Test for classification.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Importing libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import sklearn.tree as tree\n",
    "import pandas as pd\n",
    "from hazm import Stemmer ,Lemmatizer\n",
    "import multiprocessing.pool as pool\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import scipy.spatial.distance as dis"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Open CorpusText & StopWords "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "a1 = pd.read_csv('./DataSets/EconomyCleaned.csv',index_col='Unnamed: 0')\n",
    "a2 = pd.read_csv('./DataSets/PoliticCleaned.csv',index_col='Unnamed: 0')\n",
    "a3 = pd.read_csv('./DataSets/SocialCleaned.csv',index_col='Unnamed: 0')\n",
    "# a3 = pd.read_csv('./DataSets/S',index_col='Unnamed: 0')\n",
    "# corpus_original = pd.read_csv('./DataSets/EconomyCleaned.csv',index_col='Unnamed: 0')\n",
    "corpus_original = a1.append(a2).append(a3)\n",
    "\n",
    "corpus = corpus_original.copy()\n",
    "stopWords = pd.read_csv('./stopWords.txt', sep='\\t',header=None)\n",
    "stopWords = list(stopWords[0])\n",
    "stopWords.append('تر')\n",
    "stopWords.append('ترین')\n",
    "stopWords.append('ها')\n",
    "stopWords.append('های')\n",
    "stopWords.append('ای')\n",
    "stopWords.append('\\r')\n",
    "stopWords.append('ی')\n",
    "stopWords.append('خواهد')\n",
    "stopWords.append('ااب')\n",
    "stopWords.append('-')\n",
    "\n",
    "\n",
    "# stopWords.head()\n",
    "# corpus.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Get text data "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "corpus = corpus.drop(['date'],axis=1)\n",
    "corpus = corpus[['title','summery','body','cat','subcat']]\n",
    "corpus = corpus[corpus['cat'].isin(['اجتماعی','اقتصادی' , 'سیاسی'])]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>title</th>\n",
       "      <th>summery</th>\n",
       "      <th>body</th>\n",
       "      <th>subcat</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cat</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>اجتماعی</th>\n",
       "      <td>1915</td>\n",
       "      <td>1915</td>\n",
       "      <td>1915</td>\n",
       "      <td>1915</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>اقتصادی</th>\n",
       "      <td>2204</td>\n",
       "      <td>2204</td>\n",
       "      <td>2204</td>\n",
       "      <td>2204</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>سیاسی</th>\n",
       "      <td>2520</td>\n",
       "      <td>2520</td>\n",
       "      <td>2520</td>\n",
       "      <td>2520</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         title  summery  body  subcat\n",
       "cat                                  \n",
       "اجتماعی   1915     1915  1915    1915\n",
       "اقتصادی   2204     2204  2204    2204\n",
       "سیاسی     2520     2520  2520    2520"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "corpus.groupby(['cat']).count()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Let's Start\n",
    "\n",
    "We want model our document to vector\n",
    "\n",
    "We find 100 important words from our corpus"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "at first we set weight:\n",
    "- words in title: 3x \n",
    "- words in summary: 2x \n",
    "- words in body: 1x "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1- set vector dimention"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "some cleaninc in text for some chars: '\\n' , ':' , '.' , ',' and ..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# corpus = corpus_original.copy()\n",
    "def clean(text):\n",
    "    text = text.replace('\\n',' ').\\\n",
    "                replace(':','').\\\n",
    "                replace('.','').\\\n",
    "                replace('،','').\\\n",
    "                replace('/','').\\\n",
    "                replace('فیلم','').\\\n",
    "                replace('+','').\\\n",
    "                replace('»','').\\\n",
    "                replace('«','').\\\n",
    "                replace('(','').\\\n",
    "                replace(')','').\\\n",
    "                replace('\\xa0','').\\\n",
    "                replace('\\u200c',' ')\n",
    "    \n",
    "    text = text.replace('گزارش','').\\\n",
    "                replace('فارس','').\\\n",
    "                replace('نقل','').\\\n",
    "                replace('خبرگزاری','').\\\n",
    "                replace('خبرنگار','')\n",
    "    \n",
    "    \n",
    "    text = text.replace('  ',' ')\n",
    "    t = text.split(' ')\n",
    "    t = [x for x in t if x not in stopWords and\\\n",
    "                           not x.isdigit() and\\\n",
    "                           not '']\n",
    "    return t\n",
    "\n",
    "corpus['title']   = corpus['title'].apply(clean)\n",
    "corpus['summery'] = corpus['summery'].apply(clean)\n",
    "corpus['body']    = corpus['body'].apply(clean)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "dictionary = {'ahmad':12}\n",
    "\n",
    "#inner function\n",
    "stem = lambda x:Stemmer().stem(x)\n",
    "lem = lambda x:Lemmatizer().lemmatize(x)\n",
    "def cbow_adder(x,importance=1):\n",
    "#     print('      ',x,x in dictionary.keys())\n",
    "    if x in dictionary.keys():\n",
    "#         print('      ',dictionary)\n",
    "        dictionary[x] += 1*importance\n",
    "    else:\n",
    "        \n",
    "        dictionary[x] = 1*importance\n",
    "    return x\n",
    "\n",
    "#outter function\n",
    "cbow_maker     = lambda txt: (list(map(cbow_adder,txt)))\n",
    "# cbow_stemmer   = lambda txt: (list(map(stem,txt)))\n",
    "# cbow_lemmatize = lambda txt: (list(map(lem,txt)))\n",
    "# def cbow_maker(txt):\n",
    "#     print('     ',dictionary)\n",
    "#     return (list(map(cbow_adder,txt)))\n",
    "\n",
    "# def cbow_stemmer(txt):\n",
    "#     return (list(map(stem,txt)))\n",
    "\n",
    "# def cbow_lemmatize(txt):\n",
    "#     return (list(map(lem,txt)))\n",
    "\n",
    "# p = pool.Pool(5)\n",
    "#start function\n",
    "# corpus.title   = corpus.title.apply(cbow_lemmatize)\n",
    "# corpus.title = p.map(cbow_maker , corpus.title)\n",
    "# with pool.Pool(5) as p:\n",
    "# #         corpus.title =\n",
    "#     p.imap(cbow_maker , corpus.title[:100])\n",
    "map(cbow_maker , corpus.title[:100])\n",
    "#     print('    ',dictionary)\n",
    "# print('       ','lemma done!')\n",
    "\n",
    "# with pool.Pool(5) as p:\n",
    "#         p.map(cbow_stemmer,corpus.summery)\n",
    "# corpus.title   = corpus.title.apply(cbow_stemmer)\n",
    "# corpus.title   = p.map(cbow_stemmer)\n",
    "# print('       ','stm done!')\n",
    "corpus.title   = corpus.title.apply(cbow_maker)\n",
    "# corpus.title   = p.map(cbow_maker)\n",
    "# with pool.Pool(5) as p:\n",
    "#         p.map(cbow_maker,corpus.body)\n",
    "# print('       ','maker done!')\n",
    "\n",
    "dictionary = {k:v*1.5 for k, v in dictionary.items()}\n",
    "\n",
    "# corpus.summery   = corpus.summery.apply(cbow_lemmatize)\n",
    "# corpus.summery   = corpus.summery.apply(cbow_stemmer)\n",
    "corpus.summery   = corpus.summery.apply(cbow_maker)\n",
    "dictionary = {k:v*2 for k, v in dictionary.items()}\n",
    "\n",
    "# corpus.body   = corpus.body.apply(cbow_lemmatize)\n",
    "# corpus.body   = corpus.body.apply(cbow_stemmer)\n",
    "corpus.body   = corpus.body.apply(cbow_maker)\n",
    "# print(dictionary)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "def dimFinder(corpus,dimNum=100):\n",
    "    dictionary = {'ahmad':12}\n",
    "\n",
    "    #inner function\n",
    "    stem = lambda x:Stemmer().stem(x)\n",
    "    lem = lambda x:Lemmatizer().lemmatize(x)\n",
    "    def cbow_adder(x,importance=1):\n",
    "        if x in dictionary.keys():\n",
    "            dictionary[x] += 1*importance\n",
    "        else:\n",
    "\n",
    "            dictionary[x] = 1*importance\n",
    "        return x\n",
    "\n",
    "    cbow_maker     = lambda txt: (list(map(cbow_adder,txt)))\n",
    "    map(cbow_maker , corpus.title[:100])\n",
    "    corpus.title   = corpus.title.apply(cbow_maker)\n",
    "\n",
    "    dictionary = {k:v*1.5 for k, v in dictionary.items()}\n",
    "\n",
    "    corpus.summery   = corpus.summery.apply(cbow_maker)\n",
    "    dictionary = {k:v*2 for k, v in dictionary.items()}\n",
    "\n",
    "    corpus.body   = corpus.body.apply(cbow_maker)\n",
    "#     dictionary\n",
    "    di = {\n",
    "        'word':[*dictionary],\n",
    "        'frequency':list(dictionary.values())\n",
    "    }\n",
    "    cbow = pd.DataFrame.from_dict(di)\n",
    "\n",
    "    cbow = cbow.sort_values(by='frequency',ascending=False)\n",
    "    cbow = cbow[cbow.word != '']\n",
    "    dims = cbow[:dimNum]\n",
    "    return(dims)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>word</th>\n",
       "      <th>frequency</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>771</th>\n",
       "      <td>استفاده</td>\n",
       "      <td>4303.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>118</th>\n",
       "      <td>بانک</td>\n",
       "      <td>4303.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2316</th>\n",
       "      <td>توجه</td>\n",
       "      <td>4285.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>399</th>\n",
       "      <td>اقتصاد</td>\n",
       "      <td>4270.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1575</th>\n",
       "      <td>موضوع</td>\n",
       "      <td>4099.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>727</th>\n",
       "      <td>ایجاد</td>\n",
       "      <td>4059.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2484</th>\n",
       "      <td>شهر</td>\n",
       "      <td>4055.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>248</th>\n",
       "      <td>میلیون</td>\n",
       "      <td>4006.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>333</th>\n",
       "      <td>تومان</td>\n",
       "      <td>3995.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>526</th>\n",
       "      <td>استان</td>\n",
       "      <td>3950.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>524</th>\n",
       "      <td>توسعه</td>\n",
       "      <td>3937.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1226</th>\n",
       "      <td>سیاسی</td>\n",
       "      <td>3834.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>161</th>\n",
       "      <td>دست</td>\n",
       "      <td>3718.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4498</th>\n",
       "      <td>اظهار</td>\n",
       "      <td>3715.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>647</th>\n",
       "      <td>بررسی</td>\n",
       "      <td>3578.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>624</th>\n",
       "      <td>شرایط</td>\n",
       "      <td>3554.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>556</th>\n",
       "      <td>مشکلات</td>\n",
       "      <td>3522.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>135</th>\n",
       "      <td>اجرای</td>\n",
       "      <td>3512.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>209</th>\n",
       "      <td>عضو</td>\n",
       "      <td>3499.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>283</th>\n",
       "      <td>دلار</td>\n",
       "      <td>3404.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>684</th>\n",
       "      <td>سرمایه</td>\n",
       "      <td>3365.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>164</th>\n",
       "      <td>ارائه</td>\n",
       "      <td>3364.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>57</th>\n",
       "      <td>جمهور</td>\n",
       "      <td>3327.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>614</th>\n",
       "      <td>دلیل</td>\n",
       "      <td>3312.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>949</th>\n",
       "      <td>جامعه</td>\n",
       "      <td>3275.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>376</th>\n",
       "      <td>میلیارد</td>\n",
       "      <td>3264.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>186</th>\n",
       "      <td>گروه</td>\n",
       "      <td>3234.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2570</th>\n",
       "      <td>دانش</td>\n",
       "      <td>3229.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>549</th>\n",
       "      <td>نفر</td>\n",
       "      <td>3168.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2115</th>\n",
       "      <td>آموزش</td>\n",
       "      <td>3158.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>496</th>\n",
       "      <td>سیاست</td>\n",
       "      <td>3120.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>141</th>\n",
       "      <td>توسط</td>\n",
       "      <td>3115.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1046</th>\n",
       "      <td>معاون</td>\n",
       "      <td>3093.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>472</th>\n",
       "      <td>کمیسیون</td>\n",
       "      <td>3092.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>937</th>\n",
       "      <td>نامه</td>\n",
       "      <td>3086.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2334</th>\n",
       "      <td>نماینده</td>\n",
       "      <td>3080.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>857</th>\n",
       "      <td>تواند</td>\n",
       "      <td>3066.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>584</th>\n",
       "      <td>منطقه</td>\n",
       "      <td>2972.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>996</th>\n",
       "      <td>نمایندگان</td>\n",
       "      <td>2969.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>زمان</td>\n",
       "      <td>2917.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>467</th>\n",
       "      <td>جلسه</td>\n",
       "      <td>2903.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>575</th>\n",
       "      <td>جمهوری</td>\n",
       "      <td>2901.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3410</th>\n",
       "      <td>ملت</td>\n",
       "      <td>2867.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>860</th>\n",
       "      <td>مدیریت</td>\n",
       "      <td>2817.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>395</th>\n",
       "      <td>مرکزی</td>\n",
       "      <td>2808.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>273</th>\n",
       "      <td>مالی</td>\n",
       "      <td>2806.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>573</th>\n",
       "      <td>برگزار</td>\n",
       "      <td>2806.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>738</th>\n",
       "      <td>صنعت</td>\n",
       "      <td>2777.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>54</th>\n",
       "      <td>حمایت</td>\n",
       "      <td>2756.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>704</th>\n",
       "      <td>اصلاح</td>\n",
       "      <td>2752.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           word  frequency\n",
       "771     استفاده     4303.0\n",
       "118        بانک     4303.0\n",
       "2316       توجه     4285.0\n",
       "399      اقتصاد     4270.0\n",
       "1575      موضوع     4099.0\n",
       "727       ایجاد     4059.0\n",
       "2484        شهر     4055.0\n",
       "248      میلیون     4006.0\n",
       "333       تومان     3995.0\n",
       "526       استان     3950.0\n",
       "524       توسعه     3937.0\n",
       "1226      سیاسی     3834.0\n",
       "161         دست     3718.0\n",
       "4498      اظهار     3715.0\n",
       "647       بررسی     3578.0\n",
       "624       شرایط     3554.0\n",
       "556      مشکلات     3522.0\n",
       "135       اجرای     3512.0\n",
       "209         عضو     3499.0\n",
       "283        دلار     3404.0\n",
       "684      سرمایه     3365.0\n",
       "164       ارائه     3364.0\n",
       "57        جمهور     3327.0\n",
       "614        دلیل     3312.0\n",
       "949       جامعه     3275.0\n",
       "376     میلیارد     3264.0\n",
       "186        گروه     3234.0\n",
       "2570       دانش     3229.0\n",
       "549         نفر     3168.0\n",
       "2115      آموزش     3158.0\n",
       "496       سیاست     3120.0\n",
       "141        توسط     3115.0\n",
       "1046      معاون     3093.0\n",
       "472     کمیسیون     3092.0\n",
       "937        نامه     3086.0\n",
       "2334    نماینده     3080.0\n",
       "857       تواند     3066.0\n",
       "584       منطقه     2972.0\n",
       "996   نمایندگان     2969.0\n",
       "1          زمان     2917.0\n",
       "467        جلسه     2903.0\n",
       "575      جمهوری     2901.0\n",
       "3410        ملت     2867.0\n",
       "860      مدیریت     2817.0\n",
       "395       مرکزی     2808.0\n",
       "273        مالی     2806.0\n",
       "573      برگزار     2806.0\n",
       "738        صنعت     2777.0\n",
       "54        حمایت     2756.0\n",
       "704       اصلاح     2752.0"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dictionary\n",
    "di = {\n",
    "    'word':[*dictionary],\n",
    "    'frequency':list(dictionary.values())\n",
    "}\n",
    "cbow = pd.DataFrame.from_dict(di)\n",
    "\n",
    "cbow = cbow.sort_values(by='frequency',ascending=False)\n",
    "cbow = cbow[cbow.word != '']\n",
    "dims = cbow[:100]\n",
    "dims[50:100]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ahmad/myenv/lib/python3.6/site-packages/pandas/core/generic.py:4405: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "  self[name] = value\n"
     ]
    }
   ],
   "source": [
    "dim1 = dimFinder(corpus[corpus.cat=='اجتماعی'])\n",
    "dim2 = dimFinder(corpus[corpus.cat=='اقتصادی'])\n",
    "dim3 = dimFinder(corpus[corpus.cat=='سیاسی'])\n",
    "dimAll = dimFinder(corpus)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "frequentWord = (set(dim1['word'])&(set(dim2['word']))&(set(dim3['word'])))\n",
    "d = dimAll.word\n",
    "f = dimAll[d.isin(list(frequentWord))]\n",
    "# f.shape[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>word</th>\n",
       "      <th>frequency</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1665</th>\n",
       "      <td>اسلامی</td>\n",
       "      <td>10936.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>آمریکا</td>\n",
       "      <td>9219.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>97</th>\n",
       "      <td>نفت</td>\n",
       "      <td>7476.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36</th>\n",
       "      <td>قیمت</td>\n",
       "      <td>6904.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>166</th>\n",
       "      <td>افزایش</td>\n",
       "      <td>6842.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>105</th>\n",
       "      <td>شرکت</td>\n",
       "      <td>6787.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>84</th>\n",
       "      <td>هزار</td>\n",
       "      <td>6422.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>282</th>\n",
       "      <td>تحریم</td>\n",
       "      <td>5536.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>87</th>\n",
       "      <td>بازار</td>\n",
       "      <td>5187.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41</th>\n",
       "      <td>درصد</td>\n",
       "      <td>5164.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>608</th>\n",
       "      <td>نظام</td>\n",
       "      <td>4952.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>51</th>\n",
       "      <td>کاهش</td>\n",
       "      <td>4812.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>453</th>\n",
       "      <td>ماه</td>\n",
       "      <td>4680.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1001</th>\n",
       "      <td>اجتماعی</td>\n",
       "      <td>4625.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2344</th>\n",
       "      <td>دانشگاه</td>\n",
       "      <td>4573.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>52</th>\n",
       "      <td>تولید</td>\n",
       "      <td>4545.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>ارز</td>\n",
       "      <td>4528.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>66</th>\n",
       "      <td>حضور</td>\n",
       "      <td>4369.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>118</th>\n",
       "      <td>بانک</td>\n",
       "      <td>4303.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>399</th>\n",
       "      <td>اقتصاد</td>\n",
       "      <td>4270.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2484</th>\n",
       "      <td>شهر</td>\n",
       "      <td>4055.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>248</th>\n",
       "      <td>میلیون</td>\n",
       "      <td>4006.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>333</th>\n",
       "      <td>تومان</td>\n",
       "      <td>3995.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>526</th>\n",
       "      <td>استان</td>\n",
       "      <td>3950.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1226</th>\n",
       "      <td>سیاسی</td>\n",
       "      <td>3834.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>161</th>\n",
       "      <td>دست</td>\n",
       "      <td>3718.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4498</th>\n",
       "      <td>اظهار</td>\n",
       "      <td>3715.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>624</th>\n",
       "      <td>شرایط</td>\n",
       "      <td>3554.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>556</th>\n",
       "      <td>مشکلات</td>\n",
       "      <td>3522.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>135</th>\n",
       "      <td>اجرای</td>\n",
       "      <td>3512.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>572</th>\n",
       "      <td>هفته</td>\n",
       "      <td>2603.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>صادرات</td>\n",
       "      <td>2568.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>221</th>\n",
       "      <td>پرداخت</td>\n",
       "      <td>2524.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>140</th>\n",
       "      <td>اقدام</td>\n",
       "      <td>2521.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4055</th>\n",
       "      <td>شهرداری</td>\n",
       "      <td>2512.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>83</th>\n",
       "      <td>بورس</td>\n",
       "      <td>2471.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>887</th>\n",
       "      <td>تلاش</td>\n",
       "      <td>2471.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>267</th>\n",
       "      <td>جنگ</td>\n",
       "      <td>2457.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>44</th>\n",
       "      <td>مسکن</td>\n",
       "      <td>2434.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1215</th>\n",
       "      <td>توان</td>\n",
       "      <td>2427.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8051</th>\n",
       "      <td>تصریح</td>\n",
       "      <td>2422.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1774</th>\n",
       "      <td>آبان</td>\n",
       "      <td>2399.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>440</th>\n",
       "      <td>جهانی</td>\n",
       "      <td>2379.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>فعالیت</td>\n",
       "      <td>2362.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>359</th>\n",
       "      <td>مختلف</td>\n",
       "      <td>2345.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>539</th>\n",
       "      <td>هزینه</td>\n",
       "      <td>2331.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>400</th>\n",
       "      <td>پاسخ</td>\n",
       "      <td>2298.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>197</th>\n",
       "      <td>فروش</td>\n",
       "      <td>2276.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3901</th>\n",
       "      <td>الله</td>\n",
       "      <td>2272.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1749</th>\n",
       "      <td>عمومی</td>\n",
       "      <td>2265.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>198</th>\n",
       "      <td>دستگاه</td>\n",
       "      <td>2245.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>806</th>\n",
       "      <td>آب</td>\n",
       "      <td>2225.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>181</th>\n",
       "      <td>جهان</td>\n",
       "      <td>2215.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1262</th>\n",
       "      <td>المللی</td>\n",
       "      <td>2209.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>562</th>\n",
       "      <td>کمک</td>\n",
       "      <td>2200.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>208</th>\n",
       "      <td>کشورهای</td>\n",
       "      <td>2194.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>408</th>\n",
       "      <td>سامانه</td>\n",
       "      <td>2183.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>165</th>\n",
       "      <td>اطلاعات</td>\n",
       "      <td>2173.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>96</th>\n",
       "      <td>عرضه</td>\n",
       "      <td>2173.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>470</th>\n",
       "      <td>تصویب</td>\n",
       "      <td>2145.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>100 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         word  frequency\n",
       "1665   اسلامی    10936.0\n",
       "35     آمریکا     9219.0\n",
       "97        نفت     7476.0\n",
       "36       قیمت     6904.0\n",
       "166    افزایش     6842.0\n",
       "105      شرکت     6787.0\n",
       "84       هزار     6422.0\n",
       "282     تحریم     5536.0\n",
       "87      بازار     5187.0\n",
       "41       درصد     5164.0\n",
       "608      نظام     4952.0\n",
       "51       کاهش     4812.0\n",
       "453       ماه     4680.0\n",
       "1001  اجتماعی     4625.0\n",
       "2344  دانشگاه     4573.0\n",
       "52      تولید     4545.0\n",
       "3         ارز     4528.0\n",
       "66       حضور     4369.0\n",
       "118      بانک     4303.0\n",
       "399    اقتصاد     4270.0\n",
       "2484      شهر     4055.0\n",
       "248    میلیون     4006.0\n",
       "333     تومان     3995.0\n",
       "526     استان     3950.0\n",
       "1226    سیاسی     3834.0\n",
       "161       دست     3718.0\n",
       "4498    اظهار     3715.0\n",
       "624     شرایط     3554.0\n",
       "556    مشکلات     3522.0\n",
       "135     اجرای     3512.0\n",
       "...       ...        ...\n",
       "572      هفته     2603.0\n",
       "5      صادرات     2568.0\n",
       "221    پرداخت     2524.0\n",
       "140     اقدام     2521.0\n",
       "4055  شهرداری     2512.0\n",
       "83       بورس     2471.0\n",
       "887      تلاش     2471.0\n",
       "267       جنگ     2457.0\n",
       "44       مسکن     2434.0\n",
       "1215     توان     2427.0\n",
       "8051    تصریح     2422.0\n",
       "1774     آبان     2399.0\n",
       "440     جهانی     2379.0\n",
       "12     فعالیت     2362.0\n",
       "359     مختلف     2345.0\n",
       "539     هزینه     2331.0\n",
       "400      پاسخ     2298.0\n",
       "197      فروش     2276.0\n",
       "3901     الله     2272.0\n",
       "1749    عمومی     2265.0\n",
       "198    دستگاه     2245.0\n",
       "806        آب     2225.0\n",
       "181      جهان     2215.0\n",
       "1262   المللی     2209.0\n",
       "562       کمک     2200.0\n",
       "208   کشورهای     2194.0\n",
       "408    سامانه     2183.0\n",
       "165   اطلاعات     2173.0\n",
       "96       عرضه     2173.0\n",
       "470     تصویب     2145.0\n",
       "\n",
       "[100 rows x 2 columns]"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dimAll = dimFinder(corpus=corpus,dimNum=100+f.shape[0])\n",
    "dimAll = dimAll[~dimAll.word.isin(list(frequentWord))]\n",
    "# dimAll.shape\n",
    "dimAll"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "# _max = dimAll.frequency.max()\n",
    "# dimAll.frequency = dimAll.frequency/_max\n",
    "# cbow[:200].frequency.plot.bar(by='frequency',figsize=(25,10))\n",
    "# plt.show()\n",
    "# dimAll"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "vector_element = np.array(dimAll.word)\n",
    "vectors_weight = np.array(dimAll.frequency)\n",
    "# type(dims.frequency[18])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "vector_dict = [(i[1].word,i[1].frequency) for i in dims.iterrows() ]\n",
    "vector_dict = dict(vector_dict)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "import sklearn as sk\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Learn vector for NEWS"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "corpus['corpusAll'] = corpus.title*3 + corpus.summery*2 + corpus.body\n",
    "# corpus['corpusAll']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 385,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "def vectorComputer(doc):\n",
    "    points = []\n",
    "    for word in dimAll.iterrows():\n",
    "        points = [doc.count(i) for i in dimAll.word]\n",
    "        \n",
    "    return(points)\n",
    "vectors = pd.DataFrame(list(map(vectorComputer,corpus.corpusAll)))\n",
    "# idf(corpusAll[2500])\n",
    "# 2/pd.DataFrame([12,12])\n",
    "vectors.to_csv('./DataSets/Vectors.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## TF-IDF"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- TF  = freq(word)/len(doc)\n",
    "- -\n",
    "- IDF = (NumberOfDoc)/(NumberOfDocHasThisWord)\n",
    "- -\n",
    "- w = TF * IDF"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 216,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>...</th>\n",
       "      <th>90</th>\n",
       "      <th>91</th>\n",
       "      <th>92</th>\n",
       "      <th>93</th>\n",
       "      <th>94</th>\n",
       "      <th>95</th>\n",
       "      <th>96</th>\n",
       "      <th>97</th>\n",
       "      <th>98</th>\n",
       "      <th>99</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.025716</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.010716</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.016797</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.009903</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.008195</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.00247</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00903</td>\n",
       "      <td>0.009948</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.018942</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.009870</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.008226</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.006447</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.035322</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.030699</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.018288</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 100 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         0         1    2        3         4         5    6    7         8  \\\n",
       "0  0.00000  0.000000  0.0  0.00000  0.025716  0.000000  0.0  0.0  0.000000   \n",
       "1  0.00000  0.000000  0.0  0.00000  0.000000  0.000000  0.0  0.0  0.000000   \n",
       "2  0.00247  0.000000  0.0  0.00903  0.009948  0.000000  0.0  0.0  0.018942   \n",
       "3  0.00000  0.000000  0.0  0.00000  0.009870  0.000000  0.0  0.0  0.000000   \n",
       "4  0.00000  0.035322  0.0  0.00000  0.000000  0.030699  0.0  0.0  0.000000   \n",
       "\n",
       "          9 ...    90        91   92   93   94   95        96        97   98  \\\n",
       "0  0.010716 ...   0.0  0.000000  0.0  0.0  0.0  0.0  0.016797  0.000000  0.0   \n",
       "1  0.000000 ...   0.0  0.009903  0.0  0.0  0.0  0.0  0.000000  0.008195  0.0   \n",
       "2  0.000000 ...   0.0  0.000000  0.0  0.0  0.0  0.0  0.000000  0.000000  0.0   \n",
       "3  0.008226 ...   0.0  0.000000  0.0  0.0  0.0  0.0  0.006447  0.000000  0.0   \n",
       "4  0.000000 ...   0.0  0.018288  0.0  0.0  0.0  0.0  0.000000  0.000000  0.0   \n",
       "\n",
       "    99  \n",
       "0  0.0  \n",
       "1  0.0  \n",
       "2  0.0  \n",
       "3  0.0  \n",
       "4  0.0  \n",
       "\n",
       "[5 rows x 100 columns]"
      ]
     },
     "execution_count": 216,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "vectors = pd.read_csv('./DataSets/Vectors.csv',index_col='Unnamed: 0')\n",
    "idf = np.log(corpus.shape[0]/\\\n",
    "             pd.DataFrame([vectors[vectors[i]!=0].shape[0] for i in list(vectors.columns)]))\n",
    "\n",
    "idf =  pd.Series(list(idf[0]))\n",
    "# for i in range(vectors.shape[0]):\n",
    "# vectorUpdated = pd.DataFrame()\n",
    "for i in range(vectors.shape[0]):\n",
    "    tf = vectors.iloc[i] / len(corpus.corpusAll.iloc[i])\n",
    "    tf = pd.Series(list(tf))\n",
    "    vectors.iloc[i] = list(idf * pd.Series(list(tf)))\n",
    "vectors.head()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn import svm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 339,
   "metadata": {},
   "outputs": [],
   "source": [
    "d = []\n",
    "for i in vectors.iloc[:].iterrows():\n",
    "#     g = \n",
    "#     print(list(i))list(i[1])\n",
    "    d.append(list(i[1]))\n",
    "# type(g)\n",
    "# np.array(g)\n",
    "# len(d[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 347,
   "metadata": {},
   "outputs": [],
   "source": [
    "len(d)\n",
    "len(corpus.cat.tolist())\n",
    "tmp = corpus.cat\n",
    "def l(x):\n",
    "    if x == 'اقتصادی':\n",
    "        return 1\n",
    "    elif x == 'اجتماعی':\n",
    "        return 2\n",
    "    elif x == 'سیاسی':\n",
    "        return 3\n",
    "    \n",
    "lable = tmp.apply(l)\n",
    "# lable = np.array(lable)\n",
    "# d = np.array(d)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 391,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SVC(C=1.0, cache_size=200, class_weight=None, coef0=0.0,\n",
       "  decision_function_shape='ovr', degree=3, gamma='auto', kernel='rbf',\n",
       "  max_iter=-1, probability=False, random_state=None, shrinking=True,\n",
       "  tol=0.001, verbose=False)"
      ]
     },
     "execution_count": 391,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf = svm.SVC()\n",
    "clf.fit(d,lable)\n",
    "# clf.fit(np.array([[1],[2],[3]]),np.array([1,2,3]))\n",
    "# len(lable[1:3])\n",
    "# len(d[1:3])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 397,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 397,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "list(clf.predict(X=d)).count(1)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 386,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0       1\n",
       "1       1\n",
       "2       1\n",
       "3       1\n",
       "4       1\n",
       "5       1\n",
       "6       1\n",
       "7       1\n",
       "8       1\n",
       "9       1\n",
       "10      1\n",
       "11      1\n",
       "12      1\n",
       "13      1\n",
       "14      1\n",
       "15      1\n",
       "16      1\n",
       "17      1\n",
       "18      1\n",
       "19      1\n",
       "20      1\n",
       "21      1\n",
       "22      1\n",
       "23      1\n",
       "24      1\n",
       "25      1\n",
       "26      1\n",
       "27      1\n",
       "28      1\n",
       "29      1\n",
       "       ..\n",
       "1820    2\n",
       "1821    2\n",
       "1822    2\n",
       "1823    2\n",
       "1824    2\n",
       "1825    2\n",
       "1826    2\n",
       "1827    2\n",
       "1828    2\n",
       "1829    2\n",
       "1830    2\n",
       "1831    2\n",
       "1832    2\n",
       "1833    2\n",
       "1834    2\n",
       "1835    2\n",
       "1836    2\n",
       "1837    2\n",
       "1838    2\n",
       "1839    2\n",
       "1840    2\n",
       "1841    2\n",
       "1842    2\n",
       "1843    2\n",
       "1844    2\n",
       "1845    2\n",
       "1846    2\n",
       "1847    2\n",
       "1848    2\n",
       "1849    2\n",
       "Name: cat, Length: 6639, dtype: int64"
      ]
     },
     "execution_count": 386,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lable"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "vec_spc = []\n",
    "def vecCalculate():\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 360,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "vec_spc = []\n",
    "for txt in corpus.iterrows():\n",
    "    txt = txt[1]\n",
    "    tmp_body  = [txt.body.count(i) for i in dims.word]\n",
    "    tmp_title = [txt.title.count(i) for i in dims.word]\n",
    "    tmp_summ  = [txt.summery.count(i) for i in dims.word]\n",
    "    f = [tmp_body[i] + tmp_summ[i]*2 + tmp_title[i]*3 for i in range(len(tmp_body))]\n",
    "    vec_spc.append(f)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 361,
   "metadata": {},
   "outputs": [],
   "source": [
    "v = pd.DataFrame(vec_spc)\n",
    "t = v[:100]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 362,
   "metadata": {},
   "outputs": [],
   "source": [
    "d = []\n",
    "tmp = []\n",
    "for i in t.iterrows():\n",
    "    tmp = []\n",
    "    for j in t.iterrows():\n",
    "        tmp.append(dis.cosine(j[1],i[1]))\n",
    "    d.append(tmp)\n",
    "r = np.array(r)\n",
    "r[r==0] = 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 375,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([ 6, 10, 16, 20, 24, 24, 32, 32, 37, 42]),\n",
       " array([16, 20,  6, 10, 32, 37, 24, 42, 24, 32]))"
      ]
     },
     "execution_count": 375,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.where(r<.36)\n",
    "# r.min()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 132,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>body</th>\n",
       "      <th>cat</th>\n",
       "      <th>date</th>\n",
       "      <th>subcat</th>\n",
       "      <th>summery</th>\n",
       "      <th>teller</th>\n",
       "      <th>time</th>\n",
       "      <th>title</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>\\nبه گزارش خبرگزاری فارس به نقل از کنفدراسیون ...</td>\n",
       "      <td>اقتصادی</td>\n",
       "      <td>۱۳۹۷/۷/۱</td>\n",
       "      <td>\\r\\nصنعت ، تجارت ، بازرگانی\\r\\n</td>\n",
       "      <td>\\r\\nبر اساس تصمیم وزیر صنعت،‌معدن و تجارت مدت ...</td>\n",
       "      <td>بر اساس تصمیم وزیر صنعت،‌معدن و تجارت</td>\n",
       "      <td>۱۸:۴۸</td>\n",
       "      <td>مدت زمان بازگشت ارز حاصل از صادرات فرش 7 ماهه شد</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>\\nبه گزارش خبرگزاری فارس به نقل از روابط عمومی...</td>\n",
       "      <td>اقتصادی</td>\n",
       "      <td>۱۳۹۷/۷/۱</td>\n",
       "      <td>\\r\\nصنعت ، تجارت ، بازرگانی\\r\\n</td>\n",
       "      <td>\\r\\nمرکز توسعه تجارت الکترونیکی با صدور اطلاعی...</td>\n",
       "      <td>از سوی مرکز توسعه تجارت الکترونیکی منتشر شد</td>\n",
       "      <td>۱۸:۰۹</td>\n",
       "      <td>اطلاعیه‌ دیرهنگام مبنی برغیرقانونی بودن فعالیت...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>\\nبه گزارش خبرنگار اقتصادی خبرگزاری فارس، در ح...</td>\n",
       "      <td>اقتصادی</td>\n",
       "      <td>۱۳۹۷/۷/۱</td>\n",
       "      <td>\\r\\nصنعت ، تجارت ، بازرگانی\\r\\n</td>\n",
       "      <td>\\r\\nدر حالی که قرار بوددر هفته جاریاستیضاح شری...</td>\n",
       "      <td>فارس گزارش می‌دهد</td>\n",
       "      <td>۱۷:۳۳</td>\n",
       "      <td>دو راهی استیضاح و استعفای شریعتمداری</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>\\nبه گزارش خبرنگار اقتصادی خبرگزاری فارس، براس...</td>\n",
       "      <td>اقتصادی</td>\n",
       "      <td>۱۳۹۷/۷/۱</td>\n",
       "      <td>\\r\\nراه و مسکن\\r\\n</td>\n",
       "      <td>\\r\\nآسمان تهران فردا نیمه‌ابری در برخی ساعات ر...</td>\n",
       "      <td>فارس از وضعیت آب‌‌و‌‌هوا گزارش می‌دهد</td>\n",
       "      <td>۱۷:۱۵</td>\n",
       "      <td>وزش باد شدید و رعدوبرق در پایتخت/هشدار وقوع سی...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>\\nبه گزارش خبرگزاری فارس به نقل از انگجت، گوگل...</td>\n",
       "      <td>اقتصادی</td>\n",
       "      <td>۱۳۹۷/۷/۱</td>\n",
       "      <td>\\r\\nارتباطات و فن آوری اطلاعات\\r\\n</td>\n",
       "      <td>\\r\\nکاربران گوشی‌های پیکسل گوگل که به دنبال وق...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>۱۷:۱۲</td>\n",
       "      <td>تعمیر رایگان گوشی‌های آسیب‌دیده در توفان سهمگی...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                body      cat      date  \\\n",
       "0  \\nبه گزارش خبرگزاری فارس به نقل از کنفدراسیون ...  اقتصادی  ۱۳۹۷/۷/۱   \n",
       "1  \\nبه گزارش خبرگزاری فارس به نقل از روابط عمومی...  اقتصادی  ۱۳۹۷/۷/۱   \n",
       "2  \\nبه گزارش خبرنگار اقتصادی خبرگزاری فارس، در ح...  اقتصادی  ۱۳۹۷/۷/۱   \n",
       "3  \\nبه گزارش خبرنگار اقتصادی خبرگزاری فارس، براس...  اقتصادی  ۱۳۹۷/۷/۱   \n",
       "4  \\nبه گزارش خبرگزاری فارس به نقل از انگجت، گوگل...  اقتصادی  ۱۳۹۷/۷/۱   \n",
       "\n",
       "                               subcat  \\\n",
       "0     \\r\\nصنعت ، تجارت ، بازرگانی\\r\\n   \n",
       "1     \\r\\nصنعت ، تجارت ، بازرگانی\\r\\n   \n",
       "2     \\r\\nصنعت ، تجارت ، بازرگانی\\r\\n   \n",
       "3                  \\r\\nراه و مسکن\\r\\n   \n",
       "4  \\r\\nارتباطات و فن آوری اطلاعات\\r\\n   \n",
       "\n",
       "                                             summery  \\\n",
       "0  \\r\\nبر اساس تصمیم وزیر صنعت،‌معدن و تجارت مدت ...   \n",
       "1  \\r\\nمرکز توسعه تجارت الکترونیکی با صدور اطلاعی...   \n",
       "2  \\r\\nدر حالی که قرار بوددر هفته جاریاستیضاح شری...   \n",
       "3  \\r\\nآسمان تهران فردا نیمه‌ابری در برخی ساعات ر...   \n",
       "4  \\r\\nکاربران گوشی‌های پیکسل گوگل که به دنبال وق...   \n",
       "\n",
       "                                         teller   time  \\\n",
       "0         بر اساس تصمیم وزیر صنعت،‌معدن و تجارت  ۱۸:۴۸   \n",
       "1  از سوی مرکز توسعه تجارت الکترونیکی منتشر شد   ۱۸:۰۹   \n",
       "2                             فارس گزارش می‌دهد  ۱۷:۳۳   \n",
       "3         فارس از وضعیت آب‌‌و‌‌هوا گزارش می‌دهد  ۱۷:۱۵   \n",
       "4                                           NaN  ۱۷:۱۲   \n",
       "\n",
       "                                               title  \n",
       "0   مدت زمان بازگشت ارز حاصل از صادرات فرش 7 ماهه شد  \n",
       "1  اطلاعیه‌ دیرهنگام مبنی برغیرقانونی بودن فعالیت...  \n",
       "2               دو راهی استیضاح و استعفای شریعتمداری  \n",
       "3  وزش باد شدید و رعدوبرق در پایتخت/هشدار وقوع سی...  \n",
       "4  تعمیر رایگان گوشی‌های آسیب‌دیده در توفان سهمگی...  "
      ]
     },
     "execution_count": 132,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "corpus_original.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 372,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>whoTell</th>\n",
       "      <th>title</th>\n",
       "      <th>summery</th>\n",
       "      <th>body</th>\n",
       "      <th>date</th>\n",
       "      <th>day</th>\n",
       "      <th>time</th>\n",
       "      <th>year</th>\n",
       "      <th>month</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>معاون وزیر انرژی روسیه:</td>\n",
       "      <td>روسیه قادر نیست کمبود انرژی بازار آسیایی در اث...</td>\n",
       "      <td>معاون وزیر انرژی روسیه گفت: روسیه واقعا قادر ن...</td>\n",
       "      <td>\\nبه گزارش گروه اقتصاد بین‌الملل خبرگزاری فارس...</td>\n",
       "      <td>1397/07/10 :: 17:01</td>\n",
       "      <td>10</td>\n",
       "      <td>17</td>\n",
       "      <td>1397</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>یک کارشناس نفتی اعلام کرد</td>\n",
       "      <td>پیشتازی نفت به سمت 100 دلار/عربستان نتوانست خو...</td>\n",
       "      <td>یک کارشناس نفتی گفت: احتمال زیاد قمیت نفت برای...</td>\n",
       "      <td>\\nبه گزارش خبرنگار اقتصاد بین‌الملل خبرگزاری ف...</td>\n",
       "      <td>1397/07/10 :: 14:52</td>\n",
       "      <td>10</td>\n",
       "      <td>14</td>\n",
       "      <td>1397</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                      whoTell  \\\n",
       "10    معاون وزیر انرژی روسیه:   \n",
       "20  یک کارشناس نفتی اعلام کرد   \n",
       "\n",
       "                                                title  \\\n",
       "10  روسیه قادر نیست کمبود انرژی بازار آسیایی در اث...   \n",
       "20  پیشتازی نفت به سمت 100 دلار/عربستان نتوانست خو...   \n",
       "\n",
       "                                              summery  \\\n",
       "10  معاون وزیر انرژی روسیه گفت: روسیه واقعا قادر ن...   \n",
       "20  یک کارشناس نفتی گفت: احتمال زیاد قمیت نفت برای...   \n",
       "\n",
       "                                                 body                 date  \\\n",
       "10  \\nبه گزارش گروه اقتصاد بین‌الملل خبرگزاری فارس...  1397/07/10 :: 17:01   \n",
       "20  \\nبه گزارش خبرنگار اقتصاد بین‌الملل خبرگزاری ف...  1397/07/10 :: 14:52   \n",
       "\n",
       "   day time  year month  \n",
       "10  10   17  1397     7  \n",
       "20  10   14  1397     7  "
      ]
     },
     "execution_count": 372,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "h = np.where(r==r.min())[0]\n",
    "corpus_original.iloc[h]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 309,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAP4AAAD8CAYAAABXXhlaAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4wLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvqOYd8AAAIABJREFUeJztnXucldV193+r4KigeMmATkbJIIGIRMR4tOhEQ6S8JUrEmNgoyhvvqFVJjNeGErEkeKEqiWJRg1DrLUZSDVrzEiONrw3qCSKKICJODBRkMFKtMQ7q6h/nIGf/1sIZbmeG7vX9fPw4a5+9n7PPc57Nc/Z61votUVUEQZAXf9HeEwiCoPrEwg+CDImFHwQZEgs/CDIkFn4QZEgs/CDIkFj4QZAhsfCDIEO2aOGLyDAReVlElorIFVtrUkEQbFtkcyP3RKQTgCUAhgJYDuBZACer6ksbG1NbW6sNDQ0VLeucXjyfzmS/44x5g+xPO3343zg+rjeXLmQ3kV3vjGkme1ey33fGfET2n8ju7ozh+fLn8frUkP3fzpjdyObvg88JAHxA9iKyP+eM4fPQ1enzbivvw58HwJ8Xp/ZOfakDn2tvLmzv5IzZhewWsr119SbZzvwhZPP1452n33/8V1PTn7Bmzft8EIN3tbSVwwAsVdVlACAi9wEYAWCjC7+hoQHFYrGiZZXT689k15L9786YyWSPc/rwF8XHXe6MGUj2GWRf54yZQvaXyG5yxvAC/B3Z5ztj+B+73Z0+fH4/Q/ZTzphjyebvo+CMWUP2IWQ/6oxZ2objFsnmz8yfB8CSw1O7L38f/Hm8ubxKdn9nTCPZvyeb/5ECgBlk7+P04X9k+Pr5S2fMOR//VSg84bxu2ZKf+vUA/lBhL4d/CwyCoIOxzZ17InKOiBRFpNjczD+DgyBoD7bkp/4KAPtW2PuU2xJU9TYAtwFAoXCQpj8/93YO25SaT9Eep/FkZ8wIsr2fvWem5qnPpPb3nCH9plMD/Zy7bF8YRpPd+2+owfupyT/t+acm/4wE8N6pqe25KPgX9l+RXfsVO+Yp8tE2nkcdeJ8K4PVhqd1zFHXgLRJgftLe8wvb5fNkD/gaNfzQjun7K2qg8z1/uB3Dt78BfIO6wY659tzU5qn1vcqOmfbz1D7jMNsHV5PN25mfOGMqz8M3nNctW3LHfxZAHxHpJSI1AE4C8PAWHC8Igiqx2Xd8Vf1ARC4A8EsAnQBMU9WFW21mQRBsM7bkpz5U9VH4LtsgCDowEbkXBBmy2QE8m0OhMECLxdbcAA1k8zPWXzpj+Dmr44AyQST8Y4ef6wNnSRo4c4eSB2cJOWsAoO8R1PApsjk2AAAOIpsdOt7z3qbEukQONz0m6ZPUkjqpVOz8Rfk5MHvYbBzFDZI6lC7WldTjeTMG95FD8KRJtg+/96s0pje/DwDQbvMu8miO+q0z5vLUfP03qX2bM2TCc9TA15MXI/FZsjn+BAC+npoHnpaYLS/aETW6IaitUDgKxeK8VgN44o4fBBkSCz8IMiQWfhBkyBZ59Tfv7Sr20hycAwCNr1AD74s4wAEw+zoTLw8YX4H2Se06O+IOnUAtFFjzU+dtxnJ8/CNkc84AgHt2SO2RHITCAT0A7/En/dHpYgJc/jqxRC+zQ44kX8GT5CeYZgNELv4Bt6TnoEGGcQc03UoNR1xi58LuoN78fTi+A9AmeBQH43hR5T9KzZ4PpfaEp+2Q+Qen9j30+nUc0QNgHAX9XONMpYWCz16Ynpg17pKtnB8nNvnEHT8IMiQWfhBkSCz8IMiQKj/H76PF4k0VLXc7vTjPmfdX/+yMOY5sL2eZ8705ecbbsHNCBO+v2P8AAPPJ5j2+s18082U/wOPOGM6Hch7wYgjZvBF15rKGkpdq/5U6eBoKPL+/Jpv20ACwYkFqT3QOe/MsauCYDr5WAPs9slYAPScHYLUY+DvzhDi4D2szNDhjRrZhLuzP4RgPx0dUkftfKPwzisVV8Rw/CAJLLPwgyJBY+EGQIbHwgyBDqhzA8wbSxARWzgFswg0H57AjD7CRHuOdPix+yI6Vv3PGkMPmKUr4aLzUGfPNVmxW2wFMZvN5FPByqyPcOP/41B7ozYXPCzuXHKWiWnaYrXWOy5C60SKafz9W8QFQT8lLN3sONFYr4uQrz9HFzkcWKjUiUQAuSs39SDF32VHOGA4SayKbBUcBvPZ2avfyBF75O2L1H6sGNEY2qAr9wbzqE3f8IMiQWPhBkCGx8IMgQ6q8x/800mIXnhruy2TzXspkhMDu6b9velwuaUzDtdqTenjBFE2p2fiz1H7AUTQ98QVqYOEQu19/kpJYjlSq/vKIFdnAsf9EDV5gE4uAsB/DE0XZi2zeM3vKyCekZr+b6HXv3HLAy/91+rA4CgfjeMFEfEmzQjEHxADAtNRcxgFIXtAVnycWS3GWVi8+LxwEBNjr5cHEekEONCMm62sf//1UwfOBWeKOHwQZEgs/CDIkFn4QZEgs/CDIkCo79/4CadDFmU4fzohrS1XV1GHGjjwAuNZkIXIVUqdMkpkLOZecClq28i1n8NlMO1vzlQJTPB+oyXhznFZz703tQfx1e1l/7HDic3CKM4bPHQevsMMWsIEqnjIyB9uQEw6klAPAOk/587DjDrBBSp6SM8NKPuzs8xyCPBdWMAasGnQaxHSgV3E8uV7adi+PO34QZEgs/CDIkFj4QZAh7auyy6WqAeBfaEqshivdnOOmASI2OAewe3qqCLPQES3pz0qppI7yDhx4b8d7SlvmuKeSqu5YSgaa8JjzPrRHXvGS7cJiQIMWp/Zc5/wPOpYajknNx0gRGAC+SPYu37V9DF3JfsPpwwpN48i21Y+sf4ETebwKNw+STaVzWugaBICaO6iBHTFeoBD7ZbyEIT7/5LNYTMFdANJAsxbndUvc8YMgQ2LhB0GGxMIPggyp8h5/HRJF0+95fWjfxhVuVnlquJx84iWF0LNm3tP399SG7//kuc2ivTgAHPTt1O7BjoCZzvv8Y2qaUAX7pN8k5dTzPhXAubwHJn/DIFttdr6kJ3ygUtLHsKHOXDgRhp6L773EDrmebG9LvDdVq+nxGnXwlIU5WYb2zC072yE1LGJCx6jh9wVwV6/U5lCAC7nqMIDZpCzc3XbBwN2ogcVIrI8ojR350HndEnf8IMiQWPhBkCGtLnwRmSYiq0XkxYq2PUVktoi8Uv7/Htt2mkEQbE3acsefDoDLnV4B4HFV7YNSwPcVW3leQRBsQ1p17qnqb0SkgZpHABhc/nsGgDkALm/97boAGLjB7Dfd9DhLUo+HLVXtOTdYzaXJ6UPjTHAOO/IAo5D7O3IIepW6enCAEWdVeEo5pGpzPAeHsIMH4IShqXKn6TGaBIPw9bQk1QqxtcEHKiv7cBkuO//vS+q8G6+kQrSKA28APPLz1B7gqQRTUtRqcqj1WOSMIRXjp0glyVVGJrXbt8gBeJUzZDI3NqTmnC/bMUP/nhrmOgemQKYxw1Oby4sDQEtlKXDHeemwuXv8vVR1vUt4FWxqUhAEHZgtdu5pqermRitvisg5IlIUkWJzc/PGugVBUEU2d+G/IVL6nVj+/+qNdVTV21S1oKqF7t29B5dBEFSbzQ3geRjAtwBcU/6/p4jg0ATgjAqbAz+AO7SVxBhvyqbCDW9uAavQysd1Ej54T38I/bA5hKv+ADZJhJMuPDVZEn54i5RU9+BS1QBXiBmt3r/hLPSQikfU93OGmKQirvxjK+uMV/JrrKD5119o32YgN3gRPCR20YODlCjRCoBJwmkcQK87Qhwn0PXDX5H3NqYKFCXcUD4UAGDwLdTAVZYAkzDEMWKT+fMAeK/ihvqRc0iHtjzOuxfAbwF8TkSWi8iZKC34oSLySnlq17Tt7YIg6Ai0xat/8kZeYndvEATbCRG5FwQZImpEKLcdhcIALRYrKsNe5qhVnkU25+SM5WehgHnO/QBngMAKY3LuzCxY+Dn9yAOogSv5Ak0k9NmgV1EPJ4HoGdoTH8ZCIpx4AgAXp+btTlUfnm4D2d5jZD69rMdZ9wU75oR5qT2TfiSy6Cdgix2xLgcA/AfZq3pQgz3/Jh7j8QtS+zRnCBfzpSGwIRLAbLI5X6jJEU/5AsXB3e4c95AnU/s7R6a2J7Z57ob3KhQuRLG4xFGVSYk7fhBkSCz8IMiQWPhBkCGx8IMgQ6qswNOMpOz1aKdL7yNSeyyXQmbpWMAEQphS1YCtcENquKycAzgJN2lwDjvyAKDBOEs5MIVVXwEcxgovP2zlGIBJLDnbUx++LDW7j01tL96SH96aKj5OrNZMVgiiijGDnMSYteSAtdWfMZTyeGabyktcZQkwEx5Sk5jnL7cqtFNYgLkbJdhcZNV0LqfLhc60KVAOAB8+l9qdDjnR6UXXxz/wy5xEBaTXZRfndUvc8YMgQ2LhB0GGxMIPggyp8h5/VyQVbXr/jdOHRRu4wqi3e+JkB6/aKVetpWQNo4YL2GiJNOGmQXs7Y3g//jzZRztjOFKIo6E5wQiwn+dipw8FCzUPotepSg4AtHDEFEfwsK8EsOefE56c95lCe/xDfmu6zJ7EATrnk32OMxcSZVmQ7umnqJMwZKofjUlN4YQu4FrlZB+uCGx9UZ30Jmpx/D2sBr0LJ61xNWAg/U7alqUTd/wgyJBY+EGQIbHwgyBDqrzHfx+pEKYnIskKDbwP4v0YYMUivH0Q71VZtNOrcMPCkqzQ4PkSeL68p/+1M4Yq6ZhsGhbzALCGhCdrveQl9o9Q4MQap7JqLfs1WITUqzbLlYz4OzKlgQDWEz2ERUMAo4Axgub7kFM9CKek5gAWRmEBDQCjf5zaU8lP0/K6HVPDT+7J9zGHqgABwOBPU4MjCsLCq2A/UqMzZtOJO34QZEgs/CDIkFj4QZAhsfCDIEOq7Nz7CKkThJ1ygAmAuWeH1B7p/Vv1aGI9KVzxyxaa7qm/ohZ2sAHW0ULOPFbOAZyEGw7O8d7nu2RTIsxCcuQBQH+qMKRjbR/hai4UaFPrJXxwIBAFsyyy5xb/SfYQOrdjd7VjbmWFmhnOXMjh99B0et1LXuLML04qouQmAJjKx90pNWtudt6HjzstNQff54yhgKMbHGmfiw+jhotSc6CjWjX/D857fTJxxw+CDImFHwQZEgs/CDKkynv8PyHd19vkB+C41BzJe/ET7JDz0n3nkdrXOS4F1oylEiVOjImpWssVbowaLmBFNDjhpsEZw/tFCjLp/y/OGNoviteHEj5MMM5/OWM40KkhNfvtZ4f0owCee+jcTrjSeR8OhvqR04cTgth2qh+BhVv43HqJMbQ/X3laant5L/X8mbyEM2LR26l9sXOdrnkmtWvJ9zHf8x1UBvmsa30eiDt+EGRJLPwgyJBY+EGQIVXe43dHKqZgq+XaqjHsB+A9KIBbKSnnkcNtHxaNnMDPkflJP2CSiEzVWhJ9AGCfLbOIhpNww8/pzZ6eEk8A2AQc5/k0V5s1/gUvSYqFODgOgffQgInHGMnnyTm3d01M7VGewAd/1+RL+MB5pt2Z982cpONcPxxDUMd+AG8Mn28e43yePtzAwiIAalnUhI6z0hwEqGuuMLi8rk/c8YMgQ2LhB0GGxMIPggyJhR8EGVJl5946AG9sMN871XbZmdVWm8imBAoAmH98ah/rJZ9wgMhIslltB7COOXbGeGPY6UNquKycA9iEG6Mey448ALiH7ONsl3G/SO2reb6eAgyfF3ZSvemMYQcaB1lRog8AjGK1Wy+whtWAyNHb2amYpBRkJXzuSNUHAPQKGjOdOnjluPuTzQlnzufhYk03T7F9DKS4Uzfddnms+4a/37Yve8QdPwgyJBZ+EGRIqwtfRPYVkSdE5CURWSgiY8rte4rIbBF5pfz/Pbb9dIMg2BqImuqu1EGkDkCdqs4TkV1R2swcD+A0AH9U1WtE5AoAe6jq5Z90rELhc1osVuxr3rbBBpfsltqT/kgd9uD9MGCTTTwlWAoiWXFvatefbkZMlVQoYbTSv5O3O9kbpmotC1t8aMcolUQ1CTdecA5/xoedPrzXvjqx3hUrJNJVuYIrVTZaaMfsTXoZq5SEK1ZfYKfG2/Mh3j2I98ms5usp895CNvuEPAVmDmRqSM01t9shtaOogQObnKWgdL3PcqbyVQ5+StV8p8oSM2R0hahMoXA+isWXbRlnotU7vqquVNV55b/fAbAIpZCwEdgQ8jQDpX8MgiDYDtikPb6INAA4GCXh9L1UdWX5pVUA9tqqMwuCYJvR5oUvIrsAeBDAt1U1eWigpf2Cu2cQkXNEpCgixebmtVs02SAItg5tWvgisgNKi/5uVV1fcuaN8v5/vR9gtTdWVW9T1YKqFrp350yZIAjag1YDeEREUKo3tUhVKyVdHgbwLQDXlP/PUicO65A4QR61PSbpk9TSmqLN+qlU4pTSnkvOPK5ifK5Vcxn9M24hZ9IBC5y5cGklKlXtBeMYNVxWzuEsOzjBOY4ykSkLljrquqqnjMPZhZQt2d+cFKxSVp+hElo9nNLUQzjoh88bAJxJNmc2er8gWUqJg4D43AL2M5Ozr3aAM4ZLWdFcxjtZcpxYN/Iq57hLyU4Dz0Yrl37jMV7GpaUtkXuNAEYBeEFE1i+Xv0Npwf9URM5EKb+2DdpDQRB0BFpd+Kr6/wFs7PGAd/sNgqCDE5F7QZAhrQbwbE0Khf21WKxQrl1zpO1U+zVq+BLZnjIvJ5Z4gR3842Yx2V7CCqmv8l57hVMKmat8N7NKMCvaAHavymq4Dc4Y/rHlBfmwms7VZFt/w4cyPLE7KZd29kpTv0s2K+hy4A2ANRTU403/DLLnPUsNlAAFwCbL0J53/+EwLP4KNfB8bWLSs5KGrRzKl+3M9+z7nLAz9eEgIMBUzjFJRY6CU8X1USg0olj83ZYH8ARB8L+PWPhBkCGx8IMgQ6osxPHfSJJLanlvBaj8PLFF+fkuC1AAVkLXS1ghpdS5VLFk0EowK0rxSR9T3486UG4NACeM6ZjUXNNix5iqtZx05D2bTX0S78ozpod9Ts97ertf7KTPUQvvMR3F2WOoUu+jrSndAniJ7KOesH3m0X79kkNTexLPFTD+hdFUyncxVUcCgPFnpfbp/5aYSxyR4ENNteVOZDufeebJ1OBVi345NXe/5JNtAGiqjK1oW3Rs3PGDIENi4QdBhsTCD4IMiYUfBBlSZefebkgcSk9dYXqIkpPnSCqHlfr+StRywoQnDUCOrUGpY2s+OfIAYKCy042CfA5wFHPZf9NCQTS1nDQCWJUedqBxIA7AQUtd1SsBliaf2OAczzlGEUjjKEjJChUBj7LqEDlbT3dKmjWkZsuXOFEJ4HCdRuWgGFZBBkr5YhVMTZWR54gtuzWYfandUkdvX7XO1RGSfvdv0Otz9R0z5lXZNbF7s5gyAEwlp+BaliryPvP+FX97asWWuOMHQYbEwg+CDImFHwQZUuU9viIJRmk8z+lDCTZPsjCHV8mFgxa8Us4sYJAG1gxUpxKNSYShgAuverIRGeJOnIAD2EozDWR7iT1c4Wac0ycV0bAJN05VGd7TX81JXCwUAVjlWqrYc+cvnTEjEqtm3G9Mj8YmbuFz6Z0XvqTTjfRgJ07I5Bh1489jVZsfIl0XmOM+zQ3oTbE56MtBQAA+oCixzs3Ugb0JQFptygkQc4g7fhBkSCz8IMiQWPhBkCFV3uN3QSqG6O3Xad857RupfYZXCZfZ22k7JTUf2yG1hw11xnB1WfIl1H3BGcOao7wXd6r8LBqW2v32ow6ez4LO3UK7R7bCmCyi4TgpzHN63tM74hcPkO7DiZNS+5tOdbX7qVqQJx3RiwVWPLEU4nrKnLqUfBSDZ8DCm2+GqyQDOOns1N6dq+148RpEH0eQ8xV+/s8xBC86B6qMSWnbko47fhBkSCz8IMiQWPhBkCGx8IMgQ6rs3PsASZLB68NMjxso1+TiH3APL2qGK654VWWogsoX+fXfmxHfp5LE45WSUU6YZ99mJifLfJNsR3GWRGLQj/t4Si1pMsfen7fOPVvh5v+lJivnAE7CDQWzsCMPAE7kIB9KiBoByyOnprYXi1PLTk12wnFyE4BLj6AGPgeeQg1XUbouNd96G4Y9yEk7rAt1cIKjOH7qemcq5vpO53+z2ACdC7SyWlOrArsA4o4fBFkSCz8IMiQWfhBkSJUr6eyoxWJlcA1XyQHM/soErzj73UWkktrvJue4rHr6XbK9vR8FvKw4MLXrWXUDsEq2JAxhKrMCJinnHgrsGPmvzhj2Y/zI6cOJIrx/9EQbOMuIE5WcSsSJEARg1Xvn2CErSXijzgpX2ApJXHmYKvkCsElQ/BmdAJjTZ6f2nTWp/R0n8eVGDrLi75XnCuBK8mtMnGT7mKAxdn44zpAVG67/wjFA8XmNSjpBEFhi4QdBhsTCD4IMqfJz/M8BeLTC5nKoAPB8YjVI+qy/SQfYIf1Y0MPZX7WWiLH3Etu2isQt6i9M7bk/tmMGXUoNVEknSVIqMzYVYcQEroDjCWmSeMfqC2yXHjRfE0PgVHthYUwW0fASbvg5/UhWpRhsx6wje8yuts/ko6iB97deMhZXBGa/BlfyBXAnC2KQWOuNK+wYpRiUW5el9vm32TEcO7KvUxWHL8OdKf7EE0KprxBN3YH9Ij5xxw+CDImFHwQZ0urCF5GdROQZEXleRBaKyPhyey8ReVpElorI/SJS09qxgiDoGLTljv8+gKNV9SCUqi0ME5FBAK4FcKOqfhbAW7AB80EQdFBade5pKcJnvYzMDuX/FMDR2BBhMQPAVQBu/eSjvY/UObGT7XIfOfP4iCsW2DH1HFTCQUCADQbpmppewsQjVLaHiszg+86YtXSgKWQ7VZpx62PUQA6ouybaMaPIcccFVwBgCDnv1pyW2lyqGrACv+y5Y+UcwCbccHAOO/IAoCcFju3oxJxcRolHR5IC7Ve9QC1yGLeQ0pIn+NuFJHOH9Eztya/bMTuSzWI6r3a3Y/6RbOewEMrkuZ0OfDYHDgFp8JZ3si1t2uOLSCcRmY9S9ffZAF4FsFZV16duLQdQ36Z3DIKg3WnTwlfVD1V1IEoxkofBxmhuFBE5R0SKIlJsbuYiZUEQtAeb5NVX1bUolQ44HMDuIrJ+q7APAOdhJ6Cqt6lqQVUL3bvvtkWTDYJg69DqHl9EugNYp6prRWRnAENRcuw9AeAbAO5DKROF5WUduiIJYLnnF7bLSEpcOIKCHDyR0ZvZV+AEaRhFX6pI4uXODOBgHOrUlfa2AEB5PDjkt2RTpSAAJRdJJZRwM4qVegGTfDLkFqfPZam5nF4+ypaV4aq1psKNl/7BcTW1lHDjBefwnv46J1nsSurz1fuoQ39nMiSQUUPOj2f5CwJw9ShqID/BGGeZnHdkapPGCWa+Zcd0oeAn6Wv78LVwNqsae06Kyl32Ds7rlrZE7tUBmCEinVD6hfBTVZ0lIi8BuE9EJgB4DrZGVRAEHZS2ePUXADjYaV+G0n4/CILtjIjcC4IMiYUfBBlS5ey8dwEUN5ien4sbH6aXa2c5Y7jMEKumAvahw92pubfZzSBRBAZgQhX+w44YSjE/syctpB6O+qo5EezM85SFuUSTp6ZDwZRnUCDQPKtm9CzZplS1KWsFRw2X+pgsO9jgHHbkAcBEdvixY4u/d8CW7E6zMrtShS0AePdqPr8/JPt5MA1UyY1z6Dyd57kkzjTIXF+A/R6byPYcmg0Vf3NkkU/c8YMgQ2LhB0GGxMIPggxph0o6FYEzA75mu7xKyia9J1AHR4HEBOd4e6dpZJO6To/X7JDVvagPqe6u6mGGzDZJilRieYSj2PrQdGrgPb5TfQevtqEPqd3Oo53nJYeaEY36HrXw/tcrVc3qRhwN5SjDmoQbDs4B7J6e3/v/OGPoMw9Py6y/q6/YId/pk9o3UvWj060zp0k5sYr23hOsUtEg9TK0GApGu4wCjq7joDIgVav2gr0scccPggyJhR8EGRILPwgypMqVdPppsXhnRQtXbfHa+BmqV8mFn33yc2XAZhLzs34v+4eTf/gZvKfmyz4IVoIlPwEAW62G5vbBvnZIZ1besNV+Ac6G5HiBJmcMnxc+B9531lpVW0fN1yQm8TkA7HN69mNwZgxg/TtcLZdVeAEb48EVnrjiMWBVNfga9KpEcbUjrjgE2EpMfFzvXG7oUygchWJxXlTSCYLAEgs/CDIkFn4QZEgs/CDIkOoG8Px5MbCkokRTXy5dBFhHCzvdSDkHgHUueR+LkzfYSeKVXOYkFlJ3MccErPOLFX9PccaMJps+T2dHqUUpsEO+4hyXy3W9T7anVMRlvelcXu9kuVx6BDVwIonnUOPyaXxuAXt+2RnmBWqxc5KvBS4dDtjvmUuWcUIUYK9T/ozspANMINPrH9kuPTlAh4OfvONWfkZH+cch7vhBkCGx8IMgQ2LhB0GGVHePv1NfoO+UigZHSOGu4ak9qpk6kHIsAJMI4waztBYY4QRTPJUmeKCRSnQ/7pSmHkIlBBdQUs4ALwCDBYrZdsYIfx4rFmH3pnS+R/+nHTKVz+U5qXmpF/DFQTJ8bh0hEa5ww2q4AEwgECXcYJYn7MyXNAf5eNcP+VQWUKlwjtUBgBl0Xsx3NNMZRAIfPT1/AyUi1VMlnRWsCAykn6mL87ol7vhBkCGx8IMgQ2LhB0GGVFmI4yMk+8z5w22XUVR5xtTi9PZO/EzVK4vDIg703LhlZzukkZ+p0jFO4xKpwPnL0z39FKWqtlx9FoAtcdOGPbJJGPLENkkCcv95qb3YCkPMkTQhaDAX2xnMyTUAsLYV24kXYI0Np8INC2NaEQ0vGYv3zbyndyopr6SclgF0Dc6w/pMDJRUx4St5oq40Y1qkLrFr9EQ7F/5MKzjWxStDV5kI1rZKOnHHD4IMiYUfBBkSCz8IMiQWfhBkSJWde+8jUahx/9m5nGxWLWFHGABcRDYr6gLW4URKODWeeim91wnkzDvPjphihFfI2TT6x3bQ1OnUwPN3HGp6RWrLUNuHnZyL90rt8WeZEYP/ixre5R6srgOYxJjT703tO51krC7Ux5SqdircGDVcT5mXE27IWcaOPACo46AkCnx5XOiuAAAMLUlEQVQ6cIkZ8gLHPnEu2e11YGqu4RZWhQJM4tESKsXel0r4AEBLhaOxjYJacccPggyJhR8EGRILPwgypB32+BUVYAZwAg6A17undk9OxHCCQfajRJhlXrUXjhi5jWxHiOMtCurhtz7aeZtuX6aGMak51QsuIiXblaeldp0TnCPTqcETsmhFpfb0f7NDulHgSTdPbIShoJg7KVEJ5FsAgCE9qYGFOQCT1MIVblwlWxLR4IQbDs4BYJOZyI+xp+cXIMGLOg4m4usNsH4mT7G4IbFeYHeDOuIjle4E9tFshLjjB0GGxMIPggxp88IXkU4i8pyIzCrbvUTkaRFZKiL3iwj/vguCoIOyKXv8MQAWAehWtq8FcKOq3ici/wTgTAC3fvIhdkIqxHiD7cJb7wmcdMGVXQAsO4oaPIEDooWeCdc41XKvIpvzYjy9j4soq0Woqm3L63ZMzc2pbTQYvSSdham55nbbpZaEQ6iq8BLWlATQV9kvwL4DFuoA8Nbbqc26kzdyEhWAyXQexniXIiXHcNXaO70KN7QpZhENJ+HGPKfnPf2/ew/H55NNfoGF37ZD+vO90XuO3zuxTBiFl6T25pUb/i7caV93aNMdX0T2QUmi5o6yLSi5tn5W7jIDwPFtescgCNqdtv7Uvwml/Mb196JPAVirqutdvsth82cBACJyjogURaTY3Ny22t1BEGxbWl34IjIcwGpV5VjINqGqt6lqQVUL3bt7OeNBEFSbtuzxGwEcJyLHoLTB7gZgMoDdRaRz+a6/D6waRhAEHZRNKpMtIoMBXKKqw0XkAQAPVjj3FqjqlE8aXyj012Lx/g0N11rVFVz+XGrPPzi1B3oqovy2XMkFsP/G0Q+Yu2zCCkZdRQ2knjOc5gbgchK/vVbPph6eyisHKb1JtlcanCvGeA7AxsR6Vs5N7EPVJs+MkDQR6SHKpcFJ/HkA6wAkB6Eus0Omku343BooH6VJH6Me3nnh+0+6A2XlHMBJuOHgHLec+MDEOktSh+Ad+qQZsVCOTOz+65zDmvLnXGbdK+e+YS6FwpdRLD63TctkXw7gYhFZitI38JMtOFYQBFVkk0J2VXUOgDnlv5cBOGzrTykIgm1NRO4FQYZUOUmnBUnUy9e8PjSle+jlgY5KqtmDeZVvOVGEEiQ4fwIAJ0yY/aOz3bI7eBau8B5psvBGa5VpgFYFJwDwhzrUnO9OZoSpRcwqu7s7gULD2O9CSVK3Ont8FijmgjfwwrvYd+OdS44eSivcOLrO9ns0CTdWfOQsSf07dxhfmU1uMmf7CtMFmMSfiZNyPnQGrW3ldUvc8YMgQ2LhB0GGxMIPggyp8h5fkex9+l7l9KFnwtfxxrTBGcOVSr2PxYk7JIhxIW9mAcwhUY3F9HoTP1f2nixTMscc++wfgz3RyEq8vSzv+1mkFMB42kjPfI862Cq8c/UdauHzxqIVgM1eIs7nzCsAr5Lgykx+dg6cwA0T9kjtsT2cN+MqwmlSi1fhxgpjkoiGk3Bjn9Pznt5eg/srf89e5WQ+3/x5vMSkyrgJk+HlEnf8IMiQWPhBkCGx8IMgQ2LhB0GGVNm59yaSqjDTfm67nEGJI+PSxBJwdRUAeI0UYHrd5Lw3O0moQs/sBXbI0L9P7cG3pPYXhpkhH1KOUSeluQz+tDO3c1JzEX0eEgsCALC/6ccTbR8edwKpBs882Qx5VXZN7N5e4RyGP9IPyP6iM4aVcbrsYbrMpa9skHJZby+w6S/JTpV6uVQ14FW4oWguo5xjE244OMc68gDjmLvSyaWZ+FVqGE12k3PcSmfkaud1S9zxgyBDYuEHQYbEwg+CDKnyHr8GSQLNGV5W7+TU5P3X1V+3Q3pxBV3ezwPA58mmhBuKJykxl2zao91uRYU7HXIitfA+1KnycwMpo17clzo4yrY3k/jIL2xFV4y8imxS/DWJPkBvcjegL/lc+nB2DYDryZ44KbX3vcSOYbFh4c8MDDIJKsyxznEpgKVnGhBTo/z9AFbtlivcWDXc/uvIJ2QSbpzgHN7TT/REcLjy891kewFUlV/aHOd1S9zxgyBDYuEHQYbEwg+CDImFHwQZskkqu1tKobC3FouVtaaHOL1Y/YQDdry6VSPJ9koUs1Pts2Tv5oxhPZoHyfZKO7Mzj3VkWDUVMNlta55J7Vonowz3k93g9FlKNpeV9qJzyCn1Qb/U7szZe4D9jsal5ntOcNTOHPXD5xaw55LLp3mlHriNv3dPqYidiA1kz4CFlX44g5IdqUDbvjMKLDP5nt552nBeCoXTUSwu2qYqu0EQbKfEwg+CDImFHwQZUuUAnl2R7jOdOs0HUuLLC9Opg7ev4320tw/i8s+smMKvAxhDmqwcu/JrpyTxP5C9CysI9YaFgjZqeU/pKfA0kn2K04f3i+RL2N0JrFlLlVw6N1MH5zwZVeBvpebOTjn02+lkns0+F8AkpFxGlZeum+WMoSSdenqfFbZ6EJacmpgvUIyMLVUNDFJWFWI/gRdExgk3HJwD2D09V1Ui5SgA6fXBwWw+cccPggyJhR8EGRILPwgypMp7/K5I92C2zmYLVTWpMVP0xBfSPeQLYqvwHsi5DYs5KcSp+ck5OJMHpPYK5/n0LlTi1fgSeG8OYOC+qT2fRBxWOkocddMTc6rYJJ3Ryp+Jnj17e3yzV+VYBls+6GZpSewLlMUwOJ4AwNn7UYMXe0GVc667lF73VGppb71iFL3uxGv0Tb+zA5XPwUxY+DxwBRtPDbeJ7NYSbgC7p3fKCifn1y1PZYg7fhBkSCz8IMiQWPhBkCGx8IMgQ6rs3Ps9UufFD02PGv1bauGSQta5MUbSQJvJ+prz3uwUbCLbCZJpoeCV90im51xbQss6nLzgG2L+H6iBgnzqOIgGwGPpXEarE5hinGoNqdn0M2cMq800kW1Vby5QUkVaQc7KepIeBmCDi+qdPg1kc5KR5+jlS5oLlztJUi2HpjYL8b55pfM+A8nmOuteoBM7MNmR543j68dxlCYJZzs6r1vijh8EGRILPwgyJBZ+EGRIVYU4RKQZpY1+LWykSEdle5orsH3Nd3uaK7B9zPczqupqRldS1YX/8ZuKFFW1UPU33gy2p7kC29d8t6e5AtvffD+J+KkfBBkSCz8IMqS9Fj6rGHRktqe5AtvXfLenuQLb33w3Srvs8YMgaF/ip34QZEhVF76IDBORl0VkqYiYMoPtjYhME5HVIvJiRdueIjJbRF4p/3+P9pzjekRkXxF5QkReEpGFIjKm3N5R57uTiDwjIs+X5zu+3N5LRJ4uXxP3i0hNe891PSLSSUSeE5FZZbvDznVTqdrCF5FOAG4B8BUABwA4WUQOqNb7t5HpAEjtE1cAeFxV+6BUOaKj/IP1AYDvquoBAAYB+Nvy+eyo830fwNGqehBKge7DRGQQgGsB3KiqnwXwFoAz23GOzBgAiyrsjjzXTaKad/zDACxV1WWq2gLgPgAjqvj+raKqvwHwR2oegQ2lVGYAOL6qk9oIqrpSVeeV/34HpQu0Hh13vqqq6zNOdij/pwCOBrA+W6jDzFdE9kEpI+mOsi3ooHPdHKq58OsBVKahLYefktXR2EtV19ewWgVgr/acjIeINAA4GKVUxg473/JP5/kAVgOYjVKdqbWqul6frCNdEzehlNr3Udn+FDruXDeZcO5tAlp6BNKhHoOIyC4oFRL4tqq+XflaR5uvqn6oqgMB7IPSL0DOAe4QiMhwAKtV1Svi8L+CaubjrwBQmai9T7mto/OGiNSp6koRqUPpbtUhEJEdUFr0d6vqekXIDjvf9ajqWhF5AsDhAHYXkc7lO2lHuSYaARwnIsegVJGyG4DJ6Jhz3Syqecd/FkCfsme0BsBJAB6u4vtvLg9jQ2mYbwF4qB3n8jHlPedPACxS1UqZ4Y463+4isnv5750BDEXJL/EEgG+Uu3WI+arqlaq6j6o2oHSd/lpVT0EHnOtmo6pV+w/AMQCWoLS3+14137uN87sXwEoA61Daw52J0t7ucQCvAPgVgD3be57luX4RpZ/xCwDML/93TAee7wAAz5Xn+yKAceX2/QA8g5K0zAMAdmzvudK8BwOYtT3MdVP+i8i9IMiQcO4FQYbEwg+CDImFHwQZEgs/CDIkFn4QZEgs/CDIkFj4QZAhsfCDIEP+B5+8nBQiiDlMAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(r, cmap='hot', interpolation='nearest')\n",
    "plt.show()\n"
   ]
  }
 ],
 "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": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}