n",
       " '连环迷踪腿',\n",
       " '莲蓉豆泥蜜糖珍珠糊',\n",
       " '连山若波涛',\n",
       " '梁发',\n",
       " '两回玛法',\n",
       " '两仪剑法',\n",
       " '梁长老',\n",
       " '梁自进',\n",
       " '梁子翁',\n",
       " '廖自砺',\n",
       " '烈火丹',\n",
       " '林朝英',\n",
       " '林平之',\n",
       " '林兴珠',\n",
       " '林永超',\n",
       " '林玉龙',\n",
       " '林远图',\n",
       " '林震南',\n",
       " '灵鳌步',\n",
       " '灵獒营',\n",
       " '凌波微步',\n",
       " '灵龟阁',\n",
       " '令狐冲',\n",
       " '灵境寺',\n",
       " '灵鹫宫',\n",
       " '灵鹫听经',\n",
       " '灵山礼佛',\n",
       " '岭上双梅',\n",
       " '灵蛇岛',\n",
       " '灵蛇拳',\n",
       " '凌霜华',\n",
       " '凌退思',\n",
       " '凌霄城',\n",
       " '灵虚',\n",
       " '灵应寺',\n",
       " '灵智上人',\n",
       " '刘敖',\n",
       " '刘乘风',\n",
       " '刘处玄',\n",
       " '柳大洪',\n",
       " '刘芳亮',\n",
       " '刘鹤真',\n",
       " '刘菁',\n",
       " '六脉神剑',\n",
       " '刘培生',\n",
       " '刘芹',\n",
       " '流泉下山',\n",
       " '六神不安',\n",
       " '六十四路轰天雷震挡法',\n",
       " ...]"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 查看自定义词典\n",
    "with open('../data/self_dict.txt', 'r', encoding='utf-8') as f:\n",
    "    my_dict = f.read()\n",
    "my_dict = my_dict.split()\n",
    "my_dict"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[['《', '鹿鼎记', '》'],\n",
       " ['作者', '：', '金庸'],\n",
       " ['正文'],\n",
       " ['第一回'],\n",
       " ['纵横', '钩', '党', '清流', '祸']]"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 分词\n",
    "jieba.load_userdict('../data/self_dict.txt')\n",
    "data_cut = [jieba.lcut(i) for i in txt]\n",
    "data_cut[:5]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "## 绘制词云"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true,
    "hidden": true
   },
   "source": [
    "### 词频统计"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "collapsed": true,
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['《',\n",
       " '鹿鼎记',\n",
       " '》',\n",
       " '作者',\n",
       " '：',\n",
       " '金庸',\n",
       " '正文',\n",
       " '第一回',\n",
       " '纵横',\n",
       " '钩',\n",
       " '党',\n",
       " '清流',\n",
       " '祸',\n",
       " '峭茜',\n",
       " '风期',\n",
       " '月旦评',\n",
       " '北风',\n",
       " '如刀',\n",
       " '，',\n",
       " '满地',\n",
       " '冰霜',\n",
       " '。',\n",
       " '江南',\n",
       " '近',\n",
       " '海滨',\n",
       " '的',\n",
       " '一条',\n",
       " '大',\n",
       " '路上',\n",
       " '，',\n",
       " '一队',\n",
       " '清兵',\n",
       " '手执',\n",
       " '刀枪',\n",
       " '，',\n",
       " '押',\n",
       " '着',\n",
       " '七辆',\n",
       " '囚车',\n",
       " '，',\n",
       " '冲风冒寒',\n",
       " '，',\n",
       " '向',\n",
       " '北而行',\n",
       " '。',\n",
       " '前面',\n",
       " '三辆',\n",
       " '囚车',\n",
       " '中',\n",
       " '分别',\n",
       " '监禁',\n",
       " '的',\n",
       " '是',\n",
       " '三个',\n",
       " '男子',\n",
       " '，',\n",
       " '都',\n",
       " '作',\n",
       " '书生',\n",
       " '打扮',\n",
       " '，',\n",
       " '一个',\n",
       " '是',\n",
       " '白发',\n",
       " '老者',\n",
       " '，',\n",
       " '两个',\n",
       " '是',\n",
       " '中年人',\n",
       " '。',\n",
       " '后面',\n",
       " '四辆',\n",
       " '囚车',\n",
       " '中',\n",
       " '坐',\n",
       " '的',\n",
       " '是',\n",
       " '女子',\n",
       " '，',\n",
       " '最后',\n",
       " '一辆',\n",
       " '囚车',\n",
       " '中是',\n",
       " '个',\n",
       " '少',\n",
       " '*',\n",
       " '妇',\n",
       " '，',\n",
       " '怀中',\n",
       " '抱',\n",
       " '着',\n",
       " '个',\n",
       " '女婴',\n",
       " '，',\n",
       " '女婴',\n",
       " '啼哭',\n",
       " '不休',\n",
       " '。',\n",
       " '她',\n",
       " '母亲',\n",
       " '温言',\n",
       " '相呵',\n",
       " '，',\n",
       " '女婴',\n",
       " '只是',\n",
       " '大哭',\n",
       " '。',\n",
       " '囚车',\n",
       " '旁一',\n",
       " '清兵',\n",
       " '恼',\n",
       " '了',\n",
       " '，',\n",
       " '伸腿',\n",
       " '在',\n",
       " '车上',\n",
       " '踢',\n",
       " '了',\n",
       " '一脚',\n",
       " '，',\n",
       " '喝道',\n",
       " '：',\n",
       " '“',\n",
       " '再',\n",
       " '哭',\n",
       " '，',\n",
       " '再',\n",
       " '哭',\n",
       " '，',\n",
       " '老子',\n",
       " '踢死',\n",
       " '你',\n",
       " '！',\n",
       " '”',\n",
       " '那',\n",
       " '女婴',\n",
       " '一惊',\n",
       " '，',\n",
       " '哭',\n",
       " '得',\n",
       " '更加',\n",
       " '响',\n",
       " '了',\n",
       " '。',\n",
       " '离开',\n",
       " '道路',\n",
       " '数十丈',\n",
       " '处有',\n",
       " '座',\n",
       " '大屋',\n",
       " '，',\n",
       " '屋檐下',\n",
       " '站',\n",
       " '着',\n",
       " '一个',\n",
       " '中年',\n",
       " '文士',\n",
       " '，',\n",
       " '一个十一二岁',\n",
       " '的',\n",
       " '小孩',\n",
       " '。',\n",
       " '那文士',\n",
       " '见到',\n",
       " '这',\n",
       " '等',\n",
       " '情景',\n",
       " '，',\n",
       " '不禁',\n",
       " '长叹一声',\n",
       " '，',\n",
       " '眼眶',\n",
       " '也',\n",
       " '红',\n",
       " '了',\n",
       " '，',\n",
       " '说道',\n",
       " '：',\n",
       " '“',\n",
       " '可怜',\n",
       " '，',\n",
       " '可怜',\n",
       " '！',\n",
       " '”',\n",
       " '农',\n",
       " '小孩',\n",
       " '问道',\n",
       " '：',\n",
       " '“',\n",
       " '爹爹',\n",
       " '，',\n",
       " '他们',\n",
       " '犯',\n",
       " '了',\n",
       " '什么',\n",
       " '罪',\n",
       " '？',\n",
       " '”',\n",
       " '那文士',\n",
       " '道',\n",
       " '：',\n",
       " '“',\n",
       " '又',\n",
       " '犯',\n",
       " '了',\n",
       " '什么',\n",
       " '罪',\n",
       " '？',\n",
       " '昨日',\n",
       " '和',\n",
       " '今朝',\n",
       " '已',\n",
       " '逮去',\n",
       " '了',\n",
       " '三十几',\n",
       " '人',\n",
       " '，',\n",
       " '都',\n",
       " '是',\n",
       " '我们',\n",
       " '浙江',\n",
       " '有名',\n",
       " '的',\n",
       " '读书人',\n",
       " '，',\n",
       " '个个',\n",
       " '都',\n",
       " '是',\n",
       " '无辜',\n",
       " '株连',\n",
       " '。',\n",
       " '”',\n",
       " '他',\n",
       " '说',\n",
       " '到',\n",
       " '“',\n",
       " '无辜',\n",
       " '株连',\n",
       " '”',\n",
       " '四子',\n",
       " '，',\n",
       " '声音',\n",
       " '压得',\n",
       " '甚低',\n",
       " '，',\n",
       " '生怕',\n",
       " '给',\n",
       " '押',\n",
       " '囚车',\n",
       " '的',\n",
       " '官兵',\n",
       " '听见',\n",
       " '了',\n",
       " '。',\n",
       " '那',\n",
       " '小孩',\n",
       " '道',\n",
       " '：',\n",
       " '“',\n",
       " '哪个',\n",
       " '小女孩',\n",
       " '还',\n",
       " '在',\n",
       " '吃奶',\n",
       " '，',\n",
       " '难道',\n",
       " '也',\n",
       " '犯了罪',\n",
       " '么',\n",
       " '？',\n",
       " '真',\n",
       " '没道理',\n",
       " '。',\n",
       " '”',\n",
       " '那文士',\n",
       " '道',\n",
       " '：',\n",
       " '“',\n",
       " '你',\n",
       " '懂得',\n",
       " '官兵',\n",
       " '没道理',\n",
       " '，',\n",
       " '真是',\n",
       " '好孩子',\n",
       " '。',\n",
       " '咳',\n",
       " '，',\n",
       " '人为刀俎',\n",
       " '，',\n",
       " '我为鱼肉',\n",
       " '，',\n",
       " '人为',\n",
       " '鼎锅',\n",
       " '，',\n",
       " '我',\n",
       " '为',\n",
       " '麋鹿',\n",
       " '！',\n",
       " '”',\n",
       " '那',\n",
       " '小孩',\n",
       " '道',\n",
       " '：',\n",
       " '“',\n",
       " '爹',\n",
       " '，',\n",
       " '你',\n",
       " '前',\n",
       " '几天',\n",
       " '教过',\n",
       " '我',\n",
       " '。',\n",
       " '人为刀俎',\n",
       " '，',\n",
       " '我为鱼肉',\n",
       " '，',\n",
       " '就是',\n",
       " '给',\n",
       " '人家',\n",
       " '斩割',\n",
       " '屠杀',\n",
       " '的',\n",
       " '意思',\n",
       " '。',\n",
       " '人家',\n",
       " '是',\n",
       " '切菜刀',\n",
       " '，',\n",
       " '是',\n",
       " '铁板',\n",
       " '，',\n",
       " '我们',\n",
       " '就是',\n",
       " '鱼',\n",
       " '和',\n",
       " '肉',\n",
       " '。',\n",
       " '“',\n",
       " '人为',\n",
       " '鼎锅',\n",
       " '，',\n",
       " '我',\n",
       " '为',\n",
       " '麋鹿',\n",
       " '”',\n",
       " '这',\n",
       " '两句话',\n",
       " '，',\n",
       " '意思',\n",
       " '也',\n",
       " '差不多',\n",
       " '么',\n",
       " '？',\n",
       " '”',\n",
       " '那文士',\n",
       " '道',\n",
       " '：',\n",
       " '“',\n",
       " '正是',\n",
       " '！',\n",
       " '”',\n",
       " '眼见',\n",
       " '官兵',\n",
       " '和',\n",
       " '囚车',\n",
       " '已经',\n",
       " '去',\n",
       " '远',\n",
       " '，',\n",
       " '拉着',\n",
       " '小孩',\n",
       " '的',\n",
       " '手道',\n",
       " '：',\n",
       " '“',\n",
       " '外面',\n",
       " '风大',\n",
       " '，',\n",
       " '我们',\n",
       " '回',\n",
       " '屋里',\n",
       " '去',\n",
       " '。',\n",
       " '”',\n",
       " '当下',\n",
       " '父子',\n",
       " '二人',\n",
       " '走进',\n",
       " '书房',\n",
       " '。',\n",
       " '那文士',\n",
       " '提笔',\n",
       " '醮',\n",
       " '上',\n",
       " '了',\n",
       " '墨',\n",
       " '，',\n",
       " '在',\n",
       " '纸',\n",
       " '上',\n",
       " '写',\n",
       " '了',\n",
       " '个',\n",
       " '“',\n",
       " '鹿',\n",
       " '”',\n",
       " '字',\n",
       " '，',\n",
       " '说道',\n",
       " '：',\n",
       " '“',\n",
       " '鹿',\n",
       " '这种',\n",
       " '野兽',\n",
       " '，',\n",
       " '虽',\n",
       " '是',\n",
       " '庞然大物',\n",
       " '，',\n",
       " '性子',\n",
       " '却',\n",
       " '极为',\n",
       " '平和',\n",
       " '，',\n",
       " '只',\n",
       " '吃',\n",
       " '青草',\n",
       " '和',\n",
       " '树叶',\n",
       " '，',\n",
       " '从来不',\n",
       " '伤害',\n",
       " '别的',\n",
       " '野兽',\n",
       " '。',\n",
       " '凶猛',\n",
       " '的',\n",
       " '野兽',\n",
       " '要伤',\n",
       " '它',\n",
       " '吃',\n",
       " '它',\n",
       " '，',\n",
       " '它',\n",
       " '只有',\n",
       " '逃跑',\n",
       " '倘若',\n",
       " '逃不了',\n",
       " '，',\n",
       " '那',\n",
       " '只有',\n",
       " '给',\n",
       " '人家',\n",
       " '吃力',\n",
       " '。',\n",
       " '”',\n",
       " '又',\n",
       " '写',\n",
       " '了',\n",
       " '“',\n",
       " '逐鹿',\n",
       " '”',\n",
       " '两字',\n",
       " '，',\n",
       " '说道',\n",
       " '：',\n",
       " '“',\n",
       " '因此',\n",
       " '古人',\n",
       " '常常',\n",
       " '拿',\n",
       " '鹿来',\n",
       " '比喻',\n",
       " '天下',\n",
       " '。',\n",
       " '世上',\n",
       " '百姓',\n",
       " '都',\n",
       " '温顺',\n",
       " '善良',\n",
       " '，',\n",
       " '只有',\n",
       " '给',\n",
       " '人',\n",
       " '欺压',\n",
       " '残害',\n",
       " '的',\n",
       " '份儿',\n",
       " '。',\n",
       " '《',\n",
       " '汉书',\n",
       " '》',\n",
       " '上',\n",
       " '说',\n",
       " '：',\n",
       " '“',\n",
       " '秦失其鹿',\n",
       " '，',\n",
       " '天下',\n",
       " '共逐',\n",
       " '之',\n",
       " '。',\n",
       " '”',\n",
       " '那就是说',\n",
       " '，',\n",
       " '秦朝',\n",
       " '失',\n",
       " '了',\n",
       " '天下',\n",
       " '，',\n",
       " '群雄',\n",
       " '并',\n",
       " '起',\n",
       " '，',\n",
       " '大家',\n",
       " '争夺',\n",
       " '，',\n",
       " '最后',\n",
       " '汉高祖',\n",
       " '打败',\n",
       " '了',\n",
       " '楚霸王',\n",
       " '，',\n",
       " '就',\n",
       " '得',\n",
       " '了',\n",
       " '这',\n",
       " '只',\n",
       " '又',\n",
       " '肥',\n",
       " '又',\n",
       " '大',\n",
       " '的',\n",
       " '鹿',\n",
       " '。',\n",
       " '那文士',\n",
       " '提笔',\n",
       " '醮',\n",
       " '上',\n",
       " '了',\n",
       " '墨',\n",
       " '，',\n",
       " '在',\n",
       " '纸',\n",
       " '上',\n",
       " '写',\n",
       " '了',\n",
       " '个',\n",
       " '“',\n",
       " '鹿',\n",
       " '”',\n",
       " '字',\n",
       " '，',\n",
       " '说道',\n",
       " '：',\n",
       " '“',\n",
       " '鹿',\n",
       " '这种',\n",
       " '野兽',\n",
       " '，',\n",
       " '虽',\n",
       " '是',\n",
       " '庞然大物',\n",
       " '，',\n",
       " '性子',\n",
       " '却',\n",
       " '极为',\n",
       " '平和',\n",
       " '，',\n",
       " '只',\n",
       " '吃',\n",
       " '青草',\n",
       " '和',\n",
       " '树叶',\n",
       " '，',\n",
       " '从来不',\n",
       " '伤害',\n",
       " '别的',\n",
       " '野兽',\n",
       " '。',\n",
       " '凶猛',\n",
       " '的',\n",
       " '野兽',\n",
       " '要伤',\n",
       " '它',\n",
       " '吃',\n",
       " '它',\n",
       " '，',\n",
       " '它',\n",
       " '只有',\n",
       " '逃跑',\n",
       " '倘若',\n",
       " '逃不了',\n",
       " '，',\n",
       " '那',\n",
       " '只有',\n",
       " '给',\n",
       " '人家',\n",
       " '吃力',\n",
       " '。',\n",
       " '”',\n",
       " '又',\n",
       " '写',\n",
       " '了',\n",
       " '“',\n",
       " '逐鹿',\n",
       " '”',\n",
       " '两字',\n",
       " '，',\n",
       " '说道',\n",
       " '：',\n",
       " '“',\n",
       " '因此',\n",
       " '古人',\n",
       " '常常',\n",
       " '拿',\n",
       " '鹿来',\n",
       " '比喻',\n",
       " '天下',\n",
       " '。',\n",
       " '世上',\n",
       " '百姓',\n",
       " '都',\n",
       " '温顺',\n",
       " '善良',\n",
       " '，',\n",
       " '只有',\n",
       " '给',\n",
       " '人',\n",
       " '欺压',\n",
       " '残害',\n",
       " '的',\n",
       " '份儿',\n",
       " '。',\n",
       " '《',\n",
       " '汉书',\n",
       " '》',\n",
       " '上',\n",
       " '说',\n",
       " '：',\n",
       " '“',\n",
       " '秦失其鹿',\n",
       " '，',\n",
       " '天下',\n",
       " '共逐',\n",
       " '之',\n",
       " '。',\n",
       " '”',\n",
       " '那就是说',\n",
       " '，',\n",
       " '秦朝',\n",
       " '失',\n",
       " '了',\n",
       " '天下',\n",
       " '，',\n",
       " '群雄',\n",
       " '并',\n",
       " '起',\n",
       " '，',\n",
       " '大家',\n",
       " '争夺',\n",
       " '，',\n",
       " '最后',\n",
       " '汉高祖',\n",
       " '打败',\n",
       " '了',\n",
       " '楚霸王',\n",
       " '，',\n",
       " '就',\n",
       " '得',\n",
       " '了',\n",
       " '这',\n",
       " '只',\n",
       " '又',\n",
       " '肥',\n",
       " '又',\n",
       " '大',\n",
       " '的',\n",
       " '鹿',\n",
       " '。',\n",
       " '”',\n",
       " '那',\n",
       " '小孩',\n",
       " '点头',\n",
       " '道',\n",
       " '：',\n",
       " '“',\n",
       " '我',\n",
       " '明白',\n",
       " '了',\n",
       " '。',\n",
       " '小说',\n",
       " '书上',\n",
       " '说',\n",
       " '“',\n",
       " '逐鹿中原',\n",
       " '”',\n",
       " '，',\n",
       " '就是',\n",
       " '大家',\n",
       " '争着',\n",
       " '要作',\n",
       " '皇帝',\n",
       " '的',\n",
       " '意思',\n",
       " '。',\n",
       " '”',\n",
       " '那文士',\n",
       " '甚',\n",
       " '是',\n",
       " '喜欢',\n",
       " '，',\n",
       " '点',\n",
       " '了',\n",
       " '点头',\n",
       " '，',\n",
       " '在',\n",
       " '纸',\n",
       " '上',\n",
       " '画',\n",
       " '了',\n",
       " '一只',\n",
       " '鼎',\n",
       " '的',\n",
       " '图形',\n",
       " '，',\n",
       " '道',\n",
       " '：',\n",
       " '“',\n",
       " '古人',\n",
       " '煮食',\n",
       " '，',\n",
       " '不用',\n",
       " '灶头',\n",
       " '锅子',\n",
       " '，',\n",
       " '用',\n",
       " '这样',\n",
       " '三只',\n",
       " '脚',\n",
       " '的',\n",
       " '鼎',\n",
       " '，',\n",
       " '下面',\n",
       " '烧柴',\n",
       " '，',\n",
       " '捉到',\n",
       " '了',\n",
       " '鹿',\n",
       " '，',\n",
       " '就',\n",
       " '在',\n",
       " '鼎里',\n",
       " '煮',\n",
       " '来',\n",
       " '吃',\n",
       " '。',\n",
       " '皇帝',\n",
       " '和',\n",
       " '大官',\n",
       " '都',\n",
       " '很',\n",
       " '残忍',\n",
       " '，',\n",
       " '心里',\n",
       " '不',\n",
       " '喜欢',\n",
       " '谁',\n",
       " '，',\n",
       " '就',\n",
       " '说',\n",
       " '他',\n",
       " '犯了罪',\n",
       " '，',\n",
       " '把',\n",
       " '他',\n",
       " '放在',\n",
       " '鼎里',\n",
       " '活活',\n",
       " '煮熟',\n",
       " '。',\n",
       " '《',\n",
       " '史记',\n",
       " '》',\n",
       " '中',\n",
       " '记载',\n",
       " '蔺相如',\n",
       " '对',\n",
       " '秦王',\n",
       " '说',\n",
       " '：',\n",
       " '“',\n",
       " '臣知',\n",
       " '欺',\n",
       " '大王',\n",
       " '之罪',\n",
       " '当',\n",
       " '诛',\n",
       " '也',\n",
       " '，',\n",
       " '臣',\n",
       " '请',\n",
       " '就',\n",
       " '鼎锅',\n",
       " '。',\n",
       " '”',\n",
       " '就是说',\n",
       " '：',\n",
       " '“',\n",
       " '我',\n",
       " '该死',\n",
       " '，',\n",
       " '将',\n",
       " '我',\n",
       " '在',\n",
       " '鼎里',\n",
       " '烧死',\n",
       " '了',\n",
       " '罢',\n",
       " '！',\n",
       " '”',\n",
       " '那',\n",
       " '小孩',\n",
       " '道',\n",
       " '：',\n",
       " '“',\n",
       " '小说',\n",
       " '书上',\n",
       " '又',\n",
       " '常',\n",
       " '说',\n",
       " '‘',\n",
       " '问鼎中原',\n",
       " '’',\n",
       " '，',\n",
       " '这',\n",
       " '跟',\n",
       " '‘',\n",
       " '逐鹿中原',\n",
       " '’',\n",
       " '好象',\n",
       " '意思',\n",
       " '差不多',\n",
       " '”',\n",
       " '。',\n",
       " '那文士',\n",
       " '道',\n",
       " '：',\n",
       " '“',\n",
       " '不错',\n",
       " '。',\n",
       " '夏禹',\n",
       " '王收',\n",
       " '九州',\n",
       " '之金',\n",
       " '，',\n",
       " '铸',\n",
       " '了',\n",
       " '九大',\n",
       " '鼎',\n",
       " '。',\n",
       " '当时',\n",
       " '的',\n",
       " '所谓',\n",
       " '“',\n",
       " '金',\n",
       " '”',\n",
       " '其实',\n",
       " '是',\n",
       " '铜',\n",
       " '。',\n",
       " '每',\n",
       " '一口',\n",
       " '鼎上',\n",
       " '铸',\n",
       " '了',\n",
       " '九州',\n",
       " '的',\n",
       " '名字',\n",
       " '和',\n",
       " '山川',\n",
       " '图形',\n",
       " '，',\n",
       " '后世',\n",
       " '为',\n",
       " '天下',\n",
       " '之主',\n",
       " '的',\n",
       " '，',\n",
       " '便',\n",
       " '保有',\n",
       " '九鼎',\n",
       " '。',\n",
       " '《',\n",
       " '左传',\n",
       " '》',\n",
       " '上',\n",
       " '说',\n",
       " '：',\n",
       " '“',\n",
       " '楚子',\n",
       " '观兵',\n",
       " '于',\n",
       " '周疆',\n",
       " '。',\n",
       " '定王',\n",
       " '使',\n",
       " '王孙',\n",
       " '满劳楚子',\n",
       " '。',\n",
       " '楚子',\n",
       " '只是',\n",
       " '楚国',\n",
       " '的',\n",
       " '诸侯',\n",
       " '，',\n",
       " '他',\n",
       " '问鼎',\n",
       " '的',\n",
       " '轻重',\n",
       " '大小',\n",
       " '，',\n",
       " '便是',\n",
       " '心存',\n",
       " '不轨',\n",
       " '，',\n",
       " '想取',\n",
       " '周王',\n",
       " '之位',\n",
       " '而代',\n",
       " '之',\n",
       " '。',\n",
       " '”',\n",
       " '那',\n",
       " '小孩',\n",
       " '道',\n",
       " '：',\n",
       " '“',\n",
       " '所以',\n",
       " '”',\n",
       " '问鼎',\n",
       " '“',\n",
       " '，',\n",
       " '”',\n",
       " '逐鹿',\n",
       " '“',\n",
       " '便是',\n",
       " '想',\n",
       " '做',\n",
       " '皇帝',\n",
       " '。',\n",
       " '”',\n",
       " '未知',\n",
       " '鹿死谁手',\n",
       " '，',\n",
       " '就是',\n",
       " '不知',\n",
       " '那',\n",
       " '一个',\n",
       " '做成',\n",
       " '了',\n",
       " '皇帝',\n",
       " '。',\n",
       " '”',\n",
       " '那文士',\n",
       " '道',\n",
       " '：',\n",
       " '“',\n",
       " '正是',\n",
       " '。',\n",
       " '到',\n",
       " '得',\n",
       " '后来',\n",
       " '，',\n",
       " '问鼎',\n",
       " '，',\n",
       " '逐鹿',\n",
       " '，',\n",
       " '这',\n",
       " '四个',\n",
       " '字',\n",
       " '，',\n",
       " '也',\n",
       " '可',\n",
       " '借',\n",
       " '用于',\n",
       " '别处',\n",
       " '，',\n",
       " '但',\n",
       " '原来',\n",
       " '的',\n",
       " '出典',\n",
       " '，',\n",
       " '是',\n",
       " '专指',\n",
       " '做',\n",
       " '皇帝',\n",
       " '而言',\n",
       " '。',\n",
       " ...]"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_words = []\n",
    "for i in data_cut:\n",
    "    all_words.extend(i)\n",
    "all_words"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "tmp = pd.Series(all_words)\n",
    "num = tmp.value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "### 绘制词云"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {
    "collapsed": true,
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAUwAAAECCAYAAACVPO5DAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAIABJREFUeJzsnXVYVN0Tx79ICyIIYgdgFyZ2t/7sel+7sOO1Awu7AAu7u7EQEBApBVRaOqW7e3fn98fK4soCu3AJdT/Ps8/ee2LOLK6z554zZ0aCiCBGjBgxYsqmVnUrIEaMGDG/C2KDKUaMGDFCIjaYYsSIESMkYoMpRowYMUIiNphixIgRIyRigylGjBgxQiI2mGLEiBEjJGKDKUaMGDFCIjaYYsSIESMkYoMpRowYMUIiVd0K/EB8PlOMGDGVgQSTwsQzTDFiajBaRoZ895pGhrxXYnZ2NWn19yI2mGLE1GCmdugAx+/fefcNFBXxZu483Jo6DTqXLlajZn8nEjUkWlGNUIJJYjIyYBEchGc+3/AtIR4h6zciOCUZWir1yiVvbMNV0O7fBtKyUjhwfw0A4PiqG/D8FIimmupopd0cS/dMZfIjiKkBLHj+DPbh4dBp0hRbBgxA+/r1cczeHvrDhsEuPAyDWrSsbhVrOow+kteUNcw/DrfYGCzs2g37bW0AALbhYejfrHm55SkoyaODjhb6jO7CKwv1jcJdt8MV1lVMzeXW1GnFyvSHDQMAsbGsBsQzzCokMTsbarVrl6vv2IarAACTlw3D8v3TcXzVDWw4NQ9udn64vOcprn7cx6CmYsT8MTA6wxQbzCokITsL9WsrlKuvzfPPaN66IeQV5dBYo36x+o9v3dFvXNeKqihGSBxjQ9GnQUu8DPOGvKQ0xjZvV90qiRGM2GCKEVPdLLd7ikuDpsM7ORYTza8jZPbO6lZJjGDEbkVixFQ3lpEBiMhMRad6DatbFTFViHiGKUaMmAoz2fgOXqyeV91qCEL8SC5GjBhmGaWzn+/+ncsegW1+LTewdECbBmqY0KXGruGKH8nFVB6xmZkAuCdKkn46SZKQlYVFJs+rSy0xVcDRc3Nx9NxcdOjclFf2syHdc2wGX/vs/ALks1goYLOrTMfq5q83mESElf32IiY0HlvGHUVOZi7Gqy7FrmlG+G/EIcbHu2LjwrhMJtF9+QKaP47jqf5wgbr05TMUZGRwY4rYMf53wis2Dsuev8BJewcMvHgFHY3OlNq+u44muutoQqut4HXZAUPb4/JpS959ZEoa5vbphjPWHxnVuybz1xtMCQkJXPioj+TYNHg5BkBeUQ4NW9TH2IWDMGn58HLJ/BYZhy7bTyE+PRN5BSxeeXhiKk6bOcLON5RXNuzgFQBAVEp6xT4IQzSqUwfOy5bz7g0+OsI6JARyUlLQNDJE+7Nn4BUXV40a1gx6LDMqd98zz+zh/z2+3OPO1L8tVNuODdRRW1oazevWhf0KXXRv3LjEtlv3TeZdt9RS510XPoKzCtiwtfwG3XUjeXVtGqhBTkoKNxZOF/Vj/LaI1zB/YkzdxTBPu44t449BZ2QXXNd/CrOUa+WSFRibhNYNVQEAVt5BGNGpFTgcwqX3zlg5og+vnYGpPW7YfkEzVWW83DQfMlKSjHwWMYI56TsaAKAkrY6ZzY9BWaYx2FSAbFYaJCQARSm1Yn1mH7gL/4gEAICyojwyc/LAYnOKtdNspIon+vPL1KHQ2H69vKHUdv/bcQ1N1OqiVRNVvHcNQnxqZrE2MtJSUFWqDZ12zbBnwSi+ul3vrPDQwxP/DeiHie3bQU5aGuoK5fMDLomDpja45+wO3/2lf5ZqhNE1TBBRTXhVG6OVFpHVA0feNRHR9okn+O5FxT86oVhZxy2GNPvcAyIi6rzViDpuMSQH/zA6+uoDX5uMnDyRx/MOiC6Xnn8jxv4z6ITPKLoYOJtXlpYfR7E5ARSe5U6ZBcnF+qw+9axYWXddQ8rKzRdp7O66hrz37rqG9NbJlyLiU0ptv/uaWZly+646w5P9M1rHDSggIZG0jhvQ0MvXiIio1XEDkXQujSEnLpNXZCzl5BfQnpeWjMllGEZtlXiGiaKZZeH7/9R0wSpgY9OFJRg5u391qiYUM9Zdw6NTi1GrlgSi49PQWL1udatUY7GLvwaXpMfY3N6CV/Y56QnkJJUgJ6mI1nUE/3uX9QjeRK0uXh1eXGL97uvmICIcXDKWb4bZY5kRPhqvhax08bAOPZYZ8doAgJKCHNKzcvnaFM5SC9tWJSONruP6wmmw9AnC4v49qnRsERC7FTHFvWOvoFxfCeMXD8FY5SUwS+U+fq8esA/GDvuqVJf7777ihb03+nRsgV7tm2OgtiYsnP3Qt1NLSNaqBQV5GV7bnksModVEDY/2z4fph2+w+ugHlbq1kZtbgA8ugfj4aFOV6i4MN5874erjj+D8+L7VRB2F4WtAJJ7becLcxR+tmqihV7tmaNVEDTJSkkjJzMFH7zCM6NEaUwZ25vX59RFc0H3Hlg1we+dsvrF0VpyGy8X1fGU9lhlBRkoSn86v4ys/8dAGW/4ZyuyHFYL2e4xwdf5U9G/VosrHFhKxwfzTeGHnhVZN1XDivg2ICLd3z+HVzT94H+GxybA9t4ZXpn/dAqExybip9y+fnKDwBLRqUfyceU3g8EULNFZXRkhEIqSlaqF+vTpY8e+ASh2z3ywDyMtKQ1lJHvksNl5fXCGyjELjVl9ZESsm9sW43u0gIy2FHsuMYH92DWrLSgMAiABTJx/svcGduf5qHBupKuHNkSV8ZYVtPIKisfj4IwCA47m1kJPhn22WNbs9v2EaercvfySsitB+j1FNXr8ExGuYfyYfvUKJw+FeH7ljRTl5BdRjcdF6U4/FBhQSncQr+7lOWHotENznpa2XUP2nr7hMxy5Y8JXNWXedhs4yKrNvfj6L+s48Sbq77tOWYyZCjVdRJq+6TOsPPqW+M0+Sp38Uo7IFrWEWrk3+ev/r+mJpZYLWIn9tN2PvLd51eShgswWWt7h+jFpcP0bTTe+VS24NRbyGKaZ86Cw0RD2l2jA/s4KvrFWz+rh/oORjbSPnnMbWFaPQqW1j3HziBAvbb3h8XhdxSRm4eMcOHr6R2LB0OKaN7VYVH0Noxiwxhvm11eg3ywBLZvRDbEIa9FaOYUR24Qxz4NpzvDKHs2sg/2PG+SuPbTwwc6g2ry9Q+i55TFI6lp18gnYt1BGTlAHfcOFcuYRZx9S4cRwH+o6Esqw8/qfBPaHT8sbxYu3CFm0VaswajniGKab8uPlHEhGRd3AMERHdfONMRCXPPomIHr76Qm+sPemLZ3ixupV694mIqP/UE0yrWqPprmtI2SLskn8Li+WbRQozk+yua0jjt18l14DIYvXeocXlsUqYOf7MtW9fSPPGccplFdASy2fknRjLV9/i+jGhPxMT+KbEc8e9faSyhmDUVlW3ofwtDOadQBe+e//UOMoqyCP/tLhKHbfFuZPFyqIz0qnj5TPlknfP7AttMDKhp9buFVWNx2ePMMZk/U48+VC+v6G5ix/ds/zKqC5J6Vkite/z6Dzv2iYimHd9zuMTcRjQ51lQV2Jxcsk1/iA9C+rKK7eNXEwpuT7klnCYr71fSvxvYzDFj+SlQADaPj2AvuoacE+OhMfk7YjMSsX099egU78FhjZqjSkttCtl7JbGBgAABWlpfFm8EvJS0tAwNkBDxTrQVFbBvUkzypBQxD7vSZCABPZ2eoH43HDE5IbAJPIUpjfbjIS8CAxVn122EDFihICIjVehAzC+pRWcYjciPscFU7Xc8Dy4+HLNVqd/0U2tMRa164lGCkqYYnYbYfO2M62SOKdPVdH26QEETN8N7RdHoaGoivO+9hjbtAOka0licMNWSM/PE0rO9pMv4RcShzkTeiI3j4V5k3XK7CMBIHQ11/Umu6AA+HG/9p0pXgf6Cf0ZboXuwqZ2N1BHipt8TVW2Cc4HcV1SnkacRF3p+iIbzNdBflhr9QY2/y6BWUgAzn51gu/S9Wh75RTYxAGLw0HYis0iyRRT/bCIjTmf9NGxrib2dFyI4MwoaCk2wagPG0E/5jQXe26GlmKTEmVISEiCwIJUrdoo4GQBANLy/DFVyw0A8Dy4G+96qha3z/ZPZsgsyKsMY8k44hmmCHilRKOzCv953AmWl/F65LJS+/WbZcB3X14fxNTcXCjLyZWr7z7vSRjXeDl06o3j3e/r9LJcsvY4WOOOtxsODx6Ff9t3geYlAwQv3wSNiycBAJISElCSlYPbwtVlygpNSIZG/fJl0hRTNl+S36Lnj3/zA98mY1O726gtqVSs3U7PSxii3g1nAp5id8eFeBJhA4/UIBzqsgxHfO6in1pH6NTrgMHqv10aFLEf5u9Ev1kGaNZIBRExKVg7bzA8/KJg9zkIwO/rvO0UHYF/Xj2C45xlaFKn6D9ft5vGSMnNwZUxkzGyZasy5SRkZOFjYDgmde8AAOi43QjfjtZon77fBiP/RVjX5jIkJYp27fd7T+Rd7+n0Sig5o203QV5SBuMa9cWTCBtYDil/0JFqQmwwf0eemLnC0TUELp7hv62hZJqO242wc+JQzOnXFbcdXDF/QHeB7R4aW+GzjQ/2XV2KOsrly7r5t3E9ZCva1tGBddxtnnH8nGQKRel6ePL9iNAGszx4pDyDnKQS2ipxIxs5J95Ab7VFlTZeGYgN5u/IxA/H8WpIkV/bYqcLuN5nJThEqCXBrKvYn8bYlkWzTrOwqpnhxOREQlmmHnLYWagnU/z0VN/7FxGTmcFXFrZsS4XGDA9LxJL5lyAvL4OcnHx00W4Ow7M1Mu1DmVzwHwUNxQEY1GANMlkJUJdrW12qiDd9mKb7SiO4XtiA7LwC1JaVxskntohJTkdeAQuhMcno1bYZ9s0fVbYgAezyeAhlGQW8GrIV72I8MapRF3zPSkRQRixWf76GlPwsbOswCdoqNfYsbrUyu9ceTF48CAAQ5B0ldD9nr3Bce/EJl3f/wysjAsasvoDUjBwAwIIJOlg1s/jxzLWuRUdT60jXhYKkIvQ68Dt2/2osAaDl5RMVMpotWnJDy+Xk5AMAPD2+l1sWEzy7+xHT5vYrV9+Vbd/B2G8Evmd9xrI2rxnWrBph2k+pnK9q48BdS/oSEEFERIM2GPPKu60wJGvXQLr57nOFx/BMCaf4nDRyiPcjIqLe5nqUxy6gra53iYiol9mOCo9RUzjrO1Sk9lMaLKPk2FSa2nA5sVlsGiU3lxE9dOYa8L2IiL4Fx9BQ3bNERDR98/US+14KOkk7PVeVKr/FpeMCX6Xx4PstSsxLIJ+0ko+iDh94kIYPPMi7fvrIqVSZOofOCyx3Domg+dceU8c9RnTepnQZzvb+NH3oUbJ46UpERO6fQ+jA1oc0qvseWjzlNK2bf5nXdlT3PTR7bHH/4BqM2A+TSYZsuoD07FxISHBnINYnVuDWu8/4byp3VjN08wVk5xXA+ey6MiQJ5mtyCHrU0+Tdp+ZnoY60PCQlioLds4gNKYnKCRw8yHoL7Iaf4N2fDXgFNVkl5LLzMU9jOKPj5nOyIFOrfAFq/6e8CFuuLsfhecawyLlTYV3+t+4y1JQVUEdBDiGRiTA9WxRFfqn+A1zd+y96zzOE852N5ZLf8vIJgeWlzTBfRz9HH9UBUJOtDwmGnhT1X1njc3gUguOT4HuAu3TxOSwSWXn52P7MAv20msNw1vhSZbw388SwsV0AAPev2mL20sEAgNE99sLiqz4AICEuDXPHGfLux/TchwkzemH1ttJli8pmj1k4qf2ISZHiNUwxwmEW8wVHfB7BsJsutBQbY5vHNfilR/K1uaazAa3rlJy6oCrw/xKM9YP1ceT1NmwffxQWOXfgnhKGdkpNcCnQEndDHfB57GEQCP7pMdjv+Qz3B6zFvVAHzNGo3IhHpfGr0Szrcdw4yBAzms3G9dAL2N5OnxEd7ALCsOmxKTLz8lFXXg5OO1di+R0TeETEYnTH1hjaThND2mqWKedn41haWWWz2WMW6krXA4Gwp8NFJkSKDWZ5YLHYkPqL0j+kFWQhJDMW610v4ml/PUx3PAS74SfglRqGTFYO9L3vwXzIwepWkw97Exe07aEJ9eZqsIr1wsD67TDcmqujwyh97HB/AI/kcIxv0g2OCf5opqCGY92q95RSodFsUFsRznNXlktGSmIGVNTq4NYZSyz4KWcOU6y5/wrnZk8su2ENQDzDFI5KUeK/+VfQsVsL5OezMGV2HzRuJnaQZpJxM0/j2N5p2LTrMdTrK+Hu5aWMyt/r+QQLNQdjlv1puIwtyuB53OcVetbThHtKODa2Z/aRUBTu+3pgp/079GzYBE8nCm+4V3ydh4s9uMsOTh/88PLuRyzZOAZ5uQXo2J27+ffVMRBpyVnISM9B4+aqaKqhhkZN/9zv72aPWZjdfC20lfvAK80Fd8NPAwATxlMcrUhYPD6HEhHR+WNvKTaq5NwpYoqjcf9QmW0GjTsm8PpvQZiNnp8xjX5BRETXQoo2apIT0un8odcUHZFEozvs5Gt/zcCcTu+tmtihfzCM2qo/1q3I9OlnfP0UjA8WXlinNwFx0anVosegiSdg96pi/nlVzd4vFljevk/ZDf9iWl4+getjpmFY87LXBwsJzQqGZ6orFmsUPbqrqNXByp3/AwCYeR3E8kmnceklNy3F4o2jQZwa8QQo5gd/bF7y8dN7YY/BP1inNwEA0KCxstB9Jzlw00G4pfpiu6chACAyp/QArs/euJZYN2ii4B3V0hg/+6zIfZgiLCMZbCL0Njldajtb0yJHfAUF2cpWq0YRtmyLSMYSAFRl1dBFuTtMY0wE1kvUkuAZy5/L/gaGDzkM9o/UxYcPcmMcJMSn44ONL4YPOQx//5jqVI/HH2swmWCftzF2d1iJ2NxErPla+gbJ96hkRsZMSMxAbm4BZk/VwVePcEZkisrp/pMxsJEmnKesL7sxgMHjj+PtY+HalsRV+89ov8cI7fcYYYRhUS7478mpsPINgm9MfIXkVxXfsz4CABLzAhCWaYeg9KLslDObzgXAdS8Sw8/mreMhKck1R737cOMQfLDxhaYm95QVW0Ae+OpAbDB/onBmWUhdaUUoSMmjoZwaL7zVr0REpfCuB008wfcqZOaknkLrUEdRDnJy0lCtp4Ae2qKd/tF6eBjbXUyxw8UUMdnpAIDWj46g1cPDyGYVQOvhYV7b0972OO1tjzFmV4rJUZaRx1Lbx0KPO6BP6zLbuCYLnlUVUhixqJ9WC6wc0ptXLi0piX2vrZGdXyC0PpVBzyWGQrVrrsA9GaMm2wYtFQehldJoXl2tH763hRs+YrhkZeVBQkICN67Z4qNjIL58DgEAXLv6Ac1bcE8/NWmignFjRH9SYxymF0XL+fptGTjhOA2ccJwML1qS4UVL3v3ACcd59aLK23HwORERTVt0Qeh+7R4dpdfh36jrMwNq/7hoAyYoLZGIuImvejwvSlZ2P9CVDDw+kHsis8nBiIhO+IziXV8PXkr57GwyjTpGuawMMvQdJ1j/3Ya8l28MN23B/jfWvDqPiBihxh65mv/ky6N3rhQYkUDz9tyh0Kik8nwcmr3vDvVYbEAXXziW2dYntXCThkMF7JxyjVeZnN54l+/+8RluUrujy65WhzpVAaO2qroN5W9vMAsZMpl7XOy/XY8qZDDLy0FXS75714RI+p6ZQpoPDtG89/fJKS6MNB9wd74/xYXRWkfuf2z/1HiB8p4Eewg1LpvNob1HXvKVFRlMDj39rkdERHnsLLoVslLYjyMyX3y/05IDD4iI6Ksv96jrPztvUXpWLgVHJlJwZCJ98gwVWe6wddzjsoVZOhcduk8FrLJz55SXHqtPUVpWDi06+ahS5C/srkdj1JYTEdG+ucbkZOFJId8iyWj97UoZrwYgNpg1jaNnzGji3HNExG8wv0cmV5nBZBqN+4eEci06ePJNMZciDnHohM8oSszlJk0rNKBPv+8s1p8JHD1CeNe95hvQ/L13qfdCQ9p7yYx8QmKp/5LTNG0r9+x4ZnZeucZYduwxI7qWxWT9m9R1pSFtuvya8gtYjMmNCU+gmW02ERGR/ryiWbiThSfPgB7RvcLYeDUIRm3VH+24XhUUrlV269wcbl780WUUFWQxb0Yf/Du17JQUNYmtzm/gnhiNjIJcfJpcvjP01UlIVBI0m6jylWXl5OPWGxesmlF9RymFodsqIywc2RNDtVuhi2aj6lYHWfm+UJBph1/9vxOyTFFfQfChgesG5ugzrAM6dGteBRqWifikT02k0HDqzhuI/43qApW6tRnzwUwryEZd6aoNnJuSlwMVWfkqHbMmMcVxNSQlJLGx7SL0U+Um8CIQMgqyoCStWKwtAJj0N65yPSsL+7BWACQgI6kKFicd/Vv4wCVyINicLEjWqo08ViwAYGDLIL5+p3Y/x38HpmJi1z0oyGdBTl4GJl/3CRzjiq8zdNv3FljHIIwaTPEuOYPYvdqCeTP6QKUus8Ytk5WL26E2MI9xY1Ru6fz2v2HlZpsnNzcRm9gw8L+OLBY3fqYEJLDAZRss4z7y2hYay1oStTDt49qqV1YAsXFpjMiRrKUARdku6N/CBwCQx4qBZr0d0GlqB52mDsWMJQAM/R83i2qzH+5AMrLFz8bcDXRFYFoiiAgBqQk45mbDiL5VgdhgMoC3XxTsXm3BoIknMGmeMU5ftsaoGacYkR2ZnYQ3UV+QzcrHmEbFU5VWFiqyf28qiHWt52G4el9IStSCYdcdUJCSxxTH1SjgsDC20SCMbNCPZyh7qHSColRt6NTrgru9mXV7ic0NxT7vSQCAwz6zitXnsrN59T/zz4KL4FTwybGB4lT0a+6OjuqXeWUDWwahgeIMEAgpOR8E9nOx9QcAGD/n/nis+HGK6Wd2u1hgud0z3PD/gjGmV9G9ftMK6VqlML0oWs7Xb01JO+JMb/j0sdjGqLyfGWL95wQx/lN4FWVMe70m0vcsX3JLtiJDvyXF2uz1msh3n52dR0PHVt+5/pSkjGJlfh7fBba97f+1stUhYthWiWeYvxCZnYQ+FjvQx2JHsTqrqIBiZXqHX1SFWgCAT6OOVprsPqpt8SbaBR8TfTHSZleljSNGNDQUOuND/EM0km+FBRoHcCZgBW9WSQKWTe4/dsK/M3pj4/aHxeoueDnh0Bcb3PJzxbvvgWh56xgAYIvjW1hFBOGqz+cK66tcT7FYWdsuzQS2nddGcNK7Gg3TFricL0bZ5Tm53H1fR34hIqLdHg94ZSl52XTc/X2ZfQdOOE4cDv+97ceAcutSlaz8bEz9LbdQf8stNNBqGyXlpVe3SpXGW29/MrC0p3WPXvPKZly+T8EJSTTe+BZdc/xCM6/cr0YNy4+rezjNL8U9aI7FQzrl7sC7fxvGTZvS4uZRvvc/CEZtVXUbSsYNZqGxNIu+Thwqsl4c4lBGQdkh3vzTo2nChyPU23w7ERE9CHKjvV/MadOnl6T/9Z1Iupy/8YEGTzohUp/qIiq7fKdgmOKdTyAREbXdZ0hERGsevqq0sdruNaQ3nn50/J0dtd1ryCvfbmJBXQ+dpaMWtpU2dlXAZnMEls99x3WGz2OxyDWe/4TXH2goCxE/kpdGbck6OOwzD0PUZ8I7zZFXfjfsEBSllGHgt6zU/q8iPyM+Nw3DG3JznPyj1RWRWWl4EeaNPd1Fi4a9cuFgfHixWfQPUQ00lq/64LSTL95FO30jtNM3QvtGRalsLXwCYelXfAeWScZ2agMZSUmcnTUBO168g+5dE5i4f4PbzjXIystHu31Vk873V675OVdYRq0SIhzdGTkTACAjKYlu9fnTkoQt2Fbhcf8G/mg/TAIJTDaVXpAEJWlVAT2AbFYeakvJoo/FDjiNPlKsXvP+YViM18VFn08w6Fuzwv5/TgqFvKQ0Oik3hXvKd2x3ewzzYUUGe9KH03g5ZD26mu4GALiPPyBQztIzT9CmcX0oKchhxVhuXMzH9h6YOVAbwTFJ0Gok+G8nCjedXNGtaSMYWjvCOSwCEzq3AwC89vLjXctLS+OfXl3QoaF6hcf7XdB8cBgh/+4EANjFhOBekCsuDZxerK4izBh1Emmp2ZCvLYOXH7ZXWJ4g/nXagFw2N12wdC1pPO7LjNdIORBHXC+k/xTu4+4rS8/yiiiRy4H8Z7PZPxYnB788T9pPDGj464uMj1kRQjLiaYQVd3dU+80u+pwYQhf8rXn12m92kfabXZTLyqetro9omu3ZUuVprzGkpPQsIiLqu+ksrblgQvbfQsk/MoFiU4rvhIpC232G1HafIZ378Im8o2Op7T5DCohLpIC4RL7rgLhECkqo3qWCqkbQcdRb/hVP9fwz2Vl5lJdXUGL9kGOX6fZHV969uZc/ERENP8EN0KG997RQ41wKrpzz8CIiXsMs5MFLZr9IhfQx30GRWYl8ZYWBKzTulX2+WhCjas+j02uv05Yxh+n02qKc2G+uWNPaQXvp7Q2bcuv7K9eCbGmF801yTw7nKz/jV/Ya7CM7d3L4Fkraa7hrez3WnyIiomeO3B+l/luMKTopjTFdL9k78wwoUdEa5sgzJecN/5P51WB+igsjDhHpPD/F6DimJiW79My6UPKGl51/qMjBR55FWJRaz+JwyDkunK77udDneK4LUkBqPB13txFpnBIQG0wioltPnejOM+6LSRJy08k61ou2ud3hlRUaSyIiY29H8k8RHOGnNEbVnkf/q7eY1g/Rpy9WXkRE5PTWjUbVnkezWq7hM6Ll5WWEK72N8iCnhCDKZ7PoUZgzJeVlkl9aNGm/2UVERIs/XiXtN7vobogj+aRG0ZtId4GyBm49L7CciCg3v+TZiTDsfWPFM5J3Xdx5RpKI+IxnVTHl3B1qr2dIk87cpos2/N+n9npVq0tNoNNuIxpy7DJfWXs9Q9I5wI3clJ2Xz+jfJZdVQNFZaTTH+h4dcbOm2wFcTxXN+4con80NQBKbXe6nGrHBJCJKSsmkkO8JpbbpuciA2BwO7btqRj0XGdA/u2+RT2gs9VxkwNfO5msg73rAOz0ab3OIt0tORKT1k8G87udC1pGBdD/QlUTBaNU1mtsRMTQ4AAAgAElEQVT2P1qsvZUK8rlfgpPLr1CYTyTtnWFEo2rPK7Fvp5d7yCLqG72OcKexVsLPNFgcNvUzP1Bi/WpnZkN6jVZZyvcuDHksFs9ApuXk8gxmVl4+o7qVxsaHptRez5BYbDbZ+YfSPxeLXMr+RoNZHWjcP0TdnxX9rQNS44WKliUEjNqqP3rTBwBCo5Ow75oFdi4YgbbNuZsHE7dcxasTRSlhBUW3+XXTR+vhYQT/w11wn2l5B49HzqsslYux9OMt+KRFY0fncfBIjsCuLsWPm1U3Dq9dcXDBBRjb7oFWZ8GOyoWEJCbD6L0jMnLz4BQaga5NG8E9Mgbv1i1CcxVu7qV2+kW71BIAfPduqDTdD75+j/vOHlBTVEBSVha+HdiAzntOo6WaCprXqwsbP24EcJ+DladDdZCTOA4StdRRS6o1ZJS2AT9yImbFtAAgCYVGIdWqH0OIN31EITqRu97Wc5EBb2Y5YfMV0llS9Gt2+cVHkWTqPDtFi2weCt1+2YLijsTOHwMFtCyOd0oU+aYWRRvf+uWJ0OMKQ1q2FWOyRqsspRBvbvDeGweel9n+rE3pf3e9l+/orrNbuXQp6bF+v6m1wHJReBe5mm4H9CbT74voWcikCsurLrJiOggsz4xuTpnRzatYm0pD/EhenWjcO0Qa9w6RYNfg4li/86al8y5RUEAsX/nMicwu4ouKR3gLikk5Ru7hTcknqn+16lJevlh7Cywfb3yL2u4zpI1PTanzgaIdXTaHQ/tNrWmQwWWB/UTldkBv3nVIujllsxLp7ffFjMguDTabQ08/eRGbw6FOGyq+ZJAV24PvnsOKouyE6v0h6LjZkDy/x1BSRlZFRTFqq2qk4/r57Q9gctEKJheteGU3D73AxKarcWbj3Uof/3m4e4l1y9r3QTtldWjdP1xim58ZNrIjBg5ph6SkTL7yqTN1cHBP9WYPbKi8FdKSDdC+sUO16lEaUcFF6Y0X9dDjXT89ZwFpWSl4fSx+vv/hkn8AAHpjh0JOuii82PL7L5CZl4/MvPxy6xOWYQWflPt4/X0OlGRawC/1MZLy/KBRZzSehoyHtuoyvA6fU275pWH0xh6dNxpBe/MpTOvTCd22lJ4GuTRyk+eAne8EAKjd4AuyYlqAlWuGvLQtkJBsDJk6G5AV0wL5abuZUr9EUrNzAQDfIuPQaQt3Kcb7xAbctP2KbQ/M4RtVczKG1sg1zJjwBCREctPWdunfFgAwv9sOxEck4Z7XcczpvBUvI85BVl6GMQXaPd8PAKgjLYuMgjw4/W8LlGUEB9BdbvcUlwZNF3kMJ8dA9OnPn2Hxi3MIevYWLr/1FMctyGXnY6XWNExsMkjk8X+GiAUJCa4x8YseDCX54WissqdCMiuLMWrLoKxWBw/9DPjK1JvWg2Ld2jhvW7Lelx1cMKJdK2iqcU8ypeXkoq68XKXrzCSdN/KfOvIy3MBX3lxNGY1UlBCVnIbIpDS+Nr8rrqFR6K7RBADQc+dZ5Baw4H2iXJ/pz464PkZtGeqoKOBJoBHv3jzxMqZr/YfFe6di3HyuoRhbfznMEi4xpkChwWxSWxkjG7fDzSAn+E2tegNyI/Q1rOM+o66MIiQlJCEvKYNjXbixBV9F28M85iPO96iZx9hYbA6kJGuh/5LTcLzGn6d8/cnncPYOh9NN0b70/q6h2Dv7HFITM7DswExMXTkCY9S4x1slJCRQ+P01MtuO9r2E++H5XTn83AYPHNxxXncyVl15wWc4fzaQv97/5Yg3fSoLl4QwCk7nuioNNRO8xjjb6h4FpJbuzlRZzHHaTURES1xKdhWqLvILWLT9LDf6z7CVxnx1uaWcKvldiE+s/uhN1p5BfGuWhde/vu99JFqQmD+cP38Ns7ropdYCmnW4iePfj1kvsM3VwTPglxqPfA67KlUDANztzZ0FX+1V8+JVvnPyh2dgNE7eeQ/r86uw56IZry4qIQ3ewTGw+OSHqy+cqlHLknln68N3/9TUFQAwcMpJbD9sAuObttWhFh8NlIvHmvwZORkpdN5ohH0zRQsSU1G2nHqJDSdN0Hu+ocD6wujvialZuPjUEWce2mHUqvNVqSJj/FUGk83hVFiGV3IMtJRUcS/QlQGN/hzGD+gA09PLsHneMAD8f2vNJqropNUIo/u2w9LJfSo0TkiWT9mNysH3qGTcfPwJA6ecRGZWHqaP746s7DzYm2zGrvXjsG/T//DPyqulyuiyyQipWdwNjML1xV/XHyvCz6tngh67Px/lLt1MOHqTsTHLYt8lc8jKSOGjZyhG9mmLV7beAABLJ3/0WWAIw3sfcPo+98cmLikdHoHRuPf2C96dX1VlOjJJjVvDZIIBa85BXkYayRnZ+HqF+6XqoWuEOrVlUVtOBnHJGbxyUbkX6Ip3kf64NfRfJlX+a+EQG5msdDyPvIyFGvxrsxeD9yKLlY5NbblG53X0LUxovADbPGbiUJd7kJKQFmksNocgWULoM4A7m2zWWAUAUE9ZAecO/YOBU7gJ0exNhA/Tt/2eGUy/+vHun22ehzaN1Xj3ObkfEJP4L2Sk2yG/wA/1VQxQR2E2ACAqbiQUak9GctpBaDaN4ZNr6xOCNVdfQkVRHnb7V/DKBa1h2ugvg1odhVL1tHvthiOrbgIApGWk8CrYoNT2giAC+iwwhPPtjRi/7hJMzyxHZnYeAiMSsOLQY167g6vHY1A3LcjKSGHif1egrCSP2/vnijxeOfizN30qylf/SBg+tsW93fyuHT10jVBPqTayc/ORm88qt8EsxDoqEMObtC61TXZBATqfPYvgjRsrNNafDoFg6L8Jm9ryP9K5pzqiq3J/3v1Bn2XI5+RCrlZtbGhrAHnJ0g1CIeGJqYhKTkNoQgoa1FXE+juvYb97OeopMpfo7R+j+9g6aTDaNFaDopwsr/yhowf+6a9drH1IZCOoKG2GhIQ0MrNfoWmDIhe6tMzLkJFuB3lZfk8Iozf2uP7+S7FdcqD4rnh8eibUlUp/hB/bjH/Z6eijNdDuV/p3+jfkz9/08UuJp1xWAeWyKrZZ8NrxG43YWBSGzcErtELySkL73DnSNDAgTQMD0jUxISIiTQMD+hYfT5oG3NNF179WScKnEmm/s7iDs3NIBF3+4ELtdxoKrK8KLgXt411fDNpLR3xW89VvcZ9OBn4biYjINdmetrhPL9c43pGxdMDEmg6YVPykjzCceuNQan1wRENKSt1PbHYqJabs4pWFRXWg4IiGvJeYCsOoraqRM8wL3z6hg0oDDG5c891EfOLjMeFukTN94Wxywt278InnOtzKSkkhj8XClA4dcHLMGMbGfhbujinNtVFLouwfUYfAcCjXloOphx+2jRsMAAhPSkULVWV00OPOVHwO/dmuKHufWUJ/WuVtiAzYdQEa6vVwZ90s3mPyc2dvTO3dCYP3XILt/uUVHmPC0vNITs1G+1YNkZvHwl2jhRg9/yx2rh6DS/ftMXFEF4wc0B6qKkWz7wFWW7G53VSc9HsObWUNeKSGwmHE8Qrr8pvw588wW949TH2elx7gtiZSOMvUNDCgQVeu8M0wC9+ZwCTcg6yj/antM326E+RCRERDzU7Trq+vqc0zfer+6ih5JEfy2q+685I6/BR1p3A2ae8fSkREg45cou77KvfvbX7XgQLcw2jtCP4INKNVdWl8wxVk9/JLpY5flcw+VRRPcvMtU+q0wZAc/cIYke3izpXTb1rxXFEm79wFlhfS33ILzXA4wogevxF/h1tRbHYGHGPDqlsNkQneuBHBGzeiV9Oi5PT6798zOoZjfDDuBrtgrpYO5mr1AgBEZaeip1pzAMDXCdt4rhwAYDx3Ir79FGmncCY5oE1LAIDt9mX4uncNozr+yuXdj/HM2BK168hhxUB9AMD+BRdw2+0IVNTrYuDEHpU6flVyb33RhuCJ+ePgZbgB/dq24JXF5vrhXqguviY/xvesL3x9s1nJSCvg3+wpxMM3Er20uXKaNVJB/+knkZicCRabg/7TT+LpWzcAQP/pJ0vU7XH/yklJ8TMdT5/Bf6ZvoWdpySvzTUiAX0IC7MPCkZyTU2r/Oz7cz5FVUP4jrJUG0xa4nC+hee7EDbgw0+AeHTP5UGK7zi/306h3p6i9yV5aaH+TNn0uivIz3MJIlCGF5udZZOG1WQDzaXato/14122e6fOu2z3fT0REvV8zl6lyZC99mjXWgEb20qfI7+VLF2H5kBuVKDc7j0ar6tJoVV3icDg0WlWXiIj+abeJRqvq0oLuOxjTuyaTlBtGmQVJlMvOpFO+w/nq8tnZ9D1TtFirNQ3Nkwa817e4omDbiVlZZBkYVGrf+W+fkEd8DFmHB9OZr/zRrOKzMikhO1NUdf78NUwAcIwNg6ykFHrWbyqoPTpvNMLK0X2wanRfgfV73V/jSdhX+EzeB8NvVnBJDENsThrkpWRQV1oePmkxYHHY8Jm8j7EPoWVoyFvD1DIs2vEd3bo1Wquqwi8hAVbBwTCeMAFjWvPvRton2sMn3QfLNSu+zsUk8yefwcjx2oiOTIZOv9YYOrpTdav023MlcDoka3F30jMK4rC+XdEOeQEnF9K1av5Z9z2OVtjff0SJ9YOvXEVzZWXcmVEUcyEiLQ3N6tZFck4O6skLjtMAAG2uGWJK6w7o1bApprfhft+Wv3sBw6HjoCAtcvwIRtcwpcpuUvXcC3TDLhdzWE0onhKXiOsgDACrRvdFP73zUFGQh+nORXzt9LtOgH7XCQCAq4EOfIaxw4t9aKPUAC+Gray0z/Cz4bQIDIRFYCCv7ldjCQAhmSGIzY2tNH3Ki6ycNPoNbosA3xgc2f0cR/c8h4Vz5Z+xn+SwBrK1ZFBALJj0P1Pp41Uluq2fllj3OxhLALjt44a3oQFIzMlCmO4WTHl5DyaTilz5bHW5Abon3LmD1/O4wbbDU1PRrG7dUo0lAOSz2bCPDMexQWPQ8vIJXBg5CUqyclCQloFdZCjmv32KsGVbKu/DlUKNnGH+a3UPPinxmN+mBzZpF/midd5ohHqKtWG7fzmfs26hP1pNDzjwITQUQzQ0BNbdDb8L/wx/HOgkOPXt30RqQQbYHDbkJGVxNvAetrdfWnanamDewEPIycrDU9f91a1KldPyygmMaK4Fq+/BuDBiEjbbmuHbQsHHiauZP3+G2VZZHZGZaTjn7chnMEsyiDXdUMYmZ6BhvTolGksACMsOQ2pBKmJzY9FQrmGZMq/uN8HSPVP4ynKy8hAdGg+tTqWniPiVvZ6PYBFTFAN0c/uJmN5c8FJHaWj/Z4RJOh0Rn56JAhYb19bMwAkTW2Tk5OGlyzf8M0AbO6YPK1OOsnQdLP+yD6kF6bzc1kwQkpjMC/NWERYNPYobNtuholYH1yy3Ij01G0rKzDnB12RaXjkBqxmLMbBpS1h9DwYAjNVog5VWL6tZs6qhRs4wC3kXEYBRzdpUtS4i00OXO8Pt2LIBAqOS8Ok890zv4HXnkZmThw4tG+CO3uxSZRAIEpDALu9dONjpYKltVw0/jFCfKLTt3hJj5/TH6Nn9MLvLdqQlZ2H8/IHcNodnCq2/RYw7kvMzcdrPFF2UW0BZRgHHuxXPWcRmBaCWZPMfcTSL/9b23HQGXwzWQf+hJbZOHQJ5GWnMPHEXWbn5KGCz0VC5Dm7/949QOr2JtsX/Gg8W+jOURYdDp+Gjtx5tDxjBfzf/D+xs00e4P34WWl45gReT5qKreiMAwIsgH2ywMQUBCNPlfwQM9IpERnoO9BZegVlg+XwaNU8bImT9RsRlZSIiLQ09G3PjP7Y/dwZs4oDF4SBkfc08JaZ9+yzS8rjn5n/929Qw/nw/TFEYMJm7IzxjOTNpB8qDuUvRrvUnb66f3JjNl2ng2nNkYudF68+8KFNGAYd7qmmhy8Iy2943MuOO0XAVrywxJpWuH3pBYxquom3TTxfrk51TchbG3ubbyTv1O9nEepNLYiBtcS2eTTI1uhWlRDXlvdLi+NNajNjD/ft3Wc/18czMyePVFablLawThon2q8tuJAIXHZzpjosb3XEpniMoNC2ZiIhmmz6iFpeP88qjMrj5oO76uNF2O/MSZe9fdatcOt125+pi+NGRul4oColnGRxErjHR5ZJZU+g3/aTA8shMB3oUPIq8km7Qk5DxFJBqUu4xnr/mPz03Za4xDRp3jAaNO/ZzMaO2qroNZYUN5rUHjjRhgXHZDSuRoMhEIiI6dv89ERHNO8R1XD7zzJ56LjOiHrplG4rHEY+FMpa/MqH5OrJ64kwn192mdw8/0aU9T2lMw1V0Ym3Rf+LImBTeD0v54Rq97FQ9SolqSrmZzKborWxuu7hRm/2G9MrLt1jdRJPb3Pw4N08Tm8OhR36eRESkfesMrbJ6WWk6LTR5RsYuTrTA5BlpnDIgjVMGdPKjAyVmZZFbTDRpnDIgncsXy5TzxSuciIhCIhIrTVdRuWPiLLA8MdeXlwvpdfjccsn+6BJEbp7hdOvBR4qOTeXKMvegh89c6PItW7p535He2/H+ncUGs5ADp0zJJyCaBkw+QWGR5fMRdIj3K7tRJfMt7ZtIxnKpywk67f+Mpjrspo1uwv1YVNRgsllxRESUnaZPKVEtKyRLjHBMW3iB7z4hKYPYbA75BsTwlUf9MBrC4JkSQdpvuGfXv2cV/Z8562dJ6fk5xOawK6CxcNwJ7EdE/EnkRGXQuGO0YuMd2qH/rKymjNqqGr2GWRosNgfrdz/CuUP/QkICyMtnYcSsUyKF4QIAHfOd2NlxCiY36yWqCoxyPvg8VmmtQkp+ClRkVMpsv8/7JtTllLFIYyzkJWVLbbtJ/ylc3MPQWkMd1w3nM6VypfDV1g+75l0Uqc/VD3qICI6D9bPP0Lu4qOwONZzBE07A5uVmjJxmiAb16yIqJgW2r7cgKiYFGZl5ML5mg7NHi04TZWTlYvQiY9699e11kJcTHPqu+9s9uKCzAACw3Pkm3McfAIGw6ONVyEpK4XNSKFzH/R67/iFhCShgsdG2VdEm6flrNli1ZOjPzf6eNcyDlh/o2Hs73n1qTg71O3OZIlPTyvpVKZEptvwzrV5mO6iXWc05YfI6+nWZbYa938B7t4ot+wz2vefONHr2mQrrVhWMabae9/JzCxep/Zhm6yk+KqUKtBQNn7SyP8ev+P2YRX52C+WVJSVzT7kYX7Mpty7dTffwrgtnmpkFucTmsGmn2xPqZrqbviaFCS2vm6ke33tFmL32GvWfeoL6Tz1B4xacq7C8H/wdj+TnHZ1p1u2HVMAuekQwsnUkx9BwanW4/KHIRDGYbyw9aechExo44bjA+t+JFdvuVbcKQmH7SrRjgT8by5c37MruUIVsc79C42x30qWgNzTEeqPI/e89caoErZhlgAX3OG43Uz2yifWpsLxRc07TtUeOFZbzE4zaqhrphxmalIJx7dtgXPs2uOr8BSv66uDSp8+wDw3Hl4goAMAZ+09YN7BsX8He5nqgn574m9bm+uHpmO/EJR1dAIDLGME5xrt2aoZjZ8wBAJ++hKBvT9HDzcVmZqChYh0AQCtjI3RWb4CUnBwQCLbzq84h+8LR2Rg45aTISxZVzaAJ3Uqsy8nMw9QO26C7ezKuHHiBwRO7w+z7qSrUTjQaytWDc5IvRjXsgQfhogdgmT29dyVoxSz15ZTgnBgM13Glu8IJw+6Tr2Bxdx0AIDevAClp2WikXrfCcpmkRkYr0lBVQQsVZbRQUca8Hl0BAMv79sLFaRNxd84MbBs2SChjCQBTm+vgv3bjePdZrCJH6M1udwV14fHvsiu869oi5EA/7fIJw+5cBwCesQQAr+VrsKBLNyTlZKOBQh2+Pk5fQzFo4gmhxxCF6w8/Yv2ex0LFzSwPr0O6VIrcX5FXlMWgCd1w5cALPHI/hO3nauZ6bAGHhVkfDwIgbGw7HS0VuGtsh33uV69ilcDzQevRW02LEVkHNk/kXcvJStc4YwnUUIP5MwoyRYaqviI3KOrS3sKHAmsgVxezWw6AhqI6ACAlPxOvIrkhtTIKcjC7Zf8S+9q92sJ7aXcUHAREEL2bNEVqbi4mP77HK0vNzcV+uw+wCQtBPXl5nBo1lq/P9gPP8PT6il9FMcKNRx/h6vUd/+kOrxT5VckOY+6GxayuetWsSclI15LCo367kMPOw5hGRZuJOzuUfnjhb2Ou5UPsdDKHzpNz6HifPz1Jy9tHq0mr0qmRj+RMoi7H/ZWqI1UU1OCg93Pe9X/txjM+Zp8mzeCqy58VT1lODkeGCY72XTiznL74IuxeMX9q4uC2STh46i0jWTN/5XVIF0zQ9OS9iylix08G0mZYUYKxfXbvod2gIXwTE7Cz/2Dc9nLHTQ9XSEtKIi4zE+66q6tD3SrHKfY7PsWG49TAidjxybzM9mMWGyM9MxcK8jKwvLUW/WYa4OPjTVWgaRE1foZZUaKzUwAA8bnpvDKXMYdLXLfs+/giWl4/jn/MHlaJfgBw7si/fO9MM7hPa1g+XI/TV5kLZPw6pAufkSw0mlX1eP474xQVAalataDbrScy8/Mxv3NXhKelIiApkXfckAmsYq/hecQROCRU3XdZFILmbcWD0dwflTbK3Kyae5zf4WmwFwDgvNcn2EQF89qbX+f+kAzs1Qr9Zoqe4ZIJfls/TGHZ5/kEn5OCkZDHbzAB4GrQeyxtxR8MwinmO3Y7WcFyyuLKUqlE0tJzUFep9NBXTNHSuPgXLmw1s7/W1iZfod5YGQ5mnli5jxsoJCM1G0fX30VMeBKuf9ghssz83AJMarMFdVUV8dCt4hsNsxdegmbL+migroS1q0qO78gUU57cRxf1BgAALZV6eBHgi+fTuUZD45wBQtdsQgGbDWlJSZHkWj37ghHTevKVnfCdgf71Z4JNLNjF38PwBovRR21qhT+DZ0wcujRqADYRep6+ALf/qi7HeL+ZBlBVVoDurH44eskS00Z3xaYlpS41idPsioKO+U6++6a16+H5oM3QMd+JXqpaMO61hK/eIzEG097cR9BCZo3HPu9JGNd4OXTqjcMhnxnY0u4OAjK+oFPdAYyOU5OIj0pBQkwqTmx8gJt2OzFWczMO3VoGV3t/tOrcFMfW34NZiOB0CmOb/ydwB3xs8/8AgLHd8SmzzkJKShKH9adh1frbsDSt0YEkBHJhnwle3XZE/UbKyMstwKOv+nBJegXnpOdgcfKxUNMAyfnR0FJkJg2I7tMX2D50EKbcvg8OhzBQowUuTJ1YrF0uOx9ykiIH/GUascGsCL3N9TCkQQcc6zYHOuY7MbFpT+zqVPSre9vXFeHpqeii1hCTtDoIJTOPUwDZWoJPVgBALjsLR325s4h9nV7iZqgexjVejixWKjQU/rxH2LycAuiOOAY2m4M7jrtg9tAJ4+f0AwA4mHvi0KrbJRrKn5nTcw+S49OLlVeWK9HQ0cdgY7FNqLbLTj5BQGQC8vJZqK+siKjENHy9zI2CxOZwQARISVbNitetk2YID4rDoPHaGDROG7WqYNxHHl546vUN9/+dUWw2bB7riDfRtvin+VgMUCvZTayKEBvMirDx6204JPjBZcxhBGXEYrbjGYHrmQEpiWijolamvP6WRf/BHvTbjOYK9Utsu897EiY3WQf7hKfIZKUgj5MD6VoyUJdtAV2tsg1IWQwedxy2b7cKrPvX9ga+pUQjn8MGAPhNFS1q+uJTj3H9P/6QcYYmdpCRksTq//VHJXks8Qjw+I422s0rdxAR+OIfAXvPUNh7hiA8LgWdNRvh5vZ/sOn8Kxismoi7ll+hqqSAsb3blUs+mziQlBDN8G2cfha+ruFC/RgxwfvgEPRv0RyyUty942eRVpjWlLusYR7riLrSiuirql0lupQCo9/MGrfpYx3nDNNoO7yIKtqg8EkPxhSH/6D/7SIm2K+tkHzDHvN5BrJVHa5/XDYrr1i7X41lJqtoMf5ZxCdcDCra1RvXuCdO9ViKpxEfyxy/q8pwtFXSwZb2dwAAeh2eMGIsZy64iHoqCti86zHcPL8Xq38weBFeDl+BXmotBPQunbD4FLgGR6HrWiNeWde1RnAPjYa1ZxA2XH1VId2FoaqNZWDCanAoF57RowXWN6qnhM6aDaE3dwSWju+NwMgEAMAHd+4mRVpWLh7ZuAvsK4jwrCTe9c2gj0gvEH3zx/Dp2nIbS1YB94f0fw3507Zc3fMUExoJTuUyTEuTZywB8Izl5eCnGNOwf00wloxTI92Kuqu0x82wl5jchLshE54VgwOd1+BZpFUZPctHbanSg1cAgOJPbknTmnGd5l2SAmA97CDkJLmP473qFc/VIwjXlHeIzw0vh6Yl8/jWCixZcxMnDwoOHNzu+X70U9eERh1VfE4MR7vn+4WaZXZdawT3s9xHTfezG3j3vVo3g5yMFIJjkrB12hAmP0qNQE1hMvzjlyCfFSmwvkn9umhSn+uy1qNtU6ycxF1yKHwsXz25ZP9eQYyzOgsAkKolCY8Ju8urdrlg5bNwbNlVhPpEQa520ZrjmHq6GPlvPwyaLFpgmmVa08tu9JtSox7JJ9ivxYsBpyEpUYs3k+yv1hXb23M3ZlZ/PYzv2TF4PfBs9WkqJGNbbkDvER0xZlYf9BlZNZkW9x9/DesPvgIfyxNzM6EmpwgAmG9/G7cH1sxTMjUF5/CidCK9W4RWyZgXA+ygKCWLgPQ47O9afBOlsrB88BE3D5hgif50DJvRG8eXX8XWS0uRlpiBF5esISlVC3O3VZ0+DCNew/wdcHMIQIOm9bD9X2MkxKTCLMyo7E4VZPA4bqqEktYxxYiGT+wMdGj4pErGGv7OCFs6jkJH5UYYY3UGX/+nx3tyqWqOLLmMHdeWITokHo011atFBwYRG0xB/M9uL94M0mdCF8YI8Y3Ghimn8NKvfDlfxPxdGPpY4VqgA1a2HYw17YaW3UGMMPy9BvNS0FsMVu8M6XQXFVEAACAASURBVFpSkJaQhIKUHKY4HEBjeVVE5yTBbvgJDLLeAtvhxyHBcNzQ6iA5KRP1VBUF1iXEp6O+ulIVa1RxJjmsQae6rRGUGY4WtZvguPYmrPiij4lNhmJco0FY+XU/LvSo/LznNZFsVj5qS1W73+Kfxp+9S14ay1uNQzulZtBSbITmCupQlVWC3fATeNhvO6+N3fAT1WIsY7MzAACzre9hyKsLfHXjza4Wa/8hOrhY2c+M7HcQ65ffhKOdPxw++PHKr114j5H9DuK1yVcGtOYSG5cG5y8hvEf6yuTlgHMAgCUa03Bcm3s4oENdLcTnJWOT+3G0VhR9F/9PQWwsaz6/lcGsDKxivUus62GmB//0aKHkNKxdB9ZRgchhFaCxQtHM71XYN7CJ4JMSx9d+u7NpqfIsP+7CgMFtkZSQgajIZADAqAGHUPDD/cPT7TtG9qv40UAAaNigLrbueYo2rRogMSmTEZklMclhDQ51Xo9RDftjksMa5LLzUVtSDu3qaCA4MwIrWs2q1PGZIDaxuDP938gIizO4HvgJ4ZnJ1a1KlfFbPZJXBj3MuGHCvo49JFJdScx9fx8JOVmwGM8NTnzc3QZbuw6F5v3DCJnNPab5MMgds1p1BRHBJjoIw5sI544EgGckLT/uErqPKMTFp6PBb/ioXxX0n2+EhmpKYLM5eHFat9xyXnzXRQEnC5kFcaglIYkFrd6BQ2zUkpDE1YCB0FAcjNBMOyxqbQVJiZo76xxmfhqWo9dBUkJCaDe1auDvyelT2byP/Ubd3+4UWNf97c4S6wSRmJNJaXk5pbaxigwoVmYbHSz0GGKK03u24PzXTJORlUsrDzyiFQceUt85BhRaziylhXzP/EjPwxZRQg43HWx2QRJlFsQzoWqV0vaZPt97DYRRW/VXzjALZ44AML5JN0xp1hNLnbjR1Qtnkz+3+Rnbkbv5nNjFlE5WVh42bHsAAJCSksT5U/Nw+rwlmjaph+xs7gkr1XqKGDe6C9w8vqObECd6+szhRlrq11UDH91D8b/BnfDGlru04nSvcuMjstkc9J9vxNg49nHH4J/2Bkvb2PPKrgYM5LsXUyH+3l1yJkjOz8JIa+7RSBUZBXSo2wTZrHy4pYQB4BpMQY/icxyN4fdjPVOUR/S/hdy8AkREJqO1VoNidckpWYiNS4PXt0jMmqYDABg29jg4P757H8yLzuMnJGagvlqdYjIEUWg4C6lsYymGn/TkTCTFpKAgtwBrB+2DRdZtvvqtY4/guBl/CL/RCvP52h2cexZaXVrgpv5TqDdTxR0/xv2V/95dciaoJ6PAu7YavhNnei7A1T5F61ELP13E/i7T8XXsIWxyLcr5c6//3xEFWxiiolMQHpGE+ASuZ0BIWAJs7P2xdN0tAMDg8fy77fVUuH/zQmMJAK201PHBfBtWLSuKRxoVnYJPLsH45htV6vihUdxz1073NmHdnMHYuXRUxT8Ug3TeZIRPAcwefa2JOJm6YoWOHlQbq8Ai6zb8PnM9P6Y24qZaOW62A7lZefh5UvYm+TqfDIeXX3BT/yl0xnTF+nNVH4NWVP46g/kzPcz0eK9CbvZdgfFNuqGHmR4+xPlWo3Y1E+Mr79GokTKWrL2JxWtuAAA0W5YcoakQrV9OjFw+txAAMHNq0TnlJo1VMHFcV3Rs3wSpadklytJoogoOEfrMMcDscT0xcWjncnwS5vCJjCtWtuzScwEthed93CMAgFXsA75y//SvCMn0wi7PaRWSzwSj5g2CRdZtzG61Hm4fvqFdL24ytOcxFwEA/l9CoNtjOyQkJDBaYT7uHXkBaVkpHF10AaMV5mPnxOMgDsEi6zZ23VkDvUmVkwSQUZheFC3nq0oRtKEjqKynmR5fWfe3O+mgl0mFxv5f9z189xwOh3d9bPvjCsmuSuYuu0r6R1/x7t9aevGuB407Vh0qVRudNhryXj/fVwQ9j6lk6LeaHBNeUx6bu5nonfqJd73TY0rFlC4BLWND4hDRvJdPKZdVQERE+SwWERF1unSmUsasZP78vOQ1hc9jivwcC2ehep0mV0imopI8XD8FoXvfVgAAq1duaNxcFZvmX4a51++zNnrnEn+k+rEjigKM2Jr+XWfZ2zdVx+MNc3j3PTSb4GtI6csKZVFfrila1G6PvmrjeY+0Hev2qZDMsshnsyEpIQGXqEioyssjn82GrKQUPOJj0bNRE7RSUa3U8X8H/mqDWeJOeLwvBqu3592TiHtSiz4vwo1eN/jKxnTWg4ysFF590QdxCGM668HM4yBGTuqOIN9omHsdwqrp53D+6RrRP0gl0urxQchJSiGXzULQTK7vZ3R2GhrXrotpVtfRsk49GPSu2I/I70znTUZ87wDQuJ4SmqkqV0ju+janeddpBYnI5+RCXa4Zr6ytEjPpJn5GRlISynJyiM/ORLv/t3ffcVEcbRzAf3SkKU0URQEbIFYUjcZYooJd8xp7iYo1Mcbejb33gr33FlQsWBAVC6iggIogTaVIb3LAtXn/uHh4oR1whfJ8P5/75HZ2d+Yh9+Z5Z3dnZ4xNMd/zNlx79Ueb2nVgtVs5i46VN1XuKTlQ8IB0h1uL4eO0Ehqqalj4+hzuxAXBUFMXqdws8TFWeqa41OmvQus9++ks7sTfwQDzARhYp2IlkXnPr+FDRiKqqWliiHVLDKwvui/Y8MJqGGvpIjk3S5wwF7xwx7P4KLSvaYnfGjvCtkb+J+PK9PB9JDrbWBV/oCzbfBeBznbWAETJM2jLTIW2LwsfUpKhr6UFr6gIDLG1h/fnj+hS3woCxqAm7yn15YeGFSmCkDG09ch7m6adSUO4th1X6PHTX02HqZYpzKuZAwBcrFzkHqOsNbwgugXxLTF+2/7et32fs1LR2+MALPRqwFRbD8c7j8x3rKLdDAiBqb4uJhy6jMC1MxTWbvM528AYELRlJtrM34VcPr9CJkxF4fME6GsmepLukSIa/3zn9BN8/vAF6hpqqFXfBM1/bILaUjxMlAIlzP86dO4JNNRFCzGNHSzf+zxFCckMwfr36wEg3yU5yW/NgdtQU1OFtqYG3B8GwfOQaNJo5ymuMDHUQ9inRPTv2kzqYUPfEmZb67oyiI4hNsYC5nXyZlyPjTGHkfERaGs7SxwZ+DEOI3eK1v5WVVVBwKbCr0JkYVLH5YiLSkKDZhbY/u8Y1nUTD2HhQRd8CPiE+M/J+LGvfBcf43B5yMzOxdOwj1hzzQsvV0h3K8nZaCI8Ug7i2sH7eHrzNda7zQIApCakY7jNHAB5SVRG6NXI/9p59D7r+MsmFp+UUdaqinTEx6/A8karpH8impLNkdiecf86W/HUk9U/sJHVP7BRXH46+DUbdeOCRJms1Tus3KfZaZkcdudpMPP2D2fnbhX877akNlx/KJN6GGMsJro2i4muzRhjjMcLZbEx9Vj8lx9YTHRtlpEh+ZtP2HuJXXgaILO2i+NsMplFBceIt8e3XcICHoewFaNdWfCLCLm2PXa/aDSH3YKt7EloFBN8N9JDGn1rTWGMMfZb64US5U6GLszJ0EU2QeaRaa6qFOMwX78T9QJmrbwktzb67D+BC6+CMMvtJuzX7RSXt928F+6TRiMuIxOnXhS96NWQ62dhqF0NAGB5UDTmrF8DG8RzviJq4lxETZwr3vc2KQEne/+KqwNHo8Eh+awC+HuL9vCKjpBL3dLg8YU44uaD+duuwu/d5zLX13XtQXS1tS5zPbEx5mCMCwAwrxOL2BhzqKuLJkippjMUKioayMyQ/E0GOTbFykueZW67OH90W4OX998CAOpY10Q/c9ELFbnZXAj4AvT+7SfYtJHv/dtjk37F5puPAIiWEk7jlGzBNvc40fSHR/3WYsmveQ+3PFIOyrp3KXuyzsCl/JRa5Ockie0Ne28XeXx31SGsu+oQxhhjgY+CJba/ff+2/b05V26xhe632ceUVLbKw0tin6u3Dxtw8JRU8bq+8mFfsjIlyl58iWaW//Ywv/Uo1/k+YO1Ou0pVZ2k0PraFMcbYKt/7rM3ZPexCaFAxZ5Dv2W3bka/st0uX2YvoaCVEQ4og01xVKe5hlkQPtaHQ1tWCe8YJ8bZD9+ZYf3sxmJChp8YwuCUfgV4N3ULr2P/kOXrYNESNatVgpCPqMfY7cBLuk0Yr5G8gynX4pR+SOBw8jorCu4REhM+ZBQFj8AwLx/WQENiZmmJKO8fiKyKKQA99irPnqS/a17OAQ11zifKo1DRYGpZtfBwhALDF+wl0NDUw9bvEOO7yP+hmbY3RrVoqMTLyHzT5RlGabtmFthZ1sOyO5P2kN1/iK3Sy7KkxDD01hmHfnBPoqTEMO38/DAAIePAWK37dgmDfDxAKhEqOsmSEX2z/s90YLOc2GOc0wI8A+5q31MfF/fcBAEzIIFDy33kvLByzO3WUSJYAcPR/v4iTZVp2ye7rVSUcHq/QfefeBiowkpKrdAmzlr4eMnJyEZORt4xAmx17YV/LDL0OnyjizPJt6TnRuL7a1ma4wzuH6wfuAgBadGmKJ1degJfLg6paBfo5helQrRUMliJaH12YMhKqtUKhotECKtpOgLo1VPSmig+v8++YvN4NZkPw7zIdytK9YYNij6lRTTlzpjq3XAYelw/nlstw7ZwvAIDPF2Dnanc4t1yGAe1XKSWub3gCAebcu4W7kWE4ESR6SHo3Mgztju7DzLs3sdDrLqx2b4GgfFz55ifrm6Kl/MjFYo+7+cr2PXte6PFRWTEsISeZHQy/wDzivMXlQWmhLCgtlL1Jyz9juiKd23iVue26xYZbTmWMMcbJ4LAe6kPF+3uoD5WYzKO8E8Q1ZsLsu0wQ14gJM3eLCvlxjDHGhLmP8x2/adZptmLiYcYYY2/lPHSmonp87y0TCoRsbG/JYU/+PmHMqcVS9kvHNUqKLM/Gp6L/tix35c2Wf/5tELPctVmiTEZkmquUnSjlmjBLauKLJWzg42ls0oulEuV7w86ygY+nsYufPZQUWflR7/CGYsdvNp+Rf1zqvOM3St1mfEyK+PuKSUeYs+VM5mw5s9T1VQT33F6y0KDP+cqdG8wp8ryRPTexE66ejDHGLp98Ii53P+/LnFosZY/uvJFtoCXU2HUbYywvWX6fIEOSEpnt3vyjD8qInpLLS44gFw8Sn2N/+Dk00bdCaGaUeOINLVVN5ApFY/PcOu5RZphKlZzDwV8Pr+Ok05BCj+HyBfALj8YPTfKWzHWcuwu5PD4Ctsv2lUHL/ZthY2QCj19/w+pnD3Ao8CUAYE+Pfuhj3USmbSnK9AHbkZqYiYU7R8HH8x0mzO8DAODm8KCprYH5I/dhw+kpSo6ywqA3fSqba+HNxB+/+Pnisoqm+YytrO0c0ZyJAqGQNZ+xlfmHR7MfF7pKHOMZGCbXOJoekXkvpUCdth5grdbtlnm97/yjWK9Gc1nwqyiJch6Pz05s82DTB26XeZuVGPUwFc1y/2ZETZ4jsQ0AqioqEDIGPU1NvBn3p7LCq9ICEuIQlpYC02q6+MnCUvzbfP97yUOTlXnTudmbm+GyywiZ1s8Yg4qKCpiQQUU1r5O0cMwBrDsxSaZtVXI0DlPWjoT+CHUVbYxpdK/A/Zb7N8N31BSY6eoBADh8Hoa7X0DDGkZQVVHBpi7OBZ5HSmb10n/w8P47WNQzRnR0Cu54Fzxf6X9dC3uP/g1t4HDCFZqqang2ajJaHtuN17/Jd25RIWOYdPYKDo0YJNd2SJnINGFW6QmE3T6OhbmOaCLWAfUP59t/ONAPq595IWryHDDk9TSvhb2HuZ4+NFRVEZf1tdh2uDk89DNxwcVPe/Brvd9xI/UI1DVEsyu9CYqGfbO6+LnbOgCA5/2FRVVVqdUw1MXdJ0uKP/Bf33qTAPCn53UAQLvaFvCO/ij3ZAmIrjDKmiyT0rOQnMHBvL3XEZ2YBr+DNC1ceUY9zO/c+Pw7+ljI54FOX+MJqG5iAD6Xj/Xu82Blnzd79smTT3D82CP0H+CAP/8sXysgKopAIMSSueehb6CNlKSv2Ly79K+Zbn7xGLv9fQAAJtV08HLMNFmFWSCG0nVjznq+wvCfWyEiNhkHr/vCPzQatzfT5baM0SW5PHzJfg0jrYbQVNWTS/0T2yzCbu/luLrvLtr2aC5OmMnJX2FsLGrz527rMGLED5jg0gVrgxdhovUM6Knro5qaDgDRUhmbQ1YgLicajkY/op6OFToYd5ZLvIWpf0S0hO7H8VVj3Z7hPpNw3HEPxj7/HWfbHxCXL7t+D0lZHDQxM8HXXC4WO3Upcd3z9l7Hxql9AQAOE0X3RA/NG4JWjerIJHZlCE1KQkhSEuIyMuHzORp7B/SDlrpSL2TpKbmi8YQ8pbQbw/nEGGNsqt9IcdmSoL/YtlDR4OP1wUsUHlN9KcZhlle/+81kS4NWsZcp/ux9huRLCAfDjxV4DlfAZYwxNuzZxHz7snK5LJWTLftASyHyXUyB5f0tZyg4kjznAgOV1vZ3ZJqrKtC7dMpzKfoMEnK/YOmb2ZjiNxpT/BQzK9GjJE8wMLi2PoVp/qMAAIaaRvir0SIAwHybol9z42Tm4MpBL5nGFDV+XrnrXba+Kd19z92ttyKDlwEHw1Zooi+a3/Jpsi/Seel4mOhd4Dljnovmm7T59/jv5fD5ZX4FkpOZDSfdMRKfb3MCXD/oiT87L8etYw8KPd+l4wq4dFyBrX+dRF+L6fn2c3N4uLy34IeZspaQlbf+lcMeVyy6fRcNNm9VSNsKI+sMXMpPheAec5nNfDVJ2WFIZdvMk2z1hANskPVfyg5FrgRCIWt1Y3Gxx2Xx82a6H+M7kQmZkB2PPM0YY2yUzwQmEArYaJ/8s30vDVrHGGMslZsuUX7OL5CNP3W5LKFL4GRms546o8XbPjdfsYkOC9hQyz/YjulHCj1vw7SjbO3kw+KPUCD5aqxzrWkyi7GConGYysLAIGACqKuU/p7Mo4+2YEwAi+qT8OXrJTQzOwR9TfviTyylXrWm4dYXV7nVrwzfepT+vVej9c0l8O+df7E2WRjuI3oAs8xuLmwN8vcwZenmES/0Ht9VvO2kOwa3s05g6S9boK2rhVD/SBx/K7nULZ/HR0p8Bt4+D8fFPXfx82BHqKmroW6Dmlg6Mv9vrqOvjcuh8lsut6fjSrRwsISGhhpio1Ow/fB41DAsfF5ZBaF7mBXdwygb9iCyMfP53EVclvjlJxYXXZslJfQRl8VFm7P4WDuWkjiMJScOVkao5dK3HmW3u2tZam6WXNv6du+yoHuYhTkYdpMxxlhXz6Lf+67Ilvx1Rvw9MUG0llaPtitYoH8U69F2BYuLSVVWaP9F9zArsodRjfFT/WB0tgwBX5g3BZ2J2UNoarZF9Rp5PQANzVbQ1ZsCQ5OT0NTqoIxwy7VUbha63Vsr1za+PRlfbb+o2GO735+PLp5z4Fy7LZJzMyBkDOm8LIljunjK9w0kRVm1bbj4u4mpvvj7nKknYN+yHhb9eVoZYcldlR64rgydLUPF3zvW8wMAfIkxR606sQAAdY0m4m1j0+vIzjoDQB1gfGWEW259fykuz8vybxroWYq/MzCoFHCld6/bBnTxnIO6OibisuoaeZekkV+/YIHdUIRmRqOxviyWAi5f7jxfpuwQ5I56mHIQmFbwE9fCfEuWRqZXJbYBoJqu6B1lPYO5MoquZI66+eCH4aJeb/fxu/D7qgsAgIHTD0IoFN163nf+MU67v1BYTP69V8s9Qf7XNP9R2PFhHc5+Ooo//MeIRy2UxMKAI3iZEoo61UwqTU+zqqEepgw9SLiAVG4C6urI9wGBtOYEDEVTgzZgEOJdhj82tziPdF4KDDQMC+whFWTcoPYIeB8DANDX1UZSmugS88quiQCAl28/YcrQH+H3tuzL5JaWW+e/5N7Gpub7sPjNDPxadxR01HUxwLzg6e2G1st7kWCU5c8S+851zLusf/CzfJZOJvJFPUwZSuUmICDtAdoaOSk7FACAg2En2FV3QBujLmhZQ3QP1CvhGjaHzC5RPTy+AGMXnoSaqgrObxmHkMh4AABjQJum9QAAr99HF1mH3ZXl+NP3fCn+iuLV1zUp/qAy0lXXQ3vjTojI+oCeZn0LPW5qo37i7y4Nesk9LqJgsn6KVMpPheed6MYYY2zN21FsSeDAAo9xri8aEykUCtnBNVdZDidX7nFx+F/Z7Nd566zPeT2UzX49hOUIOEWcJR+2bn8rvE1FeZIYKP54xr9kvklvxfvmvd7N7n15wR4k+BdZB5dznTHGWE5GyebzbHhyY76y+sfXlaiOSozGYVZEN04/RXUjXWRn5SLwWRjuXRbd87sVta2YM0uPK8yFpqpWvvLgjFewNWiF5ylecDTqWsCZsmd3ZTkA4N3A5QppT9H6es+Be6dNOBnlgcdJATjQZgEAwDf5LZa9OQgDDV1c7LCm0PO5WcegoTMYKip64OfcByflN+ia3oKaRtMi27U8sR5RYxYUW1aYpgu34e26Sj1DEk3vVhH1GdkBvSxn4lbUNvQY7Ih7l1/INVkCKDBZAoCtQSsAUFiyfBT/AQuaOaOdiaVC2lMGVahABSp4mOAPNVU1cXk6Lwt7HOZAX12nyPM1dccgK2kYdE0ugJt1DOraP0OQ611kwrQ8sR5Huv2ar2xN+/JxS6gyooSpIItH7xMnyG+Js6LxefAezdtYQUcvLxH3b7sc3Fw+PAILf2q95NVVNKtRB26fXsOtq/Rr0Vge2ARb45q49b+xZYpbEebbiuYXGGfVFzU088Yl9qwlWrt85dsjWNZ0fKHnC/kfIBREAiwHOsbFLwdteWI9AKBb3Qb5ykY2biVxbNOF22BT2xQqKioIjk0AAHgvmYJOq/fh7bqZYAywX1Tpe5oyQZfk5YzdleWoXa064rLTxZevk5+dRjo3GwBwrrOLUuKaO/4wgl5GAgDUNdRw3W8FnJtLTnpRVNK0u7Icr/otgZaa9P8fPdPrBtqY1cFIu5alC7oCY4wDFZWCe6XZfB5sz2xB5JgF4uvNMffO41FspNSX4lWITC/J6Sl5OdO9ti1mNe2O+nrG4jLb6rUQmBqttGQJQJwsAYDPE5T4/HcDl5coWQLAtq59sPjxXVge2FTi9iq6wpIlANieEY2LtTqxHm0u7AIAPIoV/T6WJ9aLe5pE9ihhliPe8R8Qw0lDn7rN8JWXi8ScTABA77r2eDdwOc5GKm5w+H/1H94+X9lFKdfcKYuoSXMRNUk5g/YrgqScrOIPIjJDCbMc6WTWCJe7TgYAePeaA1Nt0b2wxgZmAIDhVm2VFtu0hX0xdnp3AEC3Pi0AAPrVq8GghqgnpIjkSUQiRs8Xf//Wa6dLccWge5jlmEAghJqaKo66+6JZg9poYycaJJ7JycVDvzBsP/sQ91zlu17N977ds/z+XmVBZaTsDmzxwA9dbNDMwVLZoVR0dA+zMnlw5y16Oq4scJ/bgyC0H7cNey89weGrvgCArGwucnJ56NXBFvdcp+GhX5giw6VkqQCu62/goUcQGtrUlms7PC4fvaxmYVDTBfgQpLxXWysUWY+EL+WnSunfeR178SyMMSaaQ7Ak2o7Zwo66+7J/vAIYY4y9Di14LRdZc2omOat5REgc27X6mszbWeh5hzHG2LBL59ix10W/GaMIh0IHMcYY2xXcle0PyZur9EPGA7bnfXfGF3Iljn+VfJHted+9zO2OctrM0lK+MqcWS8tcV2FCAj6yXtaz2Jldd+TWRjlAb/pUFgumn0L/wW3QobMNejquLLfTY30MT0D9BjUlynhcPjQ0ZT+Mt8vxQzDQ0sa1YSWfDaggh8LvwaVB91KfvzfECVOb3MbHrOeI4wShvekE5AoycfDDABhrWSEl9yMYhPjD5j5ORYyBo8lvaGzQDQdC+6KmdmMMrCffNW3+93QaWtaww5ecROxpvaJE5375nAIAGPfTamw6/zt0DarBysZcHmEqEy2zW9FFf0rG+MGi9c+1q2mAxxVA8O/CV4Bs5xW88z4MPW0a4uKrN/i1leyXwsjl8nHCzRe6OlrYffIhnl4o2cQeRel24gg01dTgMbLkA9cPh99DK0Nr3Ij1Qx9zB2QJctDJ1K5UceQIMpEtSIN3/B70t1iPT1nPcSN6GdRU1DGx8TWo/HtnK50Xi+oa5vCO3w1Hk7HQUtMvpuay+99T0T3sPa1XoJa2qdzbq4AoYVZkPR1Xou0PDTFzcV+M6LtdnBzl0cMMiU9CRHIKetk1xqkXr6GjqYFfWhT9bnJJdRiyBWtn94eXTyhmjuuGGgbVZFp/aY14uhUORg3hEeuPTjXt8CU7DX807gW76hbKDk2m4nISYKZlAlUVUdLmMz7UVdTxOSsVFrqGsHVbgXtOM1BHp4aSI1UaeuhTkd15vgxrdowQb/d0XCl+6NPTcSX6dVqLIU5b4NSu4AdBJdHEzAQ19XXhsFHUm5V1svxm0ZZraNOsHmatvYyJi89IfZ5AIMSvc47C/eEbJKZ+lWlM8TnpcI9+gckNe2KZ/RC8So0ot8ny8uu3+cpmXLqB8af+KfK8x0kvUVu7JlRVVMHAMCdgrXiBvrDMREx+ehpnO09A99s7wBOW/GUDkh/1MJUkNSULQ523FNirHNxjE/afmQJjU/lf0ilb1wm7MG5gO4zp5yiX+uOyU1G7miEAoP/DtbjWufi1eRTF5YwbolPTkcsXJTNDnWr4Z+IINF+7C4GL8tYYD4qNRzNzsxLVbesmeT8zeNDfZQ+4YqJLciJp5KrT6NTCGnyBEBe8XiM7l4cX+2UzC/maeRfw6O5b9B/qiMmznaGuoYa+7VaBxxWtMeQ0sDVm/T1A4pznzyOwYPEF8fb9uwUPqs7O4SElPQt1zESXi8eu+eK3/u1kEndF0GTlNqzt3xP/a9kU/fadhPsU0QQeXXcchteMCeLjXn2ORSuLkj+MyRXwIWBC6KhryizmCoguB1V6hQAADO5JREFUyYmktRN7o0UDcwzqZI8ra8bJLFkCwOKNQ2BorIfHnsFQ1xBNW/YtWQKArl7+KeSSU0SX1+3biWbSSU/PLrDuatoa4mQJoEolSwAY5tAci67dQVhiMq5OHoUmK0UzWGmqq2HY0bzZ6V9Fx5Wqfi019aqeLGWOepiVwI1nwUj9ygFfIASXJ8Ckfvnf+y6N1XPPw/dRKNx9lwIAnFr9jduvJC/1lv91Bsu3j5AoGzx0F9LSOLh3ez5+7rkennfotT2iNHRJTvIbt/4cji4YpuwwCClvKGESQoiU6B4mqTpS0mj6MlJ+0BIVShCTkg5VFVWoqamgpoGeuLzZrLxlK4K2Vu3lAkbNOIr1Cwdh2O+H4HFiOgZPOQCPk6KhNvPW/oONi35RcoTSa7B1K9rVrYtcgQAqAC4NH44GW/O/Mhk+a5ZU9b0Ki4FP8Ecc9ngOoZDBf+9MtJm2HVdXjkNKJgeGetVQ17TKDlSXK+phKkEdo+qYffw6fl5+UKJ84SDRomRHf/9VInmW1RxPD1ju2SL+fM/p7PFS1Wl9Zi2sz6zFqPtnxN9lqVZNA9StVQM1jfUx/e/z4mQJAC3s6mLZFndwsrkAgIT0DQiL64DI+H6IjO+HkBjZvwJaFuGzZsE3OhpD7O1xafhwAMD6nj0RPmsWGhgZwdrQUOpkCQATtlzAsTsvIRTm3cn6Y2BHeAdFQlNDHfuu+8j8byD/kvVsHqX8VCkjd5xlITGJ7H1MArOfuZV1WOTK7GduZYwx8T9lafGDu6z1YVc25dZV1vLQHnG57X7R+tez790qUX29bx5kVqfXSJT99eRqvrKy6vjLJsYYY7NWXWKMMcbl8llGZrZ4/7FLz1jWd2u78wXJjDHG3kc3LbLeXcFdJbYzuPEsKSec3YldK5O4v2ezfTs7+PKleNvB1ZW9io2VOGbQ6dNS15fD5bGsHC7bdeUxO+LxnLWaspXxBQJ27dlb1nrqVrb7ymOWw+WxlEzFrztfTsk0Vyk7UVbJhKloLjeuiL9/S5jD3M5LHFN/92ap6yssMco6YZbG50QX9jlpcqH7v/KS2KWo6WxfSG/GGGOecZJ/93+TKanwKGGSok1/+g/74eoO1vPmPoly63OrZVJ/eU6YRfmWDC9FTZfY/pwlmnfzyIfBygmMyJNMcxXdw6zA9rx7nK/MPykaz+KjoKWmjrCMJLxL/QIAyODloHudxjJpt0MtSzhc3i5RdvNTsEzqlqc/bO4DAP5Xf6fEdl0d0Tre4xpeVE5gpMKghFmB/W73o/j7LJ+ryBHw0dqkLrhCAX40s8KaNr1hZ1gLAGCgoQ0VGQ1JO9VtBFJzORjrdRYAMOeZO/547IaIEYqd2KLH7R0Y431MvG1/ZRUA4ES4L9YE3lJoLKRqoIHrldC7tHjY1TADh8+DjroGnid+gqNpPfF+7y8R6FTLusztTPP+Bx6f32Nko9ZY1da5zPWVRKtra/C87wIk52Zh2MNDeOAsesrc9MpK9KrTFClcDo50HK3QmMrqU1QS6lmaKDuMyobe9CHE1m0FfjJriN8a/oBWxhZIzeVgyrMz2PvDCJjrVC91vZ8jEmFhrfiZy9evuooFS0WzPvX8aS1GjumIsS6dFR5HJUQJkxB5eHgjAN4eQViyS7Se0MmddyHgC9G1X0vsXPoPtpybquQISSnQq5GkfIh48xkx4QnwvR2IA9/NfwkAn0ILnpJs5+2neBEejRfh0Uj5yhGX28+T3UD90rh1/jk692kBne+mqxv9Zw/8NssJ9RuZUbIkAChhVnrJXzmwWb4N518Gyrxua3sL7Jh5AubWNXHlgCcAYGzLBTiw+AKCX4Rj4+RDcDaaKHHOn04dMG7/RTwMjoCRno64fESHlrCftw3287YhIV22y1VIo9dQ0YzvvYe2Q68mNB0dKYSsxymV8kPkpMnfW9m1gGC51D26+XwW4h/JnAxdGGOM9a01hTHGWHpSJhvdbF6+42NS0tn4/RcZY4wN3n6KNZ27lfXecITNPX2DMcZY07lbWdO5sn/TiVRpMs1VdA+zkrNZvg1uU0bBthYtwUqqJLqHSUpm0L5Tyg6h0ulV909lh0CUgHqYVUREUgqsTYwK3e+X6g8Hw9YY+3wCJlpPwLlPF7C79fYCj/2axsHUzithbmWKDVdmyyvkcqmgRHkreqcSIiFSoh4mKbn/JsupftOxNniDeNvBsDVmvZ4LAPBJ9i00WQKAXg0d5GbnIu5jEtz23ZPYF/oqCkKBUIaRiywOlO38l3fjgjDj5QnsDLmNXl4bij0+MTYVy8cdQAfn5rBsUhsAYGZhhJY/NsbWWafx0uudTOMj5RMlzCpof/ghqKuqg8/4OBElulxfELgEW1tuAgAk5SYVW8dm93nQr6GLQVO6i8u4OTzUMNGHqpp8/me1IdhFZomzR+1mCEz7hD+bOCE59yv2hN4t8vid889j+dFJWHrIBXs9F8LAUBcOnW2x7twfCHn9EW262skkLlK+0SU5KfeWBP4PzWp0xNB6otcfFwf+gjXN/5FZ/ancLGx+dx1rWg4t9thvl+S3ondKfCflFr3pQ4iy8fkCqKurKTsMUjxKmIQowoo3I6Cjro/Wht0AAGbaFrCv3kHJUZESooc+hCjC3/ZnYKBujCb6DrDQaQwbg7bKDokoGSVMQoqgqqKKujoNcSP2CNRVNJQdDlEyuiQnBTr7NAA7bz9By/q1Ud/EEAv6d1Fo+3NH7YdRTQNMmt8HxmYG4vIA3wjUa2CK0DcxUFEBHDvbKDQuUuHQJTmRv7i0DHD5AnS3b4QfGtUr/gQZW3tkAiLex8HvyQfMHX0AAOB57RUWTTiMjDQOlk89juaORU+C3OCcbJf+JYR6mKRAj4IjcfjBC/hFxojL3mycCQDYFOCF8U0c0eHaLrz6ZTaaXdqE684uOB3mB30NLWiriS5dZ9h3KlMMve0WQUVVBe4Bq8RjO8f+vAGDJ/wE11XX0KZTY6w6MK5MbZBKT6Y9THVZVkYqj59srfCTrVWB+/YFP8PQBi3BGMOd6BAAAIfPxdmwVwgftghfOJkw09EvU/vn9nth27mp0DOohsd33sDapjbqWpniuOd8AICX+2tKlkThKGGSEgsftkh8udvcWPSaoINJXfH+Wv8my9fJsWhpbF6qNoZN7ir+XqeAdW62nqUJfYni0SU5IaQyo4c+hMhDTs4DZYdAyjlKmESh0nnJuPDZVdlh5JOT+xCamq2QkDRI2aGQcowSJlGo6hrGeJniVeC+4AxvbA8ZilRurEQ5n3EBAF4JR+UWV1KyCwABcnN95NYGqfgoYRKF8E99hI9ZoVgYOAwbW1zEvIBf8x1zNXo9TLUsofafN2puxe6ER9xucAXZRbYRn3kKDAK8im4vUS5gHGTkPMPHlBUS5VzBF/H3uuYfoKpqBIs6Ba92SQhAD31IOXL+01JEfPWDS4O9OBQ+FQvtbmJv2HhMbXgE6971Rg2NWhhttQl66sYFnu/70Qrt6keCL0yFuqohACAg9mfk8CIkjmtXPxJv4vpBS70eVFQ0UM9wATTVasn97yNKQbMVEVKQ9/GjYWN2EhzuO0Sl/A27WhcBAKmcuzDU6SFOqIAouTpYvIa6anXk8CKgrVH0W0OkwqKESaqGtm47MMm2PSbatCvReWFJM9DQZEeB+yKS58LaeCO+/+9IyHKhqqJVllBJ+UUJk1R+Nuc3wPXH/2FjgBc8ek9Udjik4qJXI0nl937ofCTlZKFtTQtlh0KIGD0lJ+XSXB93OLrtwOkP/hLlCRlflRQRIdTDJOWUtroGpjXtgDnNu4jLMrJzEJeWieiUdOhpaaFx7fzvmBMiT9TDJOXSqjbOCEuXXO7XoJo2ph29Ars6ZkjjZCOHx1dSdKSqooc+pNz6kJ6Iax/fYXbzzuKy8QcuAQA+JqfBc6GLskIjFQc99CFVg/eXSPCFAomyI5MGKykaQuiSnJRTtz6/x/OET5jfspuyQyFEjBImKZc4fB6y+Txlh0GIBLqHSSqM6eMPQ01NFfFxaRg14Sd06GwDQyNdZYdFyjd604dUXft33IGFpQl6D2it7FBIxUAJk5Ce7VcCAO74LFNyJKSco4RJCCFSojV9CCFEGShhEkKIlChhEkKIlChhEkKIlChhEkKIlChhEkKIlChhEkKIlChhEkKIlChhEkKIlChhEkKIlChhEkKIlChhEkKIlChhEkKIlMrLmj4ynVGEEELkgXqYhBAiJUqYhBAiJUqYhBAiJUqYhBAiJUqYhBAiJUqYhBAiJUqYhBAiJUqYhBAiJUqYhBAiJUqYhBAiJUqYhBAiJUqYhBAiJUqYhBAiJUqYhBAiJUqYhBAiJUqYhBAiJUqYhBAiJUqYhBAiJUqYhBAiJUqYhBAiJUqYhBAiJUqYhBAiJUqYhBAipf8Dv9Cn84eDQ8cAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x26fd7e82b00>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "from wordcloud import WordCloud\n",
    "\n",
    "pic = plt.imread('../data/aixin.jpg')\n",
    "wc = WordCloud(background_color='white', max_words=300,\n",
    "               font_path=r'C:\\Windows\\Fonts\\SIMLI.TTF', \n",
    "               mask=pic)\n",
    "wc2 = wc.fit_words(num)\n",
    "plt.figure(figsize=(5, 5))\n",
    "plt.imshow(wc2)\n",
    "plt.axis('off')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "## 去除停用词"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {
    "collapsed": true,
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[' ',\n",
       " '\\ufeff',\n",
       " '：',\n",
       " '&#',\n",
       " '】',\n",
       " '【',\n",
       " ',',\n",
       " '!',\n",
       " '会',\n",
       " '的',\n",
       " '日',\n",
       " '月',\n",
       " '\"',\n",
       " '#',\n",
       " '$',\n",
       " '%',\n",
       " '&',\n",
       " \"'\",\n",
       " '(',\n",
       " ')',\n",
       " '*',\n",
       " '+',\n",
       " ',',\n",
       " '-',\n",
       " '--',\n",
       " '.',\n",
       " '..',\n",
       " '...',\n",
       " '......',\n",
       " '...................',\n",
       " './',\n",
       " '/',\n",
       " '//',\n",
       " '0',\n",
       " '1',\n",
       " '2',\n",
       " '3',\n",
       " '4',\n",
       " '5',\n",
       " '6',\n",
       " '7',\n",
       " '8',\n",
       " '9',\n",
       " ':',\n",
       " '://',\n",
       " '::',\n",
       " ';',\n",
       " '<',\n",
       " '=',\n",
       " '>',\n",
       " '>>',\n",
       " '?',\n",
       " '@',\n",
       " 'A',\n",
       " 'Lex',\n",
       " '[',\n",
       " '\\\\',\n",
       " ']',\n",
       " '^',\n",
       " '_',\n",
       " '`',\n",
       " 'exp',\n",
       " 'sub',\n",
       " 'sup',\n",
       " '|',\n",
       " '}',\n",
       " '~',\n",
       " '~~~~',\n",
       " '·',\n",
       " '×',\n",
       " '×××',\n",
       " 'Δ',\n",
       " 'Ψ',\n",
       " 'γ',\n",
       " 'μ',\n",
       " 'φ',\n",
       " 'φ．',\n",
       " 'В',\n",
       " '—',\n",
       " '——',\n",
       " '———',\n",
       " '‘',\n",
       " '’',\n",
       " '’',\n",
       " '‘',\n",
       " '“',\n",
       " '”',\n",
       " '”，',\n",
       " '…',\n",
       " '……',\n",
       " '…………………………………………………③',\n",
       " '′∈',\n",
       " '′｜',\n",
       " '℃',\n",
       " 'Ⅲ',\n",
       " '↑',\n",
       " '→',\n",
       " '∈［',\n",
       " '∪φ∈',\n",
       " '≈',\n",
       " '①',\n",
       " '②',\n",
       " '②ｃ',\n",
       " '③',\n",
       " '③］',\n",
       " '④',\n",
       " '⑤',\n",
       " '⑥',\n",
       " '⑦',\n",
       " '⑧',\n",
       " '⑨',\n",
       " '⑩',\n",
       " '──',\n",
       " '■',\n",
       " '▲',\n",
       " '、',\n",
       " '。',\n",
       " '〈',\n",
       " '〉',\n",
       " '《',\n",
       " '》',\n",
       " '》），',\n",
       " '」',\n",
       " '『',\n",
       " '』',\n",
       " '【',\n",
       " '】',\n",
       " '〔',\n",
       " '〕',\n",
       " '〕〔',\n",
       " '㈧',\n",
       " '一',\n",
       " '一一',\n",
       " '一下',\n",
       " '一个',\n",
       " '一些',\n",
       " '一何',\n",
       " '一切',\n",
       " '一则',\n",
       " '一则通过',\n",
       " '一天',\n",
       " '一定',\n",
       " '一方面',\n",
       " '一旦',\n",
       " '一时',\n",
       " '一来',\n",
       " '一样',\n",
       " '一次',\n",
       " '一片',\n",
       " '一番',\n",
       " '一直',\n",
       " '一致',\n",
       " '一般',\n",
       " '一起',\n",
       " '一转眼',\n",
       " '一边',\n",
       " '一面',\n",
       " '七',\n",
       " '万一',\n",
       " '三',\n",
       " '三天两头',\n",
       " '三番两次',\n",
       " '三番五次',\n",
       " '上',\n",
       " '上下',\n",
       " '上升',\n",
       " '上去',\n",
       " '上来',\n",
       " '上述',\n",
       " '上面',\n",
       " '下',\n",
       " '下列',\n",
       " '下去',\n",
       " '下来',\n",
       " '下面',\n",
       " '不',\n",
       " '不一',\n",
       " '不下',\n",
       " '不久',\n",
       " '不了',\n",
       " '不亦乐乎',\n",
       " '不仅',\n",
       " '不仅...而且',\n",
       " '不仅仅',\n",
       " '不仅仅是',\n",
       " '不会',\n",
       " '不但',\n",
       " '不但...而且',\n",
       " '不光',\n",
       " '不免',\n",
       " '不再',\n",
       " '不力',\n",
       " '不单',\n",
       " '不变',\n",
       " '不只',\n",
       " '不可',\n",
       " '不可开交',\n",
       " '不可抗拒',\n",
       " '不同',\n",
       " '不外',\n",
       " '不外乎',\n",
       " '不够',\n",
       " '不大',\n",
       " '不如',\n",
       " '不妨',\n",
       " '不定',\n",
       " '不对',\n",
       " '不少',\n",
       " '不尽',\n",
       " '不尽然',\n",
       " '不巧',\n",
       " '不已',\n",
       " '不常',\n",
       " '不得',\n",
       " '不得不',\n",
       " '不得了',\n",
       " '不得已',\n",
       " '不必',\n",
       " '不怎么',\n",
       " '不怕',\n",
       " '不惟',\n",
       " '不成',\n",
       " '不拘',\n",
       " '不择手段',\n",
       " '不敢',\n",
       " '不料',\n",
       " '不断',\n",
       " '不日',\n",
       " '不时',\n",
       " '不是',\n",
       " '不曾',\n",
       " '不止',\n",
       " '不止一次',\n",
       " '不比',\n",
       " '不消',\n",
       " '不满',\n",
       " '不然',\n",
       " '不然的话',\n",
       " '不特',\n",
       " '不独',\n",
       " '不由得',\n",
       " '不知不觉',\n",
       " '不管',\n",
       " '不管怎样',\n",
       " '不经意',\n",
       " '不胜',\n",
       " '不能',\n",
       " '不能不',\n",
       " '不至于',\n",
       " '不若',\n",
       " '不要',\n",
       " '不论',\n",
       " '不起',\n",
       " '不足',\n",
       " '不过',\n",
       " '不迭',\n",
       " '不问',\n",
       " '不限',\n",
       " '与',\n",
       " '与其',\n",
       " '与其说',\n",
       " '与否',\n",
       " '与此同时',\n",
       " '专门',\n",
       " '且',\n",
       " '且不说',\n",
       " '且说',\n",
       " '两者',\n",
       " '严格',\n",
       " '严重',\n",
       " '个',\n",
       " '个人',\n",
       " '个别',\n",
       " '中小',\n",
       " '中间',\n",
       " '丰富',\n",
       " '串行',\n",
       " '临',\n",
       " '临到',\n",
       " '为',\n",
       " '为主',\n",
       " '为了',\n",
       " '为什么',\n",
       " '为什麽',\n",
       " '为何',\n",
       " '为止',\n",
       " '为此',\n",
       " '为着',\n",
       " '主张',\n",
       " '主要',\n",
       " '举凡',\n",
       " '举行',\n",
       " '乃',\n",
       " '乃至',\n",
       " '乃至于',\n",
       " '么',\n",
       " '之',\n",
       " '之一',\n",
       " '之前',\n",
       " '之后',\n",
       " '之後',\n",
       " '之所以',\n",
       " '之类',\n",
       " '乌乎',\n",
       " '乎',\n",
       " '乒',\n",
       " '乘',\n",
       " '乘势',\n",
       " '乘机',\n",
       " '乘胜',\n",
       " '乘虚',\n",
       " '乘隙',\n",
       " '九',\n",
       " '也',\n",
       " '也好',\n",
       " '也就是说',\n",
       " '也是',\n",
       " '也罢',\n",
       " '了',\n",
       " '了解',\n",
       " '争取',\n",
       " '二',\n",
       " '二来',\n",
       " '二话不说',\n",
       " '二话没说',\n",
       " '于',\n",
       " '于是',\n",
       " '于是乎',\n",
       " '云云',\n",
       " '云尔',\n",
       " '互',\n",
       " '互相',\n",
       " '五',\n",
       " '些',\n",
       " '交口',\n",
       " '亦',\n",
       " '产生',\n",
       " '亲口',\n",
       " '亲手',\n",
       " '亲眼',\n",
       " '亲自',\n",
       " '亲身',\n",
       " '人',\n",
       " '人人',\n",
       " '人们',\n",
       " '人家',\n",
       " '人民',\n",
       " '什么',\n",
       " '什么样',\n",
       " '什麽',\n",
       " '仅',\n",
       " '仅仅',\n",
       " '今',\n",
       " '今后',\n",
       " '今天',\n",
       " '今年',\n",
       " '今後',\n",
       " '介于',\n",
       " '仍',\n",
       " '仍旧',\n",
       " '仍然',\n",
       " '从',\n",
       " '从不',\n",
       " '从严',\n",
       " '从中',\n",
       " '从事',\n",
       " '从今以后',\n",
       " '从优',\n",
       " '从古到今',\n",
       " '从古至今',\n",
       " '从头',\n",
       " '从宽',\n",
       " '从小',\n",
       " '从新',\n",
       " '从无到有',\n",
       " '从早到晚',\n",
       " '从未',\n",
       " '从来',\n",
       " '从此',\n",
       " '从此以后',\n",
       " '从而',\n",
       " '从轻',\n",
       " '从速',\n",
       " '从重',\n",
       " '他',\n",
       " '他人',\n",
       " '他们',\n",
       " '他是',\n",
       " '他的',\n",
       " '代替',\n",
       " '以',\n",
       " '以上',\n",
       " '以下',\n",
       " '以为',\n",
       " '以便',\n",
       " '以免',\n",
       " '以前',\n",
       " '以及',\n",
       " '以后',\n",
       " '以外',\n",
       " '以後',\n",
       " '以故',\n",
       " '以期',\n",
       " '以来',\n",
       " '以至',\n",
       " '以至于',\n",
       " '以致',\n",
       " '们',\n",
       " '任',\n",
       " '任何',\n",
       " '任凭',\n",
       " '任务',\n",
       " '企图',\n",
       " '伙同',\n",
       " '会',\n",
       " '伟大',\n",
       " '传',\n",
       " '传说',\n",
       " '传闻',\n",
       " '似乎',\n",
       " '似的',\n",
       " '但',\n",
       " '但凡',\n",
       " '但愿',\n",
       " '但是',\n",
       " '何',\n",
       " '何乐而不为',\n",
       " '何以',\n",
       " '何况',\n",
       " '何处',\n",
       " '何妨',\n",
       " '何尝',\n",
       " '何必',\n",
       " '何时',\n",
       " '何止',\n",
       " '何苦',\n",
       " '何须',\n",
       " '余外',\n",
       " '作为',\n",
       " '你',\n",
       " '你们',\n",
       " '你是',\n",
       " '你的',\n",
       " '使',\n",
       " '使得',\n",
       " '使用',\n",
       " '例如',\n",
       " '依',\n",
       " '依据',\n",
       " '依照',\n",
       " '依靠',\n",
       " '便',\n",
       " '便于',\n",
       " '促进',\n",
       " '保持',\n",
       " '保管',\n",
       " '保险',\n",
       " '俺',\n",
       " '俺们',\n",
       " '倍加',\n",
       " '倍感',\n",
       " '倒不如',\n",
       " '倒不如说',\n",
       " '倒是',\n",
       " '倘',\n",
       " '倘使',\n",
       " '倘或',\n",
       " '倘然',\n",
       " '倘若',\n",
       " '借',\n",
       " '借以',\n",
       " '借此',\n",
       " '假使',\n",
       " '假如',\n",
       " '假若',\n",
       " '偏偏',\n",
       " '做到',\n",
       " '偶尔',\n",
       " '偶而',\n",
       " '傥然',\n",
       " '像',\n",
       " '儿',\n",
       " '允许',\n",
       " '元／吨',\n",
       " '充其极',\n",
       " '充其量',\n",
       " '充分',\n",
       " '先不先',\n",
       " '先后',\n",
       " '先後',\n",
       " '先生',\n",
       " '光',\n",
       " '光是',\n",
       " '全体',\n",
       " '全力',\n",
       " '全年',\n",
       " '全然',\n",
       " '全身心',\n",
       " '全部',\n",
       " '全都',\n",
       " '全面',\n",
       " '八',\n",
       " '八成',\n",
       " '公然',\n",
       " '六',\n",
       " '兮',\n",
       " '共',\n",
       " '共同',\n",
       " '共总',\n",
       " '关于',\n",
       " '其',\n",
       " '其一',\n",
       " '其中',\n",
       " '其二',\n",
       " '其他',\n",
       " '其余',\n",
       " '其后',\n",
       " '其它',\n",
       " '其实',\n",
       " '其次',\n",
       " '具体',\n",
       " '具体地说',\n",
       " '具体来说',\n",
       " '具体说来',\n",
       " '具有',\n",
       " '兼之',\n",
       " '内',\n",
       " '再',\n",
       " '再其次',\n",
       " '再则',\n",
       " '再有',\n",
       " '再次',\n",
       " '再者',\n",
       " '再者说',\n",
       " '再说',\n",
       " '冒',\n",
       " '冲',\n",
       " '决不',\n",
       " '决定',\n",
       " '决非',\n",
       " '况且',\n",
       " '准备',\n",
       " '凑巧',\n",
       " '凝神',\n",
       " '几',\n",
       " '几乎',\n",
       " '几度',\n",
       " '几时',\n",
       " '几番',\n",
       " '几经',\n",
       " '凡',\n",
       " '凡是',\n",
       " '凭',\n",
       " '凭借',\n",
       " '出',\n",
       " '出于',\n",
       " '出去',\n",
       " '出来',\n",
       " '出现',\n",
       " '分别',\n",
       " '分头',\n",
       " '分期',\n",
       " '分期分批',\n",
       " '切',\n",
       " '切不可',\n",
       " '切切',\n",
       " '切勿',\n",
       " '切莫',\n",
       " '则',\n",
       " '则甚',\n",
       " '刚',\n",
       " '刚好',\n",
       " '刚巧',\n",
       " '刚才',\n",
       " '初',\n",
       " '别',\n",
       " '别人',\n",
       " '别处',\n",
       " '别是',\n",
       " '别的',\n",
       " '别管',\n",
       " '别说',\n",
       " '到',\n",
       " '到了儿',\n",
       " '到处',\n",
       " '到头',\n",
       " '到头来',\n",
       " '到底',\n",
       " '到目前为止',\n",
       " '前后',\n",
       " '前此',\n",
       " '前者',\n",
       " '前进',\n",
       " '前面',\n",
       " '加上',\n",
       " '加之',\n",
       " '加以',\n",
       " '加入',\n",
       " '加强',\n",
       " '动不动',\n",
       " '动辄',\n",
       " '勃然',\n",
       " '匆匆',\n",
       " '十分',\n",
       " '千',\n",
       " '千万',\n",
       " '千万千万',\n",
       " '半',\n",
       " '单',\n",
       " '单单',\n",
       " '单纯',\n",
       " '即',\n",
       " '即令',\n",
       " '即使',\n",
       " '即便',\n",
       " '即刻',\n",
       " '即如',\n",
       " '即将',\n",
       " '即或',\n",
       " '即是说',\n",
       " '即若',\n",
       " '却',\n",
       " '却不',\n",
       " '历',\n",
       " '原来',\n",
       " '去',\n",
       " '又',\n",
       " '又及',\n",
       " '及',\n",
       " '及其',\n",
       " '及时',\n",
       " '及至',\n",
       " '双方',\n",
       " '反之',\n",
       " '反之亦然',\n",
       " '反之则',\n",
       " '反倒',\n",
       " '反倒是',\n",
       " '反应',\n",
       " '反手',\n",
       " '反映',\n",
       " '反而',\n",
       " '反过来',\n",
       " '反过来说',\n",
       " '取得',\n",
       " '取道',\n",
       " '受到',\n",
       " '变成',\n",
       " '古来',\n",
       " '另',\n",
       " '另一个',\n",
       " '另一方面',\n",
       " '另外',\n",
       " '另悉',\n",
       " '另方面',\n",
       " '另行',\n",
       " '只',\n",
       " '只当',\n",
       " '只怕',\n",
       " '只是',\n",
       " '只有',\n",
       " '只消',\n",
       " '只要',\n",
       " '只限',\n",
       " '叫',\n",
       " '叫做',\n",
       " '召开',\n",
       " '叮咚',\n",
       " '叮当',\n",
       " '可',\n",
       " '可以',\n",
       " '可好',\n",
       " '可是',\n",
       " '可能',\n",
       " '可见',\n",
       " '各',\n",
       " '各个',\n",
       " '各人',\n",
       " '各位',\n",
       " '各地',\n",
       " '各式',\n",
       " '各种',\n",
       " '各级',\n",
       " '各自',\n",
       " '合理',\n",
       " '同',\n",
       " '同一',\n",
       " '同时',\n",
       " '同样',\n",
       " '后',\n",
       " '后来',\n",
       " '后者',\n",
       " '后面',\n",
       " '向',\n",
       " '向使',\n",
       " '向着',\n",
       " '吓',\n",
       " '吗',\n",
       " '否则',\n",
       " '吧',\n",
       " '吧哒',\n",
       " '吱',\n",
       " '呀',\n",
       " '呃',\n",
       " '呆呆地',\n",
       " '呐',\n",
       " '呕',\n",
       " '呗',\n",
       " '呜',\n",
       " '呜呼',\n",
       " '呢',\n",
       " '周围',\n",
       " '呵',\n",
       " '呵呵',\n",
       " '呸',\n",
       " '呼哧',\n",
       " '呼啦',\n",
       " '咋',\n",
       " '和',\n",
       " '咚',\n",
       " '咦',\n",
       " '咧',\n",
       " '咱',\n",
       " '咱们',\n",
       " '咳',\n",
       " '哇',\n",
       " '哈',\n",
       " '哈哈',\n",
       " '哉',\n",
       " '哎',\n",
       " '哎呀',\n",
       " '哎哟',\n",
       " '哗',\n",
       " '哗啦',\n",
       " '哟',\n",
       " '哦',\n",
       " '哩',\n",
       " '哪',\n",
       " '哪个',\n",
       " '哪些',\n",
       " '哪儿',\n",
       " '哪天',\n",
       " '哪年',\n",
       " '哪怕',\n",
       " '哪样',\n",
       " '哪边',\n",
       " '哪里',\n",
       " '哼',\n",
       " '哼唷',\n",
       " '唉',\n",
       " '唯有',\n",
       " '啊',\n",
       " '啊呀',\n",
       " '啊哈',\n",
       " '啊哟',\n",
       " '啐',\n",
       " '啥',\n",
       " '啦',\n",
       " '啪达',\n",
       " '啷当',\n",
       " '喀',\n",
       " '喂',\n",
       " '喏',\n",
       " '喔唷',\n",
       " '喽',\n",
       " '嗡',\n",
       " '嗡嗡',\n",
       " '嗬',\n",
       " '嗯',\n",
       " '嗳',\n",
       " '嘎',\n",
       " '嘎嘎',\n",
       " '嘎登',\n",
       " '嘘',\n",
       " '嘛',\n",
       " '嘻',\n",
       " '嘿',\n",
       " '嘿嘿',\n",
       " '四',\n",
       " '因',\n",
       " '因为',\n",
       " '因了',\n",
       " '因此',\n",
       " '因着',\n",
       " '因而',\n",
       " '固',\n",
       " '固然',\n",
       " '在',\n",
       " '在下',\n",
       " '在于',\n",
       " '地',\n",
       " '均',\n",
       " '坚决',\n",
       " '坚持',\n",
       " '基于',\n",
       " '基本',\n",
       " '基本上',\n",
       " '处在',\n",
       " '处处',\n",
       " '处理',\n",
       " '复杂',\n",
       " '多',\n",
       " '多么',\n",
       " '多亏',\n",
       " '多多',\n",
       " '多多少少',\n",
       " '多多益善',\n",
       " '多少',\n",
       " '多年前',\n",
       " '多年来',\n",
       " '多数',\n",
       " '多次',\n",
       " '够瞧的',\n",
       " '大',\n",
       " '大不了',\n",
       " '大举',\n",
       " '大事',\n",
       " '大体',\n",
       " '大体上',\n",
       " '大凡',\n",
       " '大力',\n",
       " '大多',\n",
       " '大多数',\n",
       " '大大',\n",
       " '大家',\n",
       " '大张旗鼓',\n",
       " '大批',\n",
       " '大抵',\n",
       " '大概',\n",
       " '大略',\n",
       " '大约',\n",
       " '大致',\n",
       " '大都',\n",
       " '大量',\n",
       " '大面儿上',\n",
       " '失去',\n",
       " '奇',\n",
       " '奈',\n",
       " '奋勇',\n",
       " '她',\n",
       " '她们',\n",
       " '她是',\n",
       " '她的',\n",
       " '好',\n",
       " '好在',\n",
       " '好的',\n",
       " '好象',\n",
       " '如',\n",
       " '如上',\n",
       " '如上所述',\n",
       " '如下',\n",
       " '如今',\n",
       " '如何',\n",
       " '如其',\n",
       " '如前所述',\n",
       " '如同',\n",
       " '如常',\n",
       " '如是',\n",
       " '如期',\n",
       " '如果',\n",
       " '如次',\n",
       " '如此',\n",
       " '如此等等',\n",
       " '如若',\n",
       " '始而',\n",
       " '姑且',\n",
       " '存在',\n",
       " '存心',\n",
       " '孰料',\n",
       " '孰知',\n",
       " '宁',\n",
       " '宁可',\n",
       " '宁愿',\n",
       " '宁肯',\n",
       " '它',\n",
       " '它们',\n",
       " '它们的',\n",
       " '它是',\n",
       " '它的',\n",
       " '安全',\n",
       " '完全',\n",
       " '完成',\n",
       " '定',\n",
       " '实现',\n",
       " '实际',\n",
       " '宣布',\n",
       " '容易',\n",
       " '密切',\n",
       " '对',\n",
       " '对于',\n",
       " '对应',\n",
       " '对待',\n",
       " '对方',\n",
       " '对比',\n",
       " '将',\n",
       " '将才',\n",
       " '将要',\n",
       " '将近',\n",
       " '小',\n",
       " '少数',\n",
       " '尔',\n",
       " '尔后',\n",
       " '尔尔',\n",
       " '尔等',\n",
       " '尚且',\n",
       " '尤其',\n",
       " '就',\n",
       " '就地',\n",
       " '就是',\n",
       " '就是了',\n",
       " '就是说',\n",
       " '就此',\n",
       " '就算',\n",
       " '就要',\n",
       " '尽',\n",
       " '尽可能',\n",
       " '尽如人意',\n",
       " '尽心尽力',\n",
       " '尽心竭力',\n",
       " '尽快',\n",
       " '尽早',\n",
       " '尽然',\n",
       " '尽管',\n",
       " '尽管如此',\n",
       " '尽量',\n",
       " '局外',\n",
       " '居然',\n",
       " '届时',\n",
       " '属于',\n",
       " '屡',\n",
       " '屡屡',\n",
       " '屡次',\n",
       " '屡次三番',\n",
       " '岂',\n",
       " '岂但',\n",
       " '岂止',\n",
       " '岂非',\n",
       " '川流不息',\n",
       " '左右',\n",
       " '巨大',\n",
       " '巩固',\n",
       " '差一点',\n",
       " '差不多',\n",
       " '己',\n",
       " '已',\n",
       " '已矣',\n",
       " '已经',\n",
       " '巴',\n",
       " '巴巴',\n",
       " '带',\n",
       " '帮助',\n",
       " '常',\n",
       " '常常',\n",
       " '常言说',\n",
       " '常言说得好',\n",
       " '常言道',\n",
       " '平素',\n",
       " '年复一年',\n",
       " '并',\n",
       " '并不',\n",
       " '并不是',\n",
       " '并且',\n",
       " '并排',\n",
       " '并无',\n",
       " '并没',\n",
       " '并没有',\n",
       " '并肩',\n",
       " '并非',\n",
       " '广大',\n",
       " '广泛',\n",
       " '应当',\n",
       " '应用',\n",
       " '应该',\n",
       " '庶乎',\n",
       " '庶几',\n",
       " '开外',\n",
       " '开始',\n",
       " '开展',\n",
       " '引起',\n",
       " '弗',\n",
       " '弹指之间',\n",
       " '强烈',\n",
       " '强调',\n",
       " '归',\n",
       " '归根到底',\n",
       " '归根结底',\n",
       " '归齐',\n",
       " '当',\n",
       " '当下',\n",
       " '当中',\n",
       " '当儿',\n",
       " '当前',\n",
       " '当即',\n",
       " '当口儿',\n",
       " '当地',\n",
       " '当场',\n",
       " '当头',\n",
       " '当庭',\n",
       " '当时',\n",
       " '当然',\n",
       " ...]"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "with open('../data/stoplist.txt', 'r', encoding='utf-8') as f:\n",
    "    stop = f.read()\n",
    "stop = stop.split()\n",
    "stop = [' '] + stop\n",
    "stop"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[['《', '鹿鼎记', '》'], ['作者', '：', '金庸'], ['正文'], ['第一回'], ['纵横', '钩', '党', '清流', '祸']]\n",
      "[['鹿鼎记'], ['作者', '金庸'], ['正文'], ['第一回'], ['纵横', '钩', '党', '清流', '祸']]\n"
     ]
    }
   ],
   "source": [
    "data_after = [[j for j in i if j not in stop] for i in data_cut]\n",
    "print(data_cut[:5])\n",
    "print(data_after[:5])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x26fd39ef0f0>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "from wordcloud import WordCloud\n",
    "\n",
    "all_words = []\n",
    "for i in data_after:\n",
    "    all_words.extend(i)\n",
    "\n",
    "tmp = pd.Series(all_words)\n",
    "num = tmp.value_counts()\n",
    "\n",
    "\n",
    "pic = plt.imread('../data/aixin.jpg')\n",
    "wc = WordCloud(background_color='white', max_words=300,\n",
    "               font_path=r'C:\\Windows\\Fonts\\SIMLI.TTF', \n",
    "               mask=pic)\n",
    "wc2 = wc.fit_words(num)\n",
    "plt.figure(figsize=(5, 5))\n",
    "plt.imshow(wc2)\n",
    "plt.axis('off')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 练习：垃圾短信分类"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 文件读取"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>labels</th>\n",
       "      <th>content</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>商业秘密的秘密性那是维系其商业价值和垄断地位的前提条件之一</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>南口阿玛施新春第一批限量春装到店啦         春暖花开淑女裙、冰蓝色公主衫  ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>带给我们大常州一场壮观的视觉盛宴</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>有原因不明的泌尿系统结石等</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>23年从盐城拉回来的麻麻的嫁妆</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   id  labels                                            content\n",
       "0   1       0                      商业秘密的秘密性那是维系其商业价值和垄断地位的前提条件之一\n",
       "1   2       1  南口阿玛施新春第一批限量春装到店啦         春暖花开淑女裙、冰蓝色公主衫  ...\n",
       "2   3       0                                   带给我们大常州一场壮观的视觉盛宴\n",
       "3   4       0                                      有原因不明的泌尿系统结石等\n",
       "4   5       0                                    23年从盐城拉回来的麻麻的嫁妆"
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "message = pd.read_csv('../data/message/data/message80.csv', header=None)\n",
    "message.columns = ['id', 'labels', 'content']\n",
    "message.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 数据抽取"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [],
   "source": [
    "n = 1000\n",
    "data = message.loc[message['labels'] == 1, ].sample(n, random_state=123)\n",
    "data = pd.concat([data, message.loc[message['labels'] == 0, ].sample(n, random_state=123)])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(2000, 3)"
      ]
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [],
   "source": [
    "import re\n",
    "import jieba\n",
    "data_cut = data.content.apply(lambda x: jieba.lcut(re.sub('[^\\u4E00-\\u9FD5]+', '', x)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [],
   "source": [
    "with open('../data/stoplist.txt', 'r', encoding='utf-8') as f:\n",
    "    stop = f.read()\n",
    "stop = stop.split()\n",
    "stop = [' '] + stop"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_after = data_cut.apply(lambda x: [i for i in x if i not in stop])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "metadata": {},
   "outputs": [],
   "source": [
    "import itertools\n",
    "tmp = list(itertools.chain.from_iterable(list(data_after)))\n",
    "# list(itertools.chain(*list(data_after)))\n",
    "\n",
    "num = pd.Series(tmp).value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x26fb8aa6198>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "from wordcloud import WordCloud\n",
    "\n",
    "pic = plt.imread('../data/aixin.jpg')\n",
    "wc = WordCloud(background_color='white', max_words=300,\n",
    "               font_path=r'C:\\Windows\\Fonts\\SIMLI.TTF', \n",
    "               mask=pic)\n",
    "wc2 = wc.fit_words(num)\n",
    "plt.figure(figsize=(5, 5))\n",
    "plt.imshow(wc2)\n",
    "plt.axis('off')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.feature_extraction.text import CountVectorizer\n",
    "from sklearn.naive_bayes import MultinomialNB\n",
    "from sklearn.model_selection import train_test_split\n",
    "# data_after = data_after.loc[data_after.apply(len) > 0]\n",
    "tmp = data_after.apply(lambda x: ' '.join(x))\n",
    "x_train, x_test, y_train, y_test = train_test_split(\n",
    "    tmp, data.loc[data_after.index, 'labels'], test_size=.2, random_state=123\n",
    ")\n",
    "\n",
    "cv = CountVectorizer().fit(x_train)\n",
    "cv_train = cv.transform(x_train)\n",
    "cv_test = cv.transform(x_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8725"
      ]
     },
     "execution_count": 114,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model = MultinomialNB().fit(cv_train, y_train)\n",
    "model.score(cv_test, y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 文本向量表示"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 词频"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## TF-IDF值"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "# 常用文本分类器"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "## kNN"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "## Naive Bayes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "## SVM"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "# 模型评估"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.6"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}